1 Wednesday, 23 November 2022 2 (11.00 am) 3 (Proceedings delayed) 4 (11.18 am) 5 MR BLAKE: Good morning, sir. 6 SIR WYN WILLIAMS: Good morning, Mr Blake. I'm very sorry 7 if I've kept people waiting. 8 MR BLAKE: Sir, today's witness is Mr Sibbick. Thank you. 9 SIR WYN WILLIAMS: Yes. 10 DAVID SIBBICK (sworn) 11 Questioned by MR BLAKE 12 MR BLAKE: Thank you very much. 13 Can you give your full name, please? 14 A. My name is David Sibbick. 15 Q. Mr Sibbick, you should have in front of you a witness 16 statement. 17 A. I do. 18 Q. Is that statement dated 26 August of this year? Yes? 19 A. Yes, I have that. 20 Q. Thank you. On the final page, page 19, there's 21 a signature there. Can you confirm that that's your 22 signature? 23 A. That is indeed my signature. 24 Q. Thank you. Is that statement true to the best of your 25 knowledge and belief? 1 1 A. Yes, it is. 2 Q. Thank you very much. 3 For the purpose of the record, the statement is 4 WITN03350100. 5 Mr Sibbick, thank you very much for attending the 6 Inquiry today. As you know, I'm going to be asking 7 questions on behalf of the Inquiry. 8 Your statement and the exhibits are now in evidence, 9 so anything that I'll ask you will be supplementary to 10 that. 11 I'm going to start with your background. You 12 started your career in the Civil Service in 1960? 13 A. I did, a long time ago. 14 Q. I think you began in the Post Office itself? 15 A. That's correct. 16 Q. What did you do in the Post Office? 17 A. I started by working in what was, I think, called the 18 Establishments Division, it was concerned with postmen's 19 pay and issues like that. 20 Q. Thank you. Then you moved to what was then called the 21 Ministry of Posts and Telecommunications? 22 A. When the Post Office moved from being a government 23 department, there was a small sponsoring Ministry set up 24 to do the things that sponsoring ministries do, and 25 I moved to that rather than staying in the Post Office. 2 1 Q. Then at some stage you became the private secretary to 2 the Permanent Secretary in that Department? 3 A. Yes, I did. 4 Q. Eventually that Department became the Department for 5 Trade and Industry? 6 A. It was absorbed into the DTI as then was. 7 Q. In 1989 you became Director of Posts -- 8 A. Correct. 9 Q. -- and you retired from the Civil Service in 2000? 10 A. Yes. 11 Q. In the period that we're dealing with, in this phase, 12 procurement, acceptance and rollout, it looks as though 13 you were the most experienced member of the Civil 14 Service when it came to matters relating to the Post 15 Office. 16 A. I think that's almost certainly correct, yes. 17 Q. Thank you. 18 Before we start, do you have any general 19 observations on how the Post Office was governed during 20 your period? 21 A. Well, that is an absolutely huge question which I'm sure 22 you will be looking at in much greater detail in Phase 6 23 of this Inquiry. I suppose what I can say is that there 24 was a convention at the time that the way these 25 organisations should be governed was that the 3 1 government, a minister, would appoint the board, and the 2 board was responsible for all operational issues within 3 the Post Office, reporting obviously to the Minister, 4 who would have regular dialogue with the chairman, but 5 the overriding principle was you appoint the chairman 6 and board and then you either back them or sack them. 7 You didn't meddle day to day and part of the 8 argument for that was that, if you were going to meddle 9 day to day, if you were going to have your civil 10 servants crawling over everything that the operational 11 people in the organisation tried to do, you would not 12 attract to the top of the organisation people of the 13 sort of quality that you needed to attract. 14 Q. We'll get to the detail in due course but, in practice, 15 was the Post Office left alone to get on with its 16 business or was there meddling? 17 A. I suppose it depends on what you call "meddling". This 18 overall principle was there but, as almost everything in 19 the real world or the political world, it gets a bit 20 overtaken from time to time by events, and Horizon 21 turned out to be one such event. 22 Q. I'm going to take things chronologically, starting with 23 your first involvement. 24 At paragraph 7 of your witness statement you've 25 said, you describe the long-running sore between the 4 1 Benefits Agency and the Post Office and officials being 2 scarred by their involvement. Those are obviously 3 powerful words. Can you tell us about your early 4 involvement in the project, so the early to mid-1990s, 5 and those differences? 6 A. I was not particularly involved in the early part of all 7 of this. My close involvement really started at the 8 point when the Secretary of State for the Benefits -- 9 responsible for the Benefits Agency wrote to my 10 Secretary of State and to the Treasury Secretary of 11 State, saying "Look, we've got a problem here, we need 12 to get round a table and sort it", and that, I think, 13 was in the late summer/early autumn of 1997. 14 I was aware before then of the Horizon project, and 15 I was aware from talking perhaps informally to people 16 that there were these frictions -- 17 Q. Can I just pause you there for one second, sorry, only 18 because we've lost the Chair on the screen. 19 A. Oh. 20 Q. Sorry, we may have to wait a moment. There is 21 a transcript, so everything you've said will be 22 recorded. 23 A. Yes. 24 MR BLAKE: We're just trying to re-establish connection. 25 (Pause) 5 1 Sir, we can see you now. Can you see and hear us? 2 SIR WYN WILLIAMS: Yes, for the first time, I think, in 3 however many months this has been happening, a glitch 4 occurred. So I didn't catch what Mr Sibbick was going 5 to say in answer to your questions on paragraph 7, 6 Mr Blake. 7 MR BLAKE: So this is the long-running sore; you described 8 in 1997 your first awareness that there were -- 9 A. That was the first time that ministers were involved 10 and, therefore, I was involved in having to brief 11 ministers on how to best take all of this forward and 12 what our line might be. 13 Q. You've said officials being scarred by their 14 involvement -- 15 A. I knew, before then, but it was interesting to have it 16 confirmed by a senior official from DSS that these 17 issues had been rumbling on for a long, long time. The 18 Benefits Agency, they can obviously speak for 19 themselves, but they felt that they were being ripped 20 off, if I can put it like that, by the charges that they 21 were having to pay to get the work done by the Post 22 Office. The Post Office, of course, thought the 23 Benefits Agency were not paying nearly enough for the 24 work that they were carrying out for the 25 Benefits Agency, and I think these squabbles just went 6 1 on, I suspect, every time that the contract had to be 2 renegotiated. 3 The Benefits Agency, or DSS, also had a particular 4 problem in, as I understand it, they were not able to 5 get their accounts signed off by their auditors because 6 of the volume of fraudulent transactions around the 7 Benefit Payment Card or other payment methods, and the 8 large amount of money that went missing. 9 So they also had that interest. 10 Q. How important was the benefits card aspect of the 11 Horizon project in those early days? 12 A. I think it was absolutely crucial, from the point of 13 view of Post Office Counters. The great fear was that 14 if too many customers -- if they were forced to accept 15 automatic credit transfer of their payments, would no 16 longer go into Post Offices and, apart from withdrawing 17 their money, actually spending their money on the 18 private side of the shop, the so-called "footfall", and 19 I think this footfall aspect was hugely important to 20 subpostmasters, to the National Federation of 21 SubPostmasters, and the threat of ACT had some years 22 earlier resulted in the National Federation organising 23 a significant protest march down Whitehall, as I recall, 24 from people not so much, I think, opposed to the idea of 25 having to have their payments into bank accounts but the 7 1 fear that they would lose their village shop or their 2 corner shop, if they lived in an urban area. 3 I think, you know, this was the big concern, and why 4 politically it was so important to have in place, if you 5 will, trying to get away from the fraud-prone paper 6 based pension books, and so on, to have something else 7 in place, and the Benefit Payment Card was what they 8 came up with to serve that purpose. 9 Q. Keith Todd, the former CEO of ICL, has given evidence 10 that he didn't expect the complexities arising from what 11 he had thought were two aligned government entities, the 12 Benefits Agency and the Post Office. How obvious would 13 it have been outside of Government that there were these 14 frictions between the Benefits Agency and the Post 15 Office? 16 A. Well, I don't think it would have been evident to the 17 general public necessarily, but I'm sure people who had 18 any real involvement in either of these organisations 19 would have known that this was, as I say, a long-running 20 difficulty. 21 Q. The procurement process began in 1994. Were you 22 involved in that at all? 23 A. No, I was not. I was not. I knew that -- I knew that 24 it was going on but I wasn't involved in any way. 25 Q. Were you aware that Pathway's technical solution was the 8 1 least preferred option of the three bidders? 2 A. The Tom, Dick and Harry, as I believe it was referred 3 to? 4 Q. Yes. 5 A. Yes, subsequently, I was certainly aware that ICL 6 Pathway had been accepted over the other two bidders, in 7 part because it appeared that they were willing to take 8 the biggest element of risk in the project. 9 Q. Were you aware that their technical solution was the 10 least preferred of the three? 11 A. I'm not sure that I was conscious of this. In the early 12 days, I had no reason to believe that ICL would not be 13 capable of delivering what they had contracted to 14 deliver. 15 Q. We'll come to work that was carried out in the summer of 16 1998 but, at the procurement stage or between the 17 procurement stage and that period, were you aware, for 18 example, that concerns had been raised about a system 19 known as Riposte during the procurement stage? 20 A. My understanding was that Riposte was something that 21 emerged once it had been decided not to go ahead with 22 the Benefit Payment Card. I'm not a computer expert, 23 I'm not even terribly computer literate, but what I did 24 understand and I understood from talking to senior 25 people in ICL, that ripping out the Benefit Payment Card 9 1 part of the system and replacing it with something else 2 was absolutely a non-trivial thing to have to do and 3 would require a lot of work in a certain amount of time. 4 To answer your question about Riposte, my 5 understanding was that Riposte kind of emerged when you 6 were looking at the, I think it's called middleware or 7 shareware, for the system going forward without the 8 Benefit Payment Card. I wasn't aware that it was, in 9 any sense, an issue whilst the Benefit Payment Card was 10 still the way forward. 11 Q. We will, in due course, come to some references to that 12 system in 1998, so I don't want to -- I'll take you to 13 that. 14 Were you aware, for example, that the evaluation 15 board had concluded that Pathway required a proactive 16 management stance going forward from the procurement 17 stage? 18 A. I think probably not really. I had no reason to believe 19 that the Pathway project was not being well managed 20 until it emerged that it was running late and over 21 budget, and so on, which was -- I think the system 22 originally went live in 1996 and it was in late 1997 23 that the two parties put Pathway into a position of "We 24 can pull the plug on this now if we want to, we've got 25 the right now to pull the plug it if we want to". And, 10 1 as I say, it was about that time that ministers got 2 involved, because they could see -- well, they could see 3 a car crash coming. 4 Q. Let's move to the spring to autumn of 1998 -- you have 5 dealt with that at paragraphs 15 to 17 of your witness 6 statement -- and the Horizon Working Group. Can you 7 tell us why the Horizon Working Group was set up? 8 A. Yes. First, there's some confusion. There were two 9 Horizon working groups. The first one was the sort of 10 ministerial one, senior officials, and so on, with 11 Adrian Montague's technical group reporting to it. 12 There was a second group set up which, at the time, 13 nobody could think of an original name to distinguish it 14 from the first one, I suppose, but that was more to keep 15 some of the other parties -- so the NFSP, the CWU, and 16 so on -- to keep them involved. It was also thought 17 that, because it was their members who were going to be 18 using this system, that it would be a very useful body 19 to monitor progress of the rollout once that had 20 started. 21 So that had an ongoing role, if you like, the second 22 one. It was chaired initially by Ian McCartney, 23 I think, and subsequently by Alan Johnson. 24 Q. Yes, and I think you refer to them in your witness 25 statement as the working group 1998 and the working 11 1 group 1999? 2 A. Yes. 3 Q. Let's focus on the working group 1998 for now. 4 A. Yes. 5 Q. Can we go to HMT00000034, please. Thank you very much. 6 This is the report that they produced in July 1998. 7 Can you tell us the background to this report, please? 8 A. Well, the background to it was to look at where the 9 project had got to, what options there were moving 10 forward, and to make sure, as well, through the 11 technical subgroup that the technical aspects of it had 12 been properly explored and properly understood by 13 government. 14 Q. Can we turn to page 4, please. That's internal page 3, 15 but it's page 4 on the pdf. Thank you. This is the 16 "Summary and conclusions", and I'm just going to read 17 from 1.1. At the end of 1.1 it says: 18 "We were asked to consider: 19 "[First] whether the project is technically viable; 20 and if so, how quickly it can be completed and at what 21 cost to government; 22 "[second] the direct and indirect costs of 23 cancellation and of any alternative available to deliver 24 the project's objectives." 25 Then it says: 12 1 "We set up an Independent Panel of experts to 2 address the first question." 3 The independent panel of experts is the group that's 4 led by Adrian Montague; is that correct? 5 A. That is correct. 6 Q. Thank you. Then can we look at, down the page to 1.4, 7 please, these are the three options that this group set 8 out at this stage, so the summer of 1998. Number 1 was 9 "Seeking to continue the project"; and scrolling down, 10 the second option would be to "[Reshape] the project by 11 cancelling the Benefit Payment Card"; and the third 12 option "Terminating the whole project". 13 So those were the three options in play at that 14 stage. 15 Can we look at the "Recommendations" on page 8, 16 please, and it's at the bottom of page 8. Thank you. 17 So the recommendations, starting at 1.7: 18 "The Working Group, apart from the DSS and Treasury 19 Social Security team, recommends that ..." 20 Then over the page, thank you very much, first: 21 "in line with option 1, ICL Pathway should be given 22 terms for continuing with the project ..." 23 So essentially their first recommendation is 24 continuation. 25 Then we see, for example, the third bullet point 13 1 there: 2 "If ICL cannot accept these terms, or if the 3 negotiations cannot be satisfactorily concluded within 4 two months, Ministers should sanction the public sector 5 parties withdrawing from the contracts on the grounds of 6 ICL Pathway's non-performance and we should implement 7 option 3." 8 So the options there are continuation, give them 9 a chance, but if negotiations break down, withdrawal. 10 Is that a fair summary there? 11 A. Yes, it is. 12 Q. Thank you. Then below that at 1.8 it says: 13 "DSS recommend a similar approach with ICL, but 14 based on option 2." 15 So that's option 2 was getting rid of the benefit 16 card part of the programme? 17 A. Continuing with the project but dropping the Benefit 18 Payment Card, yes. 19 Q. Then it says: 20 "Treasury Social Security team prefers option 3." 21 Option 3 being cancellation. 22 Now, DSS and Treasury Social Security team are quite 23 significant parts of this working group, in that -- who 24 formed the working group? It was the DTI -- 25 A. No, it was set up by the Treasury. 14 1 Q. Yes. Yes, but it was the DSS and the Treasury Social 2 Security team, and the DTI were presumably members of -- 3 A. Oh, yes. 4 Q. -- the group? 5 A. Oh, yes. 6 Q. Yes. So in other words, translating 1.8, the DTI agrees 7 with continuation, the Treasury, save for their social 8 security team, agree with continuation -- you're not so 9 sure? 10 A. I think the Treasury probably, at this point, were 11 wavering. They appeared to start off being quite firmly 12 opposed to the project and seeing, understandably, the 13 benefits from ACT. I think they moved a bit, probably 14 with the personnel involved, to, I thought, 15 understanding the problems that cancellation would 16 cause, politically, both in terms of the network of Post 17 Offices but also the harm to the country's industrial 18 strategy, in particular the damage, possibly even 19 bringing collapse of ICL, Fujitsu being a major inward 20 investor, and so on, and I think those issues started to 21 weigh a bit more heavily in some Treasury minds at 22 least. 23 Q. So, at this stage, who is it that really supports 24 option 1? That's the recommendation of the group, but 25 it seems as though, even at this stage, the DSS 15 1 certainly don't and the Treasury Social Security team 2 don't. 3 A. Well, I think -- I think the group didn't succeed, if 4 indeed it was even trying to do that, in persuading 5 certain members of the group that continuation was not 6 the best way forward. And this was the issue that was 7 then -- "fought over" is perhaps putting it too 8 strongly -- for quite a few months after that. 9 Q. The section on technical viability is on page 12, and 10 can we have a look at that, please. If we could scroll 11 down to the second half of that page, thank you. So it 12 says: 13 "The Panel has concluded ..." 14 Is that the independent panel that you mentioned 15 before with Adrian Montague, or is that -- 16 A. Yes, I think that's what it's referring to. 17 Q. "The Panel has concluded that the project is technically 18 viable, although there are some risks, in particular 19 around: 20 "scalability and robustness. The programme is 21 probably the biggest of its kind and the system has had 22 to be tested at the level of its component parts. The 23 Panel is satisfied these risks are being well managed by 24 ICL Pathway, but they nevertheless remain. 25 "the system is (necessarily) heavily dependent on 16 1 a third party 'middleware' product called 'Riposte'. 2 ICL Pathway have taken steps to cover their dependency 3 on the project." 4 So that's where Riposte is mentioned. 5 A. Yes, it is. It is, and I have to confess that I don't 6 remember Riposte being such a major part of the project 7 at this stage. 8 As I said, I know it became a major issue when the 9 Benefit Payment Card part of the system had to be -- had 10 to be removed and replaced with something else, but what 11 we did know, I think, from Adrian Montague's report, was 12 that Riposte had been used by a number of other Post 13 Offices, I think, and certainly other applications. 14 So they had reason to believe that it would be fit 15 for purpose. I know that later on there was discussion 16 way above my head, in terms of technical understanding, 17 about Riposte and web-based Riposte and whether, if they 18 adopted that, it would negate some of the features of 19 Windows NT, which had hitherto been relied on to carry 20 out certain functions, and so on. So I think it was not 21 without some technical controversy. 22 Q. If we look below that: 23 "The Panel also believes that the basic 24 infrastructure is very robust for the future and is 25 generally based on industry standard products. It 17 1 should therefore allow POCL to compete for new business 2 in a variety of markets, and for example develop new 3 applications based on smart cards ... 4 "The Panel has seen no evidence to suggest that the 5 systems being developed by BA and POCL to connect up to 6 the systems being developed by Pathway will not work as 7 required." 8 Let's look at the report itself. So that section is 9 on "Technical viability", and that's the heading there, 10 and what it's highlighted is some risks around 11 robustness, that's that first bullet point, and also 12 some concerns about the use of Riposte, albeit at 3.1.5 13 it says "the basic infrastructure is very robust"? 14 A. I think it may well be that, if Riposte was a part of it 15 at that stage, it was just, as it were, a standard 16 industry application that was working perfectly well. 17 Again, my limited understanding was that the Benefit 18 Payment Card was operated within the system on a kind of 19 batch basis, so that the information would be -- from it 20 would be collected up at the end of the day and 21 presumably consolidated, and then fired off back to the 22 Benefits Agency or whoever, over lines that were leased 23 or paid for much more cheaply than if you had the thing 24 online all the time during the day. 25 I suspect that if we could think back 20 or 18 1 25 years, all of that would make a lot more sense than 2 it perhaps would seem to today, when -- but maybe you 3 wouldn't even consider these as problems. 4 Q. But you would accept that what's in front of you is at 5 least some concerns about robustness and dependency on 6 Riposte? 7 A. Yes, I would absolutely accept that. 8 Q. Perhaps we should look at the Montague report itself, 9 which may assist to develop that further. Can we look 10 at POL00028094, please. So this is the report. Can you 11 tell us a little bit about the other author -- in fact, 12 all three authors? Who was Adrian Montague? 13 A. Adrian Montague was the person who was in charge of the 14 Government's PFI initiative. 15 Q. And Bill Robins? 16 A. I don't know. 17 Q. I think he was head of the Northern Ireland Social 18 Security Agency. 19 A. Okay. 20 Q. And Alec Wylie? 21 A. Again, I don't know. 22 Q. He was Director General of Communications and 23 Information Services at the Ministry of Defence. Does 24 that help your recollection or not? 25 A. Not really, I'm afraid, no. 19 1 Q. Do you know how they were selected for that particular 2 project? 3 A. How Adrian Montague was? 4 Q. Adrian Montague or all three of them. 5 A. Well, I imagine Adrian Montague himself recruited those 6 two people, I don't know that. 7 Q. We'll be hearing from -- 8 A. Yes. 9 Q. -- Sir Adrian. 10 A. Okay. Why Adrian Montague? Well, this was -- Horizon 11 was at that point a massive PFI scheme, so it's not 12 surprising, perhaps, that he was selected for that task. 13 Q. He wasn't himself a technical expert, though? 14 A. I believe not. 15 Q. No. You've said in your statement that ministers and 16 officials were effectively reliant on these experts to 17 inform them of technical issues. 18 A. Yes. 19 Q. Was there a standing body of technical advisers within 20 Government dealing with Horizon or was it -- 21 A. No. 22 Q. -- reliant on reports such as this? 23 A. It was reliant on reports such as this. What I would 24 say is that I know that the Post Office itself had an IT 25 department headed by -- I can't remember his name but 20 1 I believe he was very well respected within the IT 2 industry, and I'm sure the Benefits Agency, likewise, 3 would have had their own technical expert. So it's not 4 that the project lacked technical people looking at it, 5 and evaluating it. 6 Q. Was Government therefore dependent on what they were 7 told by the Post Office? 8 A. Dependent what we were told by the technical 9 subcommittee. They were -- they were there to provide 10 technical expertise to all the government parties 11 involved equally, rather than each of us trying to set 12 up our own expert and have experts layered on experts 13 layered on experts, and so on. 14 Q. Can we look at page 3 of this document, which provides 15 the "Executive Summary". The third bullet point under 16 "Background" reads as follows: 17 "In the light of concerns over progress, this Panel, 18 chaired by the head of the Treasury Task Force on 19 Private Finance, was set up to make an independent 20 assessment of whether the programme was technically 21 viable, if so how quickly it could be completed and at 22 what cost." 23 Is that your recollection of its purpose? 24 A. Yes, it is. 25 Q. The issue, it seems, from the third bullet point, is one 21 1 of viability rather than, for example, reliability. 2 Would you agree with that, at that stage, in 1998? 3 A. Yes, given that that was a finding of the technical 4 committee. 5 Q. Yes. But the focus, the word that we've used, we've 6 heard, both from the overall working group report and 7 from this report, is one of "viability". 8 A. I think what was meant by that was technical viability. 9 Q. Yes. 10 A. There were financing issues, of course, and they come up 11 a bit later on in all of this but I think at this point 12 one is talking about technical viability: will the 13 system work? Will it do what it's supposed to do? 14 Q. Peter Copping of PA Consulting has given evidence to 15 this Inquiry, and he has described the task that he was 16 asked to carry out for this expert panel as "calibrating 17 the art of the possible", rather than looking at, for 18 example, technical faults and defects. Would you agree 19 with that? 20 A. Well, that was his view. I mean, I have no real basis 21 for challenging that but what we -- what Government 22 needed was advice on whether this thing can be made to 23 work and do the job that it was -- that it was set up to 24 do. 25 Q. Would you accept that it wasn't a report that addressed, 22 1 at a very detailed level, issues such as technical 2 faults and reliability of the system? 3 A. When I read the report of the Adrian Montague group, it 4 looks to me as though it does go into quite a lot of 5 technical detail, and I imagine they drew on whatever 6 they needed to draw on to come up with that. 7 I think it was quite reassuring and I don't think it 8 was kind of deliberately slanted because that's what 9 ministers or anyone else wanted to hear. 10 Q. Perhaps we can look at the "Findings". If we look at 11 the second finding there, the second bullet point, it 12 says: 13 "Our view is that the programme is technically 14 viable. There must be some risk around scalability and 15 robustness because the system has had to be tested at 16 the level of component parts, but we are satisfied these 17 risks are being well managed by Pathway." 18 So their view being expressed there is that the 19 programme itself is technically viable, albeit there are 20 risks with regard to scalability and robustness. Is 21 that a fair reading of that finding? 22 A. Yes, I think -- I think that is exactly right. If there 23 was -- if there were issues around testing, and those 24 later on as well, I think it was around the fact that 25 this is an immense project, in terms of the numbers -- 23 1 20,000 post offices, 40,000 counter positions, if 2 I vaguely remember the numbers -- and I would have 3 thought anything that you tried to put together, you can 4 test in a much more limited environment, but you don't 5 know what's going to happen when you roll it out at that 6 sort of -- at that sort of scale. 7 And I wonder whether the testing that was done at 8 the end, after the Benefit Payment Card had gone from 9 the system, whether that was sort of tested at 10 sufficient scale, I don't know. I seem to remember 11 seeing bits of paper around the place where people were 12 expressing concerns that it hadn't been. 13 I think some of these pieces of paper came from the 14 National Federation of SubPostmasters, where members who 15 had early experience of the system were finding all 16 sorts of bugs in it that they were reporting upwards. 17 I don't think it's surprising that there were such 18 bugs. My understanding of the way these huge systems 19 work, whether they're the bank's systems or government 20 systems or anything else, you know, there will be bugs, 21 and the issue is whether they can be identified quickly 22 enough and whether they can be put right quickly enough. 23 Q. Is that kind of analysis, I think your view is that that 24 should take place towards the rollout stage or later 25 down the line than, for example, 1998? 24 1 A. I think, I think what I'm saying is that you need to 2 test the system at all stages but when it's kind of 3 complete, when you're about to push the button and roll 4 it out to all of these post offices, you do want to make 5 sure, I would suggest, that you have tried to test the 6 thing at scale as thoroughly as you possibly can. 7 Q. Looking again at that second finding, where they say 8 "There must be some risk around scalability and 9 robustness" -- 10 A. Yes. 11 Q. -- would it be fair to say that the expert report wasn't 12 finding the Horizon system at that stage as robust, that 13 wasn't a finding that they were making? I will out of 14 fairness take you to the next paragraph, which does 15 say -- I'll read that: 16 "There is good evidence of future proofing at all 17 levels. The basic infrastructure is very robust for the 18 future and, in the main, industry standard products have 19 been used. The system who allow POCL to compete for new 20 business in a variety of markets, including banking and 21 financial services. New applications based on smartcard 22 technology should be relatively straightforward and 23 economic. If online applications are required, they may 24 take longer and require more investment." 25 So looking at those two paragraphs -- 25 1 A. Yeah. 2 Q. -- is a fair summary that, in terms of the actual 3 working of the Horizon system, there are certainly risks 4 around robustness, the basic infrastructure itself is 5 robust, very robust for the future, but there are 6 undoubtedly risks with regards to scalability and 7 robustness of, for example, the software that it would 8 use? 9 A. Yeah, I think that's what I was probably rather 10 inadequately trying to express, that the system had been 11 tested -- I won't say in the laboratory, because 12 obviously it was a lot more than that. The system had 13 been thoroughly tested and all the bits were shown to 14 work, and so on, but, when you start rolling it out into 15 the real world, you are bound to get problems coming up 16 with it. I think the final sentence of that "If online 17 applications are required, they may take longer and 18 require more investment", and that was also very much my 19 understanding, that it's one thing to take the Benefit 20 Payment Card out of the system, though you still need to 21 test how it works, sort of, without that, but what you 22 need to replace it with is smartcard technology and 23 that -- that still had to be developed. 24 That was a bit of an un -- as I understood it, that 25 was a bit of an unknown at that time and ICL, I think, 26 1 recognised that there was actually quite a lot of 2 development work still to be done on that. 3 Q. Would you accept that those two paragraphs there are not 4 signing off Horizon as being robust? 5 A. I think it's signing it off as being robust, as far as 6 they had got. But it wasn't finished. There was more 7 to be done, and then the scalability thing was always 8 going to be -- have a big question mark over it. You 9 know, would it really work at -- I think it's something 10 like 40,000 counter positions. So you then have not 11 just issues about the system itself but about the 12 training to use it, whether it's easy to use, whether 13 some 80-year old subpostmistress in the Orkneys is going 14 to easily get to grips with that technology. And I can 15 feel for the poor lady, because I wouldn't be very good 16 at it either. 17 Q. I think your evidence is that it was saying that it was 18 sufficient at that stage but there were undoubtedly 19 risks with regards to scalability, for example? 20 A. Absolutely, and possibly not just scalability. What 21 I recall was that, at the end, in a big rush after it 22 had been decided to drop the Benefit Payment Card, the 23 system had to be adapted to work without it and perhaps, 24 as far as possible, to make provision for things to be 25 added later on, and there was then a lot of -- I don't 27 1 like to use the word "pressure" because it sounds as 2 though somebody is sort of breathing down their necks 3 and saying "Get on with this". 4 The pressure came from the political decision to 5 finally agree that the Benefits Agency could start 6 rolling out their ACT solution from 2003, which was not 7 a very long way ahead, given the time it takes for all 8 of these things to work through. 9 So that was, that was, if you like, the pressure to 10 get the system done and rolled out. Of course, it 11 wouldn't have been accepted by POCL, if they thought 12 there were major flaws with it. I remember there were 13 a couple of flaws identified at the very end, just 14 before the sign-off, and they were put right and they 15 were tested and shown to be okay, so that POCL was 16 satisfied with that, and then they wrote a rather large 17 cheque. 18 Q. We'll get to all of that probably this afternoon, but 19 I'm happy to stay with it briefly now. Is it therefore 20 your view that, ultimately, Horizon was rushed out after 21 the Benefits Agency pulled out of the project? 22 A. I don't like to say "rushed out", because that sounds 23 like it was just kind of "Get it out of here, 24 willy-nilly". I don't think it was that at all. I do 25 think there was a lot of pressure, a lot of desire on 28 1 the part of the Post Office to get this thing moving 2 because, if you sit around too long you're not going to 3 get it all sorted by the time ACT comes along, and you 4 want to get it out there and in use in order to persuade 5 banks and other organisations to start using it because, 6 apart from anything else, you're going to need 7 additional revenue streams when the amount that the 8 Benefits Agency is asking you to do is reduced and, 9 therefore, the amount of income you get from that is 10 reduced. 11 Q. Thank you. 12 Returning to the findings, can we just scroll down 13 a little bit, please. Thank you. So it goes on there: 14 "A further nine months delay to the programme is our 15 best forecast, with September 2001 for national rollout 16 completion. Critical path issues will have to be 17 resolved fast to make this possible, and the date could 18 be brought forward with commitment and goodwill on all 19 sides." 20 Perhaps that gives an indication as to the time 21 pressures that may might be involved -- 22 A. Yes, yes. 23 Q. -- in rollout, even at that stage, while the benefits 24 card system was part of the project. Do you agree with 25 that? 29 1 A. Yes, I think so. 2 Q. Moving down to the bottom of that page: 3 "Driven mainly by timetable slippage, the sponsors' 4 business cases are eroding. The direct cost of delay is 5 estimated at £180 [million], over half of which falls to 6 the sponsors. Potential savings from fraud reduction 7 would also be delayed. Pathway, on the basis of the 8 figures it has provided, would make an overall loss if 9 the contract continued on its present terms and would 10 require an extension to break even." 11 So, I mean, perhaps even in the summer of 1998 there 12 seemed to be significant time pressures to complete the 13 project? 14 A. I think that's right. 15 Q. Can we go over the page, please. 16 A. We say "time pressures". I mean, the pressures came, at 17 least in part, from the fact that all the disadvantages 18 of the situation, as it was then, were costing everybody 19 money. You know, "We can't sit around and just let this 20 thing kind of bleed to death and us bleed to death with 21 it". 22 Q. Then the report suggests a "Possible way forward", and 23 it says, for example, in the first bullet point: 24 "Although the parties did not ... agree, we believe 25 that a restructuring of the full programme could 30 1 offer ... a way forward. 2 "The restructuring would extend the use of the card 3 beyond the current contract end date. BA and POCL would 4 prepare for a rapid increase thereafter in benefit 5 payments via the banking system ... POCL could by then 6 be ready to offer a competitively priced service for 7 customers who still wished to use post offices for 8 access to cash in this new environment. Pathway would 9 be closer to recouping its investment. 10 "A second option [so this is the alternative], less 11 risky in programme management terms, would be to descope 12 the programme by stopping the Benefit Payment Card, 13 while still allowing time for BA and POCL to prepare for 14 ACT." 15 So what were the proposals being put forward by the 16 expert group at that stage? 17 A. Well, as I understand that, what they're saying was "We 18 could propose going forward, still on the basis of the 19 Benefit Payment Card, for a limited period and then to 20 have to switch over". 21 The alternative that they were proposing, I think, 22 was, well, the one of scrapping the Benefit Payment Card 23 and then trying to move forward without it, until such 24 time as an alternative can be -- can be developed. 25 Q. Can we turn to page 11, which sets out some of the 31 1 problems that had been identified by the expert panel. 2 It's paragraph 22 that I'd like to look at. 3 So it says there -- I'm going to read it just for 4 the purposes of the transcript: 5 "However there remain problems and difficulties in 6 formally signing off requirements and solutions so that 7 delivery dates can be planned and agreed. For example: 8 "there is not yet a stable baseline requirement 9 formally agreed by all parties on which plans and key 10 milestones can be agreed; 11 "the parties have yet to sign off proposals to 12 descope Release 2 ... in New Release 2 ... a partial 13 solution ready for the start of national rollout and New 14 Release 2+ ... the full solution to be available later; 15 "there is no agreed Acceptance Plan or timescale for 16 acceptance, which puts at risk the timetable for 17 contractual acceptance of the system; 18 "there is no consensus on the length of Model Office 19 testing, live trial ... and the contingency to be 20 allowed; 21 "there is no agreement on the rate of rollout, or 22 'beat rate'; 23 "there are no agreed timescales for change control 24 decisions; 25 "version 4 of the Master Plan ... has not been 32 1 signed off and there is no formal agreement about the 2 conditions for deciding that rollout has been completed; 3 and 4 "hence the dates proposed for the start of live 5 trial and rollout to all 19,000 post offices are at 6 risk." 7 Over the page, please, to paragraph 27, this is 8 again a mention of the Riposte concerns, it says: 9 "Although we believe the architecture to be viable, 10 there is a concern that the system is (necessarily) 11 heavily dependent on the third party middleware product 12 'Riposte'. This risk will persist and steps must be 13 taken to manage this risk over the operational lifetime 14 of the system (in addition to those steps already taken 15 in the development stages by ICL and Pathway). If, as 16 is confidently predicted by ICL [and I think this is 17 a point that you were making earlier] by ICL, this 18 product becomes a Postal industry standard, this risk is 19 significantly mitigated. Pathway has also taken steps 20 to cover their dependency on Riposte by holding a copy 21 of the source code and by training their staff in its 22 use." 23 Now, again, those passages that I've just mentioned, 24 there are certainly risks that are being highlighted 25 with regards to the Horizon project? 33 1 A. There certainly are, yes. 2 Q. Do you know how your Department envisaged that 3 monitoring would take place with regards to the 4 operational lifetime of Horizon? 5 A. I don't think that we had any plans to closely monitor 6 this on a day-to-day basis. Horizon had become 7 a political issue, with ministers involved, and 8 ministers were just concerned to get a solution. We 9 were not concerned to stand over the thing and try to 10 see whether each individual little bit can get properly 11 resolved. We just wanted to know whether what ministers 12 had agreed to could be delivered in time and, if it 13 couldn't, then it comes back for some decisions on what 14 we do in those new circumstances. 15 As far as we were concerned, all of these things 16 here -- and I agree there are quite a lot of them -- 17 were for the parties involved to solve, and they were 18 not saying to us that those issues could not be 19 resolved. They'd been flagged up as things that needed 20 to be put right. 21 I, again, find this reference to Riposte a little 22 confusing because, if it had always been part of the 23 system, then I'm not sure what the -- what the issue 24 was. 25 I knew that it needed to become part of the system 34 1 in order to allow the functionality that POCL wanted to 2 see for the future. 3 Q. I mean -- 4 A. If I've got that wrong, then I apologise. 5 Q. No, I mean, sticking with Riposte, we know, for example, 6 that there was a known bug identified later on, that we 7 refer to as the Callendar Square bug, 2000 and onwards. 8 Was there anyone in Government tracking these kinds 9 of issues that were highlighted in this report? 10 A. I'm not aware that there was anyone in -- certainly in 11 DTI, I don't know what DSS were doing, but not tracking 12 it on a sort of day-by-day, issue-by-issue basis. 13 Q. Because a report of this kind, it provides a snapshot in 14 time, doesn't it? 15 A. Yes, it does. 16 Q. I think you've said that you would have relied on the 17 Post Office to have followed these kinds of matters 18 forward? 19 A. Er -- 20 Q. The highly technical matters that -- 21 A. Well, I'm sure that, if highly technical matters looked 22 as though they could be showstoppers, then they would 23 have been escalated up within the Post Office. But the 24 Post Office is, as you know, a very big organisation and 25 lots of different bits and functions of it, so I don't 35 1 think the board would have seen -- seen it as its job to 2 follow these issues on a day-by-day basis to make sure 3 that they were getting resolved. Like us, they would 4 want to know whether everything is going to be okay to 5 go ahead. 6 Q. I'm going to move on to November 1998. Can we look at 7 BEIS0000181, please. Are you content? We do usually 8 take a morning break. We've started quite late today. 9 Are you happy to -- 10 A. No, I'm content to go on if that's what you wish to do. 11 Q. Please do tell me if you would like to at any stage 12 though. 13 A. Thank you. 14 Q. So let's look at this document. One thing that you've 15 also referred to in your witness statement is a KPMG 16 report addressing technical issues, and I just wanted to 17 cover that off as well. 18 Do you recognise this document? I think it's 19 an annex to a ministerial submission. 20 A. Yes. Yes. 21 Q. It talks there of the three options: option 1, 22 continuing with Horizon; option 2, continuing with the 23 project minus the benefit card; and option 3, 24 cancellation of the project. 25 Can we look at page 3, please. Option 2 is the 36 1 continuation minus the benefit card, and it says there: 2 "KPMG have confirmed that Option 2 is technically 3 and commercially feasible." 4 Again, similar to the kind of language that we heard 5 from that expert's report: feasibility, viability. Do 6 you consider those similar terms, that the experts there 7 are looking into feasibility rather than, for example, 8 reliability? 9 A. That might be a better word, yes. 10 Q. Would it be fair to say that the focus of the Government 11 at this stage, at that time, was whether the project was 12 possible rather than whether it would be, for example, 13 reliable? 14 A. Of course, we would want to see that it was -- that it 15 was reliable. It was going to be responsible for making 16 an enormous number of, if you like, state payments or 17 benefits to a great number of people living all over the 18 country, and it needed to work, it needed to be 19 reliable, it needed to be relatively easy to use. 20 Q. But we saw the task that was set for the independent 21 group was whether the system was viable. If we look at 22 some of these KPMG documents we can see there that their 23 task was whether it was feasible. 24 Do you see the difference between that and asking 25 somebody whether the system is reliable or to analyse 37 1 and assess the reliability of the system? 2 A. I think -- of course, the -- of course, financial issues 3 came into all of this, but I don't think, in the context 4 that we've discussed so far this morning, we were 5 talking about anything other than whether the system 6 could work. 7 Q. Can we look at BEIS0000179. This is another document 8 that describes the work that KPMG were carrying out. 9 I think you wrote this ministerial submission -- 10 A. Yes, looks like it. 11 Q. -- to the Secretary of State. This is 6 November 1998. 12 It's over the page, please, paragraphs 5 and 6, 13 which discuss, to some extent, KPMG's role. It says 14 there: 15 "To assist with the first strand of work, the 16 negotiations between BA/POCL and ICL, Graham Corbett, 17 Deputy Chairman of the MMC and former Finance Director 18 of EuroTunnel was appointed to chair the negotiations. 19 KPMG were appointed to assist him, particularly in 20 understanding and validating the business cases of each 21 of the contracting parties." 22 Then the next paragraph: 23 "KPMG were also asked to undertake a major piece of 24 work on the second strand of activity, that of enabling 25 value for money comparisons to be made between the three 38 1 options." 2 It may assist if I take you to one further document 3 that relates to KPMG and that is HMT00000005. This is 4 a KPMG report or interim -- I think it's a progress 5 report, and perhaps we could -- I think you're named 6 there as one of the recipients -- just scroll to the 7 final page of that, page 5. It again addresses 8 "Option 2 viability", and it says: 9 "Having examined the high level architecture of ICL 10 Pathway it would appear that Option 2 is technically 11 feasible. Indeed Pathway are actively marketing the 12 system to overseas post offices, without the Benefit 13 Payment Card." 14 Again, I think that's something that you mentioned 15 earlier, that it was being used abroad in other post 16 offices? 17 A. Riposte. 18 Q. Riposte? 19 A. Yes. 20 Q. Again, there we see reference to technical feasibility 21 and -- 22 A. Yeah. 23 Q. -- KPMG having examined the high level architecture. 24 Did you understand anybody to be producing, at that 25 stage, a report that went into the detail of, for 39 1 example, bugs and errors or technical concerns about 2 reliability? 3 A. I don't think so. Here we have moved on, apart from the 4 high level architecture bit, we are here talking about 5 financial viability of the project and whether the 6 financial attributes would -- could be made to be 7 acceptable to ICL and to the Post Office. So now, 8 I think, Mr Corbett was brought in and KPMG, to look at 9 these things, because now we're kind of moving more 10 towards "We think we know where the system is going in 11 a physical sense, now does it make commercial sense, can 12 it be made to make commercial financial sense?" 13 Q. One thing that KPMG has looked at is the high level 14 architecture -- 15 A. Yes, yes. 16 Q. -- and it said that it's technically feasible? 17 A. Yeah. Again, they believe that it's technically sound 18 and can work. 19 Q. I suppose "technically sound" may be different to 20 "technically feasible", and I think the point I'm really 21 making is: at the DTI, were you interpreting these 22 reports as signing off the Horizon system? 23 A. No. 24 Q. No? 25 A. No, I don't think so. I think we were reassured by 40 1 these reports that there was something there that could 2 be made to fulfil the functions that we needed from it. 3 I don't think at any stage we were saying "Oh, okay, 4 this is now an absolute done deal, it's a masterpiece, 5 it will work, no problems with that, we can forget about 6 that entirely". But I think these reports were giving 7 us reassurance that the thing was going in the right 8 direction and could continue to be made sound in wind 9 and limb. 10 Q. Albeit subject to the risks that we've seen identified 11 in the experts' report? 12 A. Absolutely, absolutely. 13 Q. Can we look at a document from December 1998, and that 14 is at CBO00100001_072. This is a document that I'm 15 afraid we have only sent you quite recently. Have you 16 had a chance to have a look at that? 17 A. I've sort of skimmed it, yes. 18 Q. It's a letter or a note to the Prime Minister from Geoff 19 Mulgan. Do you remember who Geoff Mulgan was? 20 A. Er ... 21 Q. I think he was a special adviser to the Prime Minister. 22 A. I think he was a special adviser to Lord Falconer. 23 Q. Thank you. Would you have seen this document at the 24 time? 25 A. I would expect to have done, yes. 41 1 Q. I'll just read the underlined part of paragraph 1: 2 "A decision now needs to be taken on whether to 3 proceed with the Horizon project." 4 Can we go over the page, please. I'm going to read 5 paragraph 4 in its entirety for the record. It says: 6 "However, the decision is not clear cut. The 7 problems that have beset this project may well continue; 8 continuation would lock the government in for 9 10-12 years to what many see as a flawed system; 10 cancellation on the other hand would enable the Post 11 Office to take advantage of newer, cheaper or more 12 flexible technology, while the DSS could move rapidly to 13 paying benefits into people's bank accounts. 14 Cancellation would also release around £2-3 [billion] 15 over the next decade to be spent in other ways to 16 support and automate the Post Office." 17 Paragraph 6: 18 "in making a judgement, the following issues are 19 paramount ..." 20 It's that first bullet point that I want to focus 21 on. It says: 22 "The virtues of the project itself: overall, 23 Horizon, now looks increasingly flawed. It is centred 24 around a technology, the Benefit Payment Card ... that 25 is both overengineered -- and very expensive -- and 42 1 likely soon to be obsolete. Indeed, ICL acknowledge 2 that the [Benefit Payment Card] will have not commercial 3 value to them at the end of the project. Although they 4 remain underdeveloped, the alternatives, which involve 5 simpler off-the-shelf banking technology, look 6 increasingly attractive, offering a route to universal 7 banking, automated post offices and better provision of 8 government information." 9 Were you aware at that time of the suggestion that 10 the technology had been overengineered? 11 A. I don't -- I don't know exactly what is meant by that. 12 I'm sure that ICL would have wanted to do everything 13 that they could, to put everything that they could into 14 the system to make sure that when it was rolled out in 15 these huge numbers that it would -- that it would all 16 work. You can look at it afterwards, I suppose, and 17 say, "Well, you didn't really need to go to quite these 18 lengths, look, it works an absolute treat, nothing ever 19 goes wrong with it, you could have cut some corners, you 20 might have had one or two things -- one or two things 21 being thrown up, but it would have been good enough". 22 I think the -- in a way, the way that I -- that 23 I read this, this whole piece, is that it would be so 24 much easier if we weren't starting from where we are 25 starting. If we could start with a clean sheet of 43 1 paper, life would be so much simpler. But the point 2 underlying it all was: but we're not starting with 3 a clean sheet of paper, we've got ICL involved in 4 a major, major project, the collapse of which would 5 be -- have serious implications, as I've said, for them, 6 for Fujitsu, for inward investment, for the Private 7 Finance Initiative, and so on, and what do you do about 8 the whole thing of the Post Office Counters Network? 9 So I understand absolutely what this is saying. My 10 reaction to it is: yeah, but we're not starting with 11 a clean sheet of paper, we've got what we've got and we 12 probably need to try to make the best of it. 13 Q. Could we go over the page, please, to paragraph 7. In 14 fact, it may be over two pages. Thank you very much. 15 It says there: 16 "Departments remain divided. Alistair Darling 17 remains strongly opposed to continuing. Ian McCartney 18 for DTI will argue strongly for accepting a deal (Peter 19 Mandelson has largely kept out of the discussions). The 20 Treasury is divided at official level, but Stephen Byers 21 will probably, on balance, want to accept the deal for 22 pragmatic reason, even though he would prefer to 23 cancel." 24 A. Yes. 25 Q. Is that an accurate reflection of the respective 44 1 positions at that stage? 2 A. I think absolutely so, yes. 3 Q. Then moving on to the next paragraph: 4 "At first glance, most of the factors point towards 5 continuation. However my view, which Lord Falconer 6 broadly shares, is that although short-term 7 considerations and expedience point strongly towards 8 making a deal, this will in the [long term] prove 9 unsatisfactory, leaving the Post Office and government 10 dependent on a hugely expensive, inflexible, 11 inappropriate and possibly unreliable system." 12 Do you know where those concerns about reliability 13 came from? 14 A. I suppose they came from some of what we've seen, that 15 there are risks attached to -- to all of this, and the 16 Benefits Agency, which is -- which was much closer to 17 the project than we were, for example, had, I think, 18 increasingly cast doubts on ICL's ability to deliver 19 that. They didn't want the project from the word go. 20 So they were always rather hostile to it and I'm sure 21 that they would have interpreted a lot of these caveats 22 that we've seen as, "Well, look, look at all this, look 23 at all this, it's going to be awful, isn't it, and, at 24 the end of the day, we would have spent a huge amount of 25 money and we won't actually have achieved our longer 45 1 term objectives with it". 2 Q. If we go to the final page, there are some handwritten 3 notes. Now, we have a statement from Sir Tony Blair now 4 and his statement can be found at WITN06080100. I'm not 5 going to bring it up on screen but that's just for the 6 record and so that it is in evidence. 7 Sorry, could we go to the page before as well, which 8 is where the handwriting starts. I can read you, he has 9 typed out this handwriting, just in case you can't read 10 it? 11 A. I think I can, in fact, read the handwriting. 12 Q. Well, I'll read you his interpretation of it. It is: 13 "I would favour Option 1 but for Geoff's statement 14 that the system itself is flawed. Surely there must be 15 a clear view on this. Speak to me on that, ie reading 16 the enclosed paper, it all focuses on the financial deal 17 but there the risks are pretty even, probably coming 18 down on the side of continuing. The real heart of it is 19 the system itself." 20 A. Yeah. 21 Q. Now, the message seems to be getting there to the Prime 22 Minister that the system itself has flaws or it is 23 flawed. Where would he be getting that information 24 from? Is that also -- I mean, I suppose that's from the 25 author of this letter -- 46 1 A. Yes. 2 Q. -- who you've said has received it likely from the DSS. 3 Is that a fair analysis of what's happened -- 4 A. Yeah, I don't -- I mean, I think it's a fair -- a fair 5 statement of where, you know, where they were coming 6 from, and I don't think I disagree with very much of 7 what's said. But all of that, in the end, had to be 8 tempered by the political damage on the other side of 9 cancellation. 10 Q. Thank you. 11 Those kinds of concerns, though, about the system 12 being flawed, don't seem to appear to be articulated in 13 DTI correspondence. Do you think that's fair, from what 14 you've seen of the submissions and the letters at the 15 time? 16 A. Well, a lot of the technical reports, and so on, were 17 appended to briefing. I don't think that we were trying 18 to frighten ministers with some of the things that could 19 go wrong. I don't think either we were implying that 20 everything is okay and all you've got to do is take 21 a decision and everything will be wonderful. And 22 I think that our ministers, like other ministers, were 23 very well aware of the pressures that were increasing, 24 almost day by day, that some decision wasn't taken and 25 something moved forward. 47 1 I mean, it had just kind of got stuck in the mud, as 2 it were, and all of that time -- I can't remember what 3 the number was -- but several million pounds a day were 4 being thrown away. 5 Q. Because of its commitment to the project and to that 6 option 1, do you think that the DTI shied away from 7 highlighting those kinds of problems at that time? 8 A. It's possible that we were misinterpreting the gravity 9 of, of some of these things. But I don't think we were 10 any more biased one way or the other than the technical 11 reports that we were receiving. 12 Q. Can we look at BEIS0000418, please. This is a letter 13 from Peter Mandelson, who was at that time the Secretary 14 of State -- 15 A. Yes. 16 Q. -- to Stephen Byers, who was the chief secretary to the 17 Treasury. Would you have drafted this kind of letter? 18 A. Probably. 19 Q. I'm going to read the first paragraph. It says -- can 20 we just scroll down slightly, thank you: 21 "I was disappointed that our meeting yesterday was 22 again unable to reach a clear decision on the way 23 forward for the BA/POCL automation project. The 24 continued uncertainty is becoming increasingly damaging 25 for all the parties concerned. As I see it, the choice 48 1 is a straightforward one: 2 "to continue with Horizon, we will need to secure 3 heads of agreement. Within two to three years Horizon 4 will have equipped the Post Office with a modern IT 5 system, capable not only of handling the Benefit Payment 6 Card, but also front-end banking facilities in 7 conjunction with the commercial banks. This will help 8 us to implement our social banking policies, and to 9 bring modern electronic government into communities 10 which other organisations simply do not reach." 11 So a positive result of Horizon being implemented, 12 or suggested quite considerable benefits from Horizon. 13 A. That such benefits could and should flow from proceeding 14 with Horizon, yes. 15 Q. Then the next bullet point, quite a bleak picture is 16 painted: 17 "or we can take a major step into the unknown, 18 delaying the modernisation of the Post Office Counters 19 network, risking the departure of existing clients and 20 virtually eliminating the possibility of attracting new 21 ones. The damage to the confidence of subpostmasters 22 and the knock-on effect of network closures will produce 23 political fallout, no matter how carefully we try to 24 handle it. The reduction in the network will reduce our 25 ability to extend social banking and modern government 49 1 into the very communities we most wish to target. Our 2 relations with Fujitsu, a major inward investor into the 3 UK over the past decade, would be severely damaged, as 4 would the credibility of PFI." 5 Those are some of the things that you've been 6 telling us about this morning. Would you accept that 7 quite a bleak picture is being painted there? 8 A. Well, I think, I think what that was -- what this letter 9 was doing was countering the letter from, was it 10 Alistair Darling or Geoff Mulgan or someone, who seemed 11 to be painting a very, very different picture from the 12 one that we were interpreting from the evidence that we 13 had. 14 Q. Can we go over the page, please: 15 "There is still some way to go to complete the 16 Horizon project, but the basic development work has been 17 thoroughly evaluated by independent experts ..." 18 Can I pause there: is that a reference to the expert 19 group that we've seen? 20 A. Yes, it is, and I think the point there is that those 21 experts were saying that the basic development work was 22 robust, because it hadn't really moved much beyond that 23 at the point that they were looking at it. So they were 24 saying that, if you like, the nuts and bolts -- or the 25 electronic equivalent of nuts and bolts -- you know, 50 1 they all look the right ones and put together in the 2 right order, in a way that would work and work reliably. 3 Q. So I'll just read the whole of that sentence. It says: 4 "... the basic development work has been thoroughly 5 evaluated by independent experts who have pronounced it 6 viable, robust and of a design which should accommodate 7 future technological developments." 8 Do you think that fairly and accurately reflected 9 what the independent group had found or do you think it 10 was painting again a slightly rosy picture? 11 A. I think so, I mean, the technical group and other people 12 accepted that the system, if it was going to do proper 13 smartcard functions, would need further development work 14 but the -- if you like, the architecture of the whole 15 thing looked as though it was suitable for those 16 adaptions, when they came along. 17 Q. Could we bring on screen, perhaps alongside this 18 document, a document we've seen, it's POL00028094, and 19 it's -- thank you. That's the report. If we could look 20 at page 3 of the report, thank you, and if we scroll 21 down slightly, thank you, so this is where it says: 22 "Our view [of] the programme is technically viable. 23 There must be some risk around scalability and 24 robustness because the system has to be tested at the 25 level of component parts." 51 1 Do you think that that is fairly reflected in the 2 statement that it is viable, robust and of a design that 3 should accommodate future technological developments? 4 A. I think it broadly is, yes. I mean, this says that -- 5 technically viable, the system has been tested at the 6 level of component parts. So the basic bits of it, 7 which is what they had at that time, they kind of signed 8 off as being fit for purpose. But it doesn't in any 9 way, shape or form say that, you know, this means that 10 the whole project, when it's completed, will be 11 absolutely fine. It says, "As far as it's got, when 12 we're looking at it, it looks good to us". 13 Q. One of the things that the Inquiry is trying to 14 establish is where this term "robust" comes from and 15 it's a phrase that we'll then see in quite a few 16 documents from the DTI. 17 A. Yes. 18 Q. Do you think that, effectively, shorthand has been used 19 there to describe a much more complex problem, or a much 20 more complex issue? 21 A. Well, I'm sure robust doesn't mean it'll bounce if you 22 drop it on the floor; it's not that kind of robustness. 23 I think what it's saying is the way that it's being 24 built and put together should mean that, in use, in the 25 way that it's likely to be used, it should stand up to 52 1 the task being asked of it. 2 Q. Do you think that the risks that were highlighted by the 3 independent group should have been highlighted in this 4 kind of correspondence, the risk to robustness that they 5 identified? 6 A. Yes, as I say, I don't think they were trying to say 7 everything is perfect, what they're saying is "We've 8 looked at it and, as far as we can see, it looks at this 9 stage to be okay". They're not saying "We're absolutely 10 certain that it'll be okay when more work has been done 11 on it". At this stage it looks -- it looks good, or 12 good enough. 13 Q. Moving on to the next paragraph, it says: 14 "I believe the only sensible choice is to proceed 15 with the Horizon project. It is the way forward which 16 offers the least commercial and technological risk." 17 Do you know where that came from, that it offers the 18 least technological risk? 19 A. No, I think that what -- what it refers to, or what it's 20 trying to refer to is, if you cancelled the project and 21 you had to start from scratch again, you would be -- 22 because you've then got nothing, you don't know whether 23 something else could be developed that would be that 24 much better. I mean, you know what you've got, you 25 don't know what you haven't got. It might be wonderful, 53 1 it might be rubbish. 2 Q. At that stage, you were aware that ICL Pathway, out of 3 the three bidders, for example, posed the highest 4 technological risk? 5 A. I didn't know at the time, I was not involved in that at 6 the time. I've subsequently learned that the parties 7 had chosen ICL Pathway because their proposal meant that 8 ICL would be carrying the greatest risk, was my 9 understanding of the -- 10 Q. I mean, do you think in December 1998 the Secretary of 11 State, or those in high levels within the DTI, would 12 have been aware that Pathway, in fact, had been found to 13 pose the highest technological risk at the procurement 14 stage? 15 A. Probably not. 16 Q. Do you think that these kind of -- 17 A. I don't know what DSS ministers would have known from 18 the Benefits Agency, but I don't think that was anything 19 that was ever brought to our attention at the time. 20 Q. Do you think that Peter Mandelson in this letter, having 21 referred to the system to be "viable, robust and of 22 a design that should accommodate future technological 23 developments", and also highlighting that it offered the 24 least technological risk, do you think that was trying 25 to spin it a little bit and sound a bit positive in 54 1 order -- 2 A. I don't think the technological -- least technological 3 risk was spinning it. I think if you were to throw away 4 Horizon and then go out into the marketplace and try to 5 get something else, you don't know what you would get. 6 So there would be a much bigger risk around that. I'm 7 not suggesting that you might not find something better, 8 but there's a risk that you would find something worse. 9 Q. It wouldn't have posed the least technological risk, 10 though. I mean, having nothing poses the least 11 technological risk, doesn't it? 12 A. It poses the least technological risk but an enormous, 13 enormous political risk. 14 Q. Yes, but do you think the phrase there "least 15 technological risk" was really trying to make the risks 16 involved in the project seem less than they were? 17 A. No, well, I hope it wasn't interpreted like that, it 18 wasn't intended to mean that. What it was intended to 19 do, as I've just said, is to say "Well, we've got 20 something that we know here -- we think, not that there 21 are no problems with it or no technological risks left, 22 but we think that it's going to be okay; if you start 23 from scratch again, that could be a greater risk". 24 Q. Before we break for lunch, can we look at CBO00000009, 25 please. Now, this is a letter or a note from Jeremy 55 1 Heywood, it is effectively the response from the Prime 2 Minister to Geoff Mulgan's note that we saw earlier. 3 A. Yes, yes. 4 Q. I'm going to read that second paragraph, it says: 5 "The Prime Minister was concerned about your view 6 that the Benefit Payment Card is over-engineered and is 7 likely soon to be obsolete. His clear preference would 8 be to avoid cancelling the project, but to go for 9 a variant of your Option 1 and Option 2. We should 10 retain the [Benefit Payment Card] but seek to ensure 11 that over time it delivers real benefits and provides 12 an effective transition path to a satisfactory long-term 13 position. If necessary the Prime Minister thinks it may 14 be sensible to give ICL a financial incentive to improve 15 the [benefit card] project in this way." 16 So this is just a few days after that letter from 17 Peter Mandelson, again being quite frank about the 18 overengineering and likely to be soon obsolete aspect of 19 the benefit card payment aspect. Was there a feeling in 20 the Department for Trade and Industry at this time that 21 those kinds of issues just shouldn't be mentioned, or 22 should be underplayed? 23 A. Sorry, I'm not quite sure I understand. 24 Q. So we have, at exactly the same period as we have that 25 Peter Mandelson letter -- 56 1 A. Yes. 2 Q. -- where again it's referring to the issues that have 3 been highlighted by Geoff Mulgan about the 4 overengineering. 5 A. Yes, yes. 6 Q. I think his letter also referred to reliability 7 problems. That seems to be quite frank about those 8 problems. 9 A. Yes. 10 Q. Was there a lack of candour about those kinds of issues 11 arising from DTI correspondence on these issues? 12 A. Er ... 13 Q. Again, were the DTI underplaying the technical concerns, 14 were they minimising them? 15 A. Certainly -- they certainly weren't trying to talk them 16 up. I'm quite certain of that. I don't think we would 17 have argued that the Benefit Payment Card, in the light 18 of developments in the marketplace since the project had 19 been started, has turned out to be the cleverest choice 20 that could have been made, but -- I come back to this 21 point again -- we've got what we've got and we need to 22 try to make the best of it. 23 So you had an option of taking the project forward, 24 I mean, if you didn't want to abandon the -- if you 25 didn't want to abandon Horizon and, as it were drop, ICL 57 1 off a very high cliff, so you want to continue with ICL, 2 so you have a choice of doing it with the Benefit 3 Payment Card where, in a sense, the longer you stick 4 with it the more you're investing in a technology that 5 you know is not really going to be fit for purpose a bit 6 further down the road, or do you drop the Benefit 7 Payment Card at this point and say "Right, we're now 8 going to have to find something quickly that will serve 9 for the future", some form of smartcard. 10 But I think we all -- we would have all agreed that 11 the Benefit Payment Card was not the best choice that 12 could have been made, in the light of experience. 13 Q. Do you think that the DTI and the DSS were so 14 fundamentally locked into their positions that, 15 effectively, the DTI was supporting Horizon at any cost? 16 A. Well, I don't know, I don't know "at any cost". 17 I certainly think that we understood the major, major 18 problems that would arise from scrapping Horizon, 19 as I said, for ICL, for the Post Office Counters 20 Network, and so on. So I think we were very solid, if 21 you like, on the right answer -- maybe wish we didn't 22 start from here, but the right answer, given where we 23 are, has got to be to stick with this thing in one form 24 or another. 25 MR BLAKE: Thank you very much. 58 1 Sir, I think that might be an appropriate time to 2 break for lunch. 3 SIR WYN WILLIAMS: All right, that's fine by me. 4 Are we on schedule to finish Mr Sibbick at 5 a reasonable time this afternoon, Mr Blake? 6 MR BLAKE: Yes, we are. I mean, we could start, perhaps, at 7 1.50 rather than 1.55. 8 SIR WYN WILLIAMS: All right, let's do that. Fine, thank 9 you. 10 MR BLAKE: Thank you very much. 11 (12.56 pm) 12 (The short adjournment) 13 (1.50 pm) 14 MR BLAKE: Good afternoon, sir. 15 SIR WYN WILLIAMS: Good afternoon. 16 MR BLAKE: Thank you very much, Mr Sibbick, before lunch we 17 were on 14 December, I'm now moving to 21 December and 18 it seems as though it was a very busy pre-Christmas 19 period in 1998. 20 A. It was. 21 Q. Let's look at CBO00100001_057, please. This is 22 a document from the Chief Secretary to the Treasury, 23 suggesting a proposed way forward. Do you remember this 24 at all? 25 A. In general terms, yes. 59 1 Q. That Chief Secretary to the Treasury, as of that date, 2 was Stephen Byers. 3 Can we go over to page 2, please, which is where we 4 find the passage on the way forward, it's on the second 5 half of that page. So his proposal for the way forward 6 was, "Stage 1 -- stabilisation": 7 "Ministers would agree by 24 December: 8 "to continue the programme with the Benefit Payment 9 Card, with a firm re-commitment to deliver the project 10 on time by all parties; 11 "to agree commercial terms with ICL on the basis of 12 the 9 December offer updated by Keith Todd's letter of 13 18 December", and various other provisions. 14 But the key one there being to continue with the 15 Benefit Payment Card, and then there being a stage 2 16 which we will see over the page. It's the bottom of the 17 page, "Stage 2": 18 "By the end of March 1999, with the commercial 19 arrangements redrawn as above, the parties will have 20 an opportunity to see whether the programme can do more 21 to deliver the Government's present policy objectives. 22 Ministers will ask the Post Office to take forward 23 discussions between the parties in the context of the 24 PPP envisaged with ICL to explore what more could be 25 done, with the agreement of all parties, to further the 60 1 following objectives", and there are various objectives 2 on the page after. 3 Essentially, what is being proposed by the Chief 4 Secretary to the Treasury at that stage is more Horizon, 5 rather than less Horizon; do you agree with that, as in 6 to use Horizon for future additional purposes? 7 A. Yes, I think it was envisaged from quite an early stage 8 that Horizon, as it was being developed at that point in 9 time, needed to be a springboard, if you like, or 10 a platform that could be developed for wider and more 11 modern purposes. 12 Q. Then perhaps we could go to BEIS0000397 and that is 13 a ministerial submission that I believe you drafted on 14 21 December, so, again, the same day as that was 15 received from the chief secretary to the Treasury. 16 Would you have received the chief secretary's document 17 in advance? Would you have seen it before others, 18 perhaps? 19 A. I think it looks from this as though, as though when 20 I drafted this, we had sight of the chief secretary's 21 note. 22 Q. Yes. So, I mean, it says there: 23 "The Chief Secretary's office has this evening 24 circulated a note ..." 25 Would this ministerial submission perhaps have been 61 1 drafted earlier in the day and bits filled in or do you 2 think you actually urgently responded in the evening of 3 21 December -- 4 A. I think it's entirely possible that I responded that 5 evening. I do remember that around this time there were 6 an awful lot of awfully late nights. 7 Q. If we look at paragraph 2, please, it says there in bold 8 and underlined: 9 "This is the outcome for which we have fought for 10 the last nine months, and you should strongly support 11 the Chief Secretary's proposal." 12 So clearly, in terms of the DTI position as at 13 21 December, it was fully in support of that proposal? 14 A. Yes. 15 Q. Can we go over the page, please, to paragraph 7. Now, 16 we saw stage 2 mentioned in the chief secretary's 17 document, and it says here: 18 "Stage 2 has clearly been devised to try to make the 19 package more palatable to DSS. At any level much beyond 20 that of a rather obvious face-saver it seems unlikely to 21 succeed. What may also help, however, is that the 22 argument has moved quite strongly against DSS and 23 Option 3 (termination of Horizon, rapid move to ACT) on 24 two fronts in recent weeks." 25 Can you tell us about where things had reached at 62 1 that stage then, and which Government departments were 2 supportive of the DTI at that stage? 3 A. I think we knew at that stage that Number 10 was broadly 4 supportive of moving in that direction, albeit with some 5 reluctance, in that they saw it, we all saw it, as 6 a less than optimum solution. 7 Q. Perhaps we can look at a letter to Number 10, and that 8 is CBO00100001_053, please. This is just a couple of 9 days later, a letter from Ian McCartney, who was then 10 Minister of State, to the Prime Minister. 11 Is this a document, a letter, that you would have 12 drafted or been involved in the drafting of? 13 A. Yes. 14 Q. I'm going to read the first two paragraphs. It says: 15 "I have become seriously concerned at our handling 16 of the decision on the future of the BA/POCL counters 17 automation project, Horizon. The Christmas break is 18 upon us, yet despite a series of meetings and several 19 rounds of correspondence, a decision remains beyond our 20 grasp. 21 "On Monday, Stephen Byers put forward a suggested 22 compromise [that's I think the one we just saw] that 23 seemed to command a broad measure of support. Certainly 24 we in DTI would have been content to sign up to it. 25 Yesterday Alistair Darling submitted a counter-proposal 63 1 which essentially revisits an option we had already 2 discarded -- namely that of continuing with the Horizon 3 infrastructure whilst dropping the Benefit Payment 4 Card ... and introducing early compulsory ACT." 5 The next paragraph says: 6 "We should be clear that the smartcard at the heart 7 of Alistair's proposal has absolutely no direct role in 8 the delivery of welfare benefits or in the early 9 introduction of front-end banking at post office 10 counters." 11 Would it be fair to say that that was quite a tense 12 period with the Treasury and those are quite strong 13 words? 14 A. I do remember drafting this and feeling at the time -- 15 and I was absolutely not alone in this -- that this kind 16 of paralysis that seemed to have been reached at 17 ministerial level was damaging everybody and costing 18 a lot of money, and not moving anything towards 19 an outcome at all, and that somehow this logjam needed 20 to be broken, and that it was very disappointing that 21 DSS was simply trying to loop the thing back into the 22 same old arguments. 23 I think the sentence about "We should be clear that 24 the smartcard ... has absolutely no direct role in the 25 delivery of welfare benefits", I think that meant at 64 1 that point in time, not that it wasn't the right way 2 forward for the longer term. But the system at that 3 stage wasn't ready for that, and we hadn't at all worked 4 out how that was all going to be -- how that was all 5 going to be done. 6 The intention had been, from the DTI's standpoint, 7 was that it would be best to continue with the Benefit 8 Payment Card until such time as the technology had sort 9 of caught up, and the Benefit Payment Card could then be 10 converted to have smart -- I mean, not the card itself, 11 because that was just a "dumb" card, but that it could 12 be replaced by a smartcard which would do the same thing 13 as the Benefit Payment Card plus a load of other things. 14 Q. 23 December 1998 is a significant date in that it was 15 the date that Stephen Byers was appointed as Secretary 16 of State at the DTI. 17 It's referring to the proposal he made whilst Chief 18 Secretary to the Treasury. Are we to read anything into 19 his appointment at the DTI in respect of Horizon and 20 whether it was effectively a safe pair of hands in terms 21 of the continuation of the Horizon project? 22 A. I think you'd probably have to ask the Prime Minister 23 that, what his motivation in moving his ministers around 24 in the way that he did at that point in time. 25 Q. Was there a consistency of approach by his appointment? 65 1 A. Well, as it happens, yes, there was. Whether that was 2 deliberate or not, I can't -- I can't tell you. 3 Q. If we go over the page, I assume that this letter was 4 written before the change in Secretary of State, because 5 it refers to -- 6 A. Yes, I believe, I believe it was, yes. 7 Q. Is that a reason why it was sent from Ian McCartney at 8 that time, because perhaps the Secretary of State 9 himself was moving? 10 A. It might have been, I honestly don't remember. 11 Q. I'm going to read the first paragraph there, so it says: 12 "We simply cannot allow ourselves the luxury of 13 continuing to avoid a decision by tabling each time some 14 new variation on which to commission further work. The 15 continuing delay and uncertainty is already causing 16 serious damage and hardship. The 18,000 subpostmasters, 17 who have collectively sunk £1 billion of their own money 18 in the business, are finding it increasingly difficult 19 to sell their businesses when they wish to retire or 20 move on. The number of such offices remaining unsold on 21 the market is unusually high. Reinforcing this, the 22 number of net closures within the network (offices which 23 have closed and for which the Post Office has been 24 unable to find replacement subpostmasters) in the seven 25 months since the beginning of April is running at some 66 1 50% above the level of previous years. Most of them are 2 those which for social reasons we least want to lose. 3 The General Secretary of the National Federation of 4 SubPostmasters is in no doubt that the largest single 5 factor behind these depressing figures is the continued 6 uncertainty about the future of the Horizon project and 7 the associated introduction of the BPC." 8 Again, that's quite a bleak picture if Horizon isn't 9 taken forward; would you agree with that? 10 A. I would ... a slight gloss on that. It's a pretty 11 depressing picture if nothing is decided. We want -- we 12 know what we wanted/decided, but it was getting to the 13 point where almost any decision that moved the thing 14 forward would have avoided this situation which is spelt 15 out here. 16 I think the level of frustration behind this that 17 you can perhaps read into it ... I may have written the 18 words but it was a widely shared sentiment at that point 19 in time. 20 Q. But the next paragraph refers to concerns about Fujitsu 21 and Japan, and that's something that I'm going to come 22 onto, but certainly the picture that's being painted 23 there for the Prime Minister is that things are going to 24 be very bad if the Horizon project isn't taken forward? 25 A. If -- again, I would say if no decision is made. The 67 1 decision we wanted, of course, was that Horizon should 2 be taken forward, for all the reasons that we've been 3 spelling out for weeks and months, but I think more than 4 anything at this point in time was "For heaven's sake, 5 can we not take a decision on moving forward?" 6 Q. Then the paragraph after, I'm just going to read the 7 first half of that paragraph, if we could scroll down 8 slightly: 9 "The proposal in Stephen Byers' letter of 10 21 December is based on a long period of intensive 11 commercial negotiation as well as a thorough technical 12 appraisal. It offers both a way forward with the lowest 13 technical and commercial risks, and the best prospects 14 of maintaining a financially viable nationwide network 15 of post offices into the future." 16 We again there see that reference to "lowest 17 technical risk". 18 There seems to be a theme in DTI correspondence over 19 this period which is at odds with some of the risks that 20 were being presented in that expert report that we saw 21 this morning. 22 Would you agree with that, or do you still maintain 23 the position you had before lunch? 24 A. I still maintain the position. Maybe the words were not 25 as well chosen as they might have been. We believed at 68 1 that stage genuinely that the Horizon project, as 2 outlined at this point in time, was going to work and 3 offered a better and more secure way forward than 4 abandoning it and starting something completely new at 5 this point in time. 6 Q. Moving to early 1999, it seems as though there were 7 quite a few developments over Christmas and the early 8 New Year. Can we look at CBO00100001_039. 9 Thank you very much. This is a letter from the 10 Private Secretary to the Prime Minister, it's addressed 11 to Rod Clark, who I believe was at the DSS at that time, 12 and it's dated 14 January 1999. 13 Do you recognise this letter, or would you have seen 14 this letter at the time? 15 A. Yeah, I will have -- I would have seen it, yes. 16 Q. There is a very bold header that says: 17 "NO FURTHER COPIES SHOULD BE MADE of this letter, 18 and it should be made available ONLY to other Ministers 19 and officials with a STRICT NEED TO KNOW of its 20 contents." 21 Is that unusual? 22 A. I have not seen it very often, if at all. 23 Q. Do you know why it would have been written? 24 A. I imagine because some very sensitive negotiations were 25 going on at the time, and what was going on at 69 1 a political level in the background was perhaps best 2 kept away from those commercial negotiations, I guess. 3 Q. The Prime Minister's position is set out there, and I'll 4 just read those two points. It says: 5 "our key objective should be to develop the Horizon 6 Project, by negotiating with ICL the earliest possible 7 move to smart cards. It will be extremely important to 8 get the Post Office to take this negotiation seriously. 9 "but at the end of the day, if this negotiation does 10 not succeed in improving upon the existing Benefit 11 Payment Card project, it would be better to accept this 12 project than to pull out of the negotiation with ICL 13 completely, with all the damage that could do." 14 Were you aware at the time of why the Prime Minister 15 was of that opinion? 16 A. I think the Prime Minister was very well aware of the 17 damage that pulling out of Horizon, pulling out of ICL, 18 I think, would do to ICL, to Fujitsu, to that sector of 19 the UK economy, to the credibility of the PFI project, 20 as well as the political fallout from subpostmasters and 21 people who liked to use sub post offices, if something 22 was leaked or made publicly available, which meant that 23 they risked losing their village Post Office or their 24 corner shop, or whatever it is. 25 I'm quite sure from talking to people that the 70 1 demonstration in March that the National Federation of 2 SubPostmasters organised back in about 1994, people were 3 joining that not because they were so opposed to the 4 idea of having their benefits paid into bank accounts, 5 though a lot of them obviously were, but more because 6 they were told that this would pose a very real threat 7 to their village shop, to their corner shop, and so on. 8 Q. We see some more involvement from the Prime Minister on 9 1 March 1999. Can we look at that, BEIS0000375. This 10 again is a letter from the Principal Private Secretary. 11 Would you have seen this at the time? 12 A. Well, at the time or very shortly thereafter, yes. 13 Q. Can we just scroll down slightly. This addresses what 14 was then called "Option 2a". I think that was a benefit 15 account that was only accessible at the Post Office plus 16 a smartcard option. Do you remember that at all? 17 A. I confess at this distance in time my recollection of 18 exactly what the various options -- I think at one time 19 there was an option A, B1, B2, B3, C -- I don't remember 20 the detail of them all. 21 Q. Are you aware of what the Prime Minister's position was 22 around this time? Did he continue to be supportive of 23 the overall project? 24 A. I think he remained supportive of not walking away from 25 ICL, in particular, and of finding some way that 71 1 satisfied the parties who were involved in all of this 2 and allowed the project to move forward, with -- with 3 ICL, I think is the key bit of that. 4 Q. How important was the Prime Minister's guidance, 5 direction, decision, view? 6 A. I would have taken it very seriously, certainly, yes. 7 Q. How important -- were you aware of, for example, the 8 Post Office taking it seriously? 9 A. Erm ... 10 Q. Were you aware of the Post Office even being aware of 11 it? 12 A. Not aware of this in terms, no. You'd have to ask them, 13 obviously, but my recollection is that the Post Office 14 were getting more and more fed up with this. They 15 clearly -- their position was different from ours, if 16 you like, in that what they wanted was to keep the 17 Benefit Payment Card, and anything that took away from 18 that increased, in their view, the risk to their 19 customer base -- and that was obviously something they 20 didn't want to see -- and not just their customer base, 21 but also the revenue that they got from the 22 Benefits Agency for delivering those services. 23 Q. To what extent were they influenced by senior 24 politicians? 25 A. The Post Office board was clearly made aware of our 72 1 Secretary of State's thinking. They were -- they would 2 also have been aware through POCL's links with the 3 Benefits Agency and through DSS, and so on, where other 4 ministers were on all of this, and what the remaining 5 risks were that it would all go against them still. But 6 I think they shared the frustration that things had got 7 stuck and they couldn't move them forward and, you know, 8 all the time things were kind of drifting away from 9 them. 10 I think, even at the point that the project, as 11 finally agreed, started to be rolled out, they may have 12 got a bit more relaxed about it once it was going. But 13 I remember sensing still quite a lot of residual 14 bitterness, a feeling that they had been shafted, if you 15 like, and hadn't got what they wanted. Nobody had got 16 exactly what they wanted. It was a compromise. 17 Q. Moving to April 1999, can we look at BEIS0000362, 18 please. This is another ministerial submission that you 19 wrote on 16 April 1999. Can we look at paragraph 2, 20 please: 21 "After an immense amount of effort by all the 22 parties concerned over the past two or three days to 23 refine and analyse the costings associated with Option B 24 (the smartcard-based post office benefit account) ..." 25 So I think that was the option that, at some stage, 73 1 was a 2a, the one that I mentioned before: 2 "... the inescapable conclusion is, looked at 3 narrowly, this option is significantly more expensive 4 than Option A (the Benefit Payment Card)." 5 Just at the end of that paragraph: 6 "This conclusion is scarcely surprising since 7 Option B involves paying ICL for the aborted Benefit 8 Payment Card and then paying on top of that for the 9 development for the Post Office benefit account and for 10 the smartcard. 11 "This led the Treasury to produce at lunchtime today 12 a draft report by officials for Ministers concluding 13 that Option B is too expensive and should be rejected by 14 Ministers; and that Option A is undeliverable -- not for 15 technical reasons but because of the 16 'dysfunctional relationships' between the contracting 17 parties -- and should also therefore be rejected. This 18 leaves only termination which should be accepted by 19 Ministers as the least bad of three thoroughly 20 unattractive options." 21 You say at the bottom there, in paragraph 4: 22 "I said immediately that DTI officials could not be 23 party to such a conclusion." 24 Again, there seems to be quite a divide in 25 Government on the way forward -- 74 1 A. Absolutely. 2 Q. -- and the DTI's position is clearly set out there, that 3 it certainly wouldn't be a party to termination? 4 A. Yes. 5 Q. Then if we go over the page, paragraph 5, you make 6 a number of points, the first is: 7 "The Prime Minister's remit was to explore 8 an alternative way forward to Option A, not to introduce 9 the already discarded option of termination." 10 If we could look at the very final bullet point: 11 "Finally, the history of Option A is indeed one of 12 dysfunctional relationships, but tame acceptance that 13 two public sector bodies would refuse to give effect to 14 a clear collective decision by Ministers is a sad basis 15 for deciding on termination, with all the damage that 16 would do." 17 Again, strongly worded -- 18 A. Oh dear, I was upset. Yes. 19 Q. Had ministerial submissions become less strongly worded 20 over the years or was this a particularly 21 strongly-worded ministerial submission, or maybe both? 22 A. Maybe a bit of both. Again, the frustration that we're 23 just going round the same loop again and not finding 24 a way through it. Nobody's disputing that some of these 25 arguments that are put on the other side had validity 75 1 but, at the end of the day, you were in a situation 2 which had generated some political imperatives and 3 trying to pretend that they didn't exist or kicking 4 against them, if you like, was just wasting everybody's 5 time. 6 Q. I'm going to move to May 1999, and can we look at 7 BEIS0000275, please. This is a note of a document of 8 11 May 1999. It's from Katherine Hathaway, I believe 9 she was a civil servant in your Department; is that 10 right? Do you remember -- 11 A. I don't remember, actually, I'm sorry, but I don't. 12 Q. In the first paragraph it mentions a meeting with 13 George Hall? 14 A. Yes. 15 Q. Do you recall George Hall? I believe he worked at ICL. 16 A. Yes. 17 Q. This is a document that the Inquiry's seen before, it 18 was put to Keith Todd, and I'll just read to you the 19 first paragraph under "Horizon". 20 A. Yes. 21 Q. It says: 22 "We spent the first 30 minutes discussing Horizon 23 during which George confirmed that he knew that 24 Ministers were split between cancelling the project and 25 option B1 (version 2) -- he also knew exactly where that 76 1 split lay, ie HMT/DSS versus the others. When asked how 2 the deadline of 23 April had been extended he admitted 3 it was only because ICL were fudging their financial 4 reporting with potentially disastrous results as far as 5 the directors were concerned." 6 Can you offer us any insight into that at all? 7 A. Into what ICL were doing? No, I can't. 8 Q. Do you remember this allegation being -- 9 A. Well, I know that what -- they were doing everything in 10 their power to keep the project alive. I also know -- 11 I don't remember the details of it very well -- that 12 Fujitsu were looking to float Horizon -- to float -- I'm 13 sorry, to float ICL and, obviously, ICL were trying to 14 make sure that nothing awful happened that would kill 15 that. 16 Q. The next paragraph: 17 "He confirmed that Keith Todd will lose his job 18 should Horizon go down and that Fujitsu will divest 19 themselves of ICL which will be broken up and the 20 Services side will probably go to Siemens along with 21 some other Fujitsu interests on the hardware side. 22 Siemens are known to be interested in acquiring 23 a services business and are already in discussion with 24 Fujitsu." 25 Did you see that as a genuine concern or 77 1 a negotiating tactic of some sort, regarding the future 2 of ICL? 3 A. Yeah. Ultimately, I think, you know, that is what -- 4 that is what could have happened. I'm not saying that 5 I thought at the time that this is imminent, this is 6 what is going to happen. But certainly I didn't 7 discount it as, if we carried on fudging this and making 8 such a mess of it, that might well be where it all ended 9 up, and particularly so, particularly so, if we cut 10 loose ICL from going forward with the project. 11 Q. The next paragraph: 12 "He also said that he was personally hawkish about 13 the legal situation and regarded DSS as having been 14 utterly duplicitous (he suggested that ICL might accuse 15 DSS of fraud -- which I somehow doubt actually). Indeed 16 he suggested that Horizon had been a great eye-opener 17 for ICL as regards to how the Government did business 18 and that ICL would think very hard about ever 19 undertaking this kind of project again. I assume that 20 this will become common knowledge around the industry 21 eventually and that PFI in IT areas will become even 22 more difficult than it already is." 23 Why were DSS seen as duplicitous? Are you able to 24 offer any insight into that? I appreciate they're not 25 your words. 78 1 A. I don't know what he meant by that. I can only guess 2 that he may have meant that a lot of DSS's claims were 3 not well-founded and that DSS were attempting to achieve 4 their objectives using information that wasn't really 5 correct or wasn't proven, or whatever. I don't know. 6 Q. This is quite a frank conversation with somebody from 7 ICL. Did ICL see the DTI as being very much on their 8 side -- 9 A. Yes. Yes, they did. And I had that from a number of 10 sources on a number of occasions, that they were 11 grateful that we were -- we were doing our best for 12 them, and they knew that we were fighting hard against 13 a very determined opposition. And I say it again: 14 I have a great deal of sympathy with the 15 Benefits Agency's stance on this. It made perfect sense 16 from their point of view. 17 Q. This is 11 May. On 24 May, POCL and ICL reached 18 an in-principle agreement, and I'm going to go to 19 28 May, that's BEIS0000355. This is another submission 20 from yourself and this is a stage where a new working 21 group was established. This is the 1999 working group 22 I think that you mentioned earlier. 23 A. That's correct, yes. 24 Q. Can you tell us briefly why that working group was set 25 up? 79 1 A. Yes. I think it was to involve parties who hadn't been 2 directly involved in other discussions, to bring them 3 together, mainly the groups of people, NFSP and CWU, and 4 so on, who were going to be using the equipment. We 5 wanted to bring them in at that stage so that they had 6 a sense of understanding where we were trying to get to 7 and how we were trying to get there. But, equally 8 importantly, we saw that group as being a very useful 9 source of feedback as to how the rollout was actually 10 going and whether problems were being quickly 11 identified, quickly reported back, quickly acted on, 12 quickly resolved. 13 Q. This submission sets out four issues. Can we look at 14 paragraph 2. The first issue there: 15 "... there are negotiations between POCL and ICL 16 that need to take place over the next 2-3 months to put 17 in place the detailed contractual arrangement that will 18 give effect to the outline agreement reached last 19 weekend." 20 Then the next paragraph outlines a second issue, if 21 we could scroll down slightly: 22 "The second issue is to ensure that the remaining 23 development phases of Horizon, including large scale 24 live trials are completed without further slippage; and 25 most crucially that the rollout of the system following 80 1 acceptance to all offices within the network is 2 accomplished in a smooth and timely fashion. The CWU 3 and NFSP members will be in the front line of the action 4 during these phases, and both organisations have much to 5 contribute to the successful completion of these 6 phases." 7 It seems that the second issue is about prompt 8 rollout of the system, no further slippage; is that 9 right? 10 A. I -- 11 Q. Is that a fair description of that second issue? 12 A. Yes, it is. 13 Q. Over the page, please, to number 4. This is the third 14 issue: 15 "The third issue is how to maximise the commercial 16 potential of the Horizon platform." 17 Then number 5 sets out the fourth issue: 18 "The fourth issue is how POCL is to be funded in the 19 medium and longer term future once some £400 million of 20 revenue from BA ... begin progressively to walk out 21 through the door from 2003." 22 At the bottom of the page, it has the suggested 23 terms of reference and, over the page, please, there are 24 the three points there. Number 1: 25 "To oversee the negotiations between POCL and ICL 81 1 which will develop the letter of agreement signed 2 between the parties on 24 May into a Codified Agreement 3 governing the contractual relationship under which the 4 project will be taken forward; and to facilitate 5 solutions to say any problems which may arise ..." 6 Second: 7 "To oversee, to contribute actively to, and to 8 facilitate solutions where problems arise, the 9 completion of the development phases of the Horizon 10 project, and in particular the smooth and timely rollout 11 of the system to all offices within the Post Office 12 network; and [three] 13 "To contribute through ideas, contacts and other 14 practical measures, to maximising the commercial 15 potential of the Horizon infrastructure, thereby to the 16 future viability of the Post Office network as a whole." 17 Is it right to say that this 1999 working group 18 wasn't a technical group to analyse technical issues? 19 A. No, not at all. No, not at all. But, except that if 20 issues arose in sub post offices when they were trying 21 to work the system for the first time, those issues 22 would be identified and reported back. Whether they 23 were technical issues or not, I don't know, and probably 24 the people who were reporting them back wouldn't 25 necessarily know. They would just know that something 82 1 wasn't working properly. 2 Looking at these words again, I don't know, I think, 3 if I was drafting it today -- which thank goodness I'm 4 not -- rather than "oversee", which implies some kind of 5 control over it, as it were, I think the concept was 6 more to keep a very close eye on and to see whether 7 there are thoughts or whatever that you can contribute 8 to the process, as it goes on. 9 Q. Who was expected to pass on through this group those 10 kinds of issues? 11 A. I'm ... 12 Q. Who would you have expected to be contributing within 13 the group -- 14 A. Yes. 15 Q. -- to problems with the completion and development -- 16 A. To be contributing within the group? I would have 17 thought, particularly, the National Federation of 18 SubPostmasters because it was primarily their members up 19 and down the country operating 40,000 or whatever it was 20 of these terminals, who would be coming up with any 21 problems that they were -- that they were finding. 22 I mean, they would know if something wasn't working 23 properly, and if it wasn't working properly it needed to 24 be reported upwards and it needed to be jumped on very, 25 very quickly. 83 1 Q. In terms of significant technical reports to government, 2 am I right in saying that it's the 1998 working group 3 that was the significant report, as far as that is 4 concerned, and that the -- sorry, the 1998 working group 5 report, and that the 1999 working group was not looking 6 at things at a technical level but was -- 7 A. Yes. 8 Q. -- as you say, receiving information from, amongst other 9 people -- 10 A. Yes. 11 Q. -- the NFSP? 12 A. At this point in time, I'm not sure how much the 13 original working group and its technical subgroup were 14 still actively involved in all of this. My 15 recollection, which may be quite wrong, is that they 16 weren't or weren't to any very great extent. I don't 17 recall them as being. What I would have expected is, if 18 some of these things that were being identified by 19 subpostmasters as not working properly, they would have 20 been, as I say, reported up to their bosses, to 21 helpdesks or whatever and, if necessary, they would have 22 been escalated upwards, the Benefits Agency is out of it 23 at this point of course, through POCL, through ICL, and 24 escalated to whatever level was necessary to authorise 25 whatever needed to be done to put it right. 84 1 Q. We see on this page the suggested membership, so we have 2 there the DTI -- 3 A. Yes. 4 Q. -- POCL, Communications Workers Union, National 5 Federation of SubPostmasters, "ICL would not formally be 6 a member, but would be invited to attend meetings as 7 appropriate". Is this group the core group for feeding 8 back those kinds of concerns that are arising during 9 that rollout period? 10 A. It was absolutely -- it was intended to be absolutely 11 a primary channel for that purpose. It doesn't mean 12 that it would be the only one, but yes, it would be 13 a primary one and it would be perhaps the -- I was going 14 to say almost the only one that would be reporting -- 15 whose reports back would have gone to ministers. 16 Q. Can we look at NFSP00000471, please. This is a report 17 of an NFSP meeting in June 1999. 18 Would you have seen these reports at the time? 19 A. Not at all, no. 20 Q. I've taken previous witnesses to these minutes and I'm 21 going to take you to the same couple of passages at 22 page 23, please, about halfway down on that page. 23 In fact, if we look at the page before, it will make 24 it clear at the bottom of that page, 9(c), the 25 discussion here is on "Counter Automation -- BA/POCL", 85 1 and this is a meeting in June 1999. 2 If we go over the page, please, about three-quarters 3 of the way down there is a paragraph: 4 "There was general discussion on the severe 5 difficulties being experienced by subpostmasters who are 6 already running an automated system. Seven sheets of 7 comments from the North East had been passed to Mr Dave 8 Miller. The difficulties and trauma being experienced 9 by some subpostmasters were giving rise to concerns for 10 their health and emotional wellbeing. It was felt by 11 some that a tragedy was not far away if something was 12 not altered soon. The software was considered to be 13 poor quality and not intended to run such a huge 14 network. The system is based on ECCO which was 15 originally written for a network of 700 -- not 15,500." 16 Over the page, please, about halfway down it says: 17 "The general secretary assured the meeting that 18 Mr David Miller had been informed of the difficulties in 19 no uncertain terms." 20 Over the page again, we have the comments from 21 Miss Lindon: 22 "Miss Lindon commented that this seemed to be 23 a typical Post Office Counters situation and felt that 24 ICL could not be blamed for the problems." 25 It says there: 86 1 "Pointing out that it is now three years since the 2 project was first mooted, which is a very long time in 3 the technological world, she suggested that a different 4 system be tried, smaller and less complicated, which 5 would be of greater benefit to the smaller offices and 6 probably be a good deal cheaper and easier to operate. 7 POCL seem to be attempting to build an audit system into 8 the project, making the whole thing far to [sic] big, 9 too cumbersome and too complicated." 10 Were you aware of these kinds of concerns at the 11 time being raised in an NFSP meeting? 12 A. No, I read the pages that you've just referred to and 13 several of the pages that follow it and I was absolutely 14 horrified. I thought it was unbelievable that, with 15 this level of concern at this stage in the rollout, this 16 very early stage in the rollout of the project, that 17 something was not done about it. 18 Going on a few pages from here, there's the report 19 of a meeting that I attended -- 20 Q. Perhaps we could go to, is it page 27 at the bottom? 21 Sorry, do carry on. I may be pointing to the wrong 22 part. 23 A. There was a bit where the Minister was reported as 24 saying -- or the implication is "I don't care what's the 25 matter with it, this thing's got to go ahead and there 87 1 should be no slippage". I don't recollect at all the 2 Minister saying that sort of thing in those terms. This 3 is very, very stark, and absolutely not what we would 4 have wanted to see. I know there was pressure to get 5 the rollout going and even accelerating, but I can't -- 6 I find it difficult to believe that this is something 7 that we would have, as a Department, as a Government, we 8 would have countenanced just turning a blind eye and 9 saying "Get on with it". 10 It didn't come out, I think, in the meeting anything 11 like as starkly as it did in the paragraphs to which you 12 referred just now, and some of the paragraphs around it, 13 which make absolutely terrifying reading, really. 14 Q. So the meeting that you attended, which I think is at 15 the bottom of page 27 -- 16 A. Yes. 17 Q. -- that was a meeting with Mr Peberdy and the general 18 secretary. You weren't present at the larger meeting, 19 which is the subject of this report; is that right? 20 A. No, no, no, I was present at the meeting of the working 21 group. 22 Q. Yes. 23 A. Not anything else. 24 Q. The kinds of concerns that we have heard in this report, 25 were those kinds of concerns in any way brought to your 88 1 attention by Mr Peberdy at that meeting on 22 June? 2 A. I don't think that we -- I'm not saying he was trying to 3 hide anything or not say that there weren't problems, 4 but I don't ever remember getting from anyone anything 5 with the flavour of those earlier paragraphs, which are 6 really -- at the beginning of something, that you're now 7 going to multiply by 100, and 1,000, and 10,000, and so 8 on, and just leave it. 9 Q. You've said that the Horizon Working Group 1999 was -- 10 A. Yes. 11 Q. -- the forum to raise those kinds of issues? 12 A. It was -- it was a forum to raise these sort of things. 13 Obviously not the only one. Obviously, if you've got 14 a problem you report it to your supervisor or you report 15 it to a helpdesk or whatever, or whatever. But this was 16 certainly a channel for feeding back those kinds of 17 discoveries, if you like, these reactions, and it was 18 the one channel, I would have said, that has direct 19 access to a minister. 20 Q. We've seen who was at that NFSP meeting, and we've heard 21 and read that comments were fed back to David Miller of 22 POCL. Who should have been passing on those messages to 23 Government or through the Horizon Working Group? 24 A. It should have come out, in my view, much more starkly 25 at the meeting of that working group than it did. 89 1 I don't think -- I don't recall and I don't think the 2 minutes really bring out that anything was put to the 3 group in quite those stark terms. I don't know, maybe 4 I'm misremembering, but I don't recall -- I don't recall 5 that. 6 I know things were raised and, "Yes, we're going to 7 talk to the Post Office about it", and so on. Not -- 8 nothing like "This is awful, if this isn't resolved this 9 is an absolute showstopper, you know, we can't roll the 10 thing out like this to all those poor subpostmasters and 11 subpostmistresses around the country and have them 12 having nervous breakdowns, and what have you, because 13 they can't make the system work". 14 Q. Who should have been shouting that message? 15 A. Well, the representatives in particular of the National 16 Federation of SubPostmasters. I'm not saying they 17 weren't doing their job. I think -- I mean, they must 18 answer to themselves and I knew them very well and they 19 were good people, but they too needed to see the system 20 rolled out and working, and I guess they thought, "Well, 21 we don't necessarily want to start ringing alarm bells 22 too loudly in case it generates pressure to stop the 23 project, or whatever it is. You know, these are 24 concerns, they must be raised with the Post Office, they 25 must be sorted. But let's not overreact, let's not ... 90 1 let's not frighten everybody". 2 Q. Moving -- 3 A. That's what I think. I mean, I don't -- you have to ask 4 them about it. But you've got the two documents there, 5 you've got the one which records what's being said about 6 the awful situation which some subpostmasters were 7 finding themselves in, and then you've got the minutes 8 of the meeting, which I suggest don't quite reflect 9 fully the flavour of -- of that. 10 Q. Moving to July 1999, can we look at BEIS0000236, please. 11 This is another ministerial submission from yourself. 12 If we scroll down, please -- thank you: 13 "The detailed agreement between Post Office Counters 14 Limited ... and ICL for taking forward the restructured 15 Horizon project was signed by the parties this morning." 16 Over the page, please, to paragraph 5: 17 "After a slow and resentful start, POCL have I think 18 surprised themselves at the progress that they have been 19 able to make with ICL, both in contractual discussions 20 and in resolving a large number of outstanding technical 21 issues. It is very early days yet, and at risk of 22 accusations of wishful thinking, I nevertheless detect 23 in this early progress perhaps some vindication of 24 Ministers' decision to simplify the contractual 25 relationship by taking the Benefits Agency out of the 91 1 frame, and to simplify the technical content of the 2 project by removing the Benefit Payment Card." 3 Would it be fair to say that you were cautiously 4 optimistic at that stage? 5 A. Yes, I was, and I think what I was particularly 6 optimistic about was that what caused the original 7 Horizon project, if I can put it like that, to fail more 8 than anything, I think, was that the Benefits Agency and 9 POCL -- I'm not attaching blame to either side -- just 10 couldn't get on, and ICL, who probably had their faults 11 as well, were stuck in the middle of this. And it's -- 12 I think it was probably more than anything that that 13 caused the delays and eventually the, sort of, collapse 14 of Horizon Mark I, if I can call it that. 15 Q. Paragraph 6: 16 "By the same token, however, if POCL are at least 17 beginning to lift themselves out of the depression into 18 which they sank as a result of the decision to allow BA 19 to move to ACT from 2003, it will now be very important 20 to them for Ministers to reach a compromise that at 21 least gives them some modest gains over what is 22 currently on offer in their negotiations with BA." 23 A bit further down: 24 "They have a mountain to climb if the hugely 25 ambitious programme for rollout of the system to 40,000 92 1 counter positions in nearly 19,000 offices is to be 2 completed on schedule, and if they are to drive forward 3 in a positive and optimistic spirit the search for new 4 business to help plug the £400 million a year hole in 5 their finances that the loss of BA revenue will create. 6 A positive sign from Ministers now stands to produce 7 benefits well beyond its modest cost; a negative sign 8 could not fail to damage the healing process." 9 I think the "state of depression" is something that 10 you mentioned earlier. Can you give us a flavour of 11 that, please? 12 A. I just think that, at the end of all these negotiations 13 and before they'd actually got anything on the ground to 14 show for it, POCL felt that they'd been shafted. They 15 felt that they'd been kind of cut adrift with no Benefit 16 Payment Card and a system that they didn't know what 17 they could do with. 18 The negotiations with ICL for Horizon Mark I had 19 been so fractious and unproductive that they probably 20 feared that the same thing might happen with Horizon 21 Mark II, but I think they found -- or at least the 22 people who were dealing with ICL found -- that when it 23 was just a one on one and there wasn't this ridiculous 24 sort of triangular relationship, that they could 25 actually make progress and get things done, and I think 93 1 they started to feel an awful lot more cheerful. 2 That was the sense that I got from the people that 3 I spoke to but perhaps I was speaking to the wrong 4 people. 5 Q. Well, we've seen an ICL monthly report, I'm not going to 6 take you to it, I think you have been provided with it 7 recently. Its reference, for the purpose of the 8 transcript, is FUJ00058183. That's a report of 9 June 1999 and it says that POCL continued to remain 10 negative and critical towards the programme and hadn't 11 got over their bitterness in the way in which they'd 12 been treated by the public sector and that they held ICL 13 partly responsible. Is that something you would agree 14 with? 15 A. Well, it's obviously somebody's interpretation of what 16 they feel the mood in POCL to be. You know, I've said 17 that there was a great deal of bitterness around the 18 start of the rollout programme. I was certainly aware 19 of that. But the clouds seemed to be lifting, it seemed 20 to be looking an awful lot more positive, but ... 21 Q. Did you sense any mistrust between the Post Office and 22 ICL at that stage? 23 A. No, I don't think I did. I think probably POCL, as the 24 thing went along, were kind of half expecting nasty 25 surprises, because of the bad experience that they'd had 94 1 before, but I thought -- and after the end of -- or 2 towards the end of 2000, I wasn't there any more, I'd 3 retired, so I don't know how it progressed after that 4 point. But when I left it, I thought there was 5 a greater sense of optimism than there had been up until 6 that point. 7 Q. The second half of this paragraph describes the rollout 8 as "hugely ambitious". Do you think it was too 9 ambitious in respect of its timeframe, for example? 10 A. I'm sorry, what are we looking at now? 11 Q. So: 12 "They have a mountain to climb if the hugely 13 ambitious programme for rollout of the system to 40,000 14 counter positions and nearly 19,000 offices is to be 15 completed on schedule ..." 16 A. Yeah. 17 Q. Looking back at it, or perhaps even your view at the 18 time, do you or did you consider that it was too 19 optimistic and perhaps rushed? 20 A. I remember thinking that this was a huge programme to 21 try to manage, to, you know, to get 40,000 terminals in 22 and working and the people trained to use it properly 23 and the helpdesk set up and in place, and all that. 24 Yes, enormously ambitious. I don't think I had a sense 25 that, "Oh, this can't possibly work, can it?" But just 95 1 sort of looking with some awe at what they were planning 2 to achieve. 3 Q. The Benefits Agency had pulled out in May 1999, the plan 4 was to roll out pretty soon. Do you think the time 5 period that was left for, for example, all of that 6 testing that you talked about at the very beginning of 7 today, do you think there was sufficient time for all of 8 that? 9 A. I suppose, do I or did I? I don't know, I don't know at 10 the time how conscious I was of all of that. Looking at 11 it now, my understanding is that, once they had ripped 12 the guts out of Horizon, that is the removal of the 13 Benefit Payment Card part of the technology, the amount 14 of testing that was done thereafter to see whether the 15 residual bits were working properly, there was very 16 little time. I'm not saying that the testing wasn't 17 adequate. There was very little time, and I wonder 18 whether there was enough time for all of the testing 19 that should have been done, could have been done, would 20 have been desirable to do, was actually done. 21 But it's very easy always in these circumstances to 22 say "Well, in an ideal world we'd have spent another 23 six months testing this" but it wasn't -- it was 24 a somewhat less than optimum world, if you like, that 25 they were trying to do all of this in. 96 1 Q. We know from this morning that in 1998 there was this 2 thorough look into the viability of the Horizon system. 3 Do you think that that summer of 1999 would have been 4 an appropriate time to have carried out another 5 government look at the reliability, this time, of the 6 Horizon system? 7 A. Yeah, in retrospect, perhaps it would. I don't know 8 what the technical experts would have felt at the time, 9 whether it was worth going back and having another 10 sort of look at the roots, as it were, to see how they 11 were growing. Clearly, by the time it was rolled out, 12 quite a lot of time had elapsed and we know that, in 13 that time, it wasn't -- it wasn't visible to us. But we 14 knew that work was continuing, even though ICL had been 15 put in breach back in the November, or whatever it was, 16 and they must have known that that was coming down the 17 road at them anyway. 18 My understanding is that they carried on working and 19 carried on with developments, in the hope, certainly, 20 that when ministers finally sort of say got their act 21 together -- that sounds terribly patronising, I don't 22 mean that -- that once decisions had been reached they 23 would be that much further forward and be able to move 24 the thing on. 25 Everyone, I think, was conscious of time in all of 97 1 this. The Benefits Agency, very conscious that until 2 something was sorted the losses from fraud on the 3 paper-based system would continue unabated, so they too 4 needed this thing to be, to be settled. 5 Q. Can we look at a document from August 1999, that's 6 BEIS0000119, please. This is lines to take that had 7 been prepared for a meeting with Derek Hodgson. Do you 8 recall this document or the background to this at all? 9 A. I think I probably do, yes. 10 Q. It doesn't matter so much because I'm only really going 11 to take you to the background section but if there is 12 anything that you recall -- 13 A. I was just going to say about what's on the screen at 14 the moment, "in two respects the system was not yet 15 suitable for acceptance and declined to do so". What we 16 understood, that there were screen freezes and something 17 at the -- 18 Q. Perhaps we could scroll down, actually, as that's where 19 the two problem areas are highlighted. It says: 20 "The two problem areas which led POCL to refuse 21 acceptance on 18 August are: 22 "(a) some connection difficulties resulting in 23 a small number of counter transactions not being 24 properly recorded in the 'back end' systems; 25 "(b) higher than expected incidence of 'screen 98 1 freeze'. 2 "ICL confident that these are just glitches for 3 which they have an effective fix. POCL want 4 a demonstration of this by 2-3 weeks clear running 5 before accepting the system [because] (acceptance 6 triggers a [significant payout to ICL, £68 million]." 7 Where would that information have come from, do you 8 know? 9 A. I think we would have been -- we would have been told of 10 that by the Post Office, I assume. I don't know -- 11 Q. Is that the kind of technical detail that you would be 12 provided with at that stage, that limited detail, or do 13 you think -- 14 A. I don't think we would have been provided with that 15 sort of level of detail at all if it hadn't suddenly 16 cropped up as a potentially serious means of delay to 17 the programme. In the event, it wasn't and POCL 18 apparently had satisfied themselves that the patches 19 that had been put in place were good and were holding, 20 and so they went ahead with it. 21 But no, we wouldn't normally have had sight of that 22 level of detail, if it hadn't been for the context of 23 it. I believe there was something else around at the 24 time concerning helpdesks that had to be resolved as 25 well, that POCL were unhappy about. But, other than 99 1 that, at that point in time, we had no reason to believe 2 that POCL were anything -- and ICL -- were anything 3 other than satisfied that the system was in 4 a sufficiently good shape for it to be rolled out. 5 Q. You have mentioned the helpdesk. Perhaps I can take you 6 to another document that shows that, BEIS0000264. 7 A. Yes. 8 Q. This is just a front sheet. In fact, it's the annex to 9 that that I'd like to take you to, but that has 10 a separate document reference number, that is 11 BEIS0000265, and these are "Points to make". It's for 12 a meeting between Patricia Hewitt and ICL on 13 2 November 1999. 14 It says there: 15 "I was pleased to hear that the Post Office formally 16 accepted the Horizon automation system on 24 September 17 although I understand that this was delayed by one 18 month." 19 If we look over the page, it has some background 20 there: 21 "Formal acceptance of the reconfigured Horizon 22 system planned for 18 August was postponed because of 23 Post Office Counters' concerns about training, system 24 stability, data integrity (there had been 25 an unacceptably high level of screen freezes) and the 100 1 effective operation of the helpdesk. The Post Office 2 accepted the system on 24 September on the basis that 3 effective remedial action had either been completed or 4 was in hand. 5 "Currently 950 post offices are automated and the 6 Post Office expect that this will rise to 1,800 before 7 Christmas. The Post Office is committed to achieving 8 rollout of the system by March 2001 at the rate of 300 9 offices per week." 10 So, again, there there's mention in very summary 11 form of the kinds of issues that are being experienced 12 on the Horizon system. 13 Having seen those NFSP meeting minutes, for example, 14 do you feel that, at that time, so August, September, 15 October 1999, you were being provided with a frank 16 assessment of the rollout success or otherwise? 17 A. I have no reason to believe that the Post Office, that 18 POCL, would have wanted to cover up any problems that 19 there were. I can't believe that they would have wanted 20 to sign a cheque for £68 million, or whatever it was, 21 for the system if they weren't satisfied -- well, in the 22 words here, "that effective remedial action had either 23 been completed or was in hand". 24 I think -- again, I've said before, a system of this 25 size and this complexity was bound to have teething 101 1 problems showing up and the issue was: how quickly can 2 you jump on these, how quickly can you resolve them and 3 put corrections in place? 4 I don't know what the answer to that was because, as 5 I say, I wasn't there for very long after that. 6 Q. But do you think at that time you were being provided 7 with sufficient information about how it was going? 8 A. I don't think we would have been provided with a lot of 9 detailed information at all. We would have been 10 provided with information about the bigger issues and 11 told when they were -- and told when they had been 12 resolved. I'm not sure how much of this also came up 13 through the Horizon group 1999. 14 Q. I'm going to take you to one more document before we 15 have a short break, and then I've just got some 16 individual issues to take you to that shouldn't last too 17 long. 18 Let's look at this further document, it's 19 BEIS0000478. You produced some briefings and lines to 20 take in autumn of 1999. I think a few different ones 21 have been provided to you in advance of today, but it's 22 this one that I will take you to. 23 These are lines to take in respect of the Trade and 24 Industry Select Committee report. Do you remember in 25 summary what that report said or what that report was 102 1 about? 2 A. I don't remember that particular document. 3 Q. Perhaps we could scroll down slightly. 4 A. Yeah. I think this was -- I mean, this was just lines 5 to take on what the select committee had reported -- 6 Q. Yes. 7 A. -- on. 8 Q. Perhaps if we keep on scrolling -- 9 A. Yes. 10 Q. -- we can see where it goes to the Q&A brief. 11 A. Yes. 12 Q. Over the page, please. 13 A. Yeah. 14 Q. It addresses on page 5 -- 15 A. Right. 16 Q. -- halfway down -- these are simply, I think, press 17 lines or lines to take, I think the -- in fact, the 18 covering document said "To: Ms Madson News", so 19 presumably that's the newsdesk of the DTI, is it? 20 A. Yes. 21 Q. Yes. It says there: 22 "Given the failures in this project, why didn't the 23 Government just terminate and find another supplier?" 24 The line to take there is: 25 "The Government is satisfied that the agreement has 103 1 put the project back on track, this will deliver 2 a modern online IT platform to some 40,000 counter 3 positions in more than 18,000 post offices and 4 represents the best value for money for the taxpayer." 5 There are lots, if we keep on scrolling, of subjects 6 but technical robustness is only addressed at page 13. 7 Over to page 13, please, thank you. It's the second 8 point there, it says: 9 "IT consultants appointed by the Government's review 10 found the system to be technically robust and 11 deliverable?" 12 So this is a point, an anticipated point, that 13 somebody may put to the DTI? 14 A. Yes. Yes. 15 Q. The response is: 16 "They also found that its deliverability wasn't 17 dependent on a number of other factors. Given the 18 possibility of further delays with the project, and the 19 lack of assurances on cost and timetable, the Government 20 decided that a restructuring of the project had become 21 essential." 22 That reference there to IT consultants having found 23 the system to be technically robust and deliverable, 24 that seems again to be a reference back to the 1998 25 report and the DTI's interpretation of that report. Do 104 1 you -- 2 A. I think it must have been earlier than that. I think -- 3 my recollection is that the original Horizon, Horizon 4 Mark I, if you like, which was -- which was close to 5 completion, although it wasn't allowed to go ahead and 6 get properly completed, the technical experts looked at 7 that and said, in effect "If this had been allowed to go 8 ahead we think it would have worked and worked well", is 9 my recollection. 10 Q. So you don't think that that is a reference to, 11 for example, the PA Consulting report back in 1998 that 12 fed into the Horizon Working Group report? 13 A. I don't think specifically it was a PA report, no. 14 Q. No. 15 A. I mean, various -- at that stage, various people were 16 looking at various things. 17 Q. What there hadn't been, by this stage, though, so 18 1999 -- 19 A. Yeah. 20 Q. -- was an up-to-date consultant report that found the 21 system, as then being rolled out, to be technically 22 robust and deliverable? 23 A. I agree with that, yes, that's true. 24 Q. Do you think that that was perhaps a missed opportunity? 25 A. It might have been. I'm sure that the reason for it was 105 1 the pressure now to simply get this thing on the road 2 and going. In retrospect, it's possible to say that, if 3 there had been more time, things could have been done 4 differently and perhaps better. But, at the time, and 5 in the circumstances, that was the way that it was 6 handled, as I say, possibly less than optimum. 7 But it was always the time pressure, and this is not 8 just the time pressure from the Post Office or the DTI; 9 it was the time pressure from everybody, you know. The 10 Benefits Agency Agency wanted the thing moved forward so 11 that they could get closer to the point when they could 12 plug the leak of fraudulent money disappearing from the 13 system that they had to account for. So everybody 14 wanted a solution. 15 MR BLAKE: Thank you very much. 16 Sir, I think that's a convenient moment to take 17 perhaps a ten-minute break. 18 SIR WYN WILLIAMS: Yes, by all means. 19 MR BLAKE: Thank you very much. 20 SIR WYN WILLIAMS: Will that be 3.25? 21 MR BLAKE: Yes, thank you very much. 22 SIR WYN WILLIAMS: Fine. Thank you. 23 (3.14 pm) 24 (A short break) 25 (3.25 pm) 106 1 MR BLAKE: Hello, sir. 2 SIR WYN WILLIAMS: Hi there. 3 MR BLAKE: Mr Sibbick, I'm going to ask you about a few 4 discrete issues. I shouldn't be very long. I can't 5 imagine I'll be longer than half an hour. 6 A. As far as I'm concerned, as long as you need. 7 Q. Thank you. 8 The first one is Japan. There are many documents 9 that you have been provided with that address the 10 relations with Fujitsu and wider relations, commercial 11 relations. For the record, for example, they include 12 BEIS0000127, BEIS0000281, BEIS0000127, BEIS0000421. 13 Those are all from September and October 1998. 14 I'm going to start with December 1998, and can we 15 look at BEIS0000336, please. There's BEIS0000334 which 16 is just a covering telegram, and I'll skip over that. 17 Do you remember seeing this, at least in 18 preparation? If not, we can take a bit more time over 19 it. It's a telegram -- 20 A. Yeah. 21 Q. -- from Tokyo, from the British Embassy. 22 (Pause) 23 It may assist -- sorry, could we go to BEIS0000334, 24 please. So this is from Isabel Anderson, who I believe 25 worked for you or with you? 107 1 A. With me, yes, yes. 2 Q. I think you are, yes, copied in there. This is 3 attaching certain documents including a note of 4 a meeting between Sir David Wright, Ambassador in Japan, 5 and Mr Naruto, Vice Chairman of Fujitsu and Chairman of 6 ICL UK, and Mr Sakai, Senior Vice President of Fujitsu, 7 and Mr Yurino, director of ICL UK. 8 Then perhaps, if we could go to 336, this is the 9 note, this is the telegram itself. 10 I don't need to go into great detail of this 11 particular document. Perhaps we can just look at 12 paragraph 3. You have there Mr Naruto stressing the 13 difficult and serious crisis that Horizon faced at that 14 time. So that's December 1998. 15 They were worried that HMG didn't fully understand 16 the seriousness of the suggestion and perhaps, at the 17 bottom of the page, we have Mr Naruto fearing for 18 Fujitsu's domestic reputation if the project failed and 19 he repeatedly stressed that the failure of the project 20 will have serious repercussions for Fujitsu's 21 international standing. 22 Perhaps we can go down to the "Conclusion": 23 "The contents of the letter and the tone of 24 Mr Naruto's approach make it quite clear that we have 25 a major and potentially damaging problem on our hands. 108 1 The next paragraph, paragraph 8: 2 "Failure of Project Horizon and of ICL's role in it 3 would also knock a hole in the credit we get here for 4 PFI/PPP initiatives. The Fujitsu/ICL role in Project 5 Horizon has been seen as a template for high level 6 political study here of this approach to large-scale 7 public projects." 8 Paragraph 9, it continues: 9 "... any threat to ICL's continued viability would 10 have profound implications for jobs in the UK and for 11 bilateral ties." 12 Could we go to BEIS0000278. This is another 13 telegram from the British Embassy, 25 January 1999, so 14 moving on slightly. This is, if we look at paragraph 1: 15 "At his request, Keith Todd, Chief Executive of ICL, 16 called on me [that's the ambassador] on ... 23 [May] ... 17 after talks with the Fujitsu President. He expressed 18 Fujitsu's quote complete disbelief and lack of 19 understanding at HMG's decision-making process unquote." 20 That telegram continues in a similar theme. 21 Can we look at BEIS0000315, please. This is 22 a briefing for the Secretary of State on 4 June 1999, so 23 quite a bit on. 24 If we go over the page, and over the page again, to 25 the background, it sets out that there was a meeting on 109 1 14 May: 2 "... against the background of negotiations which 3 had been taking place between HMG and ICL led by 4 HM Treasury on a compromise solution ..." 5 It sets out there, in paragraph 2, that: 6 "Mr Sekizawa stressed that unless an unconditional 7 agreement could be signed by 17 May, Fujitsu would have 8 to accept a £306 [million] provision in their 9 consolidated group accounts. If that happened he would 10 find it very difficult to justify to shareholders and 11 analysts any course of action other than to walk away 12 from the project and seek to recover the £300 [million] 13 development costs already incurred. 14 Paragraph 3: 15 "Later that evening, Steve Robson wrote to ICL with 16 a counter proposal requiring ICL to accept a loss 17 estimated by ICL to be £250 million. There is no doubt 18 that when news of this offer reached Mr Sekizawa and 19 Mr Naruto on their arrival in Japan there were strong 20 feelings that the company had been betrayed by the 21 British Government. 22 "In these tense negotiations over the following days 23 we know that it was your letter to Keith Todd of 21 May 24 confirming the Government's wish to proceed with the 25 project, and later the personal appeal to Mr Naruto by 110 1 the Deputy Ambassador on your behalf that persuaded 2 Fujitsu not to abandon the negotiations." 3 Can you tell us, over this period, how significant 4 it was that -- and how much pressure there was to 5 maintain this relationship with ICL and not to damage 6 ties with Fujitsu? 7 A. I think it was, from the very beginning, seen within the 8 DTI as very important indeed, that it would have been 9 a major blow, as I think I've already described, to the 10 whole PFI concept if a project of this importance and 11 this stature, if you like, failed. But it was -- it 12 was -- however important that was, that was only one 13 element in our consideration. 14 The other was the damage to the network of post 15 offices up and down the country if the thing failed. So 16 we had these twin objectives, as it were, to keep on 17 trying to press ministers into a solution that dealt 18 with these two issues, and I think it was the 19 combination of them, the industrial one and the purely 20 political one, the subpostmasters and the network, and 21 so on, that, in the end, the force of those arguments -- 22 I would say that, wouldn't I -- but the force of those 23 arguments were what prevailed. And ... 24 Q. We see at paragraph 2 the Chairman of Fujitsu stressing 25 that an unconditional agreement needed to be signed by 111 1 17 May, and negotiations thereafter concerning the 2 date -- and May was seen as particularly important, 3 May 2000 -- sorry, May 1999. I mean, can you tell us 4 how much time pressure there was coming from the Fujitsu 5 end to get -- 6 A. Quite -- quite a lot. As we understood it, there was 7 a lot of pressure on them to get this sorted out so that 8 they could sign off their accounts, as I understand it, 9 for that year, and a lot hinged on this as to whether 10 Fujitsu would have no alternative but to kind of cut ICL 11 loose, disband it, whatever they were going to do with 12 it. 13 And we understood that there were genuine, 14 absolutely genuine, time pressures there, and this 15 wasn't just Fujitsu trying to scare us or whatever. We 16 believed that it was real, rightly or wrongly, I think 17 it was real. 18 Q. A second topic I'd like to ask you about is simply the 19 financial success or otherwise of ICL. Can we look at 20 BEIS0000255, please. 21 This is a submission to Sir Michael Scholar on 22 14 July 2000 and it relates to a draft National Audit 23 Office report on the cancellation of the Benefit Payment 24 Card project, so it's a draft of that report that had 25 been received by that date. 112 1 Can we look at page 4 of that submission, please. 2 It's paragraph 8 that I'd like to ask you about. 3 A. Right. 4 Q. It says: 5 "Given how badly wrong the project went from almost 6 day one, the [National Audit Office] report could hardly 7 fail to make uncomfortable reading to a greater or 8 lesser extent for each of the key players. Within this, 9 however, our objective has been to secure a report which 10 focuses in a positive way on the lessons to be learned 11 from the project, and which without pulling punches at 12 least avoids unnecessary criticism that could 13 gratuitously damage the commercial prospects either of 14 ICL (and through them our relations with their parent, 15 Fujitsu) or of POCL." 16 How important was avoiding damage to the commercial 17 prospects of ICL and Fujitsu, even at this stage, after 18 the rollout had taken place and the contracts had all 19 been signed? 20 A. I think it was important in the sense that ICL needed to 21 be a healthy partner going forward on all this. You 22 know, we weren't there to make ICL rich or anything like 23 that, but we did need them to be financially viable so 24 that they could take this project forward, they could -- 25 they would have money to spend on developments, further 113 1 development of the project, that was going to be 2 essential if it was to survive more than a few years 3 without completely running into the sand. 4 So, yes, we did want to see ICL financially healthy. 5 Q. Can we look at BEIS0000253, please. This is again 6 31 July 2000, so again long after the contracts had been 7 signed and rollout had commenced. This is a submission 8 from yourself to the Secretary of State. 9 A. Yeah. 10 Q. It's about the timing of the publication of the NAO 11 report. I just want to look at the second half of 12 paragraph 1, please, it says: 13 "ICL wanted early publication to minimise the 14 adverse effect on their planned flotation in the autumn 15 of a report which shows the company's performance on the 16 Horizon project in a less than flattering light." 17 Do you think that, by that stage, the DTI had become 18 too close to ICL? 19 A. No, I don't think so. There are two -- there was a part 20 of the DTI, not the part that I was in, that was 21 responsible for sponsoring the electronic sector or 22 whatever it was called. I didn't have the sense that 23 they were necessarily too close to ICL. I certainly 24 wasn't. I spoke to them periodically, I knew some of 25 the senior people there. But, no, I don't think we were 114 1 too -- I don't think that we were too close to them at 2 all. I think we had a genuine interest in seeing ICL as 3 a healthy company in a very important sector of our 4 economy. 5 Q. Do you think that it would be right for the DTI to be 6 concerned about the timing of a publication to minimise 7 the effect on ICL's planned flotation, for example? 8 A. If it was something that was going to happen anyway, and 9 you could do it at a certain time, rather than another 10 time, and that would be helpful to one of the partners 11 of a huge project, why wouldn't you do that? You know, 12 why would you go out of your way to be unhelpful if you 13 could do the reverse? 14 Q. Thank you very much. 15 The next subject I'd like to ask you about is PFI. 16 So at paragraph 33 of your statement, and I think you've 17 already mentioned it already today, you said that the: 18 "DTI avoided the loss of a major player in the 19 electronics sector, the risk of future investment 20 prospects, and the damage to the PFI brand." 21 Then at paragraph 34 of your statement you refer 22 again to "damage to the image of the UK's PFI 23 initiative". 24 How important was the PFI brand? 25 A. Erm ... it was something to which the Government 115 1 attached quite a lot of importance. It was 2 Adrian Montague's -- I was going to say "baby". He was 3 sort of in charge of promoting that. I think it was 4 just something that the UK had kind of come up with and 5 a lot of other administrations were interested in that 6 as a way of helping to finance public sector projects, 7 and people who were looking to us and saying "Wow, 8 aren't you clever, this is a wonderful way of doing 9 things". Would have -- well, we wouldn't have wanted 10 then to see the thing suddenly collapse and say "Not 11 such a good way after all, perhaps". 12 Q. Are you aware of any complications that PFI may have 13 imposed in the project itself, such as problems with the 14 sharing of information between Fujitsu and the Post 15 Office? 16 A. I'm personally not aware of that, no. 17 Q. I'm going to move on to prosecutions. 18 To what extent were you aware of any consideration 19 being given to the use of Horizon for prosecutions, or 20 Horizon data for prosecutions? 21 A. I was not aware of that at all. 22 Q. Thank you very much. 23 Mr Sibbick, is there anything else that you'd like 24 to add that may assist the Chair or that you would like 25 to say? 116 1 A. A number of things, I suppose, if I can kind of order 2 them a bit. 3 The first -- the first is: we -- with the benefit of 4 hindsight, we can all be geniuses and we wouldn't have 5 necessarily done things in the way that we did had we 6 had wider sight or wider knowledge, as it were. 7 Perhaps the second thing that I would say is that 8 you will see that I've put a lot of submissions, a lot 9 of advice to ministers over this period. Ministers I've 10 been fortunate enough to work for have been pretty 11 bright people, more than capable of making up their own 12 minds on things -- try to give them the facts -- they 13 could accept what I was suggesting they should do, what 14 the action should be, but they were clearly under no 15 obligation whatsoever to accept that. If they thought 16 I was wrong, they'd have said so, they'd have had no 17 hesitation in saying so. 18 The third thing, perhaps, if I may, is that I don't 19 think any of the parties to this were acting in bad 20 faith. I think they were promoting vigorously their own 21 vested -- and I'm sure -- their own interests, which was 22 what they were there, what they were there to do, and 23 I think all of them genuinely believed what they were 24 saying. 25 I've said several times that what the Benefits 117 1 Agency wanted, what DSS wanted, on their terms made 2 absolute sense, it was the right thing to do and, in 3 a wider sense, it was also the right thing to do. It 4 just so happened that there were some arguments on the 5 other side that were also very persuasive and which, in 6 the end, more or less prevailed. 7 Perhaps a final thought. I don't know if it's the 8 final one, but another thought is that everybody 9 acted -- this was a huge complicated project, it was, to 10 a large extent, treading ground that hadn't been trodden 11 before. It would have been wrong to expect perfection 12 first time round, you know, it was never going to be 13 like that. And then when it went wrong initially, 14 various stakes had already been put in the ground. 15 You know, we had a contract with ICL, it was a highly 16 publicised contract, the Horizon project, and you 17 couldn't -- you couldn't undo that, you couldn't sort of 18 unsee that. 19 So there was a compulsion to say, "Okay, that's 20 there, we -- it would be pretty awful just to sort of 21 abandon it, we perhaps don't have that luxury, we need 22 to take it forward, we need to find some way of taking 23 it forward". 24 I don't think at the end of this anybody got exactly 25 what they wanted. It was a compromise and a very 118 1 difficult compromise to reach. I was very pleased to 2 play some small part in all of that. 3 I'm so, so sorry that it turned out badly in the end 4 for so many people and I just wish your Inquiry every 5 success in getting to the bottom of this and, above all, 6 in making sure that nothing like this could ever happen 7 again. 8 MR BLAKE: Thank you very much, Mr Sibbick. 9 Sir, do you have any questions at all? 10 SIR WYN WILLIAMS: No, I don't, thank you. 11 Do I take it, Mr Blake, that you've have asked all 12 the questions, in this instance, that this witness is to 13 be asked? 14 MR BLAKE: Yes, I have. 15 SIR WYN WILLIAMS: Thank you. 16 Well, I'm very grateful to you, Mr Sibbick, for 17 coming to give evidence to the Inquiry. I'm sorry that 18 I caused a slight delay this morning, which I hope 19 didn't inconvenience you too much, and thanks again for 20 attending. 21 A. Thank you very much, sir. 22 SIR WYN WILLIAMS: So 10.00 tomorrow, Mr Blake? 23 MR BLAKE: Yes, thank you very much. 24 SIR WYN WILLIAMS: All right, thank you very much. 25 (3.51 pm) 119 1 (The hearing adjourned until 10.00 am 2 on Thursday, 24 November 2022) 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 120 1 INDEX 2 PAGE 3 DAVID SIBBICK (sworn) ................................1 4 5 Questioned by MR BLAKE ........................1 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 121