3 March 2023 – Andrew Winn
Hide video Show video
(10.00 am)
Mr Beer: Good morning, sir. Can you see and hear me?
Sir Wyn Williams: Yes, I can, thank you.
Mr Beer: May I call Andrew Winn, please.
Andrew Winn
ANDREW FRANK WINN (affirmed).
Questioned by Mr Beer
Mr Beer: Good morning, Mr Winn. My name is Jason Beer and, as you know, I ask questions on behalf of the Inquiry. Can you give us your full name, please?
Andrew Winn: Andrew Frank Winn.
Mr Beer: Thank you. Many thanks for coming to give evidence to the Inquiry today and for the provision of two witness statements. Can I deal with your first witness statement first, please. It should be in the hard copy bundle in front of you at tab A1. It’s 24 pages in length and is dated 26 December 2022. Do you have that?
Andrew Winn: Yes.
Mr Beer: For the transcript the URN is WITN01090100. Can you go to the 24th page of it, please.
Andrew Winn: Yes.
Mr Beer: Is that your signature?
Andrew Winn: Yes.
Mr Beer: Are the contents of that witness statement true to the best of your knowledge and belief?
Andrew Winn: Yes.
Mr Beer: Thank you. A copy of that witness statement will be uploaded to the Inquiry’s website and therefore I’m only going to ask you questions about selected parts of it. Can we turn to the second witness statement, please. That’s four pages in length and was signed by you today. Can you look at the fourth page, please, and tell me whether that’s your signature?
Andrew Winn: Just give me a moment. I’m struggling with my folder here. Sorry, where is it?
Mr Beer: I think it’s immediately after the first witness statement.
Andrew Winn: Yes.
Mr Beer: Is that your signature on the fourth page?
Andrew Winn: Yes.
Mr Beer: The URN for that is WITN01090200. Are the contents of that witness statement true to the best of your knowledge and belief?
Andrew Winn: Yes.
Mr Beer: Thank you very much.
I’m going to ask you, Mr Winn, questions about the matters that arise in what we in the Inquiry are calling Phase 3 of the Inquiry.
I’m not going to ask you any detailed questions today about cases involving the prosecution of individual subpostmasters, as that will be a matter addressed in Phase 4 of the Inquiry, or in relation to investigations that took place after the scandal broke. That may be addressed in Phase 5 of the Inquiry and you may be recalled within it. There may be some questions that do touch on those matters but only where it’s necessary to understand and explore your evidence on Phase 3 issues. Do you understand?
Andrew Winn: Yes.
Mr Beer: Thank you. Can I start with your background and experience. I think you worked for the Royal Mail Group and then the Post Office for 20 years between 1996 and 2016; is that right?
Andrew Winn: That’s correct.
Mr Beer: Have you any professional qualifications that are relevant to the issues that we’re considering today?
Andrew Winn: No.
Mr Beer: You say in your witness statement that you started working for the Post Office or Post Office Limited in around 2001. Before that time, what was your role within Royal Mail Group?
Andrew Winn: I started off as a postman, part-time postman. Then I worked in a management reporting role. Then I worked in an audit role and, from there, I moved into Post Office Limited.
Mr Beer: What was the audit role?
Andrew Winn: Initially it was with Parcelforce, which largely involved checking out individual branches, depos, checking the records, et cetera, and then we moved into a group audit. It consolidated into a group audit and that would involve going into all the different areas of the business, usually in groups and carrying out audits of what was going on there.
Mr Beer: Before 2001, before you took up your role in the Post Office in 2001, did your role require you to have any involvement with the Horizon System?
Andrew Winn: No, I believe not.
Mr Beer: Before you took up your role in the Post Office in 2001, what did you know, if anything, about the Horizon System?
Andrew Winn: Nothing.
Mr Beer: If you can just maybe move forward slightly so that the microphones pick up your voice. You will see that the Chairman is listening remotely and the proceedings are broadcast and so it’s really important that the microphones pick up what you say.
Andrew Winn: Okay, sorry.
Mr Beer: So the answer was “Nothing” I think.
Andrew Winn: Correct.
Mr Beer: Thank you. So does it follow that before you took up your role in 2001, you didn’t know anything about whether there existed any bugs, errors or defects in the Horizon System?
Andrew Winn: Correct.
Mr Beer: Now, you joined Post Office Limited, you say in your statement, in 2001 in the Network Improvement Team.
Andrew Winn: Yes.
Mr Beer: What was your job title in the Network Improvement Team?
Andrew Winn: I can’t remember.
Mr Beer: What was your function in the Network Improvement Team?
Andrew Winn: A little bit, as the name suggests, trying to find ways of improving the performance of the team. I was particularly involved in an activity based costing exercise, trying to pull together activity based costing to help make decisions and, in truth, it ended up being too high a level to have much effect.
I also got involved with the – I can’t remember what it were called but, basically, where you get mystery shoppers going into branches. So I was involved in creating the questions and writing – and that changed every month, so that would be part of my job.
Mr Beer: What level within the Network Improvement Team were you?
Andrew Winn: In terms of tiers, you mean?
Mr Beer: Well, were you at the lowest rung in the ladder? Would you a supervisor, if there were such things? Were a manager? Were you the head of the unit?
Andrew Winn: I was manager without anybody to manage. So I was what would be a CM2 grade, I believe. So there was an admin officer, who would be the admin grade, but then I think there was probably one person same grade as me and then three or one senior manager grades with a team leader on top of that.
Mr Beer: To whom did you report?
Andrew Winn: One of the senior managers.
Mr Beer: You didn’t manage a team?
Andrew Winn: No.
Mr Beer: How many people were in the Network Improvement Team?
Andrew Winn: Seven or eight.
Mr Beer: Where was it based?
Andrew Winn: In Chesterfield.
Mr Beer: Did your role in the Network Improvement Team require you to have knowledge of and understanding of the operation of the Horizon System?
Andrew Winn: No.
Mr Beer: In the course of that role, did you acquire any knowledge about whether there existed any bugs, errors or defects in the Horizon System?
Andrew Winn: No.
Mr Beer: You say in your witness statement that you moved to the problem management team in 2005. I think you’ve since seen an email which suggests that it was earlier than that because it refers to you being in that team in at least 2003?
Andrew Winn: Yes.
Mr Beer: What was your job title in the problem management team?
Andrew Winn: Problem management team leader.
Mr Beer: What was the function of the problem management team?
Andrew Winn: As I understand, it was a response to basically the Post Office moving onto an IT-type platform, rather than a manual-type platform, and it was part of the kind of plan that was laid out that, effectively, the concept was that anybody within the Post Office who got a problem reported it to the problem management team.
In reality, it should have been just an IT function but the way it was set up was that everyone who had a problem reported it in there. The IT kind of structure at the time was that the problem management team managed the problem in terms of making sure the relevant people were involved in correcting the problem, rather than actually resolving the problem themselves, which I found a difficult concept to deal with.
Mr Beer: So it was like a signposting service, was it?
Andrew Winn: Yes, that’s correct.
Mr Beer: Did you manage a team?
Andrew Winn: I did, yes.
Mr Beer: How many people were in that team?
Andrew Winn: There was probably, I think, about 12 other team members.
Mr Beer: To whom did you report?
Andrew Winn: A senior manager.
Mr Beer: Who was that?
Andrew Winn: Initially, it was Marie Cochate but she left. I can’t remember who took over from there.
Mr Beer: Where were you based?
Andrew Winn: In Dearne House.
Mr Beer: Which is?
Andrew Winn: Near Barnsley.
Mr Beer: Did that role and the role of the team that you managed require knowledge of and understanding of the operation of the Horizon System?
Andrew Winn: Yes, but I didn’t have knowledge of the Horizon System. So I would have said I was a bad placement into that role.
Mr Beer: How did you acquire, if you did, any knowledge and understanding of the operation of the Horizon System?
Andrew Winn: There were couple of members of the team who had some knowledge, basically from working in branches, but it was really a case of trying to figure it out as I went along.
Mr Beer: So you weren’t given any training at that stage?
Andrew Winn: No. I think at the time there was very few people within the Post Office who’d got much IT knowledge, to be frank.
Mr Beer: Horizon had by this time, 2003, been up and running for three years. In that time, had you acquired any knowledge or understanding of the operation of the system?
Andrew Winn: Some but very limited, I would say.
Mr Beer: You said that you were a bad fit or words to that effect.
Andrew Winn: I would have said so, yes.
Mr Beer: Why were you a bad fit?
Andrew Winn: Because I’m not a technically – I find technology quite difficult even now. I always felt as though I was playing catch up with technology, kind of still do now. So ideally somebody in that role would have had a good understanding and been able to understand problems easily, whereas I was – forever seemed to be trying to understand what it was that was meant.
Mr Beer: You said that an important part of the intended function of the problem management team was IT, information technology, although the boundaries were stretched on that.
Andrew Winn: Yes.
Mr Beer: An important part of that would have been Horizon itself?
Andrew Winn: Yes.
Mr Beer: Do you know why you were picked them if you were a bad fit, bad with technology, and had no training to be a team leader in the problem management team?
Andrew Winn: You’d have to ask the people who interviewed me but I would guess that my competition were in a similar situation.
Mr Beer: I’m sorry –
Andrew Winn: The competition for the role. There was – I don’t know how many people were interviewed for the role but I got the role on –
Mr Beer: You were all in the same boat?
Andrew Winn: I would say – I don’t particularly know the people who were competing with me but that would be my understanding, yes.
Mr Beer: You say in your witness statement that after about 18 months in the problem management team you moved to a data management team, the name of which you can’t remember.
Andrew Winn: Correct.
Mr Beer: In that 18-month period in the problem management team, did you acquire any knowledge about the number and nature of any bugs, errors and defects in the Horizon System?
Andrew Winn: To be honest, I kind of – my memory of the time is largely more about dealing with outages or breaks between the different data houses that information flew through. So it kind of feels more around checking that the different data centres were talking to each other and who was owning the problem and resolving it. There obviously must have been issues on Horizon that came through but that’s my principal memory of the role.
Mr Beer: Looking at it globally – I appreciate it may be difficult to isolate a period of time given what you then went on to do – what would your view have been of Horizon at that time, in this 18-month period before you went on to the data management team?
Andrew Winn: I don’t think I got a great view of the actual Horizon System in branches. I think I was more looking at kind of Fujitsu into other data warehouses. So I’d not got a strong view in any direct direction.
Mr Beer: What were you looking at in relation to the flow of data into data warehouses involving Fujitsu?
Andrew Winn: Whether it’s flowing as it was required.
Mr Beer: And was it?
Andrew Winn: Most of the time, yes, and occasionally, when a problem arose, it wasn’t.
Mr Beer: Did this experience tell you anything that you can now remember about the way that Horizon was operating?
Andrew Winn: With no experience of other systems of anything like similar or really any other systems, I couldn’t compare it to say whether it was good, bad or indifferent. It seemed to work most of the time fine, yes.
Mr Beer: Would that be your abiding memory, that Horizon at this time seemed to work fine most of the time?
Andrew Winn: Yes.
Mr Beer: So when you went into the role in the data management team you wouldn’t have gone into the team thinking this is a problematic system, Horizon?
Andrew Winn: No.
Mr Beer: So you moved into the data management team. What was your job title in the data management team?
Andrew Winn: I can’t remember. It was the database, whatever it was called, manager.
Mr Beer: Where was that based?
Andrew Winn: Again that was in Dearne House.
Mr Beer: What was the function of that data management team?
Andrew Winn: There wasn’t really a great function. It basically compiled data and I can’t remember what type of data it was. It was only used by one team within POL, as far as I was aware. When I got into the role, there was a reorganisation and the problem management team and the risk team, I think, were combined into one team and the risk manager took over management of the team. So I was left in a situation where was I going to go, there was a space there so I was kind of fitted in there.
Mr Beer: What data did it manage and for what purpose did it manage it?
Andrew Winn: It seems – I really – it is a blur, that role. It kind of felt like different product descriptions perhaps, promotional information.
Mr Beer: Did your undertaking of that role require knowledge of and understanding of the operation of the Horizon System?
Andrew Winn: No.
Mr Beer: In the course of that role, therefore, did you acquire any more knowledge of any errors, bugs and defects in the operation of the Horizon System?
Andrew Winn: No.
Mr Beer: So Horizon wasn’t really on your horizon in that role.
Andrew Winn: Not at that point, no.
Mr Beer: In 2007, you moved to the Finance Reporting Team?
Andrew Winn: Yes.
Mr Beer: What was your title in the Finance Reporting Team?
Andrew Winn: I can’t remember that.
Mr Beer: What was the function of the Finance Reporting Team?
Andrew Winn: It was basically producing monthly reports for different teams within the Post Office, financial management reports.
Mr Beer: Where were you based?
Andrew Winn: In Chesterfield.
Mr Beer: To whom did you report?
Andrew Winn: I can’t the lady’s name.
Mr Beer: Did you manage a team?
Andrew Winn: No.
Mr Beer: How many people were in the function of finance reporting or in the Finance Reporting Team?
Andrew Winn: I would estimate about twelve.
Mr Beer: In that role, did you require knowledge and understanding of the operation of the Horizon System?
Andrew Winn: No.
Mr Beer: Does it follow that you didn’t acquire any more knowledge in that role of any errors, bugs and defects in the Horizon System?
Andrew Winn: Yes. No, I would say not because I would have been working there when Horizon Online went live. So I would be in the same building.
Mr Beer: Just think back. In your statement, you say that you moved to the Finance Reporting Team in 2007 and I think that you’re going to tell us in a moment that in 2008 you moved to the Product & Branch Accounting team.
Andrew Winn: To my best memory.
Mr Beer: Horizon Online didn’t go live, really, until 2010.
Andrew Winn: Okay. So I was certainly in the Finance Reporting Team when Horizon Online went live, so I obviously got those dates wrong.
Mr Beer: So the date that you have given us in your statement of moving to the Product & Branch Accounting team in 2009 – sorry, 2008, might that be wrong?
Andrew Winn: Yes, I think so.
Mr Beer: We’ve certainly got documents from you in 2009, ie before Horizon Online went live, with you in the Product & Branch Accounting team. So you had certainly moved by then.
Andrew Winn: Okay. There was certainly something – my memory is that something had happened in terms of a major project around Horizon during the period that I was not in Product & Branch Accounting. So all I can say was I was aware that a major Horizon-related project had gone live and there was lots of issues flying around there.
But I’d been in the same building as a lot of people and knowing people who were involved. So, in terms of my actual role, I wasn’t involved at that point but I was aware within the business of things happening.
Mr Beer: On that awareness, were you aware of anything particularly problematic or difficult?
Andrew Winn: Yes, I was aware that the accounting in Product & Branch Accounting was causing a lot of problems in a lot of areas – not so much – I wasn’t particularly aware of issues impacting branches then because Product & Branch Accounting was based in the same building as I was in. I was aware that that there was a lot of stress coming there that needed resolving.
Mr Beer: What was the stress that was coming there?
Andrew Winn: I think feeds from branch, right, whenever things were falling into the wrong accounts and the accounts weren’t functioning as planned.
Mr Beer: How widely known was this? Even though it wasn’t your team, it was obviously being talked about?
Andrew Winn: Yes. So I would guess, if I was relatively new to the – well, no I’d worked in the building before but I wasn’t particularly well-known person within Post Office Limited, so I would guess other people would know more than me but that’s a guess.
Mr Beer: Before you moved to the Product & Branch Accounting team – we’ll try and establish the date with greater specificity in a moment – were you aware of any bugs, errors and defects in the Horizon System that affected the integrity of the data that it produced?
Andrew Winn: No. I think I was more aware that the mapping in preparation hadn’t, in terms of how it fed into P&BA accounts was the issue. I wasn’t particularly aware of what was happening in branches.
Mr Beer: What do you mean the mapping of how it went into P&BA accounts?
Andrew Winn: Part of the project planning would be to prepare data flows. So you sell a stamp in a branch and cash is paid for it and how that flows into the accounts within Product & Branch Accounting, and things were running up in unexpected areas, et cetera.
Mr Beer: So it was a mismatch between what was, in fact, going on in the branches and what the data showed at Chesterfield; is that a fair way of describing it?
Andrew Winn: Yes, I think so.
Mr Beer: Were you just picking this up before you moved to P&BA, in the noise, the conversations that you were hearing?
Andrew Winn: Yes.
Mr Beer: At what level of seriousness was this being expressed?
Andrew Winn: Oh, it was serious.
Mr Beer: Can you remember before you moved into P&BA whether this was being attributed to the way that the Horizon System was operating?
Andrew Winn: No, I can’t answer that one.
Mr Beer: So on a date you moved into Product & Branch Accounting, which as we’ve discussed already was called P&BA; what was your job title in P&BA?
Andrew Winn: Initially it was an analyst.
Mr Beer: You say in your statement that you were initially an analyst in the debt recovery team.
Andrew Winn: Debt recovery, yes.
Mr Beer: Yes, I think that’s what I said.
Andrew Winn: I heard “guess”, sorry.
Mr Beer: Debt recovery team. Your post, you say, later became described as relationship manager?
Andrew Winn: That’s correct.
Mr Beer: Can you remember when that was?
Andrew Winn: No.
Mr Beer: What was the debt recovery team?
Andrew Winn: It was the team – so branches had – one of the options with their debt was to put it into an account called “Settle centrally”. So if you had a £100, let’s say, debt at your balance period, you had a number of choices: make good cash, put the cash in to make up the deficit; make good cheque, equally put the cheque in; or settle centrally, in which case the debt would flow through to an account in Chesterfield and, hopefully, it would be a case where the debt would get balanced off against something else but, if the debt sat there, the debt recovery team would then look to recover from the subpostmaster.
Mr Beer: So the debt recovery team, would this be fair, its function was to seek to recover debts that may be owed by subpostmasters?
Andrew Winn: Yes.
Mr Beer: What was the function more broadly of the P&BA team?
Andrew Winn: To account for mainly branch activity, to consolidate it and report it.
Mr Beer: What do you mean to account for branch activity?
Andrew Winn: I’m struggling to explain that. Effectively, data would flow in from branches, it would flow in from clients, it would flow in from other parts of the business from, for example, cash centres, stock centres, all these data flows would come in and they should get matched off one against the other and accounts cleared down to zero.
Mr Beer: Thank you. If we look at your witness statement, please, WITN01090100 at page 2 – it will come up on the screen for you on the right-hand side. Look at paragraph 3.
Andrew Winn: Yes.
Mr Beer: You are talking about the role that we’re now discussing. You say:
“The role initially focused on process improvements and looking at accounting queries from branches but over time concentrated almost entirely on accounting problems in branches and as a single P&BA … point for both branches and other relevant teams…”
Yes?
Andrew Winn: Yes.
Mr Beer: Then you go on to list the other relevant teams. Starting with branches first, how would branches contact the P&BA team that were performing this role?
Andrew Winn: My role or P&BA?
Mr Beer: Your role.
Andrew Winn: Ideally in writing, explaining what the problem was.
Mr Beer: More broadly, the P&BA team, how would branches contact the P&BA team?
Andrew Winn: The P&BA team, broadly if – a lot of the teams were based in correcting – managing accounts of products and they would be looking to effectively get an input from the branch, and input from the client, match the two off, clears down to nothing.
Branches might feel as though they need a transaction correction, for example. They would contact the NBSC, the helpline, who would then direct them into the relevant team within P&BA.
Mr Beer: How would that contact be made once they had spoken to NBSC?
Andrew Winn: Normally by phone.
Mr Beer: Looking at the other relevant teams that would use P&BA as this single point of input you say “primarily Network”, just inside the brackets there.
Andrew Winn: Yes.
Mr Beer: Who or what do you mean by the word “Network”?
Andrew Winn: There was a Network team within POL at the time and that would involve – so it was primarily field support advisers, people who were in contact with branches. So it may well be that they had been in contact with the branch who then said “I’ve got this problem, what do I do?” They would direct them in towards P&BA.
Mr Beer: If Network were contacting P&BA, how would they do that: by phone or in writing?
Andrew Winn: Oh, by phone normally or email.
Mr Beer: The next relevant team that you mention is “Helpline”, the NBSC.
Andrew Winn: Yes.
Mr Beer: Was that the only helpline that would get in contact with P&BA?
Andrew Winn: As far as I’m aware, yes.
Mr Beer: How would people in the NBSC contact P&BA?
Andrew Winn: By phone – again, possibly by email but more normally by phone.
Mr Beer: Then, lastly, you mention within the brackets there “Product & Security”. Who or what is/was Product & Security”?
Andrew Winn: They would be separate teams. There would be a Product team who kind of, what it says on the can, would manage the products in terms of the relationships with the clients but also how the products are working with it at branch level.
Mr Beer: So product could be, what, Lottery or –
Andrew Winn: Yes. So there would be a Lottery product manager who would talk to Camelot and also be in contact during the network into products. So when there were products – that was a product where there was quite a lot of issues that arose they would be in communication with Camelot and different parts of POL to try and improve, smooth out the process.
Mr Beer: How would Product get in contact with P&BA?
Andrew Winn: Same again: email or phone.
Mr Beer: And Security?
Andrew Winn: Security –
Mr Beer: Who or what are you describing by the word “Security” there?
Andrew Winn: There was a Security team who were the ones who were probably, I believe – ultimately they bring prosecutions against branches but also looking after the integrity of the POL cash and products, the security of it, and they would occasionally – we’d occasionally link up where need be, again email or phone.
Mr Beer: Then, lastly, outside the brackets you say:
“… along with the National Federation of SubPostmasters (NFSP) …”
Andrew Winn: Yes.
Mr Beer: How would the NFSP contact P&BA?
Andrew Winn: They would normally ring – well, the contact would generally be to me, usually, by phone.
Mr Beer: Why would the NFSP usually come to you?
Andrew Winn: Because I’d built – that was part of the defined role when the job set up, to get a link into the network, actually, rather than the POL network, the subpostmaster network, to give them a direct line into P&BA.
Mr Beer: Was there one person that you particularly engaged with or was it a range of people?
Andrew Winn: There was one person who I dealt with, a paid officer, within the NFSP, who was –
Mr Beer: Who was that?
Andrew Winn: Sorry, I can’t remember. Oh, Stoddart … someone – something like Marie Stoddart.
Mr Beer: Marie Stoddart?
Andrew Winn: Yes, I think I’ve probably got the first name wrong –
Mr Beer: Okay.
Andrew Winn: – which apologies to her because I knew her very well, which is sad.
But also I used to go to a monthly meeting where they get together and I’d go and join in with them, make a presentation and also sit and listen to some of the other issues going round. So we do have quite a close relationship, I think.
Mr Beer: Thank you. That statement can come down from the screen now, thank you.
To whom did you report?
Andrew Winn: A senior manager named Alison Bolsover.
Mr Beer: Say that more slowly?
Andrew Winn: A senior manager called Alison Bolsover.
Mr Beer: Was she one of four senior managers?
Andrew Winn: Correct.
Mr Beer: Who were the other three senior managers?
Andrew Winn: They changed during my time there.
Mr Beer: Can you give us names that you can remember?
Andrew Winn: Yes. (Pause)
You know, I can’t remember a single name, sorry.
Mr Beer: You say in your witness statement that the four senior managers themselves reported in to the head of P&BA. Who was the head of P&BA?
Andrew Winn: Rod Ismay for most of the time that I was there.
Mr Beer: What was his title?
Andrew Winn: I believe it was head of P&BA.
Mr Beer: He reported to the Finance Director; is that right?
Andrew Winn: Correct.
Mr Beer: Who was that?
Andrew Winn: Again, that changed during the time I was there and I can’t remember either of the names, I’m afraid.
Mr Beer: Did you manage a team?
Andrew Winn: Yes, one admin officer.
Mr Beer: What was the role of the admin officer?
Andrew Winn: It was largely to document the cases that came in and, yes, that was the main part of the role.
Mr Beer: How would they document the cases that came in?
Andrew Winn: They would generally come by letter. They’d be opening a letter, they’d be trying to work out what the case related to and –
Mr Beer: Just stopping you there, you said earlier they would generally come in by phone or email?
Andrew Winn: Not to me, from the branches.
Mr Beer: Okay. I was talking about the whole range of reporting from Product and Security, from the NFSP, from the NBSC, from Network. We went through those and you said that they generally came in through phone or email contact.
Andrew Winn: Yes, yes.
Mr Beer: How would the admin officer document those?
Andrew Winn: They wouldn’t, unless I asked them to. It would be usually a case of they’d raise an issue, perhaps which might relate to a branch which may raise a case which would then get documented.
But if the NFSP rang me and said, “What’s going off with this product” or whatever, that wouldn’t be documented by the admin assistant.
Mr Beer: You describe in your statement that this role concentrated almost entirely on problems in branches and was a single point of contact.
Andrew Winn: Yes.
Mr Beer: Did your role therefore require knowledge of and understanding of the operation of the Horizon System?
Andrew Winn: Yes.
Mr Beer: Had you got any knowledge of the operation of the Horizon System by the time you joined?
Andrew Winn: It’s possible that I did because managers were typically given training to cover strike action and Christmas support and I think that would have happened before, so I would have thought I’d have a basic understanding of how the Horizon System worked, yes.
Mr Beer: In that sentence you used the word “basic understanding”. I take it you use that deliberately because that would be your level of understanding as an occasional end user?
Andrew Winn: That’s correct, yes.
Mr Beer: Wasn’t the role that you were performing a role that required much more detailed knowledge of the operation of Horizon than that?
Andrew Winn: Yes.
Mr Beer: Again, do you know why you were selected for the job if you didn’t have a detailed understanding of the way in which Horizon worked?
Andrew Winn: I would probably again point to the lack of competition in terms of advance knowledge of the Horizon System but, again, that’s probably one for the people that were interviewing, rather than – in fact, I don’t think there was an interview for that. I think it was placed in there. So that was more probably a case of Alison Bolsover, who whoever within P&BA, talking to my manager in the reporting team at the time and seeing how the fit went, after – I said I’d expressed an interest in the role when I saw the reorganisation of P&BA.
Mr Beer: Now, we’ve got a document dating from 2009 suggesting that you were in role in P&BA and that you undertook a review. Can we look at that, please. It’s POL00039029.
Do you see this appears to be a PowerPoint presentation –
Andrew Winn: Yes.
Mr Beer: – and I think you have seen it before. There’s some notes, when we get to the successive pages, underneath each slide. Maybe if we just look at an example of those, if we go to page 5 – and scroll down, please. Thank you.
So the slide that’s displayed is at the top and then some notes at the bottom; is that right?
Andrew Winn: Yes.
Mr Beer: So the people who are getting the presentation don’t see the notes. They are like a speaking note for yourself?
Andrew Winn: That’s right.
Mr Beer: Yes?
Andrew Winn: Yes.
Mr Beer: So if we just go back to page 1, please, we’ll see that the title of the presentation is “Transaction Corrections, Debt Reporting and Debt Recovery Review”, with your name and “January 2009” underneath it. Did you write this document?
Andrew Winn: Yes.
Mr Beer: So this relates, is this right, to the period when you were in P&BA?
Andrew Winn: Yes.
Mr Beer: So it’s a review conducted before the introduction of Horizon Online?
Andrew Winn: If –
Mr Beer: If my date of 2010 is correct.
Andrew Winn: Yes, yes.
Mr Beer: Can we look at page 2 of the document, please. We’ll see the “Objectives”:
“Review current ways of working and supporting operating processes in the transaction correction, debt reporting and debt recovery areas.
“Review the impact of the current ways of working and current operating processes in the transaction correction, debt reporting and debt recovery area on the POL Network.”
Then thirdly:
“Analyse and then recommend cost effective improvements to the way the correct end-to-end transaction correction, debt reporting and debt recovery operating processes work.”
So it’s focused, is this right, on looking at the way that the systems were working at that time?
Andrew Winn: Yes, it sounds very much as though that’s – I’ve been put in a role, I’ve had conversations with my boss and probably other people within P&BA, and pulled together a what do I think this role involves.
Mr Beer: Not just what does it involve, but look at ways in which it could be changed, in order to make cost effective improvements –
Andrew Winn: Yes.
Mr Beer: – to save money?
Andrew Winn: I think everything you do, you’re looking at working as efficiently as possible. I don’t think it was particularly a brief to come in and find ways of cutting down our costs.
Mr Beer: Now, the Inquiry’s heard some evidence from a previous witness, Ms Susan Harding; do you remember her?
Andrew Winn: Yes.
Mr Beer: What do you remember her as, a role that she performed?
Andrew Winn: She was actually my – for a while in the network intervention team, we talked about earlier, the previous team that I was in.
Mr Beer: She told the Chairman that subpostmasters were never forced to settle centrally. Can we just look at page 3 of this document, please. You in the first bullet point ask “What is ‘Settle Centrally’?” and then say:
“Branch Trading forces [transaction correction] acceptance.”
Andrew Winn: Yes.
Mr Beer: “Inadequate [transaction correction] evidence/instructions.
“Unclear process.
“Non-conformance not addressed.
“New subpostmasters.
“Aged/High Value/High Volume [transaction corrections].”
Overall, without coming to the detail of each bullet point first, what are you speaking about? Can you explain what is being said in this slide of yours?
Andrew Winn: I think I’m saying that I got into this role, I’ve had a look at what areas we can make improvements on and what is going to be priority ones, where it is a lack of understanding, and these are the kind of things I’m going to be trying to focus on and probably some things are going to become more important than others as we go along. But that’s probably the list of to-dos that were being agreed.
Mr Beer: What did you mean by the second bullet point “Branch Trading forces [transaction correction] acceptance”?
Andrew Winn: So every month a branch is required to carry out a branch trading process where effectively they pulled everything together, ideally everything balances, they have a nice zero at the bottom and we move on to the next trading period.
The transaction corrections can be issued at any time and there was a significant issue with the branches expressed particularly through the NFSP about – well, sorry, just re-track a little bit. Branches used to, pre-Horizon days, used to balance weekly and it was still recommended to do a balance, not a complete balance but a kind of summary balance, to try and get a view of where they were, but the primary balance was held monthly.
There was a kind of half-hearted – what seemed to be a half-hearted claim that you only issue transaction corrections on a Tuesday so we get it on a Wednesday morning when we’ve got to do the balancing, and I actually did a bit of exercise to disprove that and it turned out we did issue the most transaction corrections on a Tuesday. There may – it wasn’t significantly relevant compared to Wednesday or Thursdays.
But on branch trading, on the monthly branch trading, everything kind of has to be cleaned up, so that would mean any transaction corrections that are outstanding needed to be accepted before you could roll into the next trading period.
Mr Beer: By “accepted”, you mean accepted by the subpostmaster?
Andrew Winn: Yes.
Mr Beer: So what is the bullet point is saying is that the action of branch trading, the monthly reconciliation process, is forcing subpostmasters to accept transaction corrections that the centre is putting to them?
Andrew Winn: Yes.
Mr Beer: Were you saying that’s a good thing or a bad thing?
Andrew Winn: I was saying it was – I don’t think it was a good thing.
Mr Beer: Why wasn’t it a good thing?
Andrew Winn: Because branches would potentially walk in on a Wednesday morning of balance day, turn the Horizon on, first thing they see is a transaction correction come through, they know they have got a balance on the night, they’ve got to understand what the transaction correction is and, if it’s particularly one that’s a debt transaction correction that’s going to potentially cost them money, they need more time – potentially need more time to review it and potentially appeal against it.
Mr Beer: But the system is forcing them to accept it –
Andrew Winn: Yes.
Mr Beer: – in order to continue trading?
Andrew Winn: Yes.
Mr Beer: This was an inbuilt feature of the system, is this right, on a monthly basis, in order for a subpostmaster to continue trading?
Andrew Winn: No. No, that was an operational requirement. But in practice if you didn’t carry out a branch trading rollover Horizon didn’t mind. It was kind of quite happy to go along with that to a point – I can’t remember how far in the future when the whole thing would start falling over but, effectively, if you didn’t do your branch trading, nobody cared. A lot of branches actually did their branch trading on a Thursday for operational reasons, particularly the multiple partners, I think, if I remember correctly.
Mr Beer: But the point that you are making here is that this monthly exercise is forcing subpostmasters to accept transaction corrections without necessarily a proper exploration of the merits of the correction?
Andrew Winn: That’s correct. The system wasn’t forcing you to do that. It was the operational instructions that was saying you need to do that, which for a conscientious subpostmaster they would take it as I’ve got to do it on a Wednesday night.
Mr Beer: Thank you. Can we go over to page 4 of your document, please.
This seems to relate to issues seemingly raised by the National Federation of SubPostmasters.
Andrew Winn: Yes.
Mr Beer: Including under the first bullet point “Dispute Button” and other matters including, two bullet points from the bottom, “Horizon [transaction correction] Receipt” and “Core & Outreach Consolidated Statements”.
Were the National Federation raising with you concerns about the operation of the dispute process on balancing?
Andrew Winn: That would be one of their issues, yes.
Mr Beer: What did they say to you about it?
Andrew Winn: They were reflecting the views of the branches, which I think I’ve probably already covered, that they felt under pressure, that it was unfair to receive a transaction correction which they weren’t aware of, they weren’t expecting and potentially to have to be forced to accept it with the risk that POL may then ultimately say “No, you’ve accepted it, so you pay up”.
Mr Beer: What the Federation were saying, would this be right, was that there ought to be the facility to have a dispute button, to say “I dispute that transaction correction, I shouldn’t be forced to accept it”?
Andrew Winn: They raised it as a possibility, as an idea. Certainly, there would have been branches who were saying “We need a dispute button” but they were throwing it in, let’s look at whether that makes sense to do that.
Mr Beer: Can we see your notes underneath, please, on “Dispute Button”. Can we see your notes, the first one if that can be highlighted under “Dispute Button”. You wrote:
“… facility would be abused. POL believe they have provided evidence to support validity on issue. Dispute simply asks us to do it again. Robust dispute process is answer.”
Is that your reply, essentially, to the suggestion that there should be a dispute facility; namely, no, it would be abused?
Andrew Winn: I don’t think I would put it quite as bluntly as that. I think it was kind of looking at what the benefits and risks might be.
Mr Beer: This doesn’t say that.
Andrew Winn: No, no.
Mr Beer: “Let’s look at what the benefits and risks might be”. It says in your speaking note “Dispute button – facility would be abused”.
Andrew Winn: Yes. Sorry, these aren’t speaking notes. These are notes to make me – to lead me and don’t forget to mention this, this and this. That’s not what I would say.
Mr Beer: Why wouldn’t you put a note “Let’s explore the merits, the advantages and disadvantages of it? Why would you make a note to, remind yourself to say “facility would be abused”?
Andrew Winn: I can’t answer that. That’s the way I …
Mr Beer: Is the truth of the matter that that note is there as a prompt to remind you to say, “No, there won’t be a dispute button because the facility would be abused”?
Andrew Winn: I was going into that discussion having thought through the pros and cons and my view is that a dispute button would not improve the process. So I guess, yes, but it was a case of discussing the pros and cons and explaining why I felt, on balance, it wasn’t a good idea.
Mr Beer: On what evidence did you conclude that the existence of a dispute facility would be abused?
Andrew Winn: I think to say evidence would be pushing it; so I think an assumption.
Mr Beer: Why would you assume –
Andrew Winn: Sorry, there would be examples in the past. For example, lots of transaction corrections are effectively equal and opposite. So you might have made an error in this product set, which should have – so the transaction’s been made here, it should have been made over there, so two transactions equal and opposite value need to be issued. In an ideal world, particularly if they are in the same team, they could be issued at the same time, so they are accepted at the same time, but it might be kind of potentially a month apart from one another.
Essentially, the two transaction corrections have no impact on a branch. If you accept the credit transaction correction and make good, then you can take the cash out of the till. If you dispute the debit transaction correction, then there’s no requirement to return that cash back in there until the dispute has been resolved and, effectively, there is no dispute that we can see, it’s perfectly clear they should accept them both, but how do we get to the point where that second transaction correction is accepted?
Mr Beer: In the answer before last you said you wouldn’t say that this view was based on evidence, you used the word “assumption”.
Why would you reject the idea of a dispute button not on the basis of evidence but on the basis of assumption?
Andrew Winn: The question I guess I’d throw back: where’s the evidence that there would be any benefit in a dispute button.
Mr Beer: Haven’t you explained to us what the benefit was already, that subpostmasters were being forced into accepting a transaction correction without a full and proper exploration of the merits of the correction?
Andrew Winn: And that’s why we introduced a procedure for branches who were left in that situation, where they were able to effectively raise a dispute within POL or to raise the fact that they were accepting a transaction correction which they hadn’t fully explored which, if it kind of ended up further down the line in a debt recovery place, they could use that to support their challenge.
Mr Beer: So you make the subpostmaster accept the validity of something before a proper exploration of the dispute they wish to raise; is that the long and the short of it?
Andrew Winn: Sorry can you repeat that?
Mr Beer: Yes. You force the subpostmaster to accept the validity of the transaction correction before a proper exploration of the dispute that they wish to raise about it.
Andrew Winn: Yes.
Mr Beer: Your note continues:
“POL believe they have provided evidence to support validity on issue.”
Is that a note saying, “We’ve already explored the merits of the transaction correction. We have provided evidence already. Therefore, there’s no need for a dispute button”?
Andrew Winn: Yes, that’s – the theory behind the issue of transaction correction is that you have evidence to support that. So if you don’t have the evidence, you shouldn’t be issuing the transaction correction.
Mr Beer: The note says POL believe they’ve provided evidence. Does that mean provided evidence to the subpostmaster already of the correction?
Andrew Winn: Yes, although that evidence may and would normally just be the narrative attached to the transaction correction.
Mr Beer: The note continues:
“Dispute simply asks us to do it again.”
Is that another reason for rejecting the suggestion of a dispute facility?
Andrew Winn: Yes. As our understanding was, you press a button and the onus moves back to POL to prove the transaction correction. If we’ve already provided the evidence that we have, what are we supposed to do more to persuade the subpostmaster that it is acceptable? That would be – really that would be the point where I’m saying a robust dispute process is the answer, in that if the subpostmaster doesn’t accept the evidence, there will always be – a transaction correction will always have a contact number, they could – they would then contact the team member who’s issued the transaction correction and they can talk it through. If they don’t accept it, then we’ve got a dispute process to follow through.
Mr Beer: On that last note you say:
“Robust dispute process is the answer.”
Andrew Winn: Yes.
Mr Beer: Who was being robust or what was robust?
Andrew Winn: The aspiration for me to carry out – well, to set up the process to start with, so that it’s available to branches and the whole of POL to be able – aware of it and for how to kick off the process and then for me to ensure that the process is done as robustly as I’m able.
Mr Beer: What does “robustly” mean in this context?
Andrew Winn: Fair, accurate, timely.
Mr Beer: The notes continue, if we skip down to “TC receipt”, so three bullet points on, so “TC receipt”, so transaction correction receipt, remembering that there was a request for, I think, a Horizon transaction correction receipt, and it looks like your note-to-self is:
“… no clear benefit bar subpostmasters claiming for losses.”
Can you firstly explain, please, what the request was in relation to a transaction correction receipt.?
Andrew Winn: I’ve a very, very vague memory of this one. I think it was subpostmasters wanted a separate piece of paper to put in their accounts, to give to their accountant at the end of the year, which I think would be to claim losses on their accounts. I can’t remember exactly what they wanted, to be honest.
Mr Beer: This was, it seems, rejected too; is that right?
Andrew Winn: Yes.
Mr Beer: You said:
“… no clear benefit bar subpostmasters claiming for losses.”
Wouldn’t a benefit be subpostmasters exist in that there would be a receipt in respect losses they didn’t accept were genuine or a genuine debt? Wouldn’t that be an important benefit to them?
Andrew Winn: It never occurred to me that.
Mr Beer: Okay. Can we go on to page 5, please. “Key Issues – P&BA”. I think this is addressing the key issues for the P&BA team; is that right?
Andrew Winn: Yes, it looks like it, yes.
Mr Beer: If we just expand a little, so we can see the notes below, please. Thank you.
Looking at the notes below, these suggest that the Post Office, I think, was concerned about its own practices and the inconsistency of them towards transaction corrections, including the operation of back office systems including IMPACT; is that right?
Andrew Winn: Yes, it looks like it.
Mr Beer: We can see in the first note:
“Varying [transaction correction] routines – one of the clear ideals I had on setting out and from others who I spoke to was consistency.”
But then you noted:
“But then most teams have completely different approaches to identifying errors and resolving them – and for good reasons.”
So was there an inconsistency of approach within P&BA to identifying errors and resolving them at Chesterfield?
Andrew Winn: I don’t think it was so much identifying and resolving them – well, identifying them, I don’t think was so much of an issue. I think it was more about – if I can give one example, within I think it was the cheques team at the time where one duty would issue a transaction correction for every cheque error and another duty might issue a consolidated – so if there was a problem with a batch, they might issue a consolidated transaction correction which showed the net effect of the bulk error.
Both kind had kind of quite eloquently described their rationale behind what they were doing in terms of how the subpostmaster saw it, and I could totally understand – it wasn’t clear to me whether it was better for a subpostmaster to see everything on the one transaction correction or lots of different individual transaction corrections. So that was the issue with that one.
A better example might be the Lottery team where there was a big issue accurately recording the Lottery and a lot of that was around the fact that Lottery was typically sold a lot on the retail side, as opposed to the Post Office side, and there was the problem of getting the data across to the Post Office side in time to be reported before cut-off time, which I think was 7.00.
So it was fine for the branch offices that everyone shut up at 5.30, everything was accounted for. Branches that their retail side particularly would be up until 10.00 at night were missing the cut-off.
So what we found was that lots of branches were making lots of errors, which were really just timing errors and we could have been in a situation of issuing every day a transaction correction and then the next day a transaction correction would be coming back which would compensate for the errors.
Mr Beer: So cutting through it, what were the good reasons for the completely different approaches to identifying errors and resolving them?
Andrew Winn: The different requirements and the different attributes or problems of the products and how the team saw best to deal with them.
Mr Beer: This suggests that there was a difference between teams; is that right?
Andrew Winn: In their approach but, essentially, the difference – the teams were doing the same thing. They were looking at an account, where there was – where it wasn’t netting off to zero and their kind of goal in life, I guess, was to get that account down to zero.
Mr Beer: Moving a bullet point on underneath:
“POLFS – space/access to reference data/different transactions used by different teams.”
Can you shortly explain what POLFS was?
Andrew Winn: POLFS, I think, was a name for Product & Branch Accounting.
Mr Beer: Were you concerned about the limitations of or within Product & Branch Accounting, in being able to investigate errors and resolving them?
Andrew Winn: I think – I can’t remember exactly when this was but –
Mr Beer: January 2009?
Andrew Winn: Okay. So at that point, there would be, in many of the teams, quite large backlogs and, clearly, if you have got a large backlog of work, you’d like more resource to help clear it but there were limitations.
Mr Beer: I think that might be a different issue. Isn’t this talking about the extent to which this back office team had access to all of the data generated by both the POL back office systems and by Horizon, for the purposes of an investigation into the validity of a transaction correction or a discrepancy?
Andrew Winn: Their checking a transaction correction would be about what there is in POLFS. Sorry, POLFS is the Finance System, isn’t it? Sorry, I do beg your pardon. It would be about what’s in the Finance System not particularly what was on the Horizon System because the Horizon System should – if this has happened on Horizon, then this data should then flow into this point within POLFS.
Mr Beer: You are speaking to the limitations of POLFS here. What were the limitations of POLFS?
Andrew Winn: I’m sorry, I can’t recall what I was thinking of at that point.
Mr Beer: Next, if we skip over debt levels, we go to “Ownership”:
“Ownership – main focus has been on clearing backlog. Whether that helps the bottom line depends on whether debts generated are paid. Equally there is no incentive to seek out compensating [transaction corrections].”
Does this reflect the fact that the Post Office, through P&BA, was most concerned about getting money back into the business, debt which it considered owing and outstanding?
Andrew Winn: I think that was my – we talked before about me making assumptions about evidence and this is perhaps another example of that. But, certainly, the biggest focus was around getting these accounts down to a manageable format. So the teams were dealing with stuff in a timely manner rather than trying to pull something – work on something that was months old.
But the point I was trying to make was that, just to send out lots and lots of transaction corrections to move these values, if they all end up being disputed, and correctly so in many cases, then the stuff just still flows around within POLFS and isn’t being cleared down.
Mr Beer: Aren’t you emphasising by this that it’s no good doing transaction corrections, just to help the bottom line –
Andrew Winn: Yes.
Mr Beer: – by doing them; you have actually got to follow them up by debt recovery?
Andrew Winn: I think the first point is to make sure that they’re right and if we know that they’re right then we can focus on debt recovery where necessary.
Mr Beer: Was a focus on helping the bottom line, ie by generating money for the Post Office through debt recovery from subpostmasters?
Andrew Winn: I personally wasn’t ever given that pressure, whether somebody like Rod Ismay was, from his Finance Director, I couldn’t say.
Mr Beer: What was the purpose of mentioning whether clearing the backlog helps actually depends on whether the debts are paid by the subpostmasters? Why were you mentioning that?
Andrew Winn: I was trying to make sure that teams weren’t just ramming out as many transaction corrections as possible. The integrity of them was important – well, more than important, it was critical that things went out that were correct, and that would ultimately save because it would potentially just create a circle if we were disputing, reissuing, et cetera. We needed to be able to deal with it once, make sure we dealt with it properly, so it then doesn’t come back on us.
Mr Beer: The last sentence on this page:
“Equally there is no incentive to seek out compensating [transaction corrections].”
Were you suggesting there that there was no incentive within Chesterfield to seek out transaction corrections that would have the effect of the Post Office paying money to the subpostmasters?
Andrew Winn: Yes, I think one of the issues is, as I already mentioned, is different – so we had different teams that dealt with different products. There was two ways that we could have looked at this and the alternative way had been in operation before, where team – individuals looked after branches. So they would look after a branch across the board, which was great because they got a view of the branch but perhaps not so great in terms of knowledge of dealing with products, whereas the way we operated was, in terms of products, so teams became very knowledgeable about their products and hopefully dealt with them well. But they didn’t get a view of the branch they were dealing with.
So whereas if you are looking at a branch as a whole it might be more obvious to pick up “Well, we’ve got an error here, we’ve got an error there. Oh, look, these actually match, we can sort this branch out fine”. Whereas if you’re just looking in a single line under a limited number of products, then you won’t be aware of what’s going on there and if you’re being told “Clear this account down”, then you’re not going to trot off to the other side of the room and have a conversation “Have you got a matching item to this, because I think” – some people will do that anyway; other people wouldn’t.
Mr Beer: In this sentence, were you reflecting the fact that in the team as you saw it there was no incentive to seek out corrections that had the effect of benefiting subpostmasters?
Andrew Winn: I would say that’s correct, yes.
Mr Beer: Can we turn to page 6, please:
“Key Recommendations – Existing Processes.”
Under the third of them:
“Define ‘settle centrally’.”
Then scroll down to the notes, please, and the third point under your notes you say:
“Define Settle Centrally – legally they have accepted the debt. But would a court wear it? Need some assurances around it.”
Can you just explain what “settle centrally” was, please?
Andrew Winn: Settle centrally was an option where a discrepancy arose either through accepting a transaction correction or as a cash balance at the end of the branch trading. I think there was a limit of £150, below which you couldn’t settle centrally but, if, say, you had a debt, a loss of £200, at the end of branch trading, then you had the choice of putting the cash in, which got rid of the debt, making out a cheque, make good by cheque, which sends a cheque off for £200 to clear off the debt, or not to clear the debt at that point but to settle centrally, moves it onto a separate account within POL and, if nothing subsequently would happen from that, the team would then – the debt recovery team would then start to recover that £200.
But it may well be that the subpostmaster is fully aware that they are due a credit transaction correction, which hopefully will arrive in the next trading period, they get that credit transaction correction, settle it centrally, the two net off and everyone’s forgotten about it.
If there’s no compensating transaction correction or compensating credit at the end of the next branch trading, at some point, defined point, the debt recovery team would start looking at recovering that debt.
Mr Beer: So you have identified three options where there’s a discrepancy: firstly, pay up in cash by the subpostmaster –
Andrew Winn: Yes.
Mr Beer: – secondly, pay up by cheque; or, thirdly, settle centrally?
Andrew Winn: Yes.
Mr Beer: They were the only three options?
Andrew Winn: I think multiple branches might have had a different way of doing it and, certainly, Crown Offices had a different approach. But your bog standard subpostmaster –
Mr Beer: They were the three options?
Andrew Winn: – they were their options, yes.
Mr Beer: So the first two involved the payment or the promise of payment of money through a cheque, yes?
Andrew Winn: Yes.
Mr Beer: The third option, settle centrally, you note:
“… legally they [that’s the subpostmasters] have accepted the debt.”
Andrew Winn: That was – I’ve got no legal training whatsoever so that was my understanding of – and also like the NFSP – I think everybody’s understanding was, if you accepted that TC, you have accepted it.
Mr Beer: Where did you get that understanding; where did everyone get that understanding from?
Andrew Winn: I couldn’t specifically answer that.
Mr Beer: If you had that understanding and everyone in P&BA had that understanding, did you communicate that to subpostmasters or their representatives when they spoke to you, “But hold on, by asking to settle centrally you have legally accepted the debt”?
Andrew Winn: No, I would never say that.
Mr Beer: Why not, if that was your belief?
Andrew Winn: Because that wasn’t my – my approach was to make sure that the subpostmasters and POL were treated correctly, that things were done correctly.
Mr Beer: Why were you asking the question “would a court wear it”?
Andrew Winn: Because of some of the problems that we’ve already discussed. Branches are obliged to settle – to accept transaction corrections, which they may not think is proper to them at the time, to get – okay, they might not want to pay the £200 straight away but they will settle centrally, which defers and gives them a chance to challenge. But my non-legal opinion is that, in a court of law, if a subpostmaster went into a court saying, “Look, I was forced to settle centrally, I had to settle centrally, the Post Office is now trying to recover this money, this is the evidence that I have that says it’s not correct”, in my view, the court would say, “Well, yeah, you’re right, the Post Office is wrong”.
Mr Beer: You knew, I presume, that the IMPACT Programme had removed the facility to put any disputed funds into a suspense account where the disputed funds would be identified before the subpostmaster accepted the cash account. You knew that that was the previous position?
Andrew Winn: Sorry, could you read that again for me.
Mr Beer: Yes. That before the IMPACT Programme – did you know about the IMPACT Programme?
Andrew Winn: I recognise the name but I’m not quite sure what that did.
Mr Beer: That previously there was a facility to put disputed funds into a suspense account before the subpostmaster accepted the cash account?
Andrew Winn: All right. No, I wasn’t aware of that or I can’t remember being aware of it.
Mr Beer: But the position from when you came into post was that that wasn’t possible on rollover on branch trading?
Andrew Winn: That’s correct, yes.
Mr Beer: You had to either pay money or promise to pay money, which you understood to mean that the subpostmaster had accepted their liability to pay the debt?
Andrew Winn: Yes.
Mr Beer: The fourth bullet point that top of the page, “Disputes resolved prior to DFR”, and then your notes, which are the next paragraph on, at the bottom:
“Disputes pre-DFR – should not happen but need to lay out their responsibilities as well as ours. How does a dispute become formal? They need to present evidence.”
What does “dispute pre-DFR” mean, please?
Andrew Winn: Branches having the opportunity – DFR means “deduction from remuneration”. So we would start – the debt recovery team would start taking money from –
Mr Beer: From their pay?
Andrew Winn: From their pay, yes. So it was about no DFR should be happening – should start while a dispute is still live.
Mr Beer: So would a dispute only be treated as formal by the Post Office after the presentation of evidence by a subpostmaster?
Andrew Winn: Yes. Well, not necessarily evidence but, for example, if a subpostmaster had written in to me, then we would put a block on the debt recovery process until I’d responded, backing up or accepting – backing up POL’s position or accepting the subpostmaster’s position.
Mr Beer: Would a subpostmaster’s statement that the figures that were being produced by Horizon didn’t tally with the records that they had kept in store be sufficient?
Andrew Winn: If you are talking about paper records or – I’d certainly look at Horizon records and – well, I’d be looking – as part of my job, I’d be looking at Horizon records to see what was in it.
Mr Beer: Let’s take a different example. What about a subpostmaster that suggested that a large transaction that was shown by Horizon had, in fact, never taken place.
Andrew Winn: Right. I don’t think that that ever came up but, if that did, then that would be a massive red flag.
Mr Beer: Put another way, what evidence did you have in mind when they said “they need to present evidence”; what evidence was sufficient?
Andrew Winn: Pretty much every case is different but they would lay out what their understanding of what happened, where they thought the problem was. I would investigate that, I would look into Horizon records and try and understand what’s happened and be able to explain what’s happened and hopefully find a resolution or illustrate why this has happened.
Mr Beer: Was there any formality brought to bear, any description, on what evidence was sufficient to make a dispute formal and, therefore, stop deductions from remuneration?
Andrew Winn: No. The process was to write in to me. Once that letter arrived, and it could be a very basic – some subpostmasters would write reams of pages, others would – half an A4 but that would create a formal dispute. Any debt recovery would be paused at that point and not reinstated or started until I’d responded to the subpostmaster’s concern. It may well be, if there’s just a short amount, I might we’ll have to go back to the subpostmaster and say, “Look, I need a bit more what can you provide us”, but then the dispute will be myself and the subpostmaster trying to find the evidence that supported the issue one way or the other.
Mr Beer: What if the subpostmaster said that, “The figures in Horizon are just wrong, I can’t tell you why they’re wrong”?
Andrew Winn: Then I would look at it to try and work out what – if there was – what I’d typically do, if we’re saying that a branch, for example, has got a cash shortage and we knew – we could see that on the previous evening that they’d balanced, or whatever the discrepancy was, and then at the end of the next day they’d got this particular cash shortage, then what I would typically do is look at the transactions for that branch for that day. There may be something that just jumps out at me straight away, I can say “Oh, yes, I can seek what the issue is”. Most usually there wouldn’t be.
But I would send – potentially send an Excel spreadsheet to branches where they could look at the transactions in a better format than what Horizon reporting would come up and they could have a look and say if there’s any transaction – and I’d suggest “Look, you know, there’s this transaction, is that going to be right; is this transaction going to be right? Have a look at this one or – but have a look at them all, see if there’s any transactions that you don’t recognise that you think is suspicious, where there might be miskeying, have I missed anything?”
Mr Beer: You were working only from the data that Horizon itself produced?
Andrew Winn: Yes.
Mr Beer: What if that data was wrong?
Andrew Winn: Then I’d be looking for a branch to say something like, “The butcher always comes in at 3.30 and makes a cash deposit and there isn’t one there”. So we need to understand did the butcher actually come in that day or he did and there’s no record of it on Horizon, or there’s a transaction here which I don’t recognise, that hasn’t happened.
I can’t recall that actually happening.
Mr Beer: Can we, just before the morning break, look at the last page of this document, please, page 7. The first bullet point:
“Remove second reminder letter” and then look at the notes underneath:
“Recovery – remove one letter in process – not contractual and adds no value. People either pay up or hang out to DFR – may as well get there.”
Were you by this saying that a step in the process should be removed so that you could get to debt recovery from the subpostmaster sooner rather than later?
Andrew Winn: Yes.
Mr Beer: Is that a reflection of pressure to recover, from subpostmasters, debts?
Andrew Winn: No. No, I would say it was more a case of we’d done some analysis and found out that people paid the debts straight away, or once we sent a letter, they paid the debt. The second letter didn’t make any difference.
They obviously didn’t get the second letter if they’d already paid. If they hadn’t paid, they didn’t start paying. So there was no benefit in sending that out. It saved resource within the team and, yes, it would have had an impact on the bottom line but it would have been pretty marginal.
Mr Beer: And the “may as well get there”, you’re saying you may as well get to taking money from subpostmasters’ wages sooner rather than later?
Andrew Winn: Yes.
Mr Beer: Thank you very much. Can we take a break there for 15 minutes, please?
Sir Wyn Williams: Yes. So that takes us to 11.45; is that right?
Mr Beer: Yes, please.
Sir Wyn Williams: Fine.
(11.32 am)
(A short break)
(11.48 am)
Mr Beer: Good morning, sir. Can you see and hear me okay.
Sir Wyn Williams: Yes, I can thank you.
Mr Beer: Thank you very much.
Mr Winn, I want to look at something called the receipts/payments mismatch book. Can we start by looking at POL00028838. Thank you very much.
These appear to be notes about a meeting to discuss the receipts/payments mismatch bug. Can you see at the top “Receipts/Payments Mismatch issue notes”?
Andrew Winn: Yes.
Mr Beer: We can see that the attendees at a meeting to discuss the receipts and payments mismatch issue include you?
Andrew Winn: Yes.
Mr Beer: “Andrew Winn (AW) POL Finance”. We can also see that Mr Jenkins from Fujitsu was there, yes?
Andrew Winn: Yes.
Mr Beer: Can I just deal with the date of this meeting first. This document is not dated and does not itself identify the date of the meeting but if we look at page 3 of the document, we can see in the second paragraph Fujitsu are writing a code fix which will stop the discrepancy disappearing, et cetera, et cetera, and then there are some dates mentioned of 4 October, that’s 2010, and then a date of 11 October, 21 October, yes?
Andrew Winn: Yes.
Mr Beer: Then if we go to page 4, on this page and on the next page there are a series of actions. Would these be actions arising from the meeting?
Andrew Winn: That would seem sensible.
Mr Beer: You can see that there are target completion dates ranging between 6 and 8 October.
Andrew Winn: Yes.
Mr Beer: Would it follow from this that the meeting is likely to have taken place in September 2010 or early October 2010?
Andrew Winn: I’m not sure where the years come from. I’d say September certainly but – have we seen the year?
Mr Beer: We haven’t seen a year but we know that this bug was only discovered in that year and therefore I’m taking that as the year.
Andrew Winn: I’ve got nothing to argue with there.
Mr Beer: Okay. Can we go back to page 1, please, and look at the nature of the receipts and payments mismatch bug. If we look at page 1, underneath the table, under the cross-heading “What is the issue?” it reads:
“Discrepancies showing at the Horizon counter disappear when the branch follows certain process steps, but will still show within the back end branch account. This is currently impacting circa 40 Branches since migration onto Horizon Online, with an overall cash value of circa [£20,000] loss. This issue will only occur if a branch cancels the completion of the trading period, but within the same session continues to roll into a new balance period.”
Then if we go on to page 2, please, in the middle in bold, I think it’s in bold, anyway:
“Note the Branch will not get a prompt from the system to say there is a Receipts and Payments mismatch, therefore the Branch will believe they have balanced correctly.”
Then under lastly “Impact” at the foot of the page, the first two bullet points:
“The branch has appeared to have balanced whereas in fact they could have a loss or a gain.”
And:
“Our accounting systems will be out of sync with what is recorded at the branch.”
Does all of the information in those three places accurately and fairly describe the nature of what was then understood about the receipts and payments mismatch issue?
Andrew Winn: Yes, I think so, although my memory of the receipts and payments mismatch would be that the branch trading statements would show a compensating figure. I can’t remember where it would be. It’s not something I would expect branches to see and they would believe that they’d balanced correctly but, if my memory serves correct, on the branch trading statement there would be a value showing.
Mr Beer: If that memory is correct, why does this say the opposite?
Andrew Winn: That’s a very good question, although it doesn’t say anything about the branch trading statement.
Mr Beer: That would be the obvious place where a loss or a gain would be shown and this is saying that the branch won’t get a prompt and the branch will believe they have balanced correctly and under “Impact:
“The branch appears to have balanced, whereas in fact they could have a gain or a loss.”
Andrew Winn: I think my memory as it is would be that when you get, towards the end of your branch trading statements, your final kind of thing before rolling is to say “This is the balance” or “There is no balance”. So I think that is the point that subpostmasters would look at and say “I’m fine, I’ve got nothing to put in, take out, whatever. I’ve balanced to the penny, that’s great, press rollover”, and then a statement will roll out but whether it’s actually reviewed in detail, I would say not.
I’ve got to apologise, I may be wrong on that, but that’s my memory from the receipts and payments mismatch.
Mr Beer: Would you accept that the contemporaneous evidence suggests that it was believed that there wouldn’t be anything in branch to show the branch that they had not balanced?
Andrew Winn: I think – I’m surprised about that. I think the branch trading statements would show – whether the branch would be able to interpret it as not balancing, I’m not sure. I couldn’t say that. But I think branches would believe, from the branch trading process, they had balanced and I believe a lot of branches did not routinely check over their branch trading statements when it balanced. But that’s –
Mr Beer: Just one final point at this. We could look at another place at the top of this page:
“Note at this point nothing feeds into POLSAP and Credence, so in effect the POLSAP and Credence shows discrepancy whereas the Horizon System in the branch doesn’t. So the branch will then believe they have balanced.”
Does that help you with some further – a further contemporaneous record to suggest that the branch will believe they have balanced, whereas, in fact, they have not?
Andrew Winn: Yes, I would – that does suggest it’s not that I’m mistaken, it wasn’t shown on the branch trading statement. My main recollection is that the balancing amount showed in a discrepancy account within POLSAP. So I would accept that it may not have shown, I’m mistaken on my memory of that.
Mr Beer: I understand, thank you.
Can we look at the date of discovery of the receipts and payments mismatch bug and can we look, please – well, can you remember when you attended this meeting for how long the payments and mismatch bug had been discovered or was this the first that you became aware of it?
Andrew Winn: I must have been. I wouldn’t have just gone to a meeting without knowing anything about it. So I must have known something beforehand.
Mr Beer: Can you remember from whom you learnt that?
Andrew Winn: No, sorry.
Mr Beer: Who would it be likely to be?
Andrew Winn: Not sure.
Mr Beer: Can we look at the third page of the document, please. The top paragraph:
“The Receipts and Payments mismatch will result in an error code being generated which will allow Fujitsu to isolate branches affected by this problem, although this is not seen by the branches. We have asked Fujitsu why it has taken so long to react to and escalate an issue which began in May. They will provide feedback in due course.”
Can we firstly look at whether that’s accurate, that the problem first began in May and then what response the Post Office got when it challenged Fujitsu on why did taken so long to react and escalate the issues.
So firstly the date on which the problem was discovered. If the problem was discovered in May, that would be just before Horizon Online was accepted in June 2010; would that be right or don’t you remember when Horizon Online was accepted?
Andrew Winn: I don’t remember exactly.
Mr Beer: We know the acceptance of Horizon Online was June 2010. This record here suggests that the problem was discovered or it began in May. If it’s correct that the problem began in May, that would be about four months or so before this meeting, if we’re right that this was September/October time, yes?
Andrew Winn: Yes.
Mr Beer: Can we just look, please, at POL00029084. This is an email exchange that you weren’t copied into but I want to ask you something about the content of it.
Andrew Winn: Sure.
Mr Beer: If we look at the foot of the page, if we scroll down, please, we can see this is an email from Gareth Jenkins, on a date in September 2010, to Mark Wright. It’s cut off on the page but if we scroll up we can see the reply from mark is Mark Wright. Now, we know that Mr Jenkins attended the meeting that you attended, yes?
Andrew Winn: Yes.
Mr Beer: Just if we look at the penultimate paragraph on this page, at the foot of the page:
“Jon is easily able to reproduce the problem in a development environment and we are planning to recreate the scenario and attempt a fix in that environment in the next couple of days. However it is probably worth starting on the data extraction to ascertain the full scope of the issue … since it has probably been around since day one and data more than 6 months old is being dropped from BRSS, so the sooner we run the queries the better.”
If that’s right it’s probably been around since day one and data is being dropped after 6 months, that would put the bug’s existence before May 2010, wouldn’t it?
Andrew Winn: Correct, yes.
Mr Beer: From an email being sent in September 2010. Were you ever given that information by Fujitsu?
Andrew Winn: No.
Mr Beer: How would it have affected your conduct and thinking if you had been told that information, that the bug had probably been around since day one?
Andrew Winn: It would have been a little bit scary, I think.
Mr Beer: Turning to –
Andrew Winn: Sorry, can I just expand on that a little bit. Myself, P&BA, were pretty much dependent on Fujitsu alerting us to what branches were affected by a receipts and payments mismatch. So if we weren’t – if my memory is correct, the discrepancy would fall into the discrepancy accounts within our team. My worry from that earlier period is – when I’m saying there’s lots of things flying around into the wrong accounts, and what have you – whether any values ended up getting written off because we were – my later memory of the receipts and payments mismatch was that it sits there, they have a discrepancy of 10 quid, 10 quid sat in the discrepancy account, we just basically give that back to the branch and away we go. If we haven’t got the money in the discrepancy account, we wouldn’t know there was any issue. So yes, that would worry me.
Mr Beer: Can I go back to the second question then, what was done within POL to question why Fujitsu hadn’t reported the problem sooner. If we just go back to the document we were previously looking at, which was POL00028838, and go to page 3, please. It’s that paragraph at the top, again, in the second line:
“We have asked Fujitsu why it has taken so long to react and escalate an issue which began in May. They will provide feedback in due course.”
What was done within POL to question why Fujitsu had not reported the problem whenever it began?
Andrew Winn: I can’t answer that. That wouldn’t be me who was raising that question. I would guess that was someone from the IT team within POL.
Mr Beer: Just looking at the first page of the document, the list of the people there, which of those people would be doing the questioning of Fujitsu, why it had taken so long to report the issue?
Andrew Winn: I would guess Ian Trundell.
Mr Beer: Because he has “IT” written next to him?
Andrew Winn: Yes. But also I think he’s probably the most senior manager on that list.
Mr Beer: The first six of you are from POL – is that right – from Post Office –
Andrew Winn: Yes.
Mr Beer: – and then the last four or from Fujitsu?
Andrew Winn: Yes.
Mr Beer: He was the most senior person present?
Andrew Winn: I believe so.
Mr Beer: None of the people there appear to be from Legal; is that right?
Andrew Winn: No, but Security is probably quite close that would kind of deal closely with Legal but, yes, you’re correct, nobody from Legal.
Mr Beer: Can you recall what the response was from Fujitsu, if it was passed on to you: why has it taken so long to react to and escalate this issue which, according to them, began in May?
Andrew Winn: No, I’ve got no recollection, I’m afraid.
Mr Beer: Why would it be important to pursue that with Fujitsu or would it be important to pursue that?
Andrew Winn: Yes.
Mr Beer: Why would it be important?
Andrew Winn: Because we would need to know things in a timely manner to make sure that branches are properly supported if there are issues impacting them.
Mr Beer: We’re on page 1 of the document, can we look at the second paragraph:
“At this time we have not communicated with branches affected and we do not believe they are exploiting this bug intentionally.”
Putting aside for one moment the cases in which data produced by Horizon was being relied on in the prosecution of subpostmasters, why wouldn’t POL tell branches affected and, indeed, other branches that there may be a bug in the Horizon System that was affecting the balancing process?
Andrew Winn: I don’t know. I think there would be a concern about putting out a branch-wide notice but I think if we identified a branch had been impacted by the problem then they need to know about it as soon as possible and told that we’re sorting it out.
Mr Beer: Can we look at page 6, please. This is a document dated 29 September 2010, produced by Mr Gareth Jenkins. If we just expand it, please, to look at the whole document and just take a moment. You’ve seen this before. This has been, I think, sent to you.
Andrew Winn: No, it doesn’t ring a bell. It might do.
Mr Beer: It’s in the pack of documents that was sent to you but it’s about the same issue.
Andrew Winn: Okay.
Mr Beer: It’s a note by Mr Jenkins. Can you remember whether this was a document that was sent to you at the time or tabled at the meeting that we’re discussing?
Andrew Winn: No, I’ve got no memory of it, I’m afraid.
Mr Beer: Let’s look at page 8, please. At the foot of the page, in the last paragraph:
“It should be noted that as Discrepancies are normally Losses, then a Lost Discrepancy would normally work in the Branch’s favour and so there is no incentive for the Branch to report the problem. Also if we do amend the data to reintroduce the Discrepancy, this will need to be carefully communicated to the Branches to avoid questions about the system integrity.”
Why would anyone wish carefully to communicate information to branches to avoid questions about the integrity of the Horizon System?
Andrew Winn: I would assume to illustrate – to suggest that it’s a controlled issue rather than a kind of all-embracing, the system’s bust, issue.
Mr Beer: Can I put it another way: this bug did raise questions about the integrity of the Horizon System, didn’t it?
Andrew Winn: Yes.
Mr Beer: What proper or appropriate reason would there be for not telling people the existence of the bug and, therefore, its impact on the integrity of the Horizon System?
Andrew Winn: I can’t answer that.
Mr Beer: Wouldn’t the reason be that care would need to be taken because, if the full facts were revealed, it would raise questions about the integrity of the Horizon System and that might damage the business of Post Office and Fujitsu?
Andrew Winn: Yes, I think it would need to be presented in a way of saying “We’ve identified this, this is the impact on branches, this is what we’re doing about it”. I just think it would need to be explained carefully.
Mr Beer: Why would you want to avoid – why would anyone want to avoid questions about the system’s integrity?
Andrew Winn: It’s difficult to put that into the correct words. I’m sorry, I’m struggling to find the right words.
Mr Beer: I can understand.
Can we go back to page 2, please. This is back to the note prepared, we think, following the meeting and so of the meeting. At the foot of the page, we looked at the first two bullet points under “Impact”. Can I examine the remaining three, please, at the foot of the page. “Impact”:
“If widely known could cause a loss of confidence in the Horizon System by branches.”
Would you agree that a fairer and more balanced way of writing the sentiments behind that sentence would be: if the bug was widely known, it could ensure that branches are provided with accurate information about a known fault in the system that they are required to use?
Andrew Winn: Yes.
Mr Beer: Why wasn’t the discussion along those lines – let’s provide accurate if to subpostmasters about the facts as they are known – rather than: if this bug is widely known, it will cause or could cause a loss of confidence in the system by branches?
Andrew Winn: Yes, I think it’s a fair question. I think I was certainly of the view that impacted branches should be fully informed of what’s happening. I don’t think I ever really considered a branch-wide communication.
Mr Beer: Would a fairer approach to have been to discuss and conclude that the dissemination of such information would ensure that subpostmasters and those who were investigating them do not proceed on the false basis that the Horizon System is robust and that discrepancies are always the responsibility of the subpostmaster?
Andrew Winn: Yes, yes.
Mr Beer: Does the fact that this note was not written in either of those ways reflect the fact that that wasn’t a view expressed by anyone at the meeting?
Andrew Winn: I’ve not got a clear enough memory of the meeting. I can’t remember anybody supporting that suggestion.
Mr Beer: Why would that be? Why would anyone not support the suggestion that I’ve made: let’s reveal a known fault in the system so on that people don’t proceed on a false basis?
Andrew Winn: Yes, I think in hindsight you’re probably right. I think what you need to be able to do is “This fault has been identified, this is the impact of it, this is how you can check to ensure that your branch has not been affected. The vast majority of branches have not been affected”, but giving the branch the opportunity to check that they haven’t got that problem – something along that kind of line.
Mr Beer: Was there a definitive list of those branches that were affected?
Andrew Winn: I believe so but we depended on Fujitsu to identify these branches.
Mr Beer: Given that they had seemingly delayed in the provision of information to you promptly, was there confidence in the definitive list that they produced?
Andrew Winn: I think there was – personally, I think there was a concern “Have they identified every branch”, but kind of would have felt “Well, what can I do about it?” Fujitsu have got the information. We don’t have the information to check that. I don’t know – I wouldn’t know where to go from there. It wasn’t something where I was able to sit there with Fujitsu and say, “I don’t believe you’ve told us about every branch”.
Mr Beer: You can test their methodology, can’t you: how have you reached this; what investigations have you undertaken; who’s involved; how seriously are you considering this; what effort have you put into it?
Andrew Winn: Yes, and I think that would be something that the IT – ie Ian Trundell, that would be his area to investigate that. I could ask a question similar to that and basically not understand a word that comes back, I’m afraid. I’ve already said before, my IT knowledge is very limited.
Mr Beer: So the next bullet point:
“Potential impact upon ongoing legal cases where branches are disputing the integrity of Horizon Data.”
Did you or anyone else in the meeting, to your knowledge, consider whether information about the bug needed to go to POL Legal in order to make disclosures in ongoing court cases?
Andrew Winn: That would seem sensible.
Mr Beer: Did anyone do that which seemed sensible?
Andrew Winn: I can’t recall, I’m afraid. I wouldn’t have.
Mr Beer: Whose responsibility, on the POL side of the house, of the people that we saw in the list – do you want to go back to the list – would it be?
Andrew Winn: Can we go back to the list?
Mr Beer: Yes. It’s page 1.
Andrew Winn: I would probably say Alan Simpson.
Mr Beer: Because he’s from Security?
Andrew Winn: Yes.
Mr Beer: So we should highlight Mr Simpson as being the person responsible for making disclosures of this bug to POL Legal, in order to potentially make a disclosure in ongoing legal proceedings?
Andrew Winn: That seems sensible. I don’t know if that was the process that should or actually did take place but that sounds a sensible summation or assumption.
Mr Beer: The notes for this meeting generally suggest that the Post Office knew that knowledge of the bug should create caution over the accuracy of some of the data that Horizon was producing?
Andrew Winn: Yes.
Mr Beer: And that that could undermine confidence in Horizon?
Andrew Winn: That would be a concern, yes.
Mr Beer: You would know also, wouldn’t you, that that information would be relevant to subpostmasters who had been prosecuted?
Andrew Winn: Yes.
Mr Beer: And those who might be prosecuted in the future?
Andrew Winn: Yes.
Mr Beer: And the information would be relevant to anyone who had had proceedings brought against them by POL for the recovery of debts, civil recovery?
Andrew Winn: Yes.
Mr Beer: Or who had challenged POL –
Andrew Winn: Yes.
Mr Beer: – in the courts.
Did anyone in the meeting discuss handing that information over to the defence teams?
Andrew Winn: No, I cannot recall that being mentioned.
Mr Beer: When you attended this meeting in September/October 2010, I think you were already aware of the prosecution of Seema Misra, weren’t you?
Andrew Winn: It’s a name that rings a bell.
Mr Beer: Can we look, please, at POL00055100. Can we look, please, at the middle of the page. We can see this is an email sent to you by Jon Longman on 27 July 2010, concerning the prosecution of Seema Misra at Guildford Crown Court, yes?
Andrew Winn: Yes.
Mr Beer: We’re going to come to that in a moment but let’s get some context first. Can we look at page 2 of the document, please, and then just scroll down. This is an email from Issy Hogg, she is the defence solicitor, to Jarnail Singh. If we just scroll to the top of the page, we can see on his signature block Jarnail Singh was a senior officer in the Post Office’s Criminal Law Division, yes?
Andrew Winn: Yes, yes.
Mr Beer: If we scroll back down, please, Ms Hogg says:
“Jarnail,
“As a result of the meeting that took place between Charles McLachlan and Gareth Jenkins as directed by the judge, we now need to have:
“access to the system in the Midlands where it appears there are live, reproducible errors.
“access to the operations at Chesterfield to understand how reconciliation and transaction corrections are dealt with.
“access to the system change requests, Known Error Log and new release documentation to understand what problems have had to be fixed.
“Please you contact me with regard to these issues. Please you respond [to an email address].”
Yes?
Andrew Winn: Yes.
Mr Beer: Then if we go further up the page, we can see a forwarding:
“… please advise on the three points raised below following our telephone conversation of today.”
Then go back to page 1, please, the foot of the page, an email sent on behalf of Mr Singh to Jon Longman. Can you remember who Jon Longman was?
Andrew Winn: No, I can’t, no.
Mr Beer: And Warwick Tatford, he was prosecution counsel in the Seema Misra trial:
“I enclose a copy of an email received from Issy Hogg, the defence solicitors of 22 July 2010, the content of which is self-explanatory. Could you please be kind enough to let me have your urgent instructions as to the access and information she is requesting in respect of the system in the Midlands and the operation at Chesterfield and the errors logs. I will contact Gareth Jenkins to find out what transpired at the meeting with Charles McLachlan.”
Then scroll up, please. We can see that on the same day, about an hour later, Mr Longman forwarded the email to you, yes?
Andrew Winn: Yes.
Mr Beer: Then we can see about 33 minutes later you reply, so at 12.13 that day, and you say:
“John
“Rod Ismay the head of P&BA is not happy at the prospect of an open ended invite. He has asked the question of what are the legal parameters we are working within. Simplistically if we refuse or impose conditions do we lose the case? I think we need more guidance on how something like this might reasonably operate.
“I think Mark Burley would be the route into IT to identify who might be best placed to deal with that aspect.”
This was a reply that was essentially your boss, Rod Ismay, in P&BA, seeking to close down the disclosure request as much as possible, wasn’t it?
Andrew Winn: Yes.
Mr Beer: What reasons did he give to close down the disclosure request as much as possible?
Andrew Winn: I can’t recall any more detail than what’s shown there.
Mr Beer: What was the nature of his unhappiness about the disclosure request?
Andrew Winn: I think he didn’t feel as though it would produce anything and create more questions than it would answer.
Mr Beer: What kind of questions would it create?
Andrew Winn: I can’t recall. I don’t think he specified. I think that was probably a generic term or –
Mr Beer: Was it a generic reply “Don’t let them have access to systems or data that would reveal issues or problems with our system”?
Andrew Winn: No, I don’t think so. I don’t think – knowing Rod, I don’t think he would view it that way. I think he’d just struggle to understand what it was that they would want to see and where it might lead, and there’s also an issue of confidentiality in terms of the data we’re holding, financial information regarding other branches, for example.
Mr Beer: Why wasn’t the reply along those lines then: there are some practical hurdles to this, rather than just closing it down?
Andrew Winn: Yes, I think it’s the parameters that we’re working with, so, for example, other subpostmaster information.
Mr Beer: Why did you suggest Mark Burley would be a person who ought to be involved?
Andrew Winn: I can’t recall. I knew Mark, I worked in the product process improvement team with him. Presumably the role he was in at that point would seem the most appropriate one. I think I possibly had spoken to him about it beforehand.
Mr Beer: Can we turn, please, to POL00055225. This is an email dated 13 September 2010 and you can see the subject “West Byfleet”, that was her post office “Mrs Seema Misra”. You can see who it’s from and to, not you, but we’re going to see in a moment that it references a conversation with you. It reads:
“Hi John
“The last update I had on the above was in July, the defence solicitors had requested they had access to the operations in Chesterfield.”
I think that’s the email discussion we’ve just seen.
Andrew Winn: Yes.
Mr Beer: “This was discussed by Andy Winn/Rod Ismay. I have today spoken with Andy Winn and he has informed me that Rod had made a decision to not allow this. Therefore could you please update me with the latest progress on the case.”
Do you now recall any conversation with Mr Ismay?
Andrew Winn: Yes, I recall a conversation but the detail of it, beyond “I don’t think it’s a good idea”, and what I stated in the previous email, I can’t remember any more detail than that, I’m afraid.
Mr Beer: The record here is that Mr Ismay had made a decision not to allow it, so was bringing the shutters down on it?
Andrew Winn: I think that would be – I can’t remember what the difference between – in the timings was but, yes, that would be fair summation. I don’t know if anything more had happened in between the first document that we looked at and this one.
Mr Beer: What reasons were discussed between you and Ismay for refusing access that was sought by the defence?
Andrew Winn: Sorry, I thought we covered that in the previous –
Mr Beer: Yes, the previous email was “We need to know what the implications are. If we refuse access will we lose the case?”
Andrew Winn: I didn’t hear any more – I don’t believe I heard anything more from that up to this point, apart from the fact that Rod had – no, I can’t remember if there was anything more that developed.
Mr Beer: Did the reasons include that because if such access is given then the lack of integrity of the data that Horizon produced may be discovered?
Andrew Winn: That was never stated, no.
Mr Beer: Even though, by this time, you may have been aware that there was a bug that was producing data that lacked integrity?
Andrew Winn: Yes, but – well, yes.
Mr Beer: But even if you wouldn’t allow a defence solicitor or a defence expert into the building to examine P&BA operations or P&BA equipment, you’d presumably have discussed, wouldn’t you, “Look, there’s someone on trial for a very serious crime here based on data produced by Horizon. She’s alleging that the data’s not accurate. We know that the data produced by Horizon may not be accurate. We need to find out a way of ensuring that she knows what we know” – wouldn’t you?
Andrew Winn: Yes, I totally understand from their point of view, yes.
Mr Beer: You would agree that that would be the open thing to do.
Andrew Winn: Yes.
Mr Beer: You would agree that would be the honest thing to do?
Andrew Winn: I think, within the concerns about data protection for other branches and issues like that, yes.
Mr Beer: You would agree that that would be the thing a person with integrity would do?
Andrew Winn: I’m not – I think Rod laid out his reasons, his concerns about it. I don’t know if there was any comeback from that.
Mr Beer: Well, the comeback would have come from you, wouldn’t it, because this was a conversation between the pair of you?
Andrew Winn: We had the conversation I would have expected probably to have said, “Yes, they can come in” but I’ve got to say that I didn’t put up any impassioned disagreement with Rod. I accepted that he took a considered view of things and he was my boss and, yes, I didn’t put up a fight.
Mr Beer: If it was the open and honest thing to do, if it was the thing that a person with integrity would do, why did you and Mr Ismay not do it?
Andrew Winn: Yes, I mean I’ve got to say – refer to the previous question. That was Rod’s view. I didn’t have a strong enough view to argue against it.
Mr Beer: Were these requests by defence solicitors for access to systems and data on your mind when considering whether and how to communicate with affected subpostmasters and others the discovery of the receipts and payments mismatch bug in the September/October meeting?
Andrew Winn: No.
Mr Beer: Can we go back, please to Fujitsu 00081584. That’s an errant reference. I’m going to stick with the version of the document I’ve been using, POL00028838, the second page, please. At the foot of the page under “impact”, the third bullet point:
“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.”
This appears to be a record that those two things were brought into account at the meeting. Can I ask again: were the discussions over the defence requests for access to Horizon systems and data on your mind when considering whether to communicate with subpostmasters and others about the discovery of the receipts and payments mismatch bug?
Andrew Winn: No.
Mr Beer: What’s this referring to then?
Andrew Winn: Sorry, I don’t really understand which we’re going.
Mr Beer: We’ve seen that at two points, once in July and once in September, you were involved in conversations over defence access to systems and data on behalf of a subpostmistress who said the data provided by Horizon is not accurate. You’re in a meeting about data provided by Horizon not being accurate and there’s a record saying it has a potential impact upon ongoing legal cases where branches are disputing the integrity of the Horizon data.
I’m asking you were the requests that you received for access on your mind when you were considering the potential impact on legal cases that’s recorded at this meeting?
Andrew Winn: I would probably say no.
Mr Beer: Was the discussion at this meeting over whether Mr Jenkins was going to give evidence in Seema Misra’s trial in October 2010?
Andrew Winn: I’ve no idea.
Mr Beer: Was anything said about the Seema Misra case in this meeting?
Andrew Winn: I wouldn’t have thought so.
Mr Beer: Were any decisions made about the disclosure of the existence of the bug to the defence team?
Andrew Winn: Not that I’m aware of.
Mr Beer: If we just go on to page 6 of the document, this is the Gareth Jenkins document that we mentioned earlier. We know that this was sent on to the solicitor with the conduct of the Seema Misra prosecution, Jarnail Singh, on Friday, 8 October before her trial started on Monday, 11 October.
We can see that if we just go to the foot of the page, please. Do you see the file string at the foot of the page showing the presence of the document in a Jarnail Singh folder and that it was printed on 8 October?
Andrew Winn: Yes.
Mr Beer: I won’t turn it up now but there’s also an email. I’ll give the reference for the transcript, POL00055410, which shows this document being sent to Mr Singh.
Was there discussion at the meeting of the need to disclose information that Mr Jenkins had to the prosecution solicitor in the Seema Misra case?
Andrew Winn: I don’t recall. I don’t recall that case being raised in this meeting.
Mr Beer: We know that neither Mr Singh nor Mr Jenkins disclosed information about the receipts and payments mismatch bug to Ms Misra’s defence team or indeed to the court. Was that discussed in the course of the meeting, whether that information should be disclosed or not?
Andrew Winn: Again, I don’t remember this ex-subpostmaster being mentioned in the meeting.
Mr Beer: Can we go to page 2 of the document, please, and look at the foot of the page. We’ve dealt with the first five bullet points. Can I deal with the last – sorry, we’ve dead with the first four bullet points. Can we deal with the last, the fifth:
“It could provide branches ammunition to blame Horizon for future discrepancies.”
Was that a consideration?
Andrew Winn: Yes.
Mr Beer: So disclosure could provide branches with ammunition to blame Horizon?
Andrew Winn: Yes.
Mr Beer: It would be accurate ammunition, though, wouldn’t it?
Andrew Winn: It would be accurate in terms of there is a Horizon integrity impact which needs to be checked out before going forward. I would say that’s reasonable, yes.
Mr Beer: So if the ammunition was an arrow, it would fly true and straight, wouldn’t it?
Andrew Winn: If it was relevant to the branch, yes. I think if a branch had a discrepancy, they would be entitled to assume that before any type of prosecution, or anything like that, that it was checked that there were no error impacts on their accounts, yes.
Mr Beer: And it could be ammunition that hit the correct target; namely, a discrepancy is being caused by the Horizon System, not by the subpostmaster?
Andrew Winn: Yes.
Mr Beer: It wouldn’t be information that damaged, hurt or injured any subpostmaster? It’s not that kind of ammunition, is it?
Andrew Winn: It could end up working out the wrong way round because the receipts and payments mismatch could produce gains or losses but –
Mr Beer: That would be a good thing, though, wouldn’t it, because it would mean recovery to the Post Office of debt that was truly owing?
Andrew Winn: Yes.
Mr Beer: But it wasn’t seen in any of those ways, was it? It was seen in the light of “we can’t disclose material that might undermine our system, even if the system is in fact faulty”.
Andrew Winn: Yes, I think that’s probably a fair summation.
Mr Beer: Can we look over the page, please. Under “Proposal for affected branches”, the document reads:
“There are three potential solutions to apply to the impacted branches. The group’s recommendation is that solution 2 should be progressed.”
Can we look at solutions 1 and 3 first and then go to solution 2. So solution 1:
“Alter the Horizon branch figure at the counter to show the discrepancy. Fujitsu would have to manually write an entry value to the local branch account.
“Impact – when the branch comes to complete next trading period, they would have a discrepancy, which they would have to bring to account.
“Risk – this has significant data integrity concerns and could lead to questions of ‘tampering’ with the branch system and could generate questions around how the discrepancy was caused. The solution could have moral implications of Post Office changing branch data without informing the branch.”
So summing up solution 1 is: the branch never knows about this but there’s a fix applied; is that right?
Andrew Winn: That’s fair, yes.
Mr Beer: What were the moral implications?
Andrew Winn: I don’t understand where that came from.
Mr Beer: Can you try and help us?
Andrew Winn: It’s just not a term I would use. I would certainly say it was important that Fujitsu or – well, presumably Fujitsu, weren’t changing figures willy-nilly without the branch being aware of it. That shouldn’t happen.
Mr Beer: So it reflects a belief – is this right – that whenever Fujitsu changed branch data, the branch should be informed?
Andrew Winn: I think there’s – yes, in that kind of scenario changing branch data is a very kind of wide thing. So you change the price of a first class stamp every year. You don’t particularly ring the branch up and say we’re doing that. So in the scenario you’re talking about here, then yes, very much so.
Mr Beer: And that wouldn’t be moral in the language of the document here?
Andrew Winn: It just seems a strange terminology to me.
Mr Beer: Can you remember who made the suggestion?
Andrew Winn: No.
Mr Beer: Can you remember who rejected it?
Andrew Winn: I would have rejected it. Whether other – and people would have as well from my – not particularly a memory of that meeting but the type of thinking that NBSC tended to think in the same way as me, the Networks tended to think in the same way as me.
Mr Beer: Can we look at solution 3 then, please, at the foot of the page. It’s decided:
“… not to correct the data in the branches (ie Post Office would prefer to write off the ‘lost’” and then it’s missing.
“Impact – Post Office must absorb around £20,000 loss.
“Risk – huge moral implications to the integrity of the business, as there are agents that were potentially due a cash gain on their system.”
So this is a write off. But what about the subpostmasters that were owed money? That’s what that’s raising, isn’t it?
Andrew Winn: That’s right, yes.
Mr Beer: The answer to that isn’t really given in solution 3, is it?
Andrew Winn: I think that was probably one that we could do this but we’re never seriously going to do it.
Mr Beer: Who made the suggestion?
Andrew Winn: I don’t know.
Mr Beer: Who rejected it?
Andrew Winn: Well, I would have, along with other colleagues, I would imagine.
Mr Beer: Solutions 1 and 3, is this – I’m sorry.
Andrew Winn: I would suggest that the options were probably driven by Fujitsu’s suggestion for us to consider but that’s a suggestion.
Mr Beer: Solutions 1 and 3 keep subpostmasters in the dark, don’t they?
Andrew Winn: Yes.
Mr Beer: Both those subpostmasters who are known to be affected by the bug and the wider subpostmaster community?
Andrew Winn: Yes.
Mr Beer: Solution 2 was the recommended solution, if we read that, please:
“P&BA will journal values from the discrepancy account into the customer account and recover/refund via normal processes. This will need to be supported by an approved POL communication. Unlike the branch POLSAP remains in, balance albeit with an account discrepancies that should be cleared.
“Impact – POL will be required to explain the reason for a debt recovery/refund even though there is no discrepancy at the branch.
“Risk – could potentially highlight to branches that Horizon can lose data.”
The end of the first line/beginning of the second:
“This will need to be supported by an approved POL communication.”
Would the approved communication leave out some of the facts?
Andrew Winn: I’d probably need to see the communication to comment on that. I don’t think that would be the intention of that solution. It doesn’t suggest let’s do it. But I think it would be – what that would be saying is that it would probably need to be checked off at legal level before we send it. That’s my assumption.
Mr Beer: Was there a feeling at this meeting that the principle of the less said, the soonest mended should be applied?
Andrew Winn: No, I don’t think so. I think the less said would be the clean it up without telling the branches.
Mr Beer: The less that subpostmasters know, the better?
Andrew Winn: No. I think again it comes back to let’s let the subpostmasters that are impacted know what’s happened and what we’ve done to resolve an issue even if they potentially didn’t know they had an issue, but I don’t think it was ever suggested to do a branch-wide communication.
Mr Beer: What would POL senior management have said if the outcome of the receipts and payments mismatch meeting was a recommendation that the whole of the network should be warned of this potential problem?
Andrew Winn: I think they would have said no.
Mr Beer: Why is that?
Andrew Winn: Because they were keen to maintain the integrity of Horizon as far as possible, but –
Mr Beer: When you say, “They were keen to maintain the integrity” do you mean keen to maintain the appearance of integrity?
Andrew Winn: Yes. But again I’m assuming what the manager’s – his position might have been. I wasn’t at that kind of level so.
Mr Beer: You’re assuming it on the basis of some facts or information, though, aren’t you? You’re not just guessing?
Andrew Winn: No, I think there was a feeling (and I use that term carefully, a “feeling”) that when we got on to these issues calls, it was more the kind of ground roots, people in the NBSC, Network and myself in finance, who were, “This is what’s happened, let’s tell them, let’s tell them exactly what’s gone off and this is what we’ve done to resolve it”. Whereas there seemed to be more – possibly more strategic, dare I say, view from more senior managers of “should we really be telling everybody that this has happened?”
Now, that’s tarring every senior manager with a brush that’s unfair, but there was a feeling that there was a bit more resistance to communication kind of the higher things went, which I don’t know if there was a pressure that they were being given under or whether that was a more general way of just thinking rather than somebody else providing pressure. I don’t know.
Mr Beer: But looking at senior management collectively, and without identifying individuals, your clear understanding was that senior management expected or required you to stick to the message that Horizon had integrity?
Andrew Winn: Yes.
Mr Beer: Sir, I wonder whether we might take the lunch break early?
Sir Wyn Williams: Certainly.
Mr Beer: And come back at 1.45.
Sir Wyn Williams: Yes, that’s fine. See you then.
Mr Beer: Thank you very much, sir.
(12.49 pm)
(Luncheon Adjournment)
(1.45 pm)
Mr Beer: Good afternoon, sir. Can you see and hear me? Sir, you are still on mute. I think that was a yes.
Can you see and hear us, sir? (Pause)
Sir, can you see and hear us? Can we just pause while that’s sorted out. (Pause)
We’ll break for five minutes, if that’s possible, please, whilst we restore the link. So five minutes. Thank you.
(1.47 pm)
(A short break)
(1.50 pm)
Mr Beer: Thank you very much, Mr Winn.
Andrew Winn: Sorry to have delayed everyone.
Mr Beer: Can I just check, sir, that you can see and hear me?
Sir Wyn Williams: Well, I can hear you. I now can’t – now I can see you as well. So we’re fine.
Mr Beer: Fantastic, thank you.
Mr Winn, when I was asking questions before lunch about the date of the receipts and payments mismatch bug, the meeting about that, and one of the actions arising from it (namely, whose job it was, if anyone’s, to inform lawyers in ongoing or past prosecutions and civil proceedings about the bug), I mentioned a document that I gave the reference to but did not display. Can I display the document for you, please. It’s POL00055410.
If we look at the lower email first, please, you’ll see that it’s from Alan Simpson and it’s dated 8 October 2010 and Alan Simpson, we can see from his signature block, was the POL information security incident senior. You can see that it’s to Rob Wilson. Mr Wilson was the head of criminal law in the Criminal Law Division within Post Office at the time.
Did you know that –
Andrew Winn: No, I don’t think I came across him at all.
Mr Beer: It helps us on two things if we can just read it. It says:
“I am forwarding you the attachments above in relation to a series of incidents …”
If we just scroll up, please, we can see that under the attachments there are two that are mentioned, “ReceiptsPayments notes[version]5”. That appears to be a version of the notes that we looked at earlier concerning a meeting or meetings about the receipts and payments mismatch bug.
Then “Lost Discrepancies [29 September 2010]”, that appears to be Mr Jenkins’ document of 29 September 2010 that we looked at earlier.
Just going back down to the email:
“I am forwarding you the attachments above in relation to a series of incidents, identified by Fujitsu this week, whereby it appears that when posting discrepancies to the local suspense, these amounts simply disappear at branch level, and a balance is shown.
“The above includes Fujitsu’s initial analysis [that appears to be a reference to Mr Jenkins’ documents] and proposed solution/s, whilst the other documents the outputs from various meetings held this week.”
That appears to be a reference, would you agree, to the note of the meeting that we saw?
Andrew Winn: Yes.
Mr Beer: It would appear that the meetings, plural, were held in the week ending Friday, 8 October. Do you see that?
Andrew Winn: Yes, I can see that.
Mr Beer: “My concern is around the proposed solution/s, one or more of which may have repercussions in any future prosecution cases and on the integrity of the Horizon Online system.”
So Mr Simpson is referring the issue to the head of criminal law about future prosecutions based on data produced by Horizon Online. Do you remember whether there was any discussion about disclosure of the bug to those conducting current prosecutions, based on data produced by Legacy Horizon?
Andrew Winn: No, I wouldn’t be aware of that, no.
Mr Beer: Can you recall that being discussed in the meeting?
Andrew Winn: In this meeting?
Mr Beer: Yes, in the meeting that we saw the notes of this morning.
Andrew Winn: Not in detail but I would imagine it would be mentioned, yes.
Mr Beer: Was this seen as a bug that just afflicted Horizon Online?
Andrew Winn: As far as I can recall, yes.
Mr Beer: What exploration was there of whether this was a bug that afflicted Legacy Horizon?
Andrew Winn: I wasn’t aware of any.
Mr Beer: Thank you. That can come down.
Can we move forward, please, to December 2010 and look at POL00029718. Can we start, please, on page 8 of this collection and scroll down, please. This is an email from Emma Langfield to a number of people, including you and Mr Jenkins, dated 24 December 2010. Can you see that?
Andrew Winn: Yes.
Mr Beer: She’s asking for help there as there are some branches for whom values appear marooned in the P&BA discrepancy account. If you just scroll down. It says:
“Morning
“Please see below from [P&BA]. They have identified some branches where values appear marooned in the P&BA discrepancy account …”
Yes?
Andrew Winn: Yes.
Mr Beer: And that they:
“… either appear not to align to the values and last Fujitsu spreadsheet or have not been identified as part of this issue.”
Then there are a number of emails exchanged as part of this collection of emails, with ultimately a response from Mr Jenkins on page 1 of this collection. We see this is an email from him to you moving forward a number of years to May 2012. He says:
“Sorry it has taken me so long to get back to you.
“I’ve had a trawl back through my old emails and what I’ve found is the following …”
He sets out in his paragraphs 1, 2 and 3 what he’s discovered in relation to two branches that were in the original list, associated with the lost discrepancy issue that had occurred in 2010. Then in paragraph 4 he says:
“I can also see that although [a branch number] was in the original list, it seems to know dropped off very early. This is because it wasn’t in the original list from POL and also the precise symptoms in terms of [something] left behind were different. I can’t find any trace of any further investigation of this in my emails.
“Given that the discrepancy amounts in both cases was very large, I’d be surprised if they weren’t properly investigated at the time, but they don’t seem to be included in the lists relating to the ‘[something] Discrepancy’ investigation.
“If you want this followed up further I suggest you raise it with the POL Problem Management team (Emma Langfield), who can then get the Fujitsu Problem Management team to dig further.”
So this is correspondence under two years after the problem was discovered and there appears, would you agree, to have been some doubt as between Fujitsu and POL over which branches had been investigated as possibly impacted and which had not?
Andrew Winn: Yes.
Mr Beer: At the very least, no shared understanding of what had happened?
Andrew Winn: Yes.
Mr Beer: Who was responsible for investigating and understanding the extent of the problem and updating the branches affected?
Andrew Winn: In terms of updating the branches, I would probably suggest that that was P&BA. In terms of investigating, I would imagine that would be the IT team.
Mr Beer: But you have got in 2012 here Mr Jenkins saying that, according to the documents that he can access, there’s a branch with a very large discrepancy – two branches with a very large discrepancy and they don’t seem to be included in the list. Would you regard that as problematic?
Andrew Winn: Yes.
Mr Beer: Given the seriousness with which the mismatch bug was taken back in September/October 2010, we’ve seen the documents around that, may this suggest that the follow-up on the impact on branches was taken less seriously?
Andrew Winn: I can’t really say.
Mr Beer: Did the Post Office take seriously the need to understand the full impact of this significant issue and, in particular, the impact on individual branches and subpostmasters and ensure that all affected branches were identified and there was a proper investigation?
Andrew Winn: Yes, as far as I was aware.
Mr Beer: Can we turn, please, to POL00098016. This continues the thread of discussion on the two outstanding queries on the receipts and payments mismatch bug and we’re now ahead to April 2013. Can we go, please, to page 4 of the thread to begin it.
At the foot of the page, please, you will see an email from David Wright, who was I think an IT service adviser in service management, and it’s an email to Andy Dunks and Penny Thomas, Steve Bansal, Scott Somerside and to you?
Andrew Winn: Yes.
Mr Beer: It reads “Andy”, and I don’t think that’s you, that’s the Andy Dunks that that was addressed to; is that right?
Andrew Winn: I would think so, yes.
Mr Beer: “Andrew Winn (Relationship Manager, Financial Service Centre) has requested Service Management assistance in reopening Fujitsu investigations for two outstanding enquiries he has been dealing with for our branches.
“At our recent service review meeting held with Leighton Machin he suggested your names as the appropriate contacts. Gareth Jenkins has also been approached in the past.
“I have attached some information previously sharing via email but if you need more detail to help you resolve these incidents, please approach Andy Winn direct.”
Then they give your details.
Then if we go up the chain, please, to page 3 we’ll see that a chaser is sent on 22 February. Just scroll down, please. Keep going, please. Yes, that’s it, 22 February:
“Hi Leighton,
“I’ve just picked up a reply from an enquiry I made to Andy Winn in our Finance Service Centre (he has been on leave) and disappointingly he has not heard anything from the contracts you suggested last month.
“Can you escalate the enquiry on our behalf?”
So that’s essentially a chaser, would that be right, from somebody that worked to you, David Wright?
Andrew Winn: I think I remember David vaguely, I think he might have been working in NBSC at that time but I’m not sure he wasn’t in P&BA Or Finance Service Centre, as it was then.
Mr Beer: You can see his signature block there and he appears to be based in Dearne House in Barnsley?
Andrew Winn: Right, okay.
Mr Beer: Does that mean he was working for you or not?
Andrew Winn: No.
Mr Beer: Anyway it’s a chaser, and then if we go back to page 3, please, at the foot of the page, another chaser a month later on 5 March. It’s not month later, it’s the following month:
“Hi Leighton
“Were you able to escalate this on our behalf?”
Then a reply from Fujitsu:
“Apologies … It seems it was not cascaded at the point of your last request (I will cascade it now!)
“A response in one form or another will be provided by [close of play] Thursday.”
Then up to the bottom of page 2, please, Mr Wright:
“Hi Leighton
“Did a response get issued on Thursday? I have not received it – did it just go to Andy?”
Then up to the bottom of page 1, please, thank you. Steve Bansal to a collection of people including you:
“… accept my apologies thought as I thought I’d sent this mail some weeks ago.”
He answers a query about Paystation transactions. Do you see that in his third paragraph?
Andrew Winn: Yes.
Mr Beer: “So to answer the question very Paystation transactions”, which is something else, yes?
Andrew Winn: Yes.
Mr Beer: Then you reply that top of the page, if we go to the top of the page. On 16 April 2013, you say:
“Hi Steve
Yes it does …
“However I’m far more concerned about the outstanding enquiry relating to the receipts & payments problem back in 2010 where we have two branches [and you identify them] who appeared [to be] on the initial list of branches but not on the list of those resolved. We still have a large unexplained credit on one branch whilst we have recovered money from pay on the other despite them recording a significant surplus at the time.
“I need to be clear that there is no unresolved problem in case either branch appears in the ongoing Horizon integrity review.
“I know you have previously stated that a work plan was set up to do some further analysis on [one of the branches] but [the Post Office] requested this not to proceed but I need to understand why they disappeared from the resolved spreadsheet.”
So what you were doing, you were saying that there’s an issue that’s now two to three years old –
Andrew Winn: Yes.
Mr Beer: – and there are two branches that are maybe affected in different ways but I want a resolution?
Andrew Winn: Yes, I think from what I remember is that they were on an initial list, a kind of quite big list at the time, and Fujitsu said these have got an issue but they’re not the same as the receipts and payments, so we need to separate them, if I remember correctly.
Mr Beer: For one of them, you had made a deduction – the Post Office had made a deduction from the SPM’s remuneration?
Andrew Winn: Yes. Well, that’s what it reads there.
Mr Beer: You say:
“I need to be clear that there is no unresolved problem in case either branch appears in the ongoing Horizon integrity review.”
Why would you need to be clear that there’s no unresolved problem for that purpose, rather than because in one case the subpostmaster might owe money and in the other case the subpostmaster is owed money.
Andrew Winn: There would be no difference between the two. The issue needs to be resolved.
Mr Beer: But why was it important in case either branch appears in the ongoing integrity review? Wouldn’t it be a good enough reason that we’ve taken some money out of a subpostmaster’s wages, perhaps wrongfully, that I need an answer?
Andrew Winn: I would say any branch that we know has got a problem we need to resolve it as quickly as possible.
Mr Beer: But we’re two or three years on now?
Andrew Winn: Yes.
Mr Beer: Was it because the decks needed to be cleared because you didn’t want cases like this showing up in the integrity review, as you call it?
Andrew Winn: I don’t think it was anything to do with debts, I think it was to do with branches like that shouldn’t have issues that need resolving three weeks old, never mind three years.
Mr Beer: So the bug was a significant issue for the Post Office?
Andrew Winn: Yes.
Mr Beer: It was covered in 2010 at some point. Was it causing you some considerable concern, even aside from the integrity review, that by 2013 you still didn’t have a clear picture as to who was affected and how?
Andrew Winn: No, because my belief was that if there was a discrepancy – if the receipts and payments impacted a branch, then the balancing amount would land in our discrepancy account and we’d got the discrepancy account cleared down so that we could see any figure that was – because normally in a discrepancy account you would do your branch trading, money would go into the discrepancy account, the subpostmaster would pay it or we’d repay the postmaster, and that amount would clear down to zero. So basically the account, at that kind of time, in 2013, was very clean.
So if any branch had a receipts and payments mismatch, we would see it eventually, not in a particularly timely manner, but we start seeing figures rolling forward.
Mr Beer: Can I turn to a different issue. That document can come down, please.
Can we look – and this is to do with remote access – at POL00023432. Can we look at the second page first, please. Just to explain, the reason I’m asking you these questions is because they concern your engagement with subpostmasters and your knowledge about remote access. So the chain starts, and it’s only a two email chain, from Mr Lusher – can you see that at the foot of the page – who is a contracts adviser in the network support team in Norwich.
Andrew Winn: Yes.
Mr Beer: Just explain shortly what that role involved.
Andrew Winn: They are responsible for making sure that the contracts were applied to subpostmasters and for – by the subpostmasters. They would be, if I remember right, responsible for signing off any debt recoveries, major debt recoveries. I never really saw a job description. But contracts managers, we were regularly – they would be involved in becoming aware of debt and so there would be quite a lot of conversation, communication between contract managers and myself, some more than others.
Mr Beer: If we go to the top of the page, we can see it’s an email from Mr Lusher to you of 15 October 2008, and he says:
“Hi Andrew
“I spoke to you a few days ago about a suspension at Rivenhall. From our conversation, I believe that you had a good understanding of the problem and I would be grateful for further guidance. Rivenhall is a one position rural branch – the only abnormal product being an ATM.
“I have attached notes of the interview should you want to refer to them although they are rather long. There are two issues which the suspended subpostmaster, Graham Ward, raised …”
So just some context here, stopping at that point, Graham Ward was a subpostmaster at that branch, the Rivenhall branch, whose contract was terminated, we know whose appeal against contract termination was dismissed, who became one of the 555 Claimants in the Group Litigation and who is a Core Participant in this Inquiry. His evidence was read to the Chairman in the course of the Human Impact hearings last year.
In short, he’s a subpostmaster who lost his job, whose marriage broke down and who was left in debt with four young children.
Now, Mr Lusher encloses the interview transcript and gives a summary. Let’s read the summary together:
“1. He claims that on a number of occasions figures have appeared in the cheques line of his account. He suspects these have been input into his account electronically without his knowledge or consent. He is certain that he has cleared and remmed out cheques in the correct way and tells me that cheques must be properly cleared on the system to progress to a new account.”
Just stopping there, can you explain what you understand from what is being described there from Mr Ward’s account?
Andrew Winn: What he’s saying is that he’s seeing cheques appear on his cheque line that he doesn’t believe he has taken and input to the Horizon System.
Mr Beer: What does he seeing checks appearing on his cheque line mean?
Andrew Winn: You can pull up a cheque holdings at any point which will show the values that have been input there and that will be what he’s seeing.
Mr Beer: He is saying that he has cleared and remmed the cheques. What does he mean by that?
Andrew Winn: So at end of day you would – well, not necessarily at the end of the day, when the postman arrives for, say, 4.00, just before then you will look to see what cheques you’ve got, check them against your cheque line, make any adjustments if need be because people will make mistakes of pressing cash, like they do every time. So you need to introduce a cheque there. Rem out the cheques. I can’t remember the order of it but basically you rem out the cheques which means you’re dispatching them –
Mr Beer: You handed the cheque over?
Andrew Winn: That’s what you’re telling the system. That wouldn’t – you’d do that at the same time but not precisely the same time and you would need to clear the cheque line, cheque account.
Mr Beer: Ie I’ve received a cheque from somebody I’ve now passed the cheque on?
Andrew Winn: That’s right.
Mr Beer: In physically pass the cheque on?
Andrew Winn: Yes, that’s what he’s telling Horizon. There is a bit and I’m afraid I can’t describe it properly but there is two elements to the process. One is the remming out of the cheques and the other one is to clear the cheque line. It sounds – now it sounds – there was some logic to it but, at the moment, I can’t explain why rem out isn’t there. But it did cause problems in terms of if you get distracted and you forget you are doing it and that can result in cheques appearing when you don’t expect them to be.
Mr Beer: What he is saying is that he’s certain he has cleared and remmed out the cheques and yet there are some cheques appearing on his cheque line which suggests that he is still sitting on cheques?
Andrew Winn: Yes.
Mr Beer: He paragraph 2:
“He has made good about £10,000 and not made good about £11,000 of the shortages which arise from these figures. He claims that because of the abnormal nature of these entries, the shortages have not just rolled over from one branch trading statement to the next, but have accumulated – each being added to the last (eg if the account in period one showed a shortage of £100 which was not made good, then the shortage shown in period 2 would be £200).”
Then just scroll down a little bit:
“The subpostmaster’s contract remain suspended. I would be very grateful for your expert comment and advice.”
You would agree that the account summarised by Mr Lusher of what Mr Ward was saying was a clear and coherent one?
Andrew Winn: It appears it, yes.
Mr Beer: He’s saying he has put £10,000 of his own money in already but he’s not made good another £11,000 of loss –
Andrew Winn: Yes.
Mr Beer: – and he’s saying that the system’s magnifying the loss, even without him doing anything by simply moving from one trading period to the next.
Andrew Winn: Well, more than that. It’s doubling up every trading period.
Mr Beer: Yes it’s magnifying, doubling up. He uses the £100 to £200 example.
Andrew Winn: Yes.
Mr Beer: You would agree, I think, that this is a very serious issue being raised?
Andrew Winn: Yes, if it’s backed up, yes.
Mr Beer: It’s serious for the subpostmaster, would you agree, because on his account he’s lost £10,000 of his own money, yes?
Andrew Winn: Yes. Oh, yes.
Mr Beer: The system is showing that he owes a further £11,000. It’s serious for that reason, would you agree?
Andrew Winn: Yes.
Mr Beer: It’s serious for him because he’s been suspended?
Andrew Winn: Yes.
Mr Beer: And it’s serious for him because his contract might be terminated and he would lose his job and his livelihood?
Andrew Winn: Yes.
Mr Beer: It’s serious, would you agree, for the Post Office because if a subpostmaster’s saying that the system that’s used across the country has introduced a phantom sum into his cheque line of account, that’s very serious?
Andrew Winn: Yes.
Mr Beer: If it’s correct, it’s a serious issue with the Horizon System?
Andrew Winn: Yes.
Mr Beer: Can we turn to your response, please, page 1. You respond on 23 October, so his email was the 15th, this is 23 October 2008, and you say:
“1. The only way POL can impact branch accounts remotely is via the transaction correction process.”
Reading a couple of sentences on – sorry I will read the next sentence:
“These have to be seared by the branch in the same way that in/out remittances are I guess. If we were able to do this, the integrity of the system would be flawed. Fujitsu have the ability to impact branch records via the message store but have extremely rigorous procedures in place to prevent adjustments being made without prior authorisation – within POL and Fujitsu.”
Now, in your witness statement, you say that you were aware that Fujitsu had what you describe as remote access and this is an email suggesting you knew at least by 2008, yes?
Andrew Winn: Yes.
Mr Beer: And –
Andrew Winn: I would say that my response there was a repetition of a – I would imagine from this point I went to someone in Fujitsu or our IT side, asked them that question that Alan had posed and repeated it back there. I don’t think that kind of would have come just from me.
Mr Beer: Well, I was going to suggest this first part of paragraph 1 suggests some familiarity by you with Fujitsu’s access controls, doesn’t it?
Andrew Winn: As a result of me – Alan posed the question “Can we get – can there be remote access” and I will have asked the question based on that question.
Mr Beer: How had you satisfied yourself as to the security of the Fujitsu access controls?
Andrew Winn: I don’t think I did.
Mr Beer: How could you be satisfied that they were being monitored effectively by Fujitsu?
Andrew Winn: I couldn’t.
Mr Beer: We know from some evidence that a man called Richard Roll gave in the Group Litigation and indeed from documents disclosed by Fujitsu to us in this Inquiry, that Fujitsu’s third line of support were routinely using their ability to go into the system in a way that was described by them as “off piste”. Did you know about that?
Andrew Winn: No.
Mr Beer: That wasn’t in accordance with the regularisation controls and protocols that were in place?
Andrew Winn: Yes, I wasn’t aware of that.
Mr Beer: So what evidence did you have to believe that Fujitsu were following protocol?
Andrew Winn: Just what I’d been told.
Mr Beer: Who had told you that?
Andrew Winn: I don’t know.
Mr Beer: But you had been told something by the subpostmaster?
Andrew Winn: Yes.
Mr Beer: Why did you accept what Fujitsu were telling you rather than what the subpostmaster was telling you?
Andrew Winn: I’m not sure if I did. I’m not sure what research I did, investigation into the branch.
Mr Beer: Anyway, you continue:
“These controls form the core of our court defence if we get to that stage. He makes a casual accusation that is extremely serious to the business. As usual he should either produce the evidence for this or withdraw the accusation.”
By saying these controls form the core of our court defence, you were indicating to Mr Lusher, weren’t you, that it was okay to say to Mr Graham Ward that the Post Office was prepared to go all the way here “We’re willing to stand up in court and defend Horizon and its operation by Fujitsu”, weren’t you?
Andrew Winn: Yes, I think so, yes.
Mr Beer: You say that –
Andrew Winn: I’m saying that from – without any legal background to give me the authority to say that.
Mr Beer: You say that Mr Ward made a “casual accusation”. In what respect was it casual?
Andrew Winn: He didn’t present any evidence to support his claim.
Mr Beer: Casual means free and easy, doesn’t it?
Andrew Winn: Bad choice of words, sorry.
Mr Beer: Sorry?
Andrew Winn: A bad choice of wording at the time.
Mr Beer: Is that what this is? I mean, “casual” means without thought, free and easy, something just falls from the lips without any real thought being given to it. How did you know it was a casual observation, a casual accusation?
Andrew Winn: As I say, I cannot recall what investigation I carried out at the time.
Mr Beer: Whether you carried out an investigation or not doesn’t tell us why you suggested to Mr Lusher that this subpostmaster was making a casual accusation, does it?
Andrew Winn: I’ve already acknowledged that the term “casual” wasn’t well used.
Mr Beer: You didn’t know whether it was a casual accusation or not, did you?
Andrew Winn: I’m sorry, I feel as though we’re going round in circles. I would have carried out an investigation. I assume I would have carried out an investigation.
Mr Beer: There’s no reference to an investigation in this email by you.
Andrew Winn: Okay.
Mr Beer: Is there?
Andrew Winn: I don’t know. I can’t see the full email.
Mr Beer: Have a look at the full email. (Pause)
Andrew Winn: Can we move it down a little bit?
Mr Beer: Yes, yes, please do.
There’s no reference in that to you carrying out an investigation at all, is there?
Andrew Winn: That’s correct.
Mr Beer: So I’ll ask again: how did you know that this was a casual accusation by Mr Ward?
Andrew Winn: Reading that, I should have done an investigation.
Mr Beer: You say that that accusation, if we just go back up to paragraph 1, and in the second part of paragraph 1, starting “These controls”, in the second sentence:
“He makes a casual accusation that is extremely serious to the business.”
Can you see that?
Andrew Winn: Yes.
Mr Beer: Why was it extremely serious to the business?
Andrew Winn: Because if the accusation was correct, then that would support the lack of integrity into the system.
Mr Beer: Why was the seriousness of the business seemingly your principal concern, rather than the merits of the issue that had been raised by this subpostmaster?
Andrew Winn: I think the nature of the business at – the state of play within the business at that time, with the concern about Horizon integrity.
Mr Beer: This is 2008. This is before the Computer Weekly article had broken.
Andrew Winn: Okay.
Mr Beer: Was the integrity of the business your principal concern rather than the actual merits of an issue that had been raised by a subpostmaster?
Andrew Winn: No, I would say exactly the opposite and I’m not happy about this.
Mr Beer: You’re not happy about?
Andrew Winn: About my actions on this one –
Mr Beer: You continue –
Andrew Winn: – or lack of action.
Mr Beer: You continue:
“As usual he should either produce the evidence for this or withdraw the accusation.”
This was a subpostmaster saying that the system was introducing phantom figures information his cheque line.
Andrew Winn: Yes.
Mr Beer: How could he possibly produce evidence of that?
Andrew Winn: You can print the cheque listing reports. It can be printed out. So he could have done before and afters with the remming out of cheques. So if he remmed out £100 of cheques and then it reappears again, then those receipts –
Mr Beer: So why didn’t you say that? “Just show us these”. You are being very combative here, aren’t you? “He makes a casual accusation” –
Andrew Winn: Yes.
Mr Beer: – “that is extremely serious to the business … he should either produce evidence or withdraw it.”
Andrew Winn: Yes, I would agree that this is not the kind of letter I would typically write.
Mr Beer: You seem concerned that he was slandering the business, don’t you?
Andrew Winn: Not really.
Mr Beer: Produce evidence or withdraw that accusation, is what you’re saying here, aren’t you?
Andrew Winn: Yes.
Mr Beer: Would you put the reputation of the business ahead of investigating the merits of what was being said by a subpostmaster?
Andrew Winn: No, that would not be the way I would like to think I approached a job.
Mr Beer: Who had determined that it was for subpostmasters to prove that what they were saying was true, rather than raising an issue and it being investigated by P&BA?
Andrew Winn: The –
Mr Beer: Ie before we’ll start an investigation, you’ve got to prove what you say is true, otherwise we won’t start an investigation.
Andrew Winn: No. No, I think – I think this probably suggests that this was done in the very early days of me taking on the role. Would that tie up with dates? I would say that I got significantly better it my job as I gained experience on doing it and I’m not happy looking at this.
Mr Beer: You commenced that sentence:
“As usual he should …”
That suggests a stock line, doesn’t it?
Andrew Winn: Well, so the bulk of what you are looking at is disputing transaction corrections, that was at that time probably the bulk of my work. So if a subpostmaster said “This transaction correction isn’t correct”, we’d ask them why they’re saying that. So “we provided the evidence for the transaction correction, what are you producing to suggest it’s not correct?” I think I would have – if this would have been three or four years later, I think I would have dealt with it in a completely different way.
Mr Beer: In paragraph 2, you say:
“What ‘the abnormal nature of these entries’ means, I assume no-one knows.”
He was saying that figures appeared in his cheque line of account without his knowledge or consent. That’s pretty abnormal, isn’t it?
Andrew Winn: Mm-hm.
Mr Beer: Why were you saying “What” – quoting back to him, to Mr Lusher – “‘the abnormal nature of these entries’ means, I assume no-one knows”, when it was perfectly obvious that they were abnormal, they were phantom figures appearing in the cheque line? Why were you taking that point when it was perfectly clear what this subpostmaster was saying?
Andrew Winn: I can’t respond to that, sorry.
Mr Beer: You continue:
“The implication is that he acknowledges that when he ‘made good’ at branch trading he did not and falsified his branch trading statement and rolled the loss forward.”
You were being told by Mr Lusher that the subpostmaster had put £10,000 of his own money in, weren’t you, in the previous email?
Andrew Winn: Yes.
Mr Beer: And you here appear to be criticising the subpostmaster for doing so, saying that he’s guilty of falsification, aren’t you? When he made good, ie that’s putting the £10,000 in –
Andrew Winn: I think what I was saying is he’s told Horizon that he’s put £10,000 in but then immediately the – so he does his cash declaration at branch trading, £10,000 short, he makes good cash 10,000, so it balances. The next day the £10,000 shortage reappears again, which would suggest he’s not physically put the £10,000 into the till.
Mr Beer: That’s not what you are saying there at all.
Andrew Winn: Isn’t it?
Mr Beer: You are saying that the system is showing a £20,000 debt on the cheque line, he acknowledges some of that debt when he’s made good at branch, by putting £10,000 in, and he’s, therefore, falsified his branch trading statements, aren’t you?
Andrew Winn: No, I don’t think so. I mean, we’re not talking about the cheque line anyway here. We’re talking about the cash position. So what I’m saying is that the derived cash position was £20,000 that should be cash. He only physically had £10,000 in the till. He’s then told – he’s then increased his cash declaration to 20,000. The system is rolled forward in a balanced state but because the £10,000 didn’t physically go into the till, when he did his next cash declaration the £10,000 shortage was still there again.
Mr Beer: Moving on at the foot of the page, two paragraphs up, you say:
“If that does not satisfy him he would need to establish that his trial balance actually balances. If it does (and it will) he would need to demonstrate where balancing [£10,000] element of the loss is.
“These are all things for him to prove. If he can support any of his allegations we will investigate – and be extremely worried whilst doing so.”
In the course of these paragraphs you say, I think four times, that it’s down to the subpostmaster to prove; for the subpostmaster to establish; for the subpostmaster to demonstrate; for the subpostmaster to support what he’s saying.
Was the assumption by you that, if there was no obvious answer or cause for a discrepancy, it was assumed to be the fault of the subpostmaster unless they proved otherwise?
Andrew Winn: I think that’s probably a fair comment.
Mr Beer: And, therefore, you are applying an approach of assuming that the subpostmaster’s guilty until he proves he’s innocent?
Andrew Winn: I don’t – “guilty” doesn’t sound a –
Mr Beer: It’s not a nice word, is it?
Andrew Winn: No, no.
Mr Beer: But it’s accurate.
Andrew Winn: No, I don’t think it is.
Mr Beer: You assume that –
Andrew Winn: I believed that the loss was proper to the branch. I think guiltiness is not something that I was – would be thinking.
Mr Beer: You assume that he’s culpable for wrongdoing –
Andrew Winn: I believed he was –
Mr Beer: – unless he prove otherwise?
Andrew Winn: – culpable for the loss, yes, unless he can give me any indication he was not.
Mr Beer: Where he was saying that the problem is due to an error in Horizon, its programming, a bug or some such in the code or in the data, how was he to prove that?
Andrew Winn: I guess I can’t answer that.
Mr Beer: We’ve heard evidence from many subpostmasters that once they were suspended they were locked out of their offices. You knew that, didn’t you?
Andrew Winn: Probably. It wasn’t –
Mr Beer: This subpostmaster was suspended. If he was locked out of his office, how would he prove it?
Andrew Winn: Yes, fair point.
Mr Beer: What is the answer to it?
Andrew Winn: I don’t know.
Mr Beer: Is this email a reflection of the fact that it was important for you to maintain the position that Horizon was infallible, was foolproof, and that if he could support any of his allegations that would unfortunately undermine that position?
Andrew Winn: I think it’s a reflection of me being new to a role and not doing it very well at that point and, as I said, if I was doing it a couple of years later I would have approached it a lot better. So I think I was probably following the company line to some extent at that point.
Mr Beer: You say, at the foot of the page, in the last line:
“If he can support any of his allegations we will investigating – and be extremely worried whilst doing so.”
Is that because it would undermine the infallibility of Horizon, the line that the Post Office took?
Andrew Winn: Yes, I think if cheques were – phantom cheques were appearing on a branch’s account, then I would be extremely worried and I’ve kind of agonised over that concept for quite a while and I still cannot understand how that would happen. But if it does, it does.
Mr Beer: Wouldn’t it cause an extreme worry that a man suspended from his job, about to lose his livelihood, career and marriage was, in fact, innocent of any wrongdoing?
Andrew Winn: Yes.
Mr Beer: Was the attitude of mind that we see displayed in this email reflective of a culture in which you worked and at this time?
Andrew Winn: Possibly.
Mr Beer: Put it another way: if others in your department had been asked to respond to Mr Lusher’s email, would they have responded broadly as you did on the key issues or are you the outlier that took a particularly hard line on subpostmasters?
Andrew Winn: No, I think the former.
Mr Beer: The former?
Andrew Winn: Yes.
Mr Beer: Can I turn to a related email exchange concerning Mr Ward, POL00002268. Just forgive me whilst I catch up with my hard copy papers.
Can we turn to page 2, please. Just scroll down just so we can see the signature block, please. This is an email dated 1 February 2010 from Hayley Fowell, external relations manager at the Royal Mail Group, and it concerns Mr Ward, the man we have been looking at. It’s to David Smith, Michele Graves and Dave Hulbert:
“Hi all,
“We’ve had a media inquiry from Retail Newsagent magazine; they have been talking to a subpostmaster who has said that his branch was closed in [September] 2008 because of financial irregularities which he claims are the fault or Horizon.
“I am providing our stock line which states the system is robust but in case we get more questions on this please can you advise if you have any record of an investigation for this individual and any relevant details – Graham Ward, Rivenhall Oak Stores & Post Office in Witham.”
Can we go back over the page to page 1, please – the bottom half of the page, please. Michele Graves, the Executive Correspondence Manager for the Executive Correspondence Team, replies:
“I have been corresponding with Mr Ward for a while. You may recall he is on the spreadsheet I pulled together. I’ll send over what I have. The issue is basically the same – Horizon is at fault – and he has focused on some cheques despatched from his branch that [I think that’s] then showed up on his cheque line. The termination went to appeal and the decision to terminate was upheld. Mr Ward’s MP is Brooks Newmark who I believe has raised Parliamentary questions re Horizon integrity.
“Andy [I think that’s you now copied into this chain], you also asked me for an update on Mr Ward recently – if you have any new info, can you please let Hayley know.”
Then at the top of the page, you reply:
“Hi Hayley, we’re due to restart our former agent debt recovery process. I just wanted to check the recent communications to ensure there was nothing there to suggest we should not do these. Let me know if we should not pursue at this stage.”
In this email chain, there’s a reference to the “stock line”, the stock line which states that Horizon is robust. Is that something that you were aware of?
Andrew Winn: Yes.
Mr Beer: Would you agree that a stock line is a standard response, a hackneyed response –
Andrew Winn: Yes.
Mr Beer: – to which no real thought or attention has been given?
Andrew Winn: No.
Mr Beer: Does that not reflect then your understanding of the use of the phrase “stock line”?
Andrew Winn: I would say my understanding is that it’s used by everybody, but I don’t think it undermines the integrity behind that belief.
Mr Beer: You say you wanted to:
“… check recent communications don’t suggest that we shouldn’t restart the debt recovery process.”
Why would the recent communications affect the restarting of the debt recovery process?
Andrew Winn: It depends what the communications have been and whether the business approach had been changed.
Mr Beer: This was now some time on from your response in 2008 that there would be no further investigation by the Post Office, unless Mr Ward could prove what he was saying, and asserting your confidence in Fujitsu’s security procedures.
Were you concerned at all that Mr Ward was continuing to protest his innocence and to seek answers about why there had been unexplained entries in his accounts?
Andrew Winn: I was conscious that he was obviously still protesting his innocence, but I wasn’t doing anything about it because there was nothing I could do about it.
Mr Beer: Why was there nothing you could do about it?
Andrew Winn: Because there would be no access to Horizon information at that point.
Mr Beer: Why?
Andrew Winn: You’d have to ask Fujitsu and our IT department that. You can only – you could only go back so far.
Mr Beer: How at that stage could Mr Ward establish his innocence?
Andrew Winn: I can’t say.
Mr Beer: Yes, thank you very much, Mr Winn. Those are the only questions I ask at the moment.
I wonder whether we might take a 15-minute break and then the questions for the Core Participants can commence at 3 pm?
Sir Wyn Williams: Yes, that’s fine.
Mr Beer: Thank you very much, sir.
(2.47 pm)
(A short break)
(3.00 pm)
Mr Beer: Sir, thank you. I think Mr Stein is going to ask questions first.
Questioned by Mr Stein
Mr Stein: Good afternoon, Mr Winn, my name is Sam Stein. I represent a large number of subpostmasters and mistresses.
I am going to take you back, first of all, to the mismatch document, which I hope I’ve got the correct reference number to, which is POL00028838.
Can we go to page 3, please. Mr Winn, you were taken to this document earlier on by Mr Beer who highlighted with you the various solutions that were discussed on this page. Frankie, can you highlight Solution One, please, in yellow – thank you.
Now, let’s just remind ourselves of what Solution One referred to, Mr Winn. Solution one was:
“Alter the Horizon Branch figure at the counter to show the discrepancy. Fujitsu would have to manually write an entry value to the local branch account.”
Then it says:
“IMPACT – when the branch comes to complete next Trading Period they would have a discrepancy, which they would have to bring to account.”
Now under “RISK”:
“This has significant data integrity concerns and could lead to questions of ‘tampering’ with the branch system and could generate questions around how the discrepancy was caused. This solution could have moral implications of Post Office changing branch data without informing the branch.”
Now, you’ve just been asked before the short break, a few minutes ago, a number of questions about Rivenhall Post Office and about questions that related to interference with the data that was being seen by the postmaster there, Mr Ward, yes? At this particular meeting you are being told it is possible to tamper with branch office accounts and, indeed, the conclusion is, as regards that, that that could have moral implications of Post Office changing branch data without informing the branch.
Was this the first time that you’d learnt that Fujitsu could alter branch accounts?
Andrew Winn: I don’t know.
Mr Stein: Well, it’s a fairly significant issue to have come up in this particular meeting, bearing in mind you are dealing with Rivenhall; do you agree?
Andrew Winn: Yes.
Mr Stein: As an example did you say to yourself that we’d better do something about Rivenhall because this appears to be show that Mr Ward there may be right, that data there could be interfered with without his knowledge?
Andrew Winn: I think there’s a difference between entering data to resolve a problem rather than what implication was for Mr Ward of somebody introducing cheques for no apparent reason.
Mr Stein: I see. Did you, in any way, investigate the issue that comes from this when you thought about Mr Ward’s situation?
Andrew Winn: No.
Mr Stein: You understand the link, don’t you?
Andrew Winn: I do now, yes. I probably wouldn’t have at that point.
Mr Stein: I will take you to another document, please, which is POL00055410. Thank you. Further down the page please, Frankie, and this is a document you were shown just after the luncheon adjournment today, the one that is dated 8 October 2010 and Mr Beer took you to this and dealt with particular points.
Can I take you to the second paragraph. So this is about the mismatch meeting notes:
“The above includes Fujitsu’s initial analysis and proposed solution/s.”
Can we scroll up to the above bit please, Frankie.
So the attachments in relation to this email are the “ReceiptsPayments notes[version]5”, which we believe are the mismatch notes that you have been asked a number of questions about and then, after the colon, highlighted there, “Lost Discrepancies”, that’s a document drafted by Mr Jenkins, okay.
Let’s go back to what we’re seeing in the middle of this email. You said in your evidence today to the Chair of this Inquiry that you thought that Fujitsu had proposed the three solutions in the mismatch meeting notes?
Andrew Winn: Yes.
Mr Stein: Now, this particular email from Alan Simpson who attended that meeting is saying the above – the above attachments – includes Fujitsu’s initial analysis and proposed solutions. Does that help you in your recollection that it was Fujitsu that had put forward the solutions that you have been taken to today?
Andrew Winn: It supports what we were saying, yes.
Mr Stein: Now, as regards those solutions, did anybody consider at the meetings in relation to the mismatch bug and software error, did anybody consider the legal implications of keeping information back from subpostmasters being prosecuted?
Andrew Winn: I can’t remember that being discussed, no.
Mr Stein: As an example, did anybody suggest that might be a criminal offence of keeping that information back?
Andrew Winn: No.
Mr Stein: No?
Andrew Winn: No.
Mr Stein: Lastly in relation to this, the email is referring to:
“The above includes Fujitsu’s initial analysis and proposed solution/s, whilst the other documents the outputs from various meetings held this week.”
Let’s take that apart into two bits. So it looks as though, regarding the mismatch discussions, that there were various meetings. Is that your recollection?
Andrew Winn: I don’t remember attending various meetings – it may have been phone conferences rather than meetings.
Mr Stein: Would it be normal for these meetings to have notes taken?
Andrew Winn: Well, somebody would have been responsible, yes, I believe so, yes.
Mr Stein: Presumably, the end result of such meetings would be notes and minutes?
Andrew Winn: Yes.
Mr Stein: I am going to take you, please, to then a different document, which is POL00029224, page 5.
If you can scroll down the page. This document, I am afraid, is difficult. We might be able to improve it by expanding it slightly, Frankie. Go further down the page, please, and again a little bit further down again. You will see at the bottom of that page, if we can highlight from “Information: Ki Barnes has called in”, can we zero on that and expand and highlight.
Now, this particular document, as you can see, it’s 2001, it looks like, and the reference is to phantom transactions, okay?
Andrew Winn: Mm-hm.
Mr Stein: So I’ll read what we have here. This is a record of this matter being dealt with as a PEAK, which is part of the system being used by Fujitsu:
“I am unsure as to what to do with this call now. Romec have been to site and state that they have actually seen the phantom transactions, so it is not just the PM’s word now. They have fitted suppressors to the kit but the PM is still having problems. As yet there has been no re-occurrence to the phantom transactions but there still may be problems.”
Let’s unpick what this all means. Romec the Royal Mail or Post Office engineers; is that correct?
Andrew Winn: Yes.
Mr Stein: This is referring to phantom transactions coming up on the screen; do you agree?
Andrew Winn: Yes.
Mr Stein: Phantom transactions in relation to the date of this particular entry seems to have gone back to quite early days of Legacy Horizon; do you agree?
Andrew Winn: Yes.
Mr Stein: Were you aware of phantom transactions as being a problem?
Andrew Winn: No.
Mr Stein: Do you remember speaking to Colleen Ingham, we think, in about 2004 about phantom or ghost transactions?
Andrew Winn: No, sorry, no.
Mr Stein: So help us a little bit further in relation to this. Bearing in mind that we’re talking about phantom or ghost transactions appearing on a subpostmaster’s screen, was this some information that would have been useful for your consideration in handling postmaster issues?
Andrew Winn: Yes, I think it would be fair to say that when I left the business I didn’t accept the concept of phantom transactions and that was in 2016.
Mr Stein: Right. Now, in your evidence earlier on, when you were discussing matters with Mr Beer, you appeared to be familiar with the concept of phantom transactions being raised. You weren’t saying to him “I don’t know what you’re talking about, Mr Beer”. So when did you start to become aware of the topic of phantom transactions?
Andrew Winn: I would have said it would be when Second Sight started their investigation but that kind of feels a bit too late, to be honest. So I probably heard the term before but I struggled with the concept of it.
Mr Stein: Ms Ingham, who ran the Cockfield Post Office in County Durham recalls you speaking to her about ghost transactions in around 2004?
Andrew Winn: 2004, okay.
Mr Stein: Thinking back, do you think that’s possible?
Andrew Winn: I’m trying to think what role I’d be in in 2004. I guess that was in problem management, was it?
Mr Stein: Very early days of that, we think.
Andrew Winn: No, I don’t know if I was in problem management then or in the Transaction Improvement Team, Network Improvement Team.
Sir Wyn Williams: If it helps, I think early on in your evidence you were referred to paragraph 1 of your witness statement where you suggested that you’d gone to the problem management team in 2005 but, in fact, there was a document which showed that you were there in 2003.
Andrew Winn: Okay.
Mr Stein: So it seems likely this had been once you moved into the problem management role. Do you think then, looking back and looking at the collection of information you now have, that there was an awareness within the Post Office of phantom transactions?
A client of mine recalls speaking to you about ghost transactions, that at that time in 2004 you probably did know something about the topic of phantom transactions.
Andrew Winn: Obviously, it had been raised. I wouldn’t argue with your client’s recollections. I don’t recall it myself.
Mr Stein: Now I will ask you about a different document, please, POL00029719.
Now, this is an email, Mr Winn, from Rod Ismay. You can see the date of this. This is 3 July 2013, sent to you and the subject matter is “Branches affected by Receipts Payments and Discrepancies issue”, okay. So it’s a follow-on from such matters. If you go further down, you can see that it says:
“1 of 2 emails re the 2 branches on the other list.
“Thanks, Rod.”
So you are being referred then to the further information.
If we just go down, we’ll see therefore the connected email, which says from Pete Newsome to Rod Ismay:
“Rod
“Looks like this branch had a different problem so was removed from the original list. The email below explains what happened and how we advised Post Office on the situation. Have an answer on second branch as well, will send that through shortly.”
Again, if we can go further down the page then to the second email, headed from Mark Wright. Then that one from Mark Wright to Pete Newsome, Gareth Jenkins, involving also Steve Parker, Steve Bansal and John Simpkins:
“We’ve unearthed the following email …”
Then there’s some figures given “122946”, and then go to the second page, please, Frankie. It’s probably easiest to look at the second paragraph:
“Branch 122946 rolled from TP4 to TP5 on 11 August. They accepted a gain of £34,330.88 which was settled centrally. The BTS shows a trading position of £22,021.65.
“The branch was included in investigations into receipts payments problems at the beginning of October, and was found to be a different problem from the others also under investigation at the time, so was not included in the later spreadsheets sent to POL.”
Mr Winn, you have been asked various questions about what happened in the mismatch payments issues and about the different solutions that were proposed to the mismatch difficulties and problems. You’ve been asked questions about whether the postmasters should have been informed that “Here’s a problem, here’s a bug”, that can affect their accounts.
What happened regarding this? Did you tell postmaster branches that “There’s another problem, another bug, that can affect your figures?”
Andrew Winn: I cannot remember what the outcome from these was.
Mr Stein: Do you recall whether there were similar discussions to perhaps keep this buried and not tip off the subpostmasters?
Andrew Winn: No, I’m sorry. Do we know what the name of the branches are, see if that rings any bells with me.
Mr Stein: I’m not sure I do. If I have the information I will come back to you.
Andrew Winn: Okay.
Mr Stein: Next document, please, POL00004694, pages 1 and 2. I’m grateful.
So this email dated 10 May 2010 from Nigel Allen to you, and you will see the reference to “Barkham” and then the number appropriate for Barkham “Outstanding losses”. This is from Nigel Allen, who was he?
Andrew Winn: I believe he was a contract manager.
Mr Stein: Right. To you:
“Andy
“What is this [£25,000] of returned cash on 5 Jan?
“Has this been properly recorded on the Horizon System?
“Was it received back at the Cash Centre?”
If we go down to the second page, please, I think a little bit further down, please, thank you. Now, this particular branch concerns Pam Stubbs. Does that name ring a bell?
Andrew Winn: It doesn’t, I’m afraid.
Mr Stein: Ms Stubbs was blamed for a shortfall of £28,000. She lost her business and was pursued by the Post Office for settlement of eventually £36,000. She had been told by Fujitsu engineers that the movement of terminals, the putting in and taking out of terminals, without proper safeguards could cause faults. Were you aware of that?
Andrew Winn: That doesn’t ring any bells. No.
Mr Stein: If we can go further down to the bottom of this particular email and then the starting point for the reference that I’m about to make is “If there is anything specific to investigate”. Thank you.
So the email is saying:
“We are aware of the problems at the branch. The letter suggests she has done all the checks we (and NBSC) would have suggested. There is not a lot of value I can add as there is nothing recorded that would account for the different losses.
“There are no transaction corrections that account for the losses or that should have compensated for them. She has already checked her transactional records and can see nothing so it is unlikely that customers are suddenly going to start alerting us to strange deposits into their accounts.
“What may or may not be interesting is a [£1,000 transaction correction] issued recently for a cash shortage in a rem to the cash centre. One would have thought that with the issues involved that a mistake like this would not have been made that realising. It is possible they did realise once the rem had gone but smacks of carelessness at least.”
Now, that level of judgement in relation to an investigation as regards this branch, which is suggesting that it’s careless, is that in the same line as your views earlier, that this is a general view of the Post Office, that it’s likely to be down to subpostmaster fault where there are issues?
Andrew Winn: I don’t think it particularly states that.
Mr Stein: Then it goes on to say this:
“If there is anything specific to investigate I am happy to. It may be worth getting something in writing from Fujitsu to confirm there is nothing that could have failed to have polled/software anomalies that will come back to bite us.”
Signed “Andy”.
Is that the right approach, Mr Winn?
Andrew Winn: It feels sensible.
Mr Stein: Does it?
Andrew Winn: To me, yes.
Mr Stein: What about a deeper investigation as to polling issues?
Andrew Winn: Well, that would be Fujitsu who would do that.
Mr Stein: Given at this stage, this is after the other issues that you had been made aware of, after the mismatch bug, such discussions in relation to that, you didn’t think at this stage that the best thing to do would be to, in fact, properly investigate this?
Andrew Winn: I thought it was – obviously thought it was worth going back to Fujitsu to get a report on the polling issues but apart from that I couldn’t see any other route for investigation.
Mr Stein: Then, lastly for my part, and I take you to another document POL00090726. Page 15 of this document first. Go a little bit further down the page so we can see the letter at the bottom, please.
So we can see, Mr Winn, there’s a letter from you to Mr Afzal?
Andrew Winn: Yes.
Mr Stein: So we’re going back to the start of this letter and the date of it, please, if we just show that: 11 October 2011.
“Dear Mr Afzal
“Re Branch Discrepancy
“I’m sorry to learn that your branch has experienced connectivity problems in September. I’m afraid I don’t think I am going to be able to help you.”
You go on to say this:
“Some transactions will never be recoverable, eg stamp sales whilst others, eg card account withdrawals will be recoverable dependent upon the point at which the communications broke down. I appreciate it is difficult to know where you are if communications are lost.”
So Mr Winn, can we establish that in at least 11 October 2011 your knowledge about connectivity problems was that it could lose data?
Andrew Winn: Yes.
Mr Stein: Keep reading:
“However there is a general principle that if a transaction receipt has not been produced by Horizon, the transaction has not completed and cash should not change hands until you are certain of the transaction status. Clearly if recovery takes a period of time, the customer may have left. If the transaction is seen to be recoverable, the option not to proceed with recovery should be chosen.”
Then you go on to say this:
“Unfortunately I am not able to offer any relief to branches who may not have followed recovery procedures in full.”
So this is clearly a discussion about connectivity issues that may have caused a branch transaction difficulty; do you agree?
Andrew Winn: Yes.
Mr Stein: Why here are you saying that you are not able to help:
“I’m sorry to learn your branch has experienced connectivity problems in September. I’m afraid I don’t think I’m going to be able to help you.”
Why is that the attitude of the Post Office?
Andrew Winn: I think there was a recovery process to follow. Quite why the – if it wasn’t followed properly, it could cause issues. That must have been a business decision that we’re unable to resolve. I can’t really remember it well enough, I’m afraid.
Mr Stein: Let’s be as generous as we can to the Post Office at this juncture. We’ve got connectivity issues being caused by the system that the Post Office branches have to use, okay. That’s at least a large part of this problem. Why isn’t the Post Office saying “Sorry to hear about that problem, obviously this at the very least is partly our fault, we will help come sort it out and repay your losses”. Why isn’t that happening?
Andrew Winn: That sounds reasonable.
Mr Stein: It does. Why wasn’t it happening, Mr Winn?
Andrew Winn: I can’t say.
Mr Stein: Can we go then to the page 14, please. Society 14 October 2011 to Mr Afzal, Ferry Road Post Office.
“Dear Mr Afzal.
“Re Branch Discrepancy
“I’m sorry doubt I don’t think I can assist you any further. The process for disputing losses is via the helpline. As far as I can see there is no evidence of Horizon failure been presented which would generate an investigation.”
Then:
“There are processes in place for branches to manage any losses of connectivity. This does not represent Horizon failure and the business has been very clear that it will not compensate losses [due] to connectivity breakdown.
“As such I cannot suggest who in Post Office Limited might take a different view and be able to help you.”
Then if we scroll down to the bottom, we will see it’s signing my you:
“Yours sincerely
“Andrew Winn.”
So let’s have a look at what you said then to Mr Afzal:
“I’m sorry but I don’t think I can assist you any further. The process for disputing losses is via the helpline.”
Which helpline did you mean?
Andrew Winn: NBSC.
Mr Stein: NBSC. So via the Post Office helpline, that’s the only way that Mr Afzal should be able to dispute these matters?
Andrew Winn: That was the designated route for IT problems, yes.
Mr Stein: Then it goes on to say – or you go on to say:
“There are processes in place for branches to manage any losses of connectivity. This does not represent Horizon failure and the business has been very clear that it will not compensate losses due to connectivity break down.”
Let’s take that apart. What are the processes that were in place for branches to manage any losses of connectivity this way?
Andrew Winn: I can’t remember but they would be – on the Horizon Help you would be able to put – but once you got connectivity back, you could refer to them if you hadn’t already got the knowledge of what to do.
Mr Stein: Those processes clearly don’t involve compensation, do they?
Andrew Winn: No.
Mr Stein: Because you clear that one up:
“This does not represent Horizon failure and the business has been very clear that it will not compensate losses due to connectivity breakdown.”
Let’s just understand what you mean by this. Why would the Post Office not compensate subpostmasters for issues caused by connection problems of the Horizon equipment?
Andrew Winn: I cannot recall the process and how it was managed. I’m sorry.
Mr Stein: As an example, was this directive from the Post Office?
Andrew Winn: From the wording in my letter I’ve obviously been up and checked the business position at the time and reiterated what I’ve been told.
Mr Stein: Is this you – you were talking earlier about you settling into your job and getting better at it after a few years. Is this you having settled into the job and having got better at it after a few years, Mr Winn?
Andrew Winn: That was the business rules which I’m following.
Mr Stein: Which you are enforcing.
Andrew Winn: Yes, I’m sorry the business rules weren’t correct but –
Mr Stein: Why do connection issues not represent Horizon failure?
Andrew Winn: I would regard loss of connectivity as being like losing Wifi connection, the equivalent, as I understand nowadays. So I wouldn’t understand it as Horizon being at fault. I’d understand it as the first links which you know, a lot of rural branches did suffer from connectivity issues. I mean, my interpretation was that that was not Horizon at fault but it might –
Mr Stein: So it is just hard luck on subpostmasters; is that right, Mr Winn?
Andrew Winn: Well, there’s a process – there was a process to follow to recover the transactions.
Mr Stein: But to make sure that you are finished with the problem, you finish off by saying:
“As such I cannot suggest who in Post Office might take a different view and be able to help you.”
So you’re saying “That’s it”?
Andrew Winn: That would imply that I checked with the people who I think might have a different view and already got their opinion.
Mr Stein: Excuse me one moment.
Thank you, Mr Winn.
Thank you, sir.
Sir Wyn Williams: Thank you, Mr Stein.
Anyone else?
Questioned by Ms Page
Ms Page: Yes, please, sir.
If I can take us to a document which helps a little bit on the subject of phantom transactions. It’s POL00093133. This is an email chain, which you’re not actually copied in on but which is makes a reference to you, which may assist. Could we start at the bottom, please, the last page in this so that we read up chronologically.
Do you know either of these names Bethany or Sally Buchanan? They seem to be from Customer Service, if that means anything?
Andrew Winn: I can’t remember –
Ms Page: What we’ve got here is Bethany saying:
“Are you able to help me with this office or let me know who can help.
“The office had a major software problem back in week 41 (January 2003). After numerous phone calls to NBSC and engineer visits to change cables, monitor et cetera, a software problem was identified and the processor changed. Apparently the screen would take on a mind of its own, jump screens, add items to the sales stack, et cetera. The office balanced £422.74 short that week, which is very unusual for the office – they are normally within £20 each week.
“The subpostmistress has waited for an error notice to come back, nothing has been received yet and I’ve checked with Chesterfield several times – nothing so far. I have also checked the paperwork in the office for week 41, along with week 40 and week 42 and I can’t find anything. Is there anything you can do at Dearne Valley …”
Is that the place where your premises were when you were in problem management?
Andrew Winn: Yes, that’s correct.
Ms Page: “… to have that week’s work checked on the system to see if this has been caused by a systems fault. The only other thing I can think of is that if the system was going daft and putting things onto the sales stack, this wasn’t picked up every time by the person serving and one or more customers have been given money they were not entitled to and have just kept it.”
So that is the start and if we go up then to response, there’s a blank page so we need to go a little buyer thank you, and we can see that this is 27 June 2003 from Terry Rudd in customer relations:
“Thanks for raising this matter with us. An investigation has taken place with Julie Welsh, our contact at Fujitsu, and she accepts that the PM did call the Helpdesk to state that transactions were appearing on the sales stack, but kit was swapped out and the problem did not reoccur. As no further problems were reported, she thought that was the end of the matter.
“As the losses occurred back in January, information relating to this branch has now been archived but your concerns have today been raise with the Problem Management Team who have more experience in dealing with phantom transactions. I am unsure which member of the team will be assigned to the case, but if you have any further questions, the Line Manager for the team is Andy Winn.”
Then it sets out a reference code for queries regarding this issue and signs off:
“… Customer Relations [can’t help] but I am sure Andy and his team will do their best to resolve this.”
So what we have there is an indication that phantom transactions were something sufficiently well known in your team that this was being specifically referred to you and your team?
Andrew Winn: Yes.
Ms Page: Yes?
Andrew Winn: Yes.
Ms Page: Which ties, doesn’t it, with what Mr Stein told us, that his Core Participant had spoken to somebody about phantom transactions. So clearly this was a known problem right the way back to this early date.
What would you have done about a query like this?
Andrew Winn: Would have allocated it to a team member to raise it with Fujitsu.
Ms Page: What sort of response might you have got from Fujitsu?
Andrew Winn: A review of their – I don’t really know what they would have done. I would have hoped they’d have looked at the – identified the times and looked to see if there’s any problems arising from there and reported them back to us.
Ms Page: So following that report, let’s say they come back to you and say “Yes, we’ve identified there are phantom transactions in this particular branch or in this particular group of branches, potentially”, what would your team have done about that?
Andrew Winn: Well, we’d have escalated it probably to our IT team and said what we’re going to do about it, how we’re going to resolve it.
Ms Page: What would you have done in terms of the branches?
Andrew Winn: Resolved any financial accounting discrepancies.
Ms Page: How did you go about that?
Andrew Winn: That would have been done through a P&BA or whatever it was called at that time.
Ms Page: So there would have been quite a lot of communication, wouldn’t there, between your team and people at Fujitsu, people in P&BA –
Andrew Winn: I would have thought so, yes.
Ms Page: – people in the groups, the POL or perhaps the group, at that stage, IT directorate – was it one or the other, do you know?
Andrew Winn: I wouldn’t know.
Ms Page: Did your team escalate the issue of phantom transactions?
Andrew Winn: I would have thought so from – I presume that that was raised to our team. I can’t remember it.
Ms Page: Do you have no memory of escalating this issue of phantom transactions when you were the team leader at this time?
Andrew Winn: No.
Ms Page: Wouldn’t it have been a matter of concern to you?
Andrew Winn: Yes.
Ms Page: Wouldn’t it have been a matter of great concern to the business generally?
Andrew Winn: I don’t know how easily it was – what the Fujitsu findings were. I can’t comment.
Ms Page: If Fujitsu had come back and said “No such thing, there’s no phantom transactions, it doesn’t exist”, would you have left it there?
Andrew Winn: I don’t know where else I could have gone; so, yes, I think I would have.
Ms Page: So never mind that your team evidently has, according to this email, more experience in dealing with phantom transactions, never mind that this isn’t just a one-off, this is clearly something that’s happened on a few occasions, you would have just taken Fujitsu’s word?
Andrew Winn: I wouldn’t have known where else to go.
Ms Page: You wouldn’t have taken it to anyone in IT at POL?
Andrew Winn: If it was – yes, quite possibly. I would have imagined that the enquiry would have gone to Fujitsu and our IT teams.
Ms Page: When you were at problem management, what was the directorate that your team sat in?
Andrew Winn: I would imagine the IT directorate. I can’t actually remember.
Ms Page: So you don’t know who the director was in charge of problem management? You don’t know who it was above your team leader or your boss or whoever that may have been?
Andrew Winn: No, I can’t remember.
Ms Page: You gave us the name of one of your bosses earlier, didn’t you. Do you remember what she was the boss of?
Andrew Winn: Marie Cochate. Yes, that was when I was in Chesterfield at the process improvement team.
Ms Page: I think you gave us a name for somebody at –
Andrew Winn: Actually, I think she was my boss in problem management as well, initially.
Ms Page: So that was something in accounting, was it?
Andrew Winn: I think she moved as well. I think she moved from the accounting – from the process improvement team into the same kind of team areas as I was in within problem management, if I remember correctly.
Ms Page: Have you listened to the evidence from witnesses over the last week or so?
Andrew Winn: No.
Ms Page: Because, collectively, they’ve given evidence that problem management really was the way that Post Office monitored the performance of Horizon and indeed the performance of Fujitsu in running and maintaining the performance of Horizon. Is that how you understood the role?
Andrew Winn: No, not really.
Ms Page: Well, what did you understand it to be then?
Andrew Winn: To be resolving problems, making sure the appropriate people were resolving problems. I don’t think there was a – particularly a reporting schedule. I don’t remember that coming out of problem management when I was there.
Ms Page: Was this the business as usual way to resolve bugs, errors and defects that arose in Horizon?
Andrew Winn: It was the way it should be recorded, yes.
Ms Page: We know that by 2010 – and I’m only going forward just to kind of come back, if you like – by 2010, it was people in problem management that were the interface with Fujitsu over the receipts and payments mismatch bug, weren’t they?
Andrew Winn: Yes.
Ms Page: So the role was to be the interface on bugs and defects, and so forth, yes?
Andrew Winn: Yes.
Ms Page: So it was a crucial IT role, wasn’t it?
Andrew Winn: Yes.
Ms Page: What interested you in that role when you applied to manage that team?
Andrew Winn: To be brutally honest, a promotion.
Ms Page: What grade were you as a manager of that team of 12?
Andrew Winn: CM1.
Ms Page: TM1?
Andrew Winn: CM1.
Ms Page: Where did that place you in the grades at the time?
Andrew Winn: That was the highest grade of middle manager before you became a senior manager and from there – so there’s different grades of senior manager and then directorate level.
Ms Page: How busy was your team of 12?
Andrew Winn: I would say the team were not desperately busy and I was very busy.
Ms Page: Right. Perhaps you can pull that apart for us. Why were you busy and they weren’t?
Andrew Winn: There were plenty of problems coming in. The problem with, in my opinion, the problem management setup was that if, effectively, a member of the product team, which is nothing to do with IT, but a guy from the product team who I knew very well said to me “You know, we’re told to report any problems into your team”, you make a note of it, say “Thanks very much, let us know when you’ve resolved it” which – so the team, bear in mind I’ve got 12 people, seemed really to be employed just to record something and wait for it to be told it was cleared.
So when you went across the team “How are you doing on this one?” It was always “I’m waiting for”, “I’m waiting for”, “I’m waiting for”, and it was kind of “Oh, can you chase them”, et cetera, et cetera.
It didn’t seem to – I was quite uncomfortable with that. I kind of expected it to be much more proactive. I’d say to the teams, you know, “We need an article to resolve a product issue”.
“Well, they haven’t got time to do it”.
“Well, have a go at doing it yourself, submit it to them and they’ll probably be that horrified at what you are doing, they’ll do it themselves”, and I was trying to work on that basis of being more proactive. On the IT side we wouldn’t be able to do because we wouldn’t have had the knowledge obviously, but I found it a difficult role.
Ms Page: Yes, and one that you have already admitted you weren’t really qualified for.
Andrew Winn: Correct.
Ms Page: Given that it was a role that was crucial for the handling of the defects, bugs, errors and that that’s central to the work of this Inquiry, why didn’t you say anything about it in your witness statement?
Andrew Winn: I did make reference to it, didn’t I?
Ms Page: Why did the team get taken over by Risk, as you described to us earlier?
Andrew Winn: A routine business as usual review of the departments, as a whole. It wasn’t my decision but there was kind of a lot of change and it was decided to merge the two teams together. They were quite similar in the understanding, et cetera, et cetera. So I could understand why it was done.
Ms Page: Was there any strategic thinking at all about the business as usual management of problems, defects, bugs in Horizon?
Andrew Winn: Not that I was made aware of, no.
Ms Page: Just lastly on this topic, what we have here the one remaining, as far as we can tell document, from this period of time when you were team leader of this crucial team that had to interface on bugs, errors and defects is one email from a different team, customer relations. We have no emails from problem management itself, no records at all from your time at problem management and nothing at all about the problem of phantom or ghost transactions, which apparently your team knew about.
Can you give us any understanding or explanation for why there’s so little information about your team at that time?
Andrew Winn: No, I’m surprised there was – as far as I’m aware everything that we did was logged, so I’m quite astonished at you saying that.
Ms Page: Can we perhaps turn then a little ahead to POL00045457. Now, this is a document which seeks to define the process of settling centrally and it says at the bottom that it’s been a contribution from you. So it looks perhaps as if it might be something that’s to go into a manual or something. Does that look familiar?
Andrew Winn: Yes, yes.
Ms Page: If we just go back up again, it looks to be – it’s expressed explicitly as a clarification?
Andrew Winn: Yes.
Ms Page: It says:
“A recent audit has highlighted that many branches are unclear on how to deal with losses and gains, particularly around ‘Settling Centrally’.”
Do you think that audit might have been what you did before you did those slides in January 2009?
Andrew Winn: I wouldn’t have done the audit, I don’t think. I think that would have been feedback from network auditors. I was out in branches.
Ms Page: The reason I ask is because one of the slides that you had created in January ‘09 sort of set the task of defining the process of settling centrally and this document appears to do that?
Andrew Winn: Yes.
Ms Page: So do you think this document comes then from that process in January –
Andrew Winn: Quite possibly, without knowing the date of the document.
Ms Page: The document doesn’t appear to bear a date and –
Andrew Winn: It sounds sensible.
Ms Page: All right. Well let’s assume then that it’s part of that January 2009 reappraisal, shall we say. When we asked our client subpostmasters who had experience of running post offices before January 2009, when we asked them to have a look at this, none of them recognised this process at all and, indeed, Janet Skinner said that she was not aware that there was any dispute resolution process whatsoever. Is it right that, prior to January 2009, there really wasn’t a recognised dispute process at all?
Andrew Winn: Well, prior to me starting in the role, then that would be correct, yes.
Ms Page: So do you know why that role came up?
Andrew Winn: I think when I – I should say, we had a reorganisation in P&BA and the different teams were created and I went into the debt team and my boss Alison said, “Look, we’ve got people trying to contact teams and they’re not responding to, and we need a central point, we think it should be you, have a look at it”, and so the role kind of evolved from there.
Ms Page: You told us earlier that you didn’t have to apply or interview for this job. Was it a job that was, in effect, one that you sort of created yourself once in it?
Andrew Winn: Yes. Yes, I think when I first got there Alison and Rod Ismay’s view, they had some ideas about the role but, effectively, it was me who kind of developed it to what it was.
Ms Page: At this stage, were you still a sort of top middle management or had you entered the realms of senior management by this stage?
Andrew Winn: No, still the same grade.
Ms Page: Why do you think you were qualified to adjudicate on disputes?
Andrew Winn: I don’t think I was particularly qualified, but I probably felt that it was the role that I was best suited to working within the whole Royal Mail Group.
Ms Page: Why?
Andrew Winn: Because I think I investigated things and was prepared to look outside the box and I had a bit of an empathy with the subpostmasters, I think.
Ms Page: Do you still think that having looked back at some of your correspondence?
Andrew Winn: I’m disappointed with some of the things I wrote, yes.
Ms Page: We can take that document down and before I turn to another one, I’d just like to ask you a little bit about Mr Lee Castleton. Is that a name that rings any bells for you?
Andrew Winn: Yes.
Ms Page: Did you know about his case at the time?
Andrew Winn: I think his case was – I think his case was very topical when I first joined the team.
Ms Page: Right.
Andrew Winn: So it would be very much in my early days. I don’t think I was involved in any of the decisions around suspension or anything.
Ms Page: But it was known about, was it, in your team, quite well known?
Andrew Winn: Yes.
Ms Page: All right. So you’ll have been aware then, won’t you, that in – you won’t necessarily remember the date but it was in early 2007 that the judge awarded Post Office damages of approximately £26,000 but costs of £321,000 which you might imagine bankrupted Mr Castleton. He lost everything he’d invested in his branch, he lost his living, his family were treated like thieves and they endured years of hardship.
What we now know from documents in this Inquiry, which haven’t yet been sort of picked over but which I can quote to you briefly from – I don’t know if we necessarily need to put it on the screen – is that there was a clear intent on the part of the Post Office, with legal advice, to pursue the claim:
“… not to make a net financial recovery but to defend the Horizon System and hopefully send a clear message to other SPMs that the PO will take a firm line and to deter others from raising similar allegations.”
So that was the purpose. It was not ever envisaged that the Post Office would actually get that costs order back. That was loss leader, if you like. But the purpose was to send a firm line and a clear message to deter others.
Now, is that how the case was understood at the time? Is that something that your team would have been aware of, that it was a sort of flagship case, if you like, to try to deter others?
Andrew Winn: No, I don’t think so. I was conscious that it was probably the most high profile case at that time but I don’t think I would have picked up that message.
Ms Page: We can also see from documentation that the lawyers in charge of the case were also conscious of other cases, including one which –
Sir Wyn Williams: Hang on a minute.
Do you have any direct knowledge of Mr Castleton’s case at all, Mr Winn?
Andrew Winn: No, I don’t.
Sir Wyn Williams: I mean, I’m conscious that this is a very sensitive case, Ms Page, but don’t think it’s appropriate to use the witness just for you to read extracts of other documents.
Ms Page: I was just to about to come to the documents which he is involved with, sir. So I hope that’s.
Sir Wyn Williams: That’s fine.
Ms Page: That was a scene setting, if you like.
The related case, if you like, which you were involved with was a Mr Bilkhu. Does that ring any bells?
Andrew Winn: No.
Ms Page: So Mr Bilkhu issued a claim against POL but then withdrew it because he was threatened with costs of instructing an expert in the region of £1 million and he told you about that. Do you remember?
Andrew Winn: No.
Ms Page: Well we can perhaps then have a look at POL00001304 at page 29, bottom of the page. This is from him to you. We can perhaps go to the top of the page, just so that you can see that. He was writing from Bowburn Post Office and you can see there at the top, “Dear Mr Wynn”, he has obviously misspelt your name but the correspondence shows that he had written to you and this was part of a bit of a back and forth between the two of you.
There’s one thing that I’d like to just ask you about before we go to the bottom. We see the reference to – he says:
“Your letter is … a repeat of … previous letter and is similar in style to those I have received from Michele Graves and Philippa Wright (Flag Case Managers for Adam Crozier/Alan Cook).”
Do you know what a “flag case manager” was?
Andrew Winn: No.
Ms Page: Do you know those names?
Andrew Winn: Alan Crozier and Alan Cook, yes. The two ladies, no. I can’t remember them.
Ms Page: What about the two gentlemen?
Andrew Winn: They were heads of Post Office.
Ms Page: They were seniors?
Andrew Winn: Well, they were executive directors.
Ms Page: So you don’t recall what the flag case managers did for them?
Andrew Winn: No.
Ms Page: If we go a bit further down, we see that Mr Bilkhu tells you, “In summary”, it says:
“… POL may consider the matter closed [that’s his complaint] but I do not. According to my legal team the case can we resumed subject to legal niceties.
“In summary, the case was withdrawn [he is talking about his case] because POL’s legal team demanded that Horizon accounts at Bowburn [Post Office] for the last 4 years be examined by a forensic accountant. The cost (estimated at £1 million) be borne by me.”
Reading that letter, does that not ring any bells?
Andrew Winn: No.
Ms Page: The idea that the legal department would threaten somebody with costs estimated at £1 million?
Andrew Winn: No.
Ms Page: Looking back, do you think this is part of a culture of using legal process to threaten subpostmasters?
Andrew Winn: It’s quite possible, yes.
Ms Page: Is that perhaps part of the sort of setting of what we’ve seen in your own correspondence of this sort of putting the burden of proof on the subpostmasters –
Andrew Winn: No, I don’t think so.
Ms Page: – of using the law against them?
Andrew Winn: No, I don’t think I’ve ever threatened anything like that. I’ve tried to – where I can see a way of investigating/helping, I’ve tried to do that.
Ms Page: Some very brief questions, if I may, on the document which we’ve looked at quite a lot, the receipts and payments mismatch meeting. I just want to look at the first page of it again, if I may. It’s POL00028838. If we just look at those Post Office names, who would you say was the most senior person there?
Andrew Winn: Ian Trundell, although Alan Simpson I don’t know what grade he was at.
Ms Page: Ian Trundell and Alan – sorry, did you say?
Andrew Winn: Alan Simpson.
Ms Page: All right.
Andrew Winn: Well, I’m guessing there. I don’t actually know what grade anyone was at.
Ms Page: But your first reaction was that those were the two senior people that meeting?
Andrew Winn: Yes.
Ms Page: Who would you have reported back to about this? Presumably Mr Ismay?
Andrew Winn: Yes.
Ms Page: And did you report back to Mr Ismay about this bug?
Andrew Winn: I’m – whether I’ve talked directly to Mr Ismay or Mrs Bolsover, I’m not certain, but I would certainly have fed back as part of normal communications.
Ms Page: Thank you. The document can be taken down.
So you are saying you would have reported back to Mr Ismay and Ms Bolsover; is that right?
Andrew Winn: Yes.
Ms Page: There’s just two more issues I would like to look at, sir, if I may?
Sir Wyn Williams: Is anyone else intending to ask questions?
Mr Beer: Sir, no, they’re not.
Sir Wyn Williams: Fine. Off you go then. Five minutes, Ms Page.
Ms Page: Thank you. If I can look then, please, at POL00105280. If we could look at page 3 – and I won’t take you through the whole history up to page 3. Page 3 sort of dives in.
This is in the summer of 2013, so this is post receipts and payments mismatch bug and around the time when Second Sight’s work is pretty well known within about Post Office, yes, and you’re asking a contracts manager – this is a discussion with a contracts manager having been in touch with a branch about a loss that dates back ten years, and this is your putting three possibilities for how to deal with it to the contracts manager. You say:
“Hi Nigel, I don’t actually disbelieve the branch here but the claim that two sets of auditors have recorded missing stock as being present is a bit scary. Stamps are just pieces of paper at Swindon so there would not have been a surplus at another branch.
“I can see three options. Pay up - we don’t believe you. Create a phantom rem out, branches can now invent ten year old errors that we have to let them off on because we do not have any information about. Plus Swindon will not pick up a phantom rem, so I can rem stamps out and just sell them on my retail side.”
I think that seems to be a suggestion that you’re making for a way to sort of balance it off. Is that how I should read that?
Andrew Winn: Yes, that’s how it could have done been done remotely, yes.
Ms Page: And then network write off:
“We believe you and we’re making a gesture in recognition of long years of accurate accounting and his TC rate is excellent. However, this does leave huge question marks over the audit process. None are particularly appearing. Thoughts?”
First of all, it’s a bit striking, isn’t it, that one of the options is “we don’t believe you” when you’ve actually said at the start of the email, “I don’t actually disbelieve the branch here”?
Andrew Winn: I think what I’m trying to do there is make the first two look at what they were, totally unappealing options, and the third one is where we’re going.
Ms Page: Well, the third one is indeed where you go.
If we go up to the next response, though, page 2, what’s quite interesting is that the response from the network manager says in paragraph 3 here:
“As you say, none of the options are particularly appealing. I think the first option, making the SPMR pay up, could open up a can of worms. I’m not sure that the SPMR is a member of the NFSP but, given that the amount involved represents a significant percentage of his salary, I feel sure he would take this further. This could put us in a position of trying to defend ourselves against a charge that the auditors didn’t do their job properly and could potentially give the NFSP or an MP some useful ammunition.”
So it’s again this idea really, isn’t it, that it’s about the ammunition, it’s about MPs, it’s not really about doing the right thing?
Andrew Winn: No, I don’t think so. I think we were aiming to do the right thing.
Ms Page: Well, you did do the right thing but, rather than just doing the right thing, you’re discounting other options not on the grounds that they’re wrong or because you actually believe the branch but because it could potentially give the NFSP or an MP some useful ammunition?
Andrew Winn: And I think that would be true if we’re saying that two sets of audits were both incorrect. That’s not the kind of thing you want to advertise particularly when you don’t know if that’s the case.
Ms Page: Well, you do know it’s the case. That’s the investigation. That’s what’s happened.
Andrew Winn: Well, no, we don’t. That was the problem with this case, that we could understand what the guy was saying but he’s saying that two sets of auditors have come in, audited the branch. He said there’s an issue with these which is going to get – needs resolving. So the auditors have said, “Okay, we’ll assume that they’re there”. If they weren’t there, I mean, that’s what an auditor’s job is – to identify discrepancies. So for two of them to go in and not, would suggest that any audit cannot be relied upon.
Ms Page: So although you believe the SPMR, although you –
Andrew Winn: I don’t disbelieve the postmaster but I don’t believe that two auditors have not done their job properly. So I’m left a little bit I don’t know where we are anyway.
Ms Page: Just one last, if I may, because this brings us up a little bit more to date and it’s POL00092640 and this dates from 2015.
This seems to relate to what we’ve come to understand were weekly Horizon meetings. These were regular calls – is that right – with lawyers involved as well, people from the security team. Do you remember a weekly Horizon meeting being instituted in around 2013 and carrying on for some time apparently?
Andrew Winn: I don’t particularly remember it being a weekly meeting, though I accept that it could have been.
Ms Page: You do remember the Horizon meetings, do you?
Andrew Winn: I remember being on Horizon meetings. Whether I was on the weekly ones or not I’m not certain.
Ms Page: Well, certainly in relation to this one, it’s referring to one that’s taken place in August of 2015 and it says:
“As you will see there were numerous issues raised on the last call which are of concern.”
First of all, it says:
“Andy Winn is still receiving requests to authorise FJ to correct problems.”
Presumably that’s Fujitsu?
Andrew Winn: Mm-hm.
Ms Page: And what I understand that to mean – and you correct me if I’m wrong – is that that’s Fujitsu asking to go into Horizon to correct problems and you’re authorising them to do so. Is that what’s going on there? Is that the process?
Andrew Winn: I’m not sure what Fujitsu wanted to do but, yes, it’s asking me to authorise them to do something. I don’t know what from here.
Ms Page: Well, we’ve actually got three issues that come up. The first one is that they asked to do this when kit is removed from branches which can cause issues and then, secondly, there’s the discontinued sessions issue which has gone on to explain further, and it talks about two new products which caused the system to disconnect and recovery scripts are failing.
So these are two different things which seemingly they’re having to go into the system to correct and you’re having to sign off on the process. Is that something you recall doing?
Andrew Winn: I would imagine – I could see myself being the voice that would give POL approval, yes, if that makes sense.
Ms Page: It says in the latter part of that paragraph:
“This is apparently standard business as usual and FJ seek authorisation to correct it. It is unclear at the present time whether or not there is process assurance and documentation. I do not know whether POL have full visibility of the actions of Fujitsu and the ways in which they correct the branch data.”
Does that ring a bell?
Andrew Winn: It doesn’t ring a bell but it makes sense to me.
Ms Page: That there wasn’t really a process that was being followed?
Andrew Winn: And that I was – I think I, possibly naively, assumed that the IT department were the ones who would be kind of joined up with that, but certainly as far as I was concerned that wouldn’t be the case.
Ms Page: Then the paragraph below, just to finish off on this:
“Andy Winn also raised the issue of a computer problem with Camelot for which a fix had been issued but pointed out that branches would encounter unexplained losses that Wednesday when they conducted their BTS procedure. He went on to explain that he had received an email from Fujitsu about an incident which had occurred in June. It was termed a ‘major incident report’ and related to a branch which had an incorrect discrepancy at the time of conducting a branch trading statement. The email suggested that information had been sent to POL. Andy Winn had not previously known about this issue and so asked to whom the information had been sent. He had no response.
“Andy went on to say that he did not fully understand the issue and that a maximum of 247 branches would have been affected. 118 of those would have generated reports based on corrupted data. There was only one account in connection with which POL could have held someone responsible for the shortfall.”
So in 2015, following the receipts and payments mismatch bug, following the fact that Second Sight have become involved, following the fact that that means that it became apparent that the receipts and payments mismatch bug still hadn’t been followed up properly, you’re still having problems, aren’t you, at POL with actually getting on top of and dealing with bugs that affect corrupted data?
Andrew Winn: Is this a receipts and payments mismatch issue?
Ms Page: I don’t believe it is. I believe it seems to be a different issue. But this is evidence, is not, that POL and Fujitsu are still not working through proper procedures –
Andrew Winn: Yes, I think so, yes.
Ms Page: All right.
Sir Wyn Williams: Are you going to end on a high note, Ms Page?
Ms Page: Thank you, sir.
Sir Wyn Williams: Is that it, Mr Beer?
Mr Beer: Yes, sir. That concludes the questioning of Mr Winn.
Sir Wyn Williams: Thank you, Mr Winn, for coming to give evidence. It’s been a long day clearly.
Andrew Winn: It certainly has.
Sir Wyn Williams: I’m grateful to you that you came to give the answers to very many questions.
Andrew Winn: I hope I could be some help.
Sir Wyn Williams: Thank you.
Mr Beer: Sir, thank you. I think that’s us done now until Tuesday, 7 March when we will hear evidence from Liz Evans-Jones.
Sir Wyn Williams: Yes, all right. Thanks very much.
Mr Beer: Thank you, sir.
(4.10 pm)
(Adjourned until Tuesday, 7 March 2023 at 10.00 am)