Examination of witness (Questions 80 -
97)
WEDNESDAY 10 JUNE 1998
SIR ROBERT
WALMSLEY
80. You mentioned your negotiating power
at the start and McKinsey and others have suggested that you,
as it were, get into bed with your suppliers and McKinsey in fact,
am I right to say, work for some of these suppliers and I just
wondered in terms of the power of negotiation, et cetera, whether
in fact there are problems inherent in becoming entangled with
a supplier and, therefore, being unable—and the power of
negotiation is the power to walk away from the table, but in a
way you have got to buy these defences or whatever it is and if
you cannot buy it and then you actually get very close, so is
there not a problem there?
(Sir Robert Walmsley) I think there is a problem
and I think we have to be extremely careful not to be in bed with
our suppliers while we are negotiating a contract with them.
81. If a prospective tank producer perceived
that you were getting into bed with someone else, they would produce
something else, so the competition would go down again, would
it not, because it is a small number of players in the marketplace?
Is it your view that they are carving up the market to supply
people like you?
(Sir Robert Walmsley) I think all enterprises,
whether in government or out, would prefer to find themselves
in a monopoly position. We will try to avoid being placed in that
position and I think the statistics indicate that we have been
relatively successful.
82. Do you ever say to the politicians,
"Look, if you want to achieve this and you push us down this
route, technically this is what we need, whereas in fact there
is a cost scenario here and the players in it, so why don't you
go for this solution instead?" or do you never get into that
sort of debate?
(Sir Robert Walmsley) We very much do get into
that debate. We are looking very carefully as we start a project
at a range of solutions to a particular military need and the
balance of investment between different solutions to the same
thing is a fundamental part of our process.
83. Do you do any sort of benchmarking against
other EU purchases so that when the French buy tanks or whatever
it is, have you got good relations so that you can actually exchange
experience and costs so that you can get the best deal? Secondly,
I just wonder whether you recruit negotiators from the private
sector both from the industries you buy from and indeed professional
negotiators and what sort of training do you give people internally
and how much accountancy support do you have within your system
or how many accountants in fact? There are a few questions there,
benchmarking, private sector involvement, training and accountancy.
(Sir Robert Walmsley) As to benchmarking with
foreign countries, of course we get a good chance to do that in
collaborative programmes when we are buying the same thing in
different places, so we do get visibility of it. In terms of producing
anything remotely equivalent to this Report from other countries,
it is a source of great regret that I am unable to find such information.
It is not available. I am not sure whether it is just a question
of not wanting to give it to us, but it simply is not comparable
and available. I wish it was. I would very much like to be able
to benchmark my own organisation against other organisations undertaking
similar work and we have tried very hard to do that. In terms
of bringing in negotiators from outside, I think that is a very
serious point and I would be very happy to contemplate doing that.
What actually happens is that outsiders take my negotiators away
from me. The traffic is one-way in the other direction and I have
lost some of my most capable people to industry. This is because
they are competent and they drive a very hard bargain. In terms
of number of accountants, we have increased those very substantially
lately because of resource accounting and we have had to do that,
so we have recruited people from outside at all levels because
we are not taking them in, so to speak, in their first job in
their career. For example, my chief accountant was a new recruit
from outside and we openly advertised and we got him and we are
very pleased with what he has contributed to our work. All of
that is not to deny the absolute need to continue the professional
development of our people. Things like Earned-Value Analysis is
a type of skill, things like sending people on a negotiating course
are absolutely necessary, but I am pretty confident that my negotiators
are very, very competent and are viewed as such by the defence
industry.
Mr Love
84. Good afternoon, Sir Robert. Both of
us have made some sacrifice to be here this afternoon, although
I understand it may not have been such a great sacrifice because
Brazil are now 2-1 ahead. There has been a theme running through
this afternoon in terms of the relationship between the procurement
agency and individual companies, but also the defence company
as a group and you said earlier on that you are trying to establish
a relationship of trust and openness in one form and an arm's
length, adversarial relationship and you also mentioned that you
recognise that for a lot of these companies, their contracts with
you allow them to carry out the research and development which
then allows them to make use of that technology to sell abroad.
Can you encapsulate for me the very complex— because I think
this lies at the heart of what we are talking about this afternoon—relationship
that exists between your agency and the private sector defence
establishment?
(Sir Robert Walmsley) I think it is important
that we communicate as much information to our suppliers about
our plans and programmes as possible. We do that in many formal
sessions. Once a year I have an equipment session, or my subordinates
do, with industry. We take care to visit industrial suppliers
and they remain in touch with the operational requirements staff
of the Ministry of Defence, so they have got a good idea of our
long-term programmes, so there should be no surprises to the industry
about the type of thing that we want and the time when we expect
to require it. When it comes to characterising the relationship,
that is openness, that is plans and programmes. I think we have
a tough relationship in terms of negotiating contracts and I do
not think I would want anything else, but I think you can have
a tough relationship as well as having trust because when you
are in a NAPNOC situation you have to have equality of information.
That does require trust. I think we do get that in many cases.
Where I am seeking to deepen our relationship is during the execution
of a contract. That requires us to supply things to industry.
That is always the case. We supply test ranges, we supply people,
we supply trials facilities and all that means that we have to
understand industry's needs and they have to understand our ability
to supply them and we need to work with them to make sure that
is smoothly executed. I think there is quite a rich seam to be
mined there and I think IT links between us and our suppliers
on particular projects are beginning to demonstrate that openness,
that constructive relationship.
85. Do you think you have an objective that
takes into account the health and strength of the defence industry
in this country?
(Sir Robert Walmsley) It is an indirect objective.
It is my primary objective to obtain value for money and satisfy
the needs of the armed forces.
86. Is there any inconsistency? I understand
the Defence Export Support Organisation exists almost entirely
to promote exports for the defence industry and how both successful
and critical to Britain's manufacturing effort that is is an important
national objective. I am not decrying that. Does that mean there
is an inconsistency at the heart of your organisation, because
on the one hand here we are promoting exports and on the other
you are trying to achieve value for money?
(Sir Robert Walmsley) I very much hope not. I
think competitive contracting—and I do apologise for continuing
to repeat this—puts industry absolutely on the spot to improve
their performance and I think it is our insistence on competitive
contracting wherever possible that has done a great deal in achieving
the international success of the defence industry which does bring
direct benefits to us to reduce costs of overheads as well as
direct royalties on things that we pay them to develop.
87. I think that is a sentiment that the
National Audit Office would certainly share. Can I refer you to
figure 13, page 29. That shows a rather weak relationship between
cost variants and in-service delays. It is not a very strong relationship.
Is that a reflection of your tendency to under-estimate both the
costs and delivery dates because of all the pressures that are
on you in terms of budgeting?
(Sir Robert Walmsley) The fact is that a contract
can very tightly bind the costs. A contract is very difficult
to arrange to a very tightly bound time. Given that we attempt
to do both in our contracts it is not surprising that quite often
it is `time' that seems to give more than cost. I found this diagram
extremely interesting but quite difficult to know what it was
telling me. I know that short-term contracts——
88. I think the argument is that if you
are under-estimating both on cost and time, and it may be different
amounts that you are under-estimating, then you end up having
to re-programme and the rationale will be that at the end of the
process when you have taken in a realistic order of both the cost
and the time you will end up with more additional cost than you
would have had if you had actually done a proper realistic estimation
at the beginning of the process. Would you accept that as a criticism?
(Sir Robert Walmsley) I certainly accept the criticism
that if we do not start with realistic estimates of time and cost
we will end up failing on both. I think we have to look very carefully
at our past achievement in working with particular contractors
and decide whether our joint predictions of the outcome of a particular
contract in the past have remotely satisfied what we each convinced
ourselves would be the situation before we began. In other words,
we need to start looking very carefully at past performance before
we place another contract.
89. Can I take you up on that because you
mentioned earlier on, and I took it down, when someone asked you
about the two main areas of an increase, programme costs and inflation,
you said "better ways ahead", and those were your exact
words and I took them down. Now, I am going to ask you about the
programme costs, but in terms of inflation, inflation is shown
through as a major issue for the last five years since this Report
began and you have been well aware that the GDP deflator does
not recognise all the additional costs, so what confidence can
we have that your inflation estimates are going actually to be
more realistic in the future when you have had five years without
it being corrected?
(Sir Robert Walmsley) Well, I can only say that
the first thing to do, which is a very simple thing to do, is
to issue costing instructions to my staff so that they make correct
predictions of the effect of the indices that we use in our contracts.
It may strike you as an extremely obvious point, but it is something
that we have only very recently done in a systematic way, so there
will be no surprises any more as a result of the indices that
we use.
90. Can I stop you there because that is
one of the recommendations that the NAO Report makes and so I
would hope that when you report back next year, you can give us
some more confidence of that. I know it has been taken into account
in a number of contracts, but can we assume then that that will
be something that will be normal practice in your Department in
the future?
(Sir Robert Walmsley) I jolly well hope it will
be. Instructions have been issued and it will be my job during
the next year, together with my senior staff and together with
those who must enact it, to make sure that this is followed through
I am confident that that will happen, but that is only, so to
speak, keeping the score. The next thing which is at least as
important is to try to remove some of these rather, I believe,
onerous effects of inflation in defence procurement, and I think
we have got three things we need to look at. We need to look at
a longer period for firm price, non-variable price. We should,
as a rule, be aiming at five years, though there are occasions
when we do not want to do that. A telecommunications contract
should be variable price from day zero because the price index
of telecommunications is reducing, so when we contracted for the
defence fixed telecommunication system, we went VOP from day zero
because that was the right thing to do, but for other things we
should be attempting to go for five years or whatever represents
best value for money. The two other elements are to try to use
output indices and I have explained that we have moved in that
direction on the Eurofighter, which is fundamentally important
because of its size. The third thing is, where it is right, to
increase the proportion of the contract price which is never subject
to variation and on public-private partnership contracts where
in a sense somebody has bought a big facility and is then asking
me to pay back the interest charged on the money he has invested
in it, we have achieved as high as 70 per cent of a 20-year contract,
70 per cent on average, through the life of the contract as non-variable.
So it is those three tools, longer firm price, output indices
and a bigger proportion subject to no variation, and to always
treat this in the round as a deal and to go for best value for
money.
91. What about your programme costs? You
mentioned earlier that you were going to find a better way ahead
in terms of programme costs.
(Sir Robert Walmsley) There is a sense in which
this feeds on itself because programme changes are one of the
instabilities which will rattle through all this so that if we
are taking life in the new strategic environment where I very
much hope we are not going to be driven to absolute performance
standards, to absolute military imperatives for in-service dates,
but have a greater degree of judgment in taking those. I would
hope that the need for programme changes will reduce.
92. Earlier on you mentioned the technical
difficulties and presumably that is the major area of difficulty,
but it was mentioned earlier about trying to get a stronger handle
on the sort of technical difficulties that you currently have
difficulties with and the idea of incremental acquisition is one
that I know you have been looking at very carefully. Are there
ways ahead that will allow you to have much greater control over
those technical problems?
(Sir Robert Walmsley) I am not sure that I would
want to do that. As I mentioned before, these technical difficulties
arise in the industry and if industry freely enters into a contract
to deliver an article to us and our advisers think this is a reasonable
proposition and we place the contract and industry then stumble
across some huge new problem then we rely on the fact that they
are a competent contractor to get their way out of it, but it
is absolutely right that we also take an objective view of the
rationality of their proposition and I think we need to be very
careful before we rather complacently say that it is nothing to
do with us. So I think we need to develop our people so that they
can take a rational view of the proposition and part of that will
come from taking past performance into account, ie, how we got
on last time.
93. I suspect the word "control"
was inappropriate in that sort of context. We all know that all
contractors are not in a perfect relationship with you and if
they have under-estimated on their contract costs then a programme
change can be a way of making up for any under-estimate that they
may have made at an earlier stage. That would depend critically
on your staff being able to tell the difference between a technical
difficulty that they had and the real thing. I just wonder whether
you feel that at all stages the expert advice you are getting
on technical problems is up to the task of ensuring that it is
a genuine technical problem and not something that your contractor
trying to do to increase his contract cost.
(Sir Robert Walmsley) I think it would be dangerous
of me to sound too confident that we always got that right because
otherwise some of these difficulties presumably would not arise.
What I would say is that programme changes are nearly always the
responsibility of the Ministry of Defence reacting to changed
circumstances. I explained one of the reasons why I hoped that
would not arise. Technical difficulties happen in industry and
insofar as we have been unable to foresee them or have not taken
into account the fact that industry have always had problems in
this particular area then we should do more of that before we
place a contract with the industry. Programme changes are much
more in our own gift.
94. Can I just ask you one question following
up on a point that Mr Leslie made earlier on in relation to NAPNOC
and that is this issue about "should cost". That depends
critically on you having the information on what an efficient
company would actually charge for that. Highlighted in this report
is the inference that perhaps the information available to the
MoD was not up to the task that you are setting yourself in NAPNOC.
Is that the case? Secondly, are you taking on board the experience
that you have learned in relation to future NAPNOC contracts?
(Sir Robert Walmsley) Very much so to the last
point. For instance, the Project Director for Bowman is very much
in touch with the Project Director for Astute submarines to make
sure that the lessons learned from the Astute are transferred
into Bowman. This would have been virtually inconceivable before
we all went to Abbey Wood and were a unified organisation. As
to whether we need better information, I really have never met
a situation where we did not need better information. We do have
good information about the costs in shipyards and the way we do
it for other parts of NAPNOC contracts is to ask for a subcontract
competition. There is a sense in which that delivers the right
answer by definition. Where I think we get into most trouble is
where we are not familiar with the industry, where we are looking
at antiquated plant that has not received investment and it is
very difficult to make a judgment as to whether the factory concerned
is operating at anything remotely like the world's competitive
standard. I first of all therefore look to see if they are selling
anything to anybody else and if they are not then we have serious
doubts.
Chairman
95. Just a last few things, Sir Robert,
before we release you. The first thing is that I would not want
the Committee or you to be under any misapprehension on some of
the figures you have cited, particularly as you referred to this
as your end of term report. Figure 11 shows what you refer to
as the percentage reduction in the variances and I think the whole
Committee should understand that that arises because of the increase
in the size of the project base and since the new projects tend
not to have overruns in them, this is as much a reflection of
that as a performance measure, I think.
(Sir Robert Walmsley) Yes.
96. Similarly, in terms of overrun, the
same argument applies. The Report actually says that the 20 projects
that are common in 1996 and 1997 show an increase in overrun,
not a reduction, so I just wanted to make sure that everybody
understood quite clearly where that came from. I want to ask really
two last questions on the Report before we let you go. Some 15
years ago I wrote an article for the Harvard Business Review
about cost overruns of high-tech projects and there was an awful
lot of statistical analysis, but I could summarise it pretty much
as the better technically specified the project, and the better
the understanding of the technology involved in the project, the
better controlled the outcome will be at the end of the day. It
was almost a linear relationship between the poverty or poorness
of a technical specification and understanding and the size of
the overrun at the end of the day, and technically doubling was
not unusual in high-tech projects. Now, reading through this Report,
not the text, but just reading through the elements of this Report,
I get the very strong impression of a large number of projects
which are modified along the way and no amount of contract negotiation
will protect you against that. If you modify the project later
on, then you are going to carry the cost or, rather, the taxpayer
is going to carry the cost of that. It is a sort of technical
incompetence of a sort and even the Apache, a well-designed aircraft,
has obviously had bits latched on to it. It is like reinventing
the wheel or adding a fifth wheel. Would you say that there is
in this what looks like indiscipline and lack of foresight on
the technical front or a culture of acceptance of late modification
of projects in your Ministry?
(Sir Robert Walmsley) I think we have tendencies
in that direction. I have to say that they are always argued,
in my experience, carefully and increasingly with a view towards
rejecting changes because there is no doubt that not just in cost
terms, but in terms of the management intensity and complication
which arises from accepting even small changes, you reap a pretty
bitter harvest. That is why I mentioned before that I have started
to introduce a proportion by value of changes due to specification
alterations as part of each project's own measurement of its own
performance. With a ship like HMS Ocean, we have achieved
it in the low units of percent, though I cannot remember the exact
figure, but for Abbey Wood itself, so far we have not exceeded
the 2 per cent on project costs, so we know that it can be done
and we very much recognise the benefits of doing so.
97. I said at the beginning that we may
look at this slightly differently next year and that may be a
statistic we will be looking very closely at next year. My last
question is really rather an open-ended one and I do not want
you to replicate for us the advice you give to Ministers because
clearly you cannot do that, but perhaps you can finish by giving
us a one-minute thumb-nail sketch of which aspects of the problem
we are talking about here you think "Smart Procurement"
would help you deal with.
(Sir Robert Walmsley) I very much hope it would
lead to an improvement in all aspects. I would be surprised if
you expected me to say anything else, but working with industry
and integrated project teams should all help us to deliver things
more quickly. Openness with industry should help to avoid surprises
and should help to avoid cost increases because we would both
be aware of the technical risks that we were taking on, so if
Smart Procurement does not start to attack the problems in here,
then I think it would have failed. The only point that I would
add, Chairman, is that projects which are a long way down the
track are simply not amenable to as much change as projects which
are starting and it will take time and that is why I was talking
about the fact that there are statistics in here, like the delay
to the introduction of Cobra, which I am stuck with and which
my people are stuck with. And there is nothing I can do to eliminate
the Spearfish delay where the in-service date was achieved in
1994, but it is still buried in these statistics as being several
years late, but Smart Procurement will increasingly help to attack
all these difficulties as well as the development of my people.
That is part of it too.
Chairman: Sir Robert,
thank you very much indeed. It has been an interesting session
and I know a very heavy-duty one from your point of view, but
thank you again for your evidence.
|
