The blog post that follows was posted following the decision of the Court of Appeal in PST Energy 7 Shipping LLC v O.W. Bunker Malta Ltd [2015] EWCA Civ 1058 (‘Bunkers’).  The appeal to the Supreme Court has been expedited, and will be heard next week on Tuesday 22nd and Wednesday 23rd March.  Not only is the case of great importance for the maritime world, it has very significant repercussions for the law on retention of title clauses and the sale of goods, as explained below.

Yet again the toxic mix of a claim under section 49 of the Sale of Goods Act 1979 (SGA), a retention of title clause and the Court of Appeal has yielded unfortunate results for the law on sale of goods.   Only two years ago, the Court of Appeal (by a majority) decided that a contract for the sale of goods containing a retention of title (ROT) clause was in fact also a contract of agency, so that property did not pass to the buyer on the resale of the goods, but straight to the sub-buyer, with the result that the seller could not sue for the price under section 49 SGA (Caterpillar (NI) Limited v John Holt & Company (Liverpool) Limited [2013] EWCA (Civ) 1232 (‘Caterpillar’), see my note at [2014] LMCLQ 564).  In October, the Court of Appeal has held that a contract containing a ROT clause, whereby goods were supplied with a view to their use and destruction, is not a contract for the sale of goods at all.

For many years, sellers have incorporated ROT clauses in their contracts, to provide them with security (in the functional sense) in the case of non-payment by a buyer. The drafting of such clauses has become rather labyrinthine, in the (usually) vain attempt to achieve the ‘holy grail’ of an interest in not only the goods themselves, but in the products of such goods or in the proceeds of sale.  Such an interest has been repeatedly characterised by the courts as a charge, and is therefore void for non-registration in the insolvency of the buyer: the registration process for company charges is not practically conducive to registering such interests. Ironically, the sellers in Caterpillar appear to have achieved the ‘holy grail’, only to find that it meant that they could not sue for the price.   Sellers must be careful what they wish for….

In Bunkers, the contract in question was at the bottom of a supply chain: this is common in many industries, and particularly so in the supply of fuel oil to ships, known as bunkers. Each contract in the chain contained a ROT clause, with differing credit periods, running from the date of delivery.   Reading down the chain, with the shipowners at the bottom, contracts A (between parties 1 and 2) and B (between parties 2 and 3) were subject to a 30 day period and contracts C (between parties 3 and 4) and D (between parties 4 and 5, the Owners) were subject to a 60 day period. Party 2, who were the party who physically delivered the oil to the Owners) paid party 1 within 14 days. Party 3 became insolvent 32 days after delivery and did not pay party 2. Party 4 (who had not paid party 3) claimed the price from the Owners (party 5).  The Owners, by this time, had used all the bunkers.

The Owners resisted the claim, not because they wanted something for nothing, but because they were concerned that they would also have to pay party 2 (the physical suppliers): in many jurisdictions physical suppliers of bunkers have a direct claim against the party using the bunkers and may have a maritime lien or the right to arrest the ship. The Owners’ argument was that did not have to pay party 4, since (a) property had not passed and so party 4 could not sue under section 49(1) and (b) at the time when property should have passed under contract D, party 4 did not have good title to the goods.  This was, it was argued, a breach of section 12 SGA, and a total failure of consideration, thus there was no liability to pay the price.

The arbitrators, the first instance judge and the Court of Appeal sidestepped this argument by deciding that the contract was not a contract for sale of goods at all.   Therefore, section 49 did not apply, and passing of property was not the essence of the contract, so there was no failure of consideration.   Instead, the Owners had obtained what they contracted for, that is, the right to use and consume the bunkers, and the price was due as a matter of debt.   Moore-Bick LJ described the contract as ‘a contract under which goods are to be delivered to the owners as bailees with a licence to consume them for the propulsion of the vessel, coupled with an agreement to sell any quantity remaining at the date of payment.’   He did, however, say that ‘there is no reason why the incidents of a contract of sale of goods for which the Act provides should not apply equally to such a contract at common law, save to the extent that they are inconsistent with the parties' agreement.’

This analysis was reached on the basis of certain facts and assumptions.  1. The goods which were delivered to the Owners were to be destroyed, probably before the end of the credit period.   2. This was combined with a ROT clause, so that property would not pass until payment.   3. It seems to be assumed that payment would not be made until the end of the credit period. Therefore (the reasoning goes) the purpose of the contract cannot be the transfer of property to the buyer, since at the time when property was to be transferred the goods would have ceased to exist, and could not be the subject of such a transfer.  Since, under section 2 SGA, a contract of sale is one by which the seller agrees to transfer the property in the goods, this contract is not a contract of sale.

It may be that the Court of Appeal analysis could be said to be influenced by the very particular considerations of the market for bunkers, so that the analysis only applies to those specific type of contracts.    However, there seems to me to be a danger that it could be applied much more widely, with disturbing results.

ROT clauses are ubiquitous in contracts for the sale of raw materials and stock in trade.   In relation to the former, the intention is almost always that they will lose their identity in a process of manufacture of products or by being subsumed into a larger object or land.    In many cases this will happen, and is intended to happen, before the end of the credit period.  In relation to the latter, it is almost always the intention that the goods should be sold before the end of the credit period.  Is it the case that none of these contracts are for the sale of goods?  This would limit the scope of the SGA considerably, and introduce a large class of contracts which do not fall within that statute or the Supply of Goods and Services Act 1982 (or any class of contracts recognised by the Consumer Rights Act 2015). Let us call these ‘contracts for the use, destruction and disposal of goods’.    How could anyone be sure which terms are to be implied into such contracts or to what extent jurisprudence applying to contracts for the sale or supply of goods should apply to them?  Whether a contract falls into this category seems to depend on quite specific factual matters: this surely must produce uncertainty.

Until the decision in Caterpillar the analysis of a contract for the supply of stock in trade was that property in the goods passed to the buyer at the moment of sub-sale and then passed to the sub-buyer: this was the analysis adopted by Popplewell J and Longmore LJ in that case, though not by the majority of the Court of Appeal.   Similarly, one might have thought that where goods are consumed or materials used in manufacturing so that they no longer exist, property passes at or immediately before the moment of extinction.   The seller is then left with an (unsecured) claim for the price.   This analysis would have solved the position in the Bunkers case: each contract in the chain was a contract for the sale of goods, with an implied term that property passed (down the chain) when the bunkers were used, since at that moment each seller lost its ‘security’ for the price.  There would then be no problem for the seller (in each contract) relying on section 49 (1) to sue for the price. An implied term to this effect was rejected by Males J at first instance in Bunkers ([2015] EWHC 2022 (Comm) [67]) on the grounds that the contracts provided expressly for the passing of property on payment and no term could be implied which contradicted this.  However, the approach taken by the court in the Bunkers case does far more damage to the express wording of the contract (by completely changing its nature) than implying this sensible term, which appears to have been assumed sub silentio in many ROT cases over the years (otherwise, why weren’t the contracts recharacterised as  contracts for the use, destruction and disposal of goods’?) 

As to the goods which have not been consumed, used or sold at the expiry of the period of credit, section 49 remains problematic. The difficulty here is that, ex hypothesi, property has not passed to trigger section 49(1) but an agreement that the price should be paid a number of days after delivery does not fall within section 49(2) (and the Court of Appeal in Caterpillar decided that section 49 is exhaustive). Ideally, section 49 should be the subject of reform (see, for example, section 106 of the Canadian Uniform Sale of Goods Act, where the seller’s action is limited to after delivery has taken place).  Failing that, the indication from Males J in Bunkers ([73]) that an agreement that payment is required a certain number of days after delivery falls within section 49(2) is very much to be welcomed.

Finally, it should be noted that the analytical difficulties experienced by the courts in the cases discussed in this blog stem from the manipulation of the passing of property in sales contracts as a form of security for payment of the price. One could analyse what is happening as the grant of a floating charge by the buyer in each contract in the chain as security for payment of the price.  On this analysis, the contract would be a contract of sale, property would pass on delivery, permission to resell or to use the goods would be inherent in the nature of the charge, sub-buyers would obtain good title free of the charge and the sellers could bring an action for the price. Of course, under current English law, the seller might lose priority to a preceding secured creditor and the charge would be registrable under section 859A Companies Act 2006.  Therefore, any question of recharacterisation as things stand would be very unwelcome. However, Caterpillar and Bunkers do demonstrate the difficulties of trying to use ordinary sales law to do the job of secured transactions law.  If you try to push a square peg into a round hole, in the end, something has to give.

 

This article was first posted on the Commercial Law Centre blog. The original post can be found here