Examination of witness (Questions 40 -
59)
WEDNESDAY 10 JUNE 1998
SIR ROBERT
WALMSLEY
40. Finally, what are you doing in terms
of looking at the whole life cost of refurbishment and capital
cost as opposed to maintenance cost in order to encourage suppliers
to come up with a design that reduces refurbishment costs to a
minimum throughout the life of that project?
(Sir Robert Walmsley) The first thing is that
the people who build and design the original article should in
some way bear responsibility for the costs of supporting it. If
they do not, there is almost a guaranteed disincentive on them
to minimise the use of spare parts, et cetera, because they will
probably be the people who will make the profit from the spare
parts. The Astute contract includes a requirement for prime contract
support for the first four and a half years of the first submarine,
and I think it is two years for the second and one year of the
last, so that for the last submarine of the batch, we still have
a year of prime contract support. That means that GEC Marconi
will bear the support costs of those submarines for a total of
up to seven and a half submarine years. That is a huge incentive
on them to design a submarine so that it needs low-cost support
and there are other things too which are smaller, but just as
important, like making them own the spare parts.
Mr Leslie
41. Sir Robert, I want to talk about NAPNOC.
It is not a rude phrase, but it stands for "no acceptable
price, no contract", and it is a policy that the Ministry
of Defence have been increasingly pursuing as competition within
the defence industry gets perhaps less and less and you have to
somehow replicate competitive pricing. I think that is what it
means anyway. I am concerned that it might not be working particularly
well and the two examples that I have just spotted in the Report
are with regard to the Astute-class submarine where you have tried,
I think, your best to negotiate a bit of a better price, but because
of the delays that have occurred, you have almost painted yourself
into a corner and have had to reach a compromise over pricing,
and the other one is with regard to the Bowman radio project where
you have found, I think, that it has been difficult for the Ministry
of Defence to conduct its own equivalent pricing arrangement.
How are you going to do this NAPNOC system? How are you actually
going to make it more efficient and make it work in a better way?
(Sir Robert Walmsley) It is very difficult. There
is a distinction in character between Astute and Bowman which
might be worth my trying to set out. The great part of Astute's
cost will be incurred in the shipyard as a result of materials
and man hours in constructing the ship and of course the overhead
rates that are operating in the shipyard. What that means is that
we have to understand the costs that the shipyard will incur,
the range of overheads, how they should go about putting competitive
pressures to bear on such work as the shipyard itself can sub-contract,
and I think our experience is that we are quite good at that and
I am absolutely satisfied that we got a very good result on the
Astute-class submarine, I may say, because of two other factors
which were helping us. First of all, we had a batch order for
three which we have never done before with a nuclear submarine
order, we had always ordered them singly, and my working rule
of thumb is that we would expect to get a 10 per cent reduction
for ordering three as compared with ordering each singly.
42. Just before you go on, is it right though
that you had to compromise in a way and, okay, you say you got
a good price, but could you not have got a better price if you
had done your own costings and got a feel for what the competitive
price might have been somewhat earlier in the process?
(Sir Robert Walmsley) I cannot deny it is possible,
but if there is one NAPNOC price which I feel confident that we
made a big step forward on, then it is indeed Astute. We were
partly able to build on quite a substantial workload in the shipyard
which meant that we knew what the overheads were. We had been
around the NAPNOC system with the LPD(R)s and we were having a
competitive contract to order tankers in the one and the same
yard. I think we knew that shipyard's costs very well and we were
careful to separate the overheads as between submarines and surface
ship construction so that there was no possibility of, so to speak,
cross-subsidising surface ship construction by easing down on
the submarine overheads.
43. But you could not do that in the Bowman
radio case because you did not have enough information there?
(Sir Robert Walmsley) The character is different
in that I believe that with Bowman, in contrast to the submarines,
the great proportion of the costs are capable of sub-contract
competition so that the Bowman contract would consist of HF radios,
VHF radios, frequency management systems and a whole lot of discrete
subsystems which can each be put to sub-contract competition and
the NAPNOC process surrounding a project with that degree of sub-contract
competition consists of us making sure that these sub-contract
competitions are run properly and then sharing the benefits of
the results. We would, for instance, ask to be there when the
tenders were opened. We might ask for the right to nominate additional
bidders if we thought that the shareholders of the prime contracting
company were using this arrangement as a way of sharing out the
work amongst themselves, so with Bowman the roots will be firmly
based on sub-contract competition.
44. Are the Ministry of Defence planning
to somehow boost your own internal costing capacity? You need
this "should cost" data. Are you planning to bring in
people from private industry or whoever to somehow give you a
better flavour, a more up-to-date capacity to put a precise market
priced figure or equivalent figure on the stuff that you are buying?
(Sir Robert Walmsley) We do bring in people from
outside. We do have the Specialist Procurement Services Agency
and it is the pricing part of that which does this work. Many
of them are qualified accountants, many of them qualified engineers,
costers. Perhaps I should say that they have been protected from
run-down. So we have kept them very substantially as the centre
of most important expertise, but I should just perhaps repeat
again that we are succeeding in maintaining 75 per cent of our
contracts by value. So although there are these huge challenges
in the NAPNOC process, "should cost" data is enormously
difficult to get hold of whether you are in the industry or in
the Ministry of Defence.
45. Perhaps there is a little bit of spend
to save involved here. It might be worth a stitch in time to get
the "should cost" data right so that you can save a
lot more money in the long run and avoid being painted into a
corner by the need to replace obsolete equipment or whatever.
I wanted to move on to another point. Mr Williams picked up the
question of your summary project data for costs and timescales
and how all of them, as presented to this Committee, were incorrect
in some shape or form. What I am concerned about, though, is whether
this is somehow symptomatic of the Ministry of Defence's capacity
to monitor your whole expenditure procurement process in full
because if you cannot somehow present to us in a semi-comprehensible
form what is going on with your budget—and I know you said
the textural information is right but the numbers were a bit dodgy,
but it is the numbers that really matter—how can I have
confidence in your capacity to monitor the costs incurred by the
Department?
(Sir Robert Walmsley) Two points. First of all,
I tried to say that I think the numbers were often not very far
out. I am not trying to say that it does not matter, but we are
talking about relatively small errors. Secondly, and a much more
important point, the data on these project sheets is by and large
generated purely for the purposes of this report.
46. Do you have your own internal figures
that you are monitoring all the time?
(Sir Robert Walmsley) Yes, we do. The fundamental
reason for that is that this report measures us against the ISD
set at the very first approval of the project which is an average
across these 25 projects. So you have to go back to files in 1988
and measure from that.
47. That is another problem.
(Sir Robert Walmsley) It is a problem. The way
we manage projects is the team is set the goal which is defined
in the current approval. So all our costs are measured against
the approval that is in force governing today's expenditure. This
report goes back to the very first approval for ISD and to the
very first approval of each stage of project costs and that is
not the way that the teams manage it.
48. I would like to ask the Treasury a question.
I read in the report that of the 25 projects there is about a
9.1 per cent price rise on the initial figures and I am very concerned
that in many government departments, particularly the Ministry
of Defence, estimates come in, you start the whole thing going
and then later on down the road all these problems come up and
more money has to be found and it is a big pressure on public
service budgets. How well do the Ministry of Defence compare in
terms of capital overrun with other government departments? Are
they much worse? Are they better?
(Mr Martin) I think it would be almost impossible
to compare because the nature of MoD procurement is in a sense
one-off for central government. There is no government department
other than in, say, an information systems context which is involved
in similar types of procurement activity in terms of scale or
complexity or length of contract time as the Ministry of Defence.
So I would not feel able to make that sort of comparison, nor
would I feel that comparison would be fair.
49. I am just wondering whether there needs
to be some way of gauging how well government departments conduct
themselves in terms of reigning in over-spends on capital and
overruns. Surely it must be possible to see whether some government
departments ignore contractors who come in with late receipts
or whatever and see how well they are doing. Perhaps this is an
issue to be taken up in the resource accounting debate. I do not
know whether that is something that can be taken up.
(Mr Martin) I certainly think resource accounting
will provide some better information about performance in this
context. I would still hold to my suggestion that the MoD is dealing
with a very very different range of activities than other departments.
50. Let me just very quickly move on to
the question of the delays which we touched on before and how
22 out of the 25 projects we are looking at were late, six of
them five years late and the fact that things were not particularly
improving. First of all, can I just ask about Merlin helicopters.
I notice that the in-service datum for Merlin is March 1999, which
is a slip of five years three months. Why is that?
(Sir Robert Walmsley) This is the anti-submarine
Merlin which got its first approval in 1975. It is an immensely
ambitious thing to develop a new three-engine helicopter, as I
am sure you well know. I think it unlikely that the Royal Navy
would ever attempt to do that again with a single partner, the
Italian Navy, as it is such a huge undertaking. There have been
32 months-worth of technical difficulties with the flight control
system, the engine was more complex in terms of its integration
and all that added up to a 29-month delay even before pre-production
aircraft No.7 had crashed.
51. You have had to keep the old Sea King
helicopters. What is the equivalent cost of that delay? Has there
been a price put on that?
(Sir Robert Walmsley) I would have to give you
a note on that. We find it very difficult indeed and I would not
attempt orally to try to explain why I find it so difficult[14].
52. I think page 116 estimates that at about
£150 million. If you have got a different figure for it I
would be interested to see it. Is it not the case that in some
way having to use obsolete equipment for longer and longer periods
can in fact jeopardise your operational capacity and can create
problems? I know when we looked at the four-wheel drive utility
vehicles delay, which was a delay for more than two years, there
were questions about the fact that they may have had difficulties
in Bosnia because they had to use the old equipment longer than
necessary. Is this not a real problem that has to be addressed?
(Sir Robert Walmsley) Very much so. We set the
initial in-service date. At that stage we were very much dominated
in our thinking by the Soviet submarine threat and the Merlin
is designed to play a huge part in combating that, but that does
not mean that its operational capability is not needed and it
certainly does not mean that we look at these delays with equanimity.
That does not mean that its operational capability is not needed
today and it certainly does not mean that we look at these delays
with equanimity, but they were basically booked into this project
some considerable time ago and the progress, if it had not been,
and I know it sounds an easy thing to say, for crashes to prototype
aircraft which had technical difficulties in the programme, the
project would have, I believe now, remained on a steady in-service
date for some time, but I agree that the three aircraft crashes
which I think there have been have caused an immense difficulty
in bringing this programme to a conclusion.
Maria Eagle
53. Do you think, Sir Robert, that the Department
and the procurement element that you are responsible for in particular
has any difficulty with financial control generally?
(Sir Robert Walmsley) I think it is something
we need to work at very hard. I think the predictive tools associated
with forecasting and how much expenditure we are going to undertake
in a year, next year, the year after, is something which we still
need to work harder on, and we are doing that, but I think in
terms of controlling expenditure in the sense of avoiding fraudulent
expenditure, I think our systems are pretty robust and we have
just done our annual report on that which confirms it.
54. I am not concerned so much here about
fraudulent expenditure, but more about allowing costs to spiral
out of control. We have had the MoD in front of this Committee
on a number of occasions in the past few years and I was a Member
of the Committee on the last one, trying to find out why you have
overspent your votes. This indicates, I think, a problem with
financial control. I wonder to what extent you believe some of
the difficulties highlighted in this Report of cost overruns may
have been recognised at too late a stage or not acted upon soon
enough because of that kind of problem.
(Sir Robert Walmsley) I think it is certainly
true and I think that we do need to work harder on predicting
expenditure. When I appeared before this Committee quite recently
on the subject of the overspend, I tried to make it clear that
our ability to forecast expenditure was something we needed to
improve and I explained then how we were trying to go about it.
I think the transmission of costs into the armed services as a
result of the late appearance of equipment is something which,
as in the case of the Sea King which we have just been talking
about, resource accounting and budgeting is going to really bring
home to roost. I think that will put a new degree of discipline
into the relationship between the Procurement Executive and the
front-line services. That should help to avoid any suggestion
that we allow support costs to increase as a result of delays
or difficulties in the acquisition process.
55. Mr Williams and Mr Leslie have raised
with you the issue of your summary sheets not being accurate and
you effectively said in response to Mr Leslie, I think, and you
will correct me if I have misunderstood, that you put this information
together specifically for this Report and, therefore, the mistakes
happen because you are doing it and it is not the usual thing
that you do, apart from for this Committee. The mistakes in the
summary sheets, is that a reflection of any general problems with
control mechanisms or is it more that it is a one-off task that
you are not usually giving your attention to and, therefore, mistakes
happen for that kind of reason?
(Sir Robert Walmsley) I think I would like to
reflect quite deeply on that question[15].
My immediate answer is that it is at least in part, and certainly
the delay issue is, because it is not the normal information that
we use for managing a project. But as to whether there is something
deeper underlying our attitude to the Report, which I should say
certainly is not the case at the senior level, I regard this Report,
and so do my senior staff, as very much my end-of-term report,
if you like. It is that simple and we are using the same criteria
to monitor all the projects, not just the top 25, and we think
it is a good discipline and a good way of doing it. But as to
whether there is, so to speak, a cultural problem underlying our
approach to some of the data, I think I would like to go away
and make a real assessment of that by speaking to a few more of
the people who do the work[16].
56. I would be happy to have an answer when
you have done that. I am just wanting to draw out a few overall
lessons and points that arise out of the NAO Report, and on page
27, paragraph 2.36, this is a paragraph looking at whether the
causes of cost variances, which are increasing, have changed over
the last five years, and the NAO decided that they do not think
the causes have changed, but it is programme changes, inflation
adjustments, specification changes which are causing these variances.
It is not something surprising which has come out and leapt up
in front of you and caused problems, but it has been a trend.
Over the page at paragraph 2.39, looking at the estimates of delays
to in-service dates, and again in-service dates are slipping increasingly,
comes to a similar conclusion, that the causes of these delays
in in-service dates are well known and are broadly consistent
and have been over the past few years. In view of the fact that
these are well-known trends that you must be acutely aware of,
why have you not been able to tackle them?
(Sir Robert Walmsley) The first thing, and I hope
it does not sound too defensive, but I think the delays in terms
of `months average' is steady, not deteriorating. In terms of
the percentage cost increase, it is actually slightly less this
year than it was last year, so, as I think has been acknowledged,
the trend may have levelled out. I very much hope that is the
case and the first thing is that some of these causes of overspend,
such as changes to the specification, are now specific performance
indicators for project managers. They now focus on these, and
they minimise overspends due to changes in specification, as one
of their targets for managing the project rather than, so to speak,
count them up after they have happened. In terms of technical
difficulty, I think we have to look very carefully at how we introduce
high-risk projects into the programme. It may well be, in other
words, that we have to be prepared to abandon programmes rather
than insist on prosecuting them regardless of what the technical
advice is telling us.
57. I am not quite sure I agree with the
premise of what you have said there about the cost variances coming
down.
(Sir Robert Walmsley) I was looking on page 26
at figure 11 and I did mention "in percentage terms"
because the programme is substantially bigger because of the number
of committed contracts.
58. I think we are talking about fine distinctions
in some ways. I think certainly the NAO concludes on page 29 that
estimates of cost variance have grown over the past five years
and it may be just speaking in terms of the overall picture, but
I do not want to get bogged down in that because I have a limited
amount of time. I wanted to have a little think about what happens
when things go wrong. To what extent do you believe the MoD may
be accused of being a bit of a soft touch when it comes to contracts
which go wrong?
(Sir Robert Walmsley) I do not think we are.
59. You do not think you are at all?
(Sir Robert Walmsley) No.
14 Note: See Evidence, Appendix 1, page 19 (PAC 343) Back
15
Note: See Evidence, Appendix 1, page 19 (PAC 353) Back
16
Note by Witness: The reference to the MoD's use of the same criteria
to monitor all the projects, refers to the broader criteria of
time and cost (and the various causes of slippage or increases
used in the MPR) rather than the detailed methodology of comparison
of current estimates with the earliest approvals. Back
|
