Rockware Glass Ltd., R (on the application of) v Quinn Glass Ltd & Anor

LORD JUSTICE BUXTON: This judgment sets out no more than the background facts that are necessary to understand the issues and the judgment’s conclusions upon them. Anyone who thinks they need to learn more about the matter can safely consult the full and careful judgment of HHJ Gilbart QC [2005] EWHC 2250 Admin.

The case concerns a substantial industrial installation for the manufacture of glassware, located at a former power station at Elton, which is on the borders of the district of the city of Chester and Ellesmere Port in Neston. This is an area, as the judge pointed out, long known for its substantial industries, including petrochemicals, and refineries.

The present works have been designed, as the judge found, to be the largest glass container work factory in Europe. It is a development by Quinn Glass Limited (“Quinn”), who are a major manufacturer of glass but, as the judge understood, a relative newcomer in the United Kingdom.

The construction and operation of this plant required various authorisations, two in particular. First, the grant of planning permission under the 1990 Act and second, and this is the matter which the court is principally concerned, a permit under the Integrated Pollution Prevention and Control Regime (“IPPC”) designed for pollution control. It will be necessary to describe that regime in some more detail later on.

The challenge in these proceedings relates to an IPPC permit issued by Chester City Council (“Chester”), the designated regulator, on 2 March 2005. The permit imposed a wide range of environmental restrictions on the operation of the plant, a matter to which I shall have to return in detail. The present case is however limited to the permit’s requirements in relation to the emission from the plant of oxides of nitrogen, NO₂. The legality of the permit in that respect was challenged before the judge by a competitor of Quinn, Rockware Plc (“Rockware”). The judge quashed the permit but suspended the operation of his order pending an appeal to this court.

Having had the benefit of the judge’s judgment, Chester no longer seeks in this court to maintain its certificate in its present form but Quinn, as an obviously interested party strenuously contend before us that the judge was wrong.

The present case relates to the European Union’s regime for pollution control. Starting with article 174, as it now is, (2) and (3) of the EC Treaty:

“(2)

Community policy on the environment shall aim at a high level of protection taking into account the diversity of situations in the various regions of the Community. It shall be based on the precautionary principle and on the principles that preventive action should be taken, that environmental damage should as a priority be rectified at source and that the polluter should pay.”

[Those are sentiments that appear later in this case]

“(3)

In preparing its policy on the environment, the Community shall take account of: available scientific and technical data, environmental conditions in the various regions of the Community, the potential benefits and costs of action or lack of action and the economic and social development of the Community as a whole and the balanced development of its regions.”

Those are sentiments which, in particular the reference to the potential benefits and costs of action or lack of action, again play a role later in the appeal. That regulation in the treaty is supported by three Directives with which we are concerned. The first is Directive 96/61, the Integrated Pollution Prevention and Control Directive, which I shall refer to hereafter as the IPPC Directive. Secondly, effectively as part of the same legislative exercise though passed some three days later than the IPPC Directive, is Directive 96/62 EC, the Ambient Air Quality Assessment and Management Directive. I shall refer to that as the Air Quality Directive. Third, and playing a lesser but not insignificant role in this case, is Directive 1999/30 EC, the Emissions Limit Values Directive, which lays down within the framework of the Air Quality Directive certain minimum standards for what are described as emissions limit values (ELVs). That is the unit or standard by which the quality of air is measured.

Finally, and in the domestic sphere, we are concerned with Regulations passed in the year 2000, the Pollution Prevention and Control (England) Regulations (“the Regulations”). Although, strictly speaking, it is the Regulations that affect domestic operation, and there is no suggestion that the present Regulations do not properly implement the Directives, nonetheless it has been regarded as convenient and not inappropriate for argument to focus principally on the terms of the Directives.

In addition to the Regulations the domestic regulator, such as Chester, has the benefit of guidance issued pursuant to Regulation 37(2) of the Regulations. The two documents in question are the Secretary of State’s general guidance manual on policy and procedures of March 2003 which the judge called, and I will call, “the Manual”, and a sector guidance note, IPPC SG2, which is the Secretary of State’s Guidance on Glass Manufacturing Activities with a Melting Capacity of More Than 20 Tonnes Per Day, which describes itself in its paragraph 1(2) as giving statutory guidance:

“On the integrated pollution control standards appropriate for the generality of new and existing A2 installations in the glass manufacturing sector.”

The plant with which we are concerned does indeed have a melting capacity of more than 20 tonnes per day and therefore this guidance note is relevant to it. It will be referred to hereafter as “SG2” and is of significant importance in this case.

The responsible person for pollution control in Chester was a Mr Hosker. He reported to the chief executive, a Mr Durham. For regulatory purposes, emissions are judged according to the ELV, already referred to, of the process in question, judged by units described in terms of grams per cubic metre: g/nm³. Mr Hosker concluded, and Mr Durham accepted on his advice, that the regulatory obligations under the Directives and Regulations would be achieved by requiring of the Quinn plant emission limits in a reducing scale, reducing to the end of 2009. That which was included in the permit, set out for convenience at paragraph 127 of the judge’s judgment and elsewhere, provided for 1000 ng by the end of year one, 900 by the end of year two, 750 by the end of year three and 500 by the end of year four, those years being broadly judged from the date of the permit, so that the standard or limit of 500 mg would not be reached until the year 2009.

How Mr Hosker reached that formulation is a matter of central importance in this case and I shall have to come back to it. The acceptance of that formulation by Mr Durham was important because it was he who took the decision to issue the permit. In doing so, he acted or conceived himself to be acted under delegated powers contained in rule 29 of the council’s procedure rules. That was a power:

“To deal with matters where a decision is nominally to be made by the Council [I interpose, but] is governed substantially by matters of fact or technical factors so there is no real discretion.”

That is the case that Mr Durham thought, on advice from Chester’s lawyers, he was dealing with here. I would only comment at this stage that it is, to put it at its lowest, a striking fact that a permit for what is, as the judge found, the biggest glassworks in Europe never went anywhere near the elected representatives of the citizens of Chester, and was determined by the chief executive in an informal meeting of which no record was kept.

The recitals and general provisions of the IPPC Directive to some extent pull in two different directions, as was stressed before us by Quinn. First, there is emphasis on the need for a high level of protection for the environment. I should read at least some examples of that. Recital 8 says:

“Whereas the objective of an integrated approach to pollution control is to prevent emissions into air, water or soil wherever this is practicable, taking into account waste management, and, where it is not, to minimise them in order to achieve a high level of protection for the environment as a whole.”

And Recital 14:

“Whereas full coordination of the authorisation of procedure and conditions between competent authorities will make it possible to achieve the highest practicable level of protection for the environment as a whole.”

That is repeated in the articles themselves. For instance, article 1 says:

“The purpose of this Directive is to achieve integrated prevention and control of pollution arising from the activities listed in Annex 1. It lays down measures designed to prevent or, where that is not practicable, to reduce emissions in the air, water and land from the above mentioned activities including measures concerning waste, in order to achieve a high level of protection for the environment taken as a whole, without prejudice to Directive 85/337 and under other relevant Community provisions.”

And article 3 sets out what it describes as general principles governing the basic operations of the operator and requires member states to ensure that installations are operated in such a way as :

“(a)

All appropriate preventive measures are taken against pollution, in particular through application of the best available techniques; and

“(b)

No significant pollution is caused.”

The reference to “best available techniques” is something that we shall be concerned with later in the judgment. They are defined, or at least described, in paragraph 11 of article 2 of the Directive in these introductory terms:

“11.

‘best available techniques’ shall mean the most effective and advanced stage in the development of activities and their methods of operation which indicate the practical suitability of particular techniques for providing in principle the basis for emission limit values designed to prevent and, where that is not practicable, generally to reduce emissions and the impact on the environment as a whole.”

There is quite a bit more about best available techniques included in the legislation to which we will come at a later stage of the judgment. At the same time, and in the context of the objectives and purpose of the Directive that we have already seen, there is reference to the relevance of geographical, environmental and technical considerations and characteristics. In that connection one may refer to recital 17:

“Whereas emission limit values, parameters or equivalent technical measures should be based on the best available techniques, without prescribing the use of one specific technique or technology and taking into consideration the technical characteristics of the installation is concerned, its geographical location and local environmental conditions; whereas in all cases the authorisation conditions will lay down provisions on minimising long-distance of transfrontier pollution and ensure a high level of protection for the environment as a whole.”

Then article 9, paragraphs 3 and 4. Paragraph 3 reads:

“The permit shall include emission limit values of pollutants, in particular, those listed in Annex 3, likely to be emitted from the installation concerned in significant quantities, having regard to their nature and their potential to transfer pollution from one medium to another.”

Then paragraph 4:

“Without prejudice to article 10 [and I interpose that article 10 is an article of some importance that we will come to later] the emission limit values and the equivalent parameters and technical measures referred to in paragraph 3 shall be based on the best available techniques, without prescribing the use of any technique or specific technology, but taking into account the technical characteristics of the installation concerned, its geographical location and the local environmental conditions.”

The judge expressed his understanding of the effect of those provisions at a number of points in his judgment. It is fair to refer to three extracts. First in paragraph 71 of the judgment he referred to article 3 that I have already set out and added, having stressed the need for all appropriate preventative measures:

“I also consider that recitals 1, 6 and 8 of the Directive show that any decision must seek to prevent or minimise emissions to air. That being so, that question is as pertinent to the question of whether to grant or refuse as it is to the questions of which conditions shall be applied.”

Then in paragraph 82(a), dealing with the submission as to the overall relationship between ELV levels and general air quality, to which again I shall come, the judge said this:

“The Directive and the Regulations are explicit that the objective is prevention of an emission of NOx, or reduction as far as possible: see Directive Articles 1,3 and 6(1) [of which he quotes] and Regulations 8, 11(2) and 3(1), all of which look to achieving the lowest possible levels of emissions.”

In 82(c) he said this:

“The ‘high standard for protection of the environment as a whole’ in article 3 is not intended to be related to a particular figure. The phrase ‘high level’ is a way of describing the objective in the first recital which is to prevent or reduce to the irreducible minimum the emission of pollutants as a whole.”

Quinn raised two somewhat separate objections to that part of the judge’s analysis. First, it said that the terms of the Directive do not support the judge’s observations in terms of reduction of emissions “as far as possible” or to an “irreducible minimum”. Second, no account is taken by the judge of the balance between protection and the cost of compliance such as is mentioned in article 174/3 of the EC Treaty already read.

I do not agree with the first of those criticisms. The judge was right to stress the importance clearly attached by the Directive to seeking “as far as possible” (and that latter is of course an important qualification when considering objections in terms of unreasonableness) to reduce emissions and gasses considered harmful. I will return to the second matter of balance at a later stage of the judgment.

These are helpfully summarised by the judge in his paragraph 152:

“Conclusions on reasoning by Chester City Council

I conclude that Chester City Council

wrongly failed to consider whether alternative configurations, size or design to that proposed would have produced lower emissions of specified pollutants, and in particular oxides of nitrogen (NOx), either in the context of a BAT analysis, or in determining whether the permit should be refused or granted;

in particular, failed to consider whether the use of an oxyfuel process with a different number and size of furnaces would produce lower emission levels of oxides of nitrogen (NOx) and/or would constitute BAT;

misinterpreted and misapplied statutory note SG 2 when considering the date at which the SG 2 recommended emission limit of 500 mg/NM3 was to be applied to a new installation for the manufacture of container glass;

Took into account an immaterial consideration, namely the emission limits set at other existing UK plants, when considering whether the levels permitted at the application site were BAT for the purposes of the European and UK statutory guidance;

misinterpreted and misapplied the BREF document on achievable emission limits and achieved levels;

acted in breach of the statutory European and UK codes for pollution control by imposing emission limits for oxides of nitrogen (NOx) which were higher than those which were achievable

alternatively, failed to have regard to a material consideration, namely the evidence before it that lower emission limits were achievable for oxides of nitrogen (NOx) if alternative primary secondary techniques were used, and/or if secondary techniques were applied;

when considering the relationship between emissions of CO2 and NOx failed to take into account a material consideration, namely whether oxides of nitrogen (NOx) are global pollutants or are to be treated as greenhouse gases;

failed to give adequate or intelligible reasoning for its conclusions on the above;

failed to give adequate or intelligible reasoning for its conclusions that the storage of oxygen on site in an oxyfuel process was a matter that justified rejecting that process

failed to give adequate or intelligible reasoning for its conclusions concerning the effects on air quality.”

The judge also held that Mr Durham had not legitimately acted under his delegated powers, which was of course a reason in the judge’s view for quashing the certificate separate from any criticism that might be made of the terms thereof.

The order was in what might be thought to be a somewhat unusual form. The judge quashed the certificate but he also made Declarations which related directly to the first of his criticisms of Chester’s reasoning set out in his paragraph 152(a) already quoted. The judge’s Declarations were in the following form:

“When determining, for the purposes of the Pollution Prevention and Control (England and Wales) Regulations 2000, whether to refuse or grant a permit, and in the latter case what conditions should be included in a permit issued in respect of an installation as defined in Schedule 1, a regulator authority,

a.

when considering the application of the tests of ‘Best Available Techniques’ to the proposed installation, must have regard to, but not be confined to, the particular size, configuration, design and process proposed in the application for a permit for an installation containing an activity listed in Schedule 1;

b.

in considering an application for a permit to conduct a Schedule 1 activity in an installation requiring a permit under the Regulations, (including its decision whether to grant or refuse a permit pursuant to Regulation 10(2)), a regulator authority, when applying the test ‘Best Available Technique’ must consider whether the installation would better satisfy the relevant criteria set out in the Regulations, if the installation were devised to be of a different size, configuration, design or process from that proposed by the applicant for the permit;

is entitled, pursuant to Regulation 10(2) to refuse to grant a permit to an installation for the carrying out of an activity of a description in Schedule 1 of the Regulations, if the Best Available Technique analysis conducted by it shows that the installation would better satisfy the relevant criteria set out in the Regulations, if the installation were devised to be of a different size, configuration, design or process from that proposed by the applicant for the permit.”

The judge also gave permission to appeal but on a strictly limited basis. Leaving aside an issue on which he also gave permission as to the actual operation of his order, to which I will come at the end of the judgment, the permission in relation to the substance of the judgment is in these terms:

“Permission to appeal is given to the Interested Party, limited to a) arguing that the degree to which the technique, production capacity and process configuration proposed by the applicant is determinative of the process, criteria and considerations by which the application is to be judged by the regulator authority.”

It will however be immediately seen that even if Quinn were to succeed in this court on the matter on which the judge granted permission, that is to say the judge’s first objection in paragraph 152(a), the certificate would still stand lawfully quashed if Quinn could not dislodge the other objections in respect of which the judge did not grant permission, and also unless Quinn could establish that Mr Durham indeed had lawful delegated power to give the certificate in the first place. Of the judge’s further objections, at least those set out in sub-paragraphs (c), (g), (h) and (i) of paragraph 152 were, in the light of the judge’s exposition of them earlier in the judgment, very substantial. To some extent therefore it has to say that the judge’s grant of permission in the context of this certificate raise matters that were only theoretical.

Quinn, no doubt appreciating that difficulty for them, applied to this court for permission to appeal in effect on everything that the judge had said. Without at this stage descending into technicalities, we heard those complaints as the appeal progressed. It will however be convenient to start with an issue that, although it featured in only second place on Mr Drabble’s helpful summary list of the issues that he wished to raise before this court, is in fact of fundamental importance to the whole of his case. That concerns the implications for decision under the IPPC Directive of the Requirements of Environmental or Air Quality Standards (“EQS”) that are laid down under other parties of the community environmental regime.

As we have seen, Directive 96/62 was brought into operation at the same time as the IPPC Directive. Under it, standards are laid down for ambient air quality; that is, the air quality in an area taken as a whole rather than, as in the IPPC Directive, the contribution to that air quality or the damage to it emanating from emissions from a particular plant or institution. In that context, Quinn pointed to the frequent references already noted in the IPPC Directive to the objective of achieving a high level of protection for the environment. It set out its case in the following terms in paragraph 12 of its grounds of appeal:

Accordingly, the protection of the environment was to be judged according to whether the acts under consideration, in this case the building of the glassworks, would reduce the local EQS below the standards required by or under Directive 96/62. It was agreed, or at least not challenged, that the NO_X limit of 1000 mg per cubic metre permitted for the first year of operation would not have the effect of so reducing the local EQS. Accordingly, that limit would achieve the high level of protection that the Directive required and therefore fulfilled the terms of the IPPC Directive.

The implications of this argument are fundamental. That is because (as Mr Drabble QC for Quinn frankly agreed, and indeed urged) if it is correct the argument renders otiose the whole of the rest of this case, and indeed renders otiose most of the structures set up by the United Kingdom to operate the IPPC regime. That is because under this argument the only question that the regulator need ask himself is whether a proposed installation will raise the local EQS above the permitted level. There was therefore no need for Mr Hosker to go through the process of investigation that he described in his evidence and certainly no need for him to include the annually reducing levels of permitted emissions. The only question that the court could ask itself when reviewing his work was whether EQS had been met. The errors that the judge considered that he had made were, all of them, irrelevant.

That was the basic version of this argument. Perhaps a subset of it was that the Directive also called for a balance between level of protection and cost. There could be no justification for requiring an installer to incur extra costs on more protection, such as, in the present case, protection that would reduce levels of emission below the 1000 feature, if the existing installation before that protection already met EQS.

The judge considered this argument in detail and rejected it in a comprehensive finding in his paragraph 82. He did not give permission to appeal against that part of his decision. I would respectfully adopt what the judge said, but because of the important role this argument played before us I will say something of my own about it.

I find the argument wholly unconvincing, both when one considers the general pattern of environmental protection and when one considers the detail of the legislation. As to environmental protection in general, there is a clear, obvious and centrally important distinction between the overall requirements of air quality and management on the one hand, and on the other the precautionary requirements imposed on installations of a particular sort that have been identified as potentially causing environmental problems. The latter are subject to strict and detailed regimes, which of course share the general aspirations of community environmental protection as a whole, but which address particular cases in detail.

That is particularly so of the national legislation and guidance that fills out the broader IPPC guidance, of which an example agreed on all sides to be relevant to our case is SG2. To put it bluntly, those who for their commercial purposes introduce potentially polluting operations have to be closely controlled, and cannot freeload on non-polluting local citizens by simply claiming that the EQS to which we all contribute has not yet been damaged. That, in my respectful view, could not be put better than it is in recital 1 to the IPPC Directive:

“Whereas the objectives and principles of a Community’s environment policy, as set out in Article 130r of the Treaty, consist in particular of preventing, reducing and as far as possible eliminating pollution by giving priority to intervention at source and ensuring prudent management of natural resources, in compliance with the ‘polluter pays’ principle and the principle of pollution prevention.”

That is why detailed controls that do not refer to EQS are provided for in the IPPC Directive. When pressed on this point, Mr Drabble was reduced to saying that the detailed controls were only machinery to enable a simple answer to be given to the single question of whether EQS was to be breached. If that were indeed their sole purpose, they are puzzlingly elaborate.

There are two other structural questions relevant to this argument. First the argument assumes that EQS levels are a datum or mark of Community approval. That is not so. As recital 4 to Directive 99/30 makes clear, they are minimum requirements. That is the Directive that lays down specific air quality values and the recital reads as follows:

“Whereas the limit values laid down in this Directive are minimum requirements; whereas, in accordance with Article 130t of the Treaty, Member States may maintain or introduce more stringent protective measures; whereas, in particular, stricter limit values may be introduced to protect the health of particularly vulnerable categories of the population, such as children and hospital patients; whereas a Member State may require the limit values be attained before the dates laid down in this Directive …”

True it is that particular reference is made to potential requirements of young or infirm persons, but that is not the limit of this guidance. Recital 4 makes it quite clear that the values are minima and are not to be regarded as in themselves necessarily fulfilling Community requirements. That fact alone makes it extremely unlikely that when the legislator speaks of “a high degree of protection of the environment” he thinks that that protection begins and ends with the present numbers of EQS.

Second, the regime argued for by Quinn would be hopelessly impracticable. Mr Drabble pointed to the provisions for continued monitoring of installations and said that action could be taken against his client’s plant if EQS were breached in the longer term. But that raises the question: action against whom? The point about EQS is that we are all in it together. If three more glassworks appear in Ellesmere Port, so that EQS is breached, which of them is to be required to take measures to reduce the EQS level to that which is acceptable? It is precisely for that sort of reason that the Community legislation sets limits, more stringent than the general environmental limits, plant by plant.

Against that background, it is hardly surprising that when we look at the terms of the legislation there is absolutely nothing in it that support Quinn’s argument. Quite apart from the general point that the draftsman was largely wasting his time if all that matters is the EQS, there is almost total absence of mention of the latter in the IPPC Directive. And the only mentions that it does receive are revealing.

Those provisions reveal two things. First, the whole assumption of them is that the requirements under the Directive are something different from EQS. Second, the further assumption is that IPPC requirements will in general terms be more stringent than simply the meeting of EQS. It is the exceptional case where that does not happen that is provided for in article 10. The same is true of the national arrangements, which it is agreed are intra vires the IPPC Directive.

In SG2 one finds in paragraph 1.6 what might in modern parlance be described as a mission statement. I do not set it out. It was relied on by Quinn. It says not a word about EQS.

I would therefore reject this part of Quinn’s case. That has a considerable impact on the rest of the arguments, to which I now turn.

I have already set that out. At first sight, and by the light of nature, it appears simply to be ordinary common sense. However, there is a background. In the present case Quinn’s plant, which it planned and started to build without any approval of the regulator, consists of two cross-fired regenerative furnaces. That has to be contrasted with the original plan of three oxyfuel furnaces. We then have to turn, in that context, to the provisions on best available technique. For that we have to go back to the Directive. That is to be found in the first instance in article 2 paragraph 11. I have already set out the first paragraph of that. It continues:

“- ’techniques’ shall include both the technology used and the way in which the installation is designed, built, maintained, operated and decommissioned,

- ‘available’ techniques shall mean those developed on a scale which allows implementation in the relevant industrial sector, under economically and technically viable conditions, taking into consideration the costs and advantages, whether or not the techniques are used or produced inside the Member State in question, as long as they are reasonably accessible to the operator,

- ‘best’ shall mean most effective in achieving a high general level of protection of the environment as a whole.”

We then go to article 9, paragraphs 3 and 4, which have already been set out. In that context the judge made, as we have seen, two leading criticisms of Chester’s process of deciding on this certificate, which he set out in sub-paragraphs (a) and (b) of his paragraph 152, already quoted.

Quinn’s objections to the judge’s criticism were to some extent on two levels. First, that it was not open to the regulator to consider alternative configurations at all. The installed configuration, albeit chosen without consultation with the regulator, had nonetheless to be taken by him as a given. Second, that in any event on the facts of this case the oxyfuel process referred to by the judge was not “comparable” or “available” in the terms used by the Directive.

The first of these as a general proposition seems on its face very surprising, and it is inconsistent with the terms not only of the Directive but also of the domestic Regulations. I revert to the Directive. We have already seen how “techniques” are defined in article 2.11. They include not only the technology used but specifically the way in which installations are designed, built, maintained, operated and decommissioned. That is something that the regulator has to assess. It is difficult to see how he does that if he is limited to the installation of this plant, bearing in mind that he has to find the best available techniques.

Second, article 9.4 requires that in assessing ELVs the regulator should take into account the technical characteristics of the installation. It does not say that he is constrained by that but that he can give it weight, as the judge’s Declarations envisage. Third, in his paragraph 72, which we have already looked at, the judge referred to Regulation 3(1)(c), which accurately echoes the terms of article 2.11 of the Directive. The judge commented in his paragraph 74:

“If one were unable to examine the choice of process, configuration or size, the words in Regulation 3(1)(c) would have little effect, save to parts of the design which did not go to its fundamentals.”

I respectfully agree. The regulation repeats what the Directive says about techniques and is, for reasons I have already given, inconsistent with the regulator being limited to the particular installation.

Fourth, the judge drew attention to the Manual. He emphasised again that nobody had told him that the Manual was ultra vires and no-one suggested that to us. He read a passage from it at paragraphs 74 and 75 of his judgment:

“‘12.8 Availability: Where there is a choice, the technique that is best overall will be BAT unless it is not an “available technique”. There are two key aspects to the availability test:

a)

what is the balance of costs and advantages? This means that a technique may be rejected as BAT if its costs would far outweigh its environmental benefits; and

b)

can the operator obtain the technique? This does not mean that the technique has to be in general use. It would only need to have been developed or proven as a pilot, provided that the industry could then confidently introduce it. Nor does there need to be a competitive market for it. It does not matter whether the technique is from the outside the UK or even the EU.’

75.

Paragraphs 12.10, 12.11 and 12.12 all contain passages showing that the choice of size and configuration are matters to be considered in a BAT assessment. One should read them, remembering that emissions are affected to a very substantial degree by the design of the furnace and the nature of the process, as the distinction between air and oxygen fuelled processes indicated.”

All of these factors point against Quinn’s submission. I would therefore reject the general complaint about the judge’s declaration. Nor is right to say on the specific facts of this case that the oxyfuel process referred to by the judge was not available or comparable. It is agreed that oxyfuel process in furnaces of less than 500 tonnes per day was proven technology. That was therefore “a technique reasonable accessible to the operator” in the terms of article 2.11 of the Directive. That indeed appears to have been the technique envisaged in Quinn’s original application for planning permission for this site, filed in May 2000. That was said not to be comparable because to achieve the production that Quinn wants there would have to be three furnaces. But that assumes the argument rejected by the judge, which I also do not accept for reasons already given, that the configuration of the factory is a datum.

However, all of these arguments, interesting as they are, and potentially important as they might be in another case, seem to me to be of only limited importance in the present case. That is because best available technique, BAT, is not an end in itself but a means towards achieving protection of the environment. We have seen the definition of “best” in article 2.1: best shall mean most effective in achieving a high general level of protection of the environment as a whole. The UK regulator is given guidance in achieving that objective in SG2. Further on in his paragraph 75 the judge cites the Manual and what it says about the relevance of sector guidance notes. The passage in question reads as follows:

“Basic principles for determining BAT

“12.10

As stated above, determination of what is BAT must ultimately be made on a case-by-case basis and taking into account that individual circumstances may affect BAT judgements and what are appropriate permit conditions.”

The following paragraphs describe the steps that would be necessary if starting such an exercise from scratch. However, where sector guidance notes are available, they will have taken account of options and it may be quite adequate to rely on those notes on a baseline for what is BAT.

That theme is taken up in SG2 itself. In paragraphs 2.1 and 2.2 this is said, starting with emission limits and other provisions:

“2.1

This section contains emission limits, mass release rates and other requirements that are judged for the generality of the activities within the sector to represent BAT.

Contained emissions to air associated with the use of BAT

2.2

Guidance is given below on emission limits and other requirements which are achievable for key substances using the best combination of techniques”.

Table 3, headed “Emissions when associated with the use of BAT” sets out an annual average of 500 for NO_X. It is because of the function of BAT being to lead to pollution control that those BAT values described in SG2 are expressed in terms of emission limits. Now SG2 is merely guidance: but there should be reasons shown for departing from it. Even using the cross-fire furnaces that Quinn have installed, it is perfectly possible to achieve the SG2 level of NO_X of 500 mg per cubic metre now, and not just by 2009. That can be done by the installation of secondary adjustments to the system. That is asserted in the affidavit of a Mr Kitson, filed on behalf of Quinn, in which he sets out the circumstances in which Quinn is in the process of applying for a new permit to take the place of the one that stands quashed. He says this in paragraph 4.5:

“The new IPPC application proposes the installation of secondary catalytic reduction-denox system. This is intended to reduce the NO_X emissions from the plant to below 500 mg/m³ and it is intended that the new IPPC permit would contain a limit of 500 mg/m³ for NO_X.”

Quinn objects to doing that because it says it will cost them money; apparently calculated, as far as we can see from the papers, at a figure of 67p per tonne produced, or a cost increase of some 3%. Quinn’s objection to being asked to take that step reverted to the claim that I have rejected, that they cannot be required to do anything if emissions in the area as a whole remain below currently stated minimum EQS values. But if BAT in terms of emission standards is properly applied, no rational reason was advanced, indeed no other reason at all was advanced, for departing from the SG2 guidance that BAT is that which reduces emissions to or below 500. We know that such technology is available in this case and indeed is currently being installed.

I would therefore not interfere with the judge’s declarations, whilst noting that the future course for the regulator in this case should be guided by the provision of SG2 as explained in the preceding paragraphs.

That suffices to dispose of the appeal but I will shortly deal with some of the other issues.

I have already set out the provision in the standing orders under which Mr Durham purported to be acting. I regret that we have not seen the whole of the council’s standing orders. This rule 29 reads very much as if it were a provision to deal with matters of minor and genuinely technical importance; quite different from the detailed codes of delegation, in for instance planning matters, that one would expect to see in a local authority’s standing orders. The only way in which it could be said that the decision whether to grant this permit was governed substantially by matters of fact or technical factors would be, if Quinn were right, that measurement of the current EQS answered all the questions. But that was not in any way the issue that Messrs Hosker and Durham at their meeting thought they were addressing. As the judge pointed out, they thought they were authorising a departure from the immediate requirements of SG2, as indeed in the event they were. That was plainly an exercise of discretion; and the more it is urged upon us that the terms of the Directive are such that assessment and balance are required, the more it becomes unlikely that the decision is one that is covered by rule 29.

I would therefore hold that Mr Durham should not have taken this decision. I would venture to add (although these considerations do not affect what is an issue of the construction of the local authority’s procedures) that I do not find it satisfactory that decisions of this importance should be taken with the informality and lack of public scrutiny that occurred in this case. Mr Durham decided these issues in the course of a private meeting in which he was orally briefed by Mr Hosker, a meeting of which no note was taken, and the only evidence of which is the recollection of these two gentleman in evidence three months later; recollections that, apart from anything else, strikingly differ according to how long the meeting went on for. Mr Durham so acted after he had been advised by the council’s solicitor that he could in this case lawfully act under delegated powers, advice that again was not recorded in writing. The local authority will wish to consider whether different arrangements should be made in future, and whether its current view of the reach of rule 29 meet its public obligations of transparency and public scrutiny.

That this decision was taken by a person who was not authorised to take it is sufficient in itself to lead to the quashing of the permit, and would be sufficient even if Quinn were right in the argument that it raises on EQS. A decision that is ultra vires is unlawful even if the same outcome could have been achieved by the decision maker lawfully exercising his powers.

How did Mr Hosker reach his year-by-year reducing requirement? Mr Hosker said in the first part of his evidence that he was well aware that 500 was the SG2 limit. So why did he depart from it? An explanation was given in the acknowledgment of service, in terms set out by the judge in his paragraph 118. Those were as follows:

“The Defendant does not accept that it departed from SG2 guidance. Paragraph 1.15 of the Guidance provides that ‘Compliance with the new provisions should normally be achieved by the dates shown. Permits should be drafted having regard to this compliance timetable.’ The date shown for compliance on Table 1 of the guidance for container glass, of the type manufactured at the plant was 1^st April 2009. The annual average was fixed at 500 mg/Nm3 per cubic metre. This was precisely the condition imposed in the licence.”

As the judge pointed out, that was just a straightforward error. Table 1 in SG2 is the provision for existing installations and not for new installations such as Quinn’s. The judge commented in paragraph 119 on the evidence that he had received as to the council’s thought-processes in that respect. I do not set it out. It makes for uncomfortable reading. As the judge said, that Mr Hosker did indeed make this error is underlined by paragraph 25 of his second statement, which defends the permit by saying the standards set in SG2 will be achieved by “the due dates”; but by the correct table in SG2 the due date is now: reference to staged compliance is entirely inappropriate.

Mr Drabble asked the rhetorical question: why on earth should the regulator not do so? There is a short answer to that. He should not do so because SG2 has a different regime in respect of existing and of new plants.

The evidence even before the judge was that emissions could be brought down to 500 by secondary techniques, and as we have seen that is what Quinn is indeed doing. That was plainly a relevant consideration, even though Quinn would have wanted to argue about it on the grounds already set out earlier in this judgment, by which I have indicated I am not persuaded.

The judge also criticised failures of reasoning. Those certainly exist especially in regard to the confusion over what part of SG2 was being referred to. But I have no doubt they would not prevail if, which is not the case, everything else in the permit had been lawful.

The regulator did not take account of the effect of oxides of nitrogen on what I will say, for shorthand, is global warming. Mr Drabble said that this was a major feature in the judge’s reasoning, and it was wrong because there was almost nothing in the legislation to justify consideration in connection with nitrogen of emissions of oxygen or other factors which would contribute to global warming. I am not as persuaded as Mr Drabble that the judge was completely wrong about this but I do recognise that he may have gone too far in that direction. I do not intend to extend this judgment by examining that point further because I fear I disagree with Mr Drabble that this was a major consideration weighing on the judge. It certainly affected him, but he had set out very many more substantial questions upon which he was rightly concerned.

As I have already indicated, I would uphold the judge’s quashing of this order. That however leaves the question of the extent and time of the relief. The judge refused an application that as a matter of discretion the permit should not be quashed at all, but he suspended the effect of that quashing order until the decision of this court, and he also gave Quinn permission to appeal on the wider question of whether the quashing of the permit should be “deferred”.

Quinn’s argument, put very shortly at this stage, was that it would be excessive and unjust to impose the costs and inconvenience of quashing the permit, thus causing the works to close, in the comparatively short period before Chester has resolved Quinn’s fresh application for a new IPPC permit. Mr Drabble, speaking with proper diffidence, sought permission from this court to appeal on the question of whether the permit should be quashed at all. He contended that it was not a proportionate response, given the lack of environmental harm resulting from emissions of NO_X within the limits proposed by the permit.

That contention was based on the view of the Directive that I have already rejected when dealing with Quinn’s substantive case. Once the judge’s substantive findings are held, any challenge to his decision to issue a quashing order is hopeless. In the light of those findings, that was plainly a proper exercise of discretion on his part. Indeed, a quashing order was inevitable. Permission to appeal on this issue is refused.

There remains the question of whether the quashing of the permit should be deferred. It would be unrealistic to answer that question by reference to the circumstances that existed at the time of the judge’s order. The reality is that events have moved on and this court must decide whether, if it dismisses the substantive appeal, the stay on the judge’s order should be lifted with immediate effect, or whether, in the circumstances as they now exist, the stay should be allowed to remain in place for a further short period. That is a step which the court will not normally contemplate but the present case is unusual.

It has to be said, and I do not shrink from saying, that Quinn has not helped its cause by its building of the works before it has resolved either the planning or the IPPC issues. It appears further to have shown some insensitivity in apparently significantly increasing its activities, and the workforce who depend on them, in the period during which this appeal has been pending. In paragraph 12 of his judgment the judge said this:

“In this context, I must point out, as will become clear below, that the circumstances in which the expenditure was incurred by Quinn, and its employment of large numbers of people, are unusual, and represented the taking of a calculated risk on its part.”

I also cannot refrain from quoting the view of the member of the public, Mr Anians, who filed evidence in this case and said in paragraph 1 of his witness statement:

“I am a resident in Elton, Chester. I, my wife and children have lived in the village for over 20 years. My home is situated directly alongside the perimeter fence of a new glass plant of Quinn Glass Limited. This glass plant is, I believe, the largest in Europe, possibly the world, yet it has been built without planning permission!”

I do not normally approve of the introduction of exclamation marks into evidence but in my judgement Mr Anians is well justified in adding such a mark to the end of that paragraph. Nonetheless, despite all this we must look dispassionately at the relevant facts, which are as follows.

First, the combined effect of Regulations 9(1) and 32(1)(a) is it would be a criminal offence for Quinn to operate the installation after the permit had been quashed and before any new permit was granted. The evidence is that the furnaces could not be simply switched off but that there would be a cooling down period of about 14 days. It would be a strong step to expose Quinn to criminal liability in respect of the operation of the installation during that period. Some deferment of the quashing order, until about the end of June, seems therefore to be called for on any view.

The court must proceed on the basis that if the quashing of a permit was deferred until about the end of June to deal with the problem of the cooling down period, Quinn would shut the furnaces down and cease to operate the installation thereafter unless and until a new permit was granted. It would be too great a risk for it to expose itself to the sanctions of the criminal law and to the further enforcement powers that Chester would then have to consider using under the Regulations.

Closure of the plant, even for a relatively short period, would have potentially very serious consequences not just for Quinn but also for its workforce and third-party supplies and customers. As already indicated, the latter categories excite very substantially more sympathy than does Quinn itself, but Quinn can say that its present predicament results not just from the commercial risk that it took by proceeding as it did but also from the errors made by Chester in the discharge of its function as regulators.

A further and important consideration is that, as already stated, installation of secondary abatement plant is in hand to bring the emissions down to the level of 500 mg per cubic metre, which ought to have been imposed in the first place. That work is subject to planning permission, which we understood is expected to be uncontroversial. “Hot commissioning” is said to be scheduled for the first week in July. If the stated timetable is adhered to, and the secondary abatement plant proves effective, as the evidence that we have from Quinn said that it would be, emissions can be expected to be brought down to the appropriate level within a matter of weeks if Quinn is permitted to continue to operate the installation.

Meanwhile, Quinn’s fresh application for a permit has been under active consideration by Chester since it was received in January 2006. Chester’s evidence is that it is likely to be late July 2006 before a draft determination is made and, because of the period required for public consultation, September 2006 before a final determination is made.

Mr Tromans, for Chester, in addition to drawing our attention to that timetable, pointed to a further disadvantage of quashing the permit pending the decision on that application. The permit covers a wide range of matters, some of which would remain relevant even in the event of temporary closure of the plant. For example, provisions relating to the pools of cooling water on site. The quashing of the permit would leave a possible lacuna in Chester’s enforcement powers in relation to such matters.

In the light of all those considerations, I am satisfied that in general terms it would be wrong and disproportionate for the quashing order to take effect so as to require the closure of the plant during the short period before Chester determine the fresh application for a permit. The law is sufficiently upheld by the judgment of this court, the declarations made by the judge which we are upholding, and the fact that the existing permit will be quashed as soon as a decision is taken by Chester on whether and on what terms to grant the new permit. The relevant objective of the Directive is substantially achieved by the expected completion within a matter of weeks of installation of the secondary abatement plant. Of course, if Chester refuses a new permit the plant will have to close anyway. The situation then will be a very different one.

However Mr Gordon QC, for Rockware, sought to persuade us that our obligations under EU law prevented us from deferring the operation of the quashing order. He referred to Berkeley v Secretary of State for the Environment [2001] 2 AC 603 and in particular the speeches at page 609 of Lord Bingham of Cornhill and page 616 of Lord Hoffmann. The issue in that case was whether a planning permission should be upheld, notwithstanding substantial non-compliance with the requirement of the environmental assessment Directive and the relevant national implementing Regulations. Their Lordships emphasised the narrowness of the court’s discretion in such a case, referring inter alia to the duty laid on member states by article 10 of the EC Treaty and the obligation of national courts to ensure that Community rights are fully and effectively enforced. For those reasons they held that the planning permission had to be quashed.

In the present case, submitted Mr Gordon, the Directive requires members states to take the necessary measures to ensure that no new installation is operated “without a permit issued in accordance with this Directive”, article 4, and the court would be acting in breach of its Community obligations by allowing the installation to continue to operate if the permit had been found not to have been issued in accordance with the Directive. Indeed, Mr Gordon did not shrink from the contention that HHJ Gilbart had been in breach of his Community obligations even by staying his order pending an appeal to this court.

I am prepared to proceed on the basis that Mr Gordon’s submissions would have force if the court were contemplating a refusal of the quashing order altogether, so as to allow Quinn to operate the installation indefinitely despite the errors in the decision to grant the permit. That is far from this case. The court is concerned only with the deferment for a short period of the effect of the quashing order that it has upheld and where, as I have already said, the relevant objective of the Directive is substantially achieved by the expected completion within a matter of weeks of the installation of the secondary abatement plant.

In those circumstances, the limited exercise of discretion that is contemplated as to the timing of relief cannot in my view be said to be incompatible with the court’s obligations under Community law. The speeches in the House of Lords in Barclay plainly did not exclude entirely and as a matter of principle any such limited exercise of this court’s discretion provided the court proceeds, as this court does, with proper respect for its obligations under Community law.

Accordingly, I would dismiss the appeal, refuse the applications for permission to appeal and uphold the relief granted by the judge. But I would extend the stay that he granted on the operation of the quashing order until Chester’s decision on the new permit application. That order is made on condition that both Chester and Quinn give undertakings to this court, which counsel will or will not indicate they are able to give when I have completed the judgment, to proceed with all reasonable despatch, both in relation to the permit and in installing the secondary measures.

All parties will have liberty to apply to this court, any such application being reserved to myself.

LORD JUSTICE RICHARDS: I agree.

SIR CHRISTOPHER STAUGHTON: I also agree.