United Company Rusal Plc, R (on the application of) v The London Metal Exchange

The respondent to this appeal (“Rusal”) produces various metals which are traded through the appellant, The London Metal Exchange (“the LME”). The LME also approves the warehouses where the metals traded through it are held. The LME has proposed a new rule directed at solving a problem (“the queuing problem”) which started to occur at the start of the world recession in 2008, namely long delays in obtaining metals stored at warehouses. The LME decided that it would consult market participants on the one way which it supported for mitigating this problem, namely the adoption of a rule. The new rule would provide that the amount of metal which a warehouse, having a queue of more than 100 days, could accept (or “load in”) would be linked to the amount which it had delivered (or “loaded out”): for short, the “LILO” rule. To this end it issued a consultation notice (“the CN”) to all members, warehouse companies and London agents on 1 July 2013 inviting submissions within a three month period expiring on 30 September 2013.

The CN contained a detailed explanation of the LILO rule. It also explained that the LME had previously considered and rejected a number of options but it did not set out any of those options. These had included an option for rent caps (that is, reductions) or rent bans (that is, rent “holidays”) where delays exceeded a certain point (“the rent ban option”). The LME considered that this option carried unacceptably high legal risk on competition law grounds. On 7 November 2013, the LME announced that, following consideration of the responses to the consultation, the new LILO rule would take effect on 1 April 2014. In fact that did not happen due to the commencement of these proceedings.

On 23 December 2013 Rusal began proceedings for judicial review against the LME. Rusal’s case is that: (i) the CN should have set out an explanation for rejecting the rent ban option (“rent ban information”); (ii) the LME should have conducted a fresh competition law review before engaging in consultation and deciding to adopt the LILO option, investigations into this option with approved warehouse operators in September 2013, and (iii) the LME’s decision was vitiated by bias since it is financed by a stock levy made by it on warehouse charges and thus had an interest in maintaining the amount of the rent charged by warehouse operators (“warehouse rent”).

The case was heard by Phillips J who, in his judgment dated 27 March 2014, held that this consultation was unfair principally for two reasons: (1) it was procedurally unfair because it did not explain the principal option that had been rejected, namely the rent ban option, and (2) there had been inadequate investigation of the rent ban option prior to the consultation.

The judge’s judgment contains a careful and full explanation of the LME’s business and also trading in base metals. It also explains that the Financial Conduct Authority (“FCA”) has recognised the LME as an investment exchange for the purposes of the Financial Services and Markets Act 2000. The LME is therefore regulated by the FCA. I need not repeat most of what the judge set out by way of the background. The parties to this appeal did not claim that it was inaccurate in any material respect.

The judge explains why the queues mattered. There is a free market in base metals as well as a market in the LME. The LME price is used to help “discover” the free market price. The LME price is always at a discount to the free market price because of (a) the need to pay warehouse charges (so that the metal is, in the jargon of the market, “free on truck”) and (b) the inconvenience of having to obtain delivery from the warehouse. But, as queues have lengthened (thus creating an element of uncertainty in the warehouse rent to be paid by the holder of an LME warrant seeking to obtain physical delivery of the metal to which the warrant relates), it has become increasingly difficult to discover the free market price.    There is a divergence of view between, on the one hand, producers and warehouse operators with warehouses where there are queues, who benefit from the queues, and, on the other hand,  consumers who might have to pay higher prices and warehouse operators with warehouses without queues who face competition because producers prefer warehouses with long queues.

The background is also important in showing how the LME came to decide on the LILO option.

In 2010, the LME commissioned Europe Economics (“EE”) to advise whether the requirements in the LME warehouse contract for rates of physical delivery-out were satisfactory. In their report dated 12 May 2011 (‘the EE Report’), EE advised on five main policy options that the LME might consider to address the problem, the first four (or primary solution) seeking to impose controls on stocks or load-out rates for delivery out of physical stock. The fifth option was to ban or cap rent for metal in a long queue. The LME only published a summary of this report which did not reveal the nature of the fifth option.

EE had in fact done a report in 2007 which the LME had published and which stated that competition law prevented the LME from intervening in order to set the “free on truck” charge. This was a reference to matters analogous to banning or capping warehouse rents.

The LME adopted a modified version of the primary solution proposed in the (2010) EE Report by increasing minimum load-out rates with effect from 1 April 2012, but this was not effective to reduce queues.

The LME had considered the rent ban option but had been advised by internal and external counsel in 2000, 2005, 2007 and 2012 (and having sought the views of the European Commission in 2000 and 2007) that the rent ban option was vulnerable to challenge on competition grounds. There was a reference to the competition concerns in the context of the queuing problem in the May 2013 edition of Metal Bulletin, reporting a public statement by Mr Diarmuid O'Hegarty, the Deputy Chief Executive and former General Counsel of the LME, and in a Media Presentation accompanying but not referred to in the CN.

In late 2012, Hong Kong Exchanges and Clearing Limited (“HKEx”) agreed to acquire the LME, the acquisition being completed on 6 December 2012.

Mr Matthew Chamberlain, who had been advising HKEx on the acquisition, had already joined the LME as its Head of Strategy and Implementation in October 2012. Mr Chamberlain undertook a further review of the queuing problem in early 2013. He advised that there were 5 categories of options available but none of those categories related to the rent ban option. He presented his review to the board of the LME in April 2013. It contained a section entitled “Universe of Potential Solutions”, listing nine potential responses to queues, but not referring to the possibility of the rent ban option. The board decided in principle to consult on the LILO option. In June 2013 the board confirmed this decision but maintained the queue threshold at 100 days, rather than reducing it to 30 days suggested by Mr Chamberlain.

Thirty-three written responses were received of which 10 proposed a rent ban as well as or instead of the LILO option.

It is inevitable that Rusal should be concerned with the potential losses to producers that might result from implementation of the LILO option. It was possible that implementation would lead to an increase of metal on the free market and to a fall in the price which would be contrary to the interests of producers. In September 2013, Rusal filed a written response opposing the LILO option but not proposing the rent ban option. As the judge explains, Rusal opposed the proposed change stating “… the argument underlying the …. proposed rule change has not been clearly stated nor supported with Exchange data. Instead, the intent… to accelerate the transfer into the market of an additional two million tonnes of aluminium, accumulated and stored since the financial crisis, is an unprecedented intervention and one that Rusal strongly objects to”. Rusal suggested alternative measures designed to improve transparency in the price setting process rather than reducing queues: it did not mention the rent ban option.

There is no clarity over the issue whether Rusal were aware that the rent ban option was vulnerable to competition challenge. Mr Andryushin, Chief Operating Officer of Rusal Marketing GmbH, merely states in his evidence that Rusal did not know whether there were any viable alternatives to LILO or the reasons why all solutions, including LILO, had previously been rejected by the LME. He does not go so far as to say that he did not know that the options risked challenge by the competition authorities. Nor does he say that Rusal was unaware of the competition issues.

Nor does Rusal deny that the risk of challenge by the competition authorities to the rent ban option is a real one. On 9 October 2013, Mr Hodgson, Chief Executive Officer of Rusal Marketing GmbH, had a meeting with the LME at which the rent cap was discussed and one of his notes says that the rent cap could be considered anti-competitive, but there is no suggestion that Rusal was caught by surprise by this information.

The consultation period came to an end on 30 September 2013. Shortly before that date, the LME consulted its warehousing committee, on which a number of warehouse operators sat, for their views on the rent ban option.

On 25 October 2013, the LME decided to adopt the LILO option. The summary of responses made available to the market made it clear that, in the light of the 10 responses proposing a rent cap or ban, the LME consulted its warehousing committee about this proposal. The document issued by the LME with its response to consultation called the rent ban option “the only practical suggested alternative” to LILO (misquoted by the judge at para 74 of his judgment as “the most practical suggested alternative”). The LME had also decided to review its ability to regulate warehouse rents but that such review would be unlikely to be completed until 2014. For that reason the LME had rejected the possibility of deferring a decision on adopting the LILO option until completion of that review. Moreover, the LME had decided that any rent ban option would only be a reserve tool to use if the LILO option failed to achieve its desired effect.

On 7 November 2013, the FCA published a statement on the LME’s measures for reducing queues. It stated that it saw these measures as a first step towards strengthening the LME’s warehousing arrangements and increasing the transparency of its market. This demonstrates the FCA’s support for the adoption of the LILO option but that point is of limited significance since it is for the court to decide whether the obligation of fairness has been breached. In addition, Rusal makes the point that the evidence does not show whether the FCA also considered the rent cap option.

The course I propose to take in this judgment is to set out the judge’s reasons for holding that the consultation was unfair, explain the common law duty of fairness, and my analysis of the submissions on this appeal and reasons of the judge.

The primary holdings of the judge were as follows:

The judge further made the following material rulings:

I shall consider each of these findings when I give the reasons for my conclusions below but before that I shall consider the legal principles in more detail.

Paragraph 7 of the Financial Services and Markets Act 2000 (Recognition Requirements for Investment Exchanges and Clearing Houses) Regulations 2001 (SI 2001/995) imposes a duty on the LME to have proper procedures for amending its rules, including procedures for continuing use of the market. The scope of the LME’s duty to consult is also governed by a common law duty to act fairly. In R. v North & East Devon Health Authority Ex p. Coughlan [2001] QB 213, this court summarised the basic public law principles of lawful consultation (“the Coughlan principles”). Lord Woolf M.R., giving the judgment of this Court, held at para 108:

“It is common ground that, whether or not consultation of interested parties and the public is a legal requirement, if it is embarked upon it must be carried out properly. To be proper, consultation must be undertaken at a time when proposals are still at a formative stage; it must include sufficient reasons for particular proposals to allow those consulted to give intelligent consideration and an intelligent response; adequate time must be given for this purpose; and the product of consultation must be conscientiously taken into account when the ultimate decision is taken: R v Brent London Borough Council, Ex p Gunning (1985) 84 LGR 168.”

The word “fairness” is not used in this paragraph, but in Medway, Maurice Kay J rejected the submission that fairness was not an aspect of a lawful consultation process:

“It is an aspect of what is ‘proper’ — the word used in Coughlan (para 108). … it is axiomatic that consultation, whether it is a matter of obligation or undertaken voluntarily, requires fairness.” (Judgment, para 28)

The cases in this field demonstrate to my mind that the court should only intervene if there is a clear reason on the facts of the case for holding that the consultation is unfair. It is for the court to decide whether the obligation of fairness has been broken.

Moreover, the application of the duty of fairness is intensely case-sensitive. This is not an area of law where it is possible to provide statements of general principle. As Sullivan J held in R(Greenpeace Limited) v Secretary of State for Trade and Industry [2007] EWHC 311 (Admin):

“Judgments are not to be construed as though they were enactments of general application, and the extent to which judicial dicta are a response to the particular factual matrix of the case under consideration must always be borne in mind.”

It is also clear from the authorities that the courts have to allow the consultant body a wide degree of discretion as to the options on which to consult: as the Divisional Court held in TheVale of Glamorgan Council v Lord Chancellorand Secretary of State for Justice [2011] EWHC 1532 (Admin) at [24]:

“…there is no general principle that a Minister entering into consultation must consult on all the possible alternative ways in which a specific objective might arguably be capable of being achieved. It would make the process of consultation inordinately complex and time consuming if that were so. Maurice Kay J recognised this in [Medway], at para 26:

“Other things being equal, it was permissible for him (that is, the Secretary of State) to narrow the range of options within which he would consult and eventually decide.

Consultation is not negotiation. It is a process within which a decision maker at a formative stage in the decision making process invites representations on one or more possible courses of action. In the words of Lord Woolf MR in Ex parte Coughlan [2001] QB 23 at para 112, the decision maker's obligation “is to let those who have potential interest in the subject matter know in clear terms what the proposal is and why exactly it is under positive consideration, telling them enough (which may be a good deal) to enable them to make an intelligent response. The obligation, although it may be quite onerous, goes no further than this.”

This passage was approved by the Court of Appeal in R (Forest Heath DC) v Electoral Commission [2010] PTSR 1227 at para 54.”

Mr Michael Beloff QC, with Mr Simon Pritchard for the LME, cited a number of other authorities for this point, such as Nichol v Gateshead Metropolitan Borough Council (1989) 87 LGR 435, R (Kidderminster and District Community Health Council) v Worcestershire Heath Council [1999] EWCA Civ 1525 (refusal of permission to appeal), Tinn v Secretary of State for Justice [2006] EWHC 193 (Admin), and, at the request of McCombe LJ, R (Albert Beale and Lesley Carty) v London Borough of Camden [2004] EWHC 6 (Admin) (Munby J). I need not cite passages from these authorities save the following pertinent dictum of Auld LJ in Kidderminster:

“[Regulation 18(1) which required consultation on certain proposals] did not require it to give focus to proposals which it no longer had under consideration. In any event, the process of consultation did not, and designedly could not, preclude outright opposition to the one proposed, which opposition might prompt the authority to reconsider it and/or any of its discarded six options and/or to consider any new ones.”

In other words, there is in general no obligation on a public body to consult on options it has discarded. The statement in De Smith’s Judicial Review that there should be consultation on “every viable option”, taken on its own, is not supported by the authorities (7^th ed, paragraph 7-054).

I take each of the primary holdings of the judge in turn.

The judge’s first reason for granting judicial review was that consultees needed rent ban information. He held:

“(a)

The consultation arose because the implementation of the prime recommendation in the EE Report (contained in the published Recommendations section) had not resolved the problem of queues. Requiring rent rebates was the only other option identified in the Recommendations as capable, in principle, of addressing queues. Indeed, the EE Report had recommended that the option be “discussed”. Given its status as (apparently) the next obvious option referred to in the very report referred to in the second paragraph of the Consultation Notice, some explanation of the option, the result of the recommended discussions and the reason why the option had been discounted was necessary for a proper understanding of the LME's thinking in relation to the Proposal, just as reference to the alternative closure option was necessary in Madden.”

Mr Beloff QC submits that the fairness duty only requires the LME to consult on proposals which it proposed to implement unless there is an express promise to do more, or if there is a statutory duty to do more, or a legitimate expectation. That was not the case here. The judge went too far despite recognising the limitations of the duty of fairness by rejecting Rusal’s submission that it must consult on all viable options (judgment, para 66). Mr Beloff points out that the LME did not consider the rent ban option was viable because of the competition problems. By contrast they were advised that the LILO rule did not carry an unacceptably high level of legal risk and therefore was a viable option for implementation. He submits that, if the LME were entitled to select their preferred option, it is only necessary to include alternative options in order to make an intelligent response in exceptional circumstances, that is, where knowledge of the alternative is necessary in order to understand what is being put forward. Again he submits that that is not this case.

Mr Beloff submits that the judge held that the LME should have consulted on the rent ban option. This follows from his holding that the LME should have included that option in the consultation (see the judge’s fourth primary holding below).

Mr Beloff further submits that the CN provided a full and proper explanation of both the positive and negative impact of the LILO option. The factors relating to the LILO option were all free-standing of the rent capping option so that they were sufficient to allow intelligent response. In short, the consultees did not have to know about the rent ban option to respond on the LILO option on which the LME were consulting. Moreover, Rusal never suggested otherwise during the hearing of this appeal.

Miss Monica Carss-Frisk QC, with Mr James Segan and Ms Naina Patel, for Rusal, submits that fairness has to be judged in the round and that the judge was right to say that it was unfair to indicate that there were other options without explaining what they were. The LME should disclose the substance of its competition concerns. The LME could not logically say that the LILO option was the best course if it had not properly considered the rent ban option.

Miss Carss-Frisk submits that the information needed to enable the consultee to make an intelligent response can include information about a key alternative. That does not mean that the consultant body has to consult on every viable option but it must provide the relevant information. It would therefore have been sufficient for the LME to explain why the rent ban option was not an alternative and then Rusal could have replied that the problems were not as great as the LME thought that they were. The LME also had to set out precisely what the competition law concerns were.

I accept the submission that the judge held that the LME had to consult on the rent ban option, despite Miss Carss-Frisk’s submission that the judge merely required the CN to contain the rent ban information. That the judge required the rent ban option to be put follows from the judge’s third primary holding. Even if it were not so, Mr Beloff must surely be right that, if it was not unfair for the LME to discard options, it was likewise not unfair for it not to provide explanations of the options which it discarded.

As explained above, a public body is allowed a wide discretion in choosing the options on which to consult. However, as the judge noted, there have been exceptional cases where the courts have held that a consultation process was unfair for failure to set out alternative options. One such case was Madden, referred to at the end of (a) in paragraph 33 above.

In my judgment, the judge misinterpreted the decision in Madden. This case concerned the closure of two care homes. Richards J held that the consultation document provided to residents did not give an adequate summary of the true reasons for the proposed closures, and that “[t]he specific reasons expressed for the closure, or for the proposals to close the two individual homes, were incorrect or misleading.” In particular, he found that the survey report, on which the reasons for closure of one of the homes were based, was “presented misleadingly” in the consultation document (judgment, [58]). Residents were told that they would be assisted to find new homes, but they were not told where.

In relation to the other home, there was a misunderstanding as to whether it was required to meet certain statutory standards which led to a bald statement that it would not meet those standards. Richards J added, in relation to that closure, that “[a]ny proper understanding of the true reasons for the proposed closure would require at the least a comparison with the other home that the council thought it preferable to retain” instead one of those to be closed (judgment, [59]).

The judge relied on this passage. However, it is evident that Richards J was dealing with the hypothetical situation in which the true reasons for closure had been stated. The judge’s ruling has to be seen in its factual context. In order to explain the true reasons the authority would have to explain why some other home was preferable. In any event, there were residents for whom new homes had to be found and so one can naturally see that some alternative homes to which residents might move should in fairness be mentioned. Something had to be said about those reasons. It is easy to see that the local authority would have to explain the relative advantages of some other home which it preferred.

My reading of Madden is consistent with that given by the Divisional Court in Glamorgan at [25], where Elias LJ, giving the judgment of the court, held that Madden was exceptional and that a reference to alternatives was necessary on the facts of the case. That point is reinforced by the facts of the Glamorgan case. That case concerned proposals about court closures. The applicant contended that the consultation document issued by the Lord Chancellor should have identified possible alternative ways of achieving the savings he required and should either have consulted on those alternatives or given cogent reasons not to do so. The Divisional Court held that the Lord Chancellor was entitled to decide to consult only upon proposed closures when rationalising the court estate.

As Mr Beloff submits, this case is simply not on all fours with Madden. In Madden, Richards J found that the reasons given for the proposal were incorrect and misleading. The LME has correctly disclosed the true reasons (warehousing) for the LILO option, and it has no obligation to put forward a proposal which it is not willing to promote.

Miss Carss-Frisk cites other cases where the courts have required consultant bodies to set out alternative options, such as R (Montpeliers and Trevors Association) v Westminster City Council [2005] EWHC 16 (Admin) (where Munby J held on the facts that a general consultation of all the interested parties was intended) and R (Parents for Legal Action Ltd) v Northumberland County Council [2006] BLGR 646 (where Munby J held it was unfair for an education authority to consult on a two-tier system for schools without also consulting on the school closures involved). However, these cases clearly turn on their facts, and no useful purpose would be served by going into those facts. Inevitably there are cases which have been held to fall on either side of the line.

On this the judge held:

“(b)

Given previous references to the option, it was inevitable that responses to the consultation would propose or otherwise address banning/capping rents, but would be doing so in ignorance of (and possibly misled by the Consultation Notice as to) the very specific and technical reason why that option had been rejected. The fact that 10 responses proposed or supported the idea, but did not address possible contractual or competition law concerns, highlights the inherent unfairness of the process. It is certainly the case that the 10 responses were made without the information necessary to permit intelligent consideration and response. The procedure adopted by the LME also entailed that others who might have supported rent rebates, or contributed to the debate on legality, may have been unaware of the option and the issues which it engaged.”

Relevant to this holding is the judge’s further holding (referred to above) that the fairness of the consultation had to be assessed by reference to the CN. So it was no answer that the EE report or the Metal Bulletin or the Media Presentation had referred to the competition law issues since the EE report was not part of the CN. Mr Beloff criticises this holding as elevating form over substance.

Mr Beloff further submits that the competition difficulties were common knowledge in the market. The consultation was addressed to a limited number of highly specialist consultees. Mr Beloff submits that it was inevitable on the facts of this case in the light of the background which was well known in the market that consultees would be aware of the competition issues. In any event (he submits) it was self-evident that there were competition issues. There was no evidence that consultees were misled. There was no obligation to provide information about the options which consultees preferred, and there was no evidence to support the conclusion that the 10 responses were made without the necessary information: responses to consultation were treated as confidential. The consultees may have taken the view that it was for the LME to take the risk of challenge from the competition authorities.

Miss Carss-Frisk seeks to uphold the judge’s holding. The CN did not refer to the relevant parts of the EE Report, and the Metal Bulletin. She submits that in Madden there was a suggestion that in some circumstances a consultation document might be read against some separate document but that cannot extend to the media documents which the LME here relies on. However, Richards J held that the deficiencies in the consultation document were not salvaged by what was said at meetings with residents. It is up to the consultant body to disclose all that is material: as Collins J held in R v Secretary of State for Education ex parte Robyn Bandtock (a minor) by her father and litigation friend Tony Bandtock [2001] ELR 333 at [37], the consultant body must “not omit something which clearly is material and which cannot be expected to be identified by the reasonable reader". Some facts may be too obvious to need to be stated but that is not the case with the competition concerns which were not known to everyone. Everyone has an interest in others responding and therefore in everyone having access to sufficient information.

In my judgment, the submissions of Mr Beloff are to be preferred. The adequacy of consultation must depend on the sufficiency of information in the context in which the consultation took place. Therefore the court cannot ignore information which was well known to the consultees even if it was not set out or referred to in the consultation document. Any other conclusion would lead to cumbrous and potentially self-defeating consultation exercises where the real issue is obscured by common knowledge. There was evidence that the competition law difficulties were known in the market (see, for example, the 2007 EE Report, the Metal Bulletin, the Media Presentation and the evidence of Mr Chamberlain as to the consistent position which the LME took on rent bans from 1998 to 2014).

What was said at meetings in Madden was incomplete and accordingly was incapable of salvaging the deficiencies in the consultation document (see Madden at [62]).

I also agree with Mr Beloff’s submission that it should not be assumed from the fact that none of the ten consultees who proposed the rent ban referred to competition law concerns that they must have been ignorant of the competition law difficulties attending the rent ban option. It is striking that no consultee has filed even now evidence that it was misled into thinking that there were no competition law issues. It was open to consultees to take the view that any competition law issues were for the LME, and not them, to confront.

The judge held:

“(c)

The unfairness of the above is further illustrated and increased by the fact that the LME recognised during the consultation period that further discussion and legal review of the option of banning/capping rents was necessary (and, indeed, was commenced during the consultation period). It is unclear, and no explanation was provided by the LME, as to why such discussions and review had not commenced earlier or the consultation postponed pending their completion. In my view it was patently unfair to continue with the consultation without informing the market that it was simultaneously discussing and reviewing other options which had not even been referred to in the Consultation Notice.”

Mr Beloff criticises the inferences which the judge drew as to the reasons for this investigation. The purpose was to use the rent cap as a possible adjunct to the LILO option. That was something completely different from the proposal in the CN. The prevailing market conditions and regulatory and political pressure were constantly in flux. So there was every reason to check the position to see if an option should be reconsidered. Consultees’ responses prompted the reappraisal. The LME has to preserve and protect an orderly market and that involves reducing queues which may involve a reduction in the premium payable on the delivery of metals covered by an LME warrant over the price of free metal (that is, metal not held in an LME-approved warehouse).

Mr Beloff makes the point that legal advice as to competition law can never be definitive as to the future. The key issue on competition was justification in the public interest of the restriction (whether LILO or a rent ban). Justification changes from time to time so it cannot be assumed from the fact that there was no justification at any particular point in time that there is no justification later. The LME never arrived at the point at which it could be said they had committed themselves to further consultation because of an option raised by consultees. It did not at the material time consider it viable. The LME was right to engage with the warehousing committee as part of continuing developments. In point of fact, only one warehouse operator promoted the rent cap.

Mr Beloff submits that the judge should have considered why discussions did not start earlier and were not postponed. The reason was that the advice at the outset was clear. As the regulator, the LME was bound to keep the options under review. At the time, queues were not diminishing but increasing.

Miss Carss-Frisk submits that there was no good reason why the board of the LME could not have completed its review into the rent cap option before the consultation. It was unfair to investigate it at this stage without telling the consultees.

In addition, on Miss Carss-Frisk’s submission, the critical words in this paragraph are “without informing the market”. The problem might have been avoided if the LME had told the market about setting up this further review.

In my judgment, the judge was wrong to hold that it was unfair for LME to start these investigations during the consultation period. Consultees were not deprived of knowledge of another possible option since the option being considered by the warehousing committee was to use the rent ban option as an adjunct to, not a substitute for, LILO.

In Coughlan, the Health Authority obtained and considered a report by Dr Clark on the opinions of local clinicians which was received well after the consultation period had ended. Rejecting the claimant's complaint that the authority had acted unfairly in considering the report, Lord Woolf held at para 112:

“It has to be remembered that consultation is not litigation: the consulting authority is not required to publicise every submission it receives or (absent some statutory obligation) to disclose all its advice. Its obligation is to let those who have a potential interest in the subject matter know in clear terms what the proposal is and exactly why it is under positive consideration, telling them enough (which may be a good deal) to enable them to make an intelligent response. The obligation, although it may be quite onerous, goes no further than this.”

In the present case, the review which the LME instituted of the rent cap ban did not undermine anything that the LME had said in the CN. It was about a different use of the rent cap ban. If the LME had changed its mind and considered that the rent ban option was a viable alternative to the LILO option, it would have to have issued a fresh CN to consultees.   It follows that there was nothing unfair in the LME pursuing this review without making an announcement to consultees with respect to the CN, and that the judge was wrong to conclude that the institution of this review was “tantamount to an admission that it had failed to make sufficient inquiry and had failed to consider relevant matters prior to commencing the consultation” (judgment, paragraph 94).

The judge held:

“(d)

Given that the LME recognised that the Proposal would likely result in losses being suffered by metal producers, but decided to discount that as a factor, fairness demanded that they should have the opportunity to consider and comment on an alternative that might cause them less damage. Whilst the LME may regard banning/capping rent as a possible further step to reduce queues in the future, metal producers will by then have suffered the full losses caused by the introduction of the Rule: by failing to include banning/capping rents in the consultation, the LME has deprived the metal producers of any opportunity to reduce those losses. The situation is therefore analogous to that in Medway.”

Mr Beloff attacks this holding root and branch. He submits that it was inconsistent with para 96 of the judge’s judgment (below). Medway was not authority for the proposition that every consultee is entitled to have the option put forward which would cause them least damage. If the judge was right, the LME would have to put the entire “universe” of possible solutions into the CN.

Moreover, submits Mr Beloff, the decision in Medway does not provide any analogy. In that case, there was a risk that the objectors would be constrained by a government policy option which had not been the subject of consultation. This was contrary to the authority that in general it would be for the consultant body to select the options on which it would consult. In this case, the consultee would not be closed out because the LME had to continue to review the policy options and it was specifically reviewing the rent ban option at the date of these proceedings.

Miss Carss-Frisk submits that the LME had an obligation to take into account that the LILO option would almost certainly result in considerable damage to producers and might even force them to close smelters. The LME had to treat their interests as relevant and there was no evidence that they had done so.

In my judgment, the judge’s fourth primary holding amounts to saying that the LME had to set out an option that would be less damaging to producers even if the LME did not support it. This is plainly contrary to the consistent line of authority that the duty of fairness does not require a consultant body to put forward options which it had discarded.

I do not consider that, properly understood, Medway supports the judge’s conclusion. It was a rather unusual case of a consultation document preceding a white paper. A white paper sets out proposed government policy prior to producing a bill. So the consultation was at an early stage in the decision-making process, which may be of some relevance.

The subject of the consultation document was the future development of UK air transport. The consultation document set out a number of options, including expansion of several airports and the building of a new airport, but had specifically stated that the government would not consider any option that involved enlarging Gatwick airport. This option would, therefore, not feature in the resultant white paper.

Maurice Kay J held that it was not open for the consultant body to say that it was entitled to select the options if this was unfair, and unfairness was a question for the court. The reason for unfairness was case-specific and requires careful explanation. The judge held that while, in due course, objectors to planning applications related to expansion at other airport sites could argue before a planning inspector that expansion at Gatwick was an alternative, the exclusion of Gatwick from the white paper was likely to prove an "insurmountable hurdle" to that argument. The government was thus operating the consultation in a manner that deprived individuals of their only real opportunity to present their case for expansion at Gatwick. In my words, the government was making it impossible in practice to oppose their policy not to expand Gatwick in planning matters. In those circumstances, Gatwick could not be excluded from the consultation exercise.

That fact pattern is simply not the case here. Critics of the LILO option could put forward the rent ban option and ten of them did so. They caused the LME to review its approach and to institute a fresh review of that option as an adjunct to LILO. Consultees were not, therefore, deprived of the effective ability to promote the rent ban option.

It is not, however, right to conclude that Phillips J in this case took the view that the LME had to take into account the interests of the producers since, later in his judgment, the judge accepted that the LME’s decision could not be driven by the fact that some groups stood to gain and some to lose:

“96 In my judgment, however, the LME's approach to dealing with queues cannot be criticised in this respect. In dealing with a recognised problem in the market, it was reasonable and indeed essential to approach the issue uninfluenced by the competing interests of buyers and sellers in the effect on price, unless such effect would be such as to damage the market. Given Rusal's own evidence as to the massive fluctuations which have been seen in the price of aluminium in recent years, the relatively minor and short term effect the Proposal is likely to have on prices does not appear to be of great significance. It would be improper for the LME to take into account which category of its users would win and which would lose in any such price effect.”

Rusal attacks this holding in its respondent’s notice for the reasons already indicated.

Mr Beloff counters this by submitting that the statutory duty of the LME is to maintain an orderly market.

I agree with Mr Beloff’s submission. The LME cannot fetter the way it exercises its statutory duty by reference to what is in the commercial interests of some of the participants in the market. On the other hand, in agreement with Miss Carss-Frisk’s careful analysis, those commercial interests may be a factor relevant to the maintaining of an orderly market. This may be the case, for example, in relation to preventing warehouses with queues from withdrawing from approved status. But that is different from making decisions which are driven by a participant’s financial interest.

The judge went on to deal with a number of points, including the point made by the LME that it had rejected the rent ban option because of competition law concerns. He held that it was not open to the LME to take this position because it had not disclosed any advice it had received:

“…it is said that LME was justified in not referring to the option of banning/capping rents because, given the significant risk of regulatory action in respect of which no comfort could be obtained, it was simply not and never could be a viable option. However, given that the LME has not disclosed any advice it has received in the past and is now engaged on what must be (given the time taken) an extensive review of the competition law issues with a view to considering the option in future, it is in my judgment simply not open to the LME to take that stance.” (judgment, para 76)

On Mr Beloff’s submission, the judge failed to take into account that LME was not bound to disclose legally-privileged advice to consultees.

Miss Carss-Frisk submits that there need be no waiver of privilege. She submits that the judge is not saying that the privilege should be waived but that there should be disclosure of such rudimentary matters as the articles of the EU treaties on which the LME relies and broadly the issues that arise and some broad statement that certain options were not available “in the light of” competition issues.

In my judgment, the passage I have quoted is not open to the benign interpretation placed on it by Miss Carss-Frisk. The judge expressly referred to the disclosure of advice. To require LME to do this as the price of a lawful consultation would turn the duty of fairness on its head. The real question is whether consultees had sufficient information about potential competition difficulties. I have already held that, in the light of information in the market and other documents known to market participants, the consultees had sufficient information on what no doubt would have been a sensitive issue.

The judge held:

“85 Rusal acknowledged that the conflict of interest upon which it relies is inherent in the structure of the LME and its warehousing arrangements. Rusal further recognised the doctrine of “necessity”, as explained in section 6 of chapter 10 of De Smith (above), extends to situations where “the administrative structure makes it inevitable that there is an appearance of bias”, so that the decision-maker is not disqualified.

86 Rusal contends, however, that the LME could and should have taken steps to address the appearance of bias, for instance, by setting up an independent advisory committee or other independent structure.

87 However, the ultimate decision to accept or reject any independent recommendation would have to be taken by the LME itself (through its Board) and could not properly be delegated, entailing that any appearance of bias could not be displaced, precisely because it is inherent in the administrative structure of the LME. In my judgment the doctrine of necessity applied in this case and, given the necessity for the LME to act where it appeared to have a conflict, there was no obligation to adopt some half-way measure to mitigate the appearance of bias.

88 But in any event the LME's stock levy, although only referred to in a footnote in the consultation report, is an integral part of the LME's structure, well known throughout the marketplace and is certainly known to Rusal. In those circumstances Rusal cannot sit back and then raise the issue of bias after the event, but must be taken to have waived its objection: see De Smith (above) paragraph 10-061.”

Miss Carss-Frisk submits that this conclusion was wrong in law. She contends that it is important to maintain the high standards of public law. She submits that the doctrine of necessity should be used only sparingly. She also submits that there was no waiver here. There was no proper disclosure of the financial interest of the LME in the proposals. There was no explanation of the consequences of not objecting (cf Smith v Kvaerner Cementations Foundations Ltd [2007] 1 WLR 370 at [26] - [29].

Mr Beloff relies on the fact that the stock levy was well-known in the market. It represents 1.1% of the daily rent collectable on LME warrants. (There are also some warehouses listing fees.) The LME does not depend on higher stock levy as a part of its financial management (and a footnote in the response document issued by the LME after the consultation so states). Mr Beloff contends that Rusal waived any right to take a point on bias by not raising it until after the consultation process was completed. The LME could not abdicate its responsibility for dealing with the queuing problem. The problem would have existed even if the LME had taken independent advice. In addition, the stock levy would decrease if queuing was reduced. The judge’s conclusion was, therefore, correct.

I consider that the judge came to the right conclusion on bias. It is not necessary to rely on the controversial doctrine of necessity. The stock levy is widely-known in the market. It is referred to on the LME’s website. Rusal has clearly waived any right that it had to object to the LME making its decision by not taking its objection at the outset (see De Smith at para 10-061).

The common law duty of fairness is imposed on a public body to enable those who are affected by its decisions to respond to a particular proposal about a decision the consultant body proposes to make or may make depending on the outcome of the consultation. The duty to provide sufficient information does not in general extend to providing options or information about proposals which it is not making unless there are very specific reasons for doing so.

In my judgment, the case law shows that the explanation provided by a consultant body in its consultation document is not unfair unless something material has been omitted or something has been materially misstated.

For the detailed reasons given in this judgment, I conclude that the judge was wrong to say that the consultation in this case was unlawful. My reasons reduced to their essence are that, in my judgment, the judge was wrong because:

It is also clear that Rusal is not challenging the consultation process in order to make some response on the rent ban option which it has been unable to do for want of information but to pursue some other objection of its own. Even though Rusal has standing, its complaint is nonetheless unreal. It has no support from any consultee who feels that it was misled.

The common law duty is imposed to hold the constitutional balance between a public body and the general body of consultees. It is not, in my judgment, imposed to enable a consultee to challenge the absence of information on an option which it quite simply does not wish to have pursued.

Moreover, this is a comparatively unusual case of commercial judicial review, that is, where judicial review is sought in respect of the decision of a body with the public function of running a commercial market or exchange. Miss Carss-Frisk made a powerful submission that the standard of fairness should not be diluted simply because the issue arose in a commercial situation. But that is not the only relevant factor. Where judicial review is sought in a commercial setting of this kind, parties may make or lose money as a result of the delay caused by a legal challenge and the regulation of the market may be undermined. Accordingly, the court has to act with considerable care lest the integrity of the market is prejudiced by an unsustainable legal challenge.

Leaving aside commercial considerations, the judge’s conclusions represented a considerable extension in the duty of fairness in consultation. In particular, it would considerably increase the burden for consultant bodies if they had to consult on all the options which they were not advancing. Mr Beloff referred us to the sapient observation by Jackson LJ when granting permission for this appeal that “[t]he judge’s analysis, if correct, places onerous obligations on any public body conducting a consultation on complex issues in a politically sensitive area.”

In my judgment, the judge in his lucid and careful judgment extended the principles beyond the limits to which they can properly be stretched. On facts such as these, the Coughlan principles spelling out the duty of fairness must be kept within their proper limits.

For the reasons given in this judgment, I would allow this appeal.

I agree.

I also agree.