Featured

14th IISTL Colloquium on New Technologies and Shipping/Trade Law

Sponsored by 

The Institute’s 14th Annual Colloquium will be held on 10-11 September 2018. The subject of this year’s event is new technologies and their present and future effect on shipping and trade law.

14th IISTL Colloquium

To register for this event, please visit our Eventbrite page.

Continue reading “14th IISTL Colloquium on New Technologies and Shipping/Trade Law”

Force majeure and counterfactuals

A nice force majeure issue — and one of considerable importance — came up before Teare J yesterday in Classic Maritime v Limbungan [2018] EWHC 2389 (Comm), argued by IISTL stalwart Simon Rainey QC. Imagine you conclude a contract (in this case a CoA under which you have to provide a number of iron ore cargoes) which in the event you can’t and don’t perform, and never could have performed. An exemption clause in the contract says that if you could have performed it but a force majeure event X (inundations in Brazilian iron ore mines) then occurs that stops you performing it, you are not liable for breach. Event X occurs. Are you (a) in breach of contract, (b) on the hook for substantial damages?

On (a) the answer is Yes. You promised to perform, you haven’t performed, and because you never could have performed in any case you can’t shelter behind the exemption clause.

But what about (b)? There are two ways to look at this. One is to say: this is a simple case of unexcused non-performance, and hence you must be liable to the shipowner for his lost profits on the carriage, a figure amounting to many millions. The other point of view runs thus. If, counterfactually, you could have performed but for X, the shipowner would in the event have had no claim to performance because of the exemption clause. Hence hence it’s no skin off his nose that you didn’t perform, and damages are nominal only. Teare J plumped for the second: nominals only.

This view is highly plausible and for the moment clearly represents the law. It also dovetails quite nicely with the general rule in cases such as The Golden Victory [2007] UKHL 12, [2007] 2 A.C. 353 and Bunge v Nidera [2015] UKSC 43, [2015] 2 CLC 120, that in assessing damages we take into account later events that would have taken away the right to demand performance.

But this case, or the issue in it, may go further. There is a respectable argument, that certainly can’t be dismissed summarily, which suggests a different answer. In so far as the inability to rely on a force majeure clause is due to a party’s own default, which was the case in Classic Maritime, should it be open to that party to argue that if he had acted differently he would have been able to invoke that very same clause? Suppose a force majeure clause requires notice to be given within 7 days after the force majeure event; a party prevented by force majeure nevertheless fails to give notice for 10 days, and thus loses the protection of the clause. Is it really open to the party then to say that if he had given the proper contractual 7 days notice he would have been protected by the clause, the counterparty would have had no right to demand performance, and hence damages are nominal only? I’m doubtful. And I’m equally not sure that this scenario is that different from what happened in Classic Maritime.  It’s just a thought. Whether it’s a good one, only time will tell.

Brexit update.

 

July saw the passing of the European Union (Withdrawal) Act 2018 which will repeal the European Communities Act 1972 (ECA) but retain EU legislation as part of UK law. The Act will come into effect on “exit day” which has been defined as 29 March 2019 at 11.00 p.m. Directly applicable EU law, which is currently given effect through s.2(1) of the ECA is referred to as ‘converted legislation’ and covers: EU regulations; EU decisions; EU tertiary legislation; Direct EU legislation as it applies with adaptations to the European Economic Area; and any other rights which are available in domestic law by virtue of section 2(1) of the ECA, including the rights contained in the EU treaties, that can currently be relied on directly in national law without the need for specific implementing measures. This corpus of EU law will be converted and incorporated into UK law immediately before exit from the EU. The second category of retained legislation is ‘preserved legislation’ which comprises: regulations made under section 2(2) or paragraph 1A of Schedule 2 to the ECA; other primary and secondary legislation with the same purpose as regulations under section 2(2) ECA; other domestic legislation which relates to the above, or to converted legislation, or otherwise relates to the EU or EEA. This corpus of legislation will be preserved as it exists immediately before exit from the EU.

Section 6 of the EUWB sets out the relationship between the CJEU and domestic courts and tribunals after exit. The validity, meaning or effect of any retained EU law is to be decided in accordance with any retained case law and any retained general principles of EU law, and having regard to the limits, immediately before exit day, of EU competences. Decisions of the CJEU made after exit day will not be binding on domestic (UK) courts. Domestic courts cannot refer cases to the CJEU on or after exit day and are not required to have regard to anything done by the EU or an EU entity on or after exit day. However, domestic courts, when interpreting retained EU law, will be able to consider post-exit EU actions including CJEU case law if they consider it appropriate. The UK Supreme Court (UKSC) and the High Court of Justiciary (HCJ) are not bound by either retained general principles or retained CJEU case law. In deciding whether to depart from any retained EU case law, the Supreme Court or the High Court of Justiciary must apply the same test as it would apply in deciding whether to depart from its own case law.

Section 7 gives ministers delegated powers to correct operability problems in converted and preserved legislation by way of statutory instrument, and to transfer the functions of EU authorities to UK public authorities and of creating new UK public authorities to take on those functions.

The Act means that most existing EU law will continue to apply as domestic law after exit day, although civil servants are going to have a demanding task making the necessary amendments to make sense of this transition – for example, references in the 2012 Brussels Regulation (recast) on Jurisdiction and Judgments will need to redefine ‘Member state’ so as to include the UK. However, there is no reciprocity in this exercise and this will be particularly felt with regards to the provisions of the Brussels Regulation on reciprocal enforcement of judgments. The UK will still enforce a French judgment under these provisions but France will no longer return the favour. It is likely that some reciprocity will be regained by the UK applying to accede to the 2005 Hague Convention on Choice of Court Agreements which would take about three months, although the process cannot be initiated until after Exit Day. The current parties to the Convention are the EU, Singapore, Mexico. The UK is currently party to the Convention through its membership of the EU but will cease to be a party on Exit Day. Many areas of maritime law fall outside the Convention. Article 2 excludes, inter alia, e) insolvency, composition and analogous matters; f)  the carriage of passengers and goods; g) marine pollution, limitation of liability for maritime claims, general average, and emergency towage and salvage; h) anti-trust (competition) matters; i)  liability for nuclear damage; j) claims for personal injury brought by or on behalf of natural persons; k) tort or delict claims for damage to tangible property that do not arise from a contractual relationship;

Arbitration will be unaffected. Enforcement of awards is subject to the 1958 New York Convention. The EU Regulations on choice of law in contract and tort, Rome I and Rome II, will also continue to function in EU courts as per usual as their choice of law rules are of universal application and are not tied to EU Member States.

After Exit Day the UK will be free to amend or repeal parts of this domesticated corpus of EU law. Two pieces of legislation immediately come to mind as candidates for this exercise. First, the universally unpopular  Regulation (EU) 2017/352 on Port Services which is due to come into effect five days before Exit Day. Second, the Brussels Regulation (recast) on Jurisdiction and Judgments could be amended, by including arbitration proceedings in art. 25, so as to permit once again the use of anti-suit injunctions to restrain proceedings commenced before the courts of an EU Member State in breach of a submission to arbitration in the UK.

A tale of two targets. UK government sees off judicial review challenge to its climate change targets.

 

A recent little reported decision on the most important issue of our times (no, not ‘Brexit’). In Plan B Earth v Secretary of State for Business, Energy and Industrial Strategy and the Committee on Climate Change [2018] EWHC 1892 (Admin) an application was made by the claimants to apply for judicial review, brought by Plan B Earth and eleven other claimants, of the refusal by the Defendant (“the Secretary of State”) to revise the 2050 carbon target under the Climate Change Act 2008 (“the 2008 Act”) at the present time.

Under article 2(1) (a) the 2015 Paris Agreement on Climate Change which the UK has ratified the parties commit to: “Holding the increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial limits, recognising that this would significantly reduce the risks and impacts of climate change.” The Paris Agreement also aims for net zero emissions in CO2 by the second half of the century.

Under the 2008 Climate Change Act a duty is imposed on the  Secretary of State to ensure that the net UK carbon account for the year 2050 (“the 2050 target”) is at least 80% lower than the 1990 baseline.  Section 2(1)(a) confers a power on the Secretary of State by order to amend the 2050 target by amending the 80% figure.  The power may be exercised if it appears to the Secretary of State that there have been significant developments in— (i) scientific knowledge about climate change, or (ii) European or international law or policy, that make it appropriate to do so.” Before exercising the power the Secretary of State is required by s.3(1)(a) to obtain, and take into account, the advice of the Committee on Climate Change (“the Committee”), an independent body composed of experts, established by s.32 of the Act.

In response to the Paris Agreement, the Committee advised in October 2016 that no change should be made to the 2050 target at this time.  As regards the Paris Agreement the Committee stated

“In line with the Paris Agreement, the Government has indicated it intends at some point to set a UK target for reducing domestic net emissions to net zero.  We have concluded it is too early to do so now, but setting such a target should be kept under review. .. Emissions pathways indicate that CO2 emissions will need to reach net zero by the 2050s-70s, along with deep reductions of all other greenhouse gases, in order to stay below 2°C.  To stay close to 1.5°C CO2 emissions would need to reach net zero by the 2040s. … We currently have no scenarios for how the UK can achieve net zero domestic emissions. …”  The UK’s 2050 target under the 2008 Act was challenging but could be met in various ways using currently known technologies. Current policy in the UK was not enough to deliver the existing carbon budgets that Parliament has set.  The Committee’s assessment was that current policies “would at best deliver around half of the emissions reductions required to 2030, with no current policies to address the other half.  This carbon policy gap must be closed to meet the existing carbon budgets, and to prepare for the 2050 target and net zero emissions in the longer term.”

Supperstone J held that the claim was not arguable and refused the application. The Secretary of State had not misunderstood the Committee’s advice and the Committee’s advice did not misunderstand the Paris Agreement. It was not arguable that the Secretary of State’s refusal to amend the 2050 target was an unlawful exercise of his discretion. The Secretary of State was correct in understanding that the Paris Agreement does not impose a binding legal target on each specific contracting party to achieve any specified temperature level by 2050.

The applicants plan to appeal. Later this year the IPCC will issue its special report on global warming of 1.5C to the UK government for comment. The Secretary of State will seek the Committee’s advice “as soon as is reasonably practicable following publication of the final IPCC report”.

Elsewhere, on May 23, 2018 a similar challenge was made against the European Parliament and the Council before the EU General Court (“EGC”) to urge EU institutions to adopt more stringent greenhouse gas reduction targets than those set out in its 2030 climate and energy framework which commits the EU to a reduction of overall greenhouse gas emissions by 2030 by at least 40% compared to 1990 levels.

A point of interest. In 2017 the year following the ratification of the Paris Agreement the International Energy Agency reported that energy related emissions had climbed 1.4 percent to 32.5 gigatons.

A case of “Lord make us carbon neutral, but not yet” (pace St Augustine)?

 

 

Incorporation of Inter-Cub Agreement into time charter. Does it bring in the ICA’s security provisions?

 

If your charter incorporates the provisions of the Interclub NYPE Agreement 2011 don’t assume that everything in the agreement is brought into the charter. It all depends on the wording of incorporation, as owners found out in London Arbitration 18/18.   Owners sought counter-security from charterers pursuant to the provisions of cl.9 of the ICA 2011. Clause 35 of the time charter provided “…Liability for cargo claims, as between Charterers and Owners, shall be apportioned/settled as specified by the Interclub New York Produce Exchange Agreement effective from 1996 and its subsequent amendments  (Tribunal’s emphasis).”The Tribunal held that the incorporating words in the time charter brought into the charter only those parts of the ICA relating to apportionment and settlement.  The incorporating words did not bring in the ICA’s provision relating to security for claims.

Charter time bar not read  into charterers’ letter of indemnity.

 

In Navig8 Chemicals Pool Inc v. Glencore Agriculture BV (The Songa Winds) [2018] EWCA Civ 1901, the Court of Appeal had to consider whether a time bar clause in a voyage charter operated to bar claims under a letter of indemnity issued by charterers to owners in respect of delivery of cargo without production of bills of lading. Clause 38  of the charter required the owners to release the cargo  against charterers’ letter of indemnity  (LOI) if bills of lading were not available at the discharge port and the period of validity of any letter of indemnity was to be three months from date of issue. The LOI provided by charterers did not refer to any time bar and provided in clause 5 “. As soon as all original bills of lading for the above cargo shall have come into our possession, to deliver the same to you, or otherwise cause all original bills of lading to be delivered to you, whereupon our liability hereunder shall cease.”

Two points arose. Was the LOI provided by charterers subject to the three month time bar provided in the charterparty? If so, how did that time bar run?  Andrew Baker J, [2018] EWHC 397 (Comm) ,had found that the LOI was not subject to the time bar, and that the effect of the time bar was that the period within which the requested delivery of the cargo must take place was three months from the date of delivery.

The Court of Appeal has upheld the first finding. Charterers had the contractual right to insist that the LOI incorporated the terms set out in cl.38 but they had failed to do so. Clause 5 of the  LOIs was a self-contained provision which confines charterers’ liability, and which containeds no reference to any extraneous term which might impact on the time limit of that liability.Their LOIs were distinct agreements to the voyage charter, setting out self-contained obligations and rights which could be relied on by third parties, such as owners’ agents, as against charterers.

On the second point, however, the Court of Appeal rejected the judge’s construction  of cl. 38 and found that cl.38 meant that there was a time limit of three months for making claims.

Demurrage time bar and submission of documents. Fault of owners and consequential loss of time.

 

When dealing with a demurrage time bar clause in your charter, it pays for owners to know who charterers’ agents are, as shown by London Arbitration 19/18. Owners chartered two vessels under a contract of affreightment for a series of voyages. The charters provided for the demurrage claim and supporting documents for each voyage to be sent by owners to charterers within 30 days starting from next day after discharging at discharging port. Failure to supply the documents within this period would free charterers from any responsibility for demurrage. The charter also provided that owners were to pay commission on F/D/D to a firm of brokers, D.

Charterers declined to pay demurrage accruing on eight voyages as they had not received the documentation within the 30 days. Owners claimed that they had passed the documentation to D within the 30 days and that D were charterers’ agents. The tribunal found that D were an intermediate broker who had no principals and who could look to either charterers or owners for their commission. The charter had included a specific provision to cover their commission, making owners responsible for it. D’s duty was simply to pass messages up and down the chartering chain. There was no evidence that charterers held D out as their broker and they were clearly not charterers’ actual broker. Receipt of documents by D did not amount to receipt by charterers or their agents.

The tribunal rejected owners’ argument that D had ostensible authority to act as charterers’ agents. The fact that the charterers had paid all freight claims and many of the demurrage claims which were sent by the owners to D alone did not amount to a representation on the charterers’ part that D had authority to accept all demurrage claims under the charter on their behalf as their agent. Those claims were paid, not because they were received by D, but because they were passed by D to C and it was receipt by the latter that triggered payment by charterers. Nor was there any estoppel, either by convention or promissory. The parties proceeded on the basis that charterers would pay all demurrage claims received from D within the 30 day period. The fact that charterers had paid a demurrage claim presented out of time on one voyage out of the 87 performed did not amount to a waiver of their right to rely on the time bar clause.

The tribunal also rejected charterers’ claim on another voyage to suspend laytime for consequential delay on arrival at the discharge port because of a delay on completion of loading of nine days due to engine problems. For culpable fault resulting in time not counting, the fault and the delay had to be co-extensive.

Relective loss and the unsecured creditor.

 

Garcia v Marex Financial Ltd [2018] EWCA Civ 1468 is a cautionary tale of a creditor with a judgment against two companies being thwarted by their beneficial owner removing their assets from the jurisdiction before the judgments could be enforced. The creditor decided to sue the wrongdoer for the torts of knowingly inducing and procuring the two companies to act in wrongful violation of your and causing loss to you by unlawful means. But not so fast, what about the rule against recovery of ‘reflective loss’ established by the House of Lords in Johnson v Gore Wood [2002] 2 AC 1? The rule states that where a company suffers loss caused by a breach of duty owed to it, only the company may sue in respect of that loss. A claim is barred if the loss suffered by the claimant would have been made good by restoration of the company’s assets. The rule is subject to an exception in Giles v Rhind [2003] Ch 618, where as a consequence of the actions of the wrongdoer, the company no longer has a cause of action and it is impossible for it to bring a claim or for a claim to be brought in its name by a third party. Here, there was a clear breach of duty owed to the company by the beneficial owner who had  asset stripped them.

But does the rule also apply to unsecured creditors of the company who are not its shareholders? Until now this question was undecided, but not any longer. In Garcia v Marex Financial Ltd [2018] EWCA Civ 1468, the Court of Appeal have decided that the rule does apply to unsecured creditors of the company who are not its shareholders. On the facts the rule barred the action against the tortfeasor and the Giles v Rhind exception did not apply as the wrongdoing had not made it impossible for the companies to pursue a claim against the wrongdoer.

 

Management of the vessel, or management of cargo? Effect of s4. of US COGSA on charterers’ claim for costs of unnecessary strapping required by master.

Clearlake Shipping Pte Ltd v Privocean Shipping Ltd (15 May 2018. QB D (Com Ct) is an unreported decision of Cockerill J on the effect of cl.2 of NYPE 1946 form and s.4(2)(a) of US COGSA 1936 which is applied as a paramount clause. Charterers incurred extra expenses due to unnecessary strapping insisted on by the master with a view to the ship’s stability. The master insisted on the strapping in order to ensure the stability of the vessel. In the arbitration the charterer produced expert evidence that the cargo strapping had been unnecessary and that adequate stability could have been achieved by distributing the cargo differently or by ballasting. The arbitrators found that the master had been negligent and in breach of cl. 8. However, they rejected charterer’s contention that the cost of strapping was for owner’s account by virtue of cl.2 of the charter which provided that “Charterers are to provide necessary dunnage and shifting boards, also any extra fittings requisite for a special trade or unusual cargo…”  The shipowners, though, had a defence to the claim under s.4(2) of the incorporated US COGSA 1936, since the neglect or default of the master was “in the management of the ship”.

On appeal Cockerill J upheld both findings.(i)  Clause 2 said nothing about the position where the charterer had paid for a fitting that turned out to have been unnecessary. (ii) The master’s default was in the management of the ship and owners had a defence under s.4(2) of the incorporated US COGSA 1936. The master’s breach was not any lack of care for the cargo during loading or discharge. His intervention came before loading. Since his action in requiring the cargo to be strapped was directed at the safety of the ship it was an act in the management of the vessel within the s.4(2) exemption. It was also clear that safe stowage without strapping could have been achieved by ballasting, and the same result should be reached whether the issue was one of different distribution of the cargo or of ballasting. Ballasting would be a matter in the management of the vessel and it followed that for that reason also the exemption from liability applied.

The case provides a salutary reminder to time charterers that they may be getting more than they bargained for with a clause paramount. The US COGSA exceptions in s4(2) and the Hague Rules exceptions in art IV(2), are not limited to breaches in respect of the activities listed in s2/art II. As stated by Robert Goff LJ in The Satya Kailash [1984] 1  Lloyd’s Rep 588, 596.

[o]n the approach of the majority of the House of Lords in the Adamastos case, even such general words of incorporation can be effective to give an owner the protection of the statutory immunities in respect not merely of those matters specified in s. 2, but also of other contractual activities performed by him under the charter.”

 

All change for financier assignees — second time lucky with anti-anti-assignment provisions?

The good times seem likely to end finally on 31 December this year for anti-assignment clauses. The Government has published the draft Business Contract Terms (Assignment of Receivables) Regulations 2018, which for SMEs essentially invalidate anti-assignment clauses affecting receivables — i.e. sums payable for goods or services supplied. A few pointers:

1. The prohibition is not limited to assignment to financiers: assignment to debt-collectors, etc, also seems to be protected.

2.  There are anti-avoidance provisions. Any attempt to put conditions on the assignability of receivables is outlawed. The blurb states that a set-off clause is not such a condition: this may be important where, for example, a contract allows set-offs that would not otherwise be pleadable against an assignee. On the other hand, there is some doubt about this: the Regulations do not contain any such provision, and the blurb, of course, is not part of them.

3. There are exceptions. These include financial services, swaps, energy futures, petroleum licences, public-private partnership projects and contracts with national security implications. Importantly there are also two other carve-outs. One is contracts where one or more parties is not acting in the course of a business. This means consumers can, if there is a suitable term, continue to refuse to deal with an assignee. Another is contracts which neither party entered into in the course of a business here: so genuine international contracts remain subject to the old freedom of contract rules. Perhaps suprisingly, rental contracts are also excluded, except when connected with certain forms of financial services.

All in all, these seem an improvement on last year’s regulations (not difficult). As to their effect we’ll have to wait and see.

Can an actual carrier rely on a circular indemnity clause in a multimodal bill of lading?

 

Yes, they can, in the US.

In Royal Smit Transformers BV v Onego Shipping & Chartering, and others Case 17-30543, the US Court of Appeals for the Fifth Circuit  on 2nd August 2018 held that actual carriers could rely on a circular indemnity provision in a ‘Himalaya’ clause in a through bill of lading as a complete defence to a claim brought against them by the shipper under that bill. Royal Smit contracted with Central Oceans USA to ship its transformers from the Netherland to Louisiana. The arrangement was established by a multimodal through bill of lading between the parties. Central Oceans made separate contracts with actual carriers for the three legs of the journey; the sea voyage to New Orleans; rail carriage to St Gabriel; truck carriage to the final destination. The actual carriers were not involved in the multimodal bill of lading and Royal Smit were not involved in the contracts made by Central Oceans with the actual carriers.

The bill of lading contained a ‘Himalaya’ clause which provided:

  1. Defenses and limits for [Central Oceans], Servants, etc.

. . . .

(b) [Royal] undertakes that no claim shall be made against any servant, agent, or other persons whose services [Central Oceans] has used in order to perform the Multimodal Transport Contract and if any claim should nevertheless be made, to indemnify [Central Oceans] against all consequences thereof.

(c) However, the provisions of this Contract apply whenever claims relating to the performance of the Multimodal Transport Contract are made against any servant, agent or other person whose services [Central Oceans] has used in order to perform the Multimodal Transport Contract, whether such claims are founded in contract or in tort. In entering into this Contract, [Central Oceans] . . . does so not only on his own behalf but also as agent or trustee for such persons.

The transformers were delivered to the final destination in January 2016, where an inspection revealed that the transformers had been damaged by “excessive vibration” somewhere along the journey. When Royal Smit sued the three actual carriers they relied on cl.15(b), the ‘circular indemnity’ provision, by way of complete defence. The District Court agreed. . Relying on two Supreme Court opinions, Norfolk Southern Railway Co. v. Kirby, 543 U.S. 14 (2004) and Kawasaki Kisen Kaisha, Ltd. v. Regal-Beloit Corp., 561 U.S. 89 (2010), the court concluded that “actual carriers who fall within the scope of Himalaya Clauses can rely on those clauses to limit their liability.” This was the case with the particular Himalaya provision relied on by the actual carriers in the present case. The position was not affected by the separate contracts negotiated by the actual carriers with Central Oceans, such as the non-negotiable bill of lading issued by the sea carrier, Onego.

The Fifth Circuit has now upheld this decision. The clause did not fall foul of s.3(8) of  COGSA 46 U.S.C. § 30701.  A Himalaya Clause that protects downstream carriers from suit by a cargo owner did not, in and of itself, limit the cargo owner’s ability to receive the recovery to which it is entitled. Royal had agreed with Central Oceans to a COGSA-authorized damages limitation and the mere fact that it must recover its remedy only from Central Oceans did not prevent it from receiving the full measure of that bargain. Nothing in the Himalaya Clause precluded Central Oceans from suing the defendants to recoup its losses from Royal.

The Court also rejected Royal’s arguments that it had never agreed to be bound by the Himalaya clause in the bill of lading. Royal had based its claim on the bill of lading and had therefore accepted the terms of the bill of lading including unnegotiated clauses (Mitsui & Co. (USA), Inc. v. Mira M/V, 111 F.3d 33, 36 (5th Cir. 1997)). Furthermore,  basic principles of maritime law governing the interpretation of contracts foreclosed Royal’s argument that the court should look to extrinsic evidence to discern an intent contrary to the plain text of the bill of lading. Only if the written language of the document was ambiguous, which was not the case here, could the court look beyond the written language to determine the parties’ intent.

The decision is good news for sub-contractors under multi-modal bills of lading involving carriage to the US. The position regarding such clauses under English law is somewhat different. First, the circular indemnity provision does not provide a defence to the actual carrier but provides a right to the carrier under the bill of lading to step in and prevent suit against its sub-carrier by the shipper/holder of the bill. Second, where an actual sea carrier is involved, art III(8) of the Hague Rules has been held to render null and void a ‘Himalaya’ clause giving the sub-contractor a complete exemption from liability (The Starsin  [2004] 1 AC 715 ) and the same is probably the case with a circular indemnity clause.