Proctor & Gamble International Operations SA v The Commissioners for HMRC

With the consent of the parties, the form of the hearing was V (video) on the Tribunal video hearing system. A face to face hearing was not held because of the restrictions imposed as a result of the coronavirus pandemic that were in place at the time of the hearing.

Prior notice of the hearing had been published on the gov.uk website, with information about how representatives of the media or members of the public could apply to join the hearing remotely in order to observe the proceedings. As such, the hearing was held in public.

The appellant is Procter & Gamble International Operations SA (“P & G”). The respondents are Her Majesty’s Commissioners for Revenue and Customs (“HMRC”). P & G were represented by Mr Jeremy White. HMRC were represented by Mr Howard Watkinson.

P & G marketed a product, Metamucil Suikervrij Orange (“MSO”), at the material times (and continues to market this product). MSO contains ispaghula husk, which has certain therapeutic and prophylactic properties, which we discuss below.

This appeal concerns a disputed decision, contained in a decision on a review dated 24 December 2015 (the “Decision”). The Decision was made under the Community Customs Code (Council Regulation 2913/92/EEC), article 12(1)-(6). The appeal is made and heard under the Finance Act 1994, section 16(1C)(b), (5), which provision is applied by the Customs Reviews and Appeals (Tariff and Origin) Regulations 1997 (SI 1997/534). The Decision upheld a Binding Tariff Information decision (“BTI Decision”) dated 13 October 2015. The BTI Decision classified MSO, for the purposes of the Community Customs Code as a “food preparation”. P & G considers that MSO is properly classified as a “medicament”.

Council Regulation (EEC) 2658/87 on the tariff and statistical nomenclature and on the Common Customs Tariff (the “Tariff Regulation”) classifies goods which are imported into the European Union Customs territory. Goods are classified according to headings specified in the Tariff Regulation. The issue is whether MSO is properly classified as a “medicament” (Heading 3004), as submitted by P & G or as a “food preparation not elsewhere specified or included” (Heading 2106), as submitted by HMRC. We discuss these headings further below. Broadly the proper classification of MSO as a “food preparation” or a “medicament” is determined by whether ispaghula husk may be properly described, for the purposes of the instruments we refer to below, as either a “plant extract” or, alternatively, a “chemically defined substance”.

It is convenient to note at this point that we were presented with the minutes of the Summary Report of the 193^rd meeting of the EU Customs Code Committee, which met on the 15^th and 16^th October 2018. The Customs Code Committee decided that ispaghula husk was properly classified as a “food preparation”, under heading 2106, rather than a medicament, under heading 3004 because ispaghula husk is not a “chemically defined substance” within the meaning of the CNEN to Chapter 30 ([8.4] of the minutes). The minutes then described the matter as “closed”. We make the relevance of the term “chemically defined substance” clear below. The conclusion of the Customs Code Committee is not binding on us and we respectfully depart from its conclusion. We also respectfully observe that we did not derive assistance from these minutes. There is no reasoning as to why ispaghula husk is not a “chemically defined substance”; there is no consideration of whether ispaghula husk is a “plant extract” (again we make clear why this is relevant below) and in any event this conclusion is expressly the product of a consensus of the Member States represented at this meeting, rather than legal debate ([8.4] of the minutes: “As the indicative round of the table showed that all represented Member States would classify the herbal preparation in question under heading 2106, considering that it does not comply with all the requirements mentioned in Additional note 1 to Chapter 30, it was suggested to close the case.”).

P & G submitted a witness statement of Mr Niall Lumley, who, at the material times was the Customs Manager and External Services Provider and Governance Manager at Procter & Gamble Technical Centres Ltd. Put short, Mr Lumley’s evidence provided background to the operation of P & G’s business and P & G’s products but was not material to this appeal. As it transpired, Mr Lumley did not give oral evidence and we have not taken any of the evidence Mr Lumley gave in his witness statement into account. The same is true and we adopted the same approach in relation to the evidence contained in the witness statement of Mr Nicolas Dirk Frans Stephan Broeckx, the Senior Marketing Manager for Global New Brand Development in Healthcare at the Procter & Gamble Company, Switzerland. And the same is also true of evidence contained in the witness statement of Mr Andy Taylor, employed as a “Higher Officer” at HMRC.

Oral evidence was given by Dr Marten Otten, a senior consultant gastroenterologist at a private clinic, the medical centre de Veluwe in Apeldoorn, the Netherlands. P & G’s position was that Dr Otten was an expert witness. HMRC objected to portions of Dr Otten’s evidence (the entirety of Dr Otten’s second witness statement), which HMRC submitted encroached on questions of law, which were not properly the subject of Dr Otten’s evidence but rather matters for the Tribunal to decide. In fact, Dr Otten’s evidence, in so far as it related to matters within Dr Otten’s expertise, concerned the therapeutic and prophylactic qualities of MSO which, as we have already observed, are not in dispute. Insofar as the composition of MSO is concerned and the process by which the most important component part, ispaghula husk (see below) is obtained and used to produce MSO, we consider Dr Otten’s evidence potentially relevant (and admissible) as to particular matters of fact but Dr Otten’s evidence is not that of an expert. We make it clear when we have relied on Dr Otten’s evidence to make findings of fact. Oral evidence was also given by Dr John Sidney Dowrick, who was, in material times, engaged by Procter & Gamble Product Supply (UK) Ltd, as a part-time consultant (“Qualified Person”, which we take to be a term of art), through Dr Dowrick’s business, JD Consultancy. We consider Dr Dowrick’s evidence to be admissible and relevant on questions of fact. Both Dr Otten and Dr Dowrick were cross-examined by Mr Watkinson.

We find the following facts (set out in bold): (there was no agreed Statement of Agreed Facts and Issues: we make clear the source of each finding, generally from the witness statements of Dr Otten and Dr Dowrick; if we make a finding of fact which reflects the content of a passage in a witness statement, we record that this passage was not the subject of cross examination, in the sense that the particular witness was not taken to it and tested as to its accuracy or credibility):

We should point out that both Dr Otten and Dr Dowrick gave their evidence and were cross-examined prior to the Tribunal admitting the issue as to whether ispaghula husk may be described as a “plant extract” for the purpose of the relevant legislative provisions we discuss below. The circumstances in which P & G (eventually) made an application for the plant extract issue to be admitted are aired in the Tribunal’s decision on costs released on 7 August 2020 (TC/2026/00458). Mr Watkinson did not make any application to have either Dr Otten or Dr Dowrick recalled for further cross-examination on the plant extract issue (neither did Mr White apply to recall either witness to give further evidence). Nevertheless, we expressly disregard: first, Dr Dowrick’s evidence (at [19] of Dr Dowrick’s second witness statement) that, in Dr Dowrick’s opinion, for the purposes of the relevant legislative provisions, ispaghula husk is not a “plant extract” because a “plant extract” is (according to Dr Dowrick) a “chemical component that is extracted by means of a solvent, such as water or alcohol.” Dr Dowrick also observes, at [21] of Dr Dowrick’s second witness statement, that “…it is not feasible to extract ispaghula husk by means of water or alcohol [as] to do so would cost the ispaghula husk to swell prematurely thus compromising its affect (sic) as a bulking agent.” The term “plant extract” is a matter of law for the Tribunal to decide and Dr Dowrick’s observations are not relevant. And second, we also disregard Dr Dowrick’s evidence (at [22] of Dr Dowrick’s second witness statement) that MSO is an “active substance”, for the purpose of the relevant legislative provisions, which issue is, again, an issue of law for the Tribunal to decide.

The function of a customs duty regime is to specify goods (apply a “nomenclature”) that attract specified rates of duty.

The EU is a party to the International Convention on the Harmonised Commodity Description and Coding System of 14 June 1983 (the “Harmonised System” or “HS”). The HS is administered by the Customs Cooperation Council, informally known as the World Customs Organisation (“WCO”): see Lyons, EU Customs Law (3^rd edn, OUP 2018), page 157).

Furthermore, the fact that the EU is a party to the HS means that the provisions of the HS take effect subject to the primary law of the EU but, importantly, the HS takes precedence over EU secondary law, which includes the EU legislative instruments we discuss below: EC Treaty, article 300(7), by and large replicated in TFEU, Article 218 (which replaced Article 300 of the EC Treaty) and in particular Article 216.2 (“Agreements concluded by the [EU] are binding upon the institutions of the [EU] and on its Member States”). The HS contains headings (each heading is given a four digit numerical code, the first two digits identify the chapter in question, whereas the latter two digits show the number of the heading in the chapter) and sub-headings, which refer to categories of goods (such as “food preparations” in heading 2106 and “medicaments” in heading 3004). As Lyons observes (op. cit., at page 157): “The parties to the HS [including the EU] undertake to use all the headings and sub-headings without addition or modification, together with their numerical codes, and to follow the numerical sequence of the [HS]. They also undertake to apply the General Rules for the Interpretation of the [HS] and all of the section and sub-heading notes and agree not to modify the scope of the sections and chapters into which the [HS] is divided.”: Article 3.1 (a) of the HS.

The provisions of the HS are amplified by “Harmonised System Explanatory Notes” (“HSENs”), issued by the WCO. HSENs may be used as an aid to construction of the HS, but HSENs are not legally binding. This is trite law. The HSEN for heading 2106 provides, relevantly, that “…food supplements, based on extracts from plants, fruit concentrates, honey, fructose etc… are often put up in packagings with indications that they maintain general health or well-being. Similar preparations, however, intended for the prevention or treatment of diseases or ailments are excluded ([and are properly classified under] heading 30.03 or 30.04).” It is common ground that this HSEN does not materially affect this appeal. HMRC accept that MSO has therapeutic and prophylactic properties specific to the treatment of medical disorders, in particular constipation.

The HSEN for heading 3004 provides, relevantly, that heading 3004 “…covers medicaments consisting of mixed or unmixed products, provided that they are…[p]ut up in measured doses or in forms such as… capsules, cachets, drops or pastels…or small quantities of powder, ready for taking as single doses for therapeutic or prophylactic use.” Again, the HSEN does not materially affect this appeal. It is common ground that MSO is sold in sachets, in measured doses.

So far as the provisions of EU law are concerned, we were referred only to the English text of the relevant instruments and we decide this appeal on that basis. Put short, as we have observed, the EU is obliged, under Article 3(1)(a) of the HS, to have the EU Customs tariff and statistical nomenclatures to be in conformity with the HS, to use the headings and subheadings of the HS (without addition or modification, together with the related numerical codes) and to follow the numerical sequence of the HS. We summarise the provisions of EU law on which we rely below, together with our observations on their application. The Tariff Regulation, Annex I contains the EU Combined Nomenclature (the “CN”); Annex I is updated annually by a Commission Regulation. The Tariff Regulation, provides, relevantly:

The CN at Annex I contains “General Rules of Interpretation” (“GIRs”) and the relevant headings and subheadings. GIR 1 provides that “The titles of sections, chapters and sub-chapters are provided for ease of reference only; for legal purposes, classification shall be determined according to the terms of the headings and any relevant section or chapter notes and, provided such headings or notes do not otherwise require, according to [the GIRs]”.

The two relevant headings in the GIRs, for the purposes of this appeal, are heading 2106 (“food preparations not elsewhere specified or included”) and 3004 (“medicaments (excluding goods of heading 3002, 3005 or 3006) consisting of mixed or unmixed products for therapeutic or prophylactic uses, put up in measured doses (including those in the form of transdermal administration systems) or in the forms or packings for retail sale”).

Note 2 to Section VI provides, relevantly: “…goods classifiable in heading 3004…by reason of being put up in measured doses or for retail sale are to be classified in [that heading] and in no other heading of the nomenclature.”

Note 1(a) to chapter 30 provides that Chapter 30 “…does not cover…foods or beverages…other than nutritional preparations for intravenous administration…”.

The cumulative effect of the HS, the Tariff Regulation and the CN at Annex I (heading 2106, heading 3004, note 2 to Section VI and note 1(a) to chapter 30, in the light of the GIR 1) is that a product (here MSO), which has a therapeutic or prophylactic use, put up in measured doses, is a “medicament” and (even if potentially classifiable under another heading, such as 2106), properly classified under heading 3004. And importantly, the EU, as a party to the HS, cannot legislate to change the content of the HS headings (put another way, the EU cannot legislate to change the meaning of “food preparation” for the purposes of heading 2106, or the meaning of “medicament” for the purposes of heading 3004). This is confirmed by the terms of Article 1(2)(c) and Article 3 of the Tariff Regulation, to which we have referred above.

Commission Regulation (EC) 1777/2001 inserted an “Additional Note” to chapter 3004 (the “Additional Note”), which provides: “Heading 3004 includes herbal medicinal preparations and preparations based on the following active substances: vitamins, minerals, essential amino acids or fatty acids, in packings for retail sale. These preparations are classified in heading 3004 if they bear on the label, packaging or on the accompanying user directions the following statements of (a) the specific diseases, ailments or their symptoms for which the product is to be used; (b) the concentration of active substances or substances contained therein”.

The European Commission issues Combined Nomenclature Explanatory Notes (“CNENs”), which like HSENs in relation to the HS, are an aid to construction but are not legally binding. The CNEN to the Additional Note provides, relevantly, that “…herbal medicinal preparations are preparations based on one or more active substances, produced by subjecting a plant or parts thereof to a process such as drying, crushing, extraction or purification…” and further provides that “’active substance’ means a chemically defined substance, a chemically defined group of substances … or a plant extract… [which] have medicinal properties for the prevention or the treatment of specific diseases, ailments or their symptoms.” TheCNEN also provides that “Heading 3004 does not cover, inter alia, food supplements or dietary preparations…”. We consider that the “process” described in the CNEN relates to the description and definition of the “herbal medicinal preparation” (which is “produced” by a process such as “drying, crushing, extraction” etc), rather than “active substance”. The syntax of the relevant passage in the CNEN, we consider, makes clear that the definition of “herbal medicinal preparation[s]” is that of “preparations based on one or more active substances” (“active substances” being defined subsequently in the CNEN); that definition is then extended to include a component that the “herbal medicinal preparation” must have been “produced” by “subjecting a plant or parts thereof to a process “such as drying, crushing extraction” etc. A “process such as drying, crushing, extraction or purification” more naturally relates to the “production” of a “herbal medicinal preparation”, rather than to a “chemically defined substance” (or a chemically defined group of substances), or a “plant extract”. Indeed, if the draftsperson of the CNEN had intended the “process” condition in the CNEN to relate to the “active substance”, rather than to the “herbal medicinal preparation”, we would have expected the provisions as to the “process” to be included within the definition of “active substance”. And in circumstances in which the definition of an “active substance” in the CNEN relates to the labelling requirements in the Additional Note, we cannot readily discern why the draftsperson of the CNEN would be at all concerned with how an active substance is “produced”, whereas we can far more readily see that the draftsperson is anxious to define a “herbal medicinal preparation” as one which has been “produced” by a particular type of “process”, as opposed to comprising, say, an unprocessed substance. Furthermore, on any view, whether the relevant “process” specified in the CNEN relates (as we think) to the definition of “herbal medicinal preparation”, or to the definition of “active substance”, the “process” need not be only one of “drying, crushing, extraction or purification”. The CNEN clearly states that the relevant “process” need only be one “such as drying” etc, so that a “process” which is not one of those specified in the CNEN but similar to them may “produce” a “herbal medicinal preparation” within the meaning of the CNEN (and thus the Additional Note). The same is true if the provisions as to “process” relate to the definition of “active substance”; an “active substance” which is a “chemically defined substance/group of substances” or a “plant extract” may be “produced” by a “process” that is not identical to that specified in the CNEN.

As it happens, although this question of construction as to whether the “process” specified in the CNEN relates to the definition of “herbal medicinal preparation” or, alternatively, “active substance” is important, for the reasons we give below it makes no difference in this particular appeal and we have delivered our decision on the alternative basis that the “process” relates to the definition of “active substance”.

The Additional Note is expressed to “include” (1) “herbal medicinal preparations” and (2) “preparations [other than herbal medicinal preparations]” based on one or more of an “active substance” comprised in an exclusive list of vitamins, minerals, essential amino acids or fatty acids. The CNEN makes it clear (using the CNEN as an interpretative aid) that “herbal medicinal preparations” need not be based on an “active substance” in this latter list, since the CNEN defines a “herbal medicinal preparation” as one “based on” different “active substances” (a “chemically defined substance, a chemically defined group of substances … or a plant extract”).

Since, as we have already observed, the EU may not legislate to change the meaning of any of the terms comprising the headings in the HS (that is, the EU may not legislate to change the meaning of either “food preparation”, in the heading 2106, or “medicament” in the heading 3004), the Additional Note cannot properly be read as cutting down the meaning of “medicament” in the heading 3004; neither may the Additional Note be read as extending the term “medicament” in heading 3004 (we note, in passing, that the term “includes” is only intelligible as a term which extends a meaning). So all that is left as a description of the Additional Note’s juristic role is that of a confirmation that products which are properly labelled (or have proper accompanying user directions) as to the particular disease or ailment which the product is to be used to treat, which specify the concentration of the “active substance” the product contains, may be properly classified as “medicament” under heading 3004.

Put another way, the Additional Note, at least on its face, permits products which are labelled in a manner compliant with the additional note to be conclusively treated, at least for the purposes of the Tariff Regulation, as “medicament”, under heading 3004. But because the EU is not permitted to legislate so as to change the meaning of the term “medicament”, in the heading 3004, a product, even if labelled in a manner compliant with the Additional Note, cannot be properly classified as a “medicament” if it does not have either therapeutic or prophylactic properties.

It follows, we consider, that a product, which is a “herbal medicinal preparation” (being a “preparation” which, assuming we are correct that the “process” specified in the CNEN relates to defining the “herbal medicinal preparation” and not the “active substance”) is “produced by subjecting a plant or parts thereof to a process such as drying, crushing, extraction or purification” (the phrase “such as” demonstrates that production methods other than those specified in the CNEN may yield a “preparation” which is a “herbal medicinal preparation”), that product is conclusively a “medicament”, under heading 3004 , if: (1) the relevant labelling and/or accompanying user directions specify both the disease or ailment the product is directed to and specify the concentration of an “active substance” which is either a “chemically defined substance” or a “plant extract” and (2) the “chemically defined substance, [or] … group of substances … or plant extract” have medicinal properties for the prevention or treatment of specific diseases, ailments or their symptoms (there is no corresponding requirement for the “active substances” which relate to “other” preparations, which are not “herbal medicinal preparations”, specified in the Additional Note but this requirement is clearly implicit for this latter “other” preparations, in order to ensure that the Additional Note does not have an effect which effectively illegitimately extends the meaning of “medicament” in heading 3004). Furthermore, there is no need, seemingly, in the provisions of the Additional Note, or, indeed, in the CNEN, to specify in the label, or any other documentation, any substance which does not fall within the notion of an “active substance.” In other words, so far as “herbal medicinal preparations” are concerned, the Additional Note does not require an exhaustive description of the product but only the specification of the concentration of the “chemically defined substance, [or] … group of substances” or the “plant extract” that have the medicinal properties for the treatment of specific diseases etc.

So far as the dispute between the parties in this appeal is concerned, it is common ground that if MSO is not a “medicament”, within heading 3004, MSO is a “food preparation” within heading 2106. So the burden on the Tribunal is to determine whether or not MSO is, indeed, a “medicament” under heading 3004.

It is convenient to summarise the effect, as we see it, of the relevant EU provisions as follows:

In relation to the application of the Additional Note, the only issue in dispute is whether the “packings” in which MSO is sold (the MSO label and the patient leaflet), in specifying the “active ingredient” as ispaghula husk anhydricum, is compliant with the Additional Note.

If ispaghula husk is a “plant extract”, within the meaning of the CNEN applicable to the Additional Note, P & G’s appeal succeeds on any view. HMRC have made no submission that the ispaghula husk does not have medicinal properties which relate to specific diseases or ailments. If ispaghula husk is indeed a “plant extract”, the ispaghula husk would be an “active substance” within MSO. If the “process” condition in the CNEN relates (as we consider) to the “herbal medicinal preparation”, MSO, in turn, would be a “herbal medicinal preparation”, being “produced” by a “process” of the combination of the winnowing of the ispaghula husk, the subsequent milling of that ispaghula husk and, thereafter, the blending of the ispaghula husk with the excipients, in the process we have described above. We have already noted that the CNEN does not specify any exhaustive process or an exhaustive list of processes which must be followed to produce a “herbal medicinal preparation”. Thus the labelling and patient information for MSO which specified the “active substance” concentration by reference to “ispaghula husk” (rather than arabinoxylans) would be compliant with the Additional Note and thus properly classified under heading 3004. Equally, if the “process” condition in the CNEN relates to the “active substance”, for the reasons we give below, the process of winnowing the ispaghula husk from the plantago seed satisfies this condition and, on this alternative application of the CNEN, the ispaghula husk is a “plant extract” within the CNEN and MSO is a “medicament” within heading 3004. It is for this reason that we say that it does not matter whether we are right or wrong about whether the “process” condition applies to the “herbal medicinal preparation” or to the “active substance” in the CNEN.

P & G submit that ispaghula husk was indeed a “plant extract”. P & G submit that the term “plant extract” is a term for the Tribunal to construe.

To summarise P & G’s submissions: ispaghula husk is the seed coat of the plantago plant; ispaghula husk is separated from the plantago seed by mechanical means; P & G rely on the reference to the “winnowing machine” made by Dr Dowrick (we have already observed that Dr Otten also referred to a “winnowing machine”); P & G also rely on Dr Otten’s description of the process by which MSO is produced (milled ispaghula husk blended with excipients); P & G referred to the Oxford English Dictionary meaning of the verb “extract” meaning “to draw out of any containing body or cavity”; “to obtain (constituent elements, juices etc) from a thing or substance by suction, pressure, distillation or any chemical or mechanical operation”; P & G also referred to the noun “extract” from the Oxford English Dictionary as meaning “something drawn or taken out of a thing” and also “the substance extracted; the chief parts drawn from anything; in modern use a pharmaceutical term applied to the tough or viscid matter obtained by treating any substance with solvents and then evaporating the solvent … Also loosely used for any preparation containing the active principle of a substance in a concentrated form”. P & G also sought support from various descriptions of ispaghula husk outside customs law, specifically: “ispaghula husk is used without further processing or as preparations obtained by powdering the seed coats… Herbal medicines containing ispaghula husk are usually available in solid form is to be taken by mouth” (“Herbal medicines: Summary for the public”, produced by the European Medicines Agency); a description by the European Medicines Agency’s Community Herbal Monograph of ispaghula husk as a “herbal substance…episperm and collapsed adjacent layers removed from seeds” and a “herbal preparation… Powdered herbal substance”; and the European Pharmacopoeia’s definition of ispaghula husk as consisting of the “episperm and collapsed adjacent layers removed from the seeds of [the plantago seed]”. P & G submit that ispaghula husk “derives from the outer coat of the plantago seeds”.

We take P & G’s submissions to amount to this: ispaghula husk is separated from the plantago seed by a process which can be described as a “mechanical” process; the plantago seed is a “plant” and thus the separation of the ispaghula husk from the plantago seed means that the ispaghula husk has been “extracted” from a “plant” (the plantago seed) or, alternatively, may be described as an “extract from” that same plant. The ispaghula husk has been “drawn out of”, or “obtained from” the plantago seed by a mechanical operation. Thus the ispaghula husk is a “plant extract”. This conclusion is supported, according to P & G, by the definitions of ispaghula husk outside customs law, on which P & G rely. It is, we think, implicit in P & G’s submissions that the “process” condition in the CNEN relates to the “active substance”, rather than to the “herbal medicinal preparation”.

HMRC say that the ispaghula husk is not a “plant extract”. HMRC objected to the proposition advanced by P & G that a “plant extract” is anything separated from a plant by means of a mechanical process. HMRC object to this on the grounds that P & G had restricted its application to argue the “plant extract” issue (which P & G had to do because P & G had not originally advanced the “plant extract” issue at all), to the proposition that a “plant extract” in the CNEN includes material that has been “extracted by milling”. HMRC say that P & G should not be entitled to advance any submissions which go beyond the proposition that a “plant extract” is a substance extracted by “milling” (and not by any other process, such as winnowing). And HMRC go on to say that the separation of the ispaghula husk from the plantago seed is done by means of winnowing; the milling process was subsequent that winnowing, so that the “extraction” of the ispaghula husk from the plantago seed cannot be said to be done by “milling”. And HMRC say that the process of “winnowing” (which HMRC described as the exposure of material to an air current so that unwanted lighter particles are blown away) is not a process of “extraction”, so that the ispaghula husk cannot be described as an “extract”. HMRC also point out that, as we have already observed above, Dr Dowrick suggests that ispaghula husk is not a plant extract because a “plant extract is a chemical component that is extracted by means of a solvent” (at [19] of Dr Dowrick’s second witness statement). It is also implicit in HMRC’s submissions that the “process” condition in the CNEN relates to the “active substance”, rather than to the “herbal medicinal preparation”.

We find that ispaghula husk is indeed a “plant extract” for the purposes of the CNEN. We allow P & G’s appeal on this basis.

Put short, ispaghula husk is the outer coat of the plantago seed which is “obtained” from the plantago seed by means of a winnowing machine. It is perfectly intelligible to describe (and we find as a fact that) ispaghula husk as an “extract” from the plantago seed. If we are correct that the “process” condition in the CNEN relates to the “herbal medicinal preparation”, rather than to the “active substance”, P & G’s appeal succeeds. But even if we are wrong as to the “process” condition in the CNEN and this “process” condition relates to the “active substance”, the ispaghula husk is “produced” by the separation of the ispaghula husk from the plantago seed by the winnowing machines referred to by both Dr Dowrick and Dr Otten, which is a process of “extraction” (by winnowing). To the extent that it is any part of HMRC’s case that “extraction” must be made by means of a solvent, we reject this. There is nothing to this effect in the text of the CNEN and no reason in principle why they should be implied into the text of the CNEN. We have already observed at paragraph [5] above that we respectfully disagree with the conclusion of the Customs Code Committee, reached in its meeting of 15^th and 16 October 2018, for the reasons given there.

Incidentally, HMRC are correct that P & G had restricted their pleadings, in amending P & G’s grounds of appeal, to include the “plant extract” issue, to a submission that the ispaghula husk was a “plant extract” because the ispaghula husk was extracted by means of “milling”. We do not understand this submission made by P & G, since the ispaghula husk was separated from the plantago seed by means of winnowing, prior to milling. Neither was this submission made in P & G’s further submissions on the “plant extract” issue. HMRC are correct in their complaint that P & G’s submissions on the “plant extract” issue went beyond P & G’s amended grounds of appeal. And it was procedurally incorrect and discourteous to the Tribunal that P & G made submissions on the “plant extract” issue on the basis that the relevant “process” was “winnowing”, not “milling”, without any application for permission to do so.

However, we consider it proper to decide the plant extract issue in favour of P & G in the manner that we have set out above. First, as we have observed, several times, we do not consider that the “process” condition in the CNEN relates to the “active substance”, so that whether the relevant “process” is winnowing or milling is irrelevant (HMRC have not taken any issue with the proposition that MSO is a herbal medicinal preparation). Second, even if we are wrong, and the “process” condition in the CNEN relates to the “active substance”, HMRC themselves, in HMRC’s further submissions on the plant extract issue, addressed expressly the question of whether winnowing was a “process” within the CNEN (on the assumption, with which we disagree, that the “process” condition relates to the “active substance”, rather than to the “herbal medicinal preparation”) and no question of prejudice to HMRC arises in allowing the appeal, on the basis that the ispaghula husk is a “plant extract” that is “extracted” from the plantago seed by means of winnowing. Since, if it is relevant, the only “process” relevant to the “plant extract” issue is the process of “winnowing” (and it is not disputed that this process is by means of a “machine”) and neither party sought to have either Dr Otten or Dr Dowrick recalled to give further evidence on the “plant extract” issue, had P & G applied to amend P & G’s grounds of appeal to include the ground that the relevant “process” was that of “winnowing”, rather than “milling”, we would have granted that application. And third, this Tribunal has the jurisdiction to decide an appeal on a point not made by either party (subject, of course, to the overriding objective and potential prejudice to one or other of the parties: there is no prejudice to HMRC in this case, as we have observed).

Our conclusion on the “plant extract” issue is sufficient to dispose of this appeal. However, P & G had made submissions (indeed it was P & G’s primary case) that the ispaghula husk was a “chemically defined substance” for the purposes of the CNEN. For the reasons we give below, we reject this submission and we consider it proper to record our decision, in case we are wrong on the “plant extract” issue.

P & G submitted that the ispaghula husk was a “chemically defined substance” for the purposes of the CNEN. P & G submit that the term “active substance”, in the CNEN is not limited to active components whose action can be described solely in terms of the chemicals within the active component. This is because, according to P & G, although the action of the active component within a herbal medicinal preparation must have a curative or prophylactic action, that action need not be solely chemical. P & G also relied on a definition of “active substance” in Directive 2001/83/EEC (the “Medicines Directive”), which P & G submits supports P & G’s construction of the term “chemically defined substance” for the purposes of the CNEN. P & G says that it would be impractical, confusing and against the interests of public health for a duty provision to require a different kind of patient leaflet or label from the Medicines Directive. P & G says it is likely that the draftsperson of the Additional Note intended to use EU medicines law on patient information, and it is very unlikely that the draftsperson intended to create patient information requirements for customs purposes that were different from EU medicines law. P & G also made reference to a number of customs classifications, in other jurisdictions, which classified ispaghula husk as a “medicament”, and not as a “food preparation”.

HMRC submitted that a “chemically defined substance” is a form of matter having constant chemical composition and characteristic properties, which cannot be separated into components by physical means, that is, without breaking chemical bonds. It followed, according to Mr Watkinson, that ispaghula husk, which comprised 85% arabinxylans and 15% “secondary metabolites”, was not a “chemically defined substance”, since ispaghula husk, quite simply, contained substances which were not arabinoxylans, that is, other substances that had their own chemistry. HMRC also relied on the decision of the Customs Code Committee, in the meeting of 15^th and 16 October 2018, that ispaghula husk was properly classified as a “food preparation”, and not as a “medicament”.

We consider HMRC to be correct. The substance which comprises ispaghula husk is not, on any view, a synonym for the arabinoxylans which have therapeutic and prophylactic properties. Ispaghula husk cannot be a synonym for these arabinoxylans, since ispaghula husk comprises other substances (as to 15%) in addition to these arabinoxylans. Neither is ispaghula husk a term which connotes the chemistry of the substances which comprise it. Ispaghula husk is not defined by reference to its chemistry but rather its physical nature (it is the very “husk” of the plantago seed). Mr White’s submission that because the therapeutic or prophylactic properties of the ispaghula husk contained in the MSO need not be “chemical”, the term “chemically defined substance” need not be a reference to the chemical components comprised within an “active substance” is a non sequitur. The CNEN is simply not concerned with the therapeutic or prophylactic properties of any particular substance. Such therapeutic or prophylactic properties are a necessary condition for a substance to be classified as a “medicament” under heading 3004, but in order to fall within the terms of the Additional Note and the CNEN, the requirements as to the labelling and user information must be satisfied. Furthermore, we consider that our conclusion ought not to be displaced by definitions contained in other legislative instruments, such as the Medicines Directive, which performs a different role to that of the Additional Note and the CNEN (and indeed the Tariff Regulation and the CN, contained in Annex I, which, we have already observed, has the dual function of serving the requirements of the Common Customs Tariff and facilitating the gathering of external trade statistics of the EU and other EU policies concerning the importation and exportation of goods). There is no textual cross-reference in the CN, the Additional Note, or the CNEN to any other legislative instrument, and certainly not to the Medicines Directive. The CN is a legislative instrument which imposes customs duty by reference to the nature and function of a very wide variety of items and while a “medicament” must have therapeutic or prophylactic properties, relating to specific diseases or ailments, labelling requirements which are relevant to the imposition of one rate of customs duty, rather than another, are, at the very least, potentially entirely distinct from public health considerations. The CNEN amplifies the construction of the Additional Note, which in turn amplifies the construction of the CN. There was nothing in Mr White’s submissions which compelled us, as a matter of principle, to read across provisions of the Medicines Directive, as to the definition of “active substance”, into the CN in Annex I of the Tariff Regulation. And we consider the classification of ispaghula husk as a “medicament” in other jurisdictions to be irrelevant for the purposes of the current appeal, since (1) certain jurisdictions classified ispaghula husk as a “medicament” but other jurisdictions classified ispaghula husk as a “food preparation”, (2) we were not shown the reasoning behind any particular classification, so any particular classification cannot even be viewed, we consider, as persuasive and (3) none of these other classifications are binding upon us in any event.

Had the ispaghula husk not been properly classified as a “plant extract” (or, alternatively, on P & G’s submissions, a “chemically defined substance”) for the purposes of the CNEN, P & G would have relied on the propositions that the Additional Note was invalid, in so far as the additional note restricted the definition of “medicament”, by imposing labelling requirements which effectively restricted the definition of “medicament” in the heading 3004 and, alternatively, that the CNEN was invalid, insofar as the CNEN imposed a condition as to the requirement to specify a “chemically defined substance” (or a chemically defined group of substances), or a “plant extract” which, equally, illegitimately restricted the scope of the term “medicament” for the purposes of heading 3004. P & G’s submission effectively amounts to the proposition that the therapeutic and prophylactic properties of ispaghula husk that have a curative effect on specific diseases, ailments etc, is sufficient for ispaghula husk to be properly classified as a “medicament” and not as a “food preparation”.

Given that we have found ispaghula husk to be a “plant extract”, for the purposes of the CNEN, we do not consider it fruitful to conclude on either of these two contingent issues. We have already observed that the term “includes”, as a matter of language, extends a definition, so that it is highly unlikely that a failure to satisfy the conditions in the Additional Note would exclude a substance from being properly classified as a “medicament” under heading 3004, if it was properly so classified without reference to the Additional Note. But we have also observed that the Additional Note may not properly extend the definition of "medicament”. It follows that we consider it highly unlikely that a failure to satisfy the requirements of the Additional Note (and the CNEN) would disqualify a substance from being classified as a “medicament”, if that substance is otherwise properly classified as a “medicament” under heading 3004. The same is true of any failure to satisfy the conditions of the CNEN. But these observations go to the proper construction and interpretation of the text and function of the Additional Note and the CNEN (which we consider together confer a presumptive status of “medicament” on a substance which satisfies the requirements, which presumptive status may be removed if the substance in question does not have the required therapeutic or prophylactic properties). On that view, the Additional Note and the CNEN are perfectly valid. However, we offer this very brief summary of our view only to provide the starting point for an analysis of these two issues for an appellate court, should this matter go further. Any such analysis would, we consider, require careful consideration of the legislative history of all of the relevant legislative instruments and, perhaps, consideration and analysis of other language versions of these instruments. None of these were provided to the Tribunal and we therefore restrict our observations to these very brief obiter comments.

As a separate matter to the substantive dispute, P & G made an application for costs, dated 20 November 2018. P & G complained about HMRC’s conduct in relation to a stay application made by HMRC on 26 June 2018 (the “Stay Application”). P & G had objected to the Stay Application on 6 July 2018.

P & G rely on the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 (SI 2009/273), rule 10 (“unreasonable conduct” of one of the parties).

P & G complain that:-

HMRC say:

We dismiss P & G’s costs application. To be clear, HMRC did, indeed, rely on the minutes of the Customs Code Committee in the course of Mr Watkinson’s submissions (albeit unsuccessfully). We say nothing about HMRC’s alleged breach of directions as P & G did not make any further submission on this part of the costs application and we see no connection between this alleged breach and the Stay Application in any event.

At no stage of its costs application did P & G suggest that HMRC behaved improperly or “unreasonably” in seeking to rely on the Customs Code Committee’s conclusion on the proper classification of ispaghula husk. Rather, P & G’s costs application amounts to a complaint about HMRC’s motivation behind the Stay Application, pending the result of the reference of the proper classification of ispaghula husk to the Customs Code Committee and (it does not matter to us, for the reasons we give below, whether this is part of that same submission or a distinct submission) the timing of HMRC’s reference and withdrawal of the Stay Application.

We repeat our summary of the principles that govern the application of rule 10, which we gave in our previous costs decision, reported separately, in this case:-

We have already observed that P & G do not object to HMRC having referred the proper classification of ispaghula husk to the Customs Code Committee. And there was no objection to the admissibility of the minutes of the meeting of the Custom Code Committee of 15^th and 16^th October 2018 by P & G in the substantive hearing. Had there been, we would have rejected any objection to HMRC relying thereon. So we turn to P & G’s complaint about HMRC’s motivation in making the Stay Application, to await the Customs Code Committee’s decision. This is a very serious allegation indeed. It amounts to accusing HMRC (which, to be clear, is an accusation which must be levelled at one or more specific individuals) of misleading the Tribunal as to the motivation behind the Stay Application. Any such allegation must be particularised and P & G have not particularised any specific allegation, against any specific individual at all. The chronology of the correspondence does not come even close to (indeed it says nothing about) particularising any specific allegation against any specific individual as to motivation. We consider P & G wrong to have made such an allegation as to the motivation of HMRC in making the Stay Application (the language of P & G’s costs application, at [10] of P & G’s costs application: “[P & G] invites the tribunal to consider [HMRC’s] motives in a stay application…”). It is highly inappropriate make an inferred allegation as to motivation without making it clear as to which specific courses of conduct amounted to an effective misleading of the Tribunal. In any event, since there are no particularised allegations made at specified individuals, in relation to specific conduct, we simply record that we dismiss P & G’s costs application on the basis of this alleged “motivation” (for which we find no evidence) of disrupting the substantive proceedings, on the part of HMRC.

Insofar as the timing of HMRC’s making a reference to the Customs Code Committee (and thus the timing of HMRC’s Stay Application) and the timing of HMRC’s withdrawal of HMRC’s Stay Application, on no view does either suggest that HMRC had a motivation to disrupt the substantive proceedings or have any other motivation, other than to find support (or indeed the opposite) for HMRC’s position in the substantive dispute. HMRC sought support for HMRC’s substantive position, and HMRC ultimately did, in fact, as we have observed, several times, rely on the relevant minutes of the meeting of 15^th and 16^th October 2018. To the extent that the timing of making a reference to the Customs Code Committee is concerned, it is wholly unsurprising that HMRC wish to make such a reference when there is live litigation. As to the withdrawal of HMRC’s Stay Application, the Customs Code Committee expressly “closed the file” after the meeting of 15^th and 16^th October 2018 and so we would have found it baffling if HMRC had pursued HMRC’s Stay Application after the minutes of that meeting had been published. The view expressed in that meeting was not a “preliminary view” with a further “final” view to follow, at least on the terms of the minutes of that meeting, so that HMRC could not be (remotely) accused of having somehow overplayed the importance of any view obtained from the Customs Code Committee, in making the Stay Application in the first place.

The same reasons apply if P & G rely on the timing of the making of the reference of the proper classification of ispaghula husk to the Customs Code Committee, (and thus the making of the Stay Application) as a separate ground of complaint. We have already said that we find it wholly unsurprising (and only proper) to wait until there is ongoing, live litigation which makes such a reference relevant. There is no improper motivation which can be ascribed to such a reference by reason of the timing of that reference.

We refuse P & G’s costs application for these reasons.

We allow P & G’s appeal on the classification of ispaghula husk.

We refuse P & G’s costs application.

This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.