POL00088209 - Post Office Agent Branch Losses - Mitigation Proposal (draft). Author Angela Van Den Bogerd. Document marked strictly confidential.

Evidence on official site

POL00088209
POL00088209

POST OFFICE PAGE 10F 6

Agent Branch Losses-Mitigation Proposal

Author: Angela Van Den Bogerd Date: 01/08/2016

Executive Summary

Context

Over the last three years losses identified at branch on risk audits have increased steadily
with the value of individual losses also increasing; this is despite the number of risk audits
being less year on year. Postmaster debt recovery has also been adversely affected over the
same period.

With the recent decision to merge the Fraud team with the Support Services Resolution team
there is a further opportunity to have the end-to-end.management of branch losses within
Support Services along with the ownership for these branch losses.

To complete the life cycle of branch losses the proposal is that the branch audit function and
associated team members and budget transfer to Support Services. This paper sets out the
rationale for this.

Questions addressed in this report,

1. What is the value and trend of losses sustained at branch risk audits in recent years?

2. What is the impact on the branch loss of less risk based audits being done now than
previously?

3. How could we better monitor and respond to agent losses in branch

Conclusion and, Recommendation

This paper recommends,that the audit function and associated team and budget transfers to
Support Services from Network Operations to form specialist team that combines the
investigative skills from project Sparrow; the existing Fraud analysis skills and the existing
auditor skills. The honed approach of this newly formed specialist team will ensure that
losses are monitored and mitigated before they escalate; that a more effective and efficient
approach to risk audits is deployed and that debt is recovered from former and existing
Spmrs in an appropriate and timely manner.

Strictly Confidential Angela Van Den Bogerd Aug 2016
POL00088209
POL00088209

POST OFFICE PAGE 2 OF 6

Input Sought Input Received

The decisions we would like from the GE are: The audit data referenced in this paper has
been supplied by Sue Richardson, Network
Operations team and Kim Abbotts for the Q4
risk audit data.
1.Endorsement of the proposed approach
to risk audits
2.Endorsement of the transfer of the
branch audit function and the 18 FTEs
to Support Services
3.Agreement on an implementation date

The Report

What is the need or opportunity and why now?

The value of branch losses sustained at risk audits is increasing year on year as is the
average loss per branch. With the ongoing business focus on simpler and cheaper to run we
have been exploring a new approach that aims to respond to branch accounting anomalies
quicker by assessing the potential financial risk and appropriate intervention whether that be
telephone assistance or on-site audit.

What is the value and trend of losses sustained at branch risk audits in recent years?

As can be seen from the table below the value of losses verified at risk audits has increased
year on year despite a fewer number of risk based audits having been done. The risk here is
that in reducing the number of risk audits undertaken we are taking longer to verify the
actual loss which by default typically increases the value of that loss on audit. This is
evidenced by the increase in the average branch loss discovered when the risk audit is
actually done - up 63% in 15/16 when compared to the prior year.

Risk Based Audits Findings 2013/14 - 15/16 inclusive until P9 15/16

Financial year 13/14 14/15 15/16 *Data until P9
Number of audits 747 697 376*

Value of losses on audit £1,292,490 £1,512,275 £1,335,794

Average loss per branch £1,730k £2,169k £3,552k

Variance on prior year +23% +63%

If we responded more quickly to the risk of the branch loss would the actual loss be less?
The data flow is not currently set up within the Fraud team (but will be going forward) to
track the actual time from request to delivery of a risk audit by the field audit team so we do
not have the average time to delivery. However the process as currently constructed can
result in 10 weeks for a risk audit to be actually actioned at branch. There are branch
examples that support the need to be more responsive to identified risks and although within
the parameters of the current process the scheduling and audit teams try to be flexible,

Strictly Confidential Angela Van Den Bogerd Aug 2016
POL00088209
POL00088209

POST OFFICE PAGE 3 OF 6

given the demands placed on a multi-skilled Field Support Advisor (FSA) team - that is
required to deliver induction training for new postmasters; remedial training; transfer,
closure, random, WHS and Crown audits in addition to being flexed to support NT activity —
risk audits have not always had the responsiveness required to mitigate the actual loss at
branch.

Appendix 1 details some recent examples of where due to a failure to intervene earlier the
loss identified on audit was greater - the examples are intended to show the impact of delay
(not to apportion blame); they demonstrate why we need to tighten up on responsiveness
and also get smarter with the tools we use to raise a flag of concern.

What do we propose to do and why?

This paper sets out an end-to-end life cycle approach for branch audit losses that would
enable the complete management of the branch loss from the earliest point of discovery ie
Fraud Analysis team to the collection of the loss/debt ie Agent Debt Recovery team. To make
this process complete the following needs to happen:

e The audit function to be detached from the existing multi skilled training and audit
team to form a specialist team within Support Services that combines the
investigative skills from project Sparrow; the existing Fraud analysis skills and the
existing auditor skills all enhanced by HORIce.

e With a skilled audit/fraud/ issue resolution team branch accounting anomalies will be
resolved more quickly and will invariably reduce the need for on-site face to face
support as this team will be skilled in providing bespoke telephone support to
branches. As the approach embeds this will reduce the average loss per branch
identified through this improved branch risk approach.

e With further development around the questions we pose to HORIce and the analysis of
the data produced we will be able to move to risk audits only which in turn will ensure
we target our resource more effectively and efficiently.

e The output of this newly formed team will equip the Contract Advisor to make
informed decisions regarding precautionary suspension and termination cases. This
will result in the investigation work being undertaken at the appropriate level and by
individuals who have a working knowledge of the Horizon system and associated
processes. These new audit roles will undertake the investigation work in advance of
the risk audit taking place so that they are more targeted and efficient whilst on site
ie they understand exactly what they are looking for as they have carried out the
investigation work and are therefore better equipped to have a constructive
conversation with/provide a better informed explanation to the postmaster at the time
of the audit.

The business case
e If we were to replicate the 697 risk audits undertaken in 14/15 18 FTEs would be

required. [audit prep time; post audit report; time on site and average travel time to
verified by Sue Richardson] Risk audits are undertaken by 2-3 FSAs on site.

Strictly Confidential Angela Van Den Bogerd Aug 2016
POL00088209
POL00088209

POST OFFICE PAGE 4 OF 6

e 18 FTEs to transfer with the risk audit activity from the Network Operations team to
the Support Services team along with the associated T&S budget. 2 Field Team
Leaders should also transfer as associated line managers. This is a straight budget
transfer from Network Operations to Support Services.

e The 18 FTEs would need to be geographically dispersed so as to provide economical
coverage of the UK.

e The team of 18 FTEs will merge with the existing Support Services Resolution team
led by Kath Alexander and Shirley Hailstones.

Risks & mitigations

e 18 FTEs will potentially incur longer travelling time and an increase in T&S costs due
to a reduced population to cover the same geographical consideration.

e By using HORIce and their honed investigative experience the newly formed specialist
team will be skilled in providing bespoke telephone support to the branch resulting in
quicker branch accounting resolution and in turn reducing the need for on-site face to
face support/audit.

What options did we consider?

e An alternative option would be to baton pass the Sparrow type investigation process;
the HORIce information tool and the responsibility for reducing branch losses to the
existing Network Operations team.

e This option has been rejected as:

- spreading what is being proposed as a specialist approach across the existing FSA
team would dilute the effectiveness of the approach

- the branch audit function is a better fit with the newly formed fraud/resolution and
Contracts team rather than the training team

- the tier 2 support function is a natural extension of NBSC which is already part of
Support Services

- the Sparrow investigative knowledge and expertise is already within Support
Services

What do we need to do next to progress?
« Agreement from Network on the transfer of the audit function to Support Services
e Agree the go-live date for the new team - ideally this would be at the same time as the
transfer of the Fraud team to Support Services currently planned for 1% Oct [TBC]
e Implement the new structure and improved ways of working

What would the impact be of delaying or rejecting the decision to progress?

Strictly Confidential Angela Van Den Bogerd Aug 2016
POL00088209
POL00088209

POST OFFICE PAGE 5 OF 6
Delaying or rejecting the decision to progress would:

e Result in branch losses discovered at risk audits continuing to increase in total value
and average branch loss

e Result in more precautionary suspensions due to a delay in actioning the risk audit.
Experience tells us that the lower the loss identified in branch the higher the likelihood
of us recovering that money from the postmaster. Once the loss approaches or goes
above £10k the postmaster is usually less able to make good the loss. This then results
in a suspension and potentially a loss of service for a period of time.

Appendix 1

Carr Hill - Local

On the 23 Dec 2015 this branch posted £1k centrally; typically no concern. On the 27"
January 2016 a further £24k was settled centrally. This was at a branch with £1450 a month
fees so well in excess of its annual remuneration. The concurrence report (used to advise
Contracts Advisors (CAs) that postmasters have not made good their losses as per their
contract) passed to the CA on the 19" Feb 2016 detailed a £25k shortage. Separately

and coincidently a special audit request was made by lial as there were a number of issues
that had been raised and was enacted on the same day as the £25k concurrence report
arrived with the CA. The audit discovered a loss of £39k. Therefore within a matter of some 4

Strictly Confidential Angela Van Den Bogerd Aug 2016
POL00088209
POL00088209

POST OFFICE PAGE 6 OF 6

weeks the loss had increased by a further £14k. Had it not been for the special request audit
it may have been a further 4 weeks before a risk audit was enacted.

Shavington - Local

On the 21% May 2016 there was a settle centrally debt of £11k, and that appeared on the
concurrence report as not being made good. This branch has monthly fees of £1168. On the
14™ July [what action did the CA take on receipt of the first concurrence report?] it appeared
again on the concurrence report but was now £18k. There was a physical intervention by a
FSA as the operator was claiming that he didn’t know where the losses were. The FSA could
not identify where the loss had occurred or any weaknesses within the branch.

As of last week the loss is reported as £30k; so from the 4 week trading period the operator
declared in May he has lost a further £19k.

Because of the CA concern this has become a special request audit which is scheduled for 2"
August.

Newington - Local

Suspicion was first raised about this branch in November 2015 due to high levels of cash
holdings and requests still being made for additional funds. No action was taken in
November.

An audit was undertaken at this branch 15" April 2016 and a loss of £196k discovered.
Analysis undertaken demonstrates that had the intervention taken place in November the
loss although significant would have been £90k, so £106k less than the final figure.
Route-cause analysis uncovered a failure in communication between cash management and
the Fraud team and whilst again no reflection on scheduling team this is a good example of
how a failure to respond and act promptly when concerns are raised does add to a loss.

Good news is w/has advised that she has seen documentary evidence of a loan the
suspended agent has secured to repay the loss.

Strictly Confidential Angela Van Den Bogerd Aug 2016