APPENDIX 2
Supplementary memorandum submitted by
Sir Hayden Phillips GCB, Permanent Secretary, Lord Chancellor's
Department
PURPOSE OF
LIBRA
1. A brief history of the Libra project
is at Annex A. This note describes the reasons behind recent
events and the current position. Annex B provides information
on the technical infrastructure in response to the question about
equipment installed in offices.
2. The contract with FS (Fujitsu Services,
previously ICL) was a PFI service contract which included:
— a national IT infrastructure (a comprehensive
set of office services including desktop PCs and printers; Local
Area and Wide Area Networks, full online support; and other services
including disaster recovery);
— office automation facilities (including
standard office software such as e-mail, word processing, spreadsheets,
diaries, etc);
— a standard national application and
related services to support court work (principally case progression,
reception of parties, fine accounting and enforcement, in-court
computing, management information) to replace the existing three
legacy systems currently in the MCCs;
— Direct electronic links with the
criminal justice agencies and their strategic systems (police,
CPS, Probation, Prisons, Crown Court, DVLA).
3. By late 2000 (following the contract
renegotiation described in Annex A) the first two of these
(including secure e-mail) were being satisfactorily installed
and the last two were under development. The critical date for
the application software was July 2001 when it was due to be installed
and proved in Suffolk MCC.
4. It became apparent around that time that
development of the core application was falling behind schedule.
At first FS believed that this delay was containable and various
options for postponing part of the software by a few months were
discussed. Indeed FS delivered a substantial part of the software
into their testing area in January 2001. Close examination of
this software showed that its quality and completeness was such
that the July 2001 date would not be achieved.
5. It was at this stage that FS completely
changed their project management team and carried out a full internal
investigation into every aspect of the contract and their ability
to deliver. They reached the conclusion that, from their viewpoint,
there were serious problems with the contract. They estimated
that full delivery of the application software would take up to
a further two years, that costs would be far higher than anticipated
and that the whole process of requirements definition and development
had to change.
6. In the spring of 2001 FS approached LCD
and indicated that their estimated potential losses were so high
(figures between £100 million and £200 million were
indicated) that they could not progress with the contract unless
it was substantially renegotiated. At this stage rollout of the
infrastructure was well under way but the company had not yet
reached the contractual delivery date for the software application
so they had not yet formally defaulted on the contracted.
7. In these circumstances there were only
two options possible for the department—
— Insist on full delivery of the service
as contracted;
— Open discussions to see if an acceptable
solution could be found.
8. FS made it clear that they would not
proceed with the contract rather than accept the first option
and, given the complexity of the situation which needed extensive
analysis and discussion, there was no real alternative but to
investigate the possibility of re-negotiation. Making full use
of professional expertise both internal and external, the matter
was considered throughout the summer and possible solutions were
discussed with FS.
9. The outcome of this work was a framework
for a possible way forward in which the scope of the software
application would be reduced, the period of the contract reduced
and the delivery date put back. FS estimated a price for a revised
contract on this basis (around £283 million), which appeared
to be justifiable.
10. However this framework would require
months of work to drive down to the detailed level, fully consider
the implications, obtain a price and negotiate. The main elements
were therefore built into a Memorandum of Understanding (MOU).
Following extensive internal discussion and approvals, this was
signed on 5 October 2001. While legally binding, the MOU only
represented a basis for negotiation—not an agreed solution.
Part of the MOU was that:
— Work would continue on rolling out
the infrastructure and developing the software application;
— No remedies or other actions would
be initiated;
— LCD would share FS costs for the
period of the MOU.
11. The first months of the MOU work concentrated
on refining and agreeing the very detailed definition of requirements
to ensure that every detail was complete, accurate and fully understood.
At the same time the commercial aspects were progressively explored.
12. In parallel with these negotiations
the Department commissioned and carried out a range of studies
using independent experts including:
— FS technical capability and financial
standing;
— An assessment of risks;
— Benchmarking of costs for similar
work in other organisations;
— Development of contingency options
in the event that agreement could not be reached.
13. As the negotiations progressed, it became
evident that the cost of continuing with the development and implementation
of the software application were rising steeply. The initial FS
estimate for the revised scope and contract duration proved to
have been based on incorrect assumptions and clarification of
the requirements brought in additional costs. By February 2002
the requirements definition and clarifications had reached the
stage where FS were able to quote a firmer price. This was in
the region of £400 million, and there were additional costs
and risks, which had been transferred back to the Department.
14. Continuing work of development and testing
of the software application had also confirmed that delivery was
unlikely to start before May 2003 and there were measurable risks
of further delays, technical problems and cost increase.
15. Given this situation it was decided
that to go ahead with the development of the full Libra was just
not acceptable in either value for money or risk terms. FS were
therefore advised that we would not go ahead on this basis.
16. This decision was strongly opposed by
FS who had sunk costs on developing software in excess of £50
million. They proposed a number of alternative pricing and delivery
mechanisms but these did not materially affect the position as
decreases in FS charges were often cancelled out by increases
in LCD costs.
17. The work that had been done on the contingency
options had identified an alternative way forward in the event
of FS not delivering the Libra software application. Over recent
years, one of the current legacy system suppliers has updated
and enhanced their system to the point where, with extra development,
it could deliver similar functionality to that which the Libra
application would have provided. This application is already in
use across a third of the country and is thus tried and tested.
This would represent a much lower cost and lower risk alternative
to delivering through FS.
18. Once the negotiations with FS reached
the point that their total price for Libra was known, there were
three options. The estimated total costs for each option are
shown in bold:
— Accept the new price for the total
Libra system despite the fact that it had been assessed as being
neither value for money nor affordable. Allow FS to continue to
develop the software application despite the continuing risk that
it could be yet further delayed, could lead to further cost increases
and that there were technical issues yet to be resolved. Estimated
total cost £457 million;
— Terminate the Libra contract completely.
This would have left the MCCs with a partly rolled out infrastructure,
with minimal support and no opportunity to develop it until a
procurement had secured a new supplier, placing MCC business at
risk. It would inevitably have led to claims for damages from
both parties with a high probability of this leading to litigation.
Work on the application would cease. Because of the costs of procurement
and contract termination, this option would almost certainly cost
more than the third option and risks were high. Estimated total
cost £405 million;
— Negotiate a value for money deal
with FS for the infrastructure only and procure the software application
by building on the best of the legacy systems. Estimated total
cost £392 million.
19. It became clear that the third option
represented the lowest cost and the lowest risk while providing
MCCs with the services they required within an acceptable timeframe.
The subsequent negotiation with FS has been on this basis.
20. This conclusion has been subject to
extensive review by independent experts. As part of normal government
procedures it has also been reviewed by a government review team
and they have confirmed that the best decision has been made in
the circumstances. The business case has been approved by the
Treasury, and Office of Government Contracts (OGC) approval was
received following resolution of concerns about Parent Company
Guarantees and the use of the FS Flexible Finance arrangements.
It is clearly acknowledged that this is a considerably higher
cost than that in the original contract. But FS will not deliver
the original contract and the choice has to be made from the available
options.
21. The FS part of the option selected is
to continue to deliver, enhance and support the infrastructure.
Their total cost for this is £232 million. This price has
been benchmarked against infrastructure of similar size and type
across a number of public and private organisations and shown
to be within, but at the top end of the expected cost range. The
additional costs are for enhancing, delivering and supporting
the new software application.
CURRENT STATUS
22. We completed the negotiations with FS
on the variation to the original contract and the revised agreement
was signed on 23 July. An announcement was made by Yvette Cooper,
Parliamentary Secretary, Lord Chancellor's Department on 24 July.
23. Negotiations have begun with the chosen
legacy supplier, STL, to licence their software for national use
and to agree the necessary enhancements.
24. An advertisement was placed in the European
Journal on 30 August 2002 to begin the process of selecting
a Systems Integrator to host the new services and manage the migration
of MCCs onto the new services. Seventeen bidders have completed
the Questionnaire, which will allow a shortlist to be chosen.
PENALTIES AND
TERMINATION PROVISIONS
25. Technically, penalty clauses (clauses
which penalise a party in breach of contract rather than looking
to recompense the innocent party for loss suffered by it as a
result of the breach) are not enforceable under English law, and
as such the current Libra contract does not contain any penalty
clauses.
26. There are, however, a number of remedies
in the contract in the event of failure, including liquidated
damages and termination, depending on the default. In the case
of the core software application the liquidated damages could
have been applied from the date of failure to deliver at a rate
of £2,000 per day up to a maximum of 100 days. The Department
can, of course, sue for damages over and above this amount in
court. The real penalty for the supplier in failure to deliver
the application on time is the loss of revenue caused by the delay
and the extra costs of completing the work. Since this is a PFI
service contract, FS were not due to receive any significant payment
for their software development work until it was delivered. FS
have intimated that their costs of the software development to
date are in excess of £50 million.
27. The contract also made provision for
the Department to terminate the contract following a major breach.
Failure to deliver the core application did give the Department
the right to terminate. In that event, the Department could sue
for damages up to the Limits of Liability set in the contract
at £40 million. Equally, the company could counter-claim.
28. In agreeing to enter into negotiations
with Fujitsu, the Department suspended these provisions pending
the outcome of those negotiations. Failure to reach an agreed
variation to the existing contract would automatically bring these
provisions back into force.
29. It is part of the negotiated settlement
that the revised contract the Department has signed with FS waives
those provisions. Fujitsu will bear the sunk costs on the development
of the application, save for the £6.8 million already paid
for which we have free and unfettered use of the documentation.
| 
