[2013] UKFTT 573 (TC)

 TC02962

 

 

 

Appeal number: TC/2012/08422

 

 

BEER DUTY – whether allowed method for calculating alcoholic strength of beer requires repayment of duty where actual strength is lower than declared strength – no – whether failure of regulations to make provision for calculation of duty on that basis unfair and an abuse of power – no – appeal dismissed

 

 

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

	CARLSBERG UK LIMITED	Appellant
	- and -
	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

 

 

TRIBUNAL:	JUDGE  TIMOTHY HERRINGTON MS LESLEY STALKER

 

 

Sitting in public at 45 Bedford Square, London WC1 on 10 and 11 July 2013

 

 

Jeremy White, Counsel, for the Appellant

 

Andrew MacNab, Counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents

 

 

 

 

 

© CROWN COPYRIGHT 2013

DECISION

 

 

Introduction

1.             Carlsberg UK Limited (“Carlsberg”), the well known brewer, appeals under section 16(1) of the Finance Act 1994 (“FA 1994”) against the decision of the Respondents (“HMRC”) contained in a letter dated 1 August 2013 upholding on review a decision of HMRC to refuse to repay beer duty of £176,249.05 which Carlsberg contends was overpaid in respect of its July 2011 accounting period. Carlsberg had claimed repayment of the duty pursuant to section 137A of the Customs and Excise Management Act 1979 (“CEMA”).

2.             Carlsberg expresses the point at issue in this appeal as being:

“For the purposes of beer duty what is the Allowed Method of calculating the alcohol by volume (ABV) of beer?”

The reference to the “Allowed Method” is the method allowed by HMRC pursuant to Regulation 18 of the Beer Regulations 1993, discussed in more detail below.

3.             Carlsberg contends that the answer to the question set out in paragraph 2 above is the actual ABV of the beer (“Actual ABV”) subject to an exception, that exception being that the strength declared by the brewer on the delivery note accompanying the beer (“Declared ABV”) may be used instead of Actual ABV for no more than one consecutive month even if Actual ABV is greater.

4.             HMRC contend that the answer to the question is the Declared ABV of the beer, subject to an exception, that exception being that HMRC may substitute the Declared ABV with the Actual ABV if the latter is consistently greater than the Declared ABV and assess duty by reference to (greater) Actual ABV.

5.             Carlsberg puts forward two grounds on which it founds it appeal.  The first, which is its principal ground, is that the correct construction of the provisions relating to the Allowed Method is that described in paragraph 3 above.  The second, which is its alternative ground, is that the Allowed Method should be that for which it contends, as set out in paragraph 3 above. Carlsberg contends that in exercising its discretionary powers to prescribe the Allowed Method, HMRC must act rationally and proportionately and in a manner which results in legal certainty, no abuse of power and which respects the legitimate expectations of taxpayers.  Additionally, the exercise of the discretionary power must respect the principles of effectiveness, equivalence of equality of treatment non discrimination and fairness.  It contends that its interpretation of the Allowed Method meets these principles whereas HMRC’s does not.  Carlsberg accepts that its alternative ground brings into question the applicability of public law principles but nevertheless contends that the Tribunal has jurisdiction to hear arguments on the alternative ground.

6.             We consider below the relevant legislation and guidance in relation to the calculation of beer duty.  We then make findings of fact based on the evidence before us. We then set out our decision on how beer duty is to be calculated in this case in the light of those findings, the relevant legal principles and the submissions of the parties.

Relevant legislation and guidance

7.             The method of calculating beer duty is harmonised at EU level through the provisions of Council Directive 92/83/EEC. Article 3 of this Directive so far as relevant provides:

“1.  The excise duty levied by Member State on beer shall be fixed by reference either:

-          to the number of hectolitre/degrees  Plato, or

-          to the number of hectolitre/degrees of actual alcoholic strength by volume of finished product.

2. In assessing the charge to duty on beer in accordance with the requirements of Directive 92/84/EEC, Member States may ignore fractions of a degree Plato or degree of actual alcoholic strength by volume.”

8.             Rates of duty are not harmonised except that Member State must apply minimum rates of excise duty in accordance with the rules laid down in Council Directive 92/84/EEC.  Article 6 of this Directive provides:

“As from 1 January 1993, the minimum rate of excise duty on beer shall be fixed:

-          ECU 0,748 per hectolitre/degree Plato, or

-          ECU 1,87 per hectolitre/degree or alcohol of finished product.”

9.             A further Directive, Council Directive 2008/118 EC, sets out further provisions concerning the legal arrangements for Excise Duty. Article 9 of this Directive provides:

“The chargeability conditions and rate of excise duty to be applied shall be those in force on the date on which duty becomes chargeable in the Member State where release for consumption takes place.

Excise duty shall be levied and collected and, where appropriate, reimbursed or remitted according to the procedure laid down by each Member State.  Member State shall apply the same procedures to national goods and to those from other Member States”.

10.         The requirements of these Directives have been implemented in UK legislation through the Alcoholic Liquor Duties Act 1969 (“ALDA”).  Section 1(1) (b) ALDA brings beer within the scope of excise duty. Section 2(2) ALDA makes provision for the calculation of the strength of alcoholic liquor as follows:

“(2) For all purposes of this Act –

(a)          except where some other measure of quantity is specified, any computation of the quantity of any liquor or of the alcohol contained in any liquor shall be made in terms of the volume of the liquor or alcohol, as the case may be;

(b)          any computation of the volume of any liquor or of the alcohol contained in any liquor shall be made in litres as at 20°C; and

(c)          the alcoholic strength of any liquor is the ratio of the volume of the alcohol contained in the liquor to the volume of the liquor inclusive of the alcohol contained in it).”

Section 2(3) ALDA gives HMRC power to make regulations for the purpose of ascertaining the alcoholic strength of liquor as follows:

“(3) The Commissioners may make regulations prescribing the means to be used for ascertaining for any purpose the strength, weight or volume of any liquor, and any such regulations may provide that in computing for any purpose the strength of any liquor any substance contained therein which is not alcohol or distilled water may be treated as if it were.

(3A)  Without prejudice to the generality of subsection (3) above, regulations under that subsection may provide that for the purpose of charging duty on any … beer … contained in any bottle or other container, the strength, weight or volume of the liquor in that bottle or other container may be ascertained by reference to any information given on the bottle or other container by mean of a label or otherwise or any other documents relating to the bottle or other container.

(4)         Different regulations may be made under subsection (3) above for different purposes.

(5)         Nothing in this section shall prevent the strength, weight or volume of beer, wine, made-wine or cider from being computed for the purpose of charging duty thereon by methods other than that provided in this section.”

11.         Section 36(1) ALDA, as in force at the time of the occurrence of the events that are the subject of this appeal, imposes the “charge of excise duty” on beer (a) imported into, or (b) produced in, the UK.  Section 36(7AA)(a) states that the rate at which excise duty is to be charged on beer is:

“£18.57 per hectolitre percent of alcohol in the beer”.

Thus, for example, the duty due on a volume of 1000 litres of beer (10 hectolitres) with a strength of 4% ABV at the duty rate of £18.57 would be

10 x 4 x £18.57 = £742.80

12.         Section 36(2) ALDA provides that beer duty shall be determined and become due in accordance with regulations made under section 49 ALDA. Section 49(1) ALDA gives power to make regulations for various purposes including:

“(e) for determining the duty and rate thereof and, in that connection, prescribing the method of charging duty;

(f) for charging the duty, in such circumstances as may be prescribed in the regulations, by reference to a strength which the beer might reasonably be expected to have, or the rate of duty in force, at a time other than that at which the beer becomes chargeable.”

Paragraph (f) above is designed, so we were told, to cover cask conditioned beer, that is beer that can gain strength in the cask, because in general terms under the regulations made under section 49 ALDA the duty point for beer is the time when it is produced (which will usually be the time it leaves the brewery: see Regulation 8(2) of the regulations referred to in paragraph 13 below.

13.         The relevant regulations made under sections 2(3) and 49(1) ALDA are the Beer Regulations 1993. Regulation 18 of these Regulations (“Regulation 18”) is concerned with the strength of beer and provides as follows:

“18. Save as the Commissioners otherwise allow, the strength of the beer shall be deemed to be the greater of –

(a)          the strength determined by the method described in Schedule 4 to these Regulations;

(b)         the strength ascertained by reference to information on the label of the container of the beer;

(c)          the strength ascertained by reference to information on any invoice, delivery note or similar document issued in relation to the beer; and

(d)         the strength which any cask or bottle conditioned beer or any other unfinished beer is reasonably expected to have been sold by way of retail or otherwise supplied for consumption, save as HMRC otherwise allow, the strength of any beer is to be assessed in accordance with measurement criteria which reflect section 2 of ALDA, or, if higher, as stated on the container or in any invoice or similar document.”

Schedule 4 sets out a detailed method for testing the alcoholic strength of beer (ABV).  In accordance with the authority given by Article 3(2) of Council Directive 92/83/EEC (see paragraph 7 above). ABV is expressed in one decimal place and rounded down in all cases (e.g. 4.19% ABV is rounded down to 4.1%).

14.         Regulation 18 recognises the fact that the strength of beer can vary from brew to brew because brewing is not an exact science.  It also recognises that brewers advertise their beer as having a specific ABV which is to be found on containers of the beer when it is sold and thus operates as a provision deeming the strength to be the greater of the actual strength or what is stated on the invoice, delivery note or similar document or containers relating to the beer.  This is particularly important for mass producers such as Carlsberg, which will set a target ABV for their various brands of beer which will appear on all the containers in which its beer is sold. Smaller brewers are, we are told, content to pay duty by reference to the actual measured strength.

15.         It can be seen that accounting for beer duty on the Actual ABV of each batch brewed is an onerous task. The strength of the beer can be expected to vary from batch to batch making it difficult to state the actual ABV on the delivery note, invoice or container. Recognising therefore that brewers have a specific target ABV for each brand, pursuant to the power in Regulation 18 HMRC have allowed an alternative method, referred to in this decision as the “Allowed Method”.

16.         The Allowed Method seeks to permit duty to be calculated by reference to the Declared ABV provided certain conditions are met, even though the Declared ABV may be less than the Actual ABV.  The primary issue in dispute in this appeal is whether the Allowed Method allows the Actual ABV to be used if the Declared ABV is greater than the Actual ABV.

17.         The Allowed Method is set out in paragraph 12 of HMRC Notice 226 Beer Duty in the following terms:

“12 Alcoholic strength

12.1     What is alcoholic strength?

For duty purposes alcoholic strength is the percentage of alcohol by volume (ABV) in the beer. This should be expressed in one decimal place, for example, 4.19% ABV becomes 4.1% ABV.  Ignore figures after the first decimal place.

12.2     How can I measure alcoholic strength?

You may use any method you wish to measure the strength of beer as long as it produces results that agree with those that would be achieved using the reference method described in section 29.

If you do not have your own facilities for determining ABV by analysis and you do not add priming sugar, you may use the method (based upon the degree of attenuation in the beer) which is reproduced at section 30.

If you do add priming’s to your beers you may use the method which is reproduced at section 31.

Additionally, for these or any other methods not based on laboratory analysis, an independent analyst must test the ABV of each of your products, at least annually, to confirm consistency with calculated results. The results of the independent analyses must be held in your business records.

12.3     What alcoholic strength is used for duty purposes?

For duty purposes, the alcoholic strength of brewery-conditioned beer is the greater of the actual strength or the label/invoice/delivery note strength of beer when it passes the duty point.

For duty purposes, the alcoholic strength of cask or bottle conditioned beer or any other unfinished beer is the greater of the strength which the beer is reasonably expected to have when sold by way of retail or otherwise supplied for consumption or the label/invoice/delivery note strength of the beer.

However, if you comply with certain conditions, we will accept for duty purposes the strength stated on the package label, invoice, delivery note or similar document (this is known as the declared strength).

12.4     What are the conditions for using the declared strength?

If you wish to use the declared strength for duty purposes, you must be able to demonstrate that you have exercised due diligence in the control of your process to make sure that, on average, the actual ABV of each finished product equates to that which you are declaring on the label etc.

12.5     How can I demonstrate that due diligence has been exercised in the control of ABV?

You must continuously monitor and record your ABV results, which should normally fall randomly on either side of the target strength. The average of your results should equate closely with the target which must be the declared strength.

It is recognised that ABV may occasionally vary, but provided appropriate action is taken quickly to return the strength of the beer to within its normal specification, due diligence will have been demonstrated.  You must keep records of action taken to maintain product strength within control limits.

12.6     Do I have to measure the strength of each product?

You must establish the strength of each discrete batch for each of your products.  Where beer from one batch is packaged into different container types, for example, cans and bottles, you may combine the results.

12.7     What about infrequent or one-off brews?

If you can demonstrate to us that based on available information and experience, due care was taken when deciding target ABVs for new and/or infrequently brewed products (and that all decision, actions, etc. were properly recorded), we will accept the label/invoice/delivery note strength for duty purposes

12.8     How will the accuracy of this system be checked by HM Revenue & Customs?

Our officer will examine your results and your record of actions taken. Where the results have consistently fallen above your target, he will wish to confirm that action was taken as soon as the problem was identified to bring the process back into control or to change the declared ABV.  If you have failed to take such action, an assessment will be raised for the additional duty due.

12.9     Do I use the same arrangements for measuring strength of beer which may be delivered under duty suspension?

Yes, because it is likely that the beer will subsequently be delivered on payment of duty, the same arrangements must apply for all beer you produce.

It is the responsibility of the person holding the beer at the duty point to account for the duty. Therefore, if you despatch packages in duty suspension on which the strength is understated on the label/invoice/delivery note, you must inform the consignee accordingly.

12.10  What happens if there is a dispute over the strength?

Our officer may take samples of beer which will be analysed using the reference method described in section 29. The analysis result will establish the actual dutiable strength of the beer for legal purposes.

12.11  Cask and bottle conditioned beer

12.11.1 What alcoholic strength is used for duty purposes on cask and bottle conditioned beer?

Cask and bottle conditioned beers will continue fermenting after removal from registered premises. This will result in an increase in strength.  You must account for duty on the strength at which you expect the beer to be when it is consumed.  This is also the strength which must be shown on the label/invoice/delivery note.

12.11.2 Are there any additional due diligence requirements in relation to cask and bottled conditioned beer?

Yes.  In addition to the procedures laid out in paragraph 12.5 which establishes the ABV at packaging, you must regularly monitor and record the actual strength of each quality of cask and bottled conditioned beer at the expected time of consumption, to establish its alcoholic strength.  The precise method and frequency of checking is a matter for you but you must be able to satisfy us of the accuracy of your results.”

18.         We observe that the emphasis in the Notice is on HMRC accepting for duty purposes the strength stated on the package label, invoice delivery note or similar document; that is the declared strength, rather than the greater of the actual or declared strength as prescribed by Regulation 18: see the last sentence of paragraph 12.3 of the Notice.  It is also clear that the declared strength can only be used if due diligence is exercised in the brewing process to ensure that, on average, the actual ABV equates to the declared strength:  see paragraph 12.4 of the Notice.  This is clearly designed as a revenue protection measure so that close control of the brewing process is maintained to ensure that as far as possible the average Actual ABV correlates to the Declared ABV.  Paragraph 12.8 makes it clear that where that fails to be the case, an assessment will be raised based on the Actual ABV.

19.         We also observe that under the Allowed Method the question as to whether the Actual ABV is at variance with the Declared ABV is determined by reference to the average Actual ABV for the period in question, not the actual ABV of each batch: see paragraph 12.4 of the Notice.

20.         The Allowed Method as set out in paragraph 12 of Notice 226 is thin on detail.  The British Beer and Pub Association (BBPA) have issued interpretative guidelines which of course have no statutory effect but are expressed to have been compiled with help from and consultation with HMRC and to complement Notice 226.  These guidelines again emphasise that the Allowed Method is in effect an average ABV system.  Paragraph 2.1 of these guidelines state:

“It is recognised that under an average ABV system, any individual pack may have an ABV above or below the declared strength but that duty is based on the declared strength”.

We take the policy to be, as recognised by the BBPA in this statement, that the system is designed so that on average during the relevant period against which the average is measured, the average Actual ABV will equate to the Declared ABV so that, on average, duty will be paid by reference to the Actual ABV, as envisaged by Article 1 of Council Directive 92/83 EEC. The BBPA guidelines recognise that the due diligence process required of brewers who wish to use the Allowed Method is designed to ensure that this will be the case:  see the last sentence of paragraph 2.2 of the BBPA guidelines which states:

“If the brewer/packer can demonstrate that he has been diligent in the control of ABV and that corrective action has been applied where necessary, then HMRC should be satisfied that the ABV as declared is correct and that the right amount of duty is being accounted for.”

21.         The fact that the process is designed to ensure that the Actual ABV and the Declared ABV should not deviate significantly in practice is emphasised by paragraph 3.2 of the BBPA guidelines which state that the target ABV of the final product  must not exceed the Declared ABV.

22.         Section 4 of the BBPA Guidelines contains considerable detail as to the control methods expected in order to exercise appropriate control of the production process, so as to ensure that the average ABV coincides with the target value.

23.         Paragraph 4.1 of the BBPA guidelines recognises that variations may occur and they may be either side of the target.  It states

“In a process that is under control, there will be a variation around the target ABV.  Such variation will be normally distributed, that is to say, the variation from the target will be equally above and below the target ABV.”

The guidelines illustrate this by reference to a table showing a series of average ABVs for each month in a twelve month period, some a little below the Declared ABV and some above which leads to a conclusion that there would be no question of a surcharge in respect of the two months where the Actual ABV exceeded the target (which were not consecutive).  Neither did the guidelines refer to the possibility that a repayment may be due in respect of the six months where the Declared ABV was greater than the Actual ABV.

24.         In paragraph 4.2.3 of the BBPA Guidelines an example is given of  where the Actual ABV had exceeded the Declared ABV for two consecutive months with a comment that in this situation the brewer would need to demonstrate to HMRC that he had taken action to remedy the situation otherwise he may be at risk of a retrospective surcharge.  Paragraph 4.2.7 of the BBPA Guidelines comment that if the results are consistently and significantly higher than the Declared ABV then the brewer must either revise the Declared ABV and pay duty at a higher rate or take action to reduce the ABV back to that specified as soon as practicable by altering appropriate process parameters. 

25.         Paragraphs 4.2.14 and 4.2.19 of the BBPA give further guidance as to the consequences of Actual ABV being greater than that declared as follows:

“4.2.14 Will I be liable for excess duty if the actual ABV comes out higher than that declared?

Such a situation might arise, particularly where labels are prepared in advance of a particular production run.  In such cases, where the brewer can demonstrate that he has taken all reasonable care to ensure that the alcohol content meets the required strength, there should be no additional liability for duty.

4.2.19 Will assessments for additional duty be levied by HMRC?

The onus is on the brewer to ensure that problems are identified and action taken within a reasonable time. Provided his records demonstrate to HMRC that he has acted with “due diligence” assessments will not be levied.  It will be up to the brewer to demonstrate that he has behaved reasonably in the particular circumstance involved.

Examples of the type of evidence HMERC would expect to see include:

·         Records of the monitoring of actual ABV performance and investigations and corrective actions into adverse trends

·         Correspondence between departments within the brewery discussing a problem and resolutions to address it

·         Correspondence between a remote packer and the brewer regarding ABVs exceeding specification and actions required to address the issue

·         Records of discussions and correspondence between the brewer and HMRC on the matter

When a brewer’s records show that the strength of a product is greater than the target ABV and he fails to take action to correct the problem within a reasonable time an assessment will be issued by HMRC.”

26.         Carlsberg’s claim for repayment is made under section 137A of CEMA which is headed ‘Recovery of overpaid excise duty” and provides that:

“(1)  Where a person pays to the Commissioners an amount by way of excise duty which is not due to them, the Commissioners are liable to repay that amount …”.

27.         Section 13A (2)(d) FA 1994 provides that any decision by HMRC on a claim under section 137A CEMA is a “relevant decision” which means that it may be the subject of an appeal to the Tribunal pursuant to section 16 FA 1994.

28.         Sections 16(4) and (5) FA set out the powers of the Tribunal in respect of such an appeal as follows:

“(4) In relation to any decision as to an ancillary matter, or any decision on the review of such a decision, the powers of an appeal tribunal on an appeal under this section shall be confined to a power, where the tribunal are satisfied that the Commissioners or other person making that decision could not reasonably have arrived at it, to do one or more of the following, that is to say –

(a)          to direct that the decision, so far as it remains in force, is to cease to have effect from such time as the tribunal may direct;

(b)          to require the Commissioners to conduct, in accordance with the directions of the tribunal, [a review or further review as appropriate] of the original decision; and

(c)          in the case of a decision which has already been acted on or taken effect, to declare the decision to have been unreasonable and to give directions to the Commissioners as to the steps to be taken for securing that repetitions of the unreasonableness do not occur when comparable circumstances arise in future.

(5) In relation to other decisions, the powers of an appeal tribunal on an appeal under this section shall also include power to quash or vary any decision and power to substitute their own decision for any decision quashed on appeal.”

29.         It is common ground that a decision made by HMRC under section 137A CEMA not “a decision as to an ancillary matter” within section 16(4) and thus the provisions of section 16(5) apply so that we have power to quash or vary HMRC’s decision and power to substitute our own decision for any decision  quashed on appeal.  It is common ground that the determination of Carlsberg’s primary ground falls within the scope of this power.

30.         Mr McNab submits that the Tribunal can only exercise the power in section 16(5) where, pursuant to section 16(4), the Tribunal is satisfied that HMRC could not reasonably have arrived at the decision in question.  In other words, he submits that the provisions of section 16(4) and 16(5) are cumulative.  He submits that the exercise of the right to appeal under section 16 is the exercise of a public law right and therefore if the Tribunal is to allow the appeal public law considerations apply so that it needs to be satisfied that HMRC’s decision was one that it could not reasonably have arrived at even when exercising the power to substitute its own decision under section 16(5).

31.         Mr McNab relies on paragraph 91 of the Upper Tribunal’s decision in HMRC v Europlus Trading Ltd [2013] UKUT 108 (TCC) where Vos J said:

“… The FTT in this case was given a confined jurisdiction by s16 (4) of the Finance Act 1994 ‘where the tribunal are satisfied that [HMRC] could not reasonably have arrived at [their decision on review]’ broadly to (a) direct that the decision is to cease to have effect; (b) require HMRC a further review of the original decision; or (c) to declare the decision to have been unreasonable.  Section 16(5) allows the tribunal to quash or vary any decision and to substitute its own decision.”

32.         We reject Mr McNab’s submission.  In our view, as submitted by Mr White, Vos J was not interpreting his powers; he was merely repeating the effect of the statutory provisions which he had set out in full earlier in paragraphs 32 and 35 of his decision.  If Mr McNab were correct then where the decision, as in this case, did not relate to an ancillary matter we would have no power to substitute our own decision unless we considered HMRC’s decision was unreasonable, even if the decision was wrong in law. That cannot be right.  In our view the wording of section 16(5) is absolutely clear in establishing a stand alone jurisdiction which operates independently from section 16(4).  It starts with the words “In relation to other decisions” and the wording then makes it clear that in relation to these decisions, that is decisions that are not ancillary decisions, the Tribunal has power to make its own decision on the basis of the evidence before it and the relevant principles of law.

33.         We therefore approach our decision-making on that basis. If we were to accept Mr White’s submissions that we have jurisdiction to consider the alternative ground, it is common ground that such jurisdiction would be supervisory in nature, that is our powers would be confined to determining whether HMRC could not reasonably have arrived at the decision to refuse Carlsberg’s claim.

34.         Section 16(6) FA 1994 makes provision for the burden of proof in an appeal under section 16 in the following terms:

“(6) On an appeal under this section the burden of proof as to –

(a) the matters mentioned in subsection (1) (a) and (b) of section 8 above,

(b) the question whether any person has acted knowingly in using any substance of liquor in contravention of section 114(2) of the Management Act, and

(c) the question whether any person had such knowledge or reasonable cause for relief as is required for liability to a penalty to arise under section 22(1) or 23(1) of the Hydrocarbon Oil Duties Act 1979 (use of fuel substitute or road fuel gas on which duty not paid), shall lie upon the Commissioners; but it shall otherwise be for the appellant to show that the grounds on which any such appeal is brought have been established.”

Thus it is clear that in this case, the appeal not involving any of the matters referred in paragraphs (a), (b) or (c) above, that the burden on proof lies on Carlsberg to establish that it has a valid claim under section 137A CEMA.

35.         Against that legislative background we now turn to our findings of fact in this case.

Findings of Fact

36.         We heard no oral evidence.  We were provided with copies of correspondence between HMRC and Carlsberg relating to its claim.  In addition, we had a witness statement from Mr Neil Burton, who is employed by Carlsberg as Head of Internal Audit and Excise Taxation.  Mr Burton was unable to attend the hearing due to illness so he was not able to be cross-examined on his statement.  Mr McNab rightly submitted that a considerable part of Mr Burton’s statement consisted of Mr Burton’s views on the interpretation of Notice 226 which is a matter of law for the Tribunal and is therefore irrelevant.  Nevertheless, there are a number of statements of fact which are relevant to the issues under appeal and Mr McNab did not seek to challenge those statements.  We are therefore content to accept Mr Burton’s evidence on these factual issues.  Mr Burton has twenty seven years of experience of beer duty and the beer trade, having worked for Carlsberg since 1 February 2001 and before that worked for HMRC in various roles, including the production of training and guidance for HMRC staff engaged in brewery auditing and in the drafting of public notices and Regulations.  He can therefore speak with some authority on the practices in the trade but he was not put forward as an expert witness.  We therefore accept his evidence as a witness of fact, informed by his experience in the industry.

37.         From the evidence we saw we find the following facts.

38.         As Mr Burton stated in his witness statement, brewing is not an exact science. ABV variability is normal and should be expected.

39.         The variability is caused by fermentability variance (for example, a new season’s malt), variable yeast performance, the size and shape of fermentation vessels and the process control.

40.         There are also a number of unknowns and estimates.  At the time that each batch of a particular brand of beer is delivered out of the brewery the brewer will not know what the average ABV is for that particular brand for that month.  It is almost impossible to enter the correct monthly average ABV for any particular brand on the delivery note at the duty point and brewers can only therefore enter the target ABV (his target for that particular brand).

41.         For cask beers the ABV has to be calculated to take into account any secondary fermentation and duty accounted for on the expected strength at the time of consumption (two to three weeks after it actually passed the duty point).  The secondary fermentation may add 0.2% ABV to cask conditioned beer.  The ABV at the time of consumption has to be estimated.

42.         As a consequence, in about 1995 the brewery trade via its trade association (now the BBPA) asked HMRC for a due diligence system.  HMRC agreed and allowed brewers to use a method of determining the chargeable ABV using the declared ABV in certain circumstances subject to the brewers adopting a system of due diligence. This is what is known as the Allowed Method.

43.         As we have identified above, section 12 of Public Notice 226 is vague when it comes to determining what period the monitoring by a brewer of its brewing process should cover.  Section 12 only makes reference to measuring the ABV of each discrete batch: see paragraph 12.6 which in itself would give a variable time period dependent on the length of the production run.  According to Mr Burton, a small run might have three results over three hours, whereas a large run might have over 70 results over three days.  In practice, according to Mr Burton HMRC officers appear to adopt monthly monitoring periods because beer duty operates around monthly accounting periods and this is reflected in the BBPA guidelines.

44.         At all material times Carlsberg adopted the Allowed Method system of due diligence and the system of due diligence applied by HMRC in Carlsberg’s case was as follows.  For each brand Carlsberg is obliged to maintain a reasonable control of ABV and keep records to establish the average Actual ABV of brews of that brand within a month. Carlsberg’s allocated HMRC officer measures Carlsberg’s compliance with the Allowed Method against monthly averages and allows the Actual ABV to exceed the Declared ABV for two months consecutively for minor brands and one month for major brands before issuing an assessment for duty.

45.         The claim in this case relates to one of Carlsberg’s major brands, namely Carlsberg Lager.  Provided that the average Actual ABV each month for that brand does not exceed the Declared ABV for more than a month then duty is paid on the Declared ABV.  If the average Actual ABV each month does exceed the Declared ABV for more than a month, then duty is payable on the Actual ABV as follows. In month one, if the average Actual ABV exceeds the Declared ABV duty is payable on the Declared ABV.  In month two if the average Actual ABV exceeds the Declared ABV duty is payable on the average Actual ABV.

46.         Carlsberg’s claim which is the subject of this appeal is based on the following application of what it contends is envisaged by the Allowed Method and the system of due diligence for Carlsberg Lager.  In month one if the average Actual ABV exceeds the Declared ABV duty is payable on the Declared ABV.  In month two if the average Actual ABV exceeds the Declared ABV duty is payable on the average Actual ABV.  In month three if average Actual ABV is less than Declared ABV then duty is payable on the average Actual AB.

47.         Thus the Allowed Method, as interpreted by Carlsberg, is operated in such a way that it provides for both additions and subtractions according to Actual ABV. 

48.         Mr Burton exhibited copies of documentation giving an example of how the Allowed Method operated where a duty surcharge was raised by HMRC on Carlsberg in relation to the ABVs of certain of the products produced by it as at its Leeds Brewery in 2010 and 2011.  Taking one specific product (House Bitter), a minor brand, for 2010 the Actual ABV exceeded the Declared ABV of 3.2% in January and February, but no duty surcharge was raised by HMRC as the Actual ABV reverted to 3.2% in March.  The only surcharge raised by HMRC in 2010 on House Bitter was for September, because July and August’s Actual ABVs were above the Declared ABVs and did not revert to the Declared ABV in September (only September was surcharged, not July or August, even though in both of those months, the Actual ABV was in excess of the Declared ABV).

49.         In relation to the claim which is the subject of this appeal, the average Actual ABV for Carlsberg lager for the month of July 2011 as shown by Carlsberg’s production records was 3.79%.  Pursuant to the terms of paragraph 12.1 of Notice 226 this was truncated to 3.7%.  The target ABV for Carlsberg lager is 3.8% and this figure was therefore the Declared ABV for July 2011.

50.         Carlsberg paid the necessary duty for this month on the basis of the Declared ABV, the duty being calculated originally by multiplying the hectolitres of beer passing the duty point (that is when it left the brewery) in that month by 3.8% and then applying the duty rate in force at that time (£18.57 per hectolitre percent) to give the duty originally paid.

51.         On 13 October 2011 Carlsberg claimed repayment of the sum of £176,249.05, the claim being based on the difference in duty produced by using the same calculation referred to in paragraph 50 above, but replacing the Declared ABV of 3.8% with the average Actual ABV of 3.7%.

52.         In a letter dated 13 June 2012 HMRC refused this claim on the basis that the Allowed Method has no facility either stated or implied, for duty to be repaid when average Actual ABV is consistently below the Declared ABV.  In support, HMRC stated that the Allowed Method is a concession to facilitate the brewing industry and was designed to follow the basic principles of Regulation 18 whilst allowing tolerances to take account of variances which are often unavoidable in a commercial production environment.

53.         Carlsberg sought a review of this decision and on 1 August 2012 HMRC’s Review Officer, Mr David Paton, confirmed the original decision on the basis that the Declared ABV was the correct figure for duty calculation purposes as it exceeded the Actual ABV.  Carlsberg gave notice of appeal to the Tribunal against this decision on 23 August 2012.

54.         It is common ground that HMRC have not verified the figures that underlie Carlsberg’s claim and that what we are asked to determine is whether such a claim can in principle be maintained. Were we to find that it can, the figures would need to be agreed, or failing agreement, be the subject of a fresh hearing before the Tribunal.

Discussion

Principal ground

55.         Carlsberg’s principal ground of appeal is that the Allowed Method, when correctly interpreted, is that the Actual ABV is the strength of the beer for duty purposes subject to an exception (the “Exception”) namely that Declared ABV may be used instead of Actual ABV for no more than one consecutive month even if Actual ABV is greater.

56.         Carlsberg’s principal ground is founded upon the proposition that it was Parliament’s clear intention to tax the amount of alcohol in the beer. There was no intention, it contends, to tax a notional amount of alcohol that is not in the beer or charge duty on a fictitious amount of alcohol that does not exist.

57.         Mr White submits that the “greater of” Actual ABV and Declared ABV which is the basis of the deemed method set out in Regulation 18 was not designed to charge duty on a fictitious amount of alcohol but is intended as a revenue protection measure.  As such a measure, it is necessary to provide for the brewing outcome when Actual ABV exceeds Declared ABV, but not to create a windfall of duty for HMRC that does not relate to the actual alcohol in the beer.

58.         Mr White notes that in relation to Carlsberg’s claim that is the subject of this appeal Carlsberg does not need to have recourse to the Exception because in respect of the period covered by its claim (July 2011) Actual ABV is less than Declared ABV.  That being the case, Carlsberg is required to pay duty according to the Actual ABV of the beer and is therefore entitled to a repayment of the amount of duty overpaid.

59.         Mr White effectively analyses the process envisaged by the Allowed Method, that is calculation of duty by reference to the Declared ABV, followed by a due diligence process to establish the Actual ABV, was a process that leads to the calculation of a provisional amount of duty due, which will be finalised once the Actual ABV is established.  As a result of that process on his analysis Carlsberg may be required to pay additional duty if the Actual ABV has exceeded the Declared ABV for the period specified by the Exception, or receive a repayment if the calculation shows that the Actual ABV was less than the Declared ABV for the period in question.

60.         Mr White observes that paragraphs 12.3 to 12.8 of Notice 226 do not expressly cover every scenario or give detailed guidance as to the operation of the Allowed Method in practice.  In particular,

(i)            Although the brewer must “make sure that on average, the actual ABV of each finished product equates to” the Declared ABV (paragraph 12.4); and the average of the brewer’s results “should equate closely with the target” (paragraph 12.5); no guidance is given as to the meaning of the word “average” and what is an accepted variance. What figure of actual ABV above Target ABV is acceptable? What are the control limits?

(ii)          Although the brewer “must continuously monitor and record … ABV results (paragraph 12.5); and records must be made of the “strength of each discrete batch” (paragraph 12.6); no guidance is given as to the length of monitoring periods – at the end of which records must be reviewed and action taken.  How often and for what periods should the Actual ABV be compared to the Target ABV?

(iii)        Although where “results have consistently fallen above … target” (paragraph 12.8) “appropriate action … [must be] taken quickly to return the strength of the beer to within its normal specification,” (paragraph 12.6); no guidance is given as to the meaning of the words “consistently” or “quickly”.  No guidance is given as to the length of time that more than an acceptable variance will be tolerated. For how many monitoring periods will more than an acceptable variance be tolerated?

61.         Mr White accepts that the BBPA guidelines provide limited answers to those questions as follows:

(i)            Control limits (the range around the average or norm) are set by the brewer for each product according to normal variability for the particular brewery and product (paragraphs 3.7, 4.1, 4.2.4 and 4.2.5).

(ii)          Monitoring periods are calendar months (paragraphs 4.2.1 and 4.2.2).

(iii)        Control limits should not be exceeded in more than 2 consecutive monthly monitoring periods (paragraph 4.2.3).

However, he submits that the status of these guidelines remain uncertain.  It is therefore necessary to imply terms into the Allowed Method in order to give it its full meaning.  Mr White submits that if the Allowed Method is reduced to its bare essentials then it should be observed that it produces a duty liability based on average Actual ABV.

62.         Carlsberg’s interpretation, Mr White submits, removes the “greater of” concept from the Allowed Method which is the basis of HMRC’s interpretation.  Mr White submits that there is no justification for such a concept because the effect of the due diligence system is that duty is automatically paid on average Actual ABV in accordance with the intention of Parliament and there is no need for further measures to protect the revenue.

63.         Mr White submits that, consistent with its obligations under EU law (and in particular Article 6 of Council Directive 92/84/EEC) which requires Member States to fix the duty on beer by reference to the actual alcoholic strength of the finished product, HMRC have an obligation to publish an Allowed Method that does not have the arbitrary and fictitious features of the deemed method under Regulation 18.  Consistent with these principles the “greater than” concept cannot apply to the Allowed Method and it must be interpreted accordingly.  The discretion given to HMRC in Regulation 18 to prescribe an Allowed Method prevents Regulation 18 being incompatible with the Directive but not if the “greater of” concept is maintained.  The Declared ABV is only valid as a concept when the actual ABV is not known, as an expedient or simplification.

64.         Mr White submits that Carlsberg’s interpretation of the Allowed Method should be preferred because it is rational and results in legal certainty.  It is proportionate and respects the legitimate expectations of taxpayers.  The Allowed Method as interpreted by the Appellant also respects the principles of effectiveness, equivalence, equality of treatment, equal treatment, non-discrimination and fairness.

65.         Applying those principles as principles of interpretation, within the Allowed Method, Mr White submits that it is reasonable to imply a condition as follows:

“If Actual ABV is consistently less than Declared ABV, then duty is due on the Actual ABV; and a repayment will be due.”

and this condition can be implied from the words of paragraph 12.8 of Notice 226, so that as well as supporting an assessment for additional duty when Actual ABV is consistently greater than Declared ABV they would support a repayment  of overpaid duty when Actual ABV is consistently less than Declared ABV.  This is consistent with the rational basis for the Allowed Method which is that liability is determined by reference to average Actual ABV, even where the Declared ABV is greater.

66.         Our starting point in considering Mr White’s submissions is to determine the status of what HMRC have prescribed as the “Allowed Method”.  Regulation 18 sets out the basis on which strength of beer is to be ascertained in order to calculate the duty payable.  As we read Regulation 18, the strength of the beer in question will be deemed to be the greatest of the alternative figures arrived at after working through the four alternatives set out in Regulation 18 (a), (b), (c) and (d) unless HMRC have allowed a different basis of calculation to be used.  In our view, where HMRC have, as is the case with Notice 226, published a statement of what basis they will allow they are bound to accept calculations of duty which are made on that basis unless and until they change the basis on which they will allow the measurement of strength to be made. A brewer is therefore entitled to have his liability to duty calculated by reference to that basis rather than the default position set out in Regulation18.

67.         We accept Mr McNab’s submission that the brewer has a choice as to whether to calculate the duty for which he is liable by reference to the Allowed Method or whether he is content for the default provisions set out in Regulation 18 to apply.  We were told that many smaller brewers are content with the default position, which in practice means they pay duty by reference to Actual ABV of the beer, the information on the delivery note or other material referred to in Regulation 18(c) being ascertained by reference to the Actual ABV rather than, as is the case with lager brewers, the target ABV.

68.         This ability to choose is reflected in the wording of paragraph 12.3 of Notice 226, the first two paragraphs of which summarises the effect of the default provisions of Regulation 18, but the third paragraph of which states that if the brewer complies with certain conditions HMRC will accept for duty purposes the Declared ABV.  On our analysis, if a brewer makes his return on this basis HMRC are bound to calculate the duty accordingly and not by reference to the default provisions of Regulation 18. Paragraph 12.4 of the Notice also emphasises the question of choice.

69.         As Mr McNab correctly identifies, the issue in this appeal is whether HMRC were correct to reject Carlsberg’s claim for repayment under s 137A CEMA.  In our view if the Allowed Method provides as contended for by Carlsberg then Carlsberg will, in principle, succeed in its claim, subject to verification of the amount of claim.  On the basis of our analysis of section 16(4) and (5) FA 1994 set out above, Carlsberg merely needs to satisfy us that as a matter of law HMRC were wrong to reject the claim and that will be the case if we find that the Allowed Method is to be interpreted as Carlsberg contends, without the need to show in addition that HMRC’s rejection of the claim was unreasonable. Conversely if we find that the Allowed Method is to be interpreted as contended by HMRC then we must dismiss the appeal, subject to our findings on Carlsberg’s alternative ground.

70.         The next issue for us to consider before turning to the interpretation of the Allowed Method is what provisions constitute the Allowed Method.  As Mr White submits (see paragraph 60 above) Notice 226 is low on detail.  From our reading of Notice 226 we elicit the following core provisions which are relevant to this appeal as follows:

(1)          HMRC will accept for duty purposes the declared strength of the beer in question instead of the actual strength subject to the brewer having satisfied the conditions described in (2) below (paragraph 12.3).

(2)          The condition is that the brewer can demonstrate that he has exercised due diligence in the control of the brewing process to make sure that, on average, the Actual ABV of each finished product equates to that which is being declared. This due diligence consists of the matters set out in (3) to (5) below (paragraph 12.4).

(3)          The brewer must continuously monitor and record its ABV results and the average of these results should equate closely with the target which must be the declared strength (paragraph 12.5).

(4)          The strength of each discrete batch for each product must be measured and records kept of those results and the action taken to maintain product strength within control limits.  Where ABV varies action must be taken quickly to return the strength of the beer to within its normal specification (paragraphs 12.5 and 12.6).

(5)          Where results have consistently fallen above target, action must be taken as soon as the problem was identified to bring the process back into control or change the Declared ABV (paragraph 12.8).

(6)          Failure to meet the requirements of (5) above will result in an assessment being raised for additional duty.

71.         As submitted by Mr White these core provisions are silent on the following important matters of detail:

(1)          The meaning by “average” in the context of establishing what is the average actual ABV and over what period is that to be established;

(2)          The length of monitoring periods and how often and for what periods should Actual ABV be compared to target ABV;

(3)          The measure of variation between Actual and Declared ABV that is tolerated and for what period; and

(4)          The period allowed to return beer strength to specification so as to avoid an assessment and what constitutes “consistently fallen above target” thus potentially giving rise to an assessment.

72.         As correctly identified by Mr White and as we have found, the BBPA guidelines give answers to the questions on the parameters of the control limits, the monitoring periods (which are calendar months) and when there is a risk of an additional assessment (where the Actual ABV has exceeded the Declared ABV for two consecutive months).  The guidelines also indicate that strength is determined by reference to the average of the strength of all the batches of the relevant brand produced in a particular month so that the actual strength recorded will be the same for each batch, being an average of the strength of all the batches produced in that month.

73.         These guidelines reflect how HMRC in practice decide whether to issue an assessment where Actual ABV exceeds Declared ABV. This is as we describe in our findings of fact set out in paragraphs 42 and 43 above, which is supported by the example of a surcharge being raised as described in paragraph 46 above.

74.         We therefore conclude that these practices form part of the Allowed Method.  It would perhaps be more helpful if HMRC could expand Notice 226 to cover this detail and therefore provide a comprehensive description of the Allowed Method, it being as we have found, the legal basis on which beer duty may be calculated.

75.         We therefore need to consider whether the terms of the Allowed Method go further and incorporate as Carlsberg contends, a right to a repayment if in certain circumstances the Declared ABV is higher than the average Actual ABV.

76.         We consider that issue in two stages.  First, we consider whether such a right arises from a plain reading of the Allowed Method.  Secondly, if we find that a plain reading does not incorporate that right whether it is necessary to imply such a right, as submitted by Mr White, so as to ensure that the Allowed Method is compatible with the terms of the relevant EU Directives.

77.         We have no doubt that on a plain and ordinary reading the Allowed Method makes no provision for the repayment of any duty if the average Actual ABV is for any monitoring period below the Declared ABV for that period.

78.         In our view both the Allowed Method and the default provisions of Regulation 18 prescribe methods of calculating what is deemed to be the strength of the beer.  There is no requirement in these provisions that the strength of the beer used for the purpose of calculating the duty payable is in fact the actual strength of the beer concerned.  This would also be the case if the Allowed Method were held to have the meaning contended by Carlsberg.  There are a number of reasons why neither method results in the Actual ABV being used in any case:

(1)          As a practical matter, even though according to HMRC’s due diligence requirements every batch brewed should be tested, all the batches produced in every month are aggregated and an average ABV is determined for all of those batches which is used as the Actual ABV for all the beer of the relevant brand produced in that month. Obviously the actual strength of any particular consignment of beer may be above or below that average;

(2)          HMRC have, as permitted by Council Directive 92/83/EEC, allowed brewers to ignore figures after the first decimal place in expressing the ABV;

(3)          The Allowed Method allows the Declared ABV to be used instead of the Actual ABV where the due diligence process has been robust and does not (as we find) take account of the fact that the Actual ABV in respect of any particular month may be less than the Declared ABV.

79.         In our view, and contrary to Mr White’s submissions, the Allowed Method operates in effect to temper the default provisions in Regulation 18.  It employs the “greater of” concept that Carlsberg objects to as its starting point, that is the general rule that the strength of the beer is deemed to be the greater of the Actual ABV and the Declared ABV.  It tempers that rule by allowing the Declared ABV to be used even if the average ABV is greater as long as there is a proper due diligence process in place with a view to ensuring that the Actual ABV does not, on average, exceed the Declared ABV and any deviations above the Actual ABV do not occur consistently.  This is clearly apparent from the way paragraph 12.3 of Notice 226 is drafted; the last sentence of this paragraph clearly reveals the Allowed Method as being a declared strength method, with exceptions.  We reject entirely Mr White’s submission that it can be construed as an actual strength method with exceptions. As with the default provisions of Regulation 18, the general position is that the declared strength is used as the basis of calculations.

80.         We therefore reject Mr White’s submission that Parliament intended to tax the amount of alcohol in the beer.  It is clear from Regulation 18 that Parliament intended to tax what was deemed to be the amount of alcohol in the beer.   It appears to us that the reasons why this is the case and why the Allowed Method operates the way it does is that it is impracticable for the vast majority of brewers to have to calculate duty by reference to the actual strength of each batch of beer they produce.  Each such brewer works to a target ABV and labels and markets its products accordingly.  If a customer buys a can of Carlsberg lager he can expect its strength to be 3.8% ABV or thereabouts.  Brewers wish to be relieved from the harsh financial consequences of finding that the Actual ABV is in fact higher than the Declared ABV, which it is assumed what they have budgeted for in setting their targets, and the Allowed Method gives them relief from that. The Allowed Method therefore gives brewers more certainty as to the duty they are likely to pay and reflects the targets that they have set themselves.

81.         It is therefore the case, contrary to Mr White’s submissions, that both Regulation 18 and the Allowed Method do charge duty on a fictitious amount of alcohol, or to put it less provocatively, alcohol that might not be in the beer. That is the inevitable consequence of not charging duty by reference to the actual alcoholic context of each batch brewed.

82.         Nevertheless we are satisfied that neither Regulation 18 nor the Allowed Method is designed to achieve that result.  It appears to us that the Allowed Method is designed to operate so that close control of the brewing process is maintained to ensure that as far as possible the average Actual ABV correlates to the Declared ABV.  As we observed in paragraph 18 above, this is designed as a revenue protection measure, but it appears to us that if a brewer uses the Allowed Method the due diligence process he is required to undertake will also ensure that generally he will not produce beer whose Actual ABV is significantly lower than the Declared ABV. If it does, then it may be offset by the fact that there will be months where the Declared ABV exceeds the Actual ABV. In any event the remedy is in the brewer’s own hands by adjusting its brewing processes or its target ABV; it is clearly not in its interests to produce beer with a strength below the target and pay duty by reference to a higher strength.

83.         Indeed the BBPA guidelines recognise this in the passage from paragraph 4.1 of those guidelines that we quoted in paragraph 23 above, and the example table we referred to showing variations either side of the target ABV.  Mr McNab also observes that in respect of the example referred to by Mr Burton of where a surcharge was imposed in respect of Carlsberg’s House Bitter (see paragraph 46 above) in seven of the months in 2010 referred to in the assessment the Actual ABV exceeded the Declared ABV but because of the way the Allowed Method operated, as described by Mr Burton and referred to in paragraph 48 above, the only surcharge raised was for one month.  There is therefore plenty of “credit” there to offset any months in which the Declared ABV exceeded the Actual ABV.

84.         The above examples also show that even if Mr White were right and a repayment claim can be made where Declared ABV exceeded the average Actual ABV, we cannot see why it would operate in the manner for which he contends, which is that in respect of any month where such event occurs a repayment right arises, whereas where the average Actual ABV exceeds the Declared ABV, under the Allowed Method the right to surcharge only arises where that situation persists for more than a month.

85.         We accept Mr McNab’s submission that the Allowed Method is to be characterised as a declared strength method based on averaging the strengths of the various batches of beer produced in each month and assumes that the average Actual ABV will be approximately equal to the declared strength, with the result that duty is paid on the Actual ABV, taking into account the ups and downs of the various months. By contrast, what Carlsberg contends for is not based on the average ABV because in seeking repayment for an isolated month it takes no account of the fact that no duty is paid in respect of other months where the Actual ABV is greater than the Declared ABV.  It would mean that the lesser of the two figures was always taken and would not produce an average figure over time that tended towards the declared strength.  Consequently, we see no basis on which it is necessary to imply a condition of the nature contended for by Carlsberg to give efficacy to the Allowed Method; it can operate as intended so as to ensure that the Declared ABV and average Actual ABV correlate over time without the implication of such a condition.

86.         We therefore turn to the question as to whether the Allowed Method without the implied condition that Carlsberg contends for is compatible with the EU Directives.  In accordance with well established principles, there is an obligation upon us to construe UK domestic legislation in a manner which conforms to the UK’s obligations under those Directives and if it is necessary to write in words to the legislation to achieve that result then that is a course we should adopt.  This principle will apply to the Allowed Method which, as we have found, is part of the UK statutory framework pursuant to which liability to pay beer duty is assessed.  Article 1 of Council Directive 92/83/EEC requires Member States to levy beer duty “by reference …to the number of hectolitre/degrees of alcoholic strength by volume”.  We must construe the Allowed Method so that it conforms as far as possible to that obligation.

87.         We reject Mr White’s submission that the Allowed Method without the implied condition for which he contends would be incompatible with the Directive.  Directives operate at a high level of generality and in our view the Allowed Method, which operates in a manner that is designed to ensure that the Declared ABV and Actual ABV correlate over time, is consistent with the principle laid down in the Directive that the duty is fixed “by reference to” the actual alcoholic strength.  It cannot be the case that under the Directive duty can only be calculated by reference to the actual amount of alcohol in each batch of beer brewed so that there must be some tempering of that absolute principle in order that a practicable method of calculating and collecting duty is provided for. In our view Mr McNab is correct in his submission that procedures for payment and repayment of duty are left to Member States to be dealt with in their domestic legislation and Article 9 of Council Directive 2008/118 is, as set out in paragraph 9 above, sufficient authority for that proposition.  In our view the practical measures taken by HMRC in establishing the Allowed Method fall within the scope of that power and do not derogate from the underlying principle in Article 1 of Council Directive 92/83/EEC. We therefore reject Mr White’s submission that the Allowed Method without the implied condition for which he contends offends the principles of effectiveness, equivalence, equality of treatment, non-discrimination and fairness.

88.         Nor do we believe that without this implied condition the Allowed Method is disproportionate and lacks legal certainty. As we have found, the combination of Notice 226 and the practice of HMRC, also reflected in the BBPA guidelines, set out the basis on which a brewer can choose to measure the strength of its beer by reference to the Allowed Method.

89.         We therefore reject Carlsberg’s principal ground of appeal and turn to its alternative ground.

Alternative Ground

90.         Mr White submits that notwithstanding the recent Upper Tribunal cases in HMRC v Abdul Noor [2013] UKUT 071 (TCC) and HMRC v Hok Ltd [2012] UKUT 363 (TCC) which establish that there is no general supervisory jurisdiction by way of judicial review in this Tribunal, the Tribunal should hear Carlsberg’s alternative ground for the following reasons:

(1)          Notwithstanding Carlsberg’s reliance inter alia on public law principles, those principles also reflect principles of Community law relevant to excise duty liability under UK legislation implementing the harmonised community excise duty provisions. Such principles may be pleaded in any dispute determining the quantum of excise duty due;

(2)          The Appellant’s case is an appeal against a decision on a review upholding an initial relevant decision on a claim for repayment of excise duty under the Finance Act 1994 s 13A(2)(d).  Even if the Commissioners initial decision depended upon an earlier decision by them subject to public law principles, the actual decision appealed against is expressly subject to the jurisdiction of the Tribunal; and

(3)          At the time of their initial decision the Commissioners offered the Appellant a review of their initial decision and notified the Appellant of its right to appeal direct to the First-tier Tribunal.  The decision on a review informed the Appellant of its right to appeal to the First-tier Tribunal.  In the Appellant’s circumstances, there is no statutory prohibition against the Appellant’s appeal.  In the absence of a statutory prohibition against appeal to the Tribunal, the Commissioners may not now raise the issue of the Tribunal’s jurisdiction.

91.         Mr White relies on the statement in paragraph 31 of Noor that it was open to the Tribunal to consider public law issues if it was necessary to do so in the context of deciding issues clearly falling within its jurisdiction.  Paragraph 87 of Noor makes it clear that the exercise by HMRC of its statutory powers within the rubric of tax legislation was within the Tribunal’s jurisdiction, drawing a distinction between a claim under an agreement which HMRC had power to make and an ultra vires agreement in the following terms:

“87. In our view, the F-tT does not have jurisdiction to give effect to any legitimate expectation which Mr Noor may be able to establish in relation to any credit for input tax. We are of the view that Mr Mantle is correct in his submission that the right of appeal given by section 83(1) (c) is an appeal in respect of a person’s right to credit for input tax under the VAT legislation. Within the rubric “VAT legislation” it may be right to include any provision which, directly or indirectly, has an impact on the amount of credit due but we do not need to decide the point. Thus, if HMRC have power (whether as part of their care and management powers or some other statutory power) to enter into an agreement with a taxpayer and that agreement, according to its terms, results in an entitlement to a different amount of credit for input tax than would have resulted in the absence of the agreement, the amount ascertained in accordance with the agreement may be one arising “under the VAT legislation” as we are using that phrase.  In contrast, a person may claim a right based on legitimate expectation which goes behind his entitlement ascertained in accordance with the VAT legislation (in that sense), in such a case, the legitimate expectation is a matter for remedy by judicial review in the Administrative Court; the F-tT has no jurisdiction to determine the disputed issue in the context of an appeal under section 83.  As Mr Mantle puts it, the jurisdiction of the F-tT is appellate (i.e. on appeal from a refusal of HMRC to allow a claim).  The F-tT has no general supervisory jurisdiction over the decisions of HMRC. That does not mean that under section 83(1) (c) the F-tT cannot examine the exercise of a discretion, given to HMRC under primary or subordinate VAT legislation relating to the entitlement to input tax credit, and adjudicate on whether the discretion had been exercised reasonably (see eg Best Buys Supplies Ltd v HMRC [2012] STC 885 UT at [48] – [53] – a discretion under Reg 29(2) of the VAT Regulations).  Although that jurisdiction can be described as supervisory, it relates to the exercise of a discretion which the legislation clearly confers on HMRC.  That is to be contrasted with the case of an ultra vires contract or a claim based on legitimate expectation where HMRC are acting altogether outside their powers.”

92.         Mr White relies on this passage for the proposition that within the context of Carlsberg’s appeal, the F-tT has jurisdiction to examine the exercise of the discretion given to HMRC under Regulation 18 relating to the strength of the beer; and to adjudicate on whether the discretion has been exercised reasonably.  Although that jurisdiction can be described as supervisory, it relates to the exercise of a discretion which the legislation has clearly conferred to HMRC.

93.         We do not accept that the passage from Noor quoted supports Mr White’s proposition.  Noor recognises that where, for example, the VAT legislation gives HMRC a discretion as to accept alternative evidence to support a claim for deduction of input tax, (the discretion under regulation 29(2) of the VAT Regulations 1995) then the Tribunal has jurisdiction to consider whether that discretion has been exercised reasonably because it has a statutory jurisdiction to determine any claim for deduction of input tax. This is different to the situation with regard to the manner in which the discretion in Regulation 18 operates.  HMRC have exercised the power in Regulation 18 not by entering into individual agreements with taxpayers as is envisaged by Regulation 29(2) of the VAT Regulations but by prescribing an Allowed Method which applies to all brewers who care to take advantage of it. As we have determined, where a brewer calculates his liability to beer duty by reference to the Allowed Method, HMRC are bound to accept such a calculation, if correct, and the Allowed Method is the statutory basis for the calculation rather than the default provisions in Regulation 18.  The Tribunal has full jurisdiction, not merely a supervisory jurisdiction to consider any appeal against any determination made by reference to the Allowed Method by virtue of section 16(5) FA 1994.

94.         However, in our view that exhausts the Tribunal’s jurisdiction. Carlsberg contend that the Tribunal has a supervisory jurisdiction with regard to the terms of the Allowed Method.  That is in effect a complaint that it was unlawful for HMRC to have prescribed a method that did not permit Carlsberg to claim a repayment of duty when the Actual ABV for the period in question was less than the Declared ABV.  That is a claim that goes beyond Carlsberg’s right to have its repayment claim determined and on the basis of the distinction drawn in paragraph 87 of Noor is not a claim that falls within the jurisdiction of the Tribunal.  It is, as Mr McNab submits, a challenge to the manner in which HMRC have exercised the power to prescribe an Allowed Method which is a matter for the Administrative Court to determine following an application for judicial review.

95.         Carlsberg’s reliance on those public law principles that reflect the principles of Community law can be considered within the scope of the Tribunal’s substantive jurisdiction with regard to the interpretation of the Allowed Method, and we have followed that course in this case.  Their ability to rely on these principles does not however found a basis for a supervisory jurisdiction over the terms of the Allowed Method.

96.         It follows from that conclusion on the Tribunal’s jurisdiction that it is unnecessary for us to consider the extent to which the power to prescribe the Allowed Method has been exercised, as Carlsberg contends, unfairly and in a manner which amounts to an abuse of power.

97.         Nevertheless for completeness we deal briefly with Mr White’s submissions on this issue.  He relies primarily on the same grounds on which Carlsberg contended that the implied condition that repayment of duty must occur where the Actual ABV is lower than the Declared ABV, namely that the exercise of the power must be proportionate and respect the principles of effectiveness, equivalence and equality of treatment, non-discrimination and fairness.  We dealt with those grounds in paragraph 87 above and reiterate our conclusion that the Allowed Method does not infringe those principles.

98.         Additionally, Mr White relies on the principles expressed by the Court of Appeal in R v Inland Revenue Commissioners ex parte Unilever PLC [1996] STC 681, namely that HMRC has a duty to act fairly in accordance with the highest public standards.  That case concerned a situation where HMRC had adopted a practice for a period of over twenty years of permitting claims for loss relief after the expiry of the statutory time period but then sought to resile from it with notice to the taxpayer.  In our view the conduct of HMRC complained of in this case comes nowhere near that complained of in HMRC in Unilever which was characterised by Simon Brown LJ as “so outrageously unfair that it should not be allowed to stand” (see page 697).  Sir Thomas Bingham MR stated at page 692 of the judgment that

“The threshold of public law irrationality is notoriously high.  It is to be remembered that what may seem fair treatment of one taxpayer may be unfair if other taxpayers similarly placed have been treated differently.  And in all save exceptional circumstances the Revenue are the best judge of what is fair.  It is not however, been suggested that the detailed history described above has any parallel. The circumstances are, literally, exceptional.”

99.         It will be readily apparent from our findings on the principal ground that in our view there is no basis on which it would be maintained that the Allowed Method comes anywhere near being characterised as “outrageously unfair” or being regarded as “literally, exceptional”.

100.     We therefore reject Carlsberg’s alternative ground.

Disposition

101.     We dismiss the appeal.

102.     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.

 

 

TIMOTHY HERRINGTON

TRIBUNAL JUDGE

 

RELEASE DATE: 15 October 2013