Diesel fuel supplied for use in road vehicles (“white diesel”) is subject to excise duty at much higher rates than the materially identical product supplied for use by tractors and other agricultural vehicles, or for other forms of “home use” excepted under the Hydrocarbon Oil Duties Act 1979 (“HODA 1979”). The relief from the full rate of duty is allowed by way of a rebate when “heavy oil” (which includes diesel fuel) is delivered for home use: section 11(1) of HODA 1979. By virtue of section 12(2), no heavy oil on which rebate has been allowed shall be used as fuel for a road vehicle, or taken into a road vehicle as fuel, unless an amount equal to the rebate has been paid to the Commissioners for Her Majesty’s Revenue and Customs (“HMRC”).

Because of the obvious opportunities for fraud inherent in such a system, regulations made under enabling powers in HODA 1979 require diesel oil delivered for home use to be dyed red (hence its colloquial description as “red diesel”) and to contain specified chemical markers which can be detected by chemical analysis: see the Hydrocarbon Oil (Marking) Regulations 2002, SI 2002 No.1773. In addition, HMRC have wide powers to forfeit heavy oil taken into a road vehicle in contravention of section 12(2) of HODA 1979, as well as any “… vehicle… container... or any thing whatsoever which has been used for the carriage, handling, deposit or concealment of the thing so liable to forfeiture”: see sections 139 and 141(1) of the Customs and Excise Management Act 1979 (“CEMA 1979”).

Section 152 of CEMA 1979 provides that:

Section 139(6) and schedule 3 to CEMA 1979 contain important provisions which apply when goods are liable to forfeiture. The general effect of paragraphs 3 and 5 of schedule 3 is that a person who wishes to challenge a forfeiture on the ground that the thing seized was not so liable must give written notice of his claim to HMRC within one month of the date of the seizure, and if no such notice has been given by the expiry of that period “the thing in question shall be deemed to have been duly condemned as forfeited”. As this court explained in Revenue & Customs Commissioners v Jones [2011] EWCA Civ 824, [2012] Ch 414, the result of this statutory deeming is that, on any subsequent appeal under section 16 of the Finance Act 1994 against a refusal to restore the seized goods under section 152(b) of CEMA 1979, the validity of the forfeiture is no longer open to question and may not be revisited by the First-tier

It must also be appreciated that the jurisdiction of the FTT on an appeal against a decision by HMRC not to restore forfeited goods is, in other respects too, unusually circumscribed. The effect of the relevant provisions was concisely stated by Mummery LJ in Jones at [42] to [43]:

In view of the importance of those provisions in the present case, I will set out the relevant provisions of sections 15 and 16 of the 1994 Act:

confirm the decision; or

withdraw or vary the decision and take such further steps (if any) in consequence of the withdrawal or variation as they may consider appropriate.

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;

to require the Commissioners to conduct, in accordance with the directions of the tribunal, a further review of the original decision; and

in the case of a decision which has already been acted on or taken effect and cannot be remedied by a further review, 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.

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.

On an appeal under this section the burden of proof as to [various matters relating to penalties] 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.

It is common ground that a decision made by HMRC under section 152(b) of CEMA 1979 is an “ancillary matter” for the purposes of section 16, from which it follows that the powers conferred on the FTT on an appeal from the relevant review decision are confined to those set out in subsection (4), and are also dependent upon the FTT being satisfied that the decision is one which HMRC “could not reasonably have arrived at”. The apparent strictness of this approach has, however, been significantly alleviated by the decision of this court in Gora v Customs and Excise Commissioners [2003] EWCA Civ 525, [2004] QB 93, where Pill LJ accepted the submission of counsel for HMRC (Mr Kenneth Parker QC, as he then was) that the provisions of section 16 do not oust the power of the FTT to conduct a fact-finding exercise, with the consequence that it is open to the FTT on an appeal from a review decision to decide the primary facts and

Under further powers conferred by sections 100G and 100H of CEMA 1979, and regulations made thereunder (the Hydrocarbon Oil (Registered Dealers in Controlled Oil) Regulations 2002, SI 2002 No.3057), HMRC have at all material times operated a scheme whereby dealers in “controlled oils” (which include red diesel) must, subject to immaterial exceptions, apply to HMRC to be registered as a Registered Dealer in Controlled Oil (or “RDCO”). As the Upper Tribunal explained in the decision under appeal at [8]:

The power to revoke the registration of an RDCO is conferred by section 100G(5), which provides that:

The statutory and regulatory framework applicable to RDCOs is fleshed out in published guidance which is issued and updated from time to time by HMRC. We are concerned with two versions of the guidance contained in Public Notice 192, the first of which was issued in February 2013 (superseding previous versions) and the second of which was issued in May 2014, again superseding its predecessor. In the February

it is considered necessary for the protection of the revenue because, for example, you have been involved in the misuse of controlled oil. In such cases, we are likely to prosecute you

you persistently fail to meet the requirements of the scheme. However, this is likely to be the final step following a series of warning letters and civil penalties – see section 8.

It will be noted that, under this version of the Notice, the first type of case in which cancellation of approval is said to be “likely” is where “it is considered necessary for the protection of the revenue”. I would not read the example which is then given, namely involvement in the misuse of controlled oil, as an exhaustive statement of the circumstances in which cancellation of approval may be considered necessary on this ground, although such involvement is no doubt a paradigm case and (as the Notice said) likely to lead to prosecution. It is relevant to note in this connection that criminal liability for contravention of section 12(2) of HODA 1979, as opposed to liability for civil penalties, depends on proof of the intent that the restrictions imposed by section 12 should be contravened: see subsections 13(3) and (4).

The May 2014 version of Notice 192 differed in some material respects from its predecessor. First, it introduced a new “fit and proper” test which HMRC would apply when deciding whether to grant RDCO status. The detailed requirements were set out in paragraph 4.3, and included warnings that HMRC would be “very unlikely to give approval” if the business, or the applicant, or anyone with an important role or interest in the business, had previously “been involved in fuel-laundering” or “had oil or vehicles or any other revenue goods seized from them”. Cancellation of approval was then dealt with in section 4.8, as follows:

it is considered necessary for the protection of the revenue because, for example, we have evidence that you have been involved in the misuse of controlled oil or excise fraud. In such cases, we may also prosecute you

in light of any new information that comes to our attention, or that you notify to us, we are no longer satisfied that you are fit and proper to hold an RDCO approval as per paragraph 4.3, and

you persistently fail to meet the requirements of the scheme, for example, fail to exercise your obligations or fail to submit HO5 returns on time, However, this is likely to be the final step following a series of warning letters and civil penalties - see paragraph 6.5 and section 8

Paragraph 8.5, headed “Withdrawal of approval”, stated that:

Having sketched in the statutory and regulatory background, I can now explain what this case is about. The respondent (and appellant below), Behzad Fuels (UK) Limited

Following certain events in 2009 which I will describe later in this judgment, on 4 March 2013 road fuel testing officers of HMRC visited the Company’s premises on an industrial estate in Rainham, Essex. The officers detected the presence of red diesel in the white diesel bulk storage tank, and in the running tanks of four road tankers on site. On 5 March 2013, HMRC seized the vehicles and the fuel as liable to forfeiture under section 139 of CEMA 1979. The Company instructed solicitors, and correspondence ensued. HMRC were asked to restore the vehicles and the fuel, but declined to do so, and their refusal was subsequently upheld on review. No steps were taken by the Company to challenge the forfeiture, but in due course the Company appealed against HMRC’s review decision not to restore the vehicles and uncontaminated fuel. This was the beginning of the proceedings which have been described throughout as “the Restoration Appeal”.

On 5 November 2013, HMRC decided to revoke their approval of the Company as an RDCO. This decision was made at a time when the version of Notice 192 in force was the February 2013 version. The decision was subsequently upheld on review, and in March 2014 the Company appealed against the review decision, thereby setting in motion the second set of proceedings with which we are concerned (“the Revocation Appeal”).

In August 2014, the Restoration Appeal was heard by the FTT (Judge John Brookes and Mrs Shameem Akhtar) (“the 2014 FTT”). By its decision released on 28 August 2014 (“the 2014 FTT Decision”), the 2014 FTT found that HMRC’s decision in relation to the fuel was reasonable and proportionate, but required HMRC to conduct a further review of their decision not to restore the vehicles. This led to a further decision by HMRC, taken on 19 December 2014, not to restore the vehicles, which in turn generated a further appeal by the Company. This appeal was then consolidated with the

The Company then appealed to the Upper Tribunal, with permission granted by Judge Berner on 15 July 2016. The appeal was heard by Judge Berner, sitting with Judge Timothy Herrington, in April 2017. By its decision released on 8 August 2017 (“the UT Decision”), [2017] UKUT 0321 (TCC), the Upper Tribunal set aside the relevant decisions by HMRC (i.e. the second review decision in the Restoration Appeal, and the review decision in the Revocation Appeal), and directed fresh reviews of those decisions to be undertaken by different reviewing officers. The Upper Tribunal directed that the new reviews were to proceed on the basis of the findings of fact made by the FTT in 2014 and 2016, but subject to certain further findings made by the Upper Tribunal itself as set out in the UT Decision at [104] to [108]. In addition, the Upper Tribunal directed that the fresh review of the Revocation Decision should be made by reference to the February 2013 version of Notice 192, because that was the version in force at the time of the original revocation decision.

That, in summary, is the rather convoluted route which has led to the present appeal to this court by HMRC against the UT Decision, brought with permission granted by Newey LJ.

HMRC advance two grounds of appeal. The first (ground 1) applies only to the Revocation Decision, and raises a short point about the terms on which the Upper Tribunal directed HMRC to conduct their fresh review of that decision, namely that it should be conducted on the basis of Notice 192 as it stood at the time of the Revocation Decision (i.e. the February 2013 version), rather than the version which is currently in force. This direction, it is said, constituted an error of law.

The second ground of appeal (ground 2) relates to both the Revocation Decision and the Restoration Decision. Again, it concerns the basis on which the further reviews directed by the Upper Tribunal should be carried out. It is said that the Upper Tribunal “erred in law in holding that HMRC must consider, when making a decision on restoration or revocation of approval, that Behzad was not involved in laundering of oil, in circumstances where there had been no finding, on the balance of probabilities, that it was so involved.”

At this point, I need to explain what is meant by “laundering of oil”. The Upper Tribunal gave the following helpful explanation:

In paragraph 11 of their skeleton argument in support of the appeal, counsel for HMRC make the point (which I do not understand to be controversial) that:

Reverting to the grounds of appeal, HMRC contend that the directions given by the Upper Tribunal “in respect of the use of a now defunct policy and the need to treat a trader as honest, absent a finding to the contrary, are matters which are likely to have an effect on the approach the Commissioners must take in similar cases, involving significant amounts of duty and tax.” We are therefore invited to allow the appeal on both grounds, and in that event to determine that there is no need for either Decision to be retaken “because the outcome, on a proper approach, would inevitably be the same.”

I will now summarise those aspects of the factual history which are relevant to the grounds of appeal, drawing gratefully for this purpose on the three Tribunal decisions. In particular, a full account of the facts may be found in the UT Decision at [35] to [70].

The story begins on 12 June 2009, when the Company made a delivery of fuel to a garage near Liverpool, but the customer alleged that the fuel delivered was not roadworthy. The Company informed HMRC of the allegation and asked HMRC to carry out tests on the fuel. HMRC found no problem with the fuel in the container pots of the vehicle used to deliver the fuel, but discovered that the fuel in the running tank was contaminated with red diesel.

HMRC then seized the vehicle, but entered into a “restoration agreement” on payment of a fee. The agreement contained an acknowledgement that all traces of fuel must be removed from the vehicle’s running system within 24 hours from release, and contained a warning about the consequences of misuse of rebated fuel being detected on future occasions. The vehicle then returned to the Company’s depot, where it remained stationary from 12 to 17 June 2009.

On 16 June 2009, HMRC visited the Company’s depot and tested the running tank of the seized vehicle. Traces of red diesel were found in the running tank. HMRC again seized the vehicle and entered into a restoration agreement on payment of a further fee. The agreement contained a warning that this was the second occasion on which misuse of rebated fuels had been detected, and that if there was a further contravention, the vehicle would be seized and not returned. The Company then wrote to HMRC asking for the further fee to be returned, on the ground that the running tank had been emptied “to the extent physically possible” and refilled, and that the positive test was a result of residual contaminating particles still being present. HMRC refused the request, but the Company did not require HMRC to institute condemnation proceedings.

Shortly after the Company began trading in 2009, an employee of the Company (Mr Makkatu) conducted some experiments on the purification of a consignment of some

By 2013, the Company had a fleet of 10 tankers which made an average of 20 daily deliveries, amounting to approximately 700,000 litres of white diesel, 8.2 million litres of red diesel and a million litres of kerosene each year.

I have already referred to HMRC’s visit to the Company’s premises which took place on 4 March 2013. Apart from detecting the presence of red diesel in the white diesel bulk storage tank, and in the running tanks of four fuel tankers, HMRC’s officers also found two 25kg bags labelled “bleaching earth” or “bleaching agent” on the site. One of these bags was in the “shell” of a disused washing machine, and the other was in an intermediate bulk container found inside a locked container belonging to the owner of the site, the Company’s landlord: see the 2014 FTT Decision at [25]. During interviews which were held under caution a few days later, it was explained by the Company’s senior management that any bleaching agent found at the site was likely to have been left over from the unauthorised biodiesel experiments conducted in 2009.

A principal ground of HMRC’s subsequent review decision not to restore the seized vehicles and fuel, on 15 July 2013, was that the evidence suggested that laundering of fuel was taking place at the Company’s site, bearing in mind the presence of a bag labelled “bleaching earth”. The detailed conclusions of HMRC’s reviewing officer, Ms Bines, are set out in the 2014 FTT Decision at [34]. After considering extensive evidence during a two day hearing, including unchallenged evidence from Mr Makkatu and expert evidence on both sides, the 2014 FTT recorded at [41] the acceptance by the Company that the white diesel contained in the white bulk storage tank was contaminated with red diesel at a level of about 4%, and that a similar level of contamination was present in the running tanks of the four vehicles. Although the Company’s expert, Dr Stinton, thought that the level of contamination in the white diesel storage tank was “more likely to have been caused by errors or poor practice than by a determined effort to launder rebated fuels in order to make a profit”, he accepted in cross-examination that laundering was a possibility.

Accordingly, the 2014 FTT found that there was evidence on which the review officer could conclude that laundering had taken place. Not only was the fuel contaminated, but bags labelled “bleaching earth” were present on the site, and the analysis of the “sludge” extracted from a fuel pot of one of the seized vehicles contained statutory markers for red diesel. The decision not to restore the fuel was therefore both reasonable and proportionate: [49] and [50].

In relation to the vehicles, however, the 2014 FTT concluded that the explanations given by the Company for the contamination which had been detected were “credible and plausible”, and in some respects supported by contemporary documentary evidence. This material, they concluded, had not been given due consideration by Ms Bines, with the consequence that “the decision not to restore the vehicles could not have been arrived at reasonably”. Furthermore, the 2014 FTT considered that “a proportionate response would have been to consider restoration of the vehicles for a

The 2014 FTT therefore directed that a further review be undertaken in respect of the decision not to restore the vehicles, such review to take into account the explanations which had been given for the presence of bleaching agent at the site and the contamination detected on the vehicles, and the question whether it would be proportionate in the circumstances to return any or all of the vehicles for a fee.

As I have already recorded, the review directed by the 2014 FTT led to a further decision by HMRC not to restore the vehicles, and the Company’s appeal against that further decision was eventually heard (together with the Revocation Appeal) by the 2016 FTT.

The hearing before the 2016 FTT took place over three days in February 2016. The 2016 FTT had the benefit of oral evidence from three senior managers of the Company, all of whom it found to be honest and reliable witnesses, as well as from the same two experts who had given evidence to the 2014 FTT, and from the officer (Mrs Brown) who had performed the review of the decision not to restore the vehicles. The evidence of each expert was found to be “clear, dispassionate and useful”, and HMRC’s witnesses of fact were likewise found to be honest and reliable.

The further findings of fact made by the 2016 FTT are summarised in the UT Decision at [50] to [62], to which reference should be made for the full details. The most significant of those findings, stated in summary form in [49] of the 2016 FTT Decision and then explained in the following three paragraphs, is that all of the samples of fuel taken from the white diesel storage tank and the running tanks of the four vehicles on 4 March 2013 “contained laundered fuel”. This conclusion was founded on the chemical analysis and the evidence of the experts, and took full account of a possible alternative hypothesis advanced by Dr Stinton for the alteration in the ratio of two of the statutory markers. The significance of this finding, of course, is that the presence of laundered fuel, with altered ratios between statutory markers, could not be explained merely on the basis of poor wet line procedures, or the inadvertent mixture of red diesel with white.

The 2016 FTT then moved on to consider the separate question of whether the Company was itself involved in the laundering of the fuel: see [59]. In the next paragraph, it directed itself as follows:

The 2016 FTT then summarised some salient features of what it called the “competing evidence”, some of which suggested that the Company could have been involved in laundering (including the sludge sample taken from one of the vehicles, and the two bags of bleaching agent found during HMRC’s 2013 visit) and some of which pointed the other way (including the fact that there was no evidence on site of the environmental damage to be expected if substantial laundering operations had taken place there, the fact that no evidence of laundering or laundered fuel had been found during a previous visit by HMRC officers in 2011, and the 2016 FTT’s finding that “the Company had shown itself to be honest in its dealings with HMRC”).

The 2016 FTT then stated their conclusions, at [64] to [70]. They began by considering whether the sludge was bleaching earth, because “that would suggest (though would not conclusively prove) that the Company was involved in laundering”. They concluded, after reviewing the expert evidence on this point, that “the solid part of the sludge was bleaching earth”: see [67].

The 2016 FTT then continued, in a passage which I need to quote in full: “68. We also considered that the presence of bags marked

We did not consider that the [other items of competing evidence] pointed strongly in either direction. If the Company were engaged in laundering, it might simply have chosen a more discreet spot than its own business premises to undertake that laundering. Equally, if the Company were laundering fuel, it might reasonably be expected to hide any laundering agents before Dr Stinton’s visit which took place several months after the seizure. The fact that no evidence of laundering was found in 2011 does not preclude the possibility that the Company was laundering fuel in 2013.

Therefore, if matters stopped there, we would have concluded on a balance of probabilities (based on the evidence of the sludge and the presence of fuller’s earth at the Company’s premises) that the Company was laundering fuel particularly given the absence of any other explanation as to how laundered fuel came to be present in the Company’s white diesel storage tank. However, the Company’s evident honesty in its dealings with HMRC referred to at [63(4)] [i.e. when it specifically requested HMRC to perform tests on its fuel in 2009, and the answers subsequently given in interview], and the fact that we considered the Company’s witnesses to be honest and reliable has caused us to stop short of making a positive finding that the Company was involved with laundering. We have not made a positive finding that the Company was not involved with laundering partly because of the points made at [64] to [69] and because we consider that it was possible for the Company to be involved in laundering without [its senior management] being aware of this. Mr Menon is based in Doha and said in his evidence that he did not know about the experiments that Mr Makkatu was performing with biodiesel which demonstrated that Mr Menon could not know about everything that was going on at the Company. Both Mr Kumaran and Mr James were based in the UK but said that, while they were aware that Mr Makkatu was performing experiments on biodiesel, neither of them knew that those experiments involved bleaching earth. Therefore, even people based in the UK were not aware of precisely what was going on at the Company’s premises. Overall, therefore, the Company has not discharged the burden of proving that none of its agents or employees were involved in the laundering of fuel.”

The Upper Tribunal described the findings of the 2016 FTT in [69] and [70] of the 2016

The reviewing officer had not been wrong to conclude that the seizure of the vehicles in 2013 was the Company’s third “offence”, because the two seizures in 2009 did not have to be regarded as a single incident. The officer’s conclusion was “certainly at the tougher end of the spectrum”, but was not unreasonable.

The Company had been given clear warning of HMRC’s general policy of not restoring vehicles on a third seizure: [103].

The reviewing officer had not failed to consider exercising her discretion to restore the vehicles, nor was there any error of law in her decision: [104].

Although the officer accepted that she had not taken into account the fact that the Company had itself contacted HMRC in 2009, counsel for the Company had overstated matters when describing this as a “self-referral”, and in any event it did not demonstrate that everyone at the Company was unaware of the presence of contaminated fuel: [106]. Furthermore, the officer’s evidence that knowledge of the “self-referral” would not have made any difference to her decision was accepted by the 2016 FTT, and this was a reasonable stance for her to take.

Even if the Company had shown that it was completely unaware that it was using contaminated fuel in 2009, there were aggravating factors associated with the seizures of the vehicles in 2013. See [107]:

Accordingly, although the officer’s failure to take into account the 2009 “selfreferral” was a defect in her review decision, her decision would inevitably have been the same if she had taken it into account: [108].

The Company’s appeal to the Upper Tribunal succeeded, on the ground (shortly stated) that there had been no proper consideration by HMRC of the question of proportionality in making either of the review decisions under appeal. On that basis, the Upper Tribunal was satisfied that it could not be said that either decision would inevitably have been the same, had proper consideration been given to the question. Accordingly, the 2016 FTT had erred in law in concluding that the review decisions should be upheld. In those circumstances, the Upper Tribunal decided that the appropriate course was to re-make the decision of the 2016 FTT. Having regard to the limited jurisdiction provided by section 16 of the Finance Act 1994, this should be done by setting aside the decisions under appeal and directing HMRC to undertake a further review of each of them: see the UT Decision at [98] to [100].

Since the only challenge in this court is to two aspects of the terms on which the Upper Tribunal directed these further reviews to be carried out, I will not set out the reasoning which led the Upper Tribunal to conclude that the decision of the 2016 FTT should be set aside. It is enough to say the Upper Tribunal was satisfied that no proper consideration had yet been given, in either appeal, to the issue of proportionality,

I do, however, need to set out what the Upper Tribunal said when giving its directions as to the findings of fact that HMRC should take into account in conducting the further reviews of the relevant decisions:

We turn first to the FTT’s equivocal findings regarding the question as to whether the Company was involved in the laundering of fuel at [69] and [70] of the FTT 2016 Decision…

We agree with Mr Patchett-Joyce [counsel appearing then, as now, for the Company] that, having found on the balance of probabilities that the Company’s witnesses were honest and having made a corresponding finding in respect of the Company’s “evident honesty in its dealings with HMRC” and that therefore it could not find, on the balance of probabilities that the Company itself had been involved in the laundering of fuel, then it should have let matters rest at that point. Therefore, and also bearing in mind, as we stated at [89] above that HMRC’s case is not dependent on whether the Company itself has been involved in the laundering of fuel, the new reviews should be conducted on the basis that the Company was not involved in the laundering of fuel. Similarly, with respect to the question as to whether there was a finding in the 2014 FTT decision that any of the vehicles had been involved in the laundering of fuel, the review should proceed on the basis that none of the Company’s vehicles had been involved in the laundering of fuel.

It follows from our conclusions at [104] above that, in the light of the FTT’s findings as to the honesty of the company and its witnesses, in carrying out the reviews HMRC should take no account of the observation at [106] of the FTT 2016 Decision that it has not been demonstrated that at the time of the seizures in 2009 that everyone at the Company was unaware of the presence of contaminated fuel. The FTT did not find, on the balance of probabilities, that any member of staff of the Company was aware of that fact and the reviews should therefore proceed on that basis.”

The Upper Tribunal then gave further directions relating to two matters which are not challenged by HMRC, before turning to the version of Public Notice 192 which should be applied in the context of the further review of the Revocation Decision:

For the avoidance of doubt, I should make it clear that HMRC do not challenge the permission given to the Company to adduce further evidence relating to its due diligence policies and procedures in place in June 2013, if we conclude that the further reviews directed by the Upper Tribunal should not be disturbed on the basis of either ground 1 or ground 2 of the grounds of appeal.

It is convenient to begin with ground 2, since it applies to the further reviews of both the Restoration Decision and the Revocation Decision. The effect of the direction in [104] of the UT Decision is that “the new reviews should be conducted on the basis that the Company was not involved in the laundering of fuel”. Similarly, the reviews must proceed “on the basis that none of the Company’s vehicles had been involved in the laundering of fuel” (ibid), and they must disregard the possibility that any member of the Company’s staff was aware of the presence of contaminated fuel: [105].

Counsel for HMRC (Jonathan Kinnear QC, who did not appear below, leading Matthew Donmall) submit that the imposition of these requirements would be perverse in a situation such as the present, when no proper explanation has ever been given for the presence of laundered oil on the Company’s premises. It would risk requiring HMRC to consider the Company as “fit and proper” in that respect, even if HMRC were not so satisfied. In particular, the Company has never provided any explanation at all for the presence of laundered fuel within the white bulk storage tank, or in the four vehicles.

HMRC submit that the 2016 FTT directed itself correctly at [60] of the 2016 FTT Decision, quoted at [39] above. There are in principle three possible positions that may be adopted by a fact-finding tribunal in respect of an RDCO and its involvement in the laundering of oil detected at its premises and in its vehicles. First, it may be asserted by HMRC, and found to be proved, that the RDCO was fraudulently involved in the laundering of the oil. Secondly, the RDCO may assert that it was innocent of any involvement, and having heard the evidence the tribunal may agree. Thirdly, however, the tribunal may find itself unable to reach a positive conclusion either way, being persuaded of neither the RDCO’s guilt nor its innocence. This third position was the one adopted by the 2016 FTT, and was reflected in its findings at [68] to [70] quoted at [42] above.

In his oral submissions to us, Mr Kinnear QC also submitted that the Upper Tribunal had in effect wrongly reversed the burden of proof by requiring the further reviews to be undertaken on the basis that the Company was innocent of fuel laundering, when the onus was instead on the Company to establish that the original Restoration and Revocation Decisions were ones that HMRC could not reasonably have arrived at. Given the presence, now admitted, of laundered fuel in both the storage tank and the running tanks of the seized vehicles, there is at the lowest a heavy evidential burden on the Company to show that this occurred without its knowledge or involvement, and that it had appropriate policies and procedures in place both to prevent such contamination from happening in the first place and to prevent any recurrence of it. Since the Company has yet to adduce detailed evidence of its due diligence practice and procedures, HMRC would be unfairly handicapped if they had to consider and evaluate that evidence on the basis that the Company was not involved in fuel laundering at any material time. Similarly, consideration of a very important aspect of the Company’s RDCO status would be foreclosed if HMRC were obliged to review the Revocation Decision on that basis, whatever the further evidence adduced by the Company might show about the robustness or otherwise of its internal procedures.

Counsel for the Company (Mr Patchett-Joyce, leading Oliver Powell) submit that the Upper Tribunal came to the right conclusions for the right reasons. They emphasise the narrow scope of the two grounds of appeal, confined as they are to the basis on which the further reviews should be carried out. There is no suggestion that the Upper Tribunal

Counsel submitted that the 2016 FTT expressly considered the question of the Company’s involvement in laundering fuel, and deliberately stopped short of making any positive finding of involvement on the part of the Company. In those circumstances, the further review directed by the Upper Tribunal can only properly take place on the express footing that the Company was not so involved. It would be perverse and irrational to leave open the third position canvassed by HMRC, when the 2016 FTT, after hearing all the evidence, declined to make any positive finding that the Company was involved. Where the fact-finding tribunal has reached such a conclusion, the benefit of it must be given to the party which is at risk of losing its RDCO status and is seeking the restoration of its seized property. To do otherwise would permit the review of HMRC’s administrative decisions on those issues to proceed on a basis (namely the possible involvement of the Company in laundering) which no judicial body has found to be established.

In support of these submissions, counsel referred us to the decision of this court in Top Brands Ltd v Sharma [2015] EWCA Civ 1140, [2017] 1 All ER 854, on appeal from the decision of Judge Simon Barker QC (sitting as a judge of the High Court) in the Chancery Division: [2014] EWHC 2753 (Ch), [2015] 2 All ER 581. The case concerned a claim by creditors of a company in creditors’ voluntary liquidation (MML) against the former liquidator of the company, subsequently removed, who had made payments to various third parties allegedly involved (together with MML) in a joint enterprise VAT fraud. It was claimed that these payments had been made negligently and/or in breach of the liquidator’s fiduciary duties. For present purposes, the potential relevance of the case lies in the fact that the trial judge declined to make any positive finding that the relevant parties had been involved in a joint VAT fraud, while observing that on the available material such a possibility could not be rejected as fanciful or unrealistic: see the judgment of the High Court at [8], [2015] 2 All ER 581 at 585g-h.

On the former liquidator’s appeal to this court, the leading judgment was delivered by Sir Terence Etherton C, with whom Christopher Clarke and Lloyd-Jones LJJ agreed. In the context of a possible defence of illegality (at a time before the law had been clarified by the Supreme Court in Patel v Mirza) the Chancellor made the following observations at [47], upon which counsel for the Company now rely:

In considering these submissions, it is in my view necessary to look separately at the Restoration Appeal and the Revocation Appeal. I will begin with the Restoration Appeal.

In the context of the Restoration Appeal, I start by reminding myself of the very limited nature of the grounds upon which HMRC’s original review decision not to restore the seized vehicles may be challenged on appeal under section 16(4) of the 1994 Act. It has to be established that the decision is one which HMRC (or the person making the decision) “could not reasonably have arrived at”. The Upper Tribunal has now ruled that this condition is satisfied, because (put shortly) no adequate proportionality review has yet been conducted. The Upper Tribunal has given permission for the Company to provide HMRC with evidence regarding its due diligence policies and procedures applicable in June 2013: see [110] and [111(3)]. The Upper Tribunal has further directed that the new review must proceed on the basis of the findings of fact made by the 2014 FTT and the 2016 FTT, subject only to the qualifications set out in the UT Decision at [103] to [108].

That is the context in which the impugned directions in [104] and [105] fall to be considered. In that context, I do not consider that cases like Top Brands v Sharma provide a helpful analogy. The position here is very different. The main further question which has to be considered is one of proportionality, having regard to the findings of fact already made in 2014 and 2016, and in the light of the new evidence which the company now has permission to adduce. In those circumstances, and with all respect to the very experienced members of the Upper Tribunal, I find it hard to understand what justification there could be for requiring the review to proceed on the basis that the Company was not involved in laundering of fuel, when the 2016 FTT deliberately decided to leave the question open. Speaking for myself, I would not regard the findings of the 2016 FTT on this issue as “ambivalent” in any objectionable sense. They clearly considered the evidence with great care, and found themselves unable to reach a firm conclusion one way or the other. That was in my opinion a perfectly tenable position for the 2016 FTT to adopt, in circumstances where the question which they had to consider was whether the Restoration Decision could not reasonably have been made. In my view, fairness requires that the further review directed by the Upper Tribunal should start from the same inconclusive position, and HMRC should be at liberty both to consider the fresh evidence advanced by the Company, and to review the history of the matter from 2009 to 2013, without being obliged to assume from the outset that the Company was not involved in the laundering of fuel.

In reaching this conclusion, I do not find it helpful to rely on the burden of proof, nor would I accept HMRC’s submission that the burden of proof has somehow been reversed by the Upper Tribunal. The test which an appellate tribunal has to apply under section 16(4) is one of satisfaction that the decision in question could not reasonably have been arrived at. There is no longer any question but that this test has been satisfied, because neither ground of appeal directly challenges the Upper Tribunal’s decision to allow the Company’s appeal and direct a further review. The relevant question is, rather, about the terms on which the further review is to be conducted, and in that context I can see no good reason (as I have sought to explain) for not starting from the carefully calibrated stance taken by the 2016 FTT in the light of the evidence then available to it.

For these reasons, I am satisfied that the Upper Tribunal did err in principle, and therefore in law, by directing the further review of the Restoration Decision to be conducted on the footing that the Company was not at any material time involved in the laundering of fuel.

For similar reasons, it seems to me that the same conclusion must also be reached in relation to the Revocation appeal. Whichever version of Notice 192 falls to be applied, HMRC will be conducting the review with one hand tied behind their back if they are prevented from considering and evaluating evidence which tends to suggest involvement by the Company in fuel laundering. They are not free to go behind the findings of fact already made in 2014 and 2016, save in the uncontroversial respects indicated by the Upper Tribunal, but that is a very different matter from requiring HMRC to review the evidence afresh, together with the due diligence evidence presented to it by the Company, without an open mind on the issue of the Company’s possible involvement.

I should add that neither side invited us to distinguish between the Restoration and the Revocation appeals on this issue, and although I have considered them separately, since they are conceptually distinct, I do not ultimately think that it would be realistic to adopt differing approaches to them on a matter which goes to the heart both of the question whether the seized vehicles should be returned and the suitability of the Company to be an RDCO.

I will deal with this ground of appeal briefly, because the Upper Tribunal was in my judgment clearly correct to direct the further review of the original Revocation Decision to be conducted on the basis of the version of Public Notice 192 which was in force at the date of the original decision, that is to say the February 2013 version.

In arguing for the contrary conclusion, counsel for HMRC submit that it makes no sense to conduct such a review by reference to a policy that is no longer in force. It would have the effect, they say, of the Company being judged by a different (and possibly lower) standard than is now applied to other traders. Accordingly, if HMRC have to make what amounts to a fresh decision, they should do so on the basis of the policy prevailing at the date when the review is conducted, that is to say the post-May 2014 version. This would mean, in particular, that HMRC would be obliged to consider whether the Company now satisfies the “fit and proper” test articulated in paragraph 4.3 of the May 2014 version. Those are the criteria which would be applied to any fresh application by a trader for RDCO status, or to a decision to cancel approval as an RDCO in the light of new information that has come to the attention of HMRC under paragraph 4.8. It would make no sense, say HMRC, to require the further review of the Revocation Decision to be conducted in accordance with a policy which is now obsolete.

The short answer to these arguments, in my judgment, is that the Upper Tribunal only had jurisdiction to require HMRC to conduct “a further review of the original decision”: see section 16(4)(b) of the 2004 Act. It did not have jurisdiction to require a fresh review to be undertaken in the light of circumstances and published guidance in force at the date when the review is carried out. The original Revocation Decision was made by reference to the published guidance contained in the then current form of Notice 192,

Although, as I have said, I consider the answer to this question to be clear, I confess that I have some difficulty with the reason given by the Upper Tribunal in [110] for reaching the same conclusion. The Upper Tribunal’s reason, it will be recollected, was that “it would be appropriate to apply the previous policy because the sole reason HMRC gave for the revocation was the need to protect the revenue, which continues to be a ground for revocation under the new policy.” If this was indeed the Upper Tribunal’s only reason for so concluding – and they do not expressly mention any other reason – I would respectfully have to disagree with it. The mere fact that protection of the revenue is a ground for revocation under both versions of the Notice seems to me irrelevant. The real reason, to my mind, lies in the limited nature of the Upper Tribunal’s jurisdiction under section 16(4), and the general principle that when a decision is reviewed, the review should be conducted by reference to the facts as they existed, and the law as it stood, at the date of the original decision. That is the critical distinction between the review of a previous decision, on the one hand, and the taking of an entirely fresh decision, on the other hand.

I should also mention that, in the course of his oral submissions, Mr Kinnear QC sought to derive some support from the wording of regulation 10 of the Hydrocarbon Oil (Registered Dealers in Controlled Oil) Regulations 2002, which states that:

contravened any provisions of these Regulations, or

failed to comply with any condition, restriction, or requirement prescribed by the Commissioners under these Regulations, the extent to which he has followed any current guidance issued by the Commissioners must be taken into account.”

For the reasons which I have given, I would allow HMRC’s appeal on ground 2, but dismiss their appeal on ground 1. If the other members of the court agree, this means that the further reviews of the original decisions will have to be undertaken on the basis directed by the Upper Tribunal, but with the omission of the directions given in the UT Decision at [104] and [105].

My conclusion also means that, despite the deplorably long period during which this matter has remained unresolved, the further reviews will still have to take place unless the parties are able to come to an agreement. Even if HMRC’s appeal had succeeded on both grounds, I think it would have been very difficult, if not impossible, for us to accede to the submission that the further reviews could safely be dispensed with on the ground that the outcome would now inevitably be to uphold the original decisions. Indeed, Mr Kinnear barely mentioned this possibility in his oral submissions, although he did not formally abandon the point. But in circumstances where the Revocation Decision must now be reviewed by reference to the February 2013 version of Notice 192, and where the Company has permission to provide HMRC with fresh evidence on due diligence, it seems clear to me that we are in no position to pre-empt the outcome of either review.

I agree.

I also agree.