BEIS0000894
BEIS0000894
HORIZON COMPENSATION ADVISORY BOARD
CONCERNS ON THE SYSTEMs FOR
CRIMINAL PROSECUTIONS AND OVERTURNING CONVICTIONS
This paper summarises the concerns identified to date by of the members of the Horizon Compensation
Advisory Board (the Advisory Board) on aspects of the current system(s) in England and Wales for
investigating, prosecuting, and overturning the convictions of, the sub postmasters and -mistresses
(SPMs). This note does not deal expressly with the systems in Scotland or Northern Ireland, although
many issues are also relevant in those jurisdictions.
The Advisory Board advises ministers on the various schemes that operate for delivering compensation
to victims of the Post Office Horizon scandal. Terms of reference, minutes and other information are
published on the Department for Business and Trade’s website.' The Advisory Board comprises
independent parliamentarians and academics: Professor Christopher Hodges OBE (Chair), the Rt Hon
Lord Arbuthnot of Edrom, the Rt Hon Kevan Jones MP, and Professor Richard Moorhead.
The basic facts of the current status of criminal cases arising out of Horizon scandal are that it is
abundantly clear that a significant number (estimates range between 700 and 1000 people) have been
convicted of offences of which they are innocent — but that the relevant rules and procedures form
considerable barriers to rectifying these injustices, and certainly do doing so quickly. This situation
itself constitutes a further separate national scandal, in addition to the originating behaviour of the
organisations involved.
The Advisory Board considers that its remit of ensuring that full and fair compensation is paid to all
victims of the Post Office Horizon IT scandal necessarily encompasses it to consider the plight of the
largest currently outstanding cohort of SPMs who have not been compensated to date, namely those
who have been convicted unjustly, but whose convictions are not being overturned, and may never be
overturned unless positive action is taken by the state.
At the least, the Advisory Board considers that there are a number of serious issues where both the
investigation/prosecution and overturning aspects of the current systems give rise to serious need for
reform in relation to any other future similar mass cases. The Advisory Board considers that it is
inadequate merely to review the system for criminal appeals, without reviewing and reforming the
system for investigation/prosecution. One reason for this is that if the latter continues to be inadequate,
the number of appeals, and unjustified convictions, some of which may never be appealed or
overturned, will continue to be matters of great public concern, undermining public confidence and
producing a large number of appeals as well as some justified appeals that are never brought.
A. The Overarching Timing Issue
(1) _ The whole approach to rectifying blatant and widespread injustice is far too slow.
The pace of review is very slow. This occurs for a number of reasons, including initial inertia in
accepting that there was a major problem; and subsequent challenges of accessing evidence, as well as
the fact that the review and rectification systems are designed to be forensic and to examine cases on an
individual basis, involving a number of stages and bodies. The ‘system’ involves (a) too many
people/stages, (b) different criteria at each stage, (c) a great deal of missing evidence, both factual and
expert, and (d) an inability for someone to take holistic and speedy decisions. The State, as a whole,
has no mechanism for taking a speedy, holistic overview and initiating mass corrective action.
The result is that there is a high risk that many people whose convictions are unsafe have not only been
treated egregiously and unjustly by the State, and that this may continue for some time, lasting much of
their lifetimes, and in some cases not being corrected before they die with their convictions in place.
That is an affront to a civilised society based on the rule of law, justice and fairness.
The Advisory Board considers that there is an urgent need to apply a speedy, holistic resolution
that overturns all Post Office convictions. It may be logical to continue the current approach of
investigating individual cases, and trying to identify how many cases are unsafe, safe, or unclear. But
too many uncertainties remain (there are 5 or 6 grey areas) over whether forensically accurate
1
BEIS0000894
BEIS0000894
outcomes are actually achievable, and it’s certainly all going to take too long. The case for a systemic,
and hence swift, solution is strong, if not overwhelming. Given the evidence that the behaviours of Post
Office Limited (POL) in investigating and prosecuting people have been so egregious, and constitute an
affront to maintaining trust in the ability of a civilised state to uphold justice, the balance of
considerations in favour of implementing a systemic solution is clear. In this case, that balance clearly
favours overturning convictions of the innocent as against the risk that some guilty might go free (they
have been punished already).
B. Some Specific Issues with the Investigation and Prosecution of Cases
(2) — Separation between investigating and prosecuting authorities.
In England & Wales, POL investigated and prosecuted cases itself. Some cases involving DWP were
investigated and prosecuted before 2006, based on information provided by POL. In Scotland and
Northern Ireland, prosecutions were conducted by the public prosecution agencies.
The separation of authorities should give some degree of trust that the governance of authorities
provides some control of independent review over conduct that may otherwise become inappropriate or
entrenched. Recent expert evidence to the Inquiry suggests that the absence of this separation
contaminated and weakened PO practices and decision-making around investigation and prosecution.
We refer below to behavioural failures by the prosecuting department of POL in both investigation and
prosecuting. One aspect is the failure to take any account of various pieces of information that its
approach to relaying blindly on the accuracy of the Horizon data, and seeking to ‘defend’ Horizon
system, was wholly incorrect.
(2) Availability of evidence
Many cases in the POL saga are now mired in difficulties because evidence has been destroyed over the
passage of time, up to 20 years. As a result of the application of data retention policies, and a lack of
systematic storage and retention of papers by POL, limited evidence was available, and confusion and
delay has been caused by the late unearthing of further relevant silos of material. This was particularly
true for cases prior to 2007, where the available evidence for some cases did not even reveal the nature
of the charges. Even for later cases, evidence was often patchy. In the digital age, archiving of material
should be a given. The investigative efforts of POL and the CCRC have demonstrated that there may
be many individual circumstances that need investigation and interpretation — but which were not done
during original investigations. POL have informed the Advisory Board that in at least 118 cases they
currently have insufficient evidence to make a determination on whether a case is or is not unsafe, and
that in at least 333 cases they cannot conclude that a case should be conceded as unsafe based on the
information available to them, which they accept may well not provide a full picture of a case.
The Advisory Board considers that the difficulties in establishing the facts from available evidence
are so great in many cases that existing unjust convictions will not satisfy the current forensic
criteria for being overturned, unless a fresh approach is taken, either in individual cases or
systemically.
(2) Systemic behaviour in interviews and pursuing prosecutions by the Post Office
Concerns have been raised that interviews conducted by POL staff were oppressive, especially given
the vulnerability of some of the people accused. In discussions with the Advisory Board, the POL team
of independent counsel reviewing cases did not dispute this had happened in some cases. That team
noted that guidance required investigators to offer postmasters access to publicly funded lawyers to
advise them in the interview or have a ‘Post Office friend’ accompany them. If necessary, interviews
were adjourned until such advice could be provided. Further, many SPMs had received legal advice,
and had seemingly invariably been legally represented in court. However, SPMs have given sometimes
harrowing evidence that they were advised to produce explanations for apparent shortfalls, and to make
admissions and enter guilty pleas to lesser charges, so as to avoid more serious consequences. There is
evidence that representation at disciplinary interviews, prior to criminal investigatory interviews, was
discouraged and sometimes prevented.
BEIS0000894
BEIS0000894
Nevertheless, some stories of intimidation of suspects and behaviour by investigators that is oppressive
or deliberately false have been recorded, as have invitations from auditors and the like to ‘explain
shortfalls’ by making up a story to avoid prosecution for theft. A commonly reported approach was
investigators and others lying that “You are the only person this has happened to, no-one else has
questioned the Horizon system’.
Concerns about prosecutorial misconduct are emphasised by recent evidence to the Inquiry of the
absence of appropriate prosecution policies, controls, and practices, which failed to institutionalise the
golden rules of prosecution around investigating all reasonable lines of inquiry and the leading role of
business people (and business goals) in taking prosecution decisions. Currently the evidence is pointing
strongly towards a system riddled from top to bottom with serious systemic problems.
The Advisory Board’s concern is that this system needs serious scrutiny. The existence of guidelines
for investigators and for prosecutors, and a policy of incentivising the guilty to plead guilty so as to
indicate remorse and receive a lesser sentence, have all been shown to produce — in the case of many
wholly innocent people — mass and comprehensive injustice. This situation calls for serious research
and rethinking more generally, but it underlines, and is magnified by, the structural, cultural and
behavioural problems evidence around prosecutions conducted by the PO in the pursuit of the
business’s own interests.
The Advisory Board considers that there is considerable and mounting evidence that the
behaviour of POL in investigating and prosecuting many cases, was oppressive, egregious and
constitutes a cogent ground for overturning convictions in its own right.
(2) The issue of admissions, confessions and guilty pleas
In many cases SPMs pleaded guilty — often based on legal advice — and some made admissions in their
interviews or in writing.
The POL review has told the Advisory Board that they did not regard those confessions as
automatically reliable: it considered the reasons why they might have been made, and whether they
were corroborated by other available evidence. Some cases have been conceded despite the existence of
such confessions. If a postmaster’s evidence (either at the time of the prosecution or by way of fresh
evidence thereafter) indicated that s/he did not know how the shortfall had arisen, and there was no
reliable evidence independent of Horizon to the contrary, the team took that as an indication that the
reliability of Horizon was essential to the prosecution. Appeals in such circumstances were conceded.
There is a strong explanation and justification for why many SPMs made up false stories and made
false confessions. Some SPMs have said that they were quite unable to understand what had happened.
They could not understand why there were shortfalls, they were petrified, they wanted it all to stop.
They were advised to give explanations on what had happened to the money (eg it was used to pay off
other debts) which were fiction. They were advised, or threatened, with worse consequences if they
failed to appear contrite, cooperative, and accepting of guilt. So they were talked or frightened into
saying things so as to reduce possibly worse consequences and as a means of attracting leniency.
We have heard conflicting stories as to whether this situation was accepted by those reviewing
convictions at all levels (the POL team, the CCRC, the Court of Appeal). The situation is undoubtedly
difficult. There is logic in a legal system having clear rules that support personal responsibility and
honesty, and impede people trying things on. But there is no doubt in our minds that behavioural
science explains why people who are confused and frightened will predictably make up stories and
behave irrationally so as to protect themselves. The issue that is raised is whether the legal system (at
each of the stages of investigation, prosecution, sentencing and review of appeals) is capable of
differentiating between the innocent and the guilty — and the Horizon story raises serious concerns in
this regard. It is clear to us that there remain many Horizon cases where the appeals system has not
been capable of rectifying unjust convictions.
This seemingly widespread human response to being attacked by the state is deeply worrying in relation
to confidence in the criminal justice system as a whole. It calls for significant research into the
psychological effects of the behaviour of state authorities, and review of rules.
3
BEIS0000894
BEIS0000894
The Advisory Board considers that there is deeply worrying evidence of oppressive behaviour by
individuals and bodies, and that existing guidelines are inadequate as a means of controlling such
behaviour. It also considers that there is convincing evidence that oppressive behaviour led to fear
on the part of innocent SPMs, which induced them to make admissions, make up stories and enter
pleas that are demonstrably unreliable, understandable as human reactions, and should justify
their convictions being overturned. The implications of this for the criminal justice system as a
whole are profound and call for deep investigation and thought.
(2) The rules on acceptance of digital evidence are not fit for purpose and need revision.
The Common Law rule is that there is a presumption that computer evidence is admissible, and that
this can be rebutted of evidence to the contrary is adduced, in which case the party seeking to reply on
the computer record has to satisfy the court that the computer was working properly at the material
time.? But can a party and their lawyers effectively gain access to such evidence? The Horizon history
demonstrates extensive failure in this respect: The rule that computer evidence has to be taken at face
value has proved to be difficult to challenge in individual cases or within the culture of the Post Office
and other prosecuting authorities. If the courts had permitted objective scrutiny of POL material on the
reliability of Horizon, and the number of prosecutions based on it, the whole scandal would have
broken much earlier, and caused damage to far fewer people. The cases reveal only some requests for
such evidence, that appear nearly all not to have been satisfied, and not enforced by the judiciary. Some
defence lawyers tried to challenge the reliability of the Horizon system, and were apparently blocked by
courts or inappropriate rules. That rule/approach needs serious review.
Wider Governmental policy debates about the reliability of digital and AI systems is clearly based on
the proposition that their reliability cannot be taken for granted, that harms may occur, and need
appropriate governance, regulation and liability rules.*
The Advisory Board considers that justice can no longer be served by the historical general
presumption that computer evidence is acceptable. The theory that a defence can gain effective
access to reliable and comprehensive evidence that might enable a presumption to be rebutted has
been shown not to work. The rules and the system need revision.
It can be imagined that allowing extensive challenges to any computer or technical (AI) evidence may
raise concerns over extensive delays in process, and to ‘time wasting’ applications. A different but
trustworthy approach is needed to verify digital evidence.
We suggest the consideration of a mixed approach, involving independent verification of technical
systems, perhaps with an auditing body, and independent oversight of data — such as the ability to
identify and investigate a sudden increase in cases, or a large number of unexplained cases of the same
type. This would be supported by an improved mechanism for the defence to challenge reliability, at an
early stage. The system should be subject to scrutiny by an independent prosecuting authority.
C. Appeal Procedures — England & Wales
(2) Is it right that a prosecuting authority is involved in reviewing or making a decision on
whether to object to any cases, where there is evidence that that authority has behaved
egregiously in pursuing the initial cases? Should this not be done independently?
In the Horizon case, the POL has instituted a review of its criminal prosecutions to be undertaken by a
team of independent lawyers. The POL team have passed cases to the CCRC or engaged with other
authorities in Scotland and Northern Ireland. We stress that we have no criticism of the lawyers
involved in this process, who have assured us that they consider that their professional obligations
require them to take an independent view of cases based on the facts and the applicable legal
framework. Cases are initially reviewed by junior counsel. Those cases flagged as being potentially
appealable are being reviewed by two KCs. The review team confirmed that their overriding objective
is not to protect or defend the Post Office, or to try to uphold convictions (the Post Office had not
asked them to do so, and in any case such action would be contrary to their professional obligations),
but rather to ensure that unsafe convictions were quashed and that safe convictions were upheld.
4
BEIS0000894
BEIS0000894
However, the ongoing managerial control by POL over POL’s review exercise has failed to attract the
trust victims or the public. This is significant in itself. Moreover, the Inquiry has also raised significant
doubts over the adequacy of disclosure exercises, which would be highly pertinent to appeals, and have
marked out the need to investigate if concerns go beyond competence to something even more
concerning. It cannot be right that a body found to have prosecuted in a manner that was found to be an
affront to justice, and which continues to demonstrate real difficulties in providing proper disclosure
continues to manage critical decisions on appeals. The Advisory Board considers that there should
be an independent authority that could step in and govern such a process of investigation and
review.
(2) The processes and rules differ between Crown Court and magistrates court convictions. It is
clear that those who were convicted in magistrates courts face significantly lower barriers in
getting their convictions overturned. Why are these different, with different outcomes?
A postmaster convicted in the Crown Court can ask the Court of Appeal for leave to appeal (including
whether the normal time-limit for appeals should be waived). Alternatively, the Criminal Cases Review
Commission (CCRC) can refer a case direct to the Court, which bypasses the need to obtain leave to
appeal. A referral is required if a person has appealed once. The Court of Appeal will overturn a
conviction if it finds that it is ‘unsafe’. Before doing so it would review the detailed evidence from all
parties and form its own view based on that evidence: it would not quash convictions simply because
the Post Office asked it to do so.
Cases of postmasters who pleaded guilty in the Magistrates’ Courts can only be appealed if their case
is referred by the CCRC. This means that cases can be reopened if the CCRC considers that there is a
real possibility that the conviction would be overturned by the Court. Appeal cases are heard at the
Crown Courts where they were dealt with by way of a fresh hearing. On reviewing convictions, PO
has only 2 choices: support overturn or retry. If an appeal is referred, the Post Office would apply the
Full Code Test set out in the Code for Crown Prosecutors‘ — i.e. they would be assessed to establish
whether there was a realistic prospect of conviction and the retrial was in the public interest (or we
think it its own interest). The Post Office has to date concluded that it is not in the public interest to
retry any Horizon cases, either on evidential or public interest grounds, so the Crown Courts have been
able to overturn convictions without contest.
In Scotland, the single test for success in a criminal appeal is ‘miscarriage of justice’.*
a. Why are the procedures and rules different? Should they not be the same?
b. The burden of proof is easier for magistrates’ convictions because it involves a fresh
hearing, and the PO can in effect offer no evidence.
The Advisory Board considers that the rules and procedures for reviewing and overturning
convictions in any court or by any prosecutor should be the same.
(2) Inconsistencies between instigation of an appeal by a convicted person and referral by the
CCRC.
Where convictions were made in the Crown Court in England and Wales, a victim has in effect to take
a positive step to initiate his or her appeal. In contrast, convictions in Scotland can be referred for
appeal by the Scottish CCRC (SCCRC) to the High Court without the individual having to take any
action,® although the SCCRC does require the participation of the affected individual (or, if deceased,
their representative) in initiating an application. The rationale is stated to be that ‘an appeal with no
appellant stands no chance of success’. (The Advisory Board notes the word ‘appeal’ here: if the
process were not considered to be an ‘appeal’ then a different outcome might be achievable where an
affected person is unwilling to act.) In Scotland, the SCCRC’s test for referral to the High Court is
where it believes that there “May have been a miscarriage of justice’ and that it is ‘in the interests of
justice’ to make the referral.’ The tests applied by the SCCRC in making a referral and the High Court
in granting an appeal appear to be more favourable to victims than in England, Wales and Northern
Treland.
BEIS0000894
BEIS0000894
As discussed above, there are cogent reasons why traumatised individuals may not wish to initiate any
action in relation to lodging an appeal. The POL and the CCRC wrote circular letters to hundreds of
SPMs whom they could identify, but these generated a notably low number of responses. Even where,
in the light of the review, the Post Office had written personally to a postmaster to say that it would not
oppose an appeal, some were reluctant to take any action. POL has told us that of the roughly 700
cases where they think Horizon evidence was relied on, 248 people have so far not responded to various
attempts to contact them. 64 people have said that they do not wish to proceed or have asked for no
further contact. Further, POL have contacted 26 individuals to say that they would not contest their
case if they were to appeal but only 15 have so far responded. The CCRC has told us that in response
to their communicating with people, 21 requested no further contact from the CCRC or from POL.
This suggests more than potential misunderstanding of what the process might entail, disillusionment
with the legal system, and a desire to maintain closure in respect of Post Office issues and pressure in
other parts of their lives, but also extensive psychological harm and a worrying lack of faith in their
State.
Recent research surveyed 101 victims of the PO scandal who were wrongly accused, convicted and/or
investigated for financial ‘crimes’ that were actually caused by software errors.’ Most respondents
reported clinically significant post-traumatic stress (67%) and depressive (60%) symptoms —
irrespective of the outcome of their case. No differences were found in the severity of mental health
symptoms between respondents who were convicted of criminal offences and respondents who were
investigated, prosecuted, and/or pursued in civil court. The authors concluded that the findings provide
the first evidence that wrongful accusation may be just as damaging to mental health as wrongful
conviction.
The ability to instigate its own action to rectify injustice meted out through its own courts should be a
given for a democratic State. In the Horizon situation, the need for this ability to initiate action is
clearly demonstrated. There is extensive evidence that individuals have been so traumatised by their
experiences at the hands of the State that they often want little or nothing further to do with official
procedures. They just want to be left alone. The State should be able to instigate action to rectify
injustice where it is apparent that it has occurred.
The State should be able to instigate the overturning of a conviction without needing the convicted
person to initiate any step in any case where the State concludes that injustice has occurred.
(2) Limitations on the Post Office’s ability to concede cases — and different tests applying to
different bodies and stages of the appeals process
Different tests apply to different bodies and stages in the appeals process. Cases in this saga have to go
through reviews by POL, the CCRC, and the Court of Appeal. Each of them aim to apply legal criteria
to the facts as they appear — and also the criterion of what is, or is not, a ‘Horizon case’. The latter
definition was based on the Court of Appeal’s decision in Hamilton, itself limited by the state of
evidence on Horizon and the arguments presented to it at the time. The position is highly confusing and
appellants face a series of different hurdles that are difficult for lay people to understand. Importantly,
the result drives inconsistencies and hence perpetuates injustice.
For POL, its review has to try to predict what both the CCRC and Court of Appeal might decide. It
could ‘offer no evidence’, as it has done in appeals to the Crown Court, which results a successful
appeal. However, it must also apply the Full Code Test in the Code for Crown Prosecutors, which
includes two stages, starting with whether there is sufficient evidence for a realistic prospect of
conviction, and then whether a retrial would be in the public interest. Thus, POL may decide that a
retrial may not be in the public interest, in which case, the appeal succeeds. Given the limits of evidence
held by POL in many cases and the Court’s approach to appeals, there is an unidentified but plainly
sizeable body of cases that requires evidence from SPMs before the Court would allow them to
proceed. Without such evidence, POL lawyers would not feel able to advise that an appeal should be
conceded.
The CCRC also applies the law, including the Full Code Test, and the Court of Appeal’s decisions on
what is a ‘Horizon case’. As a matter of important detail, there is an inconsistency between the s 14 of
the Criminal Appeals Act 1995, which empowers the CCRC to refer a case without an application
6
BEIS0000894
BEIS0000894
being made to it, and s13(1)(a), which additionally requires the ‘real possibility’ test to be satisfied
before a case can be referred, but which must be passed where an individual does not participate in the
process. That anomaly should be removed.
The fact that different lawyers can reach differing decisions is evidenced by the facts that, firstly, the
CRC has to date decided not to refer 33 cases out of 101 reviews by POL and, secondly, the Court of
Appel has not overturned some 48 cases.
The test in the Court of Appeal is different, and places the burden on an appellant to show that their
conviction is unsafe. However, in successive POL cases, the Court of Appeal has developed
interpretations in applying the basic criteria to “Horizon cases’ and other cases. POL considers that it
may be in contempt if it fails to take a stance of opposing or arguing a case in the Court of Appeal, in
accordance with the adversarial principle on which the proceeds are based. We have been told by POL
that if it were to concede a wider range of appeals, there is no guarantee that the Court of Appeal
would agree to quash the convictions concerned, and apprehend that they would be subject to
significant criticism from the bench for excessive leniency.
Thus, the criteria are different, difficult to apply in these complex and uncertain circumstances,
although the conclusion that a large number of people have been wrongly convicted is pretty obvious.
The two- and three-tier systems are confusing, duplicative, and inconsistent. It is time to examine
whether the continued existence of two stages of review by the CCRC and by the Courts is justified.
As the CCRC had made clear in its recent article, published on 25 July 2023, cases which did not meet
that test could also be considered for review based on the facts of the case. However, the Court of
Appeal made clear that if a case was not a ‘Horizon case’, and an appellant wished to advance other
grounds of appeal, they would need to persuade the Court to extend the time limit for appealing, which
might be “a very difficult hurdle for an appellant to surmount”. This would particularly affect cases
where an appeal would be on the basis of information that was known to the appellant at the time of the
prosecution and conviction and therefore could have been raised at the time of the original
trial/conviction by the defendants’ lawyers.
The Advisory Board considers that a single test and a single investigation and institution should
determine whether a conviction is or is not safe. A process that is investigative (and not
adversarial) should be both quicker and more consistent than successive stages based essentially
on an adversarial and argumentative model.
(2) _ The review of evidence by the Court of Appeal — and at earlier stages by the PO, the CCRC
etc as well as on initial prosecution and conviction — depends on what evidence is available. Is
adequate evidence available or obtained and considered?
In civil cases, the adversarial principle means that claimant and defendant have an equal opportunity to
access all the evidence of the other side, and to comment on it. This system is confused in the criminal
appeal process. In theory, the CCRC acts as both an investigator of the evidence and a review body. It
appears, however, that it has inadequate powers in delivering both functions. The role of the CCRC is
then to refer cases to the Court of Appeal, in an adversarial context. But does that system adequately
investigate all the evidence that should be produced, or is it limited to aspects of the evidence as they
were produced at the original trial? For example, has comprehensive evidence been assembled as to the
reliability of a// the computer systems that were in operation at the time, and of all the behaviours of
the investigators, and the extent to which relevant technical, expert and factual evidence was revealed
to the defence?
It appears to us that the Court of Appeal has not had presented to it the full evidence, and may never
see it. So it may not create or review relevant criteria properly, or change them.
The Court of Appeal has to date established, as a result of a number of cases brought before it, a useful
categorisation that any case falling within the ‘Horizon shortfall’ definition will be overturned. In its
judgment on the Hamilton case, the Court of Appeal had indicated that it was prepared to overturn
cases in which the reliability of Horizon data was “essential to the prosecution” (which it labelled as
7
BEIS0000894
BEIS0000894
“Horizon cases’). However, the process by which this categorisation has been reached has depended on
the presentation to the Court of evidence of generic unreliability, arising from still limited number of
individual cases. At the time of writing, 91 cases have been overturned, the CA has refused to overturn
48 cases, yet something between 700 and 1000 cases remain in limbo, and subject to considerable
uncertainty over gaps in evidence and how multiple legal criteria might be applied to them.
What is not occurring is a comprehensive, independent and reliable investigation into the facts and
scientific evidence, so that a genuinely generic and consistent approach case be adopted. The policy is
being created on a piecemeal basis through reviewing the evidence that is presented in individual cases.
That approach inherently involves major risks of perpetuating injustice. The various bodies here seem
face inability to guarantee just outcomes because they only make decisions on such evidence as is
presented to them, whereas it appears in the Horizon cases that taking a wider view (eg of the totality
of the evidence) would avoid taking too limited a view (eg creating an arbitrary distinction between a
“Horizon shortfall’ case and other cases, or an incomplete view of relevant scientific evidence).
In illustrating these points, one can point to examples of cases where there is lack of clear evidence,
confusion over the relevance of supposed confessions or pleas, and of any extraneous evidence. It
appears to the Advisory Board that it is entirely possible that the Court of Appeal has not had presented
to it adequate evidence of the reliability of the inter-relation between POL’s Horizon system and the
DWP’s Order Book Control Service systems, including the non-balance shortfall aspects. This is
because there has been no independent scientific and technical review of such matters. As a result, we
have serious doubts about the Court of Appeal’s refusal to overturn the case of O'Donnell in August
2023 (which was not regarded as a ‘Horizon shortfall case’ but involved encashment of DWP
vouchers). However, no evidence appeared to have been available on which to evaluate the reliability
of the PO/DWP computer system. The Court of Appeal itself expressed disquiet at factual uncertainty
in its judgment. That is extraordinary — how can courts and juries be expected to dispense justice in the
absence of adequate evidence as to the true facts, and the reliability of such limited evidence as they
have, when it is clearly incomplete or inadequate?
At its meeting on 14 June 2023, the Advisory Board concluded:
“The Board believed that the criteria set by the Court of Appeal for Horizon cases were too tight, and
that a significant number of miscarriages of justice could be outstanding. They also believed that the
Court of Appeal’s judgment was based on a limited understanding of the extent of problems with
financial systems in the Post Office and with the extent of wrongdoing lying behind the “affront to
public justice” finding. This led to a much wider and higher level of concern about Post Office
prosecutions (and their review) with a number of critical documents not apparently disclosed and
available to the Court.”
The Advisory Board considers that a single independent investigation should occur in appropriate
cases. One way of achieving this would be to expand the remit of the CCRC. This would not
necessarily preclude instigation of an appeal process by an individual.
(2) Overall, the current system for reviewing and appealing convictions has too many stages,
involves too many different bodies, each with different criteria: Post Office review; requirement
for action by convicted individuals which many do not wish to take; CCRC; Court of Appeal.
The reality of the Horizon scandal is that it is apparent that many SPMs have been unjustly convicted,
and, further, that the State has great difficulty in delivering the overturning of their convictions. This is
itself a further scandal. The State urgently needs to create a fresh and robust mechanism to right such
wrongs.
The current situation is that it remains unclear either (a) how many are innocent, under ‘Horizon
shortfall’ criteria or (b) innocent under non-‘Horizon shortfall’ criteria, or (c) guilty. It is better for
democracy that that the innocent are restored than that some guilty have convictions quashed for which
they have, in any event, already been punished.
The Advisory Board considers that the process for reviewing and appealing convictions is too
complex and needs to be streamlined, if rectifying injustice is not to be prolonged.
8
BEIS0000894
BEIS0000894
(2) In systemic cases, should not a single independent investigation and review procedure be
available, with power to apply a systemic and swift solution?
All these difficulties strongly suggest to us that a common sense approach needs to be applied. If it
appears that many people continue to have been unjustly convicted (and consequently suffering from
ruined lives and serious financial hardship that cannot be compensated) then the State, adopting a slow,
case-by-case forensic approach, which faces various barriers of procedure and evidence, then the State
has a duty to act.
The Advisory Board considers that there should be an urgent systemic solution, involving a
simple, bold approach that bypasses the detailed bureaucratic delays of multiple individual
investigations.
D. Compensation for those whose convictions are overturned.
We record that we have a number of reservations about the system for compensating those whose
convictions are overturned, but they are outside the scope of this document. However, until unjust
convictions are overturned, the further injustice remains that victims remain unable to access
compensation.
30.10.23
" http://www.gov.uk/government/groups/horizon-compensation-advisory-board
? Post-pandemic economic growth: UK labour markets: Government response to the BEIS Committee's
Report of Session 2022-23. Twelfih Special Report of Session 2022-23. 18 July 2023.
5 See amongst many papers Policy Paper: Digital Regulation: driving growth and unlocking innovation at
hittp://www.gov.uk/government/publications/digital-regulation-driving-growth-and-unlocking-innovation
‘ hittp://www.cps.gov.uk/sites/default/files/documents/publications/Code-for-Crown-Prosecutors-October-
2018.pdf. See Extracts in the Annex to this paper.
5 The Criminal Procedure (Scotland) Act 1995, ss 106(3) and 175(5).
© The Criminal Procedure (Scotland) Act 1995, s 194D(1).
7 The Criminal Procedure (Scotland) Act 1995, s 194C.
* B Growns, J Kukucka, R Moorhead and RK Helm ‘The Post Office Scandal in the United Kingdom: Mental
health and social experiences of wrongly convicted and wrongly accused individuals’ Legal Criminol Psychol.
2023;00:1-15, at https://doi.org/10.11 1 1/lerp.12247
‘enth