Examination of Witnesses (Questions 140-159)
SIR NICHOLAS MONTAGU KCB, MR JOHN YARD CBE, AND MS LIS ASTALL
MONDAY 3 DECEMBER 2001
140. I am glad it is clear to you. What is not clear to me is why a justification of your higher costs of staff should take into account the fact you have already been doing the job. If 1.5 hours per function point is, as you have just outlined it to the Committee, simply down to the fact that you have done the job already, it seems to me that should be discounted from the productivity gain that has been calculated against those costs, otherwise what it means is that you are loading costs on what is simply an historical situation, that you happen to have been doing the job.
(Ms Astall) I am not sure I see what you are getting at. We achieved 8 to 10 so there is a differential of between 8 to 10 and 11Ö , which is attributable partly to us having
141. Let's call it 9 to 11Ö, which is a 2Ö hours per function point differential and productivity.
(Ms Astall) Yes, although as I think I pointed out earlier you can have a very great range on this. So the 8 to 10 to 7Ö is the increased productivity which we get by being more used to the system.
142. Yes. What I am keen to establish is that the 1.5 differential between 7Ö and 9 is what you get out of being used to the system, it cannot justify the higher staff costs that you pay which have already been justified by the differential between 11Ö to 9, which is the 2Ö hour productivity bonus per function point. Therefore what I want to be absolutely clear about is that the 1Ö hour differential between 7Ö to 9, which comes simply from your experience of the job and not because you have better qualified or better trained staff, is not being taken into account in using the justification that you pay those staff more when considering the productivity gain. That is the question. I have successfully confused both Ms Astall and the NAO! Perhaps it is something you could consider and then provide us a note on because I do not want to belabour the point but I do want an answer to it. The answer that you gave to Mr Steinberg was that the justification for your increased staff costs lay in the productivity differential. The point I am making is that that productivity differential can only be the differential of 2Ö points between 9 and 11Ö, not between 7Ö and 9 because that is an historical happenstance.
(Sir Nicholas Montagu) Chairman, I have to confess some of us are finding it quite difficult to see what Mr Gardiner is seeking to prove, but in order to help him perhaps we, jointly with the National Audit Office and Accenture, might let you have a note about exactly what this is about.
Chairman: That would be very helpful.
143. What, on present trends, is the volume of the work in function points likely to pan out to be? If you look at Figure 8 and the bar chart there, we have the range from 4,560 through 8,000 to 12,300, on present trends which end of that spectrum is it likely to be closest to?
(Mr Yard) It is about two-thirds of the way long, I should think, round about 9,000 function points, which is our forecast at the moment.
144. The trend is for 9,000. That will only eventually prove itself at the end of the contract, I take it?
(Mr Yard) Yes. That is what we can see at the moment.
145. So 2004 is the time point that will crystallise, is that right?
(Mr Yard) Correct.
146. With reference to 2004, what are the lessons from NIRS 2 that you are applying now, or would propose to apply, to the renewal of the contract and that process when it comes?
(Sir Nicholas Montagu) I think that successful outsourced IT depends on a genuine partnership of the sort that we have always had with EDS and we have established with Accenture. Each side needs to realise what the requirements of the other party are, and they need to be accommodated as part of the contract arrangements. It is also, I think, important to know what are the outcomes required from the contract. Another thing, which again I have learned very much from the relationship with EDS as well as with NIRS 2, is that you need solid relationships all the way down the chain, so that John and Lis can ring each other as necessary on a day-to-day basis, that the people below them can do so. All of that is summed up in a robust partnership that offers genuine wins for both parties.
147. But, with respect, those are general points which I have no doubt are valid for many contracts, my question was slightly more specific than that, it is what are the lessons out of what has happened in NIRS 2 that you would be seeking to apply to the renewal?
(Sir Nicholas Montagu) Essentially they are the same ones. Sir John's Report notes that we have addressed the difficulties that there were in the early stages of NIRS 2. This has been because we have established a proper partnership of the sort that I have described, that Lis and her colleagues are involved in all the relevant key committees and groupings in the Department, that the contract as now structured offers the prospect of wins for both sides if it is successful, and these are the lessons I have learned from NIRS 2 and from its working as well as it is doing.
148. Can I refer to paragraphs 2.5 through to 2.9 and ask Ms Astall what the revenue would have been to Accenture if they had used the framework rates for 8,000 units? Is it something like £150 million?
(Ms Astall) If we use the graph at 8,000 function points? Is that what you are asking?
(Ms Astall) Yes, about 150 million.
150. So in May 1997 when this enhanced system was worked out between the Contributions Agency and yourselves, you had quite a lucrative contract in place which would have given you £150 million which now you can provide for a mere £100 million. It is quite a juicy contract you had in place in May 1997, is it not?
(Ms Astall) I think it is the framework contract you have to bear in mind and that came from a set of rates which were negotiated by DSS which can be used by any Government department and they represent a set of rates for one person over a period of time, so there is no volume discount in there, there is no complexity allowance and no duration or anything taken into account. Typically they were not used for large pieces of work at all. What tended to happen was if it was a small piece of work for ten days, people might use the framework rates. If they had another contract going for a longer period of time then they would do a separate negotiation, which is what happened here.
151. In May 1997 you discussed the system enhancements again, there was an agreement reached in May 1997 that said up to 2000 function points a year would be for a certain amount and if it went beyond that, the price would escalate back to the framework rates?
(Ms Astall) Yes, that was set in 1995.
152. The framework rates were set in 1995?
(Ms Astall) The framework rates are changed each year but the contract for NIRS 2 was set in 1995. What was set in 1997 was the agreement on the size of the system.
153. Right. When was this agreement to limit enhancement to 2000 function points a year reached?
(Ms Astall) That was set in 1997 but it was based on a range agreed in 1995. So if the system was X size then you would go to this range, if it was Y size then you would go to that range. That was all specified in 1995 but it took until 1997 to make sure we finalised how big the system was.
154. In 1997 you readdressed this issue and agreed that 2000 function points a year would be all that
(Ms Astall) It was not readdressed, no. It was a closed thing decided in 1995. Once the size of the system was agreed that was a given, and 2000 function points was the range.
155. What was agreed in 1997?
(Ms Astall) In 1997 it was the size of the system measured by function points which took some time to agree, it was quite complex to go through. An independent company had to come in and do the assessment.
156. I am not quite following. Assessment for what?
(Ms Astall) For the size of the system, the number of function points that was in it.
That was done in 1997?
(Ms Astall) It was done by 1997. It took over a year.
158. It started in 1996 and was finished by 1997?
(Ms Astall) Yes.
159. Sir Nicholas, was it not clear in 1996, certainly by May 1997, there would be a change of Government? It is the same point that other people have gone on about but in your answer to one of my colleagues you said it was not foreseeable two years before an election that there would be an explosion of legislation after May 1997. By 1996 it was pretty clear that Labour was going to win the election and by 1997 it was certain because that was when the election was. From all the policy documents that I was forced to read as a candidate it was absolutely clear to me that there were masses of changes being proposed. My Conservative colleague, George Osborne, also made the same point as well as Labour colleagues. There is unanimity I think on this Committee, across all parties, that it was clear by 1995 but certainly by 1996, and absolutely certainly by May 1997, that there was going to be an explosion of legislation on the welfare side.
Chairman: I think it is unfair to put a question to a civil servant on the basis that they should look at the policy on the basis of a change of Government. I think that is a little bit unfair.
Mr Gibb: I am not sure I agree with you, Chairman. We are talking about a commercial contract here. Although it is hindsight, this Committee is assessing whether good judgments were made.
Chairman: Please answer the question.
(Sir Nicholas Montagu) I am very happy to, Chairman. If I may, I will answer it in the words of Sir John Bourn. "At the time of the transfer it was not clear to what extent, if at all, the legislative changes proposed, together with their respective implementation dates, would result in the annual 2000 function point limit being exceeded or, more fundamentally, whether the system had the technical capacity to absorb the level of change required."
7 Ev 23, Appendix 1. Back