UK: Property in Commingled Gas - the Legal Structures Compared

This article was first published in March 2006 on TDM www.transnational-dispute-management.com and OGEL www.gasandoil.com/ogel

The last ten years have seen some very high profile insolvencies in the international energy business.

More recently a fear of shortages has stimulated renewed interest in large scale gas storage.

Who owns the gas in store? Who stands to lose if the system operator becomes insolvent? These are not idle questions of purely intellectual interest to energy lawyers. A combination of increasing volumes held in store coupled with rising real prices means that the financial exposures arising out of a big insolvency could be significant.

The inherent flexibility of natural gas production operations coupled with the limited flexibility provided by "line-pack" can only go so far in helping physically to match supply and demand. Such methods can cater well enough for routine demand variations caused for example, by the normal exigencies of the weather. They are less adequate when supply is tight and there are commercial or political pressures to limit or divert supplies. It is in such an environment that large scale gas storage has an increasing role to play in ensuring stability and security of supply.

Large scale underground gas storage in North America dates back to the beginning of the last century. In some cases the chosen reservoir was a depleted natural gas field into which newly produced gas could be pumped under pressure for storage purposes. In other cases gas has been stored in underground salt domes. Similar developments have occurred in the UK as the North Sea gas industry has matured. In the UK an exhausted former production field at Rough was already being used for storage back in the early 1990s. I understand that other former fields in the south of England have been adapted for use as storage in recent years. This process is bound to grow in response to the commercial and political pressures for security of supply throughout Europe.

North American jurists have had occasion to consider this kind of infrastructure over a long period including developing a specific jurisprudence on the expropriation issues which arise where it is sought to create a new gas storage facility underneath privately owned land but for the general public benefit - what in the United States is called "eminent domain" and in England we call "compulsory purchase"¹. As already indicated my purpose in this paper is to look at a rather different topic, namely the ownership and security interests which can be created over the gas in store, when applying conventional English common law principles.

My limited exposure to the North American experience suggests that the terminology used to describe the physical processes involved is not always uniform and can cause confusion. For present purposes I shall refer to the previously uncaptured natural gas which lingers in the strata of a depleted field as "native" gas and to the gas which is injected in order to pressurise the storage system as "cushion gas". Such gas is to be contrasted with gas injected into the system by or for the account of a customer who wishes to store it and which is sometimes referred to as "working" gas.

When multiple owners of parcels of gas introduce them into a pipeline or storage system the parcels become commingled and lose their identity. Thereafter it is impossible for any one user to identify any particular part of the homogenous whole and say: "This is mine". If gas is lost from the system through leakage², abstraction or accident³, again it is impossible to say whose gas has been lost. Where depleted gas fields are used for storage the reverse can also happen. Native gas lingering in the rock strata can commingle with the stored gas. If there is a net gain in volume then who owns the surplus? What of the state which may have exclusive rights of capture or rights to charge royalties or levy tax on the surplus? Who should end up paying for this?

Over the years various legal solutions have been applied to some of these problems particularly in a transportation/pipeline context.

When the former British Gas Plc began to separate its supply and transportation functions in the early 1990s it devised a form of gas transportation agreement under which title in a volume of gas passed to the transporter at the system entry point and title in an equivalent quantity passed to the user at a system exit point. The result was that all the gas in the system belonged to and was at the risk of British Gas Plc. This structure did not involve any form of bailment. As a matter of juridical analysis it was probably best characterised as a form of barter in which two different parcels of gas existing at different places and at different times were exchanged.

Key elements of this structure have survived into the UK’s Network Code. In particular the principle that title passes to/from the transporter at system entry/exit points is preserved⁴.

The above is far from being a universal practice. My understanding is that the transportation agreements regulating the use of the Bacton-Zeebrugge interconnector between the UK and continental Europe provide for all system users to own a share of the pipeline stock as tenants in common. The result is much more akin to a conventional transportation agreement which involves commingling as, for example, where multiple consignors allow parcels of crude to be mixed within particular tanks on board a VLCC.

These two structures can have radically different consequences in an insolvency. Under the first model if the transporter or storage provider becomes insolvent the user has a simple contractual claim which has no greater priority than any other unsecured creditor. Under the second model the users can in principle claim that all the gas in the system is theirs and their only concern will be with the immediate practical and operational difficulties arising out of the insolvency of the system provider. I will come back to these later. They should not be overlooked.

In the case of domestic distribution systems the infrastructure tends to be owned and run by a monopoly service provider in a regulated environment. The operator of such a distribution system may not be a particularly serious credit risk. The operator of storage facilities is in a different position. It tends to deal only with business customers and for that reason may be less closely scrutinised by the regulatory authorities. It will not be surprising if the value of the gas in store exceeds the market capitalisation of the system provider possibly several times over. The ability of the system operator to withstand the financial fallout from a serious accident may not always be assured⁵.

Related to the above are the insurance consequences of the two different models. Property insurance including insurance on bulk commingled commodities in transit or in store is a standard insurance product. Under such an insurance the assured needs to have some kind of proprietary interest in the subject matter. The user of a system in which all the gas belonged to the system provider would not have an insurable interest in the gas within the system. Such a user would have an insurable interest in the solvency of the system provider but to insure this kind of risk would require a credit insurance rather than a property insurance.

I suspect that one of the reasons why British Gas adopted what I have called the "barter" model in its early transportation contracts is that the creation and operation of a workable version of the alternative gives rise, at any rate under English common law, to difficulties of analysis which need to be both watched and understood.

In what follows I propose to summarise how the common law analyses the proprietary consequences of the commingling of bulk commodities belonging to two or more people. In so doing I shall identify some of the specific difficulties which can arise with the commingling of parcels of gas in store or in transit.

Before doing this I should mention one point which is specific to the UK context. In the UK the right to search, bore for and get petroleum (including gas) existing "in its natural condition" in strata in Great Britain and beneath UK territorial sea is vested in the state: s 2 Petroleum Act 1998. The state grants licenses to perform these operations. The practical effect of these provisions is that the owner of land beneath which gas exists in its natural state has none of the usual rights of an owner in relation to that native gas. The landowner’s consent may be required in order to perform drilling and other works but as a matter of property law he has no interest with which we need be concerned in the native gas.

The same applies to cushion or working gas which is injected into a depleted field for storage purposes. In English law the question whether personal property should become part of the real property on which it is situate is determined not just by ease of severance but fundamentally by the intention of the person who puts it there. Where previously captured gas is injected into a depleted field specifically for the purposes of storage there is no intention to dedicate it permanently to the land below which it is situate. It does not in my view become part of the land and the property of the owner of the land. Nor does it become state property. It is not gas existing "in its natural condition" in the strata where it is stored. It is gas which has generally been processed (into an unnatural condition) prior to storage and ends up in its storage location as a result of human endeavour rather than natural forces.

In short stored gas does not in my view become the property of either the local land owners or the state. Subject to the consequences of commingling (see below) and any subsequent disposal of title it remains the property of the person who owned it when it was injected into the storage space.

Where two or more parcels of a homogenous commodity (including gas) belonging to different persons become inextricably commingled so as to lose their individual identity the resulting mix is treated in law as belonging to all the contributing owners as tenants in common in proportion to their contributions to the mix. The early authorities for this proposition involved the accidental or tortious mixing of goods: see for example Spence v Union Marine⁶ (loss of marks and accidental mixing of bales of cotton during storm at sea); Sandeman v Tyzack⁷ (loss of marks on bales of jute). In the latter case Lord Moulton considered that an admixture arising by "accident or other cause" and without any wrongful act, gave rise to a tenancy in common⁸. More recently the same principle has been applied to the consensual mixing of cases of wine within a merchant’s warehouse: Ellis Son & Vidler Ltd⁹. This was effective to create tenancy in common in favour of the customers which was binding on the liquidator of the wine merchant provided (a) that each customer’s stock had been physically separated before being placed in store and (b) that proper records were kept of the quantities belonging to each individual customer.

A further proposition which has recently been indorsed at any rate in England is that in the case of a tortious mixing any doubts as to the size of the innocent party’s contribution to the mix will be resolved in its favour: Indian Oil Corporation Ltd. v. Greenstone Shipping S.A. (Panama) The Ypatianna¹⁰. See also Glencore v Metro¹¹.

A tenancy in common in goods can also be created by the consent of the two or more acquiring parties. The most oft cited examples are indivisible items such as ships and racehorses¹². Each tenant in common will acquire a notional "share" in the whole as agreed between all of them.

There is no reason in principle why a tenancy in common should not be created over goods which are physically divisible albeit presently undivided, such as a bulk quantity of a homogenous product like gas¹³. The only qualification (which applies whether the subject matter is physically divisible or not) is that the shares of each tenant in common should be ascertainable at any point in time. The reason for this is that the proprietary relationships arising out of a tenancy in common bind not only as between the tenants in common but also as between the tenants in common and third parties. As between the tenants in common there could be no policy objection to an agreement under which the share of a particular tenant was to be determined retroactively by reference to events which had not occurred at the time to which the allocation related back. This would be a simple matter of freedom of contract. As between the tenants in common and third parties however a freedom to re-allocate shares retroactively could be prejudicial particularly in an insolvency context.

For this reason I would suggest that there would be real difficulty with the concept that the share of one tenant in common at time T1 should be ascertained by reference to events which could not occur until time T2 particularly if the tenant whose share was thereby reduced (or eliminated) had become insolvent between T1 and T2. I am not aware of any case in which an attempt has been made to do this with a homogenous bulk commodity. Analogies can be drawn with other situations in which insolvency lawyers apply what they call the "deprivation principle". That is the proposition that parties cannot by simple contract and without creating a valid security interest, alter the proprietary interests of the insolvent estate after the event of insolvency. Examples include British Eagle v Air France¹⁴ and more recently North Atlantic Insurance Company others v Nationwide General Insurance Company and others¹⁵.

Once a tenancy in common has been created by a process of admixture, there is no reason why a tenant in common should not transfer its share in the tenancy to somebody else separately from the shares owned by the other tenants provided that nothing agreed between the tenants in common prohibits this. There was formerly some doubt as to whether the transfer of such an undivided share by way of sale constituted a "sale of goods" governed by the UK Sale of Goods Act 1979. The definition of "goods" within the Act was amended by the Sale of Goods (Amendment) Act 1995. The definition now expressly includes an undivided share in goods.

By contrast it was not possible at common law to create a tenancy in common by purporting to sell an undivided share in a bulk cargo. This is because of the logical impossibility of passing title in an unknown: see for example Re Goldcorp Exchange Ltd¹⁶ quoting from Lord Blackburn’s Treatise on Sale of Goods. It is "the very nature of things" that you cannot pass title in a thing until you have determined what that thing is. The common law principle was codified in s 16, UK Sale of Goods Act 1893 (now s. 16 of the 1979 Act). The practical effect of s. 16 was that where an attempt was made to sell an undivided share in a bulk cargo on the high seas, property would not pass until the share was physically separated from the bulk during discharge. Physical separation could include the discharge of all the other cargo, a process known as "ascertainment by exhaustion": The Elafi¹⁷ .

To the above propositions there is now in the UK a statutory exception in s.20A of the 1979 Act as introduced by the 1995 Act. This permits the creation of a tenancy in common where (a) the bulk is identified by agreement between buyer and seller (b) the contents of the bulk are homogenous and (c) the buyer has "paid the price for some or all of the goods which are the subject of the contract": s 20A (1). In such a case the buyer becomes an owner in common in the bulk either when the above conditions are satisfied or at such later time as the parties may mutually agree: s 20A (2).

Thus apart from the case covered by s 20A of the 1979 Act the critical point is that the participants in the tenancy in common must have owned their individual and ascertained parcels before mixing took place. The point is put very clearly in the following passage from the judgment of Lord Mustill in Re Goldcorp Exchange Ltd¹⁸:

Alternatively the tenancy in common can be created by selling the whole bulk to a group of purchasers who buy the whole as tenants in common in fixed shares.

Such is the general state of the English common law on commingling. How might it apply to commingling of gas? Much depends on the commercial and factual complexity of the particular situation. If a limited number of participants were to inject identifiable wholly owned parcels of gas into the same store it is not difficult to see how a tenancy in common of the kind found in the cases could be brought about and made to work although even in such a case there are a number of points which need to be watched. I will call this the "simple" case which is to be contrasted with the "complex" case where the storage facility is linked by two way pipeline straight into a national transmission system.

Starting with what I have called the "simple" case, there are I would suggest, a number of potential pitfalls if multiple owners are to have title as tenants in common to commingled gas in store:

• The tenants in common must agree between themselves terms which enable the quantity belonging to any one of them to be ascertainable at any point in time. Care should be taken to avoid an arrangement under which an allocation can only be ascertained by reference to future events.

• There is no reason in principle why ascertainment should not be achieved either by allocating absolute volumes or by allocating percentages of the total mix. If however the former approach is adopted then it is essential to cater for the eventuality that the sum of the individual allocations may (for whatever reason) differ from the volume of the commingled whole. It will be necessary to agree how any such shortfall (or surplus) is to be allocated²³. Another solution would be to allocate a fixed volume (or volumes) to one (or more) tenants in common and the remainder, if any, to another tenant in common but if this solution is adopted there will still be a need to allocate shortfalls in the event that there is insufficient physical gas to meet the entitlements of those to whom fixed volumes are allocated.

• Where contributors are adding gas of differing qualities to the mix further complications can arise but these can be addressed by using an allocation algorithm in a similar way to a conventional industry allocation agreement of the kind used where gas from two different fields is delivered into the same pipeline network.

• Where a large subterranean storage facility benefits from commingling with native gas the mix may not be truly homogenous. If there are likely to be material differences in quality between one part of the mix and another then it may be necessary to provide for volumetric or financial adjustments to be made depending on the actual quality delivered to any one tenant in common in the exercise of its drawing rights. It is generally simpler if this kind of adjustment rests on a purely contractual rather than proprietary framework. If it is thought necessary to make changes to the proprietary shares so as to reflect the quality of particular volumes drawn off, then it is preferable to re-allocate for the future rather than attempting (possibly ineffectually) to make retroactive re-allocations.

• It will obviously be essential that the system provider keeps adequate and accurate records of the volumes belonging to the various tenants in common at any point in time. Where the intention is that the cushion gas should belong to the system provider there is the potential for conflicts of interest between the system provider in its capacity as one of the tenants in common (in respect of the cushion gas) and the system provider as the keeper of the books and records. It may be thought appropriate to engage an independent third party to maintain the records or at the very least to provide all tenants in common with rights of audit and inspection.

• Where the tenants in common intend to trade shares of the gas in store the position becomes quite complicated. If a tenant in common wishes to dispose of the entirety of a volume or volumes which it, or its predecessor in title, has introduced into the mix, then at common law there is no problem. If however, that tenant in common wishes to split its share and sell part to a third party and retain the other part, the common law sees a difficulty because the subject matter of the transaction is never physically ascertained at any rate unless and until either the part sold or the part retained is withdrawn from store.

• In common law jurisdictions which have enacted legislation the same as or equivalent to s 20A of the UK Sale of Goods Act, the above problem is mitigated but (in the UK) only as and when the buyer of the unascertained part has paid the price of that part. Remember that where the UK statute law applies, s 20A Sale of Goods Act will only provide a solution if the gas in store is truly homogenous, by which is meant that any one cubic foot is interchangeable with any other cubic foot contained within the same store: see the definition of "bulk" in s 61 (1) of the UK Sale of Goods Act 1979 (as amended). There may be problems in satisfying this condition if commingling with native gas leads to quality differences across the commingled whole.

• Finally and obviously because the tenancy in common structure involves a bailment by all the tenants in common to the system user, some attention needs to be given to the terms of the bailment and in particular the bailee’s obligations to look after the gas in store and to answer for any loss.

Even in what I have described as the "simple case" it takes considerable care in the legal drafting to make a tenancy in common work properly. In the more complex case where the storage space is linked directly into a national transmission system in which all the gas is owned by the system operator, the practical difficulties multiply.

Within the UK transmission system all the gas is owned by the transporter and drawing (and entry) rights are exercised by making nominations for draw down (and entry) at particular exit (and entry) points during particular periods. The link into storage is in effect both a system entry point and a system exit point at which nominations will be netted off by the system operator so as to produce either a physical net inflow or physical net outflow to/from the storage.

Under such an arrangement any tenancy in common of the gas within the store cannot sensibly arise out of the mere act of commingling at the point of entry into the store. The point can be illustrated by taking the case where user A nominates 100 mmcf for injection into the store during the same period as user B nominates 120 mmcf for withdrawal from store. Physically the result of the two nominations will be a net outflow from the store. User A cannot possibly acquire rights as tenant in common of gas within the store on the basis that 100 mmcf of gas belonging to A has been injected into the store because this never happens. If A is to acquire rights as tenant in common of gas within the store the only way that this can happen is by way of transfer from persons already owning gas within the store. In theory the desired result could be achieved by having A acquire rights as tenant in common in 100 mmcf from B. In practice this is not a sensible way forward. There will be no direct contractual relationship between A and B. B will only have identified himself to the transmission system operator by making nominations. The system operator will have no interest in putting B into contractual relations with A and even if this was done it would be a case involving the creation of a tenancy in common pursuant to a sale of an undivided share of a bulk to which effect would only be given if the conditions in s 20A of the UK Sale of Goods Act were satisfied.

I have in the past encouraged clients to explore alternative solutions. Drawing rights on a homogenous stock of gas in store have similarities with and are capable of being regulated in a similar fashion to money in a bank account or financial securities held in a pool. The analogy has begun to feature in both the case law and in market practice.

In the case of National Power Plc v United Gas Company Limited²⁴ Hobhouse LJ described the transactions relating to gas held by the British system operator (Transco) with the National Transmission System in these terms:

At trial Colman J used similar language. He described the rights arising once gas had been introduced into the system as "thermal credits with Transco" or "Transco gas credits"²⁵.

A similar approach was adopted by Transco itself in the Summary of its Network Code which was published on its website until late 2005 when responsibility for the code shifted to the Joint Office of Gas Transporters. At page 8 Transco said this:

The same thinking could be applied to gas in storage. One possibility is to insist that legal title in all gas within the store is vested in a large and respected financial institution which acts as security trustee. Users of the system acquire drawing rights which are enforceable against the security trustee. The content of the drawing rights can still be regulated by the terms of the storage contract but the obligation is backed by the capitalisation and resources of the financial institution. Provided the scheme is structured so that the volumetric obligations of the institution cannot exceed the total volume of gas in store, the risk being run by the institution (and its fees) can be kept to a minimum. The parties still need to address questions such as allocation of risk of loss/gain but they do not need to be troubled by esoteric issues of property law and the Sale of Goods Act. Drawing rights can be split or consolidated at will in the same way as shares or bonds or money in a bank.

If this structure is employed and it is thought appropriate to insure the gas in store against risk of loss care needs to be taken to adapt any insurance arrangements to the proposed structure. The users of the store will have no insurable interest in the commingled gas existing within the store. They will have an insurable interest in any diminution in their drawing rights which results from the risks to which the gas in store is exposed. If a user wishes to insure this risk it needs to require that its insurable interest is correctly identified in the policy documentation. An alternative would be to have the security trustee insure the stored gas. It should be made clear in the trust documentation (a) that the security trustee holds the policy as trustee for the users of the store and (b) how the proceeds of any claim are to be allocated between the users.

A related problem where gas is stored in a depleted field is in defining exactly what constitutes an event of loss. The geology of a depleted gas field cannot be predicted with complete certainty. The theory will be that the storage space is bounded by impermeable strata. The practice can be different. US experience for the purposes of the law on eminent domain, has been that the precise boundaries of the storage "space" cannot be predicted with precision. There may be a "frontier" zone where the strata become progressively less permeable. Stored gas under pressure can slowly seep into such strata and having got there will be very difficult to extract again. Commercially (but not physically) such gas is for all practical purposes "lost" but I would be surprised if an underwriter would recognise this as an insured event.

There is one potential hiccough in my suggestion of vesting all gas in a security trustee. It is one which can arise whatever legal structure is chosen. It concerns the rights of the state where previously uncaptured native gas lingering in the depleted field becomes commingled with the rest of the gas in store.

As a matter of legal analysis a number of problems arise. Firstly the assumption behind the example is that the store consists of a former gas field for which somebody will at some stage have had a production license. If the license has expired then it may be necessary to consider whether one is needed. As previously noted in the UK the right to search for and "get" petroleum existing "in its natural condition" is the exclusive prerogative of the state. The practical consequence of commingling previously uncaptured native gas with gas which has been introduced into the store is that some at least of the uncaptured gas will end up being withdrawn along with the stored gas. This may sound pedantic but in any case where the volumes of native gas are likely to be sufficient to hit the statutory radar it would be unwise to overlook the legalities of what is going on.

Secondly, the commingling of the native gas with the stored gas disturbs the legal simplicity of an arrangement under which title in all gas in the store is supposed to be vested in the security trustee. Even if a production license is still extant, a financial institution which is otherwise willing to act as security trustee may not be enthusiastic about assuming the obligations of a licensee under a production license. Quite apart from this it must be a moot point precisely when the native gas can be regarded as "won and saved" - to use the language of the UK model license clauses.

I do not pretend to have a simple answer to these problems. Self evidently the authorities where the store is located must be prevailed upon to consent to the commingling of stored gas with native gas and preferably on terms which end up vesting title in the entirety of the mix in the security trustee.

If the system operator becomes insolvent the big money will no doubt turn on who gets the gas in store. But smaller sums will also turn on the extent to which the system continues to operate without interruption to its physical processes. In the case of monopolistic domestic transmission systems the answer is generally to be found in the powers of the regulator and in license conditions under which the system operates. In the case of competitive storage facilities whose customers are business users who are expected to take care of themselves, there is obvious scope for conflicts of interest between different user groups and conflicts between users as a class and the liquidator or administrator of the former system operator.

These are conflicts which are best addressed in advance by defining the circumstances in which the users as a class may substitute a new system operator and regulating the relations between users so that a "majority" view on the way forward can be obtained and given effect to.

Two radically different legal structures have traditionally been used to regulate the transportation and storage of commingled gas. One preserves the legal concept of a bailment and purports to have all users become tenants in common in the commingled bulk. Under the other legal title passes to (and from) the service provider at points of entry (and exit) from the system. There is no bailment but rather an exchange of contractual obligations to take and make delivery at different times and/or places.

For reasons which I have identified, there are real difficulties in attempting to "mix and match" these different proprietary structures where two systems are directly linked into each other.

My own preference is to avoid being drawn into esoteric legal arguments about quasi-Roman law concepts of confusio and to treat the user’s drawing rights on the system as contractual rights which need to be protected and secured in a similar way to financial obligations. The vesting of title in all the gas in store in a single sound financial institution may be the best way of achieving a solution which is both practical and acceptable to all concerned.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.