England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM QBD, ADMINISTRATIVE COURT
MR JUSTICE BURTON
CO/2157/2006

		Royal Courts of Justice Strand, London, WC2A 2LL
		28/06/2007

Lord Justice Carnwath :

Introduction

1. The business of OSS is the collection of waste lubricating and fuel oil from places such as garages and workshops, and its conversion into marketable fuel oil. The issue in short is when the material ceases to be "waste" for the purposes of the Waste Framework Directive ("WFD"). Is it at the completion of the process of preparing it for use as fuel (as argued by OSS), or when it is actually burnt (as argued by the Environment Agency)? The judge agreed with the Environment Agency.
2. The answer is of considerable economic significance, because of the costs of complying with the higher standards required for waste processes. This has become particularly relevant to OSS' business since 2005, when the Waste Incineration Directive ("WID") was applied to existing waste operations. If the judgment is upheld, a power station using OSS fuel is subject to the more costly controls applying to a waste incinerator, which would not apply if it were burning virgin fuel oil.
3. OSS had in fact attempted to anticipate the impact of the Directive. For many years the largest and most profitable part of its business had consisted of reprocessing waste oils to produce "recycled fuel oil" ("RFO"). In preparation for the WID, it invested in new processes, enabling it to produce a higher quality product called "clean fuel oil" ("CFO"), which it was hoped would not be regarded as waste. The "cleanness" of the product is a matter of dispute. As Burton J put it:

"… there is considerable and presently irresoluble factual dispute as to whether, even on their own case, the processes which OSS carry out are indeed sufficient to render the CFO sufficiently safe to burn, or sufficiently comparable to, or materially indistinguishable from, a natural fuel (if such be required), or compliant with the BS2869, even if, which the Agency does not accept, such British Standard is relevant at all." (para 14)

4. That factual dispute, if it arises, is for another day. If the judgment is upheld, it will be academic. We are directly concerned with the limited legal issue on which the judge granted permission to appeal:

"… whether a lubricating oil, thus not originally used as fuel, which becomes waste can thereafter be burnt other than as waste."

5. The same legal issue arose in another case before the judge, relating to the business of Solvent Resource Management Ltd ("SRM"). Because the factual issues were largely agreed in that case, the judge directed that it be treated as the lead case for resolving the legal issues, on the basis of submissions from both companies. His decision in favour of the Agency rendered academic the factual issues in the OSS case. Although permission to appeal was granted to both companies, only OSS is before us.
6. Although not before us, the SRM case is a good illustration of the practical significance of the Agency's claimed distinction between use as fuel and other end-uses. In simple terms, SRM recovers waste solvent materials, from which it produces so-called "PGDs" (Product Grade Distillates) for sale on the market. It is common ground that, following those processes, the PGDs may be handled and sold on the basis that they have ceased to be waste. However, according to the Agency, and as confirmed by the judge, that approach does not apply to the same products when used as fuel in SRM's own plants. Thus it is a notable (and unashamed) feature of the Agency's position that identical substances, following completion of the same recovery processes, may lose or retain the character of "waste" depending on the purposes for which they are then used.

The legal framework

7. It is unnecessary to repeat Burton J's helpfully comprehensive exposition of the European and domestic sources of the law. Much of the same ground was covered by Stanley Burnton J in Castle Cement v Environment Agency [2001] Env LR 46. I would also pay special tribute to the exhaustive review and analysis of the relevant source material, European and domestic, in the judgment of Lord Reed in the Court of Session in Scottish Power Generation Ltd v Scottish Environment Protection Agency [2005] SLT 98 ("Scottish Power").
8. The arguments before us have turned principally on the interpretation of passages in the judgments of the Court of Justice in ARCO Chemie Nederland v Minister Van Volkshuisvesting [2002] QB 646 ("ARCO"), and in three subsequent cases: Palin Granit Oy [2002] 1 WLR 2644; R (Mayer Parry Recycling Ltd) v Environment Agency [2004] 1 WLR 538; and Niselli (2004) Case C-457/02. A short outline of the relevant European law will be sufficient as background to the discussion of the issues between the parties.
9. The governing legislation is the Waste Framework Directive ("WFD"), which dates from 1975 (now consolidated with subsequent amendments in Directive 2006/12/EC). The main definitions are in Article 1:

i) "Waste" means "any substance in the categories set out in Annex I which the holder discards or intends… to discard" (1(a)).

ii) "Holder" means "the producer of the waste or the natural or legal person who is in possession of it" (1(c)).

iii) "Producer" means "anyone whose activities produce waste ('original producer') and/or anyone who carries out pre-processing, mixing or other operations resulting in a change in the nature or composition of this waste" (1(b))

10. Categories of Waste are listed in Annex I, but that list is of limited help because it ends with a catch-all group: "any materials, substances or products which are not contained in the above-mentioned categories".
11. Annexes IIA and IIB list operations constituting respectively "disposal" and "recovery" of waste. "Disposal operations" include D10 ("Incineration on land") and D11 ("Incineration at sea"); "recovery operations" include R1 ("Use principally as a fuel or other means to generate energy"), R2 ("Solvent reclamation/regeneration"), R3 ("Recycling/reclamation of organic substances which are not used as solvents …"), R9 ("Oil re-fining or other reuses of oil"), and R10 ("Land treatment resulting in benefit to agriculture or ecological improvement").
12. It will be seen that burning or "incineration" of waste may qualify as either a disposal operation (D10 or 11) or a recovery operation (R1), depending on whether its "principal" purpose is to generate energy. The distinction was explained by the ECJ in Case C-228/00 Commission v Germany [2003] ECR I -1439:

"… the essential characteristic of a waste recovery operation is that its principal objective is that the waste serves a useful purpose in replacing other materials which would have had to be used for that purpose, thereby conserving natural resources…

The combustion of waste therefore constitutes a recovery operation where its principal objective is that the waste can fulfil a useful function as a means of generating energy, replacing the use of a source of primary energy which would have had to have been used to fulfil that function." (para 45-6)

13. The ordinary English meaning of the word "discard" is an imperfect guide to its significance in the definition of waste. Other language versions have equal status in European law, and may have a slightly different emphasis. For example, the French "se défaire de", or the German "entledigen", might perhaps be better translated as "get rid of": see my discussion in Mayer Parry Recycling v Environment Agency [1999] 1 CMLR 963 para 24-30. I there concluded on the then state of the authorities (including Vessoso & Zanetti [1990] ECR I –1461; Tombesi [1997] ECR I–3561; Inter-Environnement Wallonnie ABSL v Région Wallonne [1997] ECR I-7411)

"The general concept is now reasonably clear. The term 'discard' is used in a broad sense equivalent to 'get rid of'; but it is coloured by the examples of waste given in Annex I and the Waste Catalogue, which indicate that it is concerned generally with materials which have ceased to be required for their original purpose, normally because they are unsuitable, unwanted or surplus to requirements…"

Although much of the rest of the judgment has been overtaken by subsequent authority, that still seems to me a fair general summary of the intended meaning of the word "discard", taken on its own.

14. It is clear, however, that it is only part of the story. The following points, some of which will need further discussion, can be found in the cases:

i) The concept of waste "cannot be interpreted restrictively" (ARCO para 40).

ii) Waste, according to its ordinary meaning, is "what falls away when one processes a material or an object, and is not the end product which the manufacturing process directly seeks to produce" (Palin Granit Oy para 32).

iii) The term "discard" "covers" or "includes" disposal or recovery within the terms of Annex IIA and B (Wallonie para 27; ARCO para 47); but the fact that a substance is treated by one of the methods described in those Annexes does not lead to the necessary inference that it is waste (ARCO para 48-9).

iv) The term "discard" must be interpreted in the light of the aims of the WFD, and of article 174(2) of the treaty, respectively:

a) The protection of human health and the environment against the harmful effects caused by the collection, transport, treatment, storage and tipping of waste; and

b) Community policy on the environment, which aims at a high level of protection and is based on the precautionary principle and the principle that preventive action should be taken (Palin Granit Oy para 23).

v) Waste includes substances discarded by their owners, even if they are "capable of economic reutilisation" (Vessoso & Zanetti [1990] ECR I –1461 para 9) or "have a commercial value and are collected on a commercial basis for recycling, reclamation or re-use" (Tombesi para 52).

vi) In deciding whether use of a substance for burning is to be regarded as "discarding" it is irrelevant that it may be recovered as fuel in an environmentally responsible manner and without substantial treatment (ARCO para 73).

vii) Other distinctions, which may be relevant depending on the nature of the processes, are –

a) between "waste recovery" within the meaning of the WFD and "normal industrial treatment" of products which are not waste ("no matter how difficult that distinction may be") (Wallonie para 33);

b) between a "by-product" of an industrial process, which is not waste, and a "production residue", which is (Palin Granit Oy para 32-37 – see further below).

15. Article 3(1) requires Member States to take "appropriate measures" to encourage "the prevention or reduction of waste production and its harmfulness", and in particular (Art 3(1)(b)) to encourage:

"(i) the recovery of waste by means of recycling, re-use or reclamation or any other process with a view to extracting secondary raw materials; or

(ii) the use of waste as a source of energy."

Article 4 requires them to take necessary measures to ensure that

"… waste is recovered or disposed of without endangering human health and without using processes or methods which would harm the environment"

For that purpose, Article 10 requires any establishment carrying out "recovery operations" (as listed in Annex IIB) to obtain a permit.

Related Directives

16. We have been referred to a number of other directives dealing more specifically with various types of waste. Where relevant, the provisions of such special legislation will normally prevail over those of the WFD (see e.g. Mayer Parry para 57).
17. The Agency relies in particular on the Waste Oils Directive (consolidated in 75/439/EEC) ("WOD"). This applies to waste oils, including waste lubricating oils, the feedstock of OSS. Article 1 defines "waste oils" as:

"… any mineral-based lubrication or industrial oils which have become unfit for the use for which they were originally intended, and in particular used combustion engine oils and gearbox oils …"

It is notable that this definition is not in terms related to the definition of "waste" in the WFD (which came later).

It also defines "regeneration":

"… any process whereby base oils can be produced by refining waste oils, in particular by removing the contaminants, oxidation products and additives contained in such oils"

and "combustion":

"… the use of waste oils as fuels with the heat produced being adequately recovered."

18. Article 3 requires priority to be given, subject to "technical, economic and organisational constraints": first, to regeneration of waste oils; secondly, to combustion "under environmentally acceptable conditions"; and lastly to "their safe destruction or their controlled storage or tipping."
19. The judge noted (para 33) that it had not been suggested by OSS that its process of producing CFO from the contaminated lubricating oils, amounted to "regeneration" as defined. In this court, the respective positions were less clear. The EA sought to argue that the WFD should be interpreted in the light of the WOD, with the effect that waste oils which had not been "regenerated" could not cease to be waste. In response, OSS sought to resile from any concession that their process did not constitute regeneration within the meaning of the WOD.
20. I do not regard this debate as helpful. Since this was not a live issue before the judge, we have no findings on the issue, nor the evidence on which we could make findings. In any event, I would require persuasion that the link proposed by the Agency is sound. Although the WOD no doubt has precedence within its own terms, those terms do not indicate any intention to modify the WFD definition of waste. In any event, the issue raised by the appeal is directed solely to the latter. Our answer will not prejudge any questions which may arise thereafter under the WOD.
21. As already noted, the WID was applied to existing plants (that is, "incineration" or "co-incineration" plants) from December 2005. The plants to which it applies are widely defined. Thus the definition of a "co-incineration plant" includes a plant –

"whose main purpose is the generation of energy or production of material products and … which uses wastes as a regular or additional fuel; or in which waste is thermally treated for the purpose of disposal".

Thus it is clear that any likely purchaser and user of CFO would be caught, if CFO is indeed "waste".

22. Mr Howell QC also sought to rely on various other "daughter" directives, relating to different categories of product, in order to demonstrate that his interpretation of the WFD was consistent with the overall scheme. He referred to the well-established principle that European provisions must be interpreted in the light of the objectives of Community law as a whole, and "its state of evolution at the time on which the provision in question is to be applied" (see CILFIT v Ministry of Health [1982] ECR 3415 para 20).
23. Without doubting the principle, I do not find such comparisons helpful in the present context. Apart from providing the impetus for the present proceedings, WID does not seem to me to throw any light on the definition of waste in the WFD. The other comparisons raise as many questions as they solve, as Mr Wyatt QC for DEFRA demonstrated. Although they provide more specific rules for the situations to which they apply, they do not indicate in my view any intention to modify the general meaning of the WFD definition of waste.

The rival formulations

24. Before the judge, Mr Drabble QC for OSS proposed a carefully formulated "end of waste test" (see judgment para 35). Under that test (as broken down in to its component elements), waste materials may be transformed into a substance which is not waste if:

i) The substance has been "deliberately produced to a specification"; and

ii) it has been processed so as to be suitable for an identified use or uses -

a) without further processing,

b) in the same way as the non-waste-derived or virgin materials which it will in practice be used to replace,

c) under the same conditions of environmental protection,

d) without any greater harm to human health or to the environment when it is stored, transported, handled or used than in the case of the relevant virgin materials; and

iii) it is certain that it will be put to that identified use or one of those identified uses.

25. Mr McCracken QC, who appeared with Mr Tromans for OSS on the appeal, did not resile from that formulation. However, in his final submissions he reduced the test to two simple questions:

i) Is the material sufficiently analogous to the virgin product/raw material genuinely to replace it?

ii) Is the material analogous in terms of environmental risks in use?

26. In the Agency's submission, if the intended end-use is combustion, the materials remain waste until that is completed, regardless of any prior treatment. Mr Howell summarised the Agency's case in his skeleton:

"… materials which are contained in lubricating oils that have been discarded and have become waste and which are destined to be combusted cease to be waste when they are combusted with or without prior processing. At that point energy is recovered if the materials are used as a means to generate energy and the materials incinerated are disposed of if they are not."

27. The Agency recognises one exception to that rule, that is where waste fuel oil is recovered for use as fuel oil, provided that it is chemically and physically identical to the original material, and requires no further processing. The basis of this concession is that the process is "oil re-refining" (within recovery operation R9), which puts the oil back into its original state, so that it may replace a virgin oil which would otherwise be used. This is not relevant to the present case, which relates to the recovery of waste lubricating oils.
28. DEFRA takes a slightly broader view than the Agency. As Mr Wyatt puts it in his skeleton:

"… the burning of lubricating oil amounts to discarding it as unwanted for its normal purpose and accordingly constitutes recovery of waste. However material recovered from waste lubricating oil which is burned as fuel, or as a component in fuel, is not being discarded if it is material which has the same characteristics as a virgin material for which such use is a normal use."

The test embodied in the emphasised words is intended to be strict, and to be equivalent to words used by the ECJ: material "which is analogous to a raw material, with the same characteristics as that raw material…" (ARCO para 96), or material which "can hardly be distinguished" from virgin material (Niselli para 52).

29. Those various positions need to be related to the question posed by the appeal:

"… whether a lubricating oil, thus not originally used as fuel, which becomes waste can thereafter be burnt other than as waste."

OSS' answer is "yes", whichever of its formulations is adopted. DEFRA's answer is also "yes", although the conditions it would impose are more stringent. The oil must be processed to a standard which makes it "hardly distinguishable" from virgin oil. The Agency's answer is "no"; a waste lubricating oil which is destined for combustion remains waste until it is burnt, regardless of the standard of any prior processing.

30. Burton J's answer was also "no", although by slightly different reasoning. On his reading of paragraphs 94-97 of the judgment in ARCO (see below), even if the waste had been subject to "a complete recovery operation", and had become "ex-waste" for certain purposes, it could then revert back to being waste; "i.e. its conversion to ex-waste was not indefeasible" (para 58). Thus, in the case of SRM, although the PGDs were no longer waste when sold, in so far as SRM decided to burn them, they would become waste again:

"… What has occurred is that after the first recovery operation which has, if it has, been sufficient to render the product suitable for one purpose, there is then a further recovery operation (R1) when the energy is recovered, and the material is discarded" (para 61).

31. The principal question raised by the appeal is whether the negative answer given by the judge, with the support of the Agency, was right. That turns mainly on the interpretation of the judgment in ARCO. Although the OSS and DEFRA tests both result in an affirmative answer to the appeal question, the differences between them are significant, and it is desirable that we should express a view on them so far as we are able.

ARCO

32. The Court was concerned with two forms of waste material intended for use as fuel: "LUWA bottoms" in Arco itself, and powdered wood chips in the linked case of Epon. The substances arose as follows:

i) ARCO was a chemical manufacturer. "LUWA bottoms" were by-product of a manufacturing process by which it produced molybdenum. "Those LUWA-bottoms, which have a calorific value of between 25 and 28 MJ/kg, are destined for use as a fuel in the cement industry." (para 11)

ii) Epon was an electricity generating company. Its project involved "…wood residues from the construction and demolition sector delivered in the form of wood chips, which were to be transformed into wood powder and used as a fuel to generate electricity." (para 22) The conditions of acceptance required the wood chips to be "free of sand, paint particles, stone, glass, plastic particles, textile and fabric particles and metal parts", but allowed "a limited quantity of sleepers, water-impregnated wood and preserved (creosoted) wood" (para 28). It was alleged (by the claimant environmental group) that these conditions allowed acceptance of wood containing carcinogenic materials or dioxins, or materials such as "paint, impregnating substances, glues, plastics and solvents" (para 30).

33. Both cases were concerned with materials destined for combustion. Epon's case is the more useful in the present context, because it was a purchaser of materials admittedly derived from waste. The main question was whether they remained waste following their transformation into materials suitable for burning. To that extent it touched directly on the question in the present appeal. ARCO, by contrast, was the original producer of the material in question, and there was a possible issue whether the material was waste at all.
34. The various questions had been posed in different ways by the two national courts, and they were reformulated by the ECJ for the purpose of its judgment. Varying views on each issue and sub-issue had been expressed by the three parties, the Commission and the five intervening governments (discussed at length in the Advocate General's Opinion). Against this confusing background, the judgment requires some creative deconstruction to make it manageable. For present purposes, I will concentrate on the most relevant parts, which are those dealing with the court's first question (relating to both ARCO and Epon) and the last (relating only to Epon). The latter part includes the passage (paras 94-97) which has given rise to particular disagreement in the present case.
35. The first question, and the answer to it, were reasonably clearcut. The question was:

"… whether it may be inferred from the mere fact that a substance such as LUWA-bottoms or wood chips undergoes an operation referred to in Annex IIB to the Directive that the substance has been discarded and whether it is therefore to be regarded as waste for the purposes of that directive" (para 44).

The court answered the question in the negative. Annexes IIA and IIB were lists of "disposal and recovery operations as they occur in practice". Some were formulated in "abstract terms" which could apply equally to raw materials which were not waste. It followed that:

"The answer to the first question in both cases should therefore be that it may not be inferred from the mere fact that a substance such as LUWA-bottoms or wood chips undergoes an operation listed in Annex IIB to the Directive that that substance has been discarded so as to enable it to be regarded as waste for the purposes of the Directive." (para 51)

36. The second group of questions in each case was recast by the Court, which treated them all as concerning "the definition of the term 'discard' for the purpose of determining whether a particular substance is waste" (para 52-3). It divided them into three parts:

i) (both cases) concerning the method of using a substance (discussed at paras 54-73);

ii) (ARCO) concerning the method whereby the substance is produced (paras 74-88);

iii) (Epon) relating to recycling operations (paras 89-97).

37. The third question, as expressed by the court, was:

"… whether, in order to determine whether the use of wood chips as fuel is to be regarded as constituting discarding, it is necessary to consider whether the waste from the construction and demolition sector from which the chips were made has already undergone, prior to burning, operations which are to be regarded as a discarding of the waste, namely operations (recycling operations) to render the waste suitable for re-use as a fuel, and if so, whether that operation may be regarded as an operation for recovery of waste only if it is expressly mentioned in Annex IIB to the Directive or whether it may also be so regarded if it is analogous to an operation mentioned in that annex." (para 89)

38. The court's formal answer (which repeated paragraph 97) was:

"The fact that a substance is the result of a recovery operation within the meaning of Annex IIB to the Directive is only one of the factors which must be taken into consideration for the purpose of determining whether that substance is still waste, and does not as such permit a definitive conclusion to be drawn in that regard. Whether it is waste must be determined in the light of all the circumstances, by comparison with the definition set out in Article 1(a) of the Directive, that is to say the discarding of the substance in question or the intention or requirement to discard it, regard being had to the aim of the directive and the need to ensure that its effectiveness is not undermined."

Comment

39. I should say at once, without disrespect, that I do not understand the second part of that answer. It is apparent from the submissions before us that I am not alone. The court had already accepted that the mere fact that a material is destined for use as fuel does not mean that it is to be categorised as waste (answer 1). In the present answer, it also accepts that, if before being burnt, a substance has undergone an Annex IIB recovery operation, that is "one of the factors" to be taken into account in deciding whether it is "still waste". Impliedly, therefore, it is accepting that it is possible for a substance which was waste to cease to be waste, even if it is destined ultimately for use as fuel. So far so good.
40. The problem comes at the next stage. The court does not, as might have been expected, go on to identify the other factors which need to be taken into account, and the test to be applied, in deciding whether the material has ceased to be waste. Instead it simply asserts that the decision must be made in the light of "all the circumstances", and "by comparison with" the Article 1(a) definition of "the discarding of the substance, or the intention or requirement to discard it". As a response to the particular case, that is meaningless. Epon wanted to reuse the material, not to discard it. No amount of reference to "all the circumstances" would change that fact. No doubt the material was discarded by the original user; but the issue was not whether it was then waste; but whether it had since ceased to be waste. Paragraphs 94-96
41. I turn to the supporting reasoning in paragraphs 94 to 96. This needs to be seen in the context of the arguments of the parties, as summarised by the court (paras 90-3). The claimant environmental group had argued that the wood remained waste, because it was still impregnated with toxic substances. Epon argued that it was used "in an environmentally friendly manner", with no more polluting effect than "a primary raw material"; that it had been subject to a recovery operation within R2 of Annex IIB ("Recycling/reclamation of organic substances…"); and that in any event Annex IIB was not exhaustive of possible recovery operations. The Commission and the Governments argued that sorting and transforming into chips did not result in its losing its character as waste:

"… Such operations do not constitute a recovery operation for the purposes of Annex IIB to the Directive but a simple pre-treatment of the waste. A substance ceases to be waste only when it has undergone a complete recovery operation within the meaning of Annex IIB to the Directive, that is to say when it can be processed in the same way as a raw material or, as in this case, when the material or energy potential of the waste has been used during burning." (emphasis added)

42. The court responded:

"94. In that regard, it should first be noted that even where waste has undergone a complete recovery operation which has the consequence that the substance in question has acquired the same properties and characteristics as a raw material, that substance may none the less be regarded as waste if, in accordance with the definition in Article 1(a) of the Directive, its holder discards it or intends or is required to discard it.

95. The fact that the substance is the result of a complete recovery operation for the purposes of Annex IIB to the Directive is only one of the factors to be taken into consideration for the purpose of determining whether the substance constitutes waste and does not as such permit a definitive conclusion to be drawn in that regard.

96. If a complete recovery operation does not necessarily deprive an object of its classification as waste, that applies a fortiori to an operation during which the objects concerned are merely sorted or pre-treated, such as when waste in the form of wood impregnated with toxic substances is transformed into chips or those chips are reduced to wood powder, and which, since it does not purge the wood of the toxic substances which impregnate it, does not have the effect of transforming those objects into a product analogous to a raw material, with the same characteristics as that raw material and capable of being used in the same conditions of environmental protection."

Comment

43. Again, I find the process of reasoning extremely obscure. Stanley Burnton J aptly described this part of the judgment as "Delphic" (Castle Cement para 45). In paragraphs 94 and 95 the court appears to concede the possibility of a "complete recovery operation" within Annex IIB, as proposed by the Governments. But that is not regarded as determinative, because the holder of the recovered material may still have an intention to discard it. However, the relevance and significance of the latter observation are again wholly unclear to me.
44. Lord Reed, in the Scottish Power case, thought this was no more than an acknowledgement of the theoretical possibility that recovered material might need to be discarded if its holder could not find a market for it. Advocate General Alber noted the same possibility in relation to recycled material in Mayer Parry (Opinion para 104), but added that it would probably be extremely rare for "a holder of material recycled at considerable expense to intend to discard it again". I accept that this is a possible explanation. But I find it very unlikely that the court would have thought it helpful or necessary to mention such a remote factor, which had no relevance to the case before the court. There was no suggestion that Epon needed a market for the material, or that it intended to do other than reuse the material in its own processes.
45. Burton J (para 60) thought that Lord Reed fell into error on this point. In his view, the court was using the word "discard" in the more technical sense of undergoing an Article IIA or IIB operation (as signalled by it in paras 46-7 of the judgment); and thus the court was indicating that, even after one complete recovery operation, a material might still be "discarded" by undergoing another. I find this explanation even more difficult to accept. Although the court had indicated that the term "discard" was apt to "cover" operations in Annex IIA and IIB, it expressly did not accept that undergoing an Annex IIB operation was tantamount to discarding. Its answer to the first question was to the opposite effect.
46. Paragraph 96 is more understandable, at least as a comment on the facts of the case. Since the process did not "purge the wood of the toxic substances which impregnate it", the resulting material neither had "the same characteristics" as a raw material nor was "capable of being used in the same conditions of environmental protection". One may infer that those are at least two of the objective factors which are regarded as relevant in deciding whether a material has ceased to be waste. However, it is difficult to understand why this point was said to be a fortiori. That term implies a logical connection (and one of degree) between those factors and the reasons given in the preceding passage for holding that a complete recovery operation was not enough. But, as already noted, that apparently turned solely on the subjective issue of intention to discard, not on objective factors which were in any way comparable with those referred to in paragraph 96. Indeed the court had earlier indicated that the environmental effect of the process was irrelevant to the issue of "discarding" (ARCO para 73).
47. After reading and re-reading this passage many times, it remains a mystery to me. Before attempting to draw any overall conclusions, I shall comment more briefly on the later cases to which we have been referred.

Subsequent cases

48. We have been referred to three subsequent cases in which the principles laid down by ARCO have been applied: Palin Granit Oy [2002] 1 WLR 2644; R(Mayer Parry Recycling Ltd) v Environment Agency [2004] 1 WLR 538; and Niselli (2004) Case C-457/02.
49. Palin Granit Oy concerned a quarrying business. Stone left over after quarrying, which was the wrong shape for sale, was stored and used within the quarry for such things as side embankments and ramps, or as aggregates. The national court asked whether the material was to be regarded as waste, given in particular that it was kept for use in its original form and without adverse effect on the environment. The ECJ recalled the distinction, drawn in ARCO between "the product primarily sought" and "a production residue":

"The leftover stone is only a secondary product and the undertaking seeks to limit the quantity produced. According to its ordinary meaning, waste is what falls away when one produces a material or an object, and is not the end product which the manufacturing process directly seeks to produce." (para 32).

Such a secondary product might, consistently with the objectives of the Directive, be treated as "a by-product" which is to be exploited rather than "discarded". However, in view of the wide definition given to the term waste-

"… the reasoning applicable to by-products should be confined to situations in which the reuse of the goods, materials or raw materials is not a mere possibility but a certainty, without any further processing prior to reuse and as an integral part of the production process" (para 36).

On the facts of the case, reuse would involve potentially long-term storage operations, which would be a burden to the owner. Reuse was not "certain", but only "foreseeable in the longer term", with the result that the left-over stone must be regarded as "extraction residue" which the holder "intended to discard" (para 38).

50. This passage, as I understand it, is the source of the final condition in Mr Drabble's test. Although it was concerned with the original producer of the waste, rather than a subsequent holder, it introduces the useful concept of a "secondary product", which is to be exploited rather than discarded. In principle, it seems reasonable to apply the same concept to a subsequent process which transforms the waste from a production residue into a distinct product, which is marketed as such.
51. Both Mayer Parry and Niselli were concerned with scrap metal. They are relied on principally by DEFRA for its narrow formulation of the test. They also provide some insights in the way ARCO has been interpreted by the court itself.
52. Mayer Parry collected scrap metal, including packaging waste, from industrial sources, and by a process of inspection, sorting, cleaning and shredding, produced a product known as "grade 3B material". That material contained impurities such as paint oil, and non-metallic particles. It was sold to steelmakers who melted it down, removing the impurities, to produce ingots, sheets or coils of steel. The national court referred to the ECJ the questions whether such material had ceased to be waste for the purposes of the WFD, and whether it had been "recycled" for the purposes of the Directive (94/62) on packaging and packaging waste. Since it was common ground that the material was produced from packaging waste, the court regarded the issue of "recycling" under that Directive as the first issue. Having regard to the definition of "recycling", it was held that this would be achieved only when there was produced "a new material or a new product possessing characteristics comparable to those of the material of which the waste was derived" (para 75). The court did not give a separate answer to the question under the WFD, holding that it would follow that given under the packaging waste Directive as the "special" legislation.
53. In Niselli similar reasoning was applied to a case concerned directly with the WFD. The materials in that case were ferrous materials, obtained from dismantling machines and vehicles, which Mr Niselli was transporting without the necessary waste licence, for use as secondary raw materials in the steel industry. It is surprising that the case reached Luxembourg, since it seems that Mr Niselli was simply collecting and transporting waste materials, without any serious basis for claiming anything else. However, the second question raised by the Italian court was whether production residues could be excluded from the definition of waste if they could be reused for the same purpose without harm to the environment, and without requiring recycling operations. The court's answer was no. In the course of its response, it stated that the materials must continue to be classified as "waste" –

"… until they have actually been recycled into steel products, that is to say, until the constitution of the finished products derived from the reprocessing for which they are intended. In the earlier phases they cannot yet be regarded as recycled, since the reprocessing has not been concluded. Conversely, … the point at which the materials in question cease to be classified as waste cannot be fixed at an industrial or commercial stage subsequent to their reprocessing into steel products, because, from that point, they can hardly be distinguished from other steel products made from primary raw materials (see, for the particular case of recycled packaging waste (Mayer Parry)…" (para 52, emphasis added)

54. DEFRA rely on the emphasised words in support of their narrow definition. However, I do not think the context bears the weight put on it by them. The reasoning was closely related to that in Mayer Parry, and in particular its interpretation of "recycling" of ferrous materials. There is no indication that the court was intending to lay down a wider rule. In any event, the expression "hardly distinguishable" was used by way of comparison, not with raw materials, but with the resulting steel products.

Discussion

55. As this review demonstrates, a search for logical coherence in the Luxembourg case-law is probably doomed to failure. A fundamental problem is the court's professed adherence to the Article 1(a) definition, even where it can be of no practical relevance. The subjective "intention to discard" may be a useful guide to the status of the material in the hands of the original producer. However, it is hard to apply to the status of the material in the hands of someone who buys it for recycling or reprocessing; or who puts it to some other valuable use. In no ordinary sense is such a person "discarding" or "getting rid of" the material. His intention is precisely the opposite.
56. Understandably, the court has held that a material does not cease to be waste merely because it has come into the hands of someone who intends to put it to a new use. But that should not be because it still meets the Article 1(a) definition in his hands; but rather because, in accordance with the aims of the Directive, material which was originally waste needs to continue to be so treated until acceptable recovery or disposal has been achieved. Unfortunately the court has consistently declined invitations to develop workable criteria to determine that question. Instead, it continues to insist that the "discarding" test remains applicable, even where the "holder" is an end-user such as Epon, whose only subjective intention is to use, not to get rid of, the materials in issue.
57. In Mayer Parry Advocate General Alber frankly summed up the court's position as he saw it:

"The Court of Justice has thus refused to make classification of a material as waste dependent on its economic value, its fitness for reuse… or the environmental hazards posed by it… The holder's conduct can be appraised only with regard to his intentions, a fact which causes the body applying the law considerable difficulties.

The Court of Justice solves this problem by inferring an intention to discard the substance from objective indicators; in doing so it has regard both to all the factual circumstances and to the aim of the waste Directive…" (Mayer Parry Opinion para 108-9).

58. Similarly, in Palin Granit Oy, the Court itself, having noted the lack of precision in the definition of waste, commented:

"Directive 75/442 does not provide any decisive criteria for determining the intention of the holder to discard a given substance or object. Nevertheless, the Court, which has been asked on a number of occasions for preliminary rulings on whether various substances are to be regarded as waste, has provided a number of indicators from which it may be possible to infer the holder's intention. The classification ... must be made having regard to those factors and in the light of the aims of Directive 75/442." (para 25)

59. In other words, although the Court continues to play lip-service to the "discarding" test, in practice it subordinates the subjective question implicit in that definition, to a series of objective indicators derived from the policy of the Directive. What is required from the national court is a value judgment on the facts of the particular case in the light of those indicators.

Applying the definition in practice

60. Lord Reed's judgment in the Scottish Power case seems to me an impressively thorough example of the required approach. The particular case concerned the use by Scottish Power of "waste derived fuel" ("WDF") (treated sewage sludge obtained from Scottish Water) as an alternative to coal. The judgment contains a painstaking review of the European case-law. Paragraphs 135-7 of the judgment contain an exhaustive enumeration of the indicators which can found in the cases. On the facts of the particular case, the judge held that the product had not ceased to be waste, essentially for three reasons:

i) The WDF could not on the evidence be used under the same conditions of environmental protection as the coal which would otherwise be used;

ii) The operation which took place at the sludge treatment centre neither fell within any of the categories listed in Annex IIB, nor was it analogous to any such operation:

"It is clear that the sewage sludge is not recycled; and there is no distinct substance reclaimed from the sludge by extracting moisture from it and forming it into pellets. If one asks what is recovered from the sludge, the answer is, energy; and if one asks how it is recovered, the answer is, by burning."

iii) The project for the sludge treatment centre was driven principally by Scottish Water's need to establish a new means of disposal, to replace disposal into the sea (to comply with the Directive on urban waste water) (paras 139-141).

61. That can be contrasted with the decision of the Administrative Law Division of the Dutch Council of State, given on 14^th May 2003, Icopower BV v Secretary of State ("Icopower"). The case concerned the transport to Sweden of "energy pellets", produced in the following way (as described in the judgment):

"The appellant collects industrial waste products. These waste products consist of a mixture of paper, cardboard, untreated wood, plastic foil, textiles, possibly mixed with a limited percentage organic leftover substances (canteen waste) and/or glass, stone and cans. After acceptance a pre-selection takes place and the leftover substances are homogenised and reduced in size on a shredder, after which the ferro-metals are separated from the rest of the substances. After that low calorific fractions are separated from high calorific ones in a screen drum. Hereafter a separation takes place of non-ferro-metals and different components. The remaining flow, called 'fluff' by the appellant, is brought into the correct humidity degree, homogenised and pressed into energy pellets. The energy pellets are sold to electricity and heat-producing plants where they are used as (additional) fuels in these installations…"

62. The court noted that the pellets were produced with the sole aim of their use as fuel; that it was not claimed that they contained pollutants such as heavy metals; and that they were used in the same way as regular fuels, with no special precautionary measures being needed to protect the environment. On this basis it concluded that the energy pellets were "equivalent to regular fuels" and that, applying the reasoning of the ARCO judgment, they could not be characterised as waste under Article 1(a).
63. Although we are not called upon to decide the correctness of that decision, I see no reason to doubt it. It seems to me a practical and common sense approach to the issue, which is consistent with the letter and spirit of the Directive and with the case-law. It is also consistent with the objective of encouraging the recovery of waste materials for uses which replace raw materials. It should be enough that the holder has converted the waste material into a distinct, marketable product, which can be used in exactly the same way as an ordinary fuel, and with no worse environmental effects. It cannot be said that such a material is being "discarded" in any ordinary sense of the term, and there is nothing in the objectives of the Directive which requires any fictitious assumption to that effect. The energy pellets would perhaps have failed DEFRA's test, because they were not "hardly distinguishable" from the alternative fuels. But, as I have said, I do not think such a general test can be extracted from Niselli. Nor do I see any reason for it. The objectives of the Directive do not include mimicry.

Conclusions

64. I return to the rival formulations of the "end of waste test". The Agency's test is in my view too narrow. If it were correct it would have provided a complete answer to the ARCO and Epon cases. The mere fact that the materials in both cases were destined for combustion would have been enough to ensure their continuing categorisation as waste. However, that was not the court's view. Such test is contradicted directly by the court's answer to the first question in ARCO, and indirectly by the assumptions underlying the other answers. For the same reason, I would disagree respectfully with the Burton J's answer to the appeal question. To the extent that Stanley Burnton J adopted the same approach in Castle Cement (see para 51), I would also disagree with that part of his reasoning, although the decision may have been justifiable on its own facts (see para 52-8).
65. I have not disregarded the Agency's concerns, which impressed the judge, about the practical difficulties of applying a wider test, in particular of determining an appropriate comparator. However, I am not convinced that the difficulty is as great as they suggest. In most cases, as in Scottish Power and Icopower, the nature of the potential alternatives will be reasonably clear. In any event, as the ECJ has acknowledged, such difficulties are inherent in the imprecise nature of the Article 1(a) definition.
66. That is enough to provide an answer to the short question raised by the appeal. However, it is apparent from what I have said that I also regard the Secretary of State's formulation as too narrow, in so far as it requires the recovered material to be "hardly distinguishable" from an alternative raw material. In other respects, where the submissions of the Secretary of State differ from those of the Agency, I have found the former more persuasive. I agree, in particular, with the Secretary of State's reasons for not finding assistance in close analysis of the specific recovery operations in Annex IIB. As the court has made clear, the list is illustrative and general, and does not provide precise or exhaustive definitions. I also agree with the Secretary of State that the answer must be found within the WFD, rather than detailed comparisons with other related Directives, such as the WOD or the Packaging Waste Directive.
67. Further than this I think it would be wrong for us to go. However desirable it might be to have a definitive test, the ECJ has consistently declined invitations to provide one. It is not the function of the domestic court to fill the gap. Furthermore, the form in which the appeal has come to us makes it impossible to reach a more definite conclusion even on the facts of this particular case.
68. I hope, however, that in the light of this judgment, it may be possible for DEFRA and the Agency to join forces in providing practical guidance for those affected. It is unfortunate that the difficulties of interpreting the pronouncements from Luxembourg are compounded by the failure of the national authorities to agree a common approach. It is important that the national authorities should use their expertise and experience to assist those concerned with treatment and handling of waste, and also the courts (civil or criminal) who may well be faced with deciding individual cases without the benefit of any comparable expertise. The European Court recognises the scope for the national authorities to fill in some of the gaps left by the Directive. For example, in Niselli, while noting the lack of any "decisive criterion" for determining the holder's intention, it added:

"In the absence of Community provisions, Member States are free to choose the modes of proof of the various matters defined in the directives which they transpose, provided the effectiveness of Community law is not thereby undermined…" (para 34)

69. Finally, I should comment on the possibility of referring this case to the European court. None of the parties has formally requested a reference, although both OSS and the Agency have suggested that as preferable alternative to the rejection of their respective cases, while accepting that it is a matter for our discretion. I would not myself regard this as a suitable case for a reference, in view of the lack of definition of the facts. Furthermore, I would have doubts about the appropriateness of a further reference in this area of the law. I would hesitate to describe the matter as "acte clair", in the traditional sense. But that is not because of lack of opportunity for the Court to provide clarification if it had wished to do so. ARCO itself, in which the Court had the assistance not only of the Advocate-General and the Commission, but of five intervening states, would have been an ideal opportunity. One must assume that the decision not to do so was deliberate. Any lack of clarity is inherent in the imprecision of the test which the court has declared. I would wish to hear further argument before deciding that a reference in such circumstances would be necessary or useful.

Disposal

70. For these reasons, I would allow the appeal, and make no order other than to determine the appeal question in the affirmative.

Lord Justice Maurice Kay:

71. I agree.

The Master of the Rolls:

72. I also agree.