Lack of unambiguous drafting in a gas sales contract landed three hydrocarbon giants in the Court of Appeal today; it also raised a nice point about damages and counterfactuals.
In British Gas v Shell UK  EWCA Civ 2349, Shell and Esso agreed to supply, and BG to buy on a take-or-pay basis, a minimum daily quantity of gas (appearing in the forest of acronyms typical of hydrocarbon contracts as a TRDQ, or Total Reservoir Daily Quantity). The sellers controlled a couple of reservoirs which, together with others, were connected to the well-known Bacton terminal in Norfolk. As might be expected, gas from all the connected reservoirs was commingled before it came on shore, and the owners of the various reservoirs, including the sellers, had a practice of “borrowing” gas from one another to meet variations in demand. In order to protect BG’s interests, the sellers in addition undertook under Clause 6.4 of the contract to “provide and maintain a capacity (herein referred to as the ‘Delivery Capacity’) to deliver Natural Gas from the Reservoirs” amounting to 130% of the relevant daily quantity. If the capacity was reduced, then the sellers had a right to reduce the TRDQ proportionately.
As capacity in the North Sea ran down, the sellers’ capacity to supply from their own reservoirs dipped below the magic figure of 130%, though if you took into account their capacity to borrow gas the capacity remained adequate. BG saw an opportunity to sue the sellers. It argued that (1) “capacity” meant capacity from the sellers’ own reservoirs, excluding borrowed gas; and (2) had the sellers reduced the TRDQ to 100/130 of the reduced capacity, it would have bought in all excess requirements more cheaply elsewhere.
The Court of Appeal held for BG on (1): capacity on an ordinary interpretation meant capacity from the sellers’ own resources, not third parties’, so that the sellers were in breach. On damages, however, it held that BG had suffered no loss. The sellers had had a right, but no duty, to reduce the TRDQ in line with the total capacity; they had not done so; and the fact that they might have avoided being in breach of the 130% stipulation had they done so was beside the point.
The decision in (1) seems right as a matter of interpretation, and also sensible: apart from anything else, capacity clauses exist to assure certainty of supply, and would be somewhat devalued if they took into account possible arrangements that the seller might enter into with third parties.
The damages point is an awkward one, as is always the case with the fiendish counterfactual question “what would have happened if the defendant hadn’t been in breach?” It turns, it is suggested, on a proper interpretation of the sellers’ contractual obligation. Was it (i) to maintain a capacity to supply amounting to at least 130/100 of the TRDQ, or (ii) to set a TRDQ amounting at most to 100/130 of its capacity to supply (not quite the same thing)? Given the provision that there was a right but no duty to reduce the TRDQ in line with capacity, the latter answer seems correct. If so it follows, at least in the view of this blog, that BG’s claim against the sellers for substantial damages was rightly rejected as a claim for failing to do what they had not been bound to do in the first place.
Just one more thing. Before you file this case away as a useful piece of ammunition on the damages point, remember that in every case of this sort, the answer – and often many millions of dollars – is likely to turn on a careful reading of the underlying contract. A decision on one particular piece of wording may well not be a reliable guide to another.
Crucial measures to further reduce greenhouse gas (GHG) emissions from ships will be discussed by IMO’s Marine Environment Protection Committee (MEPC) met between 16-20 November to discuss measures to reduce further greenhouse gas emissions from shipping.
The IMO’s website notes that the MEPC is expected to adopt amendments to the International Convention for the Prevention of Pollution from Ships (MARPOL) to significantly strengthen the “phase 3” requirements of the Energy Efficiency Design Index (EEDI) – meaning that new ships built from 2022 will have to be significantly more energy-efficient. Those amendments were approved at the previous session of the Committee (MEPC 74) in May 2019.
The MEPC will also discuss two further energy efficiency requirements comprising draft amendments which were agreed by IMO’s Intersessional Working Group on Reduction of GHG Emissions from Ships (ISWG-GHG 7) in October, and would also apply to existing ships:
a new Energy Efficiency Existing Ship Index (EEXI) for all ships;
an annual operational carbon intensity indicator (CII) and its rating, which would apply to ships of 5,000 gross tonnage and above.
If approved at this session of the Committee, they could then be put forward for adoption at the subsequent MEPC 76 session, to be held in June 2021. Under MARPOL, amendments can enter into force after a minimum 16 months following adoption.
Between 4-12 November 2020 the Norwegian Supreme Court heard an appeal from environmental groups seeking the invalidation of the granting of licenses in 2016 to conduct exploratory drilling in the South and South East Barents Sea, an area on the Norwegian continental shelf spanning about 77 acres where oil and gas fields have recently been built. Companies were awarded licenses in 2016 to conduct exploratory drilling in the South and South East Barents Sea, an area on the Norwegian continental shelf spanning about 77 acres where oil and gas fields have recently been built. Parliament approved opening the area for exploration three years earlier.
Their action is brought under the Norwegian Constitution’s environmental provisions, art. 112, which were passed in 2014. They argue that exploratory drilling licenses violate a constitutional right to a healthy environment. They claim the oil-exploration plans were not fully researched before being approved and also rely on a previously unknown expert report throwing doubt on the economic benefit of drilling in the Barents Sea, which was commissioned by the government in 2013 but not passed onto the Parliament before its vote approving exploration in the Barents Sea. The Norwegian government has said that it fulfilled its constitutional duty by compensating for negative effects on the environment in other areas.
The action has failed in the lower courts, although both recognised the right of citizens to bring actions under the environmental provisions of the Constitution, with the higher court accepting that the right involved the impact from climate emissions — including those from oil and gas exported abroad, which is the case for most of Norway’s production.
The Norwegian government intends to continue requesting exploration licences and in June 2020 announced licensing awards in predefined areas (APA) 2020 which comprises blocks in the North Sea, the Norwegian Sea, and the Barents Sea.
A follow on from our blog of 15 April 2020 where we stated, as regards the climate change suits in Baltimore, and the Fourth Circuit Court of Appeal’s denial of the defendants’ application to remove it to the federal courts – where it would be dismissed due the decision of the Supreme Court in American Electric Power Co. v. Connecticut, 131 S. Ct. 2527 (2011) (AEP), and that of the Ninth Circuit in Native Village of Kivalina v. ExxonMobil Corp., 696 F.3d 849 (9th Cir. 2012), that such actions, at least when they relate to domestic GHG emissions caused by the defendant, are pre-empted by the Clean Air Act. .
On 31 March 2020 the Defendants submitted a petition for certiorari to the US Supreme Court. on the question whether 28 U.S.C. § 1447(d) permits a court of appeals to review any issue encompassed in a district court’s order remanding a removed case to state court where the removing defendant premised removal in part on the federal-officer removal statute, 28 U.S.C. § 1442, or the civil-rights removal statute, 28 U.S.C. § 1443.
Last month the United States Supreme Court stated that it will review the Fourth Circuit Court of Appeals’ ruling. This is on the procedural ground as to what can be reviewed by a federal appellate court. Three circuit courts – including the Fourth Circuit, which ruled on Baltimore’s case – have ruled that federal officer jurisdiction is the only issue that they can review when considering the companies’ appeal of a lower court’s remand order. The seventh circuit has taken the view that that the federal officer removal statute authorizes appellate review of the entire remand order.In the Baltimore case the District Court rejected a total of eight grounds for removal but the Fourth Circuit concluded its appellate jurisdiction was limited to determining whether the companies properly removed the case under the federal-officer removal statute.
The oil major defendants are banking on the hope that the more grounds for removal that the Court of Appeal must consider, the more chance of a successful remand to the federal courts where the pesky case will meet its demise.
Apache North Sea Ltd v Euroil Exploration Ltd  EWCA Civ 1397
In what circumstances will one contract be construed by reference to another? In the energy sector, the issue will often be an important one, given the prevalence of suites of contracts dealing with different aspects or layers of involvement or activity. The general rule is that “A document executed contemporaneously with, or shortly after, the primary document to be construed may be relied upon as an aid to construction, if it forms part of the same transaction as the primary document”: see Lewison, Interpretation of Contracts, 6th Edn, section 3.03. But this relates to different contracts which are “in truth one transaction” or “as it is called in modern jargon, a ‘composite transaction’” (Lewison). But what if the transactions are different ones, involving the same but also additional parties, but are related transactions?
Apache v Euroil: Summary and Take-Away Points
The Court of Appeal’s decision in Apache North Sea Ltd v Euroil Exploration Ltd  EWCA Civ 1397 addressed this question in the context of a Farm-Out Agreement (the FOA) between Apache as and Euroil for the sale and purchase of minority interests in respect of a UK Continental Shelf production licence relating to the Val d’Isere block and for Apache’s participation in the associated Val d’Isere Joint Operating Agreement (the JOA) as Operator.
The Court of Appeal (as the Commercial Court before it) held that, on the terms of the specific contracts in issue, it was wrong in principle to treat the FOA and the JOA “as entirely separate contracts with Apache wearing different hats in each” and that would “not reflect the true nature of the parties’ dealings at the time” . The contracts were to be construed together, and “in their proper context as a cohesive whole” .
While the Court stressed that it was dealing with the contracts before it and emphasised that it was not setting a “general precedent” for all FOAs and JOAs , the decision is significant in demonstrating a realistic approach to construing contracts which are meant to work together. As the Court stated, “Farm-out agreements do not typically exist in a vacuum. Where there is more than one owner, the parties will regulate their relationship in relation to that asset under a joint operating agreement. Farm-out agreements need to take account of and interact appropriately with those joint operating agreements to avoid inconsistencies and minimise the prospect of dispute.” 
The arguments in Apache v Euroil in the Court of Appeal
The issue arose out of the incurring of drilling costs by Apache in relation to an exploration “Earn-In Well”, using a drilling rig on a long-term lease. The rate for the drilling rig as incurred by Apache was one which was significantly above market rates at the time of drilling.
Apache sought payment of the drilling costs in full from Euroil in full under the FOA. In the very detailed terms of the FOA drafted, as was common ground and as the Court accepted, by “sophisticated parties represented by experienced lawyers” provision was made for the “Val d’Isere Earn-In Costs” which Euroil agreed to bear: “twenty six point twenty five percent (26.25%) of the total costs (other than the Back Costs) in relation to the Val D’Isere Earn-In Well, whensoever incurred, and in respect of all works undertaken pursuant to the Well Programme in connection with the Val D’Isere Earn-In Well”.
Euroil contended that the recovery was necessarily capped at market rates and relied upon the combination of the payment provisions in the FOA (requiring it to pay all Earn-In Costs “upon receipt of an invoice from [Apache] … in accordance with the relevant JOA within the applicable time periods as set out in the relevant JOA”), read together with provisions in the JOA to which both Apache (as Operator) and Euroil (and another) were parties. Euroil relied upon the usual ‘no gain no loss to the Operator” provision in the JOA and the detailed accounting procedure in the JOA which was used to be used for billing under the FOA, which had no billing procedure of its own. As part of that billing procedure, the cost of equipment leased by the Operator “not exceed rates currently prevailing for like…equipment”.
Apache responded that:
i. the FOA and the JOA were entirely different contracts with different mechanisms and purposes and separate parties; ii. The FOA was a bilateral sale contract with a price agreed which the purchaser is liable to pay. The JOA on the other hand was a multilateral joint venture contract with a joint venturers’ account. iii. Apache wore different hats at different times, depending on which contract is being considered. iv. To hold otherwise, would be impermissibly to incorporate a joint venture accounting convention in a multilateral joint operating agreement into a bilateral farm-out sale and purchase agreement so as to reduce the price there agreed; v. That would be “a significant development for the oil and gas industry, given that joint operating agreements are attached routinely to farm-out agreements by way of appendix”.
The decision in the Court of Appeal
The Court of Appeal rejected Apache’s arguments and held that the recovery of the drilling costs was capped at market rates given the provisions of the JOA. This was essentially for three reasons identified in the judgment of Carr L.J.
First, the artificiality of trying to construe the FOA as if it stood alone and without reference to the JOA. As the Court stated, this was “an ex post facto theoretical argument that does not reflect the true nature of the parties’ dealings at the time”  in circumstances where, by the time that the FOA was executed, the terms of the JOA, including the Accounting Procedure, had been negotiated and by the terms of the FOA they were to be deemed to be in full force and effect before and after completion of the FOA. The two contracts were “part of a package” and fell to be read together. As the Court said at the outset, FOAs do not exist in a vacuum and necessarily need to take account of and interact appropriately with those joint operating agreements to avoid inconsistencies and minimise the prospect of dispute.
Secondly, and building on that, not only was the JOA part of the “Agreement” which made up the FOA (because the JOA was attached by way of schedule to the FOA), but the FOA also contained what the Court described as a “plethora of references throughout the FOA to compliance with the provisions of the JOA” which showed that they were intended to interact with each other.
Thirdly, the argument that the FOA was an entirely separate and self-contained agreement could not sit with the parties’ express agreement for issuing AFEs, invoicing and payment under the FOA “in accordance with the relevant JOA”. The critical factor was that all billing under the FOA was to be done using the JOA accounting procedure and therefore invoicing Euroil for the Earn-In costs was subject, without qualification, to the JOA accounting procedure and the principles set out in it, in particular the ‘fair and equitable’ principle, reflected in market rates, and the ‘no gain no loss’ principle.
In one sense, it is difficult to see how the Court could have reached any other conclusion given the express inter-linking of the JOA into the FOA and the use of the JOA provisions for the accounting procedure. Looking at the language of the FOA in isolation, the Court found that Apache’s argument had at least an “initial attraction”. But the decisive factor was the fact that the proper construction of Euroil’s payment obligation fell to be determined on the basis of the text of both the FOA and the JOA, and sense made of each taken together.
The realistic approach of construing multiple contracts used in the energy sector is a continuing one. There are different routes by which the approach can be deployed, for example by treating the other contract or contracts as part of the factual matrix in which the subject contract was made and against which it must be construed, even if not part of the same transaction and even if not directly inter-related (as they were in Apache v Euroil).
The earlier decision of Teesside Gas Transportation v Cats North Sea Ltd  EWCA Civ 503 illustrates this in perhaps an extreme form. In that case, the terms of a cost sharing formula in a Capacity Reservation and Transportation Agreement dated 1990 and relevant to the usage of the pipeline were construed in the light of the concepts found in the later Transportation & Processing Agreements with third party shippers (“TPAs”) were concluded by the CATS Parties and with which it was to be assumed the CRTA was to work in the future. A “separate contracts” / “subsequent contract” argument was rejected by the Court on the basis that “the concepts used in those later contracts (such as “Daily Reserved Capacity Rate”) were within the contemplation of the parties in 1990 even if the names given to them and the detailed terms of the TPAs were not” (per Males LJ at ).
Coda: “Precedence Clauses”: any use?
As so often, reliance was placed on a conflicts or inconsistencies precedence clause in the FOA (“If there is any conflict between the provisions [the FOA and the JOA], the provisions of this Agreement shall prevail”). Apache argued that this established that the FOA ‘trumped’ the JOA. Again, as equally so often, the Court emphasised that such clause was only of any utility in the case of true conflict, which would usually not arise once the terms had been construed together and for which, in Carr L.J’s words, it had to be shown that “one clause in one document emasculates another clause in another document”. 
Earlier this year this blog reported on the implications for international shipping of the EU ‘Green Deal’, the topic of two papers at the IISTL’s recent Colloquium.
Things are now moving on apace. On 16 September the European Parliament voted in favour of a 40% reduction in CO2 by 2030 for all maritime transport and for the inclusion of ships of 5000 grt and over in the EU Emissions Trading System (ETS), with the establishment of an “Ocean Fund” to run from 2022 -2030 to contribute to protecting marine ecosystems. The Parliament is now ready to start negotiations with member states on the final shape of the legislation.
Where the EU goes, the IMO may follow – on which note in another interesting development, on 25 September, the major charterer, Trafigura, have submitted a proposal to the IMO for a partial “feebate” system to decarbonise global shipping. Trafigura’s press release states, “We propose a self-financing system where a levy is charged on the use of fuels with a CO2-equivalent intensity above an agreed benchmark level, and a subsidy is provided for fuels with a CO2-equivalent profile below that level. It is now time to put a price on carbon emissions in the shipping industry Our own in-depth analysis and commissioned independent research indicates that the levy should be between $250-$300 per tonne of CO2-equivalent. While primarily bridging the cost gap between carbon intensive and low or zero carbon fuels, this partial “feebate” would also raise billions of dollars for research into alternative fuels and could help assist small island developing states and other developing countries mitigate the impact of climate change.”
Rather more than the $2 per tonne bunker levy for financing R&D into alternative green fuels that various shipowner organisations proposed earlier this year.
On July 26 the “Wakashio” grounded off Mauritius, breaking up on 16 August. So far about 1200 tonnes of bunker fuel has been released into the sea. For Mauritius this is an environmental disaster.
Civil liability for bunker oil pollution falls under the Bunker Oil Pollution Convention 2001, to which Mauritius is a party. The good news is that under the Convention, the shipowner is strictly liable and there is mandatory insurance, with a direct right of action against the liability insurance, in this case the Japan P&I Club.
The bad news is that art. 6 provides that owners may limit their liability in accordance with the Convention for Limitation of Liability for Maritime Claims 1976 or as amended.
The 1996 Protocol, significantly increases the original limits in the 1976 Limitation Convention. However, it seems that Mauritius has not signed up to the 1996 Protocol.
Based on the gross tonnage of the vessel, apparently 101,932 tonnes, the limit for third-party claims including costs of prevention and clean up would be around $18m. Under the 1996 Protocol the limit would be $65m, based on the 2012 amendment to the LLMC 1996 limits, which entered into force in June 2015 and applied automatically unless objected to.
Had the oil spilled been from a laden oil tanker, the CLC and Fund regimes would have kicked in, with substantially higher limitation figures. Under the CLC the shipowner’s limitation figure would be around 65 million SDR, US $91.65 million, with the Fund’s limitation figure being 203 million SDR, US $ 324.3 million.
Most parties who lose English court cases or arbitrations give in (relatively) gracefully. In the long and ongoing Prestige saga, however (already well documented in this blog: see here, here, here, and here), the French and Spanish governments have chosen to fight tooth and nail, something that is always apt to give rise to interesting legal points. Last Friday’s episode before Butcher J (SS Mutual v Spain  EWHC 1920 (Comm)) was no exception, though in the event nothing particularly novel in the way of law emerged.
To recap, nearly twenty years ago the laden tanker Prestigesank off northern Spain, grievously polluting the French and Spanish coasts. Steamship Mutual, the vessel’s P&I Club, accepted that it might be potentially liable to direct suit up to the CLC limit, but pointed out that its cover was governed by English law, contained a “pay to be paid” clause and required arbitration in London. Nothing daunted, the French and Spanish governments came in as parties civiles when the owners and master were prosecuted in Spain, and claimed their full losses. The Club meanwhile protected its position by obtaining declaratory arbitration awards in England against both governments that all claims against it had to be arbitrated here; for good measure it then successfully transmuted these awards into High Court judgments under s.66 of the 1996 Arbitration Act (see The Prestige (No 2)  EWHC 3188 (Comm). These decisions the French and Spanish governments blithely ignored, however; instead they took proceedings in Spain to execute the judgments they had obtained there.
In the present litigation, the Club’s claim (slightly simplified) was against both governments for damages for continuing the Spanish proceedings, based either on breach of the arbitration agreement, or in the alternative on failure to act in accordance with the s.66 judgments. The object, unsurprisingly, was to establish an equal and opposite liability to meet any claim asserted by the governments under their judgments in the Spanish proceedings.
The Club sought service out on the French and Spanish governments: the latter resisted, arguing that they were entitled to state immunity, and that in any case the court had no jurisdiction.
On the state immunity point, the Club succeeded in defeating the governments’ arguments. The proceedings for breach of the arbitration agreement were covered by the exception in s.9 of the State Immunity Act 1978 as actions “related to” an arbitration agreement binding on the governments. Importantly, Butcher J regarded it as unimportant that the proceedings did not relate to the substantive matter agreed to be arbitrated, and that the governments might be bound not by direct agreement but only in equity on the basis that they were third parties asserting rights arising from a contract containing an arbitration clause.
The proceedings on the judgments, by contrast, were not “related to” the arbitration agreement under s.9: understandably so, since they were based on failure to give effect to a judgment, the connection to arbitration being merely a background issue. But no matter: they were covered by another exception, that in s.3(1)(a), on the basis that the breach alleged – suing in the teeth of an English judgment that they had no right to do so – was undoubtedly a “commercial transaction” as defined by that section.
The judge declined to decide on a further argument now moot: namely, whether suing abroad in breach of an English arbitration agreement was a breach of a contractual obligation to be performed in England within the exception contained in s.3(1)(b) of the 1978 Act. But the betting, in the view of this blog, must be that that exception would have been inapplicable: there is a big and entirely logical difference between a duty not to do something other than in England, and an obligation actually to do (or omit to do) something in England, which is what s.3(1)(b) requires.
State immunity disposed of, did the court have jurisdiction over these two governments? Here the holding was yes, but only partly. The claim based on the s.66 judgments was, it was held, subject not only to the Brussels I Recast Regulation but to its very restrictive insurance provisions dealing with claims against injured parties (even, note, where the claims were being brought, as some were in the case of Spain, under rights of subrogation). Since the governments of France and Spain were ex hypothesi not domiciled in England, but in their respective realms, there could be no jurisdiction against them.
On the other hand, the claims based on the obligations stemming from the arbitration award were, it was held, within the arbitration exception to Brussels I, and thus outside it and subject to the national rules in CPR, PD6B. The only serious question, given that the arbitration gateway under PD6B 3.1(10) or the “contract governed by English law” gateway under PD6B 3.1(6)(c) pretty clearly applied, was whether there was a serious issue to be tried as to liability in damages. Here Butcher J had no doubt that there was, even if the governments were not directly party to the agreements and the awards had been technically merely declaratory of the Club’s rights. It followed that service out should be allowed in respect of the award claims.
Further than this his Lordship did not go, for the very good reason that he had no need to. But in our view the better position is that indeed there would in principle be liability under the award claims. If, as is now clear, an injunction is available on equitable grounds to prevent suit in the teeth of an arbitration clause by a third party despite the lack of any direct agreement by the latter, there seems no reason why there should not also be an ability to an award of damages, if only under Lord Cairns’s Act (now the Senior Courts Act 1981, s.50). Further, there seems no reason why there should not be a an implied obligation not to ignore even a declaratory award by suing in circumstances where it has declared suit barred.
For final answers to these questions we shall have to await another decision. Such a decision might even indeed come in the present proceedings, if the intransigence of the French and Spanish governments continues.
One other point to note. The UK may be finally extricating itself from the toils of the EU at the end of this year. But that won’t mark the end of this saga. Nor indeed will it mark the end of the Brussels regime on jurisdiction, since the smart money is on Brussels I being replaced with the Lugano Convention, which is in fairly similar terms. You can’t throw away your EU law notes quite yet.
Anyone watching the news last week will have heard about ‘negative’ oil prices and producers paying people to take their oil off their hands – but what exactly does that mean, and how was it caused?
For starters, it’s a case of applying the basic economic principle of supply and demand. There is currently too much oil and nobody wants it. The reason for the latter is easy enough to identify: Covid-19 and the ensuing global shut down. People aren’t leaving their homes, no one’s flying anywhere, bulk products aren’t being shipped across the globe, factories aren’t running. Demand for oil has dried up. Oil production, however, has not, and aside from the obvious explanation of it being very difficult (sometimes impossible) and very expensive to turn off the tap, there are actually several other, complex reasons for why prices fell so drastically:
BAD TIMING AND GLOBAL POLITICS:
Covid-19 hit while global oil production was already high.
As a result of the development of fracking and shale oil in the USA over half a decade ago, there has been a glut of oil permeating the market. This (amongst other reasons) led to a significant downturn in prices in 2016 and as a result, the Saudi led OPEC – a legal cartel which aims to stabilise global oil prices – cooperated with several non-OPEC states (most notably Russia) to co-ordinate production cuts in order to counter the increase in American oil exports, and thus raise oil prices to a more stable, economic level. This alliance (known as OPEC+) fell apart once Covid-19 hit China. The Chinese shut down caused a major drop in the demand for oil and triggered a summit where OPEC agreed to further cut production, requiring also that OPEC+ members follow suit. Russia, however, refused. The official stance was that they wanted to wait for a better understanding of the pandemic before taking action. They also argued that there was already a shortfall as a result of political issues in Libya, which would help to offset the slump in prices.
Many analysts believe that Russia’s unwillingness to cut production was due in large part to it being disgruntled towards the United States, who have been one of the main beneficiaries of the OPEC+ cuts over the past four years (since they have had no such limitations on their production). Political relations between the two states have also not helped matters, especially with Trump’s general propensity for using oil as a political weapon against states and in particular the US sanctions targeting the building of the Russian Nord Stream 2 pipeline.
In retaliation for their lack of cooperation, Saudi Arabia then initiated an unexpected price war against Russia, turning on their taps and causing another massive drop in prices.
Meanwhile, producers in the United States kept producing in spite of the slump, even reaching record output highs in March 2020 – this seemed clearly counter-intuitive, not least because the US’ development of shale oil moved it from being one of the world’s largest importers to being the world’s biggest exporter (and thus it benefits far more from higher oil prices). The problem was twofold: one, production costs in the US are generally much higher than those found in rival, Middle Eastern states, and two, many American companies had secured billions of dollars’ worth of debt finance over the past few years to fund their increase in production. They simply could not afford to slow down.
The drop in market price quickly hit so low, however, that US ventures became commercially unfeasible and had no alternative but to begin shutting down operations. As a result, in early April the US, Russia and OPEC agreed to a deal to cut production. This agreement, while historic, seems to have done almost nothing to assure the market in the face of a global shut down – with many considering it too little, too late: there is a substantial amount of excess oil already on the market and even when quarantine restrictions are lifted, it will be some time before demand catches up to match/exceed supply.
A LACK OF STORAGE AND THE MAY FUTURES
At this point it is worth briefly explaining the two main oil grades which are used to set the majority of the world’s crude oil prices: Brent Crude and West Texas Intermediate (WTI). The former sets the prices of approximately two thirds of the global market, but WTI is produced in the US and is the US benchmark. It is this latter one which fell into negative prices – and while both Brent and WTI tend to move in lockstep (with Brent also having dropped to its lowest figures in over two decades) there were additional incidents relating to WTI which exacerbated the situation.
The other thing worth explaining quickly is the concept of a futures contract (often simply referred to as ‘futures’). Futures themselves are standardised, regulated, derivative financial contracts that oblige parties to transact a good at a specified price (‘strike price’) on a future date, with their specifications allowing market participants to trade them uniformly: each oil futures covers 1,000 barrels; dates for delivery are available up to nine years later; and title is officially transferred with the physical movement of the oil. They provide certainty for those who wish to sell or purchase crude physically (and who also need time to actually produce the oil/prepare to receive it), since the parties are able to contract with set, pre-determined prices that will not change based on the naturally volatile market price at the date of delivery (contracting for this purpose is known as ‘hedging’).
On the other hand, traders make a profit (and equally risk suffering losses) through market fluctuations. Some might retain the futures contract until it expires, requiring them to take delivery (these are usually traders who buy/sell for industry-related clients, such as producers or refineries), but most traders have no intention of doing so and instead sell the contracts forward to take advantage of the (hopefully) higher contango prices (i.e. when the price of futures is higher than the spot price) on later-dated contracts.
The vast majority of crude oil transactions take place via futures, but that does not mean the spot market (where trading for large, one-off transactions for near-term delivery takes place) should be underestimated, in fact it is vital: reported prices on the spot market are the basis of pricing for other forms of transaction, including futures. Additionally, as the expiration date of a futures contact approaches, it should become more liquid and the price should quickly begin to converge with the spot price.
The May futures for the WTI market were due to expire (and did) on Tuesday 21 April 2020 (the final day of trading for May). When that happened, whoever was still holding a futures contract would obliged to take physical delivery of the goods. Traders who had initially held off selling futures for this month began to panic as they realised their mistake: when futures are so close to expiration the only interested buyers tend to be companies who might otherwise have purchased oil on the spot market, i.e. they want to take possession of the oil and use it relatively quickly (like airlines, refineries etc.) Due to the glut, the earlier drop in prices and, most significantly, the sudden plunge of demand due to Covid-19 (which also severely weakened the spot market), these kinds of buyers for futures had dried up, and any who were willing to purchase the futures realised that they could take advantage of the situation by waiting until prices fell even further. Exacerbating the situation were the bulk exchange-traded fund rollovers (see below) and a lack of storage space (which continues to be a problem) – most notably in the terminals of Cushing, Oklahoma.
Cushing OK would be a tiny, inconsequential city (population: 8,000), were it not for the fact that it is the delivery point for WTI crude (and thus the pricing point for WTI futures). It is also where several main oil pipelines converge, essentially making it a key transhipment location between producers and refineries in the southern Gulf coast and buyers in the north. At full capacity (its current state) it stores up to 76 million barrels, which is over 10% of all US oil storage space.
Unlike Brent, which is a waterborne crude and does not suffer the same storage constraints (ships, after all, can come and go – to an extent), WTI is mostly onshore, with Cushing itself being a landlocked location in the centre of the United States. With the surplus in oil already filling the terminals there, not only were traders afraid of being forced to take delivery once the futures expired, they also realised they would have nowhere to store any of it (analysts have likened the situation in Cushing to a clogged bottle neck or traffic jam).
With the obligation to take delivery looming, those who still had May futures began paying companies to take the oil off their hands. The price they paid was calculated by what the buyers’ projected storage, insurance, transport costs etc. would be to do so. This was ultimately what dropped the WTI benchmark to below zero.
ETFS AND THE UNITED STATES OIL FUND
As the price war between the Saudis and the Russians drove prices low, it was natural to assume that this was an ideal time to invest in oil (based on the premise that crude prices will rise again since economies will, inevitably, have to reopen). This kind of thinking is not necessarily wrong but, as is the case with any financial investment, would-be venture capitalists should always undertake their due diligence before investing their money. Many didn’t and a great number of bullish speculators unfamiliar with the market (colloquially referred to by some as ‘oil tourists’) pumped over one and a half billion dollars into the United States Oil Fund (‘USO’), the largest exchange-traded fund (‘ETF’) in America (typically, ETFs are companies which use pooled investor money – similar in concept to a mutual fund – to invest in stocks, bonds and other assets). The USO ETF is designed specifically to follow price movements of WTI futures and, if futures are within two weeks of expiration, it will roll over the front month contracts to the second front contracts (this means that, when futures approach the expiration date they will be sold and the next month’s contracts purchased, usually simultaneously, thereby avoiding taking delivery). The USO is not a direct bet on oil prices and it incurs costs when it rolls its futures over. Not many of its new investors were aware of any of this.
The massive increase in investment quickly made the USO ETF one of the biggest players in the WTI market: according to a Bloomberg report, over the course of the last few months it held almost 30% of all WTI May and June futures. With such a huge and sudden injection of cash for the May futures (amidst the events which led to a crash in demand for oil) their prices swiftly rose; but then the USO sold all their May futures during their mandated rollover, buying June and July ones instead. When that happened, prices for the May futures dropped and, accordingly, they rose for June and July. Any traders left still holding May contracts suddenly found themselves in a state of trouble (see above) and when the market opened there was a huge differential in spread.
As an interesting side note, one of the companies that suffered unexpectedly from the USO’s actions was the Bank of China, which had pre-set the date to roll over its May WTI futures as the day before expiration (unlike other Chinese banks which had rolled over earlier in April). Specifically it was scheduled for Monday 20 April at 10:00 (ET), which was when the May futures were still trading at US$0, but the lack of demand (and thus liquidity) meant they took losses anyway. It’s unclear how many May futures they had to sell, but they suspended trading the next day and there was a flurry of angry investors on various Chinese social media platforms claiming that as a result of the day’s events they owed the Bank of China money (despite investors being forbidden from borrowing money to buy Bank of China funds). Additionally, Bloomberg reported the Bank’s oil related funds suffered losses of 600 million yuan.
WAS THIS A ONE OFF?
To a certain extent WTI prices dropping to such unprecedented levels was the consequence of a perfect storm of unfortunate events and it is therefore not unreasonable to conclude what happened was a localised, one-off incident. Having said that, until (at the very least) the storage issue is resolved we’re likely to continue seeing massive fluctuations in price. Case in point: S&P Global has instructed its clients to roll over all their WTI June futures into July to avoid What they believe will be a second plunge below zero for this front month. In doing so, June futures have now suffered a drop in prices and at the time of writing, almost 50% of them have been liquidated. July and September prices are more stable, but they will almost certainly fall into the same pattern until the global economy starts moving again, or at least until storage space is opened and the planned international production cuts reduce global supply to meet demand.
One recent idea to help alleviate the US glut – at least from a national perspective – was for the American government to purchase a large amount of the excess oil. It would do so via the Strategic Petroleum Reserve (‘SPR’), which is an inland oil reserve holding the largest supply of petroleum on the planet. The oil is owned and stored by the US government specifically with the aim of reducing disruptions in American supplies (the idea was conceived in the 1970s when the US suffered from an oil embargo resulting from difficulties in US-Arab relations). As and when it is deemed necessary by the President, the stored oil is released at a competitive rate, to avert potential short term crises (petroleum has been released, successfully fulfilling the SPR’s purpose, on a number of occasions in recent history).
The SPR currently has almost 80 million barrels’ worth of spare storage, and considering all oil purchased for the SPR is paid for with taxpayer’s money, it seemed economical to take advantage of the rock bottom prices. The Trump administration instructed the Department of Energy to start purchasing the oil in March 2020, but was made impossible when Congress refused to include funds for them to do so when they passed their most recent stimulus package (with federal funds channelled to areas of higher priority, given the effects of Covid-19 and the resulting shutdown).
Instead the government has decided to lease 23 million barrels’ worth of space to private oil companies. This will certainly help but the other factors remain unresolved, which means this might not be a one-off, but it’s hoped the lesson of what happened with the May futures will at least mitigate some of the volatility.
COULD THIS HAPPEN TO BRENT?
The fact that this happened to the American WTI benchmark, which has arguably less global influence than Brent does, and that the situation was exacerbated by the uniquely acute inland storage issue there, would suggest that this is a concern for the United States only. Brent, however, has already been affected by the fallout from the Covid-19 crisis, but the good news is it’s unlikely its prices will fall to the same extent that they have for WTI.
First, Brent futures are somewhat different in that they’re for cash, not physical, settlement, unlike WTI (physical settlement requires physical delivery of the crude once the futures expires, but with a financial settlement contract, the holder is merely either credited/debited the difference between their entry price and the final settlement). Secondly, Brent is a waterborne crude, with its pricing being based heavily on the Dated Brent index, which in turn bases its prices on crude from terminals at various coasts along the North Sea.
Using the North Sea as the primary location for which to base prices has its advantages: production there has been seeing a slow decline over the past few years (although prior to the pandemic, one of the terminals withdrew its plans to shut down for maintenance, which would actually push the North Sea supply to some of its highest levels in nearly a decade). Additionally, the North Sea terminals, unlike those in landlocked Cushing, have the infrastructure to allow for tankers to take delivery of the crude oil (i.e. they can take it away). As a result, Brent hasn’t suffered from the same storage-induced panic felt by the WTI traders in April – although that doesn’t mean it hasn’t suffered at all, there is still a global glut and therefore a storage issue that isn’t likely to improve until the agreed OPEC+ and US production cuts kick in, and even then the effects won’t be felt immediately. In the meantime, there are global reports that oil is being stored in every conceivable space, from the more conventional salt mines and caverns, to disused railcars.
Storage space if filling up fast, everywhere, the North Sea terminals have nowhere near the capacity that Cushing does and things might also be made worse if there were to be a sudden influx of cheaper, American produced ‘Brent-style’ crude flooding into Europe (indeed, why wouldn’t American companies sell to Europe and alleviate their storage issue? And being cheaper than Brent is currently, why would European companies choose not buy it?)
Of course, tanker owners have benefitted from this: the cost of chartering a VLCC has temporarily skyrocketed, with some shipowners set to experience their best quarters in history (which is welcome relief after the Covid-19 shutdown and the ensuing drop in demand for oil caused a massive slump in freight rates). How have companies been able to pay for this? The contango market has meant that oil prices have fallen faster on the spot market than the futures market (see above), in doing so, traders have been able to finance their storage by buying oil on the cheap in the spot market and then make a profit on it by selling it for higher prices in the futures market (for back month contracts).
Things, however, do seem to be looking up for Brent: in the past week prices have begun to rebound as global lockdowns and travel bans have slowly eased, but the fact still remains that the OPEC+ and US deal came too late. The lowest dip in the world’s demand for oil might be behind us, but the surplus is already out there and until the market can find that balance between supply and demand again, prices aren’t going to get much higher. The only saving grace producers – especially smaller and medium sized companies – will have, is to obtain a solid offtake agreement with a company that has a strong credit rating and a sure way of disposing of the oil… good luck getting one of those in this economy.
This blog recently featured a New Zealand decision in a strike out application in a climate change tort suit. Similar claims have also been a feature of litigation in the State courts in the US in the last few years. Why not in the federal courts? The reason goes back to two previous decisions: the decision of the Supreme Court in American Electric Power Co. v. Connecticut, 131 S. Ct. 2527 (2011) (AEP), and that of the Ninth Circuit in Native Village of Kivalina v. ExxonMobil Corp., 696 F.3d 849 (9th Cir. 2012), that such actions, at least when they relate to domestic GHG emissions caused by the defendant, are pre-empted by the Clean Air Act.
So, various municipalities have decided to sue in the State courts, claiming damages for what they estimate they will have to spend to mitigate the effects of climate change in future years. The oil majors who have been on the receiving end of these suits have sought removal of the cases to the Federal courts, where they will be dismissed. So far, the position on this is mixed.
The claims by the Cities of New York and Oakland saw their State law claims transferred to the Federal courts because of the interstate nature of the claims. Once there, Oakland sought, unsuccessfully, to distinguish Kivalina and AEP on the grounds that those decisions involved emissions directly from activities of the defendants, rather than by virtue of their sales of fossil fuels to third parties who then burn it and cause GHG emissions. This was not enough to distinguish the cases, and a further attempt, based on the effect of worldwide sales outside the reach of the Environmental Protection Agency and the Clean Air Act, also failed, running into the presumption against extraterritoriality. A further reason for dismissing the claims was that they implicated the interests of foreign and domestic governments and that the balancing of interests involved in the analysis of unreasonable interference in a public nuisance suit was best left to governments. New York has appealed the decision, as has Oakland.
By contrast, Baltimore’s tort claims in the State Court of Maryland have managed to stay there. The claims were not based on federal common law and the Clean Air Act did not show congressional intent for it to provide the exclusive cause of action, and indeed the Act contains a savings clause specifically preserving other causes of action. The Defendants then unsuccessfully applied to the Supreme Court for a stay, pending the hearing of their appeal.
On 6 March 2020 the Fourth Circuit declined to transfer the claims to the Federal Courts. They decided that the appeal was limited under 28 U.S.C. § 1447(d) to an appeal based on the Federal Officer Removal statute, one of the eight grounds for transfer argued by the Defendants in the District Court. The Statute, U.S.C. § 1442, authorizes the removal of cases commenced in state court against “any officer (or any person acting under that officer) of the United States or of any agency thereof, in an official or individual capacity, for or relating to any act under color of such office…” The Defendants argued that the statute applied because the City “bases liability on activities undertaken at the direction of the federal government”, pointing to three contractual relationships between certain Defendants and the federal government: (1) fuel supply agreements between one Defendant (Citgo) and the Navy Exchange Service Command (“NEXCOM”) from 1988 to 2012; (2) oil and gas leases administered by the Secretary of the Interior under the OCSLA; and (3) a 1944 unit agreement between the predecessor of another Defendant (Chevron) and the U.S. Navy for the joint operation of a strategic petroleum reserve in California known as the Elk Hills Reserve.
The Fourth Circuit held that none of these relationships could justify removal, either because they failed to satisfy the acting-under prong or because they were insufficiently related to Baltimore’s claims for purposes of the nexus prong.
On 31 March 2020 the Defendants submitted a petition for certiorari to the US Supreme Court. on the question whether 28 U.S.C. § 1447(d) permits a court of appeals to review any issue encompassed in a district court’s order remanding a removed case to state court where the removing defendant premised removal in part on the federal-officer removal statute, 28 U.S.C. § 1442, or the civil-rights removal statute, 28 U.S.C. § 1443.
In another suit, by San Mateo, the Defendants have appealed against the District Court’s decision not to transfer the suit from the California State Court. The appeal was consolidated with Oakland’s appeal. On 5 February 2020 the Ninth Circuit heard oral argument. They were later informed of subsequent developments in the Baltimore case.
A further success for the municipalities was in the Rhode Island suit, now subject to an appeal to the First Circuit.
It is, therefore, possible that at least one of these tort suits will see the light of trial in the next year or so. When that happens, expect some interesting arguments on causation and damages.