To See hist, Page 1 of tb
Qrgunisation :
Fax I
Copy to
from ~~” Chns Nicholson Ref cn/mp/270
Qepurnment Governfnent Consulting, London-CW
Teleanone
Fax a
Date 16 ocrdber 1998 , he
i
Subject BA/POCL
_ telesse a draft. 1 am mindful of the very ught nmescales under which the working group 1s
POL00028638
POL00028638
10th Fidor
1 Canada Square’ Mina,
Canary Wnart he .
LonaonjE14 SAG etn
United Kingdom . Kenn
i
T arach as background {to Monday’s meenng the current draft of our work on BA/POCL. I
should stress that this is bemg released in a siate which I would not normally be prepared to
operanng and therefore jt is released as a working draft which cannot at this stage be relied-on
as a basis for decision,ibut which may help to inform discussion on Monday. My pnncipal
concems are that: I 2
m_ i¢has not been subject to any quality review.
H
e work has proceeded in parallel with the result that there are some
d in particular draw émennion to the following:
m of necessity some
izconsistencies, I
- we requested POCL to prepare banking profit assumpnions on the basis that under Option
2 a banking platform would be in place by October 2001, as in the ongina] report. The
view of our banking technology consultants 1s that this could only be achieved if there
were.no pilot. It] may therefore be more prudent to assume a date 6-12 months later
which will affect the banking profit line throughout;
- the more aggressive banking profit assumptions which we had assumed Compared to
POCL were not estimated to have any rmpact on the size of the network by POCL. We
consider that this is an excessively conservanve assumption, but has not been reworked.
The jnformation intms fax action taken in reliance on where a list of partners’
is confidential and may be
legally privileged , itis
intended solely forthe
addressee and others
authprised to receive it.
If you are not tne intended
recipient, any disclosure,
copying, distribution or
'
41S contents is prohibited *
ang may be unlawful.
Member firm of KPMG
International. The principal
place of business is at 8
Salisbury Square Londo
EC4y 8BB .
Names 1s Open to inspection.
KPMG 1s registered to carry.
on audit work and authorised
10 Carry on investment
business by the Institute ot -
Chartered Accountants in
England and Wales.
(oss
POL00028638
POL00028638
“ od6Q- P.OR/13 F723
Tk has not yet been passible to attempt to estimate, given the current figures within the time
aveilable, of the cost of the subsidy necessary to keep the network at the same’ size between
Options 1, 2 and 3. Wé hope to do this early Text week.
I would also stress thaj within the time availzble and wath the information-avarlable (including
the constraints caused by limmed access to ICL Pathway) some of the assumptions made have
hecn_somewhar.“hernii
GRO
Adam Sharples, HMT !
Sarah Graham, DSS}
David Sibbick, DTI i -
Jersmy Crump, CITU !
Gevff Mulgan, No 10 . -
Jonathan Evans, POCL! : :
Geerge McCorkell, Ba: - '
POL00028638
POL00028638
. Te-t0-95 18:17 FROM . T-469P.03/13” F+723
4
i
HM Treasury
Benefits Agency/Post Office
Counters Limited automation project
Analysis of fall-back recommendations
i}
16 October 1998
i
i
i
i
!
i
i
1
t
I
i
KPMG
. Reflen/dv280/hmrbrep/hmiforep
x
a
* sib-t0-g91e18
FROW-
1
I
1
. Contents!
1 Intrdduction
1.1 Background
1.2 Terms of reference
1.3 Qualifications
1.4 Work hndertaken
2 Feasibility of fall-back options
21 Viabiljty of early shift to benefit payment by ACT ©
2.2 Banking viability: common assumptions and issues
2.3 Option 2: commercial and technical feasibility
24 Option, 3: commercial and technical feasibility
3 Dischission of key assumptions
3.1 POCL inetwork modelling
3.2 Cost of banking technology
3.3. Banking profits
3.4 Key DBS assumptions
4 Sumtnary of financial results
4.1 Definition of options 7.
4.2 Pinancjal results
44° Summary of key assumptions
POL00028638
POL00028638
oor) P.04/13° F-723
HM Treasury
Benefits AgencytPost Office Counters Limued auomanon. project
Working Draft ~ 16 October 1998
TD ae See ee 2S
Ss uw
14
15
21
21
wv
Kan
POL00028638
POL00028638
+ tbetosag 18:19 FRO oo T46o P.05/13 Foi28
kei!
HM Treasury
Benefits AgencwPos Uffice Counters Lumsted automution project
Working Draft- 16 Oxtuber 1998
1 Introduction
1.1 Background t
In May 1996, the Benefits Agency (BA) / Post Office Counters Limited (POCL) signed PFI
j contracts with ICL Pathway (ICL). ICL would set up and maintain an automated infrastructure for
POCL to enable benefits to be paid to customers using a magnetic strip card and to provide a
plattiorm for other POCLibusiness. :
Ministers have been reyiewing the future of the project m view of the serious delays to the
I implementation timetable and ICL's failure to deliver a key.contracmual milestone for which ICL
I has been placed in breach of contract.
Mini:ters have decided, (without prejudice to sponsors’ legal rights under the existing contract, to
allow. a period of one month for discussion between the parties to see whether sansfactory
commercial terms can be agreed for continuing the project, outstanding differences on the
Timetable can be resolved, and a credible programme for full implementation can be agreed.
Ministers will need to sptisty themselves that the proposed way forward offers best value for
moncy, taking account of the costs 10 the public sector as a whole, of the alternatives,
Tn broad terms, the alternptives to continuing the project are:
™ to abandon the benefit payment card elemert of the project, but continue with the rest of the
Horizon system;
1
W to terminate the whole project and for POCL to commission alternative technology.
1.2 Terms of reference !
‘The purpase of this consittancy is to assist the Treasury with the analysis of fallback options. We
‘were required to review and assess four specific :ssues:
= POCEs nenwork modelling of the impact on the post office network of the loss of BA income ss
and claimants foorfall jrcome, in particular:
= extentof’ closures and losses if no subsidy were provided;
m extent of. subsifly Tequired if closures were to be avoided...
« viability of DSS and BA’s programme for an early shift to benefit payment by automated credit
traasfer (ACT), including the impact on the banks and their charges to customers;
® timescale and costs of implementing an alternative technology platform for POCL;
© validity (in both technical and commercial terms) of pursuing the Horizon proyect without the
benefit payment card. I
!
13° Qualifications i
The timescale within whidh this work has been curried out has been extremely compressed and we
have been heavily reliant on project costs und assumptions produced by BA, POCL. and ICL
Pathway, adjusted in a number of cases by KPMG. We must emphasise that the realisation of the
projeccions is dependent qn the continuing validity of the assumptions on which they are based .
'
1
i}
POL00028638
POL00028638
Tae P.06/13 F723
sf 1
IG! Hil Treasury
Benefits Agency/Post Office Counters Limited automation project
Working Dreft + 16 October 1998-
Thi: is particularly so as concems the amount and the timing of the cash flows and the effect on
cash requirements, They have been produced for the purpose of comparison of a preferred option
(option 1) against nwo fjlback options. If Government wished to proceed with either option 2 or
3, considerable further Work would be required to test these assumptions and produce a business
case before a decision isitaken to proceed with either option .
14 Work undertaken !
24.10 POCL’s network model
We held a number ofjdiscussions with POCL to understand the process of their nerwork
modelling. In specific afeas of the modelling we queried some of the assumptions used in the
network model, and asked POCL to provide further evidence to support their rationale. We also
nates that POCL’s inputs to Treasury were based on scenarios which were not entirely consistent
with the requirements of the Working Group, and so we requested POCL to’carry out some further
modelling to represent mpre accurately the impact on the network and their profits.
14.2 Early shiftto benefit payment by ACT
H
We investigated the feabibility of ACT migration plans with the banking mdustry, through a
limited number of telephone based interviews.
!
14.3 Feasibility: option 2. (ojizon without BPC) and option 3 (alteraative technology platform)
Our assessment of the technical and commercial feasibility of options 2 and 3 involved:
m a review of POCL ssumptions underlying their banking strategy, and their approach to
benking technology costings and timescale estimates under options 2 and 3. Discussions were
. held with Keith Baines (POCL), Tim O'Leary (French Thornton), and Sarah Mullen
I (Treasury). i es
I W a seview of Honzon architecture and ICL/Pahway proposals for banking, together with issues
aftecting the inclusionIof banking technology. Incidental review of the impact of removing the
henetits payment card. from Horizon (though discussions here were subject to constraints on
consultation with ICL/Pathway).
& astessment, in consulthtion with retail banking experts, of the technical viability of the banking
opuons proposed for POCL and the impact of the associated risks on feasibility, cost and
timescales j i 3
® construction of. cosing Scenanos for aption 3.
1 :
' ‘
1.4.4 - POCL banking income assumptions
We reviewed the basis ofithe banking income assumptions POCL submitted to the Treasury. Ve
Noted areas where there were differences between what POCL submitted, and what appeared in the
Warking Group Report. For the assumptions prepared by POCL we interviewed POCL employees
to understand the basis and degree of confidence in the assumptions. We prepared an assessment
of the assumptions used, and developed revised assumptions. ~ ,
!
POL00028638
POL00028638
< AM Treasury
Benefirs AyencylPost Office Counters Limited automanon project
Worksag Draft - 16 October 1998
Feasibility of fall-back options
Viability of early shift to benefit payment by ACT
i
1
Banks’ views on the move wAcT
to universal payment of benefits by ACT. However, findings of some initial interviews undertaken
ease patria” Bv23
t *
Furtner research would ‘ needed to obtain a cumprehensive view from banks regarding the move
on behalf of the Benefits Agency suggest that, overall, universal payment of benefits by ACT is
ngt 4a area of concem for banks:
& Migranon to ACT: ‘Banks would expect that there would be formal discussions between
BaJ/DSS and the indystry, but no prablems ure anticipated for a migration period of two years
or three years. Notice}would only be required if this necessitated changes to products, payment —
infrastructures or ATMs. This is assumed not to be the case;
u Use of ACT by existing customers: No problems or issues are anticipated here Customers are
free to use ACT methods as part of their esisting account. The basis of their account 15 not
expected 10 change merely us a result of increasing use of ACT methods for benefits payments.
Enisting ATM facilities, including free withdrawals, are expected to remain unchanged.
However, it must be noted that some banks charge customers for withdrawals at any other bank
ATM, a policy which is expected to continue. Some concern was expressed regarding a
putative increase in counter cash withdrawals. {f there was a noticeably large rise-in counter
transactions, banks would be likely to review the need for charges in this area. In this
coanection, it is a common retail bank strategy to migrate customers out of branches to ATMS.
‘This trend is not ben fits related;
® Offering new accounts to the unbanked, Some banks believe that they have products well
suited to all sectors of the market, and are keen to extend these to those on benefits without
bank accounts and fhe unbanked generally. Under money laundering rules the main
requirement is an address, or a letter from a hostel director. There.are no exclusions for social
group reasons. However, in line with wider strategic considerations, a significant increase in
counter transactions may cause a review of charges;
1
= Use of POCL for cash withdrawals. In a situation where post offices would offer an ATM or
ATM-type facility, thé banks questioned.see no issue or problem with customers of the PO
continuing to use post offices for cash withdrawals, and indeed were in favour of customers
continuing to use the if that was their choice. Financial issues relating to the provision of
such services would ba dependent on a wide variety of issues, and would need to be determined
ona Proposition-by-proposition basis through discussion between individual banks and POCL,
8 Increasing volume of payments ia BACS: Banks do not have concems ‘regarding increases in
BACS volumes; :
@ Crstomers of another bunk switching their existing account to the bank: No problems or issues
are anticipated here. Some banks are positively interested in urtracting new customers, and
petieve they have arirgetive products to offer. Such products typically offer un on-line debit
card (no cheque Book, ATM withdrawal facilities at no cost (though some charge for using
other bank ATMs),
churge for counter wt via such an account). Some products offer modest credit
fac.tinies to allow customers to round up ATM withdrawals to the nearest £10,
H
WSO facilities, and counter facilities (though some. banks currently
POL00028638
POL00028638
. 16-10-98 18:20 FROM : 1-460 P.08/13 Fe723
“7 Gea AS Treasury
Benefits AgeatytPost Opice Counters Lemued auromation project
Working Draft + 16 October 1998
2.1.2 Benefits Agency 1
The DSS’s DeparimemaI Review Of Benefit Payment Options ~ Proposal for increased use of ACT
concludes that the following levels of ACT migration are feasible:
. 60% by the end of Mpreh 2002;
mu 99% by the end of 2004,
The paper, which is underpinned by a top-level project plan, identifies the following key
dependencies: ! .
a resolution of banking Issues + see 2.1.1;
m agreement on policy issues relating to benefits which cannot currently be paid by ACT and to
the unbankable; se °
a dlignment of penodicify;
™ introduction of secondary legislation to make ACT the normal method of payment (and
passibly also to alter periodicity) by March 2001; i" s
~ w implementation of IT changes including those required to CAPS,
> The tT changes ere scheHuled to complete by che end of June 2001. The changes may well be ;
significant - over one year’s development time has been allowed in the plan but we are not aware
of there being any detailed plans for the changes at this early stage. The plan shows the cost and
timescale impact of the IE changes being assessed during the first half of 1999,
Given the complexity of GAPS, this dependency Tepresents a risk. The risk is aggravated by need,
we understand, for CAPSichanges to be implemented over Bank Holiday weekends, increasing the
impact of development slippages.
2.2 Banking viability: cojnmon assumptions and issues
1
2.2.1 Scope of POCL banking fervice :
Development of a banking strafegy for POCL is at an early stage, and is unlikely to be complete
for x further 3-6 months. However, the agreed objective is for POCL to provide a delivery channel
for retail banking. POCL wall not be aiming to support all aspects of retail banking, nor will it
become a bank in its owniright. ‘The provision of retail banking services via POCL is intended to
Serve it dual purpose: H
if
m Jo drovide a continued benefits encashment service for benefits customers on migration to
ACT, and hence a means of retaining their custom;
t
@ Jo offer a commercial jervice to generate sufficient profitable business to replace lost Benefits
Agency income and the longer-term future of the post office counter network.
The proposed POCL bankjng offering has been envisaged as the equivalent of a “manned ATM”.
For the purposes of the vibility assessment, the following assumptions have been made:
WM POCL is aiming to offér services to the bulk of the personal banking marker. Agreement with
8-9 significant players is believed to be necessary 10 achieve this;
m POCL will offera standard service to the banks with whom they establish agreements,
+ 16-10-98
w
rs)
ie)
2.2.2.1
“Banking viability I
POL00028638
POL00028638
18:21 ~ FRO . - T-460 P.08/13 F+723
a] q
kei St Bentfits Agency/Post Office Counters Lunited automation proyect
. Working Draft =16 October 1993
& customers of banks which have entered into agreement with POCL would have access 'to the
core “manned ATM” gervices at any post office
The following are tegarded as the core services which must be provided as a basic banking
service: i .
™ cash withdrawal by Swueh/Delta card or by cheque. Depending on the policy regarding the
implementation of bepefirs ACT for the “unbankable”, cash withdrawal by Solo/Electron may
a130 need to be supported;
t
account balance enquiries;
mcer-account transfer
bill payments;
cheque deposits;
cash deposits;
1
!
1
!
i
I
mini-statements; i
W stationery ordering. {
From a technical perspective, support for PiNpads is likely to be required, and, longer term, the
banking infrastructure should be capable of supporting smartcards, given their likely importance in
future retail and government services.
A number of further possible banking services for POCL have been discussed, but have not been
reflected in this viability assessment. These are listed in appendix B. :
H ;
1 F
Assurtptions relating to the feasibility of banking which affect both options 2 and 3 are reviewed
below. ‘
i
!
Business case for the “mapmed ATM’ i:
Options 2 and 3 (other than the simple debit terminal variant of option 3) all involve the provision .
of services equivalent to a “manned ATM”. The business case for this will need to be
substsntiated. 1 :
Key information required includes further details of the geographic coverage and approach
envisaged by POCL to déliver 2 “manned ATM” zcross all outlets. Decisions on the number of
outlets and the form of ATM to be provided will have a significant impact on costs and timescales.
Factors to be considered include: :
™ in locations well servé by the banks and supermarkets, will “manned ATMs” within post
offices offer consumers benefits over existing alternative distribution channels?
@ in tural areas, will POdL be able to offer a profitable service which is more cost-effective for
the banks than the provision of an ATM?
W fo what extent do consumers in different types of area prefer a face-to-face service to an
automated service? .
an
HM Treasury.
2.2.2.2
2.2.2.3
* 18-10-98
POL00028638
POL00028638
1e:22 FRO ~ a Taso PLO/I3 FH728
Khu
'
I : : HM Treasury
i Beacfits Agenty/Poyt Office Counters Luntied automation project
t Working Deaft » 16 October 1998
' :
i
1
i
Banking software
! :
Op the basis of the banking service outlined-in section 2.2.1 above, the core software required is
essentially a teller appli¢anon, which would emulate an ATM to interface with banks’ systems.
Teller software would ot necessurily be available “off the shelf’, but standard development tools
are generally used in thys market, to develop it set of teller functions adapted for a customer's
particular business envithnment. Software-development of this nature should nor present major
technical problems, thohigh a number of issues would need to be explored thoroughly, in
partivular: i
m functional and security differences between “manned ATMs” and “real ATMs” would require
investigation. For example, the first use of a card generally results in the user being prompted
to change his or her ‘PIN. It is not clear how this would be handled by the POCL counter
system; ! ° . .
™ the possible need for! support of the Euro:should be considered, particularly with respect to
Tansition issues. I ;
1
Adopting a “manned ATM" approach has the advantage of theoretical simplicity based ona single -
process for POCL staff, although depending how it was implemented it might restrict furure
growth of services to Telail financial services customers. For example, some services currently
excluded from the network banking definition - eg caprure of address changes - would not be
supported by the ATM. Prptocols.
Tn addition to a front eng wller application POCL may even under this relatively simple option,
and depending on their retail banking goals, need to-consider implementing systems to handle
customer information, settlement, funds transfer. accounting end management information. POCL
will need to outline all ofthese requirements in more derail to enable credible estimates of costs or
timescales be provided. This will also drive out potential issues in other areas such as ownership of
customer data, data protection and the extent to which banks might feel a competitive threat from
PQCI. which would make them unwilling to co-cperate, .
1
Bank interfaces I
The development of the ploposed capability for FOCL is dependent on establishing interfaces with
retail banks There are three main possible approaches to this:
= POCL interfaces with banks individually: POCL would enter into reciprocal arrangements with
a number-of banks The banks would then allow 2 POCL “front end “ teller application to
emulate an ATM which interfaced directly with a bank’s core banking systems, thereby ucting
as an additional delivery channel for that bank. The practical issues around banks granting
aceess directly to a core. systems and the controls and safeguards needed to preserve
integrity are likely to result in a lengthy discussion period. Our view is that it is unlikely that
backs will accept the “black box” stance on this which Pontis advocates for POCL: as parties to
an 48reed interface, banks will expect evidence of the appropriate security and controls;
= POCL interfaces with hanks via the LINK network: POCL have indicated that they may wish to
gonsider using an established intermediary such as LINK Interchange Network Ltd (LINK),
wh:ch provides a switching service to members network membér ATMs, as a suitable vehicle
for their pseudo ATM:zpplication, However, to the best of our knowledge POCL have not
approached LINK officially to discuss this idea Based on our knowledge of LINK and
exploratory discussiong with them, they would perhaps aot readily accept a new member whose
“system” was not an ATM network irrespective of how it was disguised. In addition tt is likely
POL00028638
POL00028638
I * Tee0-98 18:22 I FROME
Akt Treasury
Benefits agencyltost Office Counters Limited automation prayect
Working Drafs» 16 October 1998
keel
' !
that LINK would have reservations about taking an organisation which had 19,000 new “ATM”
ourlets into a proven and reliable ATM network of approximately 15,000 LINK ATMs.
Comments on the vidblity and likely timescale of establishing a network via LINK are not
possible withour invdlving them in detailed discussions. LINK will require clear and detailed
definitions from POEL about how it proposes to meet LINK standards and preserve the
wntegrity of the LEINKInenwork, and strong ousiness case for why tts pseudo ATM or “manned
A&TM" should be accepted into the LINK ATM network. It 1s likely that LINK would not
avcept POCL without due consideration of its. application and consultation with the member
bunks. It is envisagedithat this process may-take at least 6 - 12 months,
um LOCL interfaces with one bank, This bank would in tum provide the interface with other
banks for POCL. POCL may feel that they could approach Alliance and Leicester on this, as a
development of their ¢xisting banking relationship with them. .
Option 2: commercipl and technical feusibility
Jt
Option 2 commercial vidbiliy
Drivers of the commerci4I viability of option 2 include:
& vaability of POCL netivork banking;
® remuneration of ICL. ;
The viability of network banking in general 1s discussed in section 2.2.3.
ICL has provided an indi¢ation of the margin which it will forego should option 2 be implemented.
Our enalysis indicates that this peaks at around £60m. The HM Treasury Working Group report
suggests that ICL may sepk compensation for this lost revenue In considering any potential claim
by ICL, consideration should be given to the fact that they have already requested an increase in
charges in order to avoid Josses. :
The situation may be somewhat ameliorated by the early introduction of banking, with associated
transection volumes. From this point of view, a two-phase approach to the implementation of
banking, with the early introduction of EFTPOS and “cashback”, may be desirable. This option is
discussed further below Gee section 2.4 1).
I
Option 2 technical viabillyy
1
Remoyal of the benefits payment card
Commercial sensitivities have meant that this issue has not been discussed directly berween POCL
and fathway. Nor haveidiscussions been “held more broadly about the provision of banking
technology by Horizon. Our own discussions with ICL have been restricted 10 more general
coverage of the Horizon technical architecture and their proposals for banking. Working
assumprions regarding af viability of removing the BPC from Horizon can be made from these
findings, but the technical implications would nesd to be reviewed more fully with Pathway.
Our high-level understanding of the architecture supports the POCL view that-Horizon has #
madular structure and that removal of the BPC is feasible. (NB. It is in Pathway’s interest to
ensure that the Benefit Entashment System (BES) module - which has the rules and options for the
T46o PIS FeT23
2.3.2.3
"e-t0-98 10:23 ROE : ; - To452 PLAR/I3_ FPR
bie: . : . Héd Treasury
Benefits dgeacy/Past Office Counters Limuted automation project
Working Draft - 16 October 1998
counter benefit card element - can be removed, as they hope to develop Horizon for foreign
marsets, where it would not be required].
The ampacts of removing the BPC are likely to .nclude.
® possible need for changes to the EPOSS counter application;
m possible need for changes to the TPS host application (reconciliation);
& revisiting of I capacityimodels and infrastructure deployment plans (number of communications
devices etc); I
&@ retesting of the modifjed system;
™ modifications to teainjng material.
It should be noted that the bulk of the benefit payment software has now been developed so no
major development cost pavings for {CL would result. From our understanding of Horizon, we do
nor expect the changes qutlined above to be.onerous, although the need.for.retesting is likely to
give cise to a slip in the — date for New Release 2.
Suitability of the Horizon architecture for development of a bankmg capabiliy
ICL Pathway appears confident of the feasibility of providing banking services over Horizon. This
view is supported by the!Pontis report. However, this assessment of feasibility is based on a high-
level technical design which, in tum, is based on @ number of assumptions about the business
Tequizement and the busjness, technical and procedural aspects of the putative interface berween
Horizon and the banks. I
1
Further definition of the business and technical requirements, and their agreement by POCL, ICL
Pathway and the relevant banks, would be required before the impact on the Horizon architecrure
conld be confirmed. We lave commented on the issues involved in section 2.2.
‘The.other area which wil] require more thorough technical review is the impact of on-line banking
aufhorisation requirements on the current Horizon architecture. One option for the transmission of
the messages constituting the authorisation dislogue is the Riposte infrastructure. Riposte is
essentially a batch system, albeit with frequent batch transmissions (at 15 minute intervals). The
ICL cesign relies on the ‘use of “priority messzges" which will cause a batch transmission to be
“forced”. ICL is confides that Riposte is sufficiently robust to be used in this. way. However, this
is an wrea of potential technical risk since we understand there to be no Riposte implementations of
this size in existence, ler ‘lone any with significant on-line elements. Pontis has highlighted this
asarisk area, i
; 1
i .
Pojeutial for early introdycuon of EFLPOS and cashback
There is already an intenhion, although not a commitment, to introduce ESTPOS capability into
Horizon. Given the risks associated with the establishment of # network banking interface, it may
be appropriate to consider introducing EFTPOS into Horizon in advance of the move to full
network banking. This would prevent the Joss of customers moving to ACT should the full
network banking implementation be delayed. : ‘
!
Option 2 timescales i
Hl
Ivis necessary to considerIboth-
“
POL00028638
POL00028638
*Ye10-08 18:24 FROME : Teas P.1a/I9. 723
POL00028638
POL00028638
. UM Treasucy
Hencfits Agency{Post Office Counters Limued wutomation project
Workwy Draft~ 16 October 1998
% the timescale for the inational rollour of a revised version of New Release 2. This is needed to
obtain benefits from fhe order book control system (electronic stop notices) and the automated
payment and point ofjsale systems. Optionally, EFTPOS capability could also be included,
the timescale for the felease of full network banking.
Without discussion witht ICL, discussion with the banks, derailed analysis of existing plans and .
detailed planning of the new activities, the estimation of timescales for option 2 is somewhat
probtematic, However, tb facilitate the comparison of options, we have established the indicarive
Tange of timescales depjeted by the tigure below based on our high-level understanding of the
project plans and methods, and the system architecture. 2 :
1
1999 2000 2001 2002 2003. 2004
Roll out revised NR2
Agree Banking Reqt
Development
Testing & Live Trial
Pilot (if required)
Rollout
Note: Light grey burs denote potential slippage either within the task or zs 4 result ‘of slippage of predecessors
i :
The individual tasks are described below.
H
Rall cut revised NR 2 The release content of NR2 must be revised to exclude the benefit
payment card. {n order to ensure that rudimentary banking capability is
frallable m time for ACT, EFTPOS may be added to the NR 2
efinition.
i
t
ye removal of the benefit payment card is likely to necessitate some
evelopment and retesting, although the impact of this may. be
Initigated by the fact thar there will be less user functionality to test.
Given that there may be a decision to include EFTPOS in the release,
we have estimated the net slippage as 3-9 months.
e achievement of a timescale in this range with EFTPOS would be
jependent on a rapid agreement of the detailed requirement.
T
Agrec banking 4 definition of the requirement for bunking agreed by POCL, ICL
requirzment Rathway and the relevant banks is an essential prerequisite for the start
df development.
We view this as a risk area. On the basis that work on the banking
strategy is already in progress, we have allowed around one year from
njow for the definition of the requirement. However, discussions with
2.4
10-88 18:24 FROM . = T4688 PLOI/I2 Fe72d
HAI Treasury
Beasfis Agenty/Post Office Counters Lumlted automanon peayect
Working Draft - 16 Ociwwber 1995
iba
POCL suggest that the actual timescale might be up to 9 months later,
for example if protracted negotiations with the banks cause delay.
Deve opment . We have estimated 6 to 9 months for the. development of the banking
° ‘unctionality. This may include derailed technical work.with the banks.
Ne timescale also assumes that there will be a single external
Interface to network baaking, eg to LINK. :
Testing and Live Trial Wwe understand that testing and live tril of current Horizon releases
fakes around 9 months The reduction in functional complexity arising
rom the replacement cf benefit payment with banking may reduce the
Timescale. However, stringent acceptance approaches on the part of the
banks may increase it, We have therefore estimated 6-12 months.
Pitot (optional) there 18 a possibility that the banks/LINK would expect to pilot the
pperation of network banking before full rollout began. Such a pilot
Would involve a larger number: of-outlers:than-the live mal. We have
Howed 6 months for this. Because this is an “optional” activity, we
ave shown the “best case” bar in the plan running concurrently with I -
the rollout. The possibility of the pilot being required is reflected in
potential slippage (the grey bar).
Rollout Ne have estimated a “rollout” rate of 300 per week. Rollout tmescale
pill be dominated by-
time to install PfNpads if these are not installed as part of the
rollout of NR 2. We would expect that 1CL Pathway would seek to
: distribute the upgraded software from the centre by electronic
means;
‘Mm time fo train users in network banking;
im tate at which LINK/a bank would accept additions to its network of
connécted systems.
!
t
Given the plan outlined ;above, it 1s hkely that POCL and ICL would wish to consider merging
New Release 2+ with thk banking release. Components of that release not directly dependent on
the banking requirement - for example the key management service --could be developed while
definition of the bankingirequirement was in progress.
Qption 3: commercial and technical feasibility
‘Two main scenarios have been considered here.
a simple debit terminalé;
m-a full Horizon replacement, including all the non-banking functionality which Horizon is
ictended to support (other than the benefit payment card and - given the likely timescale for
Procurement - the Ofder book control.sysiem), as well as network banking (the “manned
ATM”). t
‘The scenario favoured py Treasury is an innial implementation of simple debit terminals, to
provide a basic cash withdrawal service quickly, followed by the full Horizon replacement.
10
POL00028638
POL00028638
"further here.
2.4.1.1
2.4.12
POL00028638
POL00028638
18:25 FROM _ . “T-480 Poa/ig) FAT@4
sit 3 id HM Treasury
KPIIG Benefis AgeacylPost (fice Counters Limited automaton project
Worksng Droft- 16 October 1998
We have also considered 4 further option which involves the provision of network banking without
any other services. This ‘gption would delay the provision of cash withdrawal as compured with
debit cerminals and would also delay the provision of any richer infrastructure (eg support for
elegtronic access to government services) suitabl> for POCL’s longer term strategy. Consequently,
this option - which was pot carried forward into the Treasury figures - has nar been analysed
Simpte debit terminals
Simpiz debit terminals: commercial viability
i
‘The simplest “non-Horizon” banking option for POCL is to install debit terminals and provide a
service allowing customers to purchase existing goods or services from POCL and also to
wiphdraw money from their bank accounts via cashback. To provide this type of service POCL
waul need to enter into agreement with a merchant acquirer, a bank who. would act as the
intennediary berween POLL and the card issuer.
This option has not beep put forward as a long-term commercial prospect, but as an interim
solution which enables POCL to develop a basic banking service as quickly as possible, to prevent
the less of benefit customers on migration to ACT which is envisaged if no post office banking
facililies are in place in ume. Iris not, however, taking POCL into retail banking. a
‘The viability of this thergfore rests om the feasibility of system delivery within the ACT migration
_ timescales, together with the impact which it might have on the implementation of a fuller banking
service. ! .
: i
Simpie debit rermmals: technical viability
1
‘Based on our current a fronly bet this option appears relatively straightforward from a systems
perspective, with POCL simply behaving as any other retailer who takes debit cards and passes the
transactions on to a merchant acquirer.
Under a typical agreemept with a merchant acquirer the necessary equipment will be provided by
~ the acquirer on a rental hasis with a percentage charge per transaction value levied for the service
2.4.13
provided by the merchant acquirer, as intermediary. These intermediary services include the
transaction handling, authorisation and settlement. :
Simple debit terminals: thmescales
degree of certainty. He
following timescales, primarily from POCL estimates with which we have no reason to disagree.
Detasled ee would be required in order to establish timescales with any
4. I 1999 2000 2001 2002 2003 2004
Procurement Fe]
Opetational Trial : a I
Rollout ! I ee
Nore: Light grey bars denote gorential shppage cither within the task or as 4 result of shippage of predecessors,
1
wever, for planning and comparison purposes, we have derived the
i
* 18-10-98
2.4.2
POL00028638
POL00028638
18:26 © FROM T4680 -P.OS/I2 F724
ene I ; ; HM Treasury
- : Benefius Agency/Post Office Counters Limited automanon project
t Working Draft - 16 Octuber 1993
i
;
The specific tasks are described below
Procurement We have based our estimate of 12-15 months from the start of 1999 on
jhe POCL estimates provided in the paper Option 3 - -Alternanve
Technology for POCL, taking account of the fact that work on the
banking strategy has already begun.
Operational Trial We have taken our estimate of 3 months from POCL’s estimates. -
Rolleut We have taken our estumate of 12-18 months from POCL’s estimates.
The driver for the rollout timescale will be the logistics of installation
ind training.
Nore thar our costing moifels assume the earlier dates for these activities.
1
Nou-Horizon “full functionality” option
~ Under this option, POCH! provides network banking services, as described in section 2.2.1 above,
242.1
together with a full functjonal replacement for Horizon.
With respect to the banking component, the general commercial and technical issues raised in
section 2.2.2 again apply,
“Bull functionality” in 4 non-Horizon context has not been explicitly defined to date.. For the
purpuses of this pian we have assumed, in consultation with POCL, that the requirement
would be for the existing committed Horizon functionality with the benefit payment card and
order book control servides removed, und banking added. Further details of these assumptions are
provided in appendix C. §
i
Non-Horizon “full. functipnality”. techcal viadility
Whilst suppliers may prppose various technical options for this, some of the issues applicable to
opticn 2 will be relevantjhere too, m particular:
1
w the need to reconcile meeting standards for banking services (eg. the ATM interface) with
FOCL's potential need for commercial diversification;
wm the need to ensure a software basis which enables new applications to be added readily. This
suggests a PC/Windqws front end based on generic software, rather than a system based on #
more specialist banking package.
tt should also be observed thar a full-function Horizon replacement is unlikely to be significantly
simpler - or necessarily gheaper ~ than its predecessor. In particular:
w te connection of the Horizon replacemenc to banks’ systems. or LINK, and the associated
approval processes i$ analogous to the connection of ‘Horizon to BA systems, and will have
similar programme jnanagement requirements for the cureful planning and execution of
acceptance tests; ! ;
i
™ athough the banking application may be simpler than the benefit encashment application,
fauch of the complex}ty of the overall infrasiructure - eg the systems management architecture -
may remuin
POL00028638
16-10-08 ‘18:26 FRO
kahis:
TGQ P0412 F724
. HM Treasury
Benefits Aguacy/Pust Office Counters Limited automaton pryect
Working Draft - 16 Ocwwder 1998
Non-Horizon “full, I funcnenality + timesvales
* Given that the system bana - in particular the interface with the banks - has not been
defined and that the supplier and solution ace unknown, any planning at this stage must necessarily
be very approximate. Ngnetheless, for planning and comparison purposes, we have derived the
timescules shown below.i
POL00028638
1999) 2000 2001 2002 2003 2004
2005
Establish 2cquirement
Procurement
Development of Pilot
Pilot
Rolluut
peel
pee
“tT ~
Note" Light grey bers denote pptential slippage either within the task or 252 result of slppuge OF predecessors
‘The specific tasks are described below.
: 1
Establish Requirement
A definition of the requirement for banking agreed by POCL, ICL
‘Pathway arid the relevant banks is an essential prerequisite for the start
tof development.
We view this as # risk area. On the basis that work on the banking
\orrategy is already in progress, we have allowed around one year from
tnow for the definition of the requirement. However, discussions with
:POCL suggest that the actual timescale might be up to 9 months later,
:for example if protractsd negotiations with the banks cause delay.
Proc arement
‘Our timescale estimate of 18-24 months is based on our experience of
!procurements. There would be some potential for concurrency between
ithe early stages of this activity and the finalisation of the requirement.
Development of Pilot
We have assumed that it will take around one year to build a pilot for
frollout to around 1% of sites.
Pilot
‘We have assumed that operation of the pilot - including rework - will
ltake around 1 year.
Rollout
7
'We have assumed a raztd rollout at a rate of 300 per week,
Note that we have used the earlier of the timescales shown in our consolidated cost model and that
there is risk amtached to those timescales. That cverall eurlier timescale - 5.25 years - is in line with
POCL's estimates.
POL00028638
POL00028638
ie 88 18:27 FRO ° ae T-460~ PLOB/I2 F724
. HM Teeasun
Benefits AgencwPost Office Counters Limited autumution project
Working Draft ~ 16 October 1998
Discussion of key assumptions’
POCL network modelling
Profit impact and network closures ©
There is a close inter-relafionship between the figures for Ist round impact on POCL profits,
estimated profit from banking and the size of the post office nerwork. Tuble 1 summarises these
inter-relationships. The grecise impact‘on the size of the network will depend on both the effect
on sua-postmasters and whether they continue in business (“unmanaged closure”) and the effects
on POCL and the measures that they take to reduce their fixed cost base ie. close post offices in
the locality (“managed closure”). If POCL are successful in managing closure they succeed in
both reducing the fixed cost base and in maintaining their income flow. Of critical importance to
the etent to which eocr 1 forced To carry out closures, is the extent to which banking income
can replace Benefits Ageticy income and hence maintain the viability of the nerwork,
Key essumptions made by POCL aré:
wm that ma managed clogure scenario they migrate 60% of the. PO business to other offices. This
assumption is based on very little information from a small closure programme in the mid
1980s and from resulfs from office relocaticns. Whilst these are rather unreliable guides, we
have no reason to change this assumption,
a that footfall from the Post Office side of the business to the private side is in the range = 27-
38%, that this percentage of retail customers are PO dependent in footfall terms. This is based
on @ survey which interviewed customers of the retail side of the business as to whether they
had also visited the Post Office. The core assumption is that if customers had not been visiting
the Post Office they would not otherwise visit the retail side of the business, This is an extreme
assumption and is .cgntradicted by other survey evidence trom POCL which indicates that
between 20% and 65% of customers of the retail and post office side would still have visited
the shop that day even if the post office were not there. However, the footfall assumption only
affects the unmanaged closure scenario under Option 3, where as a result the estimates of Post
Office closures are liely to be overestimates.
The differential impact op the number of post offices between Options 1, 2 and 3 is generated by
POCL’s success in retaining BA recipients as benking customers rather than order book/BPC
cystemers. As discussed in section 3.3, outlining assumptions in the denvation of banking profits
is critically determined under Options 2 and 3 by the speed with which POCL can get banking,
technology and contracts with banks in place in order to achieve a seamless transition of customers
to accessing bank accousts through POCL.. : .
Compensation to sub-pdstmasters
POCL have assumed thét they wil} pay an amcunt equivalent to 26 months compensation to sub-
postinasters in respect of all network closures. This level of compensation is the current
arrangement if post offices ure compulsorily closed Ir is arguable that declining sub-postmaster
profits will lead to volintury closures, but we accepr the judgement that there will be a lot of
pressure to compensate} sub-postmasters for closures resufting, indirectly, from a goverament
acticn to move to payment of benefits by ACT. toe
i
4
I
I
dicccmewen
effects on subpmrs
reduced pay
from BA busi
effecis on POCL
reduced pay
4
from ancilliary
PO business
reduced private
business income
reduced sub-office
variable costs
uncontrolled
1 BA change to ACT
reduced income
reduced PO reduced income
ancilliary >! from ancifliary 1
business business
,I reduced pay I I
to subpmrs
; POCL cost reduction:
controlled
«network - subs
enetwork - crowns
*overheads
network POCL Bi
impact P&L impact
T
potential barrier
POL00028638
POL00028638
NB Sequencing
and timing are
I-—-- criticalto
end
result
POCL
income
generation:
new vision
=
3
Hous
_Webed 21/90°d 899-1
POL00028638
POL00028638
124
Testo-g 18:26 FROIE
keinbi
~~ Fease! par *
HMM Treasury
Benefits Agency/Post Office Counters Limited axtumanion project
. Working Draft- 16 October 1998
1
3.2 Cost of banking technology
!
3.2.1 Option 2
Senup costs are assymedIto be £20m in the early years (£10m for software development costs and
£}0m for hardware) We,continue to include these costs.
The main component off this cost is the recogrution that POCL will need to remunerate ICL for
their investment, becaus¢ the loss of revenue for the BPC significantly outweighs any cost savings
they can achieve at this ;tage. Option 2 in the Working Group Report included a cost to POCL of
£75m per annum from 2001/02 to 2006/07 regarding remuneration of ICL, assuming ICL could
reduce costs by about 20h,
We asked ICL if they equld provide us with information on.their reduction in margin if the BPC
was discontinued. We made a slight adjustment to their calculations (as they had assumed
compulsory migration of ACT from 2005/06 so therefore had included an extended. revenue, flow
from OBCS in partial offset of reduced BPC revenue). The result is a much lower cost to POCL
(NPV reduction of £112).
322 Option 3
3.2.2.) Debit terminals
The lener from Sarah Mullen (HM Treasury) to Helen Corlett (KPMG) of 7 October 1998
indicates that the costs Used for debit terminals in the HM Treasury Working Group Report were
derived from POCL estimares of a setup cost of £18M and running costs of £60M per annum. The
derivation of these is presented in the letter from Mena Rego (POCL) to Isabel Andérson{DFI),
13 July 1998. : _™
POCL quote the following setup costs for debit terminal operation:
i
Trem i Quantity Unit Cost = Total Cost £M I
Outlet modificqtions labour 19,000 8 1.43
Outlet modifications materials I 40,000 50 2
Tnstallation and training 19,000 200 3.8
ISDN connection 19,000 300 5.7
Subtotal ! : 12.93
vaT ! : 2.26
Project Management 60 50000 3
Total 18.19
+
1
!
These costs appear reasqnable for working purposes, although the instaltation-related costs may be
underestimated since: +
wwe understand that sbme outlets may need significant modification;
m jan-day rates of £140-200 are assumed for the installation and training - these might be too
low if contract staff ae being employcd.
1
I
H 1S
i
!
1
t
POL00028638
POL00028638
N
_Welo-ee ter28 Freie “Tod PLOG/I2 ” FHT24
HM Treasury
~ Bensfits Agency/Post Office Counters Limued ayronguon project
Working Droft - 16 October 1998
i
i
H
i
1
POCI- quate the following running costs for debit terminals, assuming 500M transuctions per year
tem i Quantity Unit Cost Total Cost £M
Terminal rental! 40,600 468 18.72
ISDN rental I 19,600 400 76
Call charges: 500,000,000 0025. — 12.5
Consumables! 40,600 10 0.4.
Transaction charges 500,000,000 0.04 20
Call centre i for fall-back I 40 50,000 2h.
transactions I
. Turnover training 4000 100 04
Relocation cost} 380 - I 525 0.2
Sub-total; _- 61.82
VAT H 10.82
Fraud risk : 500,000,000 0.025 - I 12.5
Total 85.14
Further work would bé required to confirm the nature of the call centre requirerient and
cartesponding cost. Our {esearch indicates the following running costs for debit terminals:
i :
Trem 7 Unit Cost
ISDN rental pen year 350
Visa debit transection cost 0,08
Switch/Solo trahsaction cast 0.045
‘Debit terminal rental per year _I 180
Sources. BT, Tnaieatfve ‘Costs obtained snformally from & merchant acquirer
. 1 :
Subssining these figures into the table and assuming an even distribution benween transaction
types, we obtain: i
!
Trem : Quantity [Unit Costé —- I Total Cost £M
Terminal rental! 40,000 180 72
ISDN rental} 19,000 350 6.65
Call charges} 500,000,000 0.025 12.5
Consumables } 40,000 10 0.4
Transaction charges $00,000,000 0.0625 31.25
Call centre I for . fall-back I 40 : 50,000 2
transactions!
Turnover training -[ 4000 100 0.4
Relocation costs 380 «I 525 0.2
Sub-total 60.6
vat 10.6
Fraud risk 50€,000,000 0.025 12.5
Total 83.7
i
H
I
POL00028638
POL00028638
Jae tego Frole i re “poate psoaie “Pere
3.2.2.2
AM Trewsury
Bensfits Agency/Post Ojfice Counters Lumied automanion project
7 Working Draft 10 October 1998
These costs are very simjlar to those quoted by POCL and are reasonable for working purposes
However, further work would be required to confirm the call centre costs Noie that these figures
assume a transaction vollime of SOOM. In our consolidated costs for option 3, we have assumed
lower transaction Volume; to reflect the phasing of the introduction of ACT
Full-junction Horizon Repent
The lemer from Mena Rego (POCL) to Isabel Anderson (DTI) of 7 August 1998 indicates that the
original POCL estimate Supplied for input to the HM Treasury Working Paper for a replacement
fujl-tunction Horizon was a one-off cost of £109M with ongoing running costs of £74M.
The lemer also indicated ither subsequent work had led to a revised estimate of a one-off cost of
£30M-60M and running gosts of £60M-100M.
We understand from discussion that at least the initial estimates-were derived from the Pathway
charges for the service which would be being provided (ie banking and no benefit payment card).
It is our view that comparison with Pathway charges may not“be’an appropriate means of
estimating the likely furure costs of a Horizon replacement. This is for the following reasons:
w The number of transgctions passing through the replacement will be lower than would have
péssed through Horizon since the number of banking transactions will be Jess than the number
ot benefit transactions. All other things being equal, this would cause the supplier 10 raise
transaction charges correspondingly.
emphasis on banking and electronic government may encourage suppliers to view the
w The nature of the gee will be different in marker terms. For example, an increased
. Opportunity as worthy of strategic investment - this would tend to lower charges. Alternatively,
* suppliers may perceive the opportunity as being high risk because of the fate of Horizon - this -
would tend to raise prices. Furthermore, the revised mix of services required may give rise to a
competition very different in character from that of the original procurement.
m Tae winning supplief might take: different approach from that adopted by Pathway, for
example making mary use of existing infrastructure, thereby allowing charges to be reduced.
mw Tne detail of the requirement is nor yet known und may differ significantly in complexity from
thatof Horizon. .
In order bemer to assess fhe potential range of costs for a Horizon replacement, we have modelled
the possible costs of a Horton replacement and: then: determined the revenue which a supplier
would require for the prdject to have an Internal Rate of Return of 15% - a (conservative) industry-
standard figure for bidiproject. evaluation.
Ty order to reflect the s4nge of influences on price, we have modelled @ number of scenarios of
which three are shown here:
m Scenario A - integrated solution ‘largely constructed from scratch for the project; high
implementation capex; significant proportions of support infrastructure dedicated to project;
rollout at 200 officesiper week; high proportion of contract staff.
uw Scenario B - significant element of reuse in development and support, implementation capex
ead ongoing operational costs reflect significant economies of scale and very strong purchasing
power; rollout at 300joffices per week; no contract staff.
'
7
POL00028638
POL00028638
I Werto-gs 1ess0 ROR” 0S qed "BAO FTE
HM Teewsury
Benefits Ageavy/Post Office Counters Limited amomation project
: os Working Draft - 16 Oztaber 1998
iene.
a Scenario C - as scenkrio B but with the contract staff proportions and equipment costs of
scenario A.
Scenurio A and scenario B represent extremes - the former would be likely to represent en
uncompetitive solution while the lamer would require an unusually strong combination of human,
knowledge and infrastrubrure assets. Scenario C represents a potentially realistic point berween
thesé extremes and has al per-terminal cost consistent with other managed service projects known
Tous. !
I
The snodels are based on: he following assumptions:
nse essere mie see
© Initial development takes 1 year and is followed by a pilot rollout to 1% of sites. ‘This
assumption is primarily driven by the issues relating to acceptance of the infrastructure by the
benks. In muking, thé timescale assumption, we have assumed that only a single extemal
interface - whether tq an intermediary such as LONK or to a single bank which then provides
the interface to other banks - is required.
= Full rollout begins one year later. This: assumption-derives: both-from consideration of the
+ Horizon experience apd, again, from the need to have acceptance from the banks prior to fill
rollout. :
= Sr is assumed that there are 19,000 post offices. with 40,000 counters. The average number of
counters in crown, urban and rural post offices is assumed to be 5, 3 and I respectively.
w Fall functionality is assumed to comprise current Honzon without the benefit card but with
network banking. Furfher derail is provided et appendix C to section 3.
& Rather than consider possible charging. models in depth, we have modelled a flat revenue
‘Stream starting from the beginning of rollout
@ oth inflation and thg deflationary pressure on technology costs are ignored.
w& 4 10-year contact is pssumed.
uw {cis assumed that the service is being purchased us @ managed service. However, we have not
assumed PE1-type risk transfer either in the costings or in the required IRR.
m Some provision has! been made for ongoing enhancement of the system but there is no
geovision for major ifehnialogy refresh. : :
It should be stressed thus the esnmates of cost bear-a-very high margin of error since no detailed
system requirement yet ¢xists.
Tris aseful to note that te principal cost drivers emerging from the modelling exercise were:
annual costs in excess of £100,000;
@ capital cost of implementation, particularly the cost of counter equipment; S
@ reanpower costs - bd oe of this size frequently involve large numbers of contract staff with
3:
® operational costs, particularly systems management, support (Ist to 4th line) and maintenance. _
Note that the costs presehted below exclude VAT. 7
“ p-t0-a8 4
3.2.3
ScenarioA !
tem 1 ou
Headquarters Functions ! 4
Desiga & Development : 17
Injua: Implementation + 30
Opsmuon & Support I 16
Toral H 67
. !
Agsunted Revenue exel VAT! 0
Cash Flow i 67
IRR I 18%
Scenario B i
Ttem u 1
Headsuarters Functions H i
Desiga & Development H 6
Taial Implementanion ' 29
Operation & Support i 8
Torat : 4
wt
Agsutzed Revenue excl VAT! u
Cush Flow Saal
TRR : 15%
Scenario C i
Aten 1
Hyadquarters Functions I -2
Denga& Development i 9
InwtiaI Implementation I 30
Operstion & Support i Fs
Tots! © i 49
Assumed Revenue excl vati Q
Cash Flow 49
IRR bY
Soe w
28
28
BS
no
25
103
is
4
67
57
awe
135
16
189
9
80
POL00028638
POL00028638
‘Teas PALI/I2 724
As Treasury
Benefits sligetiglPadt Uffice Conaters Lemuted automanon project
ounan
0
56
ouDA
a
19
48
ono
28
67
39
cues
0
56
3
Cos? of banking technolpgy - LINK churges (bath options 2 and 3)
oM a
Working Draft - 16 October 1998
8 9 10
3003 03
$568
o oOo 0
42 3036
si 4545
no Wo HO
$9 6565
8 9 10
»o-ot
2 2 °2
o 0 0
2% 620020
27 23 23
67° «6767
40 # 43
8 9 (0
2.2. 2
3.0 2.2
o o 0
2402100 2k
29° «62S 2K.
D9 79
50 54,5
Connection tothe LINK network is a potentially amtractive option for the implementation of
: nerwork anna: although the agreement of LINK to such # proposition is by no means certain
ip
For the purposes of costs we have assumed such a LINK connection. On the basis of informal
discussions with LINK, te would expect the following charges to be incurred:
19
POL00028638
POL00028638
US10-98 18:32 FROME a ; Te#e0 patesi2-
kkual
FeT24
. * HM Treasury
Benefits Algency(Post Office Counters Lumted axtonution project
‘ Working Draft 16 October 1993
= a joining fee payable bn entry to the nerwork - £250,000 approx - amount is at the discretion of
the LINK board; i :
wu a LINK switch conneftion fee is payable fer new members connecting directly to the LINK
central switch ~ £80,000 approx depending on the complexity of the technology used by the
connecting financial institution,
discretion of the LINK board - £1,000 - £2,500 depending on the size of the members ATM
® a fixed monthly mae fee, varies according to size of the instirution and is set at the
network (more expensive the fewer ATMs the member has) - assume £1,000 for POCL;
™ aimonthly processor fee 1s payable by those members whose system is connected directly to the
LINK switch - £1,000 flat fee for connection, £750 comms charge and £100 per ATM to a
maximum of £8,000; + :
™ a switch fee is paid % LONK by the card issuer for every transaction that-crosses the switch,
with significant discounts for volume - 7.18 pence to 0.49 pence sliding. scale banded rates
based on number of transactions processed. These terms will not-necessarily:affect POCL since
PUCL will not be a chrd issuer. However, this charging regime underlines the fact that POCL
would be-chartng new territory in joining LINK :
3.2.4. Consolidated costs for option 3
This section consolidatesithe costs for the implementation of simple debit terminals followed by a
fal-function Horizon repjacement. It assumes the following:
"Scenario C for cost oflHorizon replacement
w Debit terminals rolledjout over the course of year 1999/2000 to 2000/1.
m Rollout of Horizon hplacement - and hence payment of flat rate - starts in January 2003. In .
2U02/3, therefore, 25% of flat charge is incurred and 10% of banking transactions go through
Horizon replacement . :
xm Rollout completes April 2004. In 2003/4, 60% of banking transactions go through Horizon
replacement. I
mw POCL costs of procurément and management of. Horizon replacement not included,
® - Debit terminal transaction costs take account of phasing of ‘migration to ACT,
all costs in £44 7
Trem O9ruvd 2000/01 2002 2002/3 2003/4 2009/5 2005/6 2006/7 2007/8 2008/9 2009/10
Debit Terminal !
Setup of Debi Teminal incl VAT 9 9
Fixed Running Costs incl VAT 5 is 20 18 8 0 0 0 oO 0 )
Transuction Cyst incl. VAT 2916 5832 5,346 Q 0 ) 0 0 0
Horizon Replqcenent :
Honzon Replycement Cost ex VAT 1975, xy 9 79 0 7” m ° 79 9
VAT on Honzon 2cplacement “345625 13825. 13.825 13.825 13.825 13.825. 13,825 13,825 13.825
LINK Connection Fees incl VAT 0.38775 7 rs
LINK Running Costs inet VAT 0 0.152 0152 UIS2 «0.152152 -O1S2 0.152 -0.152
i
1
i
i
I
I
i
Total incl VAT \* - 24 46.51 116.809 106.323 92.977 92.977 92.977 92.977 92,977 92.977
i
!
I
!
i 20
i
!
1
“16-10-98 18:32 FROM
3.3
3.3.1
3.4
3.4.1
T4860 PLOI/IT. © F725
Piast : : . . HM Treasuey
i Benefits Ageney/Post (pice Counters Lunited auramation prayect
: Working Draft ~ 16 October 1993
i
t
Banking profits
POC: have estimated future income from banking with reference to consultancy provided by
McKinsey and to sore off their own, limited, research. The drivers of total income are the amounts
banks sre willing to pay, and the number of customers and transactions POCL are able to attract.
1 :
Payment from banks
PQCL based their calculations on an income per account per annum of £25 (which Treasury
revised downward to £20 in the Working Group Report). However this 1s value is based on
discyssions which McKinsey held with banks regarding how much they would be prepared to pay.
for thei customers 10 bg able to use POCL for banking transactions should the bank wish. to
reduce the size of its network to achieve cost savings. 7
Our discussions with a: indicated that payment on a ‘per transaction’ basjs would be more
likely and around 50p - 60p 1s a frequently quoted range. We asked POCL to assume SOp per
Transuction in their revised modelling. .
Key DSS assumptions
ts
Costs of ACT : ;
DSS administrative savings estimates include the cost of transferring cash by ACT into benefit
recipients’ bank accounts.! However, in addition to this cost there will be a cost to the banks of -
custorrers withdrawing cash from their accounts which is estimated to lie in the range of 50p-90p
(dependent on whether thé withdrawal is through an ATM or over the counter and the extent to
which fixed casts of the branch network are included in the calculation). In the case of the
“onbeaked” a provision has been made in the estimates of £32 million per annum payment by DSS
To the banks for them to take on this business without charging their customers for cash :
withdrawals. !
Discussions with the banks held by both DSS and KPMG have not indicated that banks would seek
to chazge for such services und indeed there is considerable interest in uttracting new customers
even taough they are benefit recipients There is some concern about large additional numbers of
customers using their braych network at a time when bunks are.seeking to transfer cash
withdrawals to ATMs. However, this illustrates the potential for the Post Office to develop its
bank1g business alongside the move to.ACT.
In the case of the “banked?, DSS have assumed that banks would not seek to recover uny
adqitivnal costs, There is p number of possible justifications for this.
i packs would obtain adgitional money flowing into bank accounts, the interest on which would
heip offset any udditional costs, e
™ individuals would manage their accounts and their pattern of cash withdrawals in such a way
the they do not incur afditional charges.
t
Whilst this might appear 16 be a somewhat extreme assumption, in the course of aur (limited)
interviews with the banks jhis hus not been seen as an issue. Our judgement, therefore, is that in
view cf the existing provision for possible costs of the unbanked which appears ta bé quite
prudev.t, there is not the need to make any additiozal provision for this sum. Indeed, the
POL00028638
POL00028638
POL00028638
POL00028638
7-469 P0271" F725
I "16-10-08 “18:93. ~~ FROM -
HM Treasury . I
Benelits agencylPost fice Counters Limited automanon project
Working Droft- 16 October 1993
KPMG-
prospective move to ACT for benefit recipients may be a useful trigger for the banks to consider
seriously the banking proposition being made by POCL
1
POL00028638
POL00028638
_ieiee “te: 34, FROM : TAs ‘p.aa/tt Peres
. HM Treasury
Beats AgenzylPost Office Counters Limited automation project
Working Draft 16 October 1998
Summary of financial results
Defisition of options’
: I
Optioa 1: Continue with project
Optica 1 involves continuing the project including the benefits payment card but with amendments
to the contract, in particular extension of its duration
This ‘option has been subject to refinement as a result of the work being undertaken by the
Independent Advisor, who is facilitating discussion between the parties to see Whether:
sm satisfactory commercial terms can be agreed Yor continuing the project;
i
* m outstanding differences on the timetable can be resolved;
mw acredible programme 1 Hor full implementation can be agreed.
For tie purpose of establishing a fixed position m the time available to complete our consultancy,
we will define option 1 according to the assumptions in the core case (30 Seprember) and in the
Independent Adviser's rpposal (11 October). i
Yeur snded Mar (99 00 01 ‘02 ‘03 ‘04 ‘05 -'06 ‘U7 ‘08 ‘09 ‘10
Roll out of system: I
= comaleted
PQCL introduces
banking services:
+ limited facilities
~ full Zaclities
BA moves to ACT
BA fisor payment to
POC: ends
TCL contact ends with:
-BA
=POCL__. i v,
Additional details of optyon 1 are as follows
m= ACT migration mite of '20:40.40 over 2006-2008
m® POCL to accept refresh costs
m Cancel 3% pa price réduction from 2006 onwards
i FOCL to agree to banking business guarantees at a level consistent with reasonably achievable
targels
POL00028638
POL00028638
q-460° p.oa/ii ” F725" 7
. HM Treasury
Benefits Agency/Post Office Counters Limited automanon project
Working Deaft- 16 October 1998
mw Some increase in volurpe guarantees from BA to POCL
I 4.1.2 Option 2; Continuation of Horizon without the senefits payment card
Option 2 involves restructuring of the project without the benefit payments card (BPC) to allow an
earlier move to ACT and parallel, provision of banking services in post offices. BA withdraws
from all contracts with IGL for the development of the BPC on the grounds of failure to perform,
and POCL works with IcL to implement Horizon technology
i)
Year ended Mar tp oo ‘Oe w2 U3 ‘04 "05 "06 07 "0B 09 “10
Ro}! cat of system
«stare v
- completed
PQCL mrraduces
banking services:
+ limited facilines
~ full tacilities
BamovestoACT I
BA floor payment (0
-I POCL eads
ICL contract ends with:
-BA
+ROCL
4.13 Option 3: Horizon cancel fled
Option 3 involves cancelling the. whole project (on the basis of ICL’s breach of contract for failure
To meet agreed timetable) and POCL commissions an integrated technology platform to enable it
to ofier banking servicesias soon as possible.
i
Year ended Mar [99 00 ‘Or “02 03 ‘OF "05:06 "07 "08 “09 “10
POCL introduces
banking services:
- basic technology
~ full banking services
BA moves to ACT
BA floor payment to .
POCL ends
v
ICL contract ends
i
i
i
i
1
T
I
1
I 24
i
POL00028638
Sn me: POL00028638_—
M6100 fests FAO ne Ro ee
ieee ' Benefits Agency/Post Oljice Counters Linmted cuowrsuon reaper
i Forking Draft 16,October 1998
4.2 Financial results
i
!
42.1 Finaacial results: floor Payment from BA 10 PUCL ends when ACT migration complete
NPV of the options discounted at 6% to 2009/10
Note NPV figures for oprfons 2. and 3 assume no compensation 1s paid either to or by ICL
1
£billion I : Option } Option 2 Option 3 Option 3V.
DSS administrative savings 0.4 : 1.0 11 10
DSS programme savin; ! 0.9 I. . 08 0.8 i 08
Net impact on DSS 13 18 19 18
Net impact on POCL 0.2 -0.4° -0.9 -0.8
Oyerill NPV savings ! 1.5 14 1.0 1.0
I ;
4.2.2. Financial results: floor. payment, from BA to POCL ends when ACT migration begins
1
NPV of the options discounted at 6% to 2009/10
Note NPV figures for o phons 2 and 3 assume nc compensation 1s paid either to or by ICL
I
£billion E Option 1 Option 2 Option 3 Option 3V
DSS administrative savitigs 13 13 13.
DsS programme savings} 0.8 0.8 - 08
‘Net :mpact on DSS : 21 > 21 z 24
NetimpactonPOCL } -0.6 -0.8 -08
Overall NPV savings we I" 1.5 13 : 13
1
I
t
wv
on
POL00028638
— POL00028638
kbhb)
HORIZON: SUMMARY OF CENTRAL OF TIONS - DA FLOOR PAYMENT TO POCLENDS WHEN ACT MIGRATION COMPLETE
1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006.07 2007-08 2008-09 » 2009-10 NFV' @6% NPV @6%
(02010. 102005
2 Continuation of Horizon without BPC, move 10 ACT {compulsory from 2001/02) floor paid anti! 2003704
Yo claimants paid by ACT 30% 30% A2%e 53% 1% 88% 100% 100% 100% 100% 100% 100%
% of transactions made by ACT 13% 4% 2% 34% 53% B% 96% 993 100% 100% 100% 100% :
DSS net administrative savings (Lm) 10 15 35 “15 25 Wh) 320 435 435 43s 435° (435 1,306°~ 87
contingency 10 10. -10 -10 -10 10 -10 -10 +10 10 10 -10 BA 56
DSS programme savings (£m) 0 5 0 130 335 35 335 135 135 335 135 135 847 469
Payment to banks‘POCL for “unbanked” 0 8 -20 -30 -33 32 32 +32 “32 +32 32 +32 -205 M5
a a= Total DSS-savings (fm) oo 220 APB nn BB voe a rere Bainee = DUB 1B 213,528 528 528 528 528 1,864 385
first sound impact on POCL. profits Lm ) 5 10 20 30 32 +163 198 183 168 “151 427 58
compensation to subpostmasiers 0 0 0 0 0 0 0 94 94 94 94 394 +263 0
cost of banking technology for POCL QO. 3 33 n “51 31 St 331 2 ay Jd 2 +228 . 195
profits for POCL from banking £m 0 0 0 Wo, «(38 2B 104 116 nN? nod 120 120 493 161
Total impact on POCL, 0 2 + +23 40 7 67 2. 3372 196 +159 143 126 A216 3B
Impact on PO nefwork ural 0 a 0 0 0 0 i) -200 400° -600 -800 = -1,000
urban 0 0 . 0 0 0 0 0 -1,200 «©—--2,400 = --3,600, --4,800 —-6,000
total 0 0 “ec 70. e Q @ -1An0 «2,800. -4,200 = -5,600 ~—--7,000
NETIMPACT ON PUBLIC SECTOR ' 138 410
3 Horlzon cancelled: ACT compulsory for all claimants from 2001/02, floor ends 2003/04
%of claimants paid by ACT 30% 30% A2% 53% 70% 88% 100% 100% 100% 100% 100% 100%
of transactions made by ACT 13% 4% 22% 3% 53% B% 26% 99% 100% 100% = 100% 100%
DSS nef administrative savings (£m) » -10 3S 335, 15° 25 ee) 120 435 "435 435 435 435, 1,306 87
‘contingency -10 -10 10 10 -10 -10 +10 10 10 10 +10 10 84 36
DSS programme savings (£m) Q 5 90 “130 335 135 135 135 135 135 135 135 87 469
Payment (o banks/POCL for "unbanked" o - 8 -20 -30 -33 32 32 +32 32 32 32 +32 - = +205 WS
Total DSS savings (£m) -20 -28 25 s 8 168 213 528 528. 528 528 528 1,864 “385
first round impact on POCL profits dm 0 Ss 10 20 30 40 0 336 137 +138 <135 +134, 302 2D
compensation to subpostmasters 0 108. +108 -108 108 108 0 o .. 0 a 0 0 429 429
cast of banking technology for POCL 0 “a4 +24 46 17 2106 93 93 93 03 93 93 554 293,
profits fos POCL from banking £m 0 0 Q 7 22 16 100 116 17 138 nod 120 428 98
Total impact on POCL 0 AI? 122 11 217 . 9B 7 113 13 2113 -109 107 “B56 3545,
impact on PO network rural 0 <800--1,200 © -1,800»-2,400- «3,000 -3,000--3,000 -3,000 3,000 -3,000 . -3,000
urban O° +1400, 2,800 4200-5600 -7,000-7,000 . -7,000 -7,000_-7,000 -7,000 ° -7,000..
fotal - _ 9 +2,200 © 4,000» 6,000 -8,000 -20,000 -10,000- 10,000 -10,000 -10,000 -1 0,000 -10,000
NET IMPACT ON FUBLIC SECTOR . 1,008 +160
HORIZ_DY.XLS= 161098 2083
Keue!
HORIZON: SUMMARY OF CENTRAL OPTIONS - 34 FLOOR PAYMENT TO POCLENDS WHEN ACT MIGRATION COMPLETE
Baseline
%of claimants paid by ACT
‘% of transactions made by ACT
DSS administrative costs
CAPS costs
total administrative costs inc CAPS
DSS programme savings (ESNS)
POCL profits £m
POCL network -
total
1998-99 1999-00 2000-0)- 2001-02 2002-03 2003-04 2004-0S 2005-06 2006-07
Continue with project: ACT migration 2005/06 to 2007/08, floor unti 2007/08 (per GC 11/10/98)
%of claimants paid by ACT
%of transactions made by ACT
DSS net udministrilive savsugs (Li)
contingency
DSS programme savings (£m)
payment for “unbanked"
Total DSS savings (£m)
fust round impact on POCL profits £m
compensation 10 subpostmasiers.
cost of banking technology for POCL
profits for POCL from banking £m
TotalimpactonPOCL =
impact on PO network moral
urban
total
HORIZ_D1.XLS - 16/10/98
10 2010 102005
30% 32% 33% 39% 36% BBY 59% AN% AZ
13%. *15% «=—«16%=—=«*28M_=«*:«*NDM.:SC« NM 22%D_. 24-25%
530530 525—s«i« S25 SS 520, S20 SSS
100 20 20 20 20 20 2... 2 20
630° $80 0545 SMS SMS SAD S40. 5355S
7 86 0 0 50 so 50 50 30 50
30 25 20 10 0 +10 +20 50 -50
9000 8800 8600 8400 8,200 8,000 7,800 7,600 7,400 Tomer ome
10,000 10,000 10,000 10,000 10,000 10,000 10,000" 10,000 10,000
19,000 18800 18,600 18400 - 18,200 18,000 17,800 17,600 . 17,400 17,200 17,000 16,800
30% «32% «33% ~~ 35% «Ye «BB «39% «AS Ge «BBY (100% 100% \
13% 15% 6% 18% 19% 21% 22% 25% 52% 71% 100% 100%
2 80H 40 so 40 30 ne ee 426 “49
0 0 a 0 0 0 0 0 +410 :10 10-10 22 0
0 5 0 135 ° 135 135 ° 135 I 138 1351383851358 868 490
0 0 0 t) t) 0 t] o 0 0 ) 0 0 0
25, “15 50 ns 15 395 195 175 155 165 528 525 1,272 44
0 B ‘10 , 20 30 40 50 80 32 69 -205 +204 W 3
0 0 0 0. 0 4oO ee 0 0 0 38 0
0 3003 zy a 4 4 “l 5 A a a at 8
0 0 0 3 “5 uM 23 45 A 107 123 123 289 “3h
0 2 7 22 - 34 53 “2 93 m4 175 83 82 2sl 136
0 0 0 0 0. a o 0 0 0 0 a
0 0 0 0 0 0 0 500 +1,000 3,000 1,000 +1,000
0 0 0 0 0 0 0 500-1000 «1,000 -1,000 1,000 .
NET IMPACT ON PUBLIC SECTOR 1,523 516
10f3
2007-08
POL00028638
POL00028638
2008-09 2009-10 NPV @6% NPV @ 6%
POL00028638
eens] POL00028638
HORIZON: SUMMARY OF CENTRAL. OFYTIONS - BA FLOOR PAYMENT TO YOCLENDS WHEN ACT MIGRATION COMPLETE ‘ 3 .
f es
1998-99 3999-00 2000-01 - 2003-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 NPV @6% NPY @6% * 3
102010 (02005
BV Variant - Horizon cancelled: ACT compulsory for all claimants, from 2003/04, Noor ends 2005/06 = :
% of claimants paid by ACT . 30% 30% A2% 53% 10% 88% 100% 100% 100% 100% Wu 100% ak
% of transachons made by ACT 13% 14% 22% + 34% 53% BY% 96% 99% 100% 100% 100% 100%"
DSS net administrative savings (£m) 3 0 50 20 20 70 120 435 435 435 435 435 1,298 BB
conlingency +10 10 10 10 10 -10 10 10 10 560
DSS programme savings (£m) Go 5 90 130 135 135 335
Payment fo banks/POCL for “unbanked" Oo. 8 +20 +30 +33 +32 +32 !
Fon DSS-savlngs{Emn)— evens arecemane 218. 70, U2 163213 y
first round impact on POCL profits £m 0 20 30 40 50
compensation to subposimasters: 0 +108 -108 +308 0
cost of banking technology for POCL. ° a 46 N17 406 93 . -
profits for POCL from banking £m 0 0 5 20 36 1 100 116 7 MB 334, 42
Totatimpact on FOCL 0 IH -190 -154 7 6B 127 113" AW -109 +7719 +568
impact on PO network mural 0 1,800 2,400 © -3,000 3,000 3,000 —--3,000 ' 3,000 © -3,000-—-3,000 .
urban 0 4,200 © --5,600--7,000 7,000 -7,000 .-7,000 -7,000 -7,000-7,000 5
totst 0 -6,000 -8,0N0 10,00 -10,009 -10,000 -30,000 -10,000 -10,000 10,000
NET IMPACT ON PUBLIC SECTOR 1,069 +198
Notes
1 Anegative saving means an addition fo costs
2 Baseline assumes "business a5 usual” with no Horizon
3 Each option is expressed asa change relative to the baseline
4 NPV figures for option 2 and 3 assume that no compensation Js paid either by or to ICL. .
5 VAT is included in payments to Pathwoy by both BA ‘and POCL, and 1s included in the cost of new banking, technology in option 3 i
i
Zz .
2,
$
-
4:
HORIZ_DT XLS - 16/10/98 . 30f3
POL00028638
Sees - POL00028638
e101 BOIL “SIX LA-ZTUOH
P.OS/IT
ao)
oso! 000'%ZE © —ooz‘et © COMLL © GON'LE «COSTE GKO'RI = Oz'st§COM'EL © C0N'st oD s'st o00'6t 10)
reg Doo'o! © aoo'o - caO'O © cao‘! - 000'! cao'o! © aao'or © 00't ©=—000'0!:-OU'G! = adO'at~—uO'AI waqin .
cee teceeececeee a, 0089, 000% 002% oor op" os" d00's_ 0B OO’ co's. cos's 000%. yen wows 190d I
ou 09° 09° OO RE TT yen ore sone raiyerd Tad me — I
os os os os. oF as os as os OL “OL OL (SNSa) sSupies ouuaiSeud sso
sts OES ots sts sts Os Ors sus - SBS bs oss 0&9 SAV “DUE $1509 IAlFaN}SIU}WPU [eTO} I
oz oz oz 0z oz 0z ha r4 tr 4 0v 0% Oot $1800 SUVD I
sos ols as sts sl¢ ozs ozs sts sts sts OES ots SISO DAL IBIINULWPT SSE I
OE UBT Ble UST Re eT VIZ “GI %3U bo § WSL Ed LOV Aq apaut suarorsues) jo %,
Vly “SY by “ey ly GE YRE KE SSE MEE BE KOE LDV Aq pred syrounroys jo. %
auyaseg
‘s00T oF O10Eo}
%9® AIN %9O AdN or-6007 608007 -80-L007" £0-900T 9u-So0z SOPOT FO-EdOT CO-ZWOT zO-1O0e fo-d00~ 00-6665 66-8665
- 3 . SNIDAG NOLLYUDIN. LOY AUWOS TAOS NAAN SUNE 1904. OLANINAVd HOOTE Vil * SNOLLEOIVULNAD 10 AUVINIINS (NOZRIOIE .
POL00028638
POL00028638
iis) -
HORIZON: SUMMARY OF CENTRAL OPTIONS - BA FLOOR PAYMENT TO POCLENDS WHEN COMPULSORY ACT MIGRATION BEGINS ‘ ”
1998-99 1999-00 2000-0) 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 NPV @6% NPV @6%
102030 fo 2008
81
86
2 Continuation of Horizon without BPC, move to ACT (compulsory from 2003/02), Noor pard unto 200002 4
‘Wot clmants paid by ACT 30% 30% ©«=—« 42% «= «53% «= 70% «= BBY «100% = 100% ©=—« 100% = 100% = 100% = 100% ;
of transactions made by ACT 13% 14% 22% 34% 53% 73% 96% 99% 100% += 100% += 100% = 100% z
3
DSS net administrative savings (£m) “10 “5 40 10 95 190-290 435— ABS ABS BSS 1,568 349 a
contingency’ =10 10-10 -10 +10 =10 10-10 -10 =10 100-10 “84 -56 7
DSS programme savings (Lm) 0 5 90 130 135135 135 135 135 135 $35 135 847 469 i
Payment to banks/POCL for “unbanked @ 8 20° -30 -33 32 32 -32 32 3200-32 232 206 “116 ;
wee ee met apa DSS gS aay SZ B20 HOD = ABI» 285 BI RB = BIB ABB BB a 2B or Dyk one BEB oome mo ;
first round mpact on POCL profits.<m 0 5 10 a 3000 Ml 520-213 9B 183 8B SS +603 m4 4
compensation (o subpostmasters 0 0 0) 0 0 0 0 -94 “94 94 94 94 -263 0 i
cost of banking technology for POCL 0 3 “33 n <1 51 “51 231 a] 5} “I 5) -28 -195
profits for POCL from banking £m 0 0 0 N36 B 103 116 7 ng 120120 490 159
Total impact on TOCL, 0 2 23 ro] 4S 4 GO 222° 196-189-163 130 604 -109 '
impacionPO network tural 0 t) C) 0 0 0 0 200-400 «600» +800 - 1,000 3
uiban r) 0 0 0 @ ) 0 -3,200, -2,400-3,600-4,800—-6,000
fete! ® 0 a a 0 0 0 -1400 -2,800 -4,200 -5,600 -7,000
NBT IMPACT ON PUBLICSECTOR 1821 337
3 Marizon cancelled: ACT compulsory for all claimants from 2001/02, Noor enus 2001/02
‘eof claimanss paid by ACT 30% 30% «= 42% «= «$3% © 70% «= 88% «100% © 100% © 100% += 100% «= 100% «= 100% ‘
eof transactions made by ACT 13% 14% = 22%. 34% 53% «73% «Ye «99% «100% 100% =— 100% 100% j
DSS nel administrative savings (fm) -10 15. 40 10 95 «190290 435— SIS 43S 4385 1,568 349
contingency 10-10-10 “10 10 10. 10 -10 +10 -10 “10-10 “84 “56
DSS programme savings (£m) 0 5 90 (130 13500135135 135 135 135, 135135 847 469+
Payment fo banksPOCL for “unbanked 0 B20 30 “33 “32 32 32 -32 232 “32 -32 -206 “116
Total DSS savings (£m) -20 28 20 «1002S's«s18?s«i2B3 3K} SB CS2B S28. 528 528 2,128, 646
first round impact on POCL profits 4m 0 5 19 4s 35 21 0 138-137, 4863S. #134 288 94
compensation (o subpostmasters 0 +108 «= +108, 108. 108-108, 0 0 0 0 0 0 429 429 z.
cost of banking technology: for POCL. 0 “14 124 460 -N7— 106-93 93 93° 93 93 93 =5S4 293 3
piofits for POCL. from banking Lm 0 0. 0 6 2h % 100 16 n7 8 9 «120 430 100
otal impact on POCR 0. -N7 1220 SDL 7 NSB T0910 841 “529 Pk
impactonPO network ~ muna 0 —--800-«-+1,200-1,800 , «2,400 -3,000 -3,000°§ -3,000 -3,000 -3,090. 3,000 $i
urban 0 +1400 -2800 -4,200 -5,600 -7,000 -7,000 -7,000 -7,000 -7,000--7,000 ae
fotol o t
+2,200 «4,000. -6,000 -8,000 -10,000 -10,000 -10,000 =10,000 -10,000 -10,000 -10,000 .
. NET IMPACT ONPUNLICSECTOR 1,285, 7
S2l-3
WORIZ_DT.XLS - 16096 . 2013 4 ‘ 3
POL00028638
a POL00028638
HORIZON: SUNNMARY OP CENTRAL OPTIONS - BA FLOOR PAYMENT TO POCLENDS WHEN COMPULSORY ACT MIGRATION BEGINS Be
. wrm.f
: 1998.99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2008-06 2006-07 2007-08 2008-09 2009-10 NIV @6% NPV @6% —
; 02050 © 102005 2:
3V Varinnt - Horizon cancelled: ACT compulsory for all ctnimants from 2003/04, floor ends 2003/04 B
‘eof claimants paid by ACT 30% «30% «42% «53% = 70% ©«—«BB%_—«100% © 100% © 100% © 100% © 100% 100% 4
% of transactions made by ACT 13% 14% 22% «34% «—«S3% «73% «9% «NNN «00% «100% = 100% 10 zh
DSS nef administrative savings (fm) + -10 -50 10 9 «190290 3S 35435 AS 1,565 346 2 d
contingency . -10 10-10 10 “10 -10 +10 +10 -10 -10 -10 =10 -84 56 F
DSS programme savings (Ln) 0 5 90 130 135 135 135135 335138 135 135 847 469 ‘
Payment (ol 1OCL for “unbanked 6 B 20-30 233 “32 32 -32 320032 232 +32 -206 +6 :
“EetaL DSS savings Clamp SISTERS TO TOD AB BI BBB —SBB + + B2B = = 828- = 828 --- S28... 2122...
first round impact onPOCL profitsim 0 3 10 20 300«=« BO 16134 13G1BS BA -205 136°,
‘compensation fo subpostmasters 0 108 «= 108-108, 108-08 0 0 0 0 o-: 0 2429 -439
cost of banking technology for POCL. 0 14 2A 467 +106 93 93 +93 +93 +93 93 554 1293
profits for POCL from banking £m C) 0 0 0 s 20 360 6 oo 116 7 ug 334. 42
Total impact on POCI. O° 137-122, 34-190 = 7 93 127 NB 2209 854 “SAS
impact onPO network rural 0: -800 1,200 -1,800 © -2,400-3,000-3,000 -3,000 -3,000 -3,000 -3,000 -3,000
: urban 0 -1400 «-2:800 -4200« -5,600-7,000 «7,000 -7,000» -7,000 -7,000 * -7,000 7,000 :
tote O- 2200-4000 --6,000 RN -1N,NNN 10,000 -10.000_ -10,000 -10,000 -10,000 10,000
NETIMPACT ONPUBLICSECTOR - 4,268 99
Notes :
1 A negative saving means an audition fo costs
2 Baseline assumes “business as usual" with no Horizon « ci
3 Bach options expressed as a change relative fo the baseline . t
4: NPV figures for option 2 and 3 assume that no compensation is paid either by o1 1 ICL. 7
5 VAT is meluded in payments 10 Pailiwvay by both BA and POCL, and is sncluded in the ost of new banking technology in option 3 7
UORIZ_DLALS - 16098 : 3013
POL00028638
POL00028638
r.CTCTSTS”t‘CCCTTT'?
ahs]
44
44)
Summary of kcy assumptions
i H
HM Treasury
Benefits Agency/Post Office Counters Linuted antonation project
Working Diaft- 16 October 1998
86-01-91
‘Assumptions underlying financial resujts for options - where the floor payment from BA to POCL ends when migration io ACT is complete
Option 1
Option 2
Option 3
Yoel
Underlying rate 1.5% 3.5% / 1.5%
BA moves to ACT 2005/06 - 2007/08 2001/02 - 2003/04 2001/02 - 2003/04
Rate of migtation when ‘compulsory’ I 20:40:40 34:33:33 34 33 33
% of transactions paid by ACT
Lag between migration of customers
and transaction
Periodicity mix: current average
Penodicity mux: current average
periodicity of paperbased > ACT
Periodicity mix: curient average
periodicity of papcerbased > ACT
DSS net administrative savings
perodicity of paperbased > ACT
Cost of withdrawing cash by the
“banked”, when moving to ACT
BA floor paymem to POCL ends 2007/08 2003/04 2003/04
Incremental investment in CAPs Yes - for interface with Horizon : -
Contingency
ACT costs 10% of ACT costs €10% of ACT costs 10% of ACT costs
DSS progranme savings
Benefit encashment fraud savings
{incremental fo savings from ESNS)
690% of estimated fraud - accrues
with BPC (then ACT)
90% of estimated fraud - accrues
with ACT
90% of estimated fraud - accrues
with ACT
_ 8e8
_ tu
NO 88hr
s0/t0°d
Sted
POL00028638
POL00028638
kei]
HM Treasury
Benefits Agency/Post Ofice Counters Lunted automation project
Worhmg Droft + 36 October 1998
Assutiptions unierlying finan cit results for aptions - where die Noor payment from BA to POCL ends when migration to ACT is complete
Changes to periodicity
Option 4
Option 2
Option 3
Payment for the “unbanked”
Incentivise banks (eg remunerate for
£9518. of Jow value. cash withdrawals)__I
Upfront £10/account + on-going
‘Upfront £10/accoumt + on-going,
L20pa for half.of the. \umbanked”....I
Upfront £10/account + on-going
£20pafor half of the.‘unbanked?_I..
First round impact on POCL profits
Migration of non-BA business from
closed PO to other PO
Retail custom dependant on PO,
foosfall
PO vision: new services included in
modelling
Accessibility value to customers of
loss of nearest PO (cst 70-80p/visit)
I £20na for dull ofthe “unbanked’s ..
60%
27-36%
Ranking only
Excluded
60%
27-36%
Banking only
Excluded
60% . .
27-36%
Banking only
Excluded
Compensation to subpostmasters
26 months remuneration duc on
incremental closure of POs
Level of remuneration:
Paid whether compulsory or
voluntary closure
Variable subpostimaster
temuncration = midpoint of before
and after migrates to ACT.
Paid whether compulsory or
voluntary closure
Variable subpostmastes
remuneration = midpomt of before
and after migrates 1o ACT
Paid whether compulsory or
voluntary closure
Variable’ subpostmaster
temunesation = midpoint of
before and after migrates to ACT
Costing of banking technology for"
POCL
Technology
Horizon
Honzon + POCL remunerates ICL
for ‘reduced margin’ re BPC
Debit terminals initually, then full
functionality system procured
2
89h-L
5
a
i
7
a
80/20°d
gen
POL00028638
POL00028638
kei)
: HAM Treasury 2
Benefits Agency/Post Office Counters Linuted automanon project
Working Draft - 16 October 1998
Assumplions underlying financial resulis for options = svhere the Moor payment from BA to POCL ends when migration to ACT is complete
86-01
Option } Option 2 Option 3
Profits for POCL from banking . .
Revenue . 50p per transaction 50p per transaction 50p per transaction
Transactions per customer per annum - I 30 30 30 :
Marker shareof benefit custonters~ 2001702 10%, 2002103-TS%;-~ ~*~) 2001702 40%, 200Z/0F 50%; "2001702 30%; 2002703 ATY%;,
migrating to ACT 2003/04 35%; 2004/05 50%; 2003/04 65% 2003/04 50%
. 2005/06 65%
Market share of other cash withdrawal I Increasing 10 7.5% Increasing 10 7.5% Increasing to 7 5%
Impact on PO network :
Closures Managed Managed Avoid loss of control _
Bh
80/80
2 9Ubed
POL00028638
POL Same
"Te=10=98"
ae ; Tere
ora ; . HM Triasucy
bul i Benefits Agensy/Post Office Counters Limed automation project
i Working Draft 16 Octobur 1998
Appendix A; List of documentation made available to the review
i
1
Benefits Agencly / Post Office Counters Automation Project. - a note by the Prime
Minister's Policy Unit and the Treasury
Review of the Benefi its Agency / Posi Office Counters Automation Project - Working
Group Report = july 1998
BA / POCL Aditomation Programme Review - Report of the Independent Panel - July
1998 :
Post Office Cobinters Limited - ICL Pathway Technical Architecture Study - Pontis
Consulting September 1998
POCL Strategy} Integrated Economics - Post Office Counters Limited Presentation to
HM Treasury Abgust I 1998
Horizon Banking Infrastructure - ICL Pathway Ltd - version 1.) - Sth August 1998
POCL Banking Posts (miscellaneous faxes) - HM Treasury and POCL
Option 3 sna Gonathan Evans, POCL to Adam Sharples, Treasury, 2998 and
4.9.98)
Departmental RE view of Benefit Payment Options - Proposal for increased use of ACT,
DSS, Septemben 1998
POL00028638
POL00028638
16-10-98
18:41
FROM-
keke
T-468- P.05/08 F726
. HAM Treasury
Benefits AgensylPost Opice Counters Lied automation project
- Working Drafi +16 October 1998
Appendix B:Banking functional assumptions
In addition to
je core “manned ATM” services assessed in 2.2.1 above, a number of
further services have been discussed by POCL, which have not been reflected in the
viability assess{nent, These are listed below, together with the rationale for excluding
them from the skrvice scope
2 ]
Service ‘
Comments
Information of banks’ fidancial
products
aren
Service would require careful negotiation with
banks, and may raise regulatory issues re. giving
financial advice Propose that this is not included in
the banking validation ‘scenario, (This would not
preclude provision of simple product details).
Corporate bagking for small
business users
!
Represents a distinct service area, which may. have
different market requirements. Extension into this
area may also be resisted by banks.
7
Passhook withdjawals
This raises logistical issues, particularly given the
different passbook printer standards and practices for
different institutions.
7
Opening and cldsing accounts
Banks are likely to want to retain full control of
account openings and closures. Assume thar POCL
will provide no more than a form-filling service here.