SUBS0000023
SUBS0000023
Closing Submission
on behalf of The National Federation of Sub-Postmasters
in Phase 3
of the Post Office Horizon IT Public Inquiry
Chaired by Sir Wyn Williams
Introductory remarks
1. The National Federation of Sub-postmasters (NFSP) is grateful to the Inquiry for its permission to
make a Closing Submission on Phase 3 in writing and sets out below its comments and analysis of
the evidence heard in this phase. The NFSP welcomes the careful way in which the Inquiry has
progressed through the issues and witnesses in phases 2 and 3. It notes that this approach has
brought out the evidence which will now illuminate Phase 4 and show how so many innocent sub-
postmasters, assistants and Crown Post Office employees were knowingly wrongly prosecuted at
the hands of the Post Office and had their lives ruined.
2. In the NFSP’s Closing Submission for Phase 2, having emphasized that it had entered into the
Horizon project, working groups, roll-outs and feedback, in good faith, the evidence on the
technical failings, and institutional knowledge at ICL/Fujitsu and POCL of such failings, was hard for
it to bear. The NFSP had believed the project as presented to them was to be the saviour of post
offices up and down the land. In the end, it turned out to be very antithesis of a saviour. In that
Phase 2 Closing Submission the NFSP stated that it would be of critical importance to find out who
knew what about what was wrong with the system and if, how and why that information was
suppressed, allowing sub-postmasters, manager, assistants and Post Office employees to be
prosecuted and their lives devasted as a consequence.
3. The evidence heard in Phase 3 has truly shocked the NFSP. It appears from the evidence that not
only was the NFSP being deceived by the Post Office about the design and integration of Horizon
from its earliest procurement and roll-out phases, it was also being deceived throughout by the
Post Office as to the integrity of Horizon. The continual assurances given to the NFSP that the
SUBS0000023
SUBS0000023
Horizon system was robust, had integrity, that feedback and issues being provided to the Post
Office by the NFSP as being raised by their members was being acted upon or taken into account,
were believed by the NFSP. Why would the nation’s most trusted brand rolling out the saviour of
the post office network be lying? But it was. And none of it was true. It was with genuine regret as
to what turned out to be a false belief and sincere shock at the depth of deception, that the NFSP
has had to listen to witness after witness describe how no actual investigation was carried out into
why so many subpostmasters, assistants and Crown Post Office employees had unexpected
shortfalls. Rather, for the NFSP, the evidence has been that the default position for almost
everyone, from the top to the bottom of the both organisations, some either:
e knew about the issues, including bugs, errors and defects in Horizon, and put alleged
dishonesty at the top of the agenda and ensured it remained there; or
® were conditioned to believe that dishonesty of postmasters and their employees was the
root cause.
The NFSP believes, from what it has heard in Phase 3, that not only was the myth of Horizon’s
integrity actively promulgated by the Post Office among its own staff, so was the alleged dishonesty
of sub-postmasters, assistants and Post Office employees in Crown offices. This meant that rather
than focusing on the errors, bugs and defects and trying to find out why shortfalls and other
difficulties were being encountered, the focus was on financial recovery and prosecutions. And just
in case there was to be any traction with the idea that there could be something other than
dishonesty at work, the report which was commissioned by the Post Office from and produced by
Rod Ismay sought to put that firmly to bed.
The NFSP believe that report can only ever be described, at the very least, as a complete
whitewash. The truth, which must have been known at the top of the two organisations, had to be
stopped from coming out and the NFSP believes that the instruction of Mr Ismay’s report was
designed to do just that, and especially coming at the time that it did in 2010, when more questions
than ever were being asked and the campaign by some of those already prosecuted was underway.
The NFSP believes that the deception by the Post Office took two forms. The first was knowing the
Horizon system did not, in fact, have the integrity that Post Office was proclaiming it had and was
riven with bugs, errors and defects. The second was doing all that it could to prevent this from
coming out and the reputational damage it would bring, despite the cruel misery caused to those
being prosecuted. It is said by the NFSP that the deception wrought on it throughout the period
SUBS0000023
SUBS0000023
time — an organisation which had to and still has to deal with the Post Office on business delivery
models and contracts for its members in sub-post offices across the UK — was clearly also wrought
indiscriminately across other organisations, their personnel, and even among the Post Office’s own
employees. This deception was perpetrated without regard for the effect on any or all of them of
finding out they had been deceived and, in their different roles, having acted on either belief or
instruction in ways which they would not otherwise have done. It seems clear that there will be
evidence regarding this in Phase 4.
In order not to take up the Inquiry’s time repeating itself, the NFSP would particularly refer the
Inquiry to paragraphs 1- 14 of its Opening Statement which set out the background, history and
role of the NFSP more broadly and referred to the other representative organisations involved with
Horizon and their members, due to the fact that more than 40% of those prosecuted by the Post
Office were not sub-postmasters and not all the sub-postmasters were members of the NFSP. It
also described how, on the basis of the evidence from the Post Office, Mr Justice Fraser was left
with an erroneous impression of the role of the NFSP and the relationship between the two
organisations, which the NFSP hopes to correct in due course during the present Inquiry. The
Opening Statement is referred to, because when taken with the scale of the deception and breach
of trust described here, heard in the evidence and in the issues with current Post Office disclosure,
the NFSP of today simply does not know the extent to which it can trust the Post Office in its
present day dealings for its members. This is said to show the wider ripple effect which continues
to this day, another element of the Horizon ‘legacy’.
It appears to the NFSP that the Post Office’s deception was, at least in part, motivated by the desire
to keep the ability to recover debt and litigate, in both civil and criminal courts, against
postmasters. Fujitsu/ ICL’s (referred to throughout here as Fujitsu for ease) deception may have
been motivated by the fact that they knew the product was not fit for purpose (as shown by the
evidence of David McDonnell in Phase 2) and were concerned about the financial repercussions
and reputational damage.
Phase 3 evidence has shown that there was knowledge about Horizon integrity issues at the top
levels of the Post Office and Fujitsu, and an almost complete, possibly blind, acceptance of the
myth of Horizon integrity in the middle to lower levels of these organisations. The NFSP was
disturbed to hear how the myth of Horizon acceptance was actively encouraged by the Post Office
to its own employees and those external to the organisation in the face of contrary evidence, as
evidenced by the Rod Ismay report, mentioned above, and the discussions surrounding that.
SUBS0000023
SUBS0000023
10. The NFSP is shocked to hear the evidence of various Post Office and Fujitsu employees indicating a
culture where postmasters were assumed to be either dishonest or incompetent (to the extent of
those seeking help effectively being mocked, either directly or behind their back). It was especially
shocking to hear Amandeep Singh’s evidence about the toxic and at times ageist and racist culture
of the Horizon Helpdesk Support Desk (HSH), given that these were the very people who were
supposed to be helping sub-postmasters and others, and particularly given the demographics
involved.
11. As touched upon in Mr Stein KC’s submissions, the evidence has made clear that the Post Office
used the IMPACT programme to further place the onus of identifying and rectifying bugs, errors
and defects upon postmasters, managers, Post Office employees and others, whilst simultaneously
removing the ability for them to dispute discrepancies. The Post Office and Fujitsu were in control
of all other sources of evidence that someone could use to dispute a discrepancy, but did not
routinely provide access or disclose evidence such as ARQ data in relation to investigations or
litigation. The NFSP looks forward to the Inquiry investigating this further in Phase 4. In terms of
disclosure, the NFSP raises the question as to whether the Post Office have a pattern of failing to
fulfil their disclosure obligations that (with reference to the continued non-disclosure of clearly
extant HSH scripts) which continues to today.
12. The current issues with disclosure may cause the Inquiry to question whether the culture issues
identified of Phase 3 within the Post Office and Fujitsu of dishonesty and concealment of the truth
continues to this day, and the NFSP notes that the Inquiry will examine this in Phase 7.
13. As always, the NFSP remains ready to assist the Inquiry in any way it can and looks forward to being
able to provide information that reflects the reality of its involvement.
High levels of Post Office and Fujitsu aware of bugs, errors, and defects
14 The Phase 3 evidence has made it clear that the highest levels of Post Office and Fujitsu were aware
that there were bugs, errors and defects in Horizon from the initial stages, throughout the project.
There are various instances highlighting this, including David Smith’s evidence, which provided an
example of an issue regarding Horizon not passing customer data, which “undoubtably... would
have been reported through to board level... there was a process of directorate reporting into the
1 POLO0028591, POLO0028589, POLO0090839, POLO0043705, WITN04600104 p10, POLO0055410,
SUBS0000023
SUBS0000023
board... But I would have expected it, in any case, to have been raised by the Ops Director with the
Managing Director anyway... [the board] would be the executive management team of the Post
Office.”* Both Mark Burley and Alan D’Alvarez made it clear that they were aware of errors, bugs
and defects within Horizon before and throughout rollout of HNG-X. The Second Sight Interim
Report stated that Post Office disclosed and therefore were aware of defects discovered in 2011-12
that had impacted 76 branches.’ The meeting discussing the receipts and payments mismatch bug,
noting its potential impact on accounts and prosecutions was attended by high level employees of
both the Post Office and Fujitsu. ‘Finally, as discussed in further detail below, it is clear that specific
bugs, errors and defects were brought to the attention of high-ranking members of the Post Office
in relation to prosecutions.>
15. It is submitted that the evidence shows it was known within Post Office and Fujitsu that EPOSS was
a problematic system, but suggestions to rewrite this were not taken up due to costs and delays, a
decision which was taken by the management team of ICL Pathway with the promise that ongoing
monitoring would be implemented.° It is submitted that this highlights two issues, firstly that the
management team were aware of ongoing issues, and secondly, and Phase 3 evidence has shown,
no such monitoring actually occurred, allowing these problems to go unchecked.’ Post Office were
also aware of the inherent weakness of what was designed, and “you could put a plaster over them
but if you really wanted to put something different in place, then you would have to start again.”*
16. On the basis of Phase 3 evidence, both the Post Office and Fujitsu at high levels of the organisations
were aware of issues with Horizon’s integrity. It is submitted that it was also clear that the Post
Office and Fujitsu sought to suppress the knowledge of this, as discussed further below. The NFSP
suggests the Inquiry may wish to consider why such a position was taken in relation to this
knowledge, and if this was because:
2 David Smith transcript, page 33
3 POLO0029744
*POL00028838
5 POLO0091384
6 Stehpen Muchow transcript, pp 58 and WITNO4600104, page 10
7 Mik Peach transcript, pages 138 - 141
8 David Smith transcript, page 24
SUBS0000023
SUBS0000023
a. the Post Office wished to maintain its ability to recover debt and litigate, and addressing
Horizon’s integrity issues would undermine this; and/or
b. Fujitsu, knowing that the product was defective, sought to minimize their reputational
damage, and financial implications.
Post Office narrative of Horizon Integrity
17. The NFSP considers the Phase 3 evidence appears to show that the Post Office promoted a self-
fulfilling myth of Horizon integrity. A clear example of the Post Office actively perpetuating this
narrative is the Rod Ismay Report, as commissioned by David Smith, Managing Director of Post
Office Limited.’ It is submitted that the evidence shows this report was instructed in order to be a
one-sided list of the reasons that people could have faith in Horizon’s integrity — and this was Mr
Ismay’s understanding of the scope of this report.’ Despite this, the report describes itself as an
“objective, internal review of POL’s processes and controls around branch accounting.” Mr Ismay is
clear now that the assertions made in the report were untested and were drawn from
conversations with various departments, rather than formal analysis."! The report was then widely
circulated, including to the board of the Post Office.!? Shortly after this report was finalised, Mr
Ismay was made aware of the payments and receipts mismatch bug and yet did not amend his
report.” In fact, individuals in senior management of the Post Office on the distribution list for the
Ismay Report were made aware that, contrary to the report, Fujitsu could insert entries into a
branch account remotely, and yet there was no move to amend this to reflect the reality of the
situation.“*
18. It is submitted for the NFSP that the Ismay Report demonstrates that the Post Office were
essentially commissioning what might fairly be described as their own propaganda about Horizon
integrity, creating the impression for their employees to rely upon that concerns raised had been
® POLO0026572, Rod Ismay transcript, 11 May, page 101
10 Rod Ismay transcript, 11 May, pages 101 - 104
11 Rod Ismay transcript, 11 May, page 108
2 Rod Ismay transcript, 11 May page 120
3 Rod Ismay transcript,12 May, page 52 and FUJO0081544
+4 Rod Ismay transcript, 12 May, page 74 and POLO0091384.].
SUBS0000023
SUBS0000023
investigated, when, in fact, no actual analysis had been undertaken. It is submitted this also shows
that high ranking individuals within the Post Office had knowledge of bugs, errors, and defects
within Horizon, but instead of addressing these, chose to cover these up by encouraging the
narrative of Horizon integrity. It is unclear why Mr Ismay, who had a background in audit,
considered this an appropriate report to write and to not revise in the light of contradictory
evidence.
19. What is clear now to the NFSP is that the mantra that Horizon was robust and had integrity was
pervasive and actively promulgated at a senior level, especially to the middle layers of the
management structure and below. This narrative seems to have been accepted without question.
This has been echoed by individuals working within audit,° training,© NBSC,”” business process,”*
network,’? problem management”°and contracts and services.” As stated by Shaun Turner, “I think
there was a baseline assumption within POL that Horizon was a robust accounting system... I believe
that this underlying assumption meant that POL was not always as attuned to the concerns raised
by Postmasters as it should have been.””*
20. It also appears as if Post Office employees were concerned that stepping out of line with that
message would have a detrimental effect to their careers.”? This seems to have produced a culture
of assuming that someone more qualified has looked into Horizon, and therefore it can be relied
upon [Brian Trotter, page 19]. **
21. The NFSP is concerned that the Post Office deliberately promoted this narrative despite evidence to
the contrary. For example, the messaging that was circulated in light of the Panorama
45 Chris Gilding transcript, page 18
16 Rita Palmer transcript, page 33
17 WITNO4640100, page 88
18
I page 8
19 Gayle Peacock transcript, pages 115 - 116
20 Gary Blacburn transcript, page 197
21 Brian Turner transcript, page 16
2 WITNO4640100, page 88
23 Gary Blacburn transcript, page 235
24 Gayle Peacock transcript, pages 111 and 145
SUBS0000023
SUBS0000023
programme,”® was contrary to the knowledge that the Post Office had of extant defects and that
Post Office employees then relied upon in their interactions with postmasters.?°
22. Additionally, the interim report of Second Sight noted that Post Office had disclosed that it had
discovered defects in Horizon Online that had impacted 76 branches.?’However, Gayle Peacock, the
Head of the Branch Support Programme, which was set up to address some of the conclusions of
the Second Sight Report, was not provided with a copy of the Second Sight Report, and instead
based her understanding of the report on a summary provided by Angela van den Bogerd. Various
issues flagged in the Second Sight report had not been brought to her attention, such as the
defects.”* As such, the Branch Support Programme did not look into these defects. The NFSP
consider this is an example of the Post Office selectively providing information to people, in order
to maintain the myth of Horizon integrity and not to draw attention to the known defects in
Horizon.
23. The NFSP considers that Fujitsu is also complicit in propagating, promulgating and perpetrating the
myth of Horizon integrity. The evidence of Andy Dunks, which makes clear, despite his belief that
he was technically qualified to provide statements in relation to a number of criminal investigations
and prosecutions that there was nothing leading him to believe that the system was operating
improperly:
a. He did not have the knowledge to make those statements;
b. he treated the SSC as “subject matter experts” on Horizon, but did not declare that he was
basing his statement regarding the content of relevant helpdesk calls on their advice;”°
c. he prepared his statements in line with templates provided by his predecessor, and
assumed he could use a standard line saying that none of the calls to HSH relate to faults
that would have had an effect on the integrity of the information held on the system;?°
25 WITNO6380102
26 Anne Allaker transcript, page 54
27 POLO0029744, page 5
28 Gayle Peacock transcript, pages 106 - 111
29 andy Dunks transcript, page 24
30 [bid at page 63
SUBS0000023
SUBS0000023
d. he did not undertake any analysis of his own; and
e. even though he was aware of software bugs within Horizon, he did not think those caused
reconciliation errors, and continued to produce statements despite this?
24. The NFSP believes that it is simply unacceptable that Fujitsu instructed an individual who was
unthinkingly using a standard line about faults not affecting integrity to produce statements in
various criminal investigations and prosecutions.” It could be argued that this was a placing of an
underqualified person in a role to produce witness statements without any guidance regarding
duties to the court, so that Fujitsu could support Post Office with prosecutions without bringing
attention to the known defects in Horizon. The fact that Mr Dunks produced these statements
feeds into the narrative that no one believed there were defects affecting integrity in Horizon, and
that multiple investigations apparently uncovered no issues. In reality, the evidence was that Mr
Dunks’ ‘investigations’ consisted of asking the same team that likely investigated any issues
thought to be systems errors to explain what the call meant. In effect, no independent
investigation was done in order to prepare these statements to the court other than any initial
investigation by support lines. This shows a complete disregard for the truth in favour of prioritising
litigation, as further discussed below.
25. The NFSP believes that the Phase 3 evidence shows that it is clear the higher levels of the Post
Office were aware of the defects in Horizon and that there was an unwavering belief in Horizon by
individuals within the middle layers of the Post Office structure and those who reported to them. It
can be said this shows an active effort on the part of the Post Office to suppress the knowledge of
defects instead of addressing issues. Given that many of these same individuals were
communicating with external organisations such as the NFSP, (e.g. Brian Trotter), or in charge of
programmes (e.g. Gayle Peacock), or managing the teams that were supporting or investigating
postmasters, managers, Post Office employees, and others (e.g. Gary Blackburn and Brian Trotter),
the NFSP suggests that Inquiry may wish to consider if there was a deliberate effort on the part of
the Post Office to appoint people to these positions who would not ask questions, or who were
thoroughly convinced of the narrative of Horizon’s integrity.
26. The NFSP considers that it could be argued that the evidence of Phase 3 has shown examples of the
Post Office and Fujitsu intentionally filtering information/whitewashing matters to those in charge
31 [bid at pages 86 and 90
32 [bid at pages 62 - 63
SUBS0000023
SUBS0000023
of running programmes/ producing statements so that those individuals did not have concerns
about Horizon’s integrity, or deliberately appointing people who are unsuitable for a role so they
can continue to propagate that belief without question.
Lack of information sharing both internally and externally
27. The NFSP believes that the Phase 3 evidence has made it clear there was widespread and
systematic suppression or filtering of information both between teams and organisations. There do
not appear to have been consistent and formal monitoring processes or follow up processes. It can
be said from the evidence that the information regarding:
a. the issues with Horizon;
b. the extent and effect of problems; and
c. how branches and those working within them had been and would be affected by issues
and fixes
was filtered at every level, leading to a lack of understanding about these issues.
28. It is submitted the evidence in Phase 3 has shown that processes within Fujitsu aimed to filter the
number of calls reaching 3 and 4" line support, i.e. the lines of support with sufficient technical
understanding to be able to identify and address systems problems. In other words, the system did
not easily allow for callers with problems to reach those who might actually be able recognise it
was a problem with the system, as opposed to a problem caused by the caller themselves.
29. According to Amandeep Singh’s evidence,*? HSH staff were dissuaded from escalating calls to
engineers due to feeling belittled and there being animosity between these lines of support.
Furthermore, if HSH could not get to the bottom of a discrepancy, rather than escalating this to a
higher level of support, the advice of HSH colleagues would be to write off the issue as a
discrepancy.*
30. There was a similar filtering process at the NBSC helpdesk, as shown by documents
POLO0114858 — Checking a Discrepancy (which states, “Please note: Discrepancies are not usually
caused by system problems they are more likely to be caused by user error... checks need to be
33 Amandeep Singh transcript, pages 67 - 68
34 Ibid at pages 75 - 76
10
SUBS0000023
SUBS0000023
made before referring SPMR to HSH to investigate system fault.” After listing all the checks, it
continues, “If SPMR has checked all above advise to roll over with the discrepancy and make good.
The discrepancy may come back in the form of an error notice at a later date,” with no reference to
escalation to HSH or other support lines) and POLO0114814 — Instructions on discrepancies (which
states: “/f nothing can be found the office will have to accept the discrepancy. Make good and roll
over.” Again, this does not refer to other support lines). It is submitted that the instruction to
accept the discrepancies and make good by both first line helpdesks may have served to prevent
discrepancies caused by errors, bugs, and defects in Horizon from being escalated to higher levels
of support, preventing these levels from having sight of the full extent of issues.
31. It is submitted that there was a lack of information sharing between teams, and a culture where
items were not monitored or followed up when passed to another team which seems to have been
pervasive throughout Fujitsu. Anne Chambers agreed, to an extent, that there was a series of
hermetically sealed cells within Fujitsu, the principal aim of which was to get a service ticket off
their desk.*> When problems were identified, these were not communicated between teams
consistently, e.g., Richard Roll provided an example where an engineer made a mistake with a
batch of laptops, which were sent out to branches, but no one outside of the laptop building team
was informed of this known problem. Even after one of these laptops was identified and replaced,
there was no movement to recall the faulty laptops.*°
32. FUJO0070841 provides an example of PEAK, where it was decided that action to address the issue
would be taken should the frequency of occurrence becomes more than once per month. The NFSP
considers it was disturbing to hear then, that to Anne Chambers’ knowledge, no such monitoring
was carried out, and at a wider level she was unaware of any formal process of recording further
occurrences of PEAKs, other than logging on a KEL, which was up to the discretion of SSC staff?”
This highlights that when unexplained errors were found, and further investigation was dependent
on recurrence, there was no formal process to detect recurrence. This means that further
investigation would be unlikely; even if there was recurrence, there was no formal process for SSC
to record this, or even identify that they were supposed to be recording this. This could lead to a
35 Anne Chambers transcript, 2 May, page 154
36 Richard Roll transcript, pages 42 - 43
37 Anne Chambers transcript, 3 May, pages 111-112
11
SUBS0000023
SUBS0000023
false conclusion that this PEAK was a one-time occurrence, and further investigation and a fix was
not necessary.
33. Monitoring of the scale of a problem,** whether to check the message store for the entire estate if
a problem was affecting other counters (at least, until Horizon Online),?? and metanalysis of
recurrences” were all left to individual discretion of SSC staff ad hoc, rather than having any formal
process. For the NFSP, this lack of consistency all seems to suggest that Fujitsu’s support lines were
not getting an accurate picture, or at least were not accurately recording the extent of issues with
Horizon. This, added to the fact that HSH may have been dismissing calls relating to Horizon errors
as discrepancies, seems to show that due to its own processes Fujitsu had an inaccurate picture of
the extent and effects of Horizon errors.
34. It is submitted that this idea of hermetically sealed units, and suppression of information also
affected Fujitsu’s communication with the Post Office. It is clear from the evidence that there were
instances when Fujitsu would hold off telling the Post Office about issues with Horizon until they
found a workaround. The evidence of Alan D’Alvarez was that presenting Post Office with the issue
and the workaround would be better than just presenting them with the issue, so they did not
receive a barrage of questions." The specific instance referred to in the evidence of Mr D’Alvarez is
especially concerning, as it affected ARQ data, including where a court action was ongoing,*? and
occurred just before acceptance of Horizon Online. It was indicated by Counsel to the Inquiry that
there is an email dated 30 June 2010, which indicates that the Post Office was not notified of the
problem with the ARQ data by Fujitsu until that day, the day upon which rollout was entered into. It
is submitted that an interpretation of this is that Fujitsu deliberately did not disclose this problem
until after acceptance, which is concerning. It is also concerning for the NFSP to think that people
working in branches may have been affected by such errors, with no knowledge of what was
happening, and this data may have been relied upon in prosecutions and civil court actions due to
this failure in communication.
38 Anne Chambers transcript, 2 May, pages 114 - 115
39 Ibid at page 105
40 Anne Chambers transcript, 3 May, pages 111
41 Alan D’Alvarez transcript, pages 137 - 138
42 bid at page 130
12
SUBS0000023
SUBS0000023
35. This is further exemplified by the fact that Fujitsu were aware of the Riposte lock problem since at
least the end of 2000,*? but the Post Office were kept in the dark, whether intentionally or not,
about this for nearly six years.
36. Another example is the receipts and payments mismatch bug, which had probably been present
since day one of Horizon Online.**Post Office did not have a meeting about this until September
2010, despite the problem likely pre-dating May 2010, as Fujitsu had taken so long to react and
escalate the issue to them.** Given that, according to Andrew Winn, that “[Product & Branch
Accounting] were pretty much dependent on Fujitsu alerting us to what branches were affected by
a receipts and payments mismatch”,’” it is submitted this is another concerning example of Fujitsu
delaying in telling Post Office about matters that may have had an impact on civil and criminal
cases against postmasters, managers, Post Office employees and others. Two years after this issue
was identified, there was no shared understanding between Post Office and Fujitsu about which
branches had been investigated and impacted, and what had happened.”®
37. It is submitted that this further demonstrates that Fujitsu was aware of ongoing issues, and also —
due to their own processes — uninformed about the extent and impact of these issues as they were
not properly followed up or investigated. It was troubling to hear from Andrew Winn that the
evidence suggests that Post Office and Fujitsu were still not working through proper procedures in
2015.8
38. As an aside, the NFSP notes that the 5-year delay in communicating the Riposte lock error was
contributed to by a failure in communication with Escher, the owner of the Riposte software,°° and
questions whether Escher should be asked to give evidence to the Inquiry about their involvement.
+3 Anne Chambers, 3 May, page 146
44 Ibid at page 152
45 POLO0029084, page 1
46 POLO0028838, page 3
47 Andrew Winn transcript, page 67
“8 Ibid at page 102
“9 Ibid at page 180
59 Anne Chambers, 3 May at page 150
13
SUBS0000023
SUBS0000023
39. It appears from the evidence that Fujitsu seem to have assumed that once a system error was
identified as the cause of discrepancy, the Post Office would ensure that no action was taken
against postmasters in relation to the discrepancy. However, there were instances where a system
error had been identified as the cause of a discrepancy but the postmaster continued to be subject
to an investigation regarding this.*! It is submitted that there was a culture of assuming Post Office
would inform postmasters, managers, Post Office employees and others, and that the effect of this
assumption enabled the Post Office to conceal errors from them and to continue with unjust
prosecutions and debt recoveries.
40. It is submitted that Fujitsu also failed to communicate proactively regarding workarounds where it
was identified branches would likely require them, instead of waiting for the branch to call HSH.*
In one case, Fujitsu did not tell the postmaster about the fact that data had been inserted into their
counters to correct an error, and a discrepancy of the same size as the correction was now
showing.
41. Even if there was understanding of the issues with Horizon, notwithstanding that these were not
investigated or monitored as they should have been, the NFSP considers the evidence
demonstrates that there seems to have been a policy instigated by the Post Office where users of
Horizon should generally not be told about systems errors. While Fujitsu’s general attitude seems
to be that it was the Post Office’s responsibility to communicate with users of Horizon,* it is clear
that it was aware that the Post Office did not want to disclose information about systems errors,
even if this had affected the branch.
42. This is not to say that the Post Office was not aware that Horizon lacked integrity. The evidence in
Phase 3 has shown that senior management and the Board of the Post Office were evidently aware
of errors, bugs, and defects having an effect on Horizon integrity. The evidence of Alan D’Alvarez
made it clear that where workarounds were proposed, Post Office had to agree whether they were
acceptable,®© and therefore even though there may have been delays in informing the Post Office
5 Anne Chambers, 2 May, pages 136 - 138
52 Alan D’Alvarez's transcript, page 48
53 Anne Chambers, 3 May, pages 49 -52
5 Anne Chambers, 3 May, pages 85 - 86, Alan D’Alvarez, page 48
55 FUJ00086462, page 1
56 Alan D’Alvarez transcript, page 16
14
43.
44.
45.
SUBS0000023
SUBS0000023
of issues, the Post Office would have still been aware of them, and the fact that a fix had not been
applied.
It is submitted that evidence demonstrates that the Post Office also had a culture of heavily
filtering knowledge through various organisational layers. This is self-evident by the suppression of
the knowledge of issues with Horizon integrity, and the corresponding propagation of the myth of
integrity in the middle and lower levels of the organisation, but also extends to the Post Office’s
communications with external individuals.
The NFSP is very concerned to hear that the policy was to not tell users of Horizon when they had
been affected by a bug, error, or defect.*” The evidence from Phase 3 shows that the Post Office
deliberately avoided telling users of Horizon when their branch had been affected by a bug, error or
defect in order to avoid questions about Horizon integrity, as per Andrew Winn’s evidence.* The
NFSP is especially concerned that the Post Office, knowing that postmasters could be liable for
discrepancies, did not inform postmasters about how to identify and workaround such identified
bugs, errors, and defects in order to maintain the illusion of Horizon integrity.°? It is the NSFP’s
position that, especially given that the Post Office was actively moving the onus of identifying
issues, and understanding the reasons for discrepancies onto users of Horizon (as discussed below),
the lack of communication regarding these errors was, in the view of the NFSP, underhand at best.
The Post Office also did not disclose such bugs in court proceedings, for example, the lack of
disclosure of the receipts and payments mismatch bug to Seema Misra’s defence team or the court,
where the Post Office was concerned that, “/t could provide branches ammunition to blame Horizon
for future discrepancies.”lt is submitted for the NFSP that this amounts to a fundamental
dishonesty for the Post Office to conceal this information, and to not make users of Horizon of
aware of bugs, errors and defects that could affect the integrity of their accounts, and that they
therefore could be liable, or prosecuted for.
57 Richard Roll transcript, pages 25 - 26
58 Andrew Winn transcript, pages 70 - 73
59 [bid at page 73
6° POLO0028838 page 2
15
SUBS0000023
SUBS0000023
Culture of assuming user error or postmaster dishonesty, and the moving of the onus onto postmasters
46. It is submitted that it is clear from the Phase 3 evidence both the Post Office and Fujitsu had
knowledge of bugs, errors and defects within Horizon from the pilot stages, throughout rollout, and
throughout the Horizon project. As well as never having been made aware of this, the NFSP was
extremely concerned to hear the pervasiveness of the culture throughout the Post Office, and
extended to Fujitsu in assuming that discrepancies were due to user error and/ or postmasters
being dishonest.
47. The NFSP was absolutely shocked at the evidence of Susan Harding, that in her role of Business
Process Architect for IMPACT, and in her role in the Post Office more generally, she operated on
the basis that postmasters must have been placing sums on the suspense account for nefarious
reasons,® and it seemed from her evidence that — despite what we now know - this seemed to still
be her view today.
48. The NFSP considers the language used, such as that the use of the suspense account “covered up
losses”, and “historically it was used by SPMs to ‘hide’ discrepancies” shows to the NFSP that
behind the scenes of an organisation, which they worked with in good faith to benefit sub-
postmasters across a range of business and practical issues, there was a deliberate culture of
mistrust of postmasters/their assistants and even their own employees. Rather than postmasters
and others being a group of people trying to understand a new accounting system throwing up
unexpected shortfalls and other issues, there was instead a built-in assumption of their guilt or
dishonesty from the outset. And it turned out this was an accounting system was one where the
postmasters and others were not being informed of factors that may affect their ability to
understand what the issue with their accounts was. The fact that this was the position adopted by
the Programme Manager of the IMPACT project, and that this was her understanding of the
requirements provided to her by the programme board, shows that the belief of guilt and
dishonesty among postmasters and others existed within the wider Post Office organisation. This is
also seen in the evidence of Stephen Grayston.” The NFSP considers on the evidence that the Post
Office evidently knew, and perhaps intended, that the IMPACT programme had the effect of placing
81 Susan Harding transcript, page 32
82 Stephen Grayston transcript, page 27
16
49.
50.
51.
SUBS0000023
SUBS0000023
the onus of identifying and rectifying errors onto postmasters, managers, Post Office employees
and others.
The NFSP was also shocked to read and hear the evidence of Amandeep Singh regarding the toxic
environment of the HSH, and the racist, ageist, and overall unacceptable comments made. His
evidence shows that the assumption of postmaster dishonesty was shared amongst the very
people who were supposed to help postmasters and others working with Horizon to understand
what was going wrong with their counters.” It therefore seems that postmasters had to fight a
battle for their problems to be taken seriously by the people who were supposed to be their first
point of contact for issues.
The evidence also shows that throughout Fujitsu and the Post Office, where support or others
could not find the underlying reason for an issue, the default position was that it was due to user
error/ there was no system fault, and recorded as such.® If there was insufficient evidence of a
system fault, the call would be closed with a reference back to the postmaster for further
information.® This put the onus back on the non-technically trained postmaster to find and resolve
the fault, and shows a cyclical structure of support.
The NFSP is troubled to note that this assumption of user error/ no system fault seems to have
persisted even when evidence gathered from the branch indicated otherwise. For example, Anne
Chambers talked about a PEAK that she closed as being due to user error, despite an auditor and
trainer coming in and experiencing the same error. Ms Chambers closed this PEAK under the user
error category without trying to find out further information from the trainer and auditor who had
been at the branch and experienced the issue. It was her evidence that it was “not an option” to
work with them to get to the bottom of this issue.” This is further exemplified by a PEAK in which
phantom transactions were appearing, engineers were sent to the branch and saw these
transactions, but despite this, it the call was closed as no fault in product as: "Phantom
®3 Stephen Grayston transcript, pages 18 - 19
6 WITN06660100 and Amandeep Singh transcript, page 72.
6 Rod Ismay transcript, 11 May, page 37; Gayle Peacock, page 85; Gary Blackburn, page 218; Amandeep Singh, pages 75 — 76; Anne
Chamers, 2 May, page 188; Andrew Winn transcript, page 126; and POLOO114858
86 Anne Chambers, 2 May, page 82; ad Stephen Parker transcript, apges 42 - 43
87 Anne Chambers, 2 May, pages 191 - 185,
68 POLO0028743, page 5
17
52.
53.
SUBS0000023
SUBS0000023
[transactions] have not been proven in circumstances which preclude user error. In all cases where
these have occurred, a user error relate cause can be attribute to the phenomenon.I am therefore
closing this call as no fault in product’ .©°This shows that even when other people witnessed the
bugs, errors and defects that were causing discrepancies, Fujitsu still did not investigate this
properly. The NFSP is very concerned that the attitude seems to have been “we can’t find the issue,
and therefore there is no system error,” especially since the Post Office relied on such conclusions
to hold Horizon users liable and raise civil/ criminal proceedings against them.
This seems to have been accompanied throughout the Post Office by an attitude that no matter
how implausible the loss seemed, the postmaster was still liable for the loss if a discrepancy was
showing.”° This seems to the NFSP to be an example of not caring about the human impact, or
whether debt was truly owed to the Post Office, as long as it could recover the debt apparently
owed. It is clear that it was not in the Post Office’s interest to properly investigate the discrepancies
or the causes of these, as further discussed below. It is concerning that the Post Office’s Conduct of
Criminal Investigations states that the Security Manager “must not overlook the fact that a fair
investigation is there to establish the truth as well as substantiate the allegation,” (emphasis
added)”*, showing the Post Office’s approach of ‘guilty until proven innocent’ when it comes to
investigations and postmasters generally.
These attitudes of:
a. assuming that postmasters are dishonest;
b. assuming that discrepancies were due to user error; and
c. assuming that, no matter how implausible a loss, the postmasters was contractually liable
seem to have formed the basis for the systematic removal of the postmaster’s ability to interrogate
Horizon and provide evidence that a discrepancy was not real or not their fault. There was an
increase of the onus onto the users of Horizon to identify and rectify errors”? and decrease in
facilities that enabled users of Horizon to do so, such as the removal of the suspense account, the
reduction of staff at Chesterfield, and the lack of availability of evidence that a user of Horizon
6 Ibid at pages 9 - 10
79 Anne Allaker transcript, pages 56-57; Brian Trotter transcript, page 50; POLOO114814; and WITNO3980100, page 7
74 POLOO038853
” Philip Boardman transcript, pages 51, and 71-72
18
54.
55.
56.
57.
SUBS0000023
SUBS0000023
could use to prove that the discrepancy was not their fault (whether by removal from the branch
due to suspension, or the limitation of the available data).
As highlighted by John Peberdy,”? by March 2001 there was £10 million in suspense accounts
whereas 18 months before there was £2 million. The NFSP is concerned that, rather than
investigating this issue and whether this 5-fold increase was due to system errors, the Post Office
was more concerned with ensuring prompt debt recovery, and therefore removed the suspense
account. This essentially created a situation where Horizon users had to accept a discrepancy in
order to roll over, and the manner in which Horizon users could raise errors became integral for the
accuracy of the data produced by Horizon.’* However, as discussed above, the errors raised by
Horizon users were not properly investigated and were assumed to be the fault of the system user.
The IMPACT programme in removing the suspense account essentially removed the key facility that
allowed the Horizon user to investigate the issue before being subject to the Post Office’s debt
recovery processes.
As Counsel for the Inquiry has identified, the Post Office put the onus on the postmaster to provide
evidence that the discrepancy was not their fault, but— as they were locked out of their branch
upon suspension — there was no way for them to recover evidence to support their position.
This lack of available evidence available to Horizon users was also caused by Fujitsu and Post
Office’s policies on data retention and the retrieval of ARQ data. The Inquiry has heard that, as
counters did not keep data that was over circa five weeks old, the only source of data when
investigating older issues was ARQ data.’° The Post Office did not routinely retrieve this data, due
to cost” and it is clear that not all of the relevant evidence was disclosed in the process of
investigations and litigation.
It is submitted that Fujitsu themselves are complicit in this lack of evidence. After providing
evidence regarding the Lee Castleton case, Anne Chambers identified and flagged to the SSC
Manager, Mik Peach, and the Head of the Security Team, Brian Pinder, that Fujitsu had not
disclosed much of the relevant data, and she questioned whether there was an obligation to
NFSPo0000513 ‘Page 15
74 Philip Boardman transcript, page 51
75 Anne Chambers, 2 May , pages 163 - 164
76 David Smith transcript, page 52
19
58.
59.
60.
SUBS0000023
SUBS0000023
disclose this.’” She suggested ways in which Fujitsu’s process in similar cases could be improved,
but this seems not have been acted upon, other than to forward her concerns to the Security
Manager with no indication of follow up or alteration to the process.”
It is submitted that the evidence shows the Post Office did not train postmasters sufficiently
regarding balancing, and the identification and rectification of discrepancies, which has become
apparent throughout the evidence, and of which the Post Office was aware due to the Christmas
Report.’? The Horizon user was therefore not provided with the tools required to identify,
understand, and ensure rectification of discrepancies, but was liable for their inability to do so. The
organisation that failed to train them and removed facilities enabling disputes was the same
organisation that stood to benefit from a postmaster’s inability to prove their position.
Returning briefly to the HSH, the Inquiry may want to question the limitations of the training that
was provided to staff, and whether they were equipped to be advising this group — given their
variations in digital literacy and considering the limitations in digital literacy at the time. Given that
when faced with discrepancies, HSH staff were essentially assisting Horizon users with
bookkeeping, despite having no training in this, the NFSP considers the Inquiry may also want to
consider whether HSH had sufficient training to be advising on balancing, given the lack of training
about Post Office processes and balancing, with the focus being on basic transactions.®° The fact
that Elizabeth Evans-Jones told the Inquiry that Knowledge Base articles were present so that HSH
staff did not require in-depth knowledge to determine whether a call was caused by user error or
software is concerning,*/as it indicates that the first line of support for Horizon users were
screening out calls for escalation without a proper understanding of what they were screening out,
and on a tick box basis.
The NFSP’s position is that all of this meant that the Post Office was — through its own policies and
reforms of Horizon — able to place the onus for the identification and rectification of errors on
Horizon users while severely limiting the user’s ability to investigate this. The Horizon user’s routes
for recourse were run by individuals who thought they were dishonest, assumed the discrepancies
77 FUJ00152299
78 Mik Peach transcript, pages 92 - 94
79 NFSP00000261, POLO0104482 at page 32,
80 Amandeep Signh transcript, pages 62 and 74; and WITNO6660100, page 1
51 Elizabeth Evans-Jones transcript, page 49
20
SUBS0000023
SUBS0000023
were due to their error, and themselves were poorly equipped to advise Horizon users. The Post
Office evidently either did not consider or did not care about the human impact of structuring a
system so they and their provider, Fujitsu, controlled the ability to access the information required
to investigate matters, and the people who they made liable for things going wrong in the systems
had little to no access to tools to help them investigate matters.
61. For the NFSP, the outcome of this is that a Horizon user with a discrepancy essentially had no
source of information to dispute the discrepancy other than the Horizon data that showed the
discrepancy, other sources of information being available. Only the Post Office could request this
further information from Fujitsu, essentially giving them a monopoly on the evidence available for
investigations and litigation.
Prioritising meeting targets, lowering costs, litigation, and debt recovery over the truth
62. For the NFSP, this all demonstrates that throughout the relevant period, the Post Office and Fujtsu
prioritised the meeting of targets, the ability to litigate and the ability to recover debt from Horizon
users over the truth. It is clear from the above that both Post Office and Fujitsu knew about errors,
bugs, and defects within Horizon that could cause discrepancies. Instead of properly investigating
and addressing these, it is submitted that the evidence shows that knowledge of these issues was
likely suppressed and certainly was not widely communicated, favouring an approach of asserting
the Horizon was robust.
63. The evidence indicates that more focus was placed upon meeting targets and keeping costs down,
rather than ensuring that the right thing was done. Brian Trotter described pressure from the Post
Office to recover debt due to performance-related targets,®* and Alan D’Alvarez described a similar
pressure about ensuring that the pilot went according to timescale.®*
64. The NFSP was concerned to hear that the primary aim of the SSC was to “keep the system up and
running so that it worked and so that Fujitsu didn’t suffer any penalties’ and that “it was widely
accepted that the underlying or root cause was that the system was crap,” but the necessary
rewrite would not happen due to cost and resource restrictions.™ This indicates that the concern
82 Brian Trotter transcript, page 52
83 Alan D’Alvarez transcript, pages 22 - 23
% Richard Roll transcript, pages 21 - 22
21
65.
66.
67.
SUBS0000023
SUBS0000023
for both Fujitsu and the Post Office was more about the cost of fixing errors, the meeting of targets,
and the financial implications of issues, rather than the human impact.
It was the evidence of Alan D’Alvarez that the integrity of the solution and its effect on the ability to
support the Post Office in litigation was “the number one priority that was given to [him] by Mark
Burley.”®® This is disturbing to the NFSP on two counts: firstly, the Post Office seem to be more
concerned with ensuring integrity to prosecute Horizon users than ensuring integrity to guarantee
that the accounts are correct and to enable the business to function correctly; and secondly, as
referred to above, Fujitsu would delay telling the Post Office about integrity issues. It is submitted
that as Fujitsu were aware that Horizon users were being prosecuted on the basis of this data, such
a delay (which was justified by it on the basis of wanting to avoid lots of questions from the Post
Office, but also in the example above meant that Post Office were not aware of issues until after
acceptance) must have been decided upon with the knowledge of the risk that Horizon users would
be prosecuted on the basis of incorrect data. It can therefore be argued that this shows that Fujitsu
were more concerned with ensuring they did not suffer any financial or relationship damage than
actually ensuring the integrity of the product, as supported by Richard Roll’s evidence.**
It had also become increasingly clear to the NFSP throughout Phase 3 that the decision to not
communicate issues to Horizon users was made to ensure debt recovery and the ability to raise
litigation against them, and make sure that they did not blame Horizon for discrepancies. The
deliberate lack of communication of issues to Horizon users was justified in Post Office
documentation on the basis that it could provide others with “ammunition” to blame Horizon,®”
ignoring the fact that if this ammunition was correct, it would show that the Horizon user was not
guilty, and the debt was not truly owed.
The NFSP is concerned that the evidence indicates that the Post Office and Fujitsu cared more
about retaining the ability to prosecute and recover debt, rather than ensuring that these
prosecutions were sound. This was enabled through the IMPACT programme despite Post Office
knowing that only 10% of discrepancies were real debt.®* This is shown throughout the evidence.
Examples include: a Fujitsu review of HNG-X discussed the Derby Error and noted that, “It would
5 Alan D’Alvarez transcript, page 95
86 Richard Roll transcript, page 31
87 Gary Blackburn transcript, page 198; POLO0105280, page 2; and POLO0105280, page 2
8 POLO0038878
22
SUBS0000023
SUBS0000023
seriously undermine Post Office credibility and possibly historic cases if it could be shown that a
discrepancy could be caused by a system error rather than a postmaster/ clerkaction. Most
importantly, the central database as the system of recorded would be called into question’. ®° An
email from the National Contract Manager for the North states, “/ am concerned if we swap the
processor now and the errors stop this could lead to (i) a claim that Horizon has problems in its
accuracy and fuel some of the recent press articles and (ii) the SPMR will claim that all previous
errors are down to Horizon and we have no way to disprove this if everything is resolved when the
new processor is installed... I have no doubt this individual is not the only one that uses Smartpost in
this way so we could end up with other claims in respect of this issue where we have insisted the
SPMR makes the loss good.”*° Also, an internal Post Office document discussing the payment and
receipts mismatch issue referred to the impact of the issue as “If widely known, could cause a loss
of confidence in the Horizon System by branches; Potential impact upon ongoing legal cases where
branches are disputing the integrity of Horizon Data; It could provide branches ammunition to
blame Horizon for future discrepancies” but does not mention that the Post Office may falsely try to
recover money from, or prosecute Horizon users.®! David Smith’s speculation as to why the Post
Office’s senior management did not want to know about issues flagged by their PR department
gives an indication of Post Office’s attitude towards prosecutions:
all that action [against
subpostmasters] was successful, why would you change anything?”®?
68. Additionally, it seems to the NFSP that the Ismay Report referred to above was commissioned in
somewhat dubious circumstances. It is submitted that the evidence shows in response to
challenges to Horizon, initial and full investigations into integrity issues were proposed.” In
response to this, the Head of Criminal Law for the Post Office advised that if such an investigation
was undertaken, it would not be appropriate for the Post Office to commence or continue with
prosecutions; that the investigation would be disclosable as undermining evidence; and that the
defence would say that such an investigation would show a lack of confidence in Horizon. This
advice was given in March 2010, with the Ismay Report — a one sided, but ostensibly objective
®9 FUJ00093031
%° po00021163
51 POLO0028838, page 2
°2 David Smith transcript, page 74
93 POLOO106867, page 3
23
SUBS0000023
SUBS0000023
report about Horizon integrity - being produced in August 2010. The NFSP believes that this makes
two things clear:
1. the Post Office avoided doing the right thing, and investigating Horizon integrity in order
to proceed with prosecutions; and
2. as a result of this, the Post Office then commissioned a report, which to an outsider may
seem like an objective review of Horizon’s integrity but was instructed with the apparent
aim of supporting the Horizon narrative.
70 The NFSP is very concerned, given the provision by it to the Post Office of issues reported by its
members, about the evidence that has emerged throughout Phase 3, as it was consistently assured
the issues were being passed on and that Horizon was robust — an assurance that they accepted in
good faith.
71. Unfortunately, the NFSP did not have, nor were they given, a clear view of the problems Post Office
and Fujitsu knew existed with Horizon. What is clear to the NSFP is that it and other unions were
victim to the filtering of information through Fujitsu and Post Office processes, all as described
above. As outlined by John Peberdy in his evidence during Phase 2, the NFSP were unable to
contact Fujitsu directly, and the NFSP could only have input through contact with the Post Office.*
The NFSP were therefore dependent on the Post Office to action any feedback they had.
72. It is also clear from the evidence in Phase 3 that the information given to the NFSP by Fujitsu and
the Post Office, meant the NFSP’s involvement was actively managed and limited by those
organisations, even although the NFSP did not realise this at the time. POLO0028137 states that
“the NFSP involvement would be kept to a minimum.” Andrew Winn’s evidence discussed an
example where the decision to force a sub-postmaster to pay may “open up a can of worms” and
“could potentially give the NFSP or an MP some useful ammunition.”®° The evidence of Steve Bansal
was that he was asked to give a talk to the NFSP in response to a period in which there had been a
number of incidents, but chose to talk about a problem that had been addressed, rather than the
significant receipts and payments mismatch bug. This was because, as he said: “/ was trying to give
some closure and some level of confidence that we were over that period.”
9 John Peberdy transcript, Phase 2, page 65
95 POLO0105280, page 2, and Andrew Winn transcript, page 175
24
SUBS0000023
SUBS0000023
73. The NFSP worked with the information it had and the assurances it was getting. For example, on
receiving complaints from members, the NFSP flagged issues with the training, especially on
balancing from at least 1998.°° The NFSP asked the Post Office for a ‘dispute button’ facility, so that
Horizon users would not be forced to accept transaction corrections that they did not agree with.
Unfortunately, this was rejected as this “would be abused”, and a robust dispute resolution process
was given as an alternative. Evidently, this did not come to fruition.” While the NFSP was in
contact with Don Grey, and did raise issues with him,®* his evidence was that if he received issues
concerning the system itself or integrity, he would pass it to his team to follow up with Business
Service Management, and that Post Office had processes in place to tackle problems.°? The NFSP
now knows that issues it passed on were dismissed, left unaddressed and/or passed from team to
team as part of such processes. Sometimes the NFSP’s calls with concerns would not even be
documented.’
74. While the NFSP was in contact with the Post Office, the individuals it was in contact with often
believed what turns out to be the myth of Horizon integrity, and therefore assured the NFSP of
this.1°? Where the NFSP managed to get a meeting with a high-ranking member of the Post Office
(who would have been aware of the bugs, errors, and defects in Horizon), it is clear that such
individuals were not being transparent to the layers within their own organisation about the
integrity of Horizon, and information appears to have been concealed from them. What chance
then was there that the truth, the whole truth and nothing but the truth would be told tothe
102 I,
NFSP? The Post Office informed the NFSP that issues were down to user errors,” issues with
%6 Bruce McNiven transcript, page 72, Don Grey transcript, page 18, and NFSP00000340, page 2
37 Andrew Winn transcript, pages 34 - 38
98 e.g. NFSP00000348, and Don Grey transcript, page 16
°° Don Gray transcript, page 21
10 Andrew Winn transcript, page 25
191 Colin Baker transcript, Phase 2, page 74
102 NFSP00000020
25
SUBS0000023
SUBS0000023
training;*or any problems were being dealt with in an appropriate manner.’™ The NFSP took this
in good faith.
75. It is submitted that in the context of evidence in Phase 3 which it can be argued identifies a
significant senior culture and organisational attitude that was apparently cultivated to prevent the
full picture of Horizon from coming out, it is unlikely that the NFSP or other external organisations
would ever have been given the true and accurate picture about Horizon, its errors, bugs and
defects.
76. The NFSP reiterates is concern, shock and disappointment at the evidence which has come out in
Phase 3 and sincerely regrets the reliance it placed on being told issues it was providing regarding
Horizon would be passed on, and that it believed in good faith that the Post Office was truthful
with it when it said that the Horizon system had integrity and was robust. As it turns out, that was
very far from the truth.
National Federation of Sub-postmasters
22 Windlesham Gardens
Shoreham-on-Sea, BN43 5AZ
193 NFSP00000261, page 3
104 Don Grey transcript, page 22
26