|
Oral evidence Taken before the Committee of Public Accounts on Wednesday 29 October 2003 Members present: Mr Edward Leigh, in the Chair __________ SIR JOHN BOURN KCB, Comptroller and Auditor General, further examined. MR BRIAN GLICKSMAN, Treasury Officer of Accounts, HM Treasury, further examined. REPORT BY THE COMPTROLLER AND AUDITOR GENERAL Purchasing and Managing Software Licences (HC 579) Examination of Witnesses Witnesses: MR PETER GERSHON CBE, Chief Executive and MR HUGH BARRETT, Chief Executive, OGCbuying.solution, Office of Government Commerce (OGC), examined. Q1 Chairman: Good afternoon, ladies and gentlemen, welcome to the Committee of Public Accounts. Today we are looking at the Comptroller and Auditor General's report on purchasing and managing software licences and we welcome back Mr Peter Gershon, who is Chief Executive of the Office of Government Commerce. Thank you for coming to speak to us this afternoon. May I start by referring you to page 4, paragraph 5? You will see there that it says "... the new arrangements could deliver savings of some £100 million over the next three years to the whole of the public sector". How much of this has actually been achieved so far by departments? Mr Gershon: I can answer that question in the context of the whole public sector. As of the end of September, in terms of direct price savings over what we were paying before the Memorandum of Understanding (MOU) was put in place, we had achieved £49 million. Q2 Chairman: Are you on target? Are you satisfied with progress so far? Are you going to meet this target? Mr Gershon: I am satisfied with progress. As the report identified, at the time the report was done, we were behind track, but I am satisfied now that we have caught up as a result of a number of actions which have been taken since the report has been published. Q3 Chairman: If you are going to meet this target that leads on to the obvious question: was the target too easy in the first place? Mr Gershon: Certainly at the time this did not feel at all easy. We were faced with the prospect of proposed changes in Microsoft's licensing agreements which we estimated would have cost the public sector somewhere in the order of £40 million to £60 million per annum. We set ourselves the target not only to avoid paying those charges, but also to end up paying less than we were paying before. That involved us having to mobilise the whole of the public sector, not just central government but the devolved administrations, local authorities and the police force, to get a consistent negotiating position with Microsoft and as far as we can tell, at the time no other public sector in the world was able to achieve that sort of arrangement. Q4 Chairman: How did you come up with the figure? It sounds quite a convenient sum. Mr Gershon: No, that was an estimate based on the deal we had done with Microsoft that it would deliver the £100 million. Q5 Chairman: Would it help you if we recommended in our report a more difficult target, a higher sum? Mr Gershon: Let me put it this way. We have already opened up discussions with Microsoft. Clearly this is a very fast moving market and we think now that there is a basis for looking again with Microsoft at the agreement in the light of changes in technology, changes in market conditions and we are about to re-open discussions with Microsoft to see whether we can obtain further improvements in the MOU. Q6 Chairman: Good. I shall take that as an answer saying yes. Thank you very much. Could you look now, please, at page 14, paragraph 1.15? You will see there that it says "Microsoft requires its customers not to reveal the level of discount received by third parties", therefore there is obviously an absence of information. How do you know whether you got a good deal or not? Mr Gershon: The thing we know for certain is that we ended up paying significantly less than we were paying before the MOU was put in place. That we can be certain about. Q7 Chairman: That is fair enough. Of course we have no idea what sort of deal the likes of ICI and Boeing are getting, do we, to take two examples? Mr Gershon: No, but we have sought intelligence from a number of sources to try to verify the nature of the deal we have undertaken and we continue to track market intelligence to assess the competitiveness of the deal. Q8 Chairman: Are you telling us that somehow having tracked this, you are pretty confident that you are on a par with people in the private sector? Mr Gershon: We have to put this into context. If you are a big corporation, you would probably be able to go to Microsoft and commit a defined volume of purchases at the time you sign the deal. This arrangement is different because we are not able to commit the whole public sector to a particular set of volumes over a given time. It is a framework agreement in which there are discounts, depending upon the numbers of licences which are taken out, but we do not give an upfront commitment to the volumes which a big corporation would tend to do. The arrangements in that sense are not quite comparable. Q9 Chairman: No; I can understand that. I suppose that leads me on to my question. If you go on reading through that paragraph it says "Departments are not obliged to use the Memorandum; for example if they have outsourced contractors purchasing software on their behalf they can use any agreements that the contractor has negotiated with Microsoft". How are you going to encourage their use? Mr Gershon: Through Mr Barrett's organisation there is now extensive promotion of the arrangements and that is an ongoing activity. For example, during the course of November and December, there will be a series of road shows aimed at a particular class of customer where we think knowledge and awareness of the MOU is not as high as we would like it to be. So there is a set of activities like that which we have to do on an ongoing basis to raise awareness of the MOU and communicate its benefits. Q10 Chairman: If you look now at paragraph 2.19 on page 22, what worries me is that it says that departments are not required to report specifically the level of savings they achieve using each of the Memoranda. "As departments have been unable to provide OGC with sufficient accurate data ..." etcetera. Is that a worry? Do we have sufficient information to be able to judge the worth of the Memoranda? Mr Gershon: We get very accurate information from the suppliers because the suppliers have a strong vested interest in maintaining very accurate records for themselves about where their licences are and the take-up of them, not least to help them control and monitor potentially illegal use of licences. Yes, we do have to rely on them. We do also periodically check with some key departments about their level of take-up which does enable us to get some sort of calibration on the data we get from suppliers, but I cannot decipher that. We are principally dependent on the suppliers for the provision of regular information about take-up under the MOU. Q11 Chairman: That is perhaps something we might look at later. It sounds an interesting point. We know that Microsoft dominates the world market; everybody knows that. What can we do to make sure that departments are not so reliant on Microsoft? Mr Gershon: As the report indicates, there is an emerging potential alternative to Microsoft called Open Source software, which we have been tracking for some time. We recently announced that we are going to run a set of proof of concept trials with selected departments and other public sector organisations to get a proper economic and technical understanding of this alternative, to see whether, at this stage, it can provide an effective alternative on value for money grounds to Microsoft. Q12 Chairman: It is obviously a question which worries you and which you are seeking to address. Mr Gershon: Yes. Looking backwards, government has also had significant benefits from the widespread adoption of Microsoft software. We take certain things for granted in terms of inter- and intra-organisational communication, but if you try to communicate with some organisations outside who do not have a Microsoft domain, you get glitches and inter-operability problems. We take all that pretty well as a given. Having said that, however, clearly being in a position where we are so dependent on one supplier is not a good position to be in. We need to look continually at whether there are competitive alternatives, as are a number of other public sector and private sector organisations. Q13 Chairman: That is a very fair answer. Could you look, please, at page 21 and paragraph 2.18? If you look at the end of that last bullet point, you will see that other departments "... used agreements negotiated by their outsourced contractors which they considered provided better value for money than the arrangements established by OGC". Why do you think that is? Mr Gershon: If you take somebody like Compaq, Compaq will do a huge volume of business per annum with Microsoft. So it is also benefiting from an arrangement with Microsoft which is giving it access to discounted software. Its volumes are likely to be bigger than ours, so I would not be surprised to see that somebody like Compaq was able to offer a more attractive arrangement. Q14 Chairman: Are you worried about the fact that some departments may have illegal software in their systems? Mr Gershon: That is something we always have to be vigilant about and certainly I have made it very clear to Microsoft and other suppliers that if they can ever highlight to us instances where they are concerned about potentially illegal use of software, we will investigate that very vigorously. Q15 Chairman: Just a general question before I end. Other services to government. We live in an increasingly globalised age with dominant suppliers. Does this worry you that the choice available to the departments is declining? Mr Gershon: In this area it is encouraging to see the emergence of something like Open Source software as creating potentially a new source of competition to a supplier who has a very significant market share. As I said earlier, to have a market in which that is the case, is not necessarily particularly good for consumers. Yes, we need to be mindful of that. Q16 Mr Jenkins: When I read this report it was like going back through the history of the development of IT systems and software procurement. It was very interesting. It is a very complicated and complex picture and I do not envy you your task of unravelling the agreements departments have and have established over the years. Are you quite confident yourself that you have the team and the resources in place to tackle the scale of the job you have? Mr Gershon: In this particular area? Q17 Mr Jenkins: Yes. Mr Gershon: It is not so much a matter of unravelling the agreements. These arrangements for a department will tend to trigger when they move to a new release of software and they have to take out new licences. They do not then necessarily have to unravel. The arrangements we put in place with Microsoft cover whether the department in effect is acquiring those licences directly or is getting the benefit of those because it has outsourced the whole of its IT to a third party. Q18 Mr Jenkins: We spend about £610 million on software, do we not? Mr Gershon: Yes. Q19 Mr Jenkins: We spent £100 million just on licences. Mr Gershon: Yes. Q20 Mr Jenkins: Do you not think this is a tremendous waste of public resources? Mr Gershon: I do not quite follow you. Q21 Mr Jenkins: I refer back to what you said to the Chairman and I have your Open Source document here. One of the things you said was that you were setting up some systems to look at the possibility of using Open Source. Mr Gershon: Yes; concept trials. Q22 Mr Jenkins: What would say if I quoted to you "Our decision to use Open Source software was based on its proven reliability, portability and lower licensing costs. Overall it represented best value for money in its applications"? That is from somebody within the Civil Service. Somebody from business is quoted as saying "By selecting Linux for purchase and pay - - - Mr Gershon: Let us be quite clear. We are talking about two different markets. The case for using Open Source software, for what are called servers, that is the computers to which PCs are attached to run applications such as the one you are quoting, which was a decision made by Mr Barrett, is much clearer cut and much more well proven using things like Linux. The proof of concept trials I referred to are about using Open Source software to replace the desktop environment which runs on PCs, where the case is far from clear cut at the moment and that is where we need to do more detailed economic and technical assessments about the viability of Open Source for the desktop. There is extensive use in government at the moment of Open Source software like Linux in a server environment or for web based applications. Q23 Mr Jenkins: You think that the Linux system is not robust enough to take over from Microsoft in most of the desktop applications we are now using. Mr Gershon: No, it is not just a case of it being robust. We have to understand issues like what the migration costs are, whether there are retraining costs, what the support costs are? There is a series of costs. Yes, the licence fee per se for the software is clearly much cheaper but there is a whole set of other costs which contribute to the cost of ownership of an Open Source solution which have to be taken into consideration. Mr Barrett: Would it be helpful if I talked about my personal experience? Q24 Mr Jenkins: Yes; it was your quote. I am sure you recognised it. Mr Barrett: Yes. This is even closer to home, this is me as an individual user. I thought I would investigate the use of Open Source software on my personal computer at home. I got a licence, free of charge, but after three hours of struggling to retrain myself from many years of using Microsoft, I abandoned it. On a sample of one, me personally, it was more cost effective to stay with Microsoft. Those arguments, when magnified across the whole public sector, may well sway the balance against the Open Source software. They may not, but we do need to take them into account, which is why we are putting in place these proof of concept trials. Q25 Mr Jenkins: We definitely want to see the outcome of these concept trials. Effectively what you are saying now is that you can get a team together to give you the same sort of regime as a manufacturer, such as Microsoft, when they bring in their consultancy teams and when they put in their particular systems. Do you have a similar sort of team to develop and use and push for the Open Source software in your own department? Mr Gershon: What we are talking about are standard applications typically, which would be the equivalent of the Microsoft Office Suite, which is a standard package which runs on your PC. You do not need a lot of consultancy to put that in. Q26 Mr Jenkins: You do not? Mr Gershon: No. Q27 Mr Jenkins: What about the other costs such as training? Do you need a lot of training? Mr Gershon: That is one of the things we have to assess, as to whether there would be, in the desktop environment, significant retraining costs to use Open Source equivalents to Microsoft Office and other commonly used products from the Microsoft world. That is one of the things we have to assess. Q28 Mr Jenkins: Are you now doing this? Do you have a team in place to do this at the present time? Mr Gershon: We have announced the nine pilots, the proof of concept trials which are going to be done and we are working with IBM who are doing this free of charge, no cost to the taxpayer, and they are providing the expert resource to work with a client under our overall direction for these trials, in order that we can then pull all the data together and make a proper evidence based assessment. Q29 Mr Jenkins: Over what timescale? Mr Gershon: About one year. Q30 Mr Jenkins: In a year's time we shall have the results of your study to show whether we can go fully into Open Source software. Mr Gershon: Or whether there are certain classes of customer for whom it might be sensible to go or whether, yes, it was possible that the whole of government should move over or that it is not yet mature enough and that we need to come back and revisit the whole situation again in 18 months or 24 months' time. This is a very fast moving area of technology. There is no doubt that the viability of Open Source is increasing all the time. Q31 Mr Jenkins: I for one will be very keen to look at the results of your surveys. I have believed for a long time that Linux and Open Source software are going to be a tremendous advantage and cost saving for different departments. So together we await the outcome. Mr Gershon: I was reading today that the Danes have done quite a lot of work in this area and the report they did, which was produced in October 2002 was published in Danish and has only just been translated into English. Its conclusions are really remarkably similar to the policy we issued jointly with the Office of the e-envoy last summer and their way forward is again very similar to the approach we are taking which is the need to do some more trials to begin to get a more comprehensive assessment about the applicability of this technology in public sector environments. Q32 Mr Jenkins: If it does prove to be worthwhile to go down that route, what power will you then have to ensure that different departments and agencies consider this source rather than remaining with conventional suppliers. Mr Gershon: Firstly, I have no power of direction, but what I can tell you is that there is a lot of interest in what we are doing. For some time there has been what is known as a special interest group with departmental representatives on it, which has been looking at the whole area of Open Source. So there is a group of people who have a strong interest in this and if there are clear value for money benefits in shifting, we will see people move. The case will have to be clear cut. Q33 Jon Cruddas: Going back to the Microsoft negotiations, on page 14, paragraph 1.12 it points out "In May 2001 Microsoft announced their new licensing arrangements to operate worldwide from August 2002". The estimated increases which were likely, if everything else stayed the same, because of them increasing their costs and then producing great pressure for more frequent upgrades, were around £40 million to £60 million. Given the fact that they had 95% of the market in terms of department work, I cannot quite understand why they have negotiated the deal, because they had such strong leverage. Why were they so willing to negotiate the Memorandum? Mr Gershon: At the beginning they were not willing. We opened up discussions with them in September 2001 and we did not finally agree a resolution of the issue with them until February 2002. Q34 Jon Cruddas: The scale of the savings are significant amounts of money and given their position it seems to me that on the one hand that needs to be commended but it also needs to be understood in terms of the processes that went through. Obviously, with other departments having their own agreements outside of the Memorandum, you would look for ways of getting greater traction with them in future negotiations, would you not? Therefore for us as a Committee it is quite interesting to see the processes within that negotiation which led you to such a successful outcome. Mr Gershon: I got into trouble for this the last time. I am very willing, if the Committee are interested, to share with the Committee the negotiating techniques we used, but I would prefer to do that in private, because I may like to use those techniques again on a similar occasion in the future and I would not like my hand to be disclosed publicly. Chairman: We will go into private at the end of the session. Q35 Jon Cruddas: In terms of the anticipated productivity gain, could you give us some information about how you estimated the £64 million? Mr Gershon: I think it was £56 million actually, from memory. We estimated it on the basis of using some external data which indicated that with more capable software and common software platforms it would lead to something like one hour per week per employee of improved productivity, but we actually then said we would discount that by two thirds. That is how we estimated it. Q36 Jon Cruddas: In terms of output per person hour. Mr Gershon: Yes, that new releases of software do have features in which enhance productivity. Q37 Jon Cruddas: Without doubt. Presumably there would have been upgrades, notwithstanding the negotiation around the Memorandum and what I am interested in is whether that is factored into the total productivity outputs here. How much of that would have occurred without the Memorandum? Mr Gershon: The answer is that we do not know that, but what was clear in the runup to the start of negotiations was that these levels of increases for many departments and public sector organisations would be such that for significant periods of time they would undoubtedly have deferred future upgrades because they did not have the funds to afford these increased charges. Q38 Jon Cruddas: So part of that accounts for the decision making lag which would have occurred without that umbrella. Mr Gershon: Yes. We made an estimate. We think that discounting it by two thirds was reasonable. Q39 Jon Cruddas: A conservative estimate. Mr Gershon: We thought it was reasonable to do that in light of the fact that there would clearly have been some upgrades, but they would probably have occurred later than they would have done with the MOU. Q40 Jon Cruddas: And anticipated future productivity movements? In terms of the different savings in different departments in the report here on page 21 it seems to me that there are quite significant estimated savings in specific departments using the Memorandum, specifically the National Probation Directorate, Figure 14. Why is that such a significant portion of the total? Does that just reflect those areas which were surveyed? Mr Gershon: Under the MOU there are two basic ways you can license Microsoft software. One is what is known as an enterprise agreement, where in effect you take out a licence for your whole organisation. The other one is in effect an arrangement which allows you to buy individual licences where individuals or a small group of individuals may have an ad hoc requirement for it. Whereas in the National Probation Directorate they made a wholesale move to this enterprise agreement, some of the others, for example the Department for Work and Pensions, would reflect the benefits gained by buying some individual licences under the other half of the MOU. Clearly if DWP were to move on that, the benefits would be huge. Q41 Jon Cruddas: That is right. So some departments have moved en masse; others have not. So there are huge anticipated savings, if some of the bigger departments were to move in a similar fashion to the National Probation Directorate. Mr Gershon: Yes. Before the MOU was in place the NHS and the MOD had already entered into agreements. Although they did that before the MOU, the volumes under those agreements count towards the volume in the MOU. So other users in the public sector in a sense are being able to piggyback on the back of those volumes. The MOU also has a provision in it for other very large users such as DWP and the Inland Revenue, that if at the outset they commit these big volumes for their whole departments, then they can get better terms than the MOU. Again, the volumes count towards the volume thresholds in the overall MOU, so that all public sector users benefit from those increases in volume. Q42 Jon Cruddas: Which are the big departments still left to make those big transitions? Mr Gershon: DWP, Revenue, more to come in the NHS. Mr Barrett: At the moment the most recent information we have on departments which have made significant savings is that the top three are the Prison Service, the wider Home Office and the Scottish Executive. Q43 Jim Sheridan: May I just labour the point in paragraph 1.16 on page 15, where you estimate significant savings under productivity improvements and efficiency gains. With the experience of my past life, could you define what you mean by "productivity improvements"? In my language, that usually means that people who are currently working are not working hard enough and "efficiency gains" usually means job losses. Is either of those correct? Mr Gershon: No. What it means is that you are either able to do more with your personal computer or, with a new release of software, you can do something which might have taken you an hour to do in a shorter time. Q44 Jim Sheridan: So "efficiency gains" are just as a direct result of the upgrade of the systems and that will include training costs as well. Mr Barrett: "Efficiency gains" in this particular instance were about the cost of doing business between government and Microsoft. Putting in place an MOU reduced the cost of doing business. That is what that particular phrase refers to in this context. Q45 Jim Sheridan: In your Memorandum of Understanding with Microsoft I notice that the MOD have decided to continue using their own agreement. Given that the MOD are at the sharp end of technology, why is that? Have you taken any steps to encourage the MOD to come on board? Mr Gershon: The volumes the MOD contracted for in its agreement count towards the overall volumes in our MOU. At the moment, for the deal that the MOD did, and remember they bought a whole lot of licences, the next point at which they could consider coming into the MOU would be at the end of their current agreement when I would hope they would consider either switching into the MOU or, if they were able to commit a significant volume, coming under that part of the MOU which says that if you are above a certain threshold of commitment then you can negotiate better terms with Microsoft. Q46 Jim Sheridan: Do you anticipate that happening or do you anticipate seriously the MOD coming under the MOU? Mr Gershon: At this stage it is too early to tell. There is still some way to go before the end of their current agreement with Microsoft. When they put their deal in place, they did consult with us some time before the MOU. We had visibility on what the MOD were doing and were satisfied that for their business needs at the time, that was the right thing for them to do. Q47 Jim Sheridan: Moving on, page 22, paragraph 2.22. Given the extensive work that the OGC have put in in terms of providing market intelligence for the IT industry, less than half the departments have taken advantage of that. Explain why. Mr Gershon: That was specifically through a thing known as the OGC Foundation, which we scrapped at the end of March this year because it was a subscription based activity and we have now moved much more to holding free of charge events and running user groups and special interest groups which are at known cost to departments. Having to pay to belong to the OGC Foundation was clearly acting as a deterrent for some organisations. Q48 Jim Sheridan: Do you anticipate other departments coming on board in the near future? Mr Gershon: It tends to be a much more flexible set of arrangements. Departments tend to come on board, depending on their specific areas of interest. Q49 Jim Sheridan: How do you consider that much can still be saved by departments if indeed they co-operate more and take up the agreements you have negotiated? Do you have a figure for how much that would bring in if more departments take up the agreements you have negotiated? Do you have a figure for how much that would save? Mr Gershon: In terms of direct price reductions, we think there is at least another £20 to £30 million to come. Q50 Mr Bacon: May I quickly go back to paragraph 2.22 on page 22? At the beginning of this paragraph it says "Fifty-nine department (90 per cent) indicated that they did seek reliable market information particularly on suppliers' performance". Clearly it is a majority but nevertheless it suggests that 10% do not. It is slightly surprising that anybody involved in the computer side of a department would not be seeking reliable market intelligence. Is there a simple explanation for the fact that 10% do not? Are they for some reason getting it another way? Mr Gershon: Some of these organisations are very small. Do they then have a critical mass of expertise to be able to do this sort of thing on an ongoing basis? That could be one reason why you would not find 100% take-up. It also falls to the OGC. Having something of a critical mass at the centre of government does enable us to be a source of collecting market information which we can then disseminate widely within government, so at least those who could not do this on their own would get the information anyway without having to be particularly proactive. Q51 Mr Bacon: I shall move on because there are other things I want to ask. Thank you for that. The NAO tells us that the departments were not taking up the terms offered by Microsoft, that it was slower than the agency had anticipated, that they might be waiting until existing agreements expired. I was just wondering what the terms of those agreements were. Are they obliged to wait until those agreements expire or is there a provision for moving across sooner if they wish to? Mr Gershon: If you bought a licence under the old regime - - - Mr Barrett: You can upgrade at any time. Q52 Mr Bacon: You can upgrade it at any time effectively. Mr Gershon: Yes. Q53 Mr Bacon: It was just the phrase "might be waiting". Would they save money if they did not wait? Mr Gershon: No, they would potentially have to pay more money by moving early. Then it would depend on the business case but clearly you would have to weigh the cost of going early against the benefits you get by moving earlier. Some departments have not seen a compelling case to move yet. The report refers to the fact that we did anticipate at the time we did the MOU that there would be a bigger surge earlier than we actually saw in practice. Q54 Mr Bacon: In paragraph 1.19 on page 16 the report says "In February 2003 OGC agreed a three-year Memorandum of Understanding with Oracle" and goes on to say that it is too early to be able to assess the impact of this Memorandum. This report was published in May and it is now late October. How much further along are you in being able to assess the impact of the deal with Oracle? Mr Gershon: Clearly we have not shared this data with the NAO, but as at the end of September, under the Oracle MOU, there have been something like £25 million worth of database products ordered under the MOU and the saving the government achieved under that was in the order of £2.7 million. Q55 Mr Bacon: So would you say it has been quite a helpful development? Mr Gershon: Yes, because as a result of further work we did, we did identify Oracle as the next company that we needed to establish an MOU with based on the penetration Oracle had within the public sector marketplace. Q56 Mr Bacon: Could I ask you to turn back to page 22? In paragraph 2.20 it says that 18 departments used outsourced contractors and that five of these had arrangements for ensuring that savings were identified and that they received their share of them. What are you doing about the other 13? Plainly if there is a saving with an outsourced contractor, it is right that the department should get its share of any benefit, but it looks as though the majority of departments are not ensuring that they get them. Mr Gershon: We believe that at the time ten had not yet used the MOU. We think that the point where we have to be very active with departments in this area is when we have reached the next volume threshold in the MOU, which I would expect to be in the first part of calendar 2004. At that point we recognise we have to be very proactive with departments, pointing out to them that they should then be benefiting from the lower prices which will have been triggered and they will need to take action with their providers to ensure that they are seeing the benefit of those reduced licences. Q57 Mr Bacon: Could I ask you about the NPFIT in the Health Service? How involved is the Office of Government Commerce with that? Mr Gershon: The elements of the programme are subject to OGC gateway reviews. Q58 Mr Bacon: When you say "the elements of the programme" do you mean that the whole programme must go through an OGC review? Mr Gershon: The whole programme and individual projects within the programme. We have done programme level reviews of it and also looked at individual components like patient booking. For example, I have regular meetings with the Director General of IT, Richard Grainger, about the programme. We have recently agreed to work collaboratively because they are looking to see how they can extend their agreement with Microsoft, which is referred to in the report and also some arrangement with Oracle and I have agreed with Richard that we are going to second one of his staff into that team for a period of time to help them before they return to the OGC. That is illustrative of the nature of our relationship. Q59 Mr Bacon: I understand that it is a series of contracts amounting roughly to £120 million a year for up to seven years for five suppliers, so roughly £4.2 billion. Does that sound in the right ballpark? £120 million a year for up to seven years to each of five suppliers yet to be decided. Mr Barrett: I cannot remember. Q60 Mr Bacon: It is big money, is it not? Mr Gershon: It is serious amounts of money. In SR2002 alone £2.3 billion has been set aside for IT in the Health Service. Q61 Mr Bacon: From 2003 to 2006. Mr Gershon: Yes. Q62 Mr Bacon: The National Audit Office looked at the Libra project in the then Lord Chancellor's Department and stated that it was worried that unless the IT was designed in parallel with the business processes one would not get full benefit of the technology. Are you confident that the NPFIT is not running counter to that advice? Mr Gershon: No, I do not think it is running counter to that advice. If you go down the route of having standard systems, which are widely implemented across a dispersed organisation and with standard systems come standard processes, Richard recognises that clearly there will be a big training challenge associated with this programme at each point where these systems are implemented to help the users of the systems operate the new processes. That is why he is looking at both national providers for core applications and local service providers who will work closely with each of the trusts in the designated geographic areas to help both with the training and with the implementation of these systems and their integration with existing systems in the trust. Q63 Mr Rendel: Paragraph 1.14 makes it clear that the level of discounts varies with the volume of software purchased. Paragraph 10 on page 33 says the MOD does not currently use the MOU. What estimate have you made of the saving made by the MOD, which is obviously considerable and I quite understand why the MOD would want to go down that route, because the saving to them is considerable, as compared with the saving which would have been made by the whole public sector if the MOD had come inside the MOU? Mr Gershon: The whole public sector benefits because the volumes that MOD have with their own arrangements with Microsoft count towards the volumes in our MOU. In that sense the public sector is benefiting. The NHS volumes count towards the overall volumes in the MOU although the terms are different. The rest of the public sector benefits from those volumes. That was one of the things we negotiated in the MOU. Q64 Mr Rendel: Excellent. I am sorry, I had not understood that from the report. In that case why is it ever going to be worth signing the MOU, if you can get some sort of other negotiation? Mr Gershon: As I indicated earlier, there is a provision within our MOU that if you are a very large organisation which can commit very significant volumes, then it is possible to get a better deal than the MOU, but the volumes you get in those special situations still count towards the volume thresholds of the OGC MOU. Q65 Mr Rendel: It does that whether you have signed the MOU or not, is what you are now saying. If you can negotiate a better deal - - - Mr Gershon: Yes, but you could only do that if you could commit very, very significant volumes. Yes, for people like the Revenue and DWP, who clearly bring very, very significant volumes, this MOU allows them to get potentially better terms. Let us suppose Revenue commits to 70,000 new licences, although the 70,000 are on better terms than under the MOU they still count within our MOU towards the achievement of the next volume threshold. Q66 Mr Rendel: I understand that. What I am saying is why, if you are a department of that size, is it ever sensible to come within the MOU? Are you saying it is not? Mr Gershon: The MOU recognised that by having this specific clause in it which allowed very big organisations to come to a separate arrangement. What we were concerned about in the negotiations was that those separate arrangements in volume terms counted towards a volume threshold for the whole public sector. Q67 Mr Rendel: You have done that. May I just be clear? What you are saying is that a department which is of that size would never want to sign the MOU as such because signing the MOU would simply mean they would get a lower level of discount. Mr Gershon: I mentioned that there are these two routes in the MOU, the so-called enterprisewide agreements and where you need to buy individual licences to do specific things, but you do not need every one in the organisation to have it. I admit that for the enterprisewide agreement Revenue could have a different arrangement, but they might well use, as the report shows DWP have used, the other half of the MOU where they wanted to buy licences either for a particular individual or a small group of individuals who did not go organisationwide. There could be a mix and match. Q68 Mr Rendel: So the licences you have with Microsoft may be various different licences. Some of them can be done on an enterprisewide basis, some of those licences might be bought by the same department which in other cases had an enterprisewide basis, that same department could sometimes buy a different licence on a person by person basis. Mr Gershon: Yes. For example, if on an enterprisewide basis you wanted to give everyone in your organisation the Microsoft Office Suite, but Microsoft also has a project planning tool, you might not need to give everyone in the organisation that tool, only some particular professionals who needed access to that. That would not then be covered in the enterprisewide agreement, it would be covered by this other sort of arrangement and I could envisage Revenue, as DWP have done, using this other agreement for those selected purchases. Q69 Mr Rendel: I think you said the agreement still has some time to go, but as far as I can see it runs out in May 2004. Mr Gershon: No, our MOU runs from March 2002 to March 2005. Q70 Mr Rendel: I am sorry. I understand the MOD's agreement was a three-year agreement running out in May next year. Mr Gershon: Yes. Q71 Mr Rendel: So that does not have that long to run. Mr Gershon: No and I would envisage that early in 2004 we would be opening up discussions with the MOD about what should happen next. Just as we have entered into discussions with the NHS, because they want to extend their agreement with Microsoft, on what would be the appropriate terms to do that. That is what is about to happen. There will be some dialogue between us and the MOD early in 2004. Q72 Mr Rendel: If they can get an agreement which is better than the MOU you would expect them to take that, but their volumes would still come within the MOU for other departments. Mr Gershon: Yes. Q73 Mr Rendel: Some departments are delaying signing. What effect does that have on the discounts, if they delay, for example, until the next contract comes up for renewal? Mr Gershon: You do not hit the higher volume threshold under the MOU. Q74 Mr Rendel: So they could sign in the middle of a prior agreement, if they have an agreement which runs now. Paragraph 1.15 says "... others may be delaying signing up until the second and third year of the agreement to coincide with their own purchasing cycle". Mr Gershon: Yes. Q75 Mr Rendel: That presumably means that they are waiting until their current agreement runs out, then they might come within the MOU. Or have I misunderstood the paragraph? Mr Gershon: No, that is a fair interpretation. Very often, if you do an upgrade, you may have to buy new hardware and there would be installation costs. Q76 Mr Rendel: Are you saying they could be signing up straightaway, they are just delaying signing because it is convenient for them to do it when the contract ends. Mr Gershon: Or the business case is not clear-cut. If you want to go to a new release of software, you may have to buy new hardware to support that new software. You may have installation costs. Q77 Mr Rendel: So you are saying that signing up to the MOU in the middle of a previous contract could actually be damaging to them in terms of their costs and that is why they are not doing it. Mr Gershon: They should only do it if there is a clear-cut overall business case for moving. Q78 Mr Rendel: For them? Mr Gershon: Yes. Q79 Mr Rendel: But there could be a clear-cut overall business case for the public sector, even if there is not for them. Mr Gershon: Yes, but this has to be driven by the needs of the individual client organisations. The scope of this agreement is so wide. Q80 Mr Rendel: What you seem to be saying is that when departments are discussing whether or not they should go ahead with an agreement under the MOU, the one thing you do not consider is the possible effect that would have on all other departments across the public sector. Mr Gershon: No. Q81 Mr Rendel: Clearly the fact that the MOD and other departments, DfES and so on, have not come within the MOU does not matter because you are telling me the volumes are included. What you now seem to be saying is that there are other departments where it might not - - - Mr Gershon: No, may I just clarify. DfES is different. DfES benefits from a completely different set of arrangements which Microsoft have for educational establishments and government departments involved in the provision of education. Q82 Mr Rendel: So their volumes are not included in the MOU. Mr Gershon: No, because that is a completely different set of arrangements. Q83 Mr Rendel: Has anyone looked into whether or not the saving for the whole public sector would be greater if DfES came within the MOU than the saving is to DfES from remaining outside? Mr Gershon: No, it is absolutely clear cut. The deal that Microsoft offer to educational users is significantly better than our MOU. Q84 Mr Rendel: I do not doubt that for a moment, but that is not the question. The question is what happens to the rest of the public sector. Mr Gershon: No, they would not benefit. The public sector would not benefit from that. Q85 Mr Rendel: You have done the calculation which assures you. Mr Gershon: I am satisfied on that point. To come back to your earlier point. When we get within sight of the next volume threshold - - - Q86 Mr Rendel: Which you say you are going to do early next year. Mr Gershon: Then there is no doubt we will engage with a number who we think are the most likely candidates to help the public sector cross the threshold, to understand what might be done to accelerate the process. Q87 Mr Rendel: Presumably you will be calculating the exact volumes to the extent that, for example, if you were 10,000 licences short and you had two departments who might be going into the MOU, one of which was looking at 7,000 licences and the other of which was looking at 12,000, you would tell the department which was going for 12,000 to sign up first to take you over the trigger point, otherwise the 7,000 one would for at least a year have to pay the higher rates. Mr Gershon: That would be logical. Q88 Mr Rendel: I did not ask whether it would be logical or not. I know it is logical. The question I was asking was: are you going to do it? Mr Gershon: That would clearly be a factor, assuming, all other things being equal, that they were both ready to go. At the end of the day no department should go if it does not have a clear-cut business case, because there are so many other factors they have to take into account other than the basic cost of the licence. Q89 Mr Rendel: I understand all that. It just seems to me that the whole point of having a central public sector buying department such as I understand you to be is to make sure that the benefit comes out in the best possible way for the public sector as a whole, rather than just for individual departments. It seems to me that what you are saying to me is that you leave it up to the individual departments to decide in each case what is best for them, what is the best way that they can minimise their costs, rather than try to look over the public sector as a whole, which is surely what you are all about. Mr Gershon: There have been instances where there have been particular areas either where there have been difficulties between the supplier and the client in reaching closure on a particular deal where we have intervened to help resolve the issues and by doing that, get a few more users on the meter. We do intervene where there are problems, but we will not get within sight of the next volume threshold until we have done the NHS deal. That is the reason why Hugh and I were very willing to put one of our experienced staff into the NHS to help them do that: it would take us a long way towards the next threshold. Q90 Mr Davidson: May I ask Mr Barrett just to clarify a point I thought he was making earlier on but perhaps I misheard? I think it was in response to a question from Mr Cruddas about departments taking advantage of a system. I think it was mentioned that the Prison Service, the Home Office and the Scottish Executive were the departments which were not accessing. Could you just clarify what you were saying? Mr Barrett: No, the information I was giving was that the three departments against which the biggest savings have currently been reported are Prisons, Home Office and the Scottish Executive. Q91 Mr Davidson: Fine; that is good. Which of the departments have reported the least? Mr Barrett: I would imagine that it would be the very small departments. I do not have that information to hand. Q92 Mr Davidson: If you know the best ones, why do you not know the worst ones? Mr Barrett: The information is available, I do not have it. It would presumably be the smallest department. I do not know, perhaps the Treasury Solicitor's Department, because presumably they have only bought a handful of licences, so they will have made smaller savings. Q93 Mr Davidson: Which departments are making the least gains from the work you have done, which have been slowest at picking up the advantages of working with yourselves? Mr Barrett: I understand the question and I do not think there is a department I could characterise in that way. Q94 Mr Davidson: There is nobody of whom justifiable criticisms could be made. I do not know whether I find that cheering or depressing actually. Mr Barrett: That is the case. Q95 Mr Davidson: May I ask about the question of unlicensed and illicit software? Do any of the companies involved have inspection rights to clarify whether or not departments are using illegal software. If they do have that, can you just clarify how that operates and whether there are risks to commercially sensitive information held in departmental files and secure military material and so on? Mr Barrett: May I give you an example of the rights which exist within the Microsoft MOU? Microsoft have the right to audit. If the audit reveals that there is more than 5% adverse variance, in other words that the department or agency has more than 5% too many users for the number of licences it has, then Microsoft have the right to recover the unlevied licence fees, but also to charge the department the cost of the audit. That is how an atypical audit process would work. In terms of confidentiality, I would have to give you a note on that because I could not answer that. I do not know whether Peter can. Mr Gershon: No; we should offer to give you a note on that. Q96 Mr Davidson: May I clarify another point about Microsoft and education and the rest of the public sector? You seem to be saying that Microsoft has a policy which gives educational establishments more discount than they could ever afford - - - Mr Gershon: It is not just Microsoft; it is very common practice in the whole software industry. Q97 Mr Davidson: Presumably if they changed their policy, then we would virtually have no ability to change that in a sense. We would not really be in a negotiating position, since they are the sole holders of the material. If they change the policy on the discount that they give either through yourselves or through educational establishments, we would then just be caught by a much higher price. Is that correct? Are there really no alternatives to their proprietary brand of software without enormous disruption costs? Mr Gershon: If our work on Open Source software for the desktop environment proves it is both economically and technically viable, then that clearly becomes a contender for the education environment as much as it does for the rest of the public sector or the private sector. Q98 Mr Davidson: Am I right in thinking that there are quite substantial changeover costs involved in any - - - Mr Gershon: We are going to try to bottom out what these costs are. Q99 Mr Davidson: Could the material we get from Microsoft be in any way subject to a decision by the United States Government on anything to do with trade sanctions where the American company could be forbidden to provide us with material or could be obliged to change its price as part of any sanctions arrangement that there could be between the EU and the US, for example? Mr Gershon: We would have to take that one away and give you a note on it. I am not aware of it, but I would want to go and consult with experts before I gave you a definitive answer. Q100 Mr Jenkins: Forgive me labouring this point. Am I right to say that the minister announced in a publication the government's OSS policy in July of last year which sets up a level playing field for procuring such solutions? Mr Gershon: Yes. Q101 Mr Jenkins: You were charged with implementing that policy, yet that was in July 2002 and at the present time you are undertaking a study which is not reporting until 2004. Am I correct? Mr Gershon: Yes. Some other things happened since that. In September my office produced more detailed guidance to departments on the implementation of policy. For example, it gave advice on what to look out for to avoid the dangers of greater lock-in to proprietary solutions. The special interest group which I mentioned earlier was formed and it was partly as a result of the work of the special interest group that led them to this proof of concept trials. Q102 Mr Jenkins: He actually went on to say that you are continuing to support departments to ensure that OSS solutions demonstrate best value for money. Could you tell me how you are supporting these departments? What sort of manpower do you have working in these departments? How many people are you using to undertake this review? If you cannot tell me now, could we have a note please? I find this strange because I was given an assurance that the minister had a policy in July 2002 and yet we are not going to be delivering information on this particular area until November 2004. There is a level of urgency here and how many people do you have working on this? Mr Barrett: May I answer the question on why it has taken 18 months for the trial to take place? I have had these conversations with companies like IBM and other people who are involved in the Open Source arena. In 2002, their view was that Open Source software was not robust enough for use on the desktop. Therefore we would frankly have been irresponsible to have recommended its use. Eighteen months on, software industry has matured, the products have matured and our judgment was that now was the time to run those pilots. Eighteen months ago the software industry said it was not fit for purpose. Q103 Mr Jenkins: So the Minister was plainly misled. Mr Gershon: No, he was announcing a policy. We do want to create a level playing field for Open Source software. He did not say that from the day we announced the policy, overnight things would change. We have to take into account the maturity of the technology. We think we were ahead of the game. By laying down the policy in advance of the maturity of the technology we can intercept when the technology becomes mature and robust for the desktop environment. Q104 Mr Jenkins: If we could have that note please, so we can look at that, I should be grateful. Thank you very much. Mr Gershon: Yes. Q105 Jim Sheridan: Assuming that big departments like Education and MOD take advantage of the MOU, do you have any concerns that some of our highly educated students could hack into the system to eradicate top-up fees or, more importantly, given the global threat of international terrorism from an MOD perspective? Are you happy that secure systems are in place or proposed to be in place to deal with any situations like that? Mr Gershon: We do not rely solely on the security of the Microsoft products. For example, within government we have a system known as the government secure intranet which has a very robust firewall, so that with all the recent rounds of viruses which you read about as far as I know a virus has never penetrated the firewall. There are other measures in place to deal with anticipated threats like hackers, but we will never rely solely on the security of the Microsoft products. Q106 Mr Rendel: May I just check that I have understood correctly. When the MOU was negotiated with Microsoft, the MOD volumes were included but the DfES were excluded. That was part of the negotiations, part of what was written into the MOU. Is that right? Mr Gershon: Yes. The net effect is that the MOD volumes are in, the NHS volumes are in, the DfES and the wider educational arrangements are out, such as all the schools. Q107 Mr Rendel: That was part of the negotiations. That was written in. When you made the Memorandum it was written in that that would be the way the volumes were treated. Mr Gershon: Yes. Q108 Mr Rendel: Two department in, one department out. Secondly, the MOU discount, as I understand it, is actually greater than the MOD discount, but the reason they did not go into the MOU was because there were other costs. Paragraph 6, page 33, indicates that the MOD was given the chance to come into the MOU but decided not to because "The Ministry found that its agreement provided greater value for money to meet its specific needs" because it - six lines before - "... would have incurred additional costs through having to purchase extra software". Mr Gershon: I do not think you should deduce anything from that about the respective discount levels between the two arrangements. Q109 Mr Rendel: Are you saying the discount level was greater under the MOD's original agreement anyway? Mr Gershon: I would have to go away and check. It would not surprise me. Remember they gave an absolute commitment in their agreement to 150,000 licences. Q110 Mr Rendel: What I am looking at is the wording here which says "If the Ministry had signed up to the Memorandum, however, it would have incurred additional costs through having to purchase extra software" and later on "The Ministry found that its agreement provided greater value for money". Presumably, that implies to me - it may not mean but it implies - the reason the Ministry decided to stick with the original agreement and not sign the MOU was because the cost of the extra software was going to make it more expensive to go that route, but it would not have been as expensive had it not had those extra costs. Mr Gershon: I will have to offer to give you a note on that. Q111 Mr Rendel: Figure 12 on page 20 shows what the departments are considering when they are looking into whether or not to buy particular software. In thinking about the discount they might get and therefore the cost of the software, do the departments take into consideration at all whether that cost is likely to reduce later on, because other departments will buy in and take you above another trigger point? Mr Gershon: In the communication we are now doing about the MOU, which began in the early summer and is ongoing, that is certainly something we are drawing to the attention of departments, that there is the potential in the future that the price they pay will go down when the next volume threshold is triggered. Q112 Mr Rendel: So it would be something they would take into consideration when deciding whether to buy it or not, in deciding whether there was a good business case for purchasing something. Mr Gershon: Yes. Of course in their business case is whether they factor that in as a banker or it is upside to the base business case. That would be to some extent for the department. Q113 Mr Rendel: That was going to be my supplementary question. To what extent do you advise them that they can take into account potential future benefits? Mr Gershon: No, if they asked us for advice at the moment, we would say they should not assume that. Q114 Mr Rendel: Clearly they should not assume it, but in the normal case you would assume that there was a probability of such and such, that you would gain a certain amount of money and you would take that into account. Mr Gershon: That the next volume threshold will be triggered and therefore it will occur. When we get much closer to the next volume threshold than we are, then our advice to departments would be that they could bank on it. Q115 Mr Rendel: So you are giving them advice about the potential benefit and the likelihood. Mr Gershon: Yes. Q116 Mr Bacon: I should like to go back to the Health Service briefly. Could you briefly describe the relationship between the OGC and the National Health Service? What projects of the NHS fall within the ambit of the OGC and what projects of the National Health Service fall outside? Mr Gershon: The Health Service per se is outside my formal remit, but the Health Service has recognised the benefits of some of the things the OGC is doing. For example, it voluntarily, not because it had to, imported the OGC gateway review process. In doing that it was agreed, not because it was part of the remit but in a separate agreement, that NHS national programmes would be subject to gateway reviews which were organised by the OGC rather than organised by the NHS's own gateway team, whereas projects which occur at an individual trust level will be organised by a gateway unit which is now sitting inside the NHS based in Leeds. Q117 Mr Bacon: The reason the National Programme for IT in the National Health Service worries me is first, it is very, very big and second, it is very innovative. There are shelf-loads - including from this Committee - of advice that very big projects should be cut up into much smaller bite-sized chunks and also there has been lots of advice from different sources, including from the National Audit Office in its recent report on the Libra project that business processes should be designed in parallel, if you are to have any chance of achieving the benefit. I am worried that the national programme for IT in the Health Service breached both those principles. It is not being cut into smaller bite-sized chunks and the business processes are not being designed in parallel. Am I wrong? Mr Gershon: Firstly, there are different projects within the overall programme, such as the electronic patient care record, electronic bookings, so in that sense there is. With electronic patient care records, the whole record is not computerised on day one, it is envisaged that that specification will evolve over time and the implementation of these systems occurs on a trust by trust basis or a GP surgery for bookings. There is no big bang implementation, which is another thing the NAO and ourselves agreed should be avoided if at all possible. So in that sense I do think there is modularity of approach. The other thing is that because this is a national level programme, the programme is also governed by the further set of actions that ministers agreed to last December - and there were six of them - one of which is about the fact that if people want to do big bang implementations it now has to come to a special scrutiny group which is made up of ministers and officials. A department is not left to its own devices to decide it is going to go big bang. One of the other things it is bound by, because the NAO and ourselves, looking back over a whole series of projects which have been the subject of NAO investigation and PAC inquiry, agreed on a list of eight common weaknesses, is that now, certainly for the so-called key mission critical projects of government, of which clearly the NHS programme is part, the accounting officer has to sign a letter of assurance at particular phases in the project that the project does not exhibit any of these common causes of weakness. The gateway reviews test whether there is a strong basis for that letter of assurance having been signed. What we are at least trying to do is make sure that the fact that we see this boringly repetitious set of weakness is a thing of the past. If projects fail in the future they will be for novel reasons and not the old familiar reasons. The NHS programme is challenging, is risky and it is ambitious; there is no point hiding that, because it is. That is why we brought in what we thought was a pretty good guy from the private sector to lead it. He has built up a very strong team around him. We are giving him all the support we can, but we will not know we have won until we cross the finishing line. The trouble with projects is that there is no guarantee of success until you have finished. Chairman: We are now at the finishing line as far as the public is concerned. We will now move to private session. |