BEIS0000384 - Submission from David Sibbick to Secretary of State re Horizon

Evidence on official site

BEIS0000384
BEIS0000384

‘@ los.

CONFIDENTIAL - COMMERCIAL

To: ci: Sir M Scholar

SECRETARY OF STATE _} separate bie Meedonala

MRMCCARTNEY _} copies Roars CGBPS
Mr Macintyre cil

From: Mr Fraser IBB

DAVID SIBBICK Dr Hopkins cll

DIRECTOR POSTS Dr Sklaroff CoM

Mr Osborne Legal C
151 Buckingham Palace Road Ms ton PORT
 ApNn I Mr Whitehead CGBPS 1
GRO L Me, kane Ms Anderson CGBPS1

12 February 1999 2 Me Mr Corry SpAdv

HORIZON

1. At your meeting yesterday with the Chief Secretary, Lord Falconer and the
Secretary of State for Social Security you were handed a paper setting out a possible
way forward for the Horizon project following the discussions between the Treasury
(Steve Robson) and ICL (Keith Todd).

BA/POCL AUTOMATION PROJECT,

Issue

Recommendation

2: That you reply to the Chief Secretary in terms of the attached draft, accepting
that the Treasury proposal has a number of attractions, but arguing that the prolonged
period of further uncertainty that it would entail, the real risk that at the end of the day
all the necessary elements might not come together in an acceptable way, and the fact
that we may very well have to defer publication of our White Paper on the future of
the Post Office until this issue is settled, still cumulatively argue in favour of

continuing with the benefit payment card (Option 1). The issues are however detailed
find ad ina brief di ion with

EF URGENT. Your colleagues have pressed you for your reaction to the proposal
by close of play today.

DATEMP\TOZ\SOSBRFOS.D0C 16
BEIS0000384
BEIS0000384

CONFIDENTIAL - COMMERCIAL

Summary

4. The Treasury proposal, if all the elements could be brought together and made
to work, offers POCL most of the advantages of the benefit payment card, together
with some additional and potentially valuable wider advantages (including early and
excellent positioning in the smart card market). It is also a better fit with wider
Government policy objectives including social/universal banking, single Government
account, and electronic Government. There is little doubt that ICL could be persuaded
to go along with it if the price (which would need to include recovery of most of the
£250 million of what under the Treasury proposal would become nugatory expenditure
on development of the bpc so far) could be made sufficiently attractive.

5. If we were starting today with a clean sheet of paper, the Treasury option offers
the promise of a number of clear advantages over the bpc option (Option 1). But we
do not have that luxury. We have instead on the one hand Option 1, where a great deal
of the development work has already been done and independently validated, and
where the technical, commercial and financial issues are known and understood. The
whole process could be put into gear and moved forward rapidly, ending months of
uncertainty and effective planning blight for the main protagonists.

6. Against this on the other hand is the Treasury option, which offers a number of
potential advantages, but which represents in a number of respects a huge leap into the
unknown, and thereby extends the uncertainty and planning blight for a minimum of
several more months. It is far from clear, for example, whether POCL could operate a
Post Office bank account - whether a physical account with a commercial banking
partner, or some form of “virtual account” within Horizon, without seriously damaging
its prospects of becoming an agent for all the main banking groups to provide front
end banking. A fundamental renegotiation of contract terms with ICL would also be
needed, and ICL’s price at the end of the day might be more than we are prepared to

pay.

7. An important related consideration is that given the central importance of the
Horizon project to the future of POCL, it will be difficult, and may in practice be
impossible, to publish our White Paper on the future of the Post Office until we are
clear on the way forward for Horizon. Treasury officials agree with this conclusion.
Adoption of the Treasury option could therefore delay publication of the White Paper
by several months.

Detail

DATEMP\TOASOSBRFOS.DOC 26
BEIS0000384
BEIS0000384

CONFIDENTIAL - COMMERCIAL

e We agree that the proposal to migrate benefit recipients from a payment book to a
Post Office credit-only account accessed by a Post Office smart card useable,
initially at least, only at post offices should keep as many recipients - and therefore
the footfall that is so important to subpostmasters - within the Post Office system as
migration to a benefit payment card (bpc) would do.

© The proposal would also give POCL entry into the smart card market at least two or
three years earlier than would migration via the bpc, and with a significantly larger
installed base of perhaps 15-20 million cards containing an important live
application. It is unclear how the market for smart card applications will have
developed by then (and also unclear which of the competing technologies is likely
to win out), but at least POCL would be superbly positioned in it.

e The proposal would also contribute substantially to the achievement of wider
Government objectives including universal banking, the reduction of financial,
retail and iT exclusion, and the creation of a channel for the delivery of Government
Direct integrated services into communities not served by other organisations.

BUT

© The proposal represents a huge leap into the unknown. It would be a major initial
task for POCL to scope and research the optimum bank account. What kind of
account would carry least risk of alienating the commercial banks, for all of whom
POCL has ambitions to offer front end banking? Should it be a physical account
with a commercial partner, or a “virtual” account within Horizon? If the latter, how
long would it take to develop the facilities which presumably would need to include
such basics as direct debiting? How would the financing work, and would it offer
POCL the prospect of a viable business case?

¢ Could an acceptable commercial deal be done with ICL? From every viewpoint it
would be essential to go ahead with rolling out the Horizon infrastructure rather
than waiting until all other issues have been settled. ICL’s insistence (para 13 of the
Treasury proposal) on splitting the contract is not unreasonable given the
uncertainty over the precise scale, nature and timing of the replacement (for the bcp)
services. They will clearly expect the initial contract for the core infrastructure
work to incorporate significant reimbursement for the £250 million spent on the
bpc, and would then expect to negotiate separately for the addition of new services
once these had been defined. The overall price might be more than Government is
willing to pay.

e Given the length and complexity of the research and negotiations that would be
needed before decisions could be taken on the Treasury’s suggested way forward,

DATEMP\TOZSOSBRFOS.D0C 368
BEIS0000384
BEIS0000384

CONFIDENTIAL - COMMERCIAL

and given No10’s insistence that our fallback position should be to proceed with
Horizon with the bpc, we would almost certainly need to ask ICL to continue work
on the bpc, at least until we were clear that the alternative way forward could be
made to work at the technical, commercial and financial levels. ICL would take the
view that if we wanted this level of insurance, we should have to pay the premium.
But it could also limit in a practical sense the resources that ICL could devote to
developing the new concepts in association with POCL.

e Two further and related issues are first that given the uncertainty of the nature,
scope and timing of the new proposals, ICL would be most unlikely to agree to take
this work forward on a PFI basis, but would instead require some form of supply
contract. And second, the fundamental changes from the original contract raise
serious questions about the application of EU procurement laws, and whether a full
re-tendering for the work would be needed.

e There would be a whole raft of subsidiary issues to be discussed and decided. An
important one would be POCL’s relationship with BA. Who for example would be

responsible for covering losses, whether fraudulent or otherwise, within the system
once payments had been handed over to POCL?

DAVID SIBBICK

DATEMP\TO2SOSBRFOS.DOC +