POL00330037
POL00330037
POST OFFICE LIMITE
D DISCUSSION PAP
THE BOARD OF DIRECTO ER
The Background to Recusal and other
AGSUCSreron, Jane MacLeod Meeting date: 20 March 2019
Context
The Board is meeting to decide whether to apply for the Judge in the GLO to recuse
himself from the proceedings on the grounds of apparent bias. The Board requested
a broader paper setting out the litigation strategy and the wider implications for the
business, putting this decision into context.
Through this paper, we have considered impact through three lenses: the current
claimants; the risk of future claimants; and the risk to ongoing operations.
We have also considered three core outcomes: the original legal strategy, working
through the four trials with a Judge that we would consider neutral; the original legal
strategy played through with this Judge following the recent judgment; and
settlement.
Questions
The questions we are seeking to answer are:
. We went into this litigation believing we were fundamentally “right”. Given the
Judge's criticisms, do we still feel that is the case?
What is the range of outcomes we now see?
Do we believe an appeal is appropriate?
Do we believe recusal will make things better or worse?
What are the implications for the CCRC?
What decisions are coming up?
Conclusions
1. We believe that our systems and processes work today. It is of course harder to
demonstrate that they worked 10 or 15 years ago. We do believe that the way we
work doesn’t always get the balance right between us and Postmasters and there
are. times when the tone can be too self-serving. Strategies to address that are
underway and will be accelerated.
eceee
2. Where the claimants formed part of a previous mediation process, our people
have worked through cases and believe that where we reached a settlement,
justice was served. However, some 450 claimants are new and we have not had
the opportunity to re-assess them.
3. In our original strategy, working through the 4 trials was expected to cost us £25-
30m! in legal fees with the possibility of an earlier settlement reducing the trial
costs. This strategy would have resolved the ‘big’ questions such that there
would have been little incentive for further trials with new claimants, and our own
plans to improve operational processes would have been sufficient to correct the
legacy issues.
4. With the result of the first trial and the current Judge, high level estimates are
very much worse. The cost of the current litigation (including potential awards of
+ On top of that, significant costs (c£12m) were incurred prior to the commencement of the
litigation taking the total to over £40m.
POL00330037
POL00330037
damages and costs against us) could be over £100m, and we could see a
significant contractual challenge from other postmasters enabled by the new
contractual interpretation, adverse findings on Horizon, and a rejection of any
limitation arguments: potentially a further c£E70m and we will incur operational
costs of c.£5m a year. Total costs could get towards £200m and in the event
of a very adverse Horizon verdict we may struggle to maintain 11,500 branches.
Overall, we might need Government support in this outcome.
5. If we were able to reach a settlement, the cost of the current case might be less
(£50-70m) but future claimants and operational changes would be unaffected.
6. Given the strength of our legal advice, we are therefore likely to recommend an
appeal on contractual interpretation to limit the cost both of this case and any
future case.
7. Recusal is the subject of this meeting and separate papers have previously been
shared. A separate call has also been set up for those Board members who are
available to discuss it with Lord Grabiner QC. Clearly, a successful recusal would
give us a chance to fight again and reduce the risk of an adverse Horizon
judgment.
8. The CCRC is not expected to announce findings until after the GLO is complete,
however adverse findings on the contract and Horizon may well have a
detrimental impact on their assessment as to whether there the cases under
review had been properly prosecuted.
9. Recusal has to be decided and, if approved, implemented this week. We now
have until 16 May to lodge an appeal against the judgment. Operationally we
have a number of workstreams underway with some urgent priorities and some
substantial decisions to come.
Input Sought
The Board is asked to note the report, commenting and requesting further information
as appropriate, and make a recommendation as to the application for recusal.
POL00330037
POL00330037
The Report
Are we right?
10. Much of our effort across the litigation has been to demonstrate that our business
“works”: the systems work; our processes work; we apply them in line with our
contracts and so on.
11. We have never suggested that we don’t make mistakes or that there has never
been an individual miscarriage of justice. However it is clear that. historically we
have taken a very black and white interpretation of the contract with an intense
focus on protecting our cash, and with an implicit, and possibly institutionalised
assumption that any losses must be due to ‘operator errors’, and which we
therefore expected the postmaster to make good. This also meant that some
postmasters were prosecuted and the CCRC is reviewing approximately 30 cases
to determine whether those convictions were sound.
12. The Complaints Investigation and Mediation Scheme (“Scheme”) set up in 2013
and which came to an end in 2015 reviewed all 141 cases in the Scheme and
offered mediation. The outcome of these reviews - including by Second Sight, did
not identify any systemic problems with Horizon, but demonstrated that in many
cases we did not provide appropriate training or support. One of Second Sight’s
key criticisms was that we did not sufficiently investigate losses in branch. The
mediation failed overall - participants in the scheme withdrew support in part due
to the fact that many of them had unrealistic expectations of the financial
outcomes.
13. In recent years, we have chosen not to prosecute cases which relied on evidence
derived from Horizon. However, we have said would like to, as we have seen losses
increase from £2m a year to £5m as a result of the change, in spite of a lot of
work to manage the issue.
14. The legal strategy has therefore reflected a belief that we can reach the right
conclusion in individual cases, that we need to protect the business and that the
systems work.
15. The current claimant population is approximately 560 strong, but they are not an
homogenous group. There are approximately 150 former Postmasters who have
joined the litigation in spite of having agreed a full and final settlement either
under the Scheme, or under Network Transformation where they received exit
payments of up to 26 months’ pay. Clearly, in these cases we believe that justice
has been done.
16. However, there are also some 450 claimants who joined the litigation after the
closure of mediation and we do not yet know whether they deserve some
restitution or not. The position on the individual claimants is therefore uncertain.
17. On the “does the system work” question, we believe that Horizon works today. We
process some 0.5b transactions through it every year and manually correct less
than 100k. Many of those transaction corrections are caused by errors by
Postmasters which are easily resolved and for which they are not liable.
18. We settle with commercial clients based on Horizon transactions and whilst each
client is different, if we were systematically failing to process transactions with
banking customers or bill payment clients accurately, then the mis-match with
19.
20.
21.
22.
POL00330037
POL00330037
4
their own customers would generate significant noise and attention. We do have
differences. We have a team of 7 in Chesterfield whose job it is to correct
banking differences but the number is limited and the vast majority reflect
obvious errors around counting the cash (picked up by the second count in
Supply Chain) or mis- keying. We are not seeing anything that suggests a
system flaw.
In addition, on appointment the Chairman commissioned an independent review
from Deloitte to look into this question. While it was overtaken by the litigation,
we are not aware of any materially adverse findings.
Obviously, this does not mean that our systems and processes worked in the same
way 10 or 15 years ago: that is extremely hard to demonstrate.
In our current strategy, we absolutely believe we need to go much further in how
we manage the relationship with agents, shifting from managing our own issues
first to supporting a partner. This encompasses a lot of different changes some of
which are underway and others will need to be accelerated. The point of stressing
it here is that we do not yet believe we are in the right place culturally and in the
way we work.
In conclusion, we would say that the systems work but that the way we deal with
our Postmasters can be further improved to demonstrate fairness and to ensure
we cannot be accused of onerous behaviour.
What is the range of outcomes?
Original Strategy
23.
24.
25.
26.
Our original strategy was based on demonstrating that:
. the burden of proof in cases of difference largely rested with the Postmaster
because only he or she could know what had happened in branch;
. the systems and processes fundamentally work and worked during this
period;
. any specific bugs or issues did not cause losses of the sort that are being
claimed, mainly because they were identified and Postmasters weren’t held
liable;
. the claimants include people who started a claim after the normal 6 year
limit, or had already settled and therefore the true number of claimants
is much smaller; and
. any legitimate claims could therefore be resolved in accordance with normal
principles.
Assuming this strategy was successful, we could envisage a further £10m of legal
costs in 2019-20, £1-5m of specific payments and perhaps £5-10m if we lost on
notice periods. A total cost of £20-30m plus some possible costs against us. In
the January Board we set out change plans that assumed we would retain £30-
50m to cover the GLO outcome.
If this was the outturn, we might not expect a significant body of additional future
claims: the trial process would have been long and expensive, but would have
addressed the core challenges relating to the contract and Horizon, such that
individual cases could then be considered.
Clearly, there would have been limited impact on our operational processes and
our agent strategy would improve matters in future beyond any legal or
contractual requirement.
POL00330037
POL00330037
The original strategy with the current judge and recent judgment
27. Clearly, the original strategy appears severely damaged for two reasons: we lost
on the contractual interpretations; and the Judge is presenting himself as someone
who believes we are conspiring to hide an injustice, appearing to believe that there
are fundamental flaws in Horizon; that we know this, and are trying to hide it.
28. In this scenario, the £20-35m estimate must look light. We should assume that
the Judge would link his conspiracy perspective and deny legal limitation in the
third trial. This will also enable future claims. Clearly, for all the claimants, the
way we work now will be in breach of the new contractual interpretation.
29. It is very hard to even guesstimate how this would work through financially. We
know that the funders will receive a significant proportion of any claim. In 2016
we believed that this was the first £20m of any award or settlement, however in
light of the legal costs that they have now funded plus a private equity margin, it
seems likely that this amount would now be higher- perhaps in excess of £30m.
Therefore a calculation based on each claimant receiving £50k (500 claimants
@£50k = £25m) would mean that after the lawyers and funders’ costs that the
actual claimants would receive relatively little; so their expectation must be for
an average award of closer to £100k per claimant - which would be c£50-60m in
total, on top of our own legal costs, and a contribution to theirs. On this basis a
total bill of at least £80m is foreseeable before we add on the operational impact.
For financial planning purposes we are working on £100m so that we have a
significant margin for error should our estimate of claims be too low.
30. More worryingly, this outcome (whether achieved via court processes or a
settlement) would open the door to material future litigation: if the verdict stands
then anyone who has been required to repay losses, been suspended, terminated
or who has resigned rather than return cash will be able to argue that the way we
have dealt with them was in breach of our contract, as now interpreted. Whether
this takes the form of litigation or some form of “PPI” equivalent scheme remains
to be determined.
31. We haven't tried to assess this as a potential cost - but say 1,000 agents (we
suspend c. 100 a year) at £50k each is another £50m plus legal costs-So we could
be at risk for a further £70m and this takes us into the territory where we might
need further Government support.
32. In terms of ongoing operations, we are already working through how we behave
in a manner that is consistent with how we want to work with agents and the new
contractual interpretation. It is early days and it includes an opportunity to be a
better business, accelerating work we wanted to do anyway. We could assume
£10m of one-off costs and £5m a year on higher run costs in more people in the
field and higher losses.
33. Our greatest threat however is if the Judge finds a way of saying that Horizon
wasn't, or isn’t, reliable. He shouldn’t do so - both expert witnesses have said that
they agree that the system is “relatively robust”. However, recent experience
suggests that the Judge may reach beyond this. That decision will be essentially a
question of fact and therefore is unlikely to be appealable. This could shape or
undermine confidence amongst our agents, our suppliers, our clients and our
people. Of course, we would seek to manage that but it could seriously undermine
our ability to maintain a network of 11,500 branches and meet customer
expectations.
34. So the total cost over the next two years could easily exceed £150m combined with
POL00330037
POL00330037
6
the failure to meet our core targets. We would struggle to afford that without
further damaging the business or getting substantial, additional support.
Settlement
35. Within our strategy, we had assumed there would be moments when we could and
probably would settle. The first of these is this Summer: following the Horizon
verdict both sides are required to go through a period of mediation.
36. Settlement would bring this trial to a close and while there would be some brand
damage from doing so, it would be relatively time limited. In light of the recent
judgment there must be a risk that a settlement would cost us a similar amount
to the figures set out in paragraph 29 and there will still be a material risk of
claims from other current and former postmasters who are not part of The
Claimant Group; as set out above, there will be many more of these.
37. Settling before the verdict on the first trial would have left contractual
interpretations uncertain, and the challenges around Horizon itself unresolved, as
well as not resolving the issues of anyone outside the claimant group. That was
costly and not desired. However, given the verdict we have, the uncertainty may
have been preferable especially when considering the risk of future claimants.
Do we believe an appeal is appropriate?
38. Our legal advice strongly supports an appeal on the contractual interpretation: our
advisors, including a former President of the Supreme Court, believe we have at
least a reasonable chance of success, overturning a bad legal judgment. Clearly,
all pre-court advice is uncertain. The cost of an appeal would be c. £im. Appealing
on the points of law is consistent with our public statement and the letter from our
Minister.
39. A better outcome on appeal would be materially desirable. Firstly, it should reduce
the ultimate cost of reaching resolution with the current claimants, including
reducing the likely cost of settlement. Secondly, it makes both a new case and
further. material losses from new claimants less likely. Thirdly, it gives us more
flexibility to make the operational changes we think right rather than being guided
by this interpretation.
40. On that basis, we recommend an appeal on the contractual variations.
Do we believe recusal will make things better or worse?
41. The Board has seen a separate paper on recusal with advice from Lord
Neuberger and an opportunity is available to discuss it further with Lord Grabiner
ahead of the Board call. We will not rehearse the arguments again here. A
successful recusal would mitigate against another unfair verdict on Horizon and
any subsequent trial would be expected - but not guaranteed - to produce a
better outcome for the litigation and for any subsequent settlement.
42. Further, as flagged by Lord Neuberger on the call last night, making the
application for recusal - even if not successful, will assist an appeal on the grounds
procedural unfairness.
43. In re-trials and in further trials we should review our tone and approach to ensure
we don’t create the same reactions again.
POL00330037
POL00330037
7
44. Clearly, it is uncertain whether an unsuccessful recusal would antagonise the Judge
further or make him more cautious.
What are the implications for the CCRC?
45. The CCRC’s views are not known and we do not expect them to reach a conclusion
on a potential miscarriage of justice in the 30 cases under review until after the
Horizon trial is concluded and the judgment handed down. The effect of the
judgment as its stands probably means that the branch trading statement would
not now be held to have been proper evidence of loss, and we are more likely to
have been in breach of contract such that there would be a greater risk of the
decision to prosecute not having been justified. If the Horizon trial also has an
adverse outcome, then it is highly likely that the CRC will have to review their
conclusions. While we are still unsighted on these, the risk of an adverse
outcome from the CCRC must now be significantly higher.
What decisions are coming up?
46. We have to reach a decision at this meeting on whether to ask the Judge to
recuse himself, and to appeal if he disagrees, and to make the application this
week.
47. Normally we would have some 21 days from last Friday to request leave for appeal,
detailing the basis of such an appeal. At this stage the judge has indicated that
consequential orders will be made on 16 May, and that would be the point at which
we would seek leave to appeal.
48. We have to respond as.a business. We have a number of workstreams underway
as follows:
Legal (Jane)
Operations (Julie Thomas)
Agents (Amanda Jones)
Communications (Mark Davies)
Stakeholders (Al and others)
IT/Horizon verdict (Rob Houghton)
Brand (Emma Springham)
° Financials (Micheal Passmore)
49. Many of the workstreams have immediate priorities and then longer term
questions and we will need to be alert to changing circumstances and trends. The
most urgent questions are around how we operate now in terms of signing up new
postmasters, how we should deal with differences and how soon we could
introduce reasonable contract variations that protect us and clarify things for
Postmasters in future.
eceeceee
50. Beyond immediate working patterns, we need to determine how far we want to go
and how quickly as we accelerate our strategy of making it easier for Postmasters.
This is underway and to reassure the Board on this, our draft budget already
assumes a net investment in the Retail field teams to improve face to face contact,
the development of the branch hub, plans to reduce errors and repeats, the
operationalising of the work we have done to create a data set for each branch
to highlight issues that can then lead to training or other support and efforts to
change the tone from audit and fraud to support and partnership.
51.
52.
53.
54.
55.
POL00330037
POL00330037
However, current plans are not enough and in particular the process to manage
differences will need to be very carefully designed, implemented and overseen and
we will have to decide how much independence we want to introduce into the
process. Transparency will be critical. This is also relevant for our work with the
NFSP which forms part of the agents’ workstream.
We then have choices to make about whether now is a good time to make some
deliberate shifts in “fairness” between us and Postmasters. Should we resolve
the stuck Postmasters issue, either by getting permission to miss 11,500 or by
putting our own staff in, albeit under new contracts? Have we got the profit share
right on Banking Framework 2 income? And so on.
We are planning to revert to the Board with recommendations before end April.
By that time, we must have a clear plan for managing an adverse Horizon trial
verdict and a plan for a mediation process.
We are also updating our longer term view of our financial situation and the
availability of funding. We have a number of workstreams already underway to
optimise funding: offering to buy our way out of Santander security; requesting
Bank of England help to improve funding available from the NCS arrangement;
potentially requesting slower payment terms from banks etc. We will cover these
at the Board meetings in May and July but none of them are straightforwardly in
our gift.