The Commissioner of Police of the Metropolis, R (on the application of) v The Police Appeals Tribunal

This is a claim for judicial review brought by the Commissioner of Police of the Metropolis challenging the 30 September 2021 decision of the Police Appeals Tribunal (“PAT”) to quash the 24 July 2020 findings of the Misconduct Panel (“the Panel”) that the allegations against the first and the second interested parties were proven and amounted to gross misconduct. The PAT substituted the Panel’s decision with a finding that the allegations were not proven. The PAT has played no active part in these proceedings, which are contested by the two interested parties. Permission to apply for judicial review was granted on the papers by Cutts J in a decision communicated on 26 April 2022.

The allegations arose from a road accident on 4 December 2016 involving a marked police vehicle. PC Michel was the driver of the police car and PC Charnock was the operator. PC Michel proceeded against a red light signal and caused a collision with a vehicle driven by a member of the public who was proceeding in accordance with a green light. In the aftermath of the accident both officers gave accounts that later turned out to be inaccurate. In particular, both officers had indicated that their vehicle moved forwards in accordance with a green light in their favour. Subsequent investigations established this was not the case and the officers accepted as much. The disciplinary charges alleged breaches of the Standards of Professional Behaviour relating to honesty and integrity and to discreditable conduct. The key question for the Panel was whether the officers’ inaccurate accounts were the product of deliberate lies or honest mistakes.

The Claimant’s four grounds of challenge are as follows:

The road traffic accident was captured on CCTV. Information was also obtained from the police vehicle’s Incident Data Recorder. The police vehicle was on the Uxbridge Road travelling south. This becomes the Hampton Road, West after the road crosses the junction with the Hounslow Road (“the junction”). The accident occurred at about midday in the middle of the junction. The police car collided with a VW Polo being driven along the Hounslow Road by Raj Mehra, who had the right of way at the time. The junction is a crossroads controlled by automatic traffic signals (“ATS”), which were functioning correctly on the day.

As the police vehicle approached the junction, PC Michel accepted an immediate response call requiring the officers to attend a suspect vehicle in the vicinity. He then came to a stop at a red light at the junction. The police vehicle was the first car stopped at these lights. After waiting for a period of time, PC Michel drove forwards before the traffic light had changed and whilst it was still showing red. Around a second later he activated his front and rear blue lights and headlamp flash, but the siren was not switched on. The vehicle moved off at a slow rate before colliding with the other car. The nearside of the police vehicle hit the offside of Mr Mehra’s car. Mr Mehra had to be cut from his vehicle and was taken to hospital with minor injuries.

Very shortly after the accident PC Michel reported to the police operator via his radio that he had had a “PolCol”, saying he had proceeded through a green light.

PS Harvey was the garage sergeant responsible for the area on the day. He attended the scene. He asked PC Michel what had happened. PC Michel said he had been stationary at a red traffic light and had waited for the lights to turn green before activating his warning equipment and proceeding; that there were two vehicles in front of him, which moved to the side, allowing him to proceed; and once through the junction the VW Polo had gone through a red light and collided with his vehicle.

PS Harvey then spoke with PC Charnock, asking him whether the version of events relayed to him by PC Michel was correct. PC Charnock indicated that he agreed with PC Michel’s account.

Mr Mehra was reported for driving without due care and having no insurance.

PC Michel made a witness statement setting out his account of the accident dated 9 December 2016. He signed this indicating that the contents were true to the best of his knowledge and belief. His account included: that he was stationary at the lights in the nearside lane with two stationary vehicles in front of him; due to the traffic lights being red he decided to wait for the phasing to turn green before moving off and activating his blue lights; when the light turned from red to green and the two vehicles in front of him began to move off, he moved into the offside lane; and the only apparent factor that caused the collision was the other driving having run a red light.

PC Charnock made a witness statement dated 6 December 2016 setting out his account of the incident. He signed it indicating that the contents were true to the best of his knowledge and belief. His account included: the traffic lights governing their junction were showing red; the police vehicle stopped behind two cars that were in front of them; when the lights turned green, the two vehicles in front filtered left and the police vehicle drove straight ahead; and a grey VW Polo then approached from their left hand side at speed.

It is unnecessary for me to detail the subsequent investigations, since neither officer challenged that it was thereby established that the police vehicle had in fact been at the front of the junction when the light was red and not behind any other vehicle; and that it had proceeded through a red light. It also followed that Mr Mehra had driven forwards in accordance with a green light.

Notices serviced pursuant to regulation 21 of the Police Conduct Regulations 2012 (2012/2632) (“the 2012 Regulations”) alleged that both officers had breached the Standards of Professional Behaviour set out in regulation 3 and schedule 2, namely (i) honesty and integrity; and (ii) discreditable conduct. The notices also contended that the breaches were so serious that dismissal could be justified.

The allegations in respect of PC Michel were:

The allegations in respect of PC Charnock were:

PC Michel and PC Charnock denied all of the allegations in their respective Regulation 22 Notices. PC Michel denied saying anything to fellow officers or in his earlier witness statement that he knew or believed to be false. As regards his account of proceedings on a green light, he said that he now realised that he had mistakenly reacted to seeing the secondary lights ahead that controlled a pedestrian crossing and which were green. In so doing he had experienced the recognised phenomenon of “see through” (looking ahead to a different set of lights and mistaking them for the relevant ones) whilst under the pressure of answering an I grade call. The notice pointed out that the secondary green light could be seen in photos taken on the day and that “see through” had been a recognised problem at this junction.

As regards his account of there being two cars in front of his vehicle when he was stopped at the red light, PC Michel said that he still had this picture in his head, but human memory and recall “can be fragile, misleading and unreliable yet honest”. The Notice said that there were a number of ways in which this mistaken recollection may have been formed. It observed that memories were open to change and distortion, as memory is a fragmentary and reconstructive process, which can be distorted or altered by reference to outside sources. When PC Michel spoke to PS Harvey he gave no details as to what the “two cars in front” meant and he may have been referring at that stage to two vehicles which passed in front of him onto Hampton Road West from the Hounslow Road (as shown on the CCTV footage); or he could have been drawing on one of the many other occasions when he had driven that junction.

PC Michel also said that he had no motive for giving an untrue account; that as a police response driver he was trained to look further ahead in the road; and he had stopped at the red light, so there would have been no point in failing to wait for it to change.

In his Regulation 22 Notice, PC Charnock said that he accepted that at the scene he had agreed with what was put to him by PS Harvey, but that this was based on his honest belief as to the events. To the extent that his recollection proved wrong, it was the result of honest errors. He also denied giving a knowingly false account in his statement. He emphasised that immediately prior to the accident he was primarily focused on operator tasks rather than the road conditions; and that as they had approached the junction he perceived that there were two vehicles in front of them and when he next looked up they were gone, leading him to believe they had turned left at the junction; and he saw a green light which he now believed was a see through error to the lights at the pedestrian crossing on the other side of the junction.

Given the central role that his evidence played in the reasoning of the PAT, it is necessary for me to refer in some detail to the report dated 25 May 2020 prepared by Professor Christopher French. Professor French is a Professor of Psychology at Goldsmiths, University of London and a member of the Scientific Advisory Board of the British False Memory Society. He has given expert evidence in several cases involving allegations of sexual abuse and at the Gross Misconduct hearing concerning officers involved in the death of Sean Rigg.

In section 3 of his report, Professor French described his instructions in the following terms:

In section 4 of his report, Professor French listed the materials he had seen. He set out his opinion in section 6 of the report. It included the following:

The Appropriate Authority’s (“AA”) case at the hearing was presented by Mr Thacker. PC Michel was represented by Mr Baumber and PC Charnock by Ms Williamson. The Panel heard evidence on 20, 22 and 23 July 2020 and submissions on 24 July 2020.

The witnesses called by the AA included Mr Mehra, PS Harvey and PC Francis. PC Francis had taken some photographs whilst standing at the junction. A zoomed in image from one of these photographs was particularly relied upon by the officers as supporting the “see through” phenomenon. PC Archer, who had analysed the IDR, gave evidence and the Panel adopted the sequence and timings of events set out in his report. The Panel also heard from Mr Scriven who is an expert in traffic signals. He described how louvered hoods had been added at this junction in October 2000 to reduce the acknowledged risk of “see through”. He accepted that from the position of the officers in their police car, they may well have been able to see a green light from the secondary traffic light over the junction, although it would have been dimmer than was shown in the photograph.

Both PC Michel and PC Charnock gave evidence and were cross examined on the basis that they had given deliberately untrue accounts in the aftermath of the accident. Professor French was called by the officers. He gave evidence in line with his report. He also explained how “memory traces” are laid down and the concept of “inattention blindness”. He said that the latter was “a posh way of saying we only actually take in information that we pay attention to at the time the event is taking place”. He also indicated that there was “lots of evidence that if you don’t take the information in in the first place you’re not going to be able to accurately recall it…If you’re not looking for things you often don’t see them even if your eyes are pointing directly open”.

The Panel also viewed the CCTV footage of the accident. On 21 July 2022 the Panel members went on a site visit to the junction. As set out at para 20 of the Panel’s decision, they took into account the fact that the alleged “see through” light had been subjected to alteration since the incident.

The Panel found that all of the allegations were proven, that the conduct in question amounted to gross misconduct and that the appropriate sanction in relation to both officers was dismissal without notice.

The Panel’s written decision was dated 24 July 2020. The Panel correctly directed itself that the burden of proving the charges was on the AA and that the applicable standard of proof was the balance of probabilities (para 4). At para 5 the Panel set out the staged approach that it had adopted. At para 7 the Panel listed what it had taken into account; this included all of the material submitted and the evidence that had been heard. It then said at para 8 that whilst it had taken all this into account, it would “only refer to what it considers to be the essential elements when giving its findings”. At para 9 the Panel listed the witnesses it had heard from, noting that it had “been able to assess their evidence” and that this gave it “a unique position”.

After summarising some of the evidence given by the witnesses called by the AA, the Panel referred to PC Michel’s evidence, indicating at para 30:

The Panel then said that it found the charges that PC Michel faced were proved (para 31). It is necessary to set out its reasoning in para 32 in full:

In para 33 the Panel addressed Professor French’s evidence. Again it is necessary to set its reasoning in full:

The Panel then discussed the account of an eye-witness to the accident, Mr Doyle, who was plainly mistaken in various elements of his recollection, including that Mr Mehra’s vehicle was stationary when the police car collided with it. PC Michel and PC Charnock’s representatives had highlighted this as an example of an innocently given erroneous recollection. The Panel noted that he was at a different location to the officers and had a different perception and in any event “taking into account all of the evidence against PC Michel the Panel find the allegations against him proven” (para 34). The Panel also indicated that it had taken into account the character references that PC Michel had provided (para 35).

The Panel then expressed a similar conclusion in respect of PC Charnock, finding that he was not “a credible or reliable witness” and that the allegations he faced had been proven on the balance of probabilities (paras 37 and 38). The Panel listed its reasons for coming to these conclusions in a series of sub-paragraphs at para 39. The contents largely replicated sub-paragraphs (d) – (n) of para 32 in relation to PC Michel, with the Panel making equivalent findings in respect of PC Charnock. The Panel also found: that the junction was a place PC Charnock knew well from passing through it as a result of his police duties (sub-para (a)); and that “PC Charnock at some point post collision heard PC Michel lie and whether it be through misguided loyalty or some other reason dishonestly and deliberately supported PC Michel’s deceit”. The Panel repeated its earlier observations in respect of Professor French and Mr Doyle (sub-paras (t) and (u)) and indicated it had taken the character references into account (para 41).

Accordingly the Panel found that both officers had breached the Honesty and Integrity standard by “not being truthful, lying and making false, misleading or inaccurate oral or written statements in their role as serving police officers” and that this also amounted to Discreditable Conduct (paras 42 and 43).

The officers’ appeal to the PAT was heard on 30 September 2021. The representation by counsel was the same as before the Panel. After hearing oral submissions, the PAT deliberated and then its decision was announced by the Chair. Written reasons were provided dated 4 October 2021.

After setting out the allegations and making reference to the Panel’s conclusions, the PAT set out the terms of rule 4(4) of the 2012 Rules and the relevant Standards of Professional Behaviour (para 8). The Panel then turned to the submissions. Mr Baumber’s submissions were described at paras 9 – 20. They were summarised in para 9 as “the findings of the Panel were unreasonable, as it failed to deal with the evidence supporting the view that the first appellant’s account had been truthful, and it made findings without adequate reasoning”. The PAT then referred to Ms Williamson’s submissions at paras 21 – 27, indicating that she said the Panel’s decision was unreasonable because there was unreasonable and inadequate consideration of: (a) the evidence of collusion; (b) the expert evidence of Professor French; and (c) the fact that PC Charnock was the operator of the vehicle (rather than the driver). Mr Thacker submitted that the appeal should be dismissed; and his contentions were set out at paras 28 – 40. At para 41 the PAT noted that it had been provided with the documentation and other material available to the Panel, together with the transcripts of the hearing, the Panel’s decisions, the grounds of appeal and the AA’s response.

The PAT made a number of general observations at paras 42 – 45 and then at para 46 summarised some of the conclusions that it would go on to detail. As material, these paragraphs said:

At paras 47 – 63 the PAT addressed the Panel’s conclusions in respect of the evidence of Professor French. After referring to the contents of Professor French’s report, his evidence and the Panel’s reasons for finding that it was of “little assistance”, the PAT set out its conclusions in respect of this area as follows:

After citing a passage from paras 66 – 68 the judgment of Leggatt J (as he then was) in Blue v Ashley (Rev 1) [2017] EWHC 1928 (Comm) about the potential unreliability of memory, the PAT’s analysis continued as follows:

The next section of the PAT’s decision appeared under the heading “Credibility” (paras 64 – 75). At para 64 the PAT said:

The PAT then summarised some of the Panel’s reasoning, indicating:

The PAT then listed evidence to which it said the Panel had given “no consideration”. These were that: the officer had stopped at the red light although he did not have to; he had pulled away after waiting at the red light for about 26 seconds; he was anticipating that the lights would change to green at any moment; he pulled away smoothly at a modest speed; he used his blue lights but not his siren; and he had said if he was found to have caused the collision he might get points on his licence but would be unlikely to lose his driver status.

The PAT cited paras 63 – 64 from Nicklin J’s judgment in R (Dyfed Powys Police) v Police Misconduct Tribunal [2020] EWHC 2032 (“Dyfed Powys”) (see para 73 below) before setting out its conclusion as follows:

Between paras 76 and 92, the PAT set out its findings in respect of the evidence. The majority of the findings listed in para 76 related to the circumstances of the accident and thus were in keeping with the Panel’s earlier findings. The PAT also included the following points:

The PAT said that they had found the expert evidence of Professor French “to be of assistance…His evidence was particularly helpful, in that it explained how a memory can be constructed of pieces of information, some of which are unwittingly incorporated into the memory by suggestion or assumption, as well as direct observation” (para 82). The PAT noted that PC Charnock had been multi-tasking at the material time (para 83). The PAT went on to say that it found the evidence of the officers to be credible and reliable. It explained this as follows:

The PAT made further reference to Professor French’s evidence about memory conformity in relation to PC Charnock, saying:

The Police Appeals Tribunal Rules 2020 came into force on 1 February 2020. However, the 2012 Rules continued to have effect in relation to appeals brought under the 2012 Regulations (made pursuant to the power in section 85(1) of the Police Act 1996).

The effect of Rule 4(1) and 4(2)(a) of the 2012 Rules is that a police officer “against whom a finding of misconduct or gross misconduct had been made at a misconduct hearing” may appeal to a tribunal in reliance on one or more of the grounds referred to in para (4). In turn, para (4) states:

As I have indicated, the appeal in this case was solely brought under limb (4)(a). There is only a power for the PAT to hear witness evidence in respect of appeals brought under limb (4)(b). Rule 22(1) provides that the tribunal “shall determine whether the ground or grounds of appeal on which the appellant relies have been made out”. Significantly, the PAT only has the power to set aside the relevant decision and remit the matter for it to be decided again where limb (b) or (c) of rule 4(4) has been made out. There is no power to remit a case in relation to a successful limb (4)(a) appeal.

In addition to there being no power to remit, at least in the ordinary course of events, the AA has no power to bring further disciplinary proceedings in respect of the same allegations if a rule 4(4)(a) appeal against findings of misconduct / gross misconduct succeeds. In Chief Constable of Nottinghamshire Police v R (Trevor Gray) [2018] EWCA Civ 34 (“Trevor Gray”) Sir Terence Etherton MR said at para 73:

By virtue of section 85(2) of the Police Act 1996, a PAT “may on the determination of an appeal under this section, make an order dealing with the appellant in any way in which he could have been dealt with by the person who made the decision appealed against”.

Given the finality for the disciplinary proceedings if a PAT concludes that the Panel’s finding of misconduct / gross misconduct was “unreasonable” it is important that the correct test is applied and that the test does not become diluted in its application because a PAT disagrees with aspects of a Panel’s reasoning or feels that it would have arrived at a different conclusion if it had been sitting as the tribunal of fact.

Counsel agreed with my proposition that the rule 4(4)(a) reference to “the finding” being unreasonable must be a reference to the “finding of misconduct or gross misconduct” in relation to which rule 4(2)(a) (or, as the case may be, rule 4(2)(b) or 4(2)(c)) confers a right of appeal.

The courts have considered the meaning of “unreasonable” in the context of rule 4(4)(a) (and its predecessor provision) on a number of occasions. In R (The Chief Constable of Wiltshire) v Police Appeals Tribunal and Woollard [2012] EWHC 3288 (Admin) (“Wiltshire”) Wyn Williams J considered the earlier authorities and identified the correct meaning as follows:

In R (Chief Constable of the Derbyshire Constabulary) v Police Appeals Tribunal [2012] EWHC 2280 (Admin) (“Derbyshire”) at para 37, Beatson J (as he then was) had decided that the PAT had erred in expressing in unqualified terms that the test to be applied was the Wednesbury test. However, he said that the PAT was “essentially correct” in describing it as “Wednesbury-type test” and identifying the question for them as “whether the decision on finding or outcome was within the range of reasonable findings or outcomes to which the Panel could have arrived” (para 37).

In Collins J’s earlier decision in R (Montgomery) v Police Appeals Tribunal [2012] EWHC 936 (Admin) (“Montgomery”) he said at para 18:

Accordingly, consistent with this case law and the consequences of upholding a rule 4(4)(a) appeal, in determining whether a Panel’s finding of misconduct / gross misconduct was “unreasonable” within the meaning of rule 4(4)(a):

Proceedings in the Administrative Court challenging the decision of a PAT arise by way of a claim for judicial review, rather than as a second appeal. Accordingly, the claimant must establish a public law error before the PAT’s decision may be quashed: R (Chief Constable of Dorset) v Police Appeals Tribunal [2011] EWHC 3366 (Admin) per Burnett J (as he then was) at para 19.

In the same case at para 33, Burnett J observed that the court should be slow to interfere with the decision of an expert tribunal, given that the court does not share that expertise. He described this as “a proper recognition of the need for caution before disagreeing with someone making a judgment on a matter for which he is especially well qualified, when the court is not”. However, in R (Wilby-Newton) v Police Appeals Tribunal [2021] EWHC 550 (Admin) at para 86 Julian Knowles J explained that these observations applied with particular force where, for example, the court was asked to overturn a particular sanction imposed for misconduct, as the Panel and the PAT are better placed than the court to determine what is required to maintain confidence in policing. But that “[t]hey may apply with lesser force…where the Court is asked to intervene because of an alleged error of law by the PAT…In such a case there is less need for caution or ‘deference’ because the Court is the ultimate arbiter of what is lawful”.

Ms Checa-Dover submits that in deciding that the Panel’s findings of gross misconduct were “unreasonable”, the PAT failed to direct itself in accordance with or apply the requisite test, namely whether they were outside of the range of reasonable findings that the Panel could have arrived at. Mr Baumber and Ms Williamson dispute this and submit that the PAT’s reasoning shows that it applied the correct test.

The PAT did refer to the terms of rule 4(4)(a) (para 36 above) and it made frequent references to the Panel’s approach, reasoning and/or conclusions being “unreasonable”, for example at paras 42, 59, 60, 62, 63, 64, 71 and 74 of its decision (paras 37, 39 – 41 and 43 above). It is common ground that the PAT’s decision contained no express self-direction as to the meaning of “unreasonable” in this context. Mr Baumber submits that it was unnecessary for the PAT to do more than this as “unreasonable” is an ordinary word and the meaning is “obvious”. I do not accept that submission. As shown by the discussions in the case law (paras 54 - 56 above) it is apparent that the word “unreasonable” may bear a number of different interpretations. By way of example, absent authority to clarify the meaning, an unreasonable decision could mean one that is unreasonable in the Wednesbury public law sense; one that is outside the range of findings that a Panel could reasonably come to; or one that the PAT disagrees with.

In the absence of a self-direction on this subject, it is necessary to examine the PAT’s reasoning in some detail to see what test it actually applied in practice. If the correct test was applied by the PAT, then failure to identify the test in its decision would not constitute a material public law error (although identifying the test that is being applied is always desirable, both in the interests of clarity and because it helps to focus the decision-maker’s mind). On the other hand, application of an incorrect interpretation of “unreasonable” would be a material misdirection and a public law error.

Before I come to PAT’s reasoning I will consider what was said to the PAT about the test that it should apply, since all counsel sought to derive some support from this in their submissions to me.

None of the written submissions from counsel that were filed in advance of the hearing, and which were available to the PAT when it was pre-reading the documents, made reference to the range of reasonable findings test. (I was told by counsel that the usual expectation is that the PAT will have pre-read the materials in advance of the hearing.) It would be desirable for the PAT to be given this kind of assistance, not least because two of the three tribunal members will have a police background; only the Chair is a lawyer.

Only Ms Williamson’s written submissions addressed the nature of the rule 4(4)(a) test. They did so by setting out an incomplete citation from para 18 of Collins J’s judgment in Montgomery (para 56 above) with underlining included to add emphasis, as follows:

In terms of counsels’ oral submissions, Mr Baumber informed the PAT that “the test is not one of Wednesbury unreasonableness. It is a much lower hurdle than that”. He made no reference to the range of reasonable findings test. Regrettably, this was not an accurate way of characterising the approach that the PAT should take. Ms Williamson did not address the PAT orally on the test that it was to apply. The PAT was then advised of the range of reasonable findings test by Mr Thacker, who said:

I return to the crux of Ground 1, namely the reasoning employed by the PAT. Having reviewed the decision, I am clear that the PAT did misdirect itself and failed to apply the range of reasonable findings test. The combined weight of the points I identify in the paragraphs that follow have led me to this conclusion.

Firstly, as I have already indicated, the PAT’s decision contains no explicit reference to the test that it understood it was applying. Even when setting out quite a detailed summary of Mr Thacker’s submissions at paras 28 – 40, the PAT did not mention the range of reasonable findings test.

Secondly, at para 42 (para 37 above) the PAT said that counsel had set out “the ways in which they consider” the findings to be unreasonable and this “constitutes a valid basis for an appeal under Rule 4(4)(a)”. The PAT summarised the submissions made by Mr Baumber and Ms Williamson at paras 9 – 27. There is no reference in that summary to a submission that the Panel’s findings of gross misconduct were outside the range of reasonable findings that the Panel could make. Rather, it contained specific criticisms of the Panel’s approach that were characterised as unreasonable.

Thirdly, after setting out extracts from Professor French’s report at paras 49 – 53, the PAT began its analysis of this evidence by saying: “The Tribunal considered Professor French’s evidence to be directly relevant to the decision as to the appellants’ credibility.” In other words, the very starting point of the PAT’s reasoning was to express disagreement with the Panel as to the weight to be attached to Professor French’s opinion. Moreover, it is apparent from the structure and content of what followed that the PAT’s different view as to the weight to be attached to Professor French’s evidence heavily influenced its conclusion that the rule 4(4)(a) appeal test was established.

Fourthly, on several occasions the PAT used the word “unreasonable” in a different sense to the rule 4(4)(a) test that it had to apply (whether the findings of gross misconduct were outwith the range of reasonable findings the Panel could make). Ms Williamson realistically accepted that this was the case in respect of both paras 59 and 60 of the decision (set out at para 39 above). I bear in mind her submission that these were simply steps in the PAT’s reasoning and that it did also apply “unreasonable” in the rule 4(4)(a) sense when it came to its conclusion. I will address those later passages shortly, however, I note at this stage that there is no explicit acknowledgement or suggestion by the PAT that it is using the word “unreasonable” in two different senses.

Fifthly, in para 71 (para 41 above) the PAT said that it “considers it unreasonable for the Panel to find the appellants would have seen the red light at the time the first appellant drove away from the light”. However, it is unlikely that here the PAT was using “unreasonable” in the sense of indicating a conclusion that this finding was one that was outside the range of reasonable findings a Panel could make. In light of the matters I identify in paras 81 - 82 below, I do not see how the PAT could have arrived at that position. Rather, the PAT was critiquing the Panel’s reasoning and its alleged failure to take certain matters into account and expressing its own disagreement with the Panel’s factual conclusion that the officers had seen the red light at this time.

Sixthly, whilst I agree with Mr Baumber and Ms Williamson that for present purposes the crucial passage in terms of the PAT’s reasoning is para 74, I consider that its contents reinforce, rather than negate, the proposition that the PAT applied the wrong test. I have set out the PAT’s para 74 at my para 43 above. Immediately prior to this, the PAT included an extract from paras 63 – 64 of Nicklin J’s judgment in Dyfed Powys, addressing when appellate courts can interfere with findings of fact made at trial. Dyfed Powys was concerned with the Administrative Court’s powers on a judicial review of a Misconduct Panel’s decision. Within the passage cited by the PAT, Nicklin J referred to para 67 of Lord Reed JSC’s judgment in Henderson v Foxworth Investments Ltd [2014] 1 WLR 2600 (“Henderson”) where he said:

In this passage Lord Reed identified a series of disjunctive reasons that could lead to an appellate court overturning factual findings made by a lower court, including where there was a demonstrable misunderstanding of relevant evidence or a demonstrable failure to consider relevant evidence. Lord Reed was not addressing the circumstances in which a Misconduct Panel’s finding that gross misconduct was proven, will be “unreasonable” within the rule 4(4)(a) meaning. The tests are not analogous; the circumstances listed by Lord Reed encompassed circumstances that would not necessarily lead to a conclusion that the lower court’s outcome was beyond the range of reasonable findings open to it. Furthermore, an appellate court (and the court hearing a judicial review) has powers to remit the case to the fact finding / decision making body that the PAT does not have in relation to rule 4(4)(a) appeals (para 49 above).

However, despite para 67 of Henderson not being directly on point, the terms of para 74 of its decision indicate that: (a) the PAT found that two of the disjunctive situations identified by Lord Reed were established, as if this was the test it had to apply and (b) it then concluded that as a result of this, the Panel’s findings were unreasonable. The first of these propositions is apparent from the PAT’s italicised quotes in para 74, finding that in this case there was a “demonstrable misunderstanding of relevant evidence” and “a demonstrable failure to consider relevant evidence”; the passages are italicised because they are direct quotes from para 67 of Henderson. The second of these propositions is apparent from the fact that immediately afterwards the PAT said: “It was therefore satisfied that the Panel’s findings, that the appellants had lied, were unreasonable” (my emphasis added). In turn, immediately after that in para 75, the PAT said: “The Tribunal therefore allowed the appeal under Rule 4(4)(a)” (my emphasis added).

Mr Baumber submitted that the PAT had applied the part of para 67 of Henderson where Lord Reed indicated that factual findings made by the lower court could be overturned where an appellate court is “satisfied that his decision cannot reasonably be explained or justified”. The difficulty with that submission is that this is not what the PAT said in para 74. If the PAT was applying that basis for finding the Panel’s decision to be “unreasonable” it would have said so, rather than quoting other bases for overturning factual findings that were identified by Lord Reed. In any event, that is not the same as the rule 4(4)(a) test, given, for example, that it encompasses a decision that lacks reasonable explanation.

It is apparent from the PAT’s citation of Dyfed Powys and then its application of quoted passages from Lord Reed’s judgment in Henderson, that this was the test the PAT thought it had to apply, rather than the outside the range of reasonable findings test of which it made no mention.

Seventhly, if the PAT had understood its conclusion in paras 74 and 75 that rule 4(4)(a) was made out to be a decision that finding the charges proven was not an option that was within the range of findings reasonably open to the Panel, it is surprising that after reaching this point it then turned to a detailed analysis of the evidence in its paras 76 – 92, explaining why it considered that the AA had not proved that the two officers had acted dishonestly (paras 44 – 46 above). The PAT’s approach from para 76 onwards is more consistent with an understanding that its conclusion thus far on rule 4(4)(a) unreasonableness was due to identified inadequacies in the Panel’s assessment of the evidence and its reasoning, rather than a determination that the finding was not open to the Panel.

Eighthly, there are no indications that the correct test was applied, capable of offsetting the concerns that I have identified. Mr Baumber and Ms Williamson submit that in paras 53 – 74 the PAT focused on the Panel’s reasoning, rather than simply expressing disagreement with the Panel’s conclusion and substituting its own view. To some extent that is correct (albeit, as I have already indicated, there are expressions of disagreement with the Panel’s approach within these paragraphs, which the PAT appeared to treat as a significant part of its reasoning). However, in so far as the PAT’s focus was upon the Panel’s reasoning, this was, as I have concluded, examined through the prism of determining whether the Panel had misunderstood relevant evidence and/or failed to consider relevant evidence and/or adequately reasoned its decision, the PAT did not address the range of reasonable findings test.

As I indicated in para 72 above, I do not consider that the PAT was in a position to decide in its para 71, that the Panel’s conclusion that the officers were aware of the red light when PC Michel drove forwards (and thus had lied in giving accounts that the light was green) was one that was outside of the range of reasonable findings that was open to the Panel. It is clear that the PAT would have arrived at a different conclusion if it had been the tribunal of fact, for the reasons that it identified (see in particular paras 44 - 46 above). However, this was a case in which there were factors pointing in both directions. The weight to be given to any particular factor was pre-eminently a matter for the tribunal of fact. It was a classic example of a case where opposing conclusions were reasonably available to the tribunal of fact, depending upon what it made of the evidence and in particular the view it formed of the credibility of PC Michel and PC Charnock.

By way of example only, some of the factors that were cumulatively capable of supporting the Panel’s finding that, on a balance of probabilities, the officers knew that the primary lights were red when PC Michel drove forwards were as follows:

I address the Panel’s conclusion that it derived little assistance from Professor French at paras 108 – 116 below. In any event, Professor French’s expert evidence was largely concerned with the officers’ respective explanations for why they had both wrongly said in their earlier accounts that there were two cars in front of their police car when it was stopped at the junction. By contrast, both officers’ explanation for their earlier incorrect accounts that the police car moved forwards after the lights had turned to green was largely based on the simple, factual proposition that they had seen and reacted to the wrong ATS. This explanation did not rely upon nonbelieved memory, memory conformity or the other phenomena described by Professor French. In so far as the PAT reasoned in para 71 that the PAT had failed to have regard to the Professor’s opinion that “if you’re not looking for things you often don’t see them”, PC Michel, as the driver, would have been expected to be looking at the lights he had stopped at and both officers had described the ATS turning to green in their earlier accounts (and had not suggested at that stage that they were not looking at the lights).

Accordingly, for the reasons I have identified, I uphold Ground 1; the PAT misdirected itself in failing to apply the range of reasonable findings approach to the Panel’s finding of gross misconduct when it allowed the officers’ appeals under rule 4(4)(a). Accordingly, it follows that its decision to allow the appeal under rule 4(4)(a) was flawed by public law error. The PAT’s decision therefore falls to be set aside.

The claimant alleges that the PAT: (a) erred in conflating the question of whether the Panel’s finding was “unreasonable” with its substituted determination that the AA had not shown that it was more likely than not that the interested parties had acted dishonestly, and/or (b) acted unfairly in failing to allow the parties the opportunity to make submissions before it arrived at its substituted determination. Ms Checa-Dover referred to the Panel’s finding of unreasonableness as the “gateway” question and the Panel’s subsequent findings as the “substantive determination”.

I reject these propositions for the reasons that I will explain.

Given the nature of the rule 4(4)(a) test (as discussed at paras 52 – 57 above), there is a risk of causing confusion by describing a conclusion that a Panel’s finding of misconduct / gross misconduct was “unreasonable” as a “gateway”; since, in at least the majority of instances, this conclusion will mean that the end of the line has been reached for the disciplinary proceedings. This is because to permissibly uphold a rule 4(4)(a) appeal against a finding of misconduct / gross misconduct, a PAT must have already considered the evidence and decided that the Panel’s finding was outside the range of reasonable findings that a Panel could make. In turn, it would therefore usually follow from such a conclusion that the only option for the PAT would be to find that the charge/s were not proven (since it had already decided that it would be outside the range of reasonable findings to conclude otherwise). I touched on this point when I considered Ground 1 (para 78 above). However, it is also relevant to Ground 2, because, as a matter of logic, it should mean that in relation to a rule 4(4)(a) appeal, in the usual course of events there are no distinct “gateway” and “substantive determination” stages at the hearing and I do not propose to adopt that terminology.

I emphasise that I am only addressing rule 4(4)(a) appeals against findings of misconduct or gross misconduct; the position in relation to appeals against sanction does not arise in this case. (And the position in relation to appeals under rule 4(4)(b) and/or (4)(c) is distinct, as different tests apply and the PAT has the power to remit if the appeal succeeds). I have referred to what I would “usually” expect to occur, given the nature of the test. I do not suggest that this would invariably be the case; there may be circumstances that are not before me in this case that would call for a different approach. During the course of submissions I asked counsel to assist me as to when it would or might be appropriate in a rule 4(4)(a) appeal for a PAT to go on to assess the evidence at a second stage for the purposes of its substituted determination after it had already concluded that a finding of misconduct / gross misconduct was outside the range of reasonable findings open to the Panel. Ms Williamson accepted that as a matter of logic it was “very unlikely” that this second stage would arise, but she suggested two scenarios where it might do so. Firstly, where the PAT sought to arrive at findings on evidence that (although it was before the Panel) had not been addressed at all below. Secondly, where a Panel’s decision was so badly flawed that it was very obvious that it was “unreasonable” and the PAT could reach that conclusion quickly and then move on to its own analysis of the evidence. In addition, Mr Baumber noted that a PAT might want to set out its own analysis of the evidence to ensure no further disciplinary proceedings were brought over the same matters (albeit this would usually be ruled out by Trevor Gray (para 50 above)). Even if these are examples of instances where a PAT might want to separately set out its own analysis of the evidence, albeit I have my doubts since the PAT would have been expected to consider the evidence in concluding that the “unreasonable” test was met, I do not consider that they give rise to a conceptually distinct second stage of the process, separate to the rule 4(4)(a) question and requiring a separate round of submissions and decision-making at the appeal.

Furthermore, I can see no reason in the present case for having two such stages. The rule 4(4)(a) question for the PAT was whether the findings of gross misconduct were within the range of reasonable findings that the Panel could have made. Answering that question in the negative (if the correct test was applied), would in turn supply the only permissible answer to the question of whether the charges were proven. Equally answering the rule 4(4)(a) question in the affirmative would mean that the appeal failed. Pursuant to rule 16 of the 2012 Rules (save where the rules impose specific requirements), it is for the PAT to determine its own procedure. Whilst there may be circumstances where a PAT could consider it desirable to split submissions on the rule 4(4)(a) issue from submissions on its own evaluation of the evidence and the order it should make (if it found the rule 4(4)(a) test to be satisfied), there is plainly no requirement to do so in a case of this nature and there was no error of law on the part of the PAT in not doing so.

Nor do I consider that there was any procedural unfairness, as alleged. The PAT did not suggest that it would restrict submissions to the rule 4(4)(a) question in the first instance. If Mr Thacker was in any doubt, he could have checked with the PAT what was envisaged. For the reasons I have explained, I can see no logical reason in this case for splitting the submissions into two parts; the PAT had to consider all of the material that was before the Panel below in determining whether its finding was unreasonable (in the sense I have identified) and therefore the parties would be expected to address this in their submissions. No restriction was placed by the PAT upon the oral submissions that Mr Thacker was able to make. In so far as the transcript suggests that he focused his submissions upon the Panel’s reasoning and the criticisms that the officers’ counsel made of that reasoning, that was his choice to do so.

Accordingly, I do not consider that the contentions in Ground 2 gives rise to a free-standing basis for impugning the PAT’s conclusion expressed in para 92 of its decision that the AA had not shown that it was more likely than not that the officers had acted dishonestly.

However, for the avoidance of doubt, it appears to me to follow from my conclusion on Ground 1, that the conclusion expressed at para 92 cannot stand in any event. The PAT had no jurisdiction or lawful basis to proceed to substitute its own determination, as its antecedent conclusion that the rule 4(4)(a) appeal test was met was flawed by misdirection.

Ground 3 is also focused upon the PAT’s substituted determination in para 92 of its decision that the AA had not shown on a balance of probabilities that the officers had acted dishonestly. This conclusion is said to be irrational because the PAT failed to consider all of the evidence and / or it involved a breach of procedural fairness. I have already rejected the procedural fairness point in respect of Ground 2 and Ground 3 does not materially add to that contention.

I do not consider that the specific complaint of irrationality that is made in Ground 3 is made out. The fact that the PAT did not have a chance to read all of the material that was before the Panel between hearing submissions and making its decision is irrelevant, given the expectation that the PAT would have pre-read the documents. There is nothing to suggest that the PAT had failed to read the material in advance; in paras 75 and 77 of its decision it said it had considered all the available evidence.

Nonetheless and for the reason I have identified in para 91 above in respect of Ground 2, the PAT’s substituted determination cannot stand in any event, given that it did not lawfully arrive at that stage as its antecedent conclusion that the rule 4(4)(a) test was made out was flawed by legal error.

The claimant contends that the PAT’s conclusion that the rule 4(4)(a) test was made out was irrational. In other words, it is said that no reasonable PAT could have concluded that the Panel’s findings of gross misconduct were outside of the range of reasonable findings available to the Panel. In support of this proposition, the claimant makes a number of specific criticisms about the way the PAT addressed the Panel’s assessment of Professor French’s evidence as well as some more general points.

As regards the PAT’s approach to the Panel’s evaluation of Professor French, the claimant contends that:

The claimant also contends that the PAT failed to have regard to aspects of the evidence heard by the Panel, identifying alleged instances of this at para 53 of the Statement of Facts and Grounds.

However, the over-arching point made is that the Panel had the advantage of hearing and assessing the witnesses across three days of evidence, including the evidence of both PC Michel and PC Charnock. The Panel also had the advantage of undertaking a site visit. In all the circumstances, there was no legitimate basis for concluding that the Panel’s finding that the charges were proven was unreasonable.

In addressing Ground 1 I have found that the PAT did not apply the correct test when concluding that the rule 4(4)(a) criterion was established and thus its conclusion in that respect was flawed by public law error. In these circumstances I have considered whether it is in fact necessary for me to also address Ground 4. To add anything to Ground 1 and in order for the court to properly assess the alleged irrationality, this contention must be considered on the basis of whether this conclusion was rationally open to the PAT if the correct rule 4(4)(a) test was applied. I consider it is appropriate to address Ground 4 on this basis, as the rationality or otherwise of the PAT’s conclusion will likely bear on the form of relief that the court should grant.

Accordingly, the question for me is whether a conclusion that the Panel’s findings of gross misconduct were outside the range of reasonable findings that the Panel could have made, was a conclusion that no reasonable PAT could have come to.

In these disciplinary proceedings the question of whether gross misconduct was established turned on one central issue, namely whether the AA had shown that the officer in question had given a dishonest account in the respects alleged. In turn, this largely rested on the Panel’s assessment of PC Michel and PC Charnock. Both officers accepted that their earlier accounts were inaccurate in the respects identified in the disciplinary charges; the question was whether the AA had proved that these inaccuracies stemmed from deliberate lies in circumstances where the officers said they had made honest mistakes. Accordingly, this was a case that was pre-eminently about these officers’ credibility. In such circumstances it would usually be expected that there would be two possible options open to the tribunal of fact, both coming within the range of reasonable outcomes, namely that the Panel could find it more likely than not that the officers had been dishonest or it could find that this had not been established. In this case, those acting for the interested parties did not submit after the evidence had been heard that the only lawful outcome for the Panel was to find that the charges were not proven.

In these circumstances it would be surprising if an appellate tribunal, who did not have the advantage of assessing the witnesses giving their evidence, were to conclude that the fact-finding body who did have that advantage, did not have the option of disbelieving PC Michel’s and PC Charnock’s accounts and finding that the charges were proven, as one of the reasonable outcomes available to it. Having listened to these two officers give their evidence and undergo cross-examination, the Panel found that they were not credible or reliable (paras 30 and 37 of its decision). I do not see how the PAT was rationally able to conclude that this was not open to the Panel. As I have already observed at para 80 above, this case appears to be me to be a paradigm example of one which the Panel could lawfully have decided either way, in terms of whether the charges were found proven or not. There were factors that supported each of these outcomes and it was for the Panel to attribute such weight to the competing factors as it saw fit.

For present purposes I am particularly concerned with assessing whether there were factors that were capable of supporting the Panel’s findings that the alleged gross misconduct was found proven, in order to determine whether the PAT could rationally decide that such a conclusion was beyond the range of reasonable findings available to the Panel.

I have already identified evidence that was capable of supporting the finding that the officers had seen the red light before moving forwards and had lied in their earlier accounts to the effect that it was green at paras 81 – 82 above. As regards the question of whether the officers had lied in both saying that there were two cars in front of the police vehicle when it was stopped at the lights, the Panel was plainly entitled to view the officers’ credibility as a whole, so that rejecting the honesty of their case on the red light issue bore directly on its assessment of the honesty of their accounts in relation to seeing the two cars, not least since this was part and parcel of the same incident. For reasons I identify at paras 108 – 116 below, I conclude that the way the PAT dealt with Professor French’s evidence was seriously flawed and that the Panel were entitled to conclude that it was of “little assistance”. Furthermore, it was permissible for the Panel to attach weight to the undisputed fact that the officers got the same notable features wrong in their accounts and to form the view that this was indicative of collusion rather than coincidence (para 32(q) and 39(s) of its decision). Contrary to Ms Williamson’s suggestion, I do not understand the Panel’s finding at para 39(v) to be inconsistent with a finding of collusion; and nor was it a necessary element of finding the charges proven that the Panel had to be able to pinpoint the precise time when collusion occurred.

I therefore conclude that it was open to the Panel to find that the alleged gross misconduct was proven in relation to both of the officers. Mr Baumber suggested during the hearing that such a conclusion on my part would be tantamount to my (wrongly) finding that the Panel had to find that the charges were proven. It is not. For the avoidance of doubt, finding the alleged gross misconduct proven in relation to the officers was an option that was within the range of reasonable findings that were open to the Panel. I accept that if they had formed a different view of the officers’ credibility, the Panel could have found that the charges were not proven and that this would also have been a finding that was reasonably open to the Panel in the circumstances.

Accordingly, a conclusion that the rule 4(4)(a) test was made out because finding the alleged gross misconduct proven was not an option that the Panel could reasonably arrive at was irrational; there was simply no proper basis for it and the PAT did not identify why this was not a permissible option for the Panel. In arriving at this decision, I have also considered the PAT’s specific reasoning, to which I now turn.

I have largely dealt with two of the key paragraphs in terms of the PAT’s reasoning, namely paras 71 and 74, when I considered Ground 1 (paras 72 - 78 and 80 - 82 above). I also note the following in respect of those paragraphs:

Professor French was very clear about the task that he had been asked to undertake and the limitations of that task. As he indicated in his report, he was asked to address whether it was possible that PC Michel gave an account that he sincerely believed to be true at the time even though objective evidence subsequently showed it to be inaccurate (para 21 above). In turn, the conclusions expressed by Professor French were couched in terms of this being a possibility (para 22 above). Professor French did not say that he considered it likely that PC Michel or PC Charnock had told the truth and he did not suggest that he was able to discount the proposition that they had lied. He reaffirmed this in his oral evidence. The following exchange occurred during his cross examination:

Accordingly, even if the Panel accepted Professor French’s evidence it did not in and of itself negate the ability of the AA to prove on a balance of probabilities that the officers had given dishonest accounts, since all he could say, as he fairly acknowledged, was that it was possible or plausible that they had provided honestly held recollections. He was quite clear that he was unable to say whether they had lied. Allied to this, Professor French had not assessed the officers personally and he had not heard them give their evidence to the Panel, nor heard the other evidence. This was the central point that the Panel was making at para 33 of its decision in terms of why it found his evidence to be “of little assistance” (para 31 above). In my judgment, this was a conclusion that was open to the Panel.

It is apparent from the contents of paras 53 – 71 and 74 of its decision that the PAT’s analysis of the Panel’s approach to Professor French’s evidence was a central plank of its conclusions. In the following paragraphs I address the claimant’s specific criticisms of the PAT’s analysis.

Firstly, the PAT criticised the Panel for “disregarding” the expert evidence, for arriving at its findings “without consideration” of the expert evidence and for not taking it into account. This was not an isolated instance of imprecise language, the PAT said this a number of times, at paras 60, 62, 63, 64 and 71 of its decision (paras 39 - 41 above). However, the PAT was simply incorrect in this regard; as para 33 of its decision showed, the Panel did consider Professor French’s evidence, but concluded that it was of little assistance for the reasons it gave. The PAT’s real disagreement was with the weight that the Panel gave to his evidence, after it had considered it.

The claimant’s second criticism is also well founded. At paras 59 – 62 of its decision (para 39 above), the PAT focused individually on some of the Panel’s reasons for finding that Professor French’s evidence was of little assistance, without reflecting upon the cumulative weight of the Panel’s reasons or indeed even referring to what appears to have been the central reason for the Panel’s assessment (para 109 above).

Furthermore, the PAT plainly misunderstood the point that the Panel was making at its para 33(c), namely that Professor French could not assist with whether or not the officers had lied in the accounts they gave, which was for the Panel to determine. In the course of making this point, the Panel contrasted its remit with the position of the Professor, who was largely reliant upon the written account of MM. This observation was accurate (Professor French gave the opinions expressed in his report by reference to PC Michel’s account in his Regulation 22 Notice) and in any event, the precise material that the Professor had regard to was not the central point being made by the Panel in this sub-paragraph. However, the contents of para 61 of the PAT’s decision indicate that it mischaracterised – and thus inaccurately criticised - this part of the Panel’s reasoning. It points out that Professor French did not only have access to PC Michel’s account and it “did not see how the expert evidence could be undermined by taking into account the information provided by the first appellant”; a proposition that the Panel had not suggested.

The contents of para 62 of the PAT’s decision indicate that it also misunderstood the Panel’s para 33(d). In para 33(d) the Panel explained that it had heard the officers give evidence and was in a position (unlike Professor French) to assess that they had lied. It was not saying that it had decided the officers had lied before it had any regard to Professor French’s evidence.

The claimant’s fourth criticism, namely that the PAT wrongly elevated the significance to be attached to the expert evidence, is also well founded. At para 53 the PAT explained why it considered Professor French’s evidence to be directly relevant, saying that the “expert evidence dealt with the ways in which these inaccurate memories were likely to have been formed” (para 38 above; emphasis now added). In fact Professor French had not advanced this proposition, as I have already noted he was careful to identify the limits of his expert opinion in both his report and his oral evidence (para 108).

Given that it had permissibly found that Professor French’s evidence was of little assistance, I do not consider that it was incumbent on the Panel to explicitly discuss in its reasons the particular processes, such as memory conformity, which he had identified as providing possible explanations for the officers’ accounts.

For the reasons I have identified, a conclusion that the rule 4(4)(a) test was made out because finding the alleged gross misconduct proven was not an option that the Panel could reasonably arrive at was irrational.

The PAT failed to apply the correct test when it concluded that the Panel’s findings that PC Michel and PC Charnock were guilty of gross misconduct were “unreasonable” within the meaning of rule 4(4)(a) of the 2012 Rules. The PAT’s decision contained no express self-direction as to the meaning of “unreasonable” in this context and my examination of its reasoning shows that it failed to apply the meaning that has been established by the case law, namely that a finding of gross misconduct was not within the range of reasonable findings open to the Panel. Accordingly, Ground 1 is well founded.

Although I do not accept the specific criticisms advanced by the claimant in respect of Grounds 2 and 3, it in any event must follow from my decision that the rule 4(4)(a) conclusion was flawed by public law error, that the PAT never lawfully arrived at a point where it could substitute its own determination.

Ground 4 is well founded in the sense that if the correct rule 4(4)(a) meaning of “unreasonable” is applied, the conclusion that the Panel’s findings of gross misconduct were outside of the range of reasonable findings available to the Panel was irrational. This was a case where the Panel could lawfully have found that the charges were proved or that they were not proved; the outcome essentially turned upon its assessment of the credibly and reliability of PC Michel and PC Charnock, whom the Panel had heard give evidence. Furthermore, the Panel permissibly concluded that Professor French’s evidence was of “little assistance” in the circumstances.

Following circulation of a draft of this judgment, the parties have agreed that it follows from my conclusions that the PAT’s decision to overturn the Panel’s finding and substitute it with a finding that that the allegations were not proven must be quashed and substituted with a decision that the appeals of both the interested parties are dismissed.