Examination of Witnesses (Questions 220-239)
DEPARTMENT FOR
ENVIRONMENT, FOOD
& RURAL AFFAIRS
AND RURAL
PAYMENTS AGENCY
MONDAY 9 JULY
2007
Q220 Mr Curry: Because they never
knew where any claim was in the process.
Johnston McNeill: Yes. And we
moved away from a high-class service, with one-to-one contact
at regional level with staff that customers had perhaps worked
with over the years—not in any inappropriate way. So it
was a significant culture change, coming through a separate customer
service centre. I think it was the fact that it was the first
time we had actually operated a brand-new scheme on a brand-new
system, using a brand-new technique, and we had not got the experience
to profile out what would be a normally acceptable progress of
clearing up the tasks as they were raised.
Q221 Mr Curry: Right. The one thing
that the Government have huge experience of is computer schemes
that go wrong. After all, your catastrophe was minor compared
with the tax credit debacle, which is still going through MPs'
surgeries right across the country. Given that so many of the
elements were high-risk because of no experience, who at any stage
said to the policy makers—I suppose that that is a euphemism
for the Secretary of State—"A huge amount could go
wrong here"? Are you conscious that anybody said that?
Johnston McNeill: All I can say
is that you have had the Permanent Secretary, Sir Brian Bender,
here talking openly about 40% confidence levels, which was a less
than an even chance of the project being delivered, about 50%
confidence levels being sought to try to give us some more reassurance
and about a 70% confidence level as a probable peak. We Reported
to the Department that this was extremely high risk. We monitored
the risk very carefully. I do not know whether the Committee has
had a chance to see the risk Reports, which were top class. They
were assessed by the OGC and found to be in line with best practice.
We noted that.
Q222 Mr Curry: But the Department—
Johnston McNeill: The Department
was fully aware. As the Permanent Secretary himself said in evidence
to the EFRA Select Committee, the risk levels were less than evens
for delivery.
Q223 Mr Curry: Yes, but even though
that was so, the Department sat there and watched it.
Johnston McNeill: That is not
the case. Sir Brian Bender was fully engaged, and we were all
working to try to identify the best way and the means by which
to deal with the matter. I have mentioned the discussions with
Accenture, the manual work-arounds and the ways of working
that we were looking at to see how we might resolve things. We
were working very hard together, in partnership, trying to find
out what else we could do to improve the level of confidence in
delivery.
Q224 Mr Curry: But the problem seems
to be that, somehow, a train crash happened with none of the passengers
realising. At some stage, somebody must have said, "Things
are going badly wrong; we're not going to be able to deliver."
Somebody presumably said to Ministers at that time, "This
is the point at which you need to bail out." In fact, the
Department removed the management system from the programme, didn't
it? At one stage it cancelled the actual bit of the computer programme
designed to identify things going wrong.
Johnston McNeill: No. First, I
was the one who informed the Secretary of State at the meeting
on 15 March—having written to Lord Bach on 10 March, and
having started payments, as I say, on 20 February—that I
could see that we were not, unfortunately, going to be able to
make the bulk of payments by the end of March. That resulted in
the meeting on 14 March, which resulted in my being stood down
on 15 March. Eventually, it became clear that, unfortunately,
despite the fact that we had defined entitlement and payments
had started, we were not going to achieve enough to get the bulk
of the payments out.
Q225 Mr Curry: So trying to sum up
again, because I must finish, if you could identify one crucial
thing—the pea as in The Princess and the Pea—
would you say that it was the task-based computer system? Would
you also say that instead of that being ascribed to personal failure
by you, it should more effectively be described as an institutional
failure by DEFRA?
Johnston McNeill: There has not
been, in any of the Reports I have read, and nor is there in the
NAO Report, any personal criticism of me. In the Office of Government
Commerce Reports, the leadership on the programme was constantly
noted and praised; the relationship with staff was identified
as being excellent. I am not aware of any personal criticism of
me whatsoever. I maintain the point that we tried our damnedest
within the RPA and DEFRA to deliver this. We worked pretty much
seven days a week, particularly in the last six months, and it
did not happen. Whether you want to call it institutional failure,
I do not know, but certainly our inexperience in dealing with
a task-based approach was a fundamental factor.
Q226 Mr Curry: Finally; the matter
is still going on now, because it is still not right. I am still
getting letters from the people who act on behalf of farmers,
who claim that even simple things that are supposed to go right
are not going right, and that there are huge misunderstandings
between the people at the RPA and the farmers out in the sticks.
This train crash was not just a derailment, it was a fairly heavy
pile-up.
Johnston McNeill: I do not see
this quite as a train crash.
Q227 Mr Curry: I am following a trains
analogy.
Johnston McNeill: Sorry; the train
was not destroyed. You now have a rural land register that was
built as part of the investment and that contains 2.4 million
parcels of land. It is one of the biggest—and to be frank,
it is heading for being one of the best—in Europe. We have
gone through hell getting the additional land loaded on, because
farmers had not told us about the additional level before and
because we had 40,000 new customers. It is now pretty much there.
Things are back down, as I understand it, to about 10,000 changes
a year to the RLR, which was the historic level. You have got
customer registration, and it is the first time in the whole of
DEFRA's history that there is one registration system for all
the customers of the whole organisation. It has been built; it
is in place; and it works. Again, we went through hell with 40,000
new customers and all the information that had to be put in. That
work is now complete.
RITA, the RPA information technology application
that delivers SPS, was used again this year, and it will be better
next year. The optical character recognition was used this year
for the 2006 scheme—we could not use it for 2005, because
it was too much of a challenge for the IT system, which could
not maintain it. You have a document management unit, which means
that all the correspondence that flows around the RPA is electronic.
That is in and is working very well. It is not a train lying at
the bottom of a ravine, because many aspects of the programme
are there now.
Yes, we went through hell in the first year:
40,000 new customers, lots of land we had never heard of before,
even though we should have under previous common agricultural
policy schemes, and all the other challenges that we faced including
delay in policy. We just did not make it at the last fence. For
that I, if I may take this opportunity to do so, apologise profusely.
We worked our damnedest to get it done, and it did not happen
at the eleventh hour. I apologise for that, but many good things
have been delivered, and the investment is still there to be seen.
Chairman: That is a very full and fair
answer, Mr McNeill, from your point of view. Thank you.
Q228 Mr Dunne: I am interested, Mr
McNeill, that you referred to your meetings with the Secretary
of State. You, I think, identified that you had one meeting on
14 March, the day before your role was suspended. Could you tell
us when your first meeting was with the Secretary of State?
Johnston McNeill: I am sorry;
I cannot recollect that date. It was a brief informal meeting,
where I was actually at an Executive Review Group with Brian Bender
and was asked to join him in a meeting with the then president
of the NFU—I think it was Mr Bennett—to discuss certain
issues regarding SPS 2005.
Q229 Mr Dunne: Do you remember what
month that was?
Johnston McNeill: I am sorry,
I cannot remember, but I recollect that that was the first time
I met the Secretary of State.
Q230 Mr Dunne: So you had no direct
Reporting relationship. It was always done through the Department.
Did you receive queries from the Secretary of State or any responsible
Ministers direct, as to progress of the single farm payment?
Johnston McNeill: All the Reports
that were sent to the top of the office were also copied to the
Secretary of State's private office as they were to Lord Whitty
initially, and then to Lord Bach's office. We had regular meetings
with Lord Bach. At the latter stages of delivering this, he was
obviously so concerned that we were having pretty much daily contact
with him, updating him and sending him the operational stats in
terms of the number of tasks that were being cleared on a daily
basis—and the graphical representations of that and so on.
Q231 Mr Dunne: Thank you. There was
significant anxiety from representatives of the NFU in particular
about the level of errors or matters that were not being correctly
transcribed into the computer systems. I forget the technical
term that you were using at the time. I believe that you were
producing weekly Reports, which were made available to the NFU,
from about October or November 2005 onwards. Would you care to
comment on the Reports that you passed up to the Department, the
escalation of error messages and, in particular, the extent to
which items were being cleared off the error messages list? The
impression that I got from reading the NFU transcripts was that
errors were materially misstated.
Johnston McNeill: I can assure
you, Chairman, that there was never any intention on the part
of the RPA to misstate any information going forward to Ministers.
For example, we copied in all of the OGC Reports untouched. We
were not under any obligation to do so, as they are normally for
the Chief Executive's eyes only. We forwarded on untouched all
Reports from independent consultants and others, and other information
went up, again, without any influence being applied to it. I was
not aware of that problem.
I am sorry, but this happened some time ago.
I cannot quite get my mind on exactly what you are referring to.
It could be that we were Reporting on the clearance of level 1
validation strikes around that time. I can only think that your
question is about that, but I am sorry that I just cannot remember
the details. We are talking about events of 15 months ago.
Q232 Mr Dunne: Indeed, we are. One
of the issues that caused a lot of problems was the mapping exercise,
which you touched on earlier. Is it the case that the mapping
exercise was initially done on a two-dimensional basis, rather
than a three-dimensional basis?
Johnston McNeill: I am sorry,
but I just cannot respond to that level of technical detail. All
I can say is that I understand that the rural land register was
built to a very high standard. In fact, the criticism has been
that perhaps the tolerance levels that we were working to were
too high, although now that the investment has been made, that
criticism has gone away.
Q233 Mr Dunne: Table 5 on page 21
of the Report states that the costs of the rural land register
increased from an estimate of £6.8 million to out-turn costs
at the end of March 2006 of £16.1 million. The footnote says
that those costs covered the mapping exercise, so there was a
significant increase in its cost. My point is that most fields
are not perfectly two-dimensional—most are three dimensional.
If you are determining historic claims and you measure hectarage
or acreage on a two-dimensional basis, you will obviously get
a different measure than if you were to measure on a three-dimensional
basis. Many of the complaints that I have received from constituents
are about land that was incorrectly mapped because the Agency
did not take a three-dimensional measurement. Are you familiar
with that issue?
Johnston McNeill: I am familiar
with the issue, obviously. In terms of actual surface area, if
a field is flat, that is one thing; if it has a hill in the centre,
that is another thing. To the best of my knowledge, we were able
to take account of that with the mapping system that we were using.
Q234 Mr Dunne: From your recollection,
was that issue raised with you as a problem?
Johnston McNeill: I do not recognise
it as having been a significant issue. I was about to say that
the reason for the additional cost is that we originally mapped
1.7 million parcels of land, but, on conclusion of building the
rural land register, the actual number of parcels of land was
2.4 million. In addition, we had some 100,000 IACS22s notifying
us not only of additional land but of changes to land that farmers
wished to make us aware of. Those changes may or may not have
interfered with the area measurement, but farmers wanted to take
the opportunity to reaffirm particular boundaries, to identify
additional fence lines or whatever. I am sorry that the issue
of taking a three-dimensional view of a field does not spring
to mind as something that was ever raised with me at an industry
forum or elsewhere.
Q235 Mr Dunne: I am surprised to
hear that, because many of the complaints coming to me are about
that.
Johnston McNeill: It may well
have been an issue for some of your constituents. All I can suggest
is that, obviously, there are still experts in the RPA who would
be happy to address the issue for you.
Q236 Mr Dunne: With the benefit of
hindsight, do you think that it was appropriate to undertake a
complete remapping of England in the midst of all the other changes
that were being made?
Johnston McNeill: There was absolutely
no choice. It was an EU requirement that the UK have a digitised
mapping system for 2005. It had to be done. It was then attached
to the change programme, and there was a move to a task-based
approach following the PricewaterhouseCoopers review. It was something
that just had to be put in place.
Q237 Mr Dunne: Did Scotland and Wales
undertake the same exercise?
Johnston McNeill: Scotland and
Wales had actually prepared a digitised map some years earlier.
Q238 Mr Dunne: They already had one?
Johnston McNeill: They had one.
They had made the investment some considerable time before. We
were still using paper maps when Scotland had a digitised map
system.
Q239 Mr Dunne: Do you think it was
appropriate that the maps should be handled by one office of the
Agency while the claims were handled by another office?
Johnston McNeill: Actually, we
had this debate at some length at the time. We started by having
a mapping unit at each office, so that we could bring local knowledge
and context to bear in dealing with farmers on mapping applications.
The difficulty that arose from that was in the performance that
was required and in obtaining a sufficiently centralised approach.
At the end, we had to bring in Infoterra, the mapping organisation
that had put the original 1.7 million maps on the system under
contract, to deal with the large and unexpected additional demand
for mapping. At that time, it was decided that it would be much
better to have things in one place where they could be supervised
and where Infoterra staff could work closely with our staff.
| 
