The Bounce Back Loan Scheme: an Update — UK National Audit Office, HC 861 (December 2021)

Record facts

Full text

A picture of the National Audit Office logo
SESSION 2021-22
3 DECEMBER 2021
HC 861
REPORT
by the Comptroller 
and Auditor General
The Bounce Back Loan 
Scheme: an update
HM Treasury, Department for Business, Energy & 
Industrial Strategy, British Business Bank plc

The National Audit Office (NAO) scrutinises public spending 
for Parliament and is independent of government and the civil 
service. We help Parliament hold government to account and 
we use our insights to help people who manage and govern 
public bodies improve public services. 
The Comptroller and Auditor General (C&AG), Gareth Davies, 
is an Officer of the House of Commons and leads the NAO. 
We audit the financial accounts of departments and other 
public bodies. We also examine and report on the value for 
money of how public money has been spent. 
In 2020, the NAO’s work led to a positive financial impact 
through reduced costs, improved service delivery, or other 
benefits to citizens, of £926 million.
We are the UK’s 
independent 
public spending 
watchdog.
We support Parliament 
in holding government 
to account and we 
help improve public 
services through our 
high-quality audits.

Report by the Comptroller and Auditor General
Ordered by the House of Commons 
to be printed on 1 December 2021
This report has been prepared under Section 6 of the 
National Audit Act 1983 for presentation to the House of 
Commons in accordance with Section 9 of the Act
Gareth Davies 
Comptroller and Auditor General 
National Audit Office
30 November 2021
HC 861  |  £10.00
The Bounce Back Loan 
Scheme: an update
HM Treasury, Department for Business, Energy & 
Industrial Strategy, British Business Bank plc

The material featured in this document is subject to National Audit 
Office (NAO) copyright. The material may be copied or reproduced 
for non-commercial purposes only, namely reproduction for research, 
private study or for limited internal circulation within an organisation 
for the purpose of review. 
Copying for non-commercial purposes is subject to the material 
being accompanied by a sufficient acknowledgement, reproduced 
accurately, and not being used in a misleading context. To reproduce 
NAO copyright material for any other use, you must contact 
copyright@nao.org.uk. Please tell us who you are, the organisation 
you represent (if any) and how and why you wish to use our material. 
Please include your full contact details: name, address, telephone 
number and email. 
Please note that the material featured in this document may not 
be reproduced for commercial gain without the NAO’s express and 
direct permission and that the NAO reserves its right to pursue 
copyright infringement proceedings against individuals or companies 
who reproduce material for commercial gain without our permission.
Links to external websites were valid at the time of publication of 
this report. The National Audit Office is not responsible for the future 
validity of the links.
010814  12/21  NAO
This report updates on government activities to protect 
public money since the government’s Bounce Back 
Loan Scheme closed to applicants and loan repayments 
started. It reviews the total number of loans made; the 
status of the loans at September 2021; the levels of 
identified fraud and the counter-fraud activity taking 
place; and the estimated level of potential fraud and 
credit losses for the Scheme overall.
© National Audit Office 2021

Contents
Key facts  4
What this report is about   5
Summary  8
Part One
Loan disbursements 
and repayments  15
Part Two
Counter-fraud measures  27
Part Three
Other changes and activities 
under the Bounce Back Loan 
Scheme (the Scheme)  42
Appendix One
Our approach  51
Appendix Two
Roles and responsibilities 
for the Bounce Back Loan 
Scheme (the Scheme)  57
Appendix Three
Number and value of loans at 
September 2021  59
Appendix Four
Our fraud evaluative framework  60
If you are reading this document with a screen reader you may wish to use the bookmarks option to navigate through the parts. If 
you require any of the graphics in another format, we can provide this on request. Please email us at www.nao.org.uk/contact-us
The National Audit Office study 
team consisted of:
Gregor Botlik, Rachel Burden, 
Rachael Savage and 
Nadia Xarcha, with assistance 
from Lewis Ferguson, under 
the direction of Simon Reason.
This report can be found on the 
National Audit Office website at 
www.nao.org.uk
If you need a version of this 
report in an alternative format 
for accessibility reasons, or 
any of the figures in a different 
format, contact the NAO at 
enquiries@nao.org.uk
For further information about the 
National Audit Office please contact:
National Audit Office 
Press Office 
157–197 Buckingham Palace Road 
Victoria 
London 
SW1W 9SP
020 7798 7400
www.nao.org.uk
@NAOorguk

4  Key facts  The Bounce Back Loan Scheme: an update 
Key facts
1.5 million
total number of loans issued 
under the Bounce Back Loan 
Scheme (the Scheme)
£47 billion
total value of loans issued 
under the Scheme
£17 billion
the Department’s estimated, 
but highly uncertain, value of 
losses from loans that will not 
be repaid (31 March 2021)
£4.9 billion 
the Department for Business, Energy & Industrial Strategy’s 
(the Department’s) most likely, but highly uncertain, estimate of 
the value of fraudulent loans in the Scheme as of 31 March 2021, 
based on a sample of 1,067 loans  
£6 million 
the minimum value of fraudulent loans that the Department 
expects the National Investigation Service (NATIS) to recover 
from organised crime over three years 
£67 million
forecast administrative costs for the Scheme for 2020-21 to 
2024-25, as of September 2021 
£32 million  
the Department’s forecast additional budget to cover the 
costs of its ‘enhanced’ counter-fraud operations, regardless 
of the scale of fraud  

The Bounce Back Loan Scheme: an update  What this report is about  5 
What this report is about 
1	
The Chancellor of the Exchequer announced the Bounce Back Loan Scheme 
(the Scheme) on 27 April 2020. The Scheme was the largest of three COVID-19 
related business support loan schemes (see Appendix Three) and sought to provide 
to the smaller end of small and medium-sized enterprises (SMEs) a “simple, quick, 
easy solution for those in need in of smaller loans”. The Department for Business, 
Energy & Industrial Strategy (the Department) launched the Scheme on 4 May 2020, 
offering loans of up to £50,000, or a maximum of 25% of annual turnover, to support 
businesses during the pandemic.
2	
The loans were provided by commercial lenders (for example, banks, building 
societies and peer-to-peer lenders) directly to businesses who are expected to 
repay the debt in full. Government provided lenders a 100% guarantee against the 
loans (both capital and interest). This means if the borrower does not repay the 
loan, government will step in and pay the lender. The loans have a fixed interest 
rate of 2.5% and a maximum length of 10 years. In the first year of the loan there 
are no capital repayments due, and the government pays the interest – making it 
interest‑free for the borrower.
3	
HM Treasury developed the Scheme with the Department and the British 
Business Bank (the Bank), which the Department owns. HM Treasury and the 
Department monitor the Scheme and set its overarching terms. The Bank was 
responsible for implementing the Scheme, for example accrediting lenders to the 
Scheme before they made loans, and its ongoing administration in consultation with 
HM Treasury and the Department. 
4	
We reported on the Scheme in October 2020.1 At that point, it had delivered 
more than 1.2 million loans totalling around £37 billion. We concluded that 
government moved very quickly to set up the Scheme, prioritising one aspect of 
value for money – payment speed – over all others and was prepared to tolerate a 
potentially very high level of losses as a result. The Scheme aimed, in most cases, 
to deliver money to borrowers within 24 to 48 hours of application. It achieved this by 
removing the requirement for banks to conduct credit and affordability checks before 
making loans and allowing potential borrowers to ‘self-certify’ the information they 
provided on their application. Lenders were required to conduct some counter‑fraud 
checks before making loans.
1	
Comptroller and Auditor General, Investigation into the Bounce Back Loan Scheme, Session 2019–2021, HC 860, 
7 October 2020.

6  What this report is about  The Bounce Back Loan Scheme: an update
5	
The Department recognised the risks to the taxpayer of prioritising speed. Its 
Accounting Officer (AO) sought and received a Ministerial Direction on the grounds 
that the spending proposal would breach HM Treasury rules on regularity, propriety, 
value for money and feasibility. On 1 May 2020, the Secretary of State for Business, 
following approval from the Chancellor, directed the AO to proceed with the Scheme. 
The Bank raised similar concerns in a Reservation Notice to the Department and 
received a written directive from the Department to proceed with the Scheme on 
3 May 2020.
6	
The Department, with the Chancellor’s agreement, extended the original closing 
date for loan applications from 4 November 2020 to 31 March 2021 to support 
businesses during the ongoing pandemic. It also introduced options for borrowers 
to increase the size of existing loans to the maximum amount to which they are 
eligible (‘top-up’), as well as providing the borrowers more time and flexibility to repay 
(‘Pay As You Grow’). These repayment options include extending the duration of the 
loans from six to 10 years, moving to interest-only payments for six months, and 
taking a repayment holiday for six months. The Scheme closed for new applications 
on 31 March 2021, and repayments for the first borrowers started in May 2021. 
The Department sought, and was granted, approval to amend the Scheme using 
the same Ministerial Direction and written directive as at Scheme launch.
Report scope
7	
This report follows on from our October 2020 investigation into the Scheme. 
In Part One, we provide a factual update on the Scheme performance information. 
This includes, for example, the number of loans issued, their total value, borrower 
characteristics and the Department’s estimate of potential Scheme losses. 
This section does not evaluate the Scheme’s performance as the full repayment 
of loans will not be known for some time. 
8	
In Part Two, we review the Department’s activities to protect public money 
through its counter-fraud activity for the Scheme. This section provides an update 
on the Department’s latest estimates of Scheme fraud, drawing upon our financial 
audit of the Departmental accounts. It also evaluates the government’s counter‑fraud 
strategy for the Scheme and its activities to date against: our good practice guide 
on fraud and error;2 and the guidance from the Government Counter Fraud Function. 
This section is primarily evaluative in nature because since we last reported the 
government set out how it will pursue fraud and we have some evidence of how its 
approach is working.
2	
National Audit Office, Good practice guidance: Fraud and error, March 2021.

The Bounce Back Loan Scheme: an update  What this report is about  7 
9	
In Part Three, we report factually on the process to recover outstanding loans, 
which had not been finalised when we last reported. We update on the costs of 
administering the Scheme and the Department’s and Bank’s plans to evaluate 
scheme performance. We also draw together information, where available, on the 
borrower’s experience of the Scheme, which was a particular area of interest when 
the Committee of Public Accounts reported on this Scheme in December 2020. 
The report does not assess the Scheme’s value for money as the outcome for 
businesses, lenders, and the taxpayer will not be known for some time. 
10	
The study draws on six main datasets from three different sources, 
including the Department, the Bank and the Government Counter Fraud Function. 
These datasets differ in nature and timing and are described in the Appendix One 
(Figure 15). As a result, not all figures in our report reconcile.

8  Summary  The Bounce Back Loan Scheme: an update
Summary
Key findings
Loans and repayments
11	
The Scheme made 1.5 million loans worth £47 billion, mainly to small and 
micro-businesses, as of 13 September 2021. This exceeds the Department’s and 
Bank’s estimates at launch that the Scheme would support between 800,000 and 
1.2 million businesses with between £18 billion and £26 billion of loans. Around one 
quarter of all UK businesses received a Bounce Back Loan. Most of the loans – more 
than 90%, or £39.7 billion – went to micro-businesses (turnover below £632,000), 
a similar rate to when we reported in October 2020. The seven main UK banks 
provided most of the loans (90% of the total value of loans made). The remaining 
10% of loans by value were provided by other banks and non-bank lenders, such as 
peer‑to‑peer lenders (paragraphs 1.3 to 1.5, 3.9 and Figures 3 and 12).
12	
The Department and the Bank have limited evidence on how businesses have 
used the loans, and their experiences of the application process. Anecdotal evidence 
from business representative groups suggests that businesses have found the loans 
useful to address cashflow shortages during the pandemic but there were delays in the 
administration. In March 2021, the Bank surveyed 1,700 borrowers, finding that about 
70% used the funds for working capital and day-to-day expenses, and about 30% 
took out the loans for ‘financial security’. Around 2,000 businesses complained to the 
Financial Ombudsman in August 2021, with more than one third of complaints being 
about a delayed administration of their loan application (paragraphs 3.6 and 3.7).

The Bounce Back Loan Scheme: an update  Summary  9 
13	
Most businesses have started to repay loans, but as of March 2021, the 
Department estimated that 37% of loans worth £17 billion will not be repaid, 
although this estimate is highly uncertain. The Department estimated in March 2021 
that between 31% and 48% of loans will not be repaid, with its ‘most likely’ estimate 
of 37%. This represents both credit losses – among borrowers who want to repay 
but are unable to – and fraud losses – among those who either were not eligible for 
a loan or took out a loan with no intention of paying it back. Credit losses are the 
greatest component. The estimate is highly uncertain as it is based on modelling 
assumptions with limitations: there is no credit score data for the borrowers as this 
was not a scheme requirement; not much repayment data are yet available; and 
repayment rates will be affected by future macroeconomic conditions, which are 
themselves uncertain. As of 30 September 2021, Bank data shows that borrowers 
have repaid £2 billion worth of loans and have defaulted on £1.3 billion of loans; 
the Department has paid lenders £19 million in guarantee claims. These data, 
however, are used to manage the guarantees rather than monitor loan performance 
and are not real-time, but are the best information available to us. The same data 
show that 73% of the 1.5 million loans have either made repayments as planned 
or not yet reached the date to start repayments, and around 21% of loans have 
taken‑up a ‘Pay As You Grow’ option (paragraphs 1.11 to 1.13, and Figure 7).
14	
The Department estimates, as of 31 March 2021, that 11% of loans, worth 
£4.9 billion, are fraudulent but this estimate is highly uncertain. The Department 
recognised before launching the Scheme that its design, with limited verification 
and no credit checks performed by lenders, made it vulnerable to losses, including 
fraud. As the Scheme progressed and the immediate need for finance lessened, 
the Scheme continued to rely on businesses self-certifying their application 
details. In October 2020, the Department stated that the Scheme was “currently 
the leading public sector fraud risk”. The Department in its audited annual report 
estimated that the level of Scheme fraud was 11% of loans, equal to £4.9 billion 
at 31 March 2021 – based on a sample of 1,067 loans. The estimate assumed that 
any fraud leads to a total loss of the loan, which is likely to overestimate losses as 
some funds should be recoverable. The estimate excludes some types of possible 
fraud – for example, where an applicant overstates its turnover and gets a larger 
loan either deliberately or inadvertently. In October 2021, the Department’s advisers, 
PricewaterhouseCoopers LLP (PwC), revised down this estimate to 7.5% of loans 
by looking more closely at the evidence on those loans previously classified as 
‘possible’ fraud including repayment data only available from May 2021; we provide 
no assurance on this revised estimate as it only became available to us in late 
October 2021. Where borrowers have made no attempt to repay, the Department has 
started to review the loans for fraud risk indicators to help it focus its recovery efforts 
on the loans at highest risk of fraud (paragraphs 2.5 to 2.11 and 2.25 to 2.26).

10  Summary  The Bounce Back Loan Scheme: an update
Scheme counter-fraud measures
15	
At the start of the Scheme the priority was to deliver loans as quickly as 
possible, and the Department’s strategy was for the lenders to tackle fraud. 
The Department is responsible and accountable for counter-fraud activity, and the 
Bank manages most of the fraud risks on its behalf. At Scheme launch, in May 2020, 
the Department recognised that prioritising payment speed by allowing potential 
borrowers to self-certify application details made the Scheme vulnerable to fraud. 
The Bank required lenders to make counter-fraud and ‘know your customer’ 
checks, and to make loans within 24 hours, or in some cases 48 hours if additional 
counter‑fraud checks were required. These lender checks of applications 
were the only counter-fraud activities in place when the Scheme launched. 
In December 2020, the Department advised HM Treasury that there was still a 
high level of residual fraud risk even after lenders had conducted these up‑front 
checks (paragraphs 1.4, 2.5, 2.22 and 2.29).
16	
As the Scheme progressed government departments introduced 13 additional 
counter-fraud measures, but most came too late to prevent fraud and were focused 
instead on detection. It is better to prevent fraud than to seek to recover funds later. 
However, the Bank did not introduce straightforward measures to identify duplicate 
applications until June 2020 after 61% of the loans by value had already been 
made. This meant most of the additional counter-fraud activity was focused on the 
detection of fraud that has already taken place, for example creating a fraud hotline. 
The Department stated that introducing more counter-fraud checks at the start 
of the Scheme would have slowed the delivery of loans to businesses. It also told 
us it did not begin to analyse fraud data until several months after Scheme launch 
because its focus was on getting loans to businesses quickly. Delivering financial 
support quickly to smaller SMEs was the main objective at the outset of the Scheme 
(paragraphs 2.22 to 2.25, and Figure 10).
17	
The Department’s counter-fraud strategy evolved over time but lacked clear 
governance at the outset and sufficient resources. We did not find any documents 
which set out the Department’s long-term ambitions, objectives, or financial 
metrics to measure the impact of counter-fraud activity. In October 2020 – five 
months after Scheme launch – the Department set out in a ministerial submission 
the resources it would need to expand its counter-fraud activity. It requested an 
additional £32 million, or 0.07% of the total value of loans, to cover the costs of 
enhanced counter-fraud operations within the Department. In the same month, 
the Bank created its first fraud risk assessment, to identify in detail the Scheme 
risks. The Bank’s internal audit of fraud, in July 2021, concluded that the way it 
manages fraud was “reasonable” but found issues with oversight and governance 
of the Scheme’s fraud risk reporting processes, and slow progress with establishing 
internal fraud responsibilities. It took some time for the Department and Bank to 
build capacity, initially reallocating internal resources and using secondees and 
external parties: since March 2021 the Bank has a dedicated financial crime team 
and is aiming for six full-time equivalent staff by the end of 2021; the Department 
is increasing its counter-fraud team from two to 10 full-time equivalent staff by 
January 2022 (paragraphs 2.12 to 2.14, 2.19 to 2.21, and Figures 9 and 10).

The Bounce Back Loan Scheme: an update  Summary  11 
18	
Owing to the scale of possible fraud, and limited resources, the Department 
is prioritising the pursuit of organised crime. The Department prioritised its 
fraud response by grouping loans into three tiers and concentrating on the 
highest risk areas. Top-tier loans are those involving organised crime groups with 
sums of more than £100,000; mid-tier – some evidence that borrowers acted 
dishonestly but not on a large scale; and bottom-tier – where individuals might 
have dishonestly received loans. The Department could not tell us how many cases 
were in each tier. It decided not to focus its investigative resource on borrowers 
who overstated turnover by less than 25%, provided there were no other fraud 
risk indicators. It expects lenders to pursue upfront recovery where feasible 
and cost‑effective to do so, and to always seek recovery action where loans are 
not repaid. This prioritisation approach risks diminishing the deterrent effect of 
counter‑fraud activity (paragraphs 2.13 to 2.15 and Figure 8).
19	
The Department uses enforcement agencies to investigate, prosecute and 
recover funds for large-scale fraud cases, but the agencies’ capacity is stretched. 
The Department uses enforcement agencies, such as the National Investigation 
Service (NATIS), to investigate, prosecute, and recover funds for top-tier and the 
‘higher end’ of mid-tier loans. Given the potential fraud levels across all COVID-19 
schemes, however, enforcement agencies are stretched. NATIS, for example, 
received more than 2,100 intelligence reports by October 2021 but only had capacity 
to pursue a maximum of 50 cases per year. The Department set NATIS a target of 
recovering at least £6 million of fraudulent loans from organised crime over three 
years, equal to the amount the Department has invested into NATIS. NATIS’ work up 
to October 2021 has resulted in 43 arrests across 33 investigations and more than 
£3 million of recoveries (paragraphs 2.16 to 2.18).
20	
The Department relies on lenders to investigate ‘mid-tier’ and ‘bottom-tier’ 
fraud, but lenders’ commercial incentives to do so are limited. If a lender follows 
the Scheme rules, including regulatory obligations, and makes a loan which 
is subsequently identified as fraudulent, it can reclaim the funds through the 
government guarantee. This offers limited commercial incentive for the lender to 
go beyond the Scheme rules in seeking a full recovery of overdue loans, fraudulent 
or otherwise. HM Treasury has asked lenders to prioritise investigating those 
applications which appear deliberately intended to mislead the Scheme over those 
who overstated turnover by less than 25%. As of April 2021, lenders claim to 
have prevented £1.97 billion of fraudulent applications and detected £5.3 million 
of fraudulent loans. We can provide no assurance over these figures. In July 2021, 
the Financial Conduct Authority, the main regulator for lending, reminded Scheme 
lenders of their wider obligations to report fraudulent activity, and highlighted 
the importance of keeping resourcing levels under review. The Department 
does not know how much lenders are spending on counter-fraud activity 
(paragraphs 2.13, 2.15 and 2.27 to 2.31).

12  Summary  The Bounce Back Loan Scheme: an update
Scheme changes, administrative costs and evaluation
21	
The new repayment options will benefit both borrowers and the taxpayer, 
but increase the overall costs to borrowers and lenders. Around 21% of eligible 
borrowers have taken advantage of at least one of the new measures to delay or 
reduce loan repayments as at 30 September 2021. The two most popular options 
were to extend the term of their loan from six to 10 years (14% of borrowers), and 
taking a repayment holiday (9%). These measures offer borrowers more choice 
on how to repay their loans and increase the chances of business survival, which 
benefits the taxpayer by reducing the potential Scheme losses. However, it creates 
extra ongoing costs – for borrowers who are repaying loans over a longer period and 
thus accruing more interest, and for lenders who must administer the loans over a 
longer period and who already make a lower profit on each loan compared with their 
standard loans. The Department did not quantify these costs and benefits when 
it introduced the new repayment options. It extended the timetable for closing the 
Scheme under the same Ministerial Direction issued when the Department launched 
the Scheme in May 2020 (paragraphs 1.6 to 1.10, and Figures 6 and 7).
22	
The process for recovering overdue loans means borrowers with different 
lenders may be treated differently. Lenders are responsible for collecting capital and 
interest payments and recovering any overdue loan repayments. In doing so, lenders 
must follow the Scheme’s detailed recovery principles, set in December 2020, and 
industry regulations. Each lender is also able to follow their ‘business as usual’ 
approach where it does not contravene the Scheme principles. For example, a 
lenders’ usual approach may include using debt collection agencies, which the 
Scheme also allows if it does not include doorstep visits. A borrowers’ experience on 
loan default will depend, therefore, on their lender’s standard approach, which differs 
across the industry. Most lenders we surveyed said that these Scheme principles are 
clear and that their internal business-as-usual recovery policies were appropriate. 
The Bank audits the lenders for compliance with Scheme principles and can claw 
back any guarantee payments where errors are found, for example where the lender 
has not followed the correct recovery protocols before claiming on the guarantee. 
The Bank has engaged RSM and KPMG to conduct these audits on its behalf 
(paragraphs 3.2 to 3.5 and Figure 11).
23	
The Bank’s Scheme administration costs for 2020-21 are expected to double 
because of the extension to the Scheme, additional repayment options, and the 
increased cost of lender audits. When we reported in October 2020, the Bank 
forecast the Scheme administration costs at £32 million by 2024-25; the Bank 
estimated in September 2021 that this will double to £67 million. The Bank said that 
the flexible repayment options have added extra costs, including more extensive 
lender audit work and additional counter-fraud analysis. The Bank expects the cost 
of lender audits and external operational support across all three COVID-19 debt 
schemes to rise to £75 million by the end of 2024-25; compared with £55 million in 
its October 2020 forecast. HM Treasury agreed to cover “all direct and ‘reasonable 
and justifiable’ indirect operational expenditure” in relation to the Scheme 
(paragraphs 3.11, 3.12, Figures 13 and 14).

The Bounce Back Loan Scheme: an update  Summary  13 
24	
The Department and the Bank have developed some metrics for measuring 
the Scheme’s impact. Our previous report found that the Scheme did not have 
a business case and lacked, at its launch, clear objectives beyond the aim of 
fast financial support for smaller SMEs. When we reported on the Scheme in 
October 2020, the Bank told us that it was in the process of developing metrics 
for measuring the performance of the loan support schemes. The Bank has since 
commissioned an external evaluation study, which will report in stages over the 
next three years. This has some metrics for measuring Scheme impact, including 
business survival rates. It will be challenging for the Bank to measure survival rates, 
as there is a lack of reliable data to compare the recovery of businesses that used 
the Scheme with a similar group that did not (paragraphs 3.13 and 3.14).
Concluding remarks
25	
To achieve the policy intention of supporting small businesses quickly 
during the pandemic, the government prioritised payment speed over almost 
all other aspects of value for money. The Scheme facilitated faster lending by 
removing credit and affordability checks and allowing businesses to self‑certify 
their application documents. As the Scheme progressed, it continued to rely 
on businesses self‑certifying their application details, even as the urgent 
need for finance reduced. Government ruled out options for additional upfront 
counter‑fraud measures when the Scheme was extended. The impact of prioritising 
speed is apparent in the high levels of estimated fraud. Counter‑fraud activity was 
implemented too slowly to prevent fraud effectively and the Department’s focus is 
now on detection and recovery of fraudulent loans.
26	
The Department needs to improve upon its identification, quantification, and 
recovery of fraudulent loans within the Scheme. Compared with the scale of its 
‘most likely’ estimate of £4.9 billion of fraudulent loans, both the £32 million additional 
budget for counter-fraud operations, and its target to recover at least £6 million of 
fraudulent loans from organised crime, are inadequate. The Department has given low 
priority to tackling ‘bottom-tier’ fraud, including those loans where borrowers misstated 
turnover by less than 25%, owing to resource constraints. It expects lenders to focus 
on this fraud tier, but they have limited commercial incentives to do so.

14  Summary  The Bounce Back Loan Scheme: an update
Recommendations
27	
In managing the Bounce Back Loan Scheme, by April 2022 the 
Department should:
a	
produce a formal strategy that sets out the longer-term ambitions, objectives 
and metrics for the impact of successful counter-fraud activity, and brings 
together its existing fraud risk assessment and counter-fraud actions; fraud risk 
appetite; and prioritisation of counter-fraud activities and resources based on 
evidence of cost-effectiveness – updating it as required to take account of any 
new fraud risks;
b	
develop a robust business case for detecting, preventing, and recovering 
fraudulent loans, including the economic rationale of its choices for 
counter‑fraud activity; and the scale of resources needed for each activity;
c	
publish the level it is aiming to reduce fraud losses to, and report against this 
metric. This level should be based on the Department’s expectation of the 
intended impact of its counter-fraud controls for the detection, prevention 
and recovery of fraudulent loans over time;
d	
refresh its fraud risk assessment at least every six months using the best 
available evidence, including with input from the Government Counter 
Fraud Function;
e	
evaluate options for controls against any new fraud risks on a cost-benefit 
basis, using this evidence to introduce controls within two months of identifying 
any new fraud risks; and
f	
set-out key performance measures for each fraud control, and measure 
performance against them regularly, adapting the approach where necessary.
In making changes to business support schemes the Department should:
g	
revisit the scheme’s business case, paying particular attention to the impact 
of changes on fraud and value for money. Where the Department accepts an 
increased risk of fraud and error as a trade-off with other policy objectives, the 
Department must lay this out explicitly. The requirement to consider trade-offs 
applies equally to new schemes.

The Bounce Back Loan Scheme: an update  Part One  15 
Part One
Loan disbursements and repayments
1.1	
This part provides a factual update on the loans issued and changes to 
the Bounce Back Loan Scheme (the Scheme) to further help businesses repay; 
repayment data so far; and the estimated loss due to credit and fraud. It does 
not evaluate the Scheme’s performance as the full repayment of loans will not be 
known for some time.
Background to the Bounce Back Loan Scheme
1.2	 The Department for Business, Energy & Industrial Strategy (the Department) 
launched the Scheme on 4 May 2020, offering loans of up to £50,000, or a 
maximum of 25% of annual turnover, to support businesses during the COVID-19 
pandemic. The loans were provided by commercial lenders (for example, banks, 
building societies and peer-to-peer lenders) directly to businesses who are expected 
to repay the debt in full. Government provided lenders a 100% guarantee against 
the loans (both capital and interest). This means if the borrower does not repay the 
loan, government will step in and pay the lender. The loans have a fixed interest rate 
of 2.5% and a maximum length of 10 years. In the first year of the loan, there are 
no capital repayments due, and the government pays the interest – making it 
interest‑free for the borrower.3 Roles and responsibilities for the Scheme are 
discussed in Appendix Two.
3	
Comptroller and Auditor General, Investigation into the Bounce Back Loan Scheme, Session 2019–2021, HC 860, 
National Audit Office, October 2020, paragraphs 2 and 4.

16  Part One  The Bounce Back Loan Scheme: an update
Loan disbursements
Number and value of loans
1.3	 Around one quarter of all UK businesses received a Bounce Back Loan 
(1.5 million loans out of a total of 6 million UK businesses).4 The Scheme is the 
largest of government’s three business loan support schemes, agreeing 1.5 million 
loans worth £46.8 billion at the time it closed in March 2021. When the Scheme 
launched in May 2020 the Department expected to support 800,000 to 1.2 million 
businesses with loans totalling between £18 billion and £26 billion. The average 
(mean) loan is £30,340, with nearly 35% of loans being at the Scheme’s £50,000 
maximum, as of 13 September 2021 (Figure 1).
1.4	
The Scheme focused on getting money to borrowers as quickly as possible – within 
24 hours, or in some cases 48 hours if additional counter-fraud checks were required. 
The Scheme disbursed more than 40% of the money in the first month of its operation 
(May 2020), with the rate slowing thereafter (Figure 2). It made more than 90% of loans 
by value by the end of November 2020 – the Scheme’s original closing month.
4	
Six million UK businesses is as of October 2020. Available at: www.gov.uk/government/statistics/business-
population-estimates-2020.
Figure 1
Value of loans provided by the Bounce Back Loan Scheme, grouped
by loan size, September 2021
Value of loans approved (£bn)
Most loans (58%) are at the maximum value of £50,000
Notes
1 
Loans with a value of £50,000 represent £27.2 billion out of £46.8 billion which is 58.2%.
2 
Figures as of 13 September 2021. 
Source: National Audit Office analysis of British Business Bank data
0.3
1.7
5.4
5.0
3.8
3.3
27.2
0
5
10
15
20
25
30
Less than or
equal to £5,000
£5,001 to
£10,000
£10,001 to
£20,000
£20,001 to
£30,000
£30,001 to
£40,000
£40,001 to
£49,999
£50,000
Loan size

The Bounce Back Loan Scheme: an update  Part One  17 
Figure 2
Value of loans provided per month since the launch of the Bounce Back Loan Scheme (the Scheme), September 2021 
Value of Bounce Back Loans drawn (£bn)
Cumulative value of Bounce Back Loans drawn 
as a proportion of total value (%)
 
Value of Bounce Back Loans drawn (£bn) 
19.99 
10.19 
3.97 
2.55 
2.23 
2.15 
1.83 
0.69 
0.85 
0.77 
1.21 
0.23 
0.08 
0.01
 
Cumulative value of Bounce Back Loans
 
drawn as a proportion of total value (%) 
42.8 
64.6 
73.1 
78.5 
83.3 
87.9 
91.8 
93.3 
95.1 
96.7 
99.3 
99.8 
100.0 100.0
Notes
1 
The Scheme was launched on 4 May 2020 and closed for new applications on 31 March 2021. Lenders could process outstanding applications after the Scheme closed in March 2021. 
Borrowers continued to draw loans until September 2021 at a very low level, with a total of 55 loans made between July and September 2021. We have excluded these numbers from 
July 2021 to September 2021 for presentational purposes.
2 
Our last report, ‘Investigation into the Bounce Back Loan Scheme’, included data up to 7 September 2020, when £36.9 billion of loans were drawn (79%).
3 
Figures as of 13 September 2021. 
Source: National Audit Office analysis of British Business Bank data 
May
Jun
Jul
Aug
2020
2021
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
43% of loans by value were provided to businesses in the first month of the Scheme and more than 75% in the first four
0
10
20
30
40
50
60
70
80
90
100
0
2
4
6
8
10
12
14
16
18
20
22
Month drawn

18  Part One  The Bounce Back Loan Scheme: an update
Distribution of loans
1.5	 The loan recipients, and their location, as at 13 September 2021, are similar in 
make-up to those when we last reported:
•	
Business size: more than 90% of loans by number went to micro-businesses 
(turnover below £632,000) – a total of 1.4 million loans worth £39.7 billion 
(Figure 3). This is in-line with the Scheme’s aim of providing support to smaller 
SMEs (small and medium-sized enterprises).
•	
Business legal form: Almost all loans (94% by number) went to private limited 
companies and sole traders – 1.4 million loans worth £43.7 billion.
•	
Geographic distribution: Broadly in line with the overall distribution of businesses 
across the UK, with the highest proportion found in London (Figure 4).
•	
Sector: Wholesale and retail sector was the largest user of loans by value 
at 17.4% of the total (Figure 5 on pages 20 and 21).
Figure 3
Total value and number of loans provided by the Bounce Back Loan Scheme 
(the Scheme), by business size, September 2021
Micro-businesses received the highest number of loans (1,393,506) at a value of £39.7 billion
Business size1
Total value of loans provided 
Total number of loans provided
(£m)
Micro
39,654
1,393,506
Small
6,805
140,572
Medium
138
3,235
Mid-sized
148
3,427
Large
7
176
Notes
1 
Business sizes are defi ned by turnover: micro-businesses have a turnover below £632,000; small businesses have 
a turnover between £632,000 and £10.1 million; medium businesses are between £10.2 million and £24.9 million; 
mid-sized businesses have a turnover between £25 million and £500 million; and large business have a turnover 
greater than £500 million.
2 
The Scheme was launched on 4 May 2020.
3 
Figures as of 13 September 2021.
Source: National Audit Offi ce analysis of British Business Bank data

The Bounce Back Loan Scheme: an update  Part One  19 
Figure 4
A comparison of the share of loans provided by the Bounce Back Loan 
Scheme (the Scheme) against the small- and medium-sized enterprise (SME) 
population, by UK region, September 2021
The geographic distribution of loans is broadly in line with the overall distribution of businesses
across the UK
Notes
1 
The percentage of Scheme lending does not sum to 100% because the regional data for some loans were not valid.
2 
An SME is defined as any business with fewer than 250 employees.
3 
The value of loans per region data are at 13 September 2021. The business population estimates per region are at 2021. 
Source: National Audit Office analysis of British Business Bank loan data and the Department for Business, Energy
& Industrial Strategy business population estimates
Percentage of Scheme lending
Percentage of SME population
Region
0
5
10
15
20
25
20.6
18.7
London
13.9
15.7
South East
10.5
9.4
North West
9.6
10.2
East of England
8.1
9.3
South West
8.0
8.0
West Midlands
7.0
7.4
Yorkshire and The Humber
6.3
6.6
East Midlands
6.1
6.1
Scotland
3.8
3.7
Wales
3.1
2.8
North East
2.7
2.2
Northern Ireland
Percentage (%)

20  Part One  The Bounce Back Loan Scheme: an update
Figure 5
A comparison of borrowing market share between the Bounce Back Loan Scheme (the Scheme) 
and small- and medium-sized enterprise (SME) pre-COVID-19 borrowing levels, by industry sector, 
September 2021, UK
The wholesale and retail trade sector’s share of borrowing is twice as much under the Scheme than prior to the COVID-19 outbreak
Scheme market share (%)
Pre-COVID-19 SME borrowing market share (%)
Industry sector
0
5
10
15
20
25
30
35
40
45
0.4
0.1
Public administration and defence
0.6
0.1
Mining and quarrying
1.8
0.7
Electricity, gas and water supply
2.0
1.7
Education
10.0
3.2
Agriculture, forestry and fishing
7.6
4.0
Health and social work
5.9
6.1
Manufacturing
2.6
6.9
Recreational, personal and
community service activities
6.6
9.2
Accommodation and food service activities
4.1
9.4
Transport, storage and communication
9.1
16.5
Construction
8.9
17.9
Wholesale and retail trade
40.5
24.4
Real estate, professional services and
support activities
Percentage (%)

The Bounce Back Loan Scheme: an update  Part One  21 
Scheme extensions and repayment options
1.6	 The Department, with the Chancellor’s agreement, extended the Scheme’s 
original closing date for loan applications from 4 November 2020 to 31 March 2021 
to support businesses during the ongoing pandemic. It extended the Scheme three 
times, to 30 November 2020, 31 January 2021 and 31 March 2021, owing to many 
smaller businesses reporting continued cashflow issues and finance requirements 
following further national COVID-19 restrictions, which extended beyond the original 
Scheme end date (Figure 6 overleaf). The Department and HM Treasury made these 
decisions in consultation with lenders who, for example, cautioned that a calendar 
year deadline following the first extension could lead to a surge of applications, 
which might cause them operational difficulties.
1.7	
Alongside the Scheme extensions, the Department and HM Treasury 
introduced new options for borrowers to help them deal with the continued impact 
of the pandemic. These options included allowing borrowers to ‘top-up’ the size 
of existing loans to the maximum amount to which they are eligible. Other options 
offered more time, and flexibility, to repay, known as ‘Pay As You Grow’, and enabled 
businesses who have started repaying their loan to request:
a	
an extension to the loan duration from six to 10 years, at the same fixed 
2.5% interest rate;
b	
interest-only payments for six months, available up to three times; or
c	
a repayment holiday for six months, available for use once.5
5	
Borrowers could take a repayment holiday initially only after having made six full or interest-only repayments,
but the Chancellor removed this requirement from 8 February 2021.
Figure 5 continued
A comparison of borrowing market share between the Bounce Back Loan 
Scheme (the Scheme) and small- and medium-sized enterprise (SME) 
pre-COVID-19 borrowing levels, by industry sector, September 2021
Notes
1 
Data as of 13 September 2021. 
2 
Pre-COVID-19 SME borrowing market share is an average of the 12-month period from January 2019. Pre-COVID-19 
borrowing levels are based on Bank of England data which uses the defi nition of an SME as any business with 
annual turnover below £25 million. 
3 
Figures do not sum to 100% as not all industry sectors have been included owing to incomplete information, or a 
lack of lending comparison data.
4 
The Scheme market share includes loans to all business sizes. This includes loans to large and mid-sized businesses 
which represent 0.3% of loans by value and 0.2% of loans by number. 
5 
Industry sectors are based on Standard Industry Classifi cation (SIC) codes adapted to bring them in line with the 
categories used in the Bank of England’s Bankstats. As such, the industry sectors and percentages differ slightly to 
those published by the British Business Bank.
Source: National Audit Offi ce analysis of British Business Bank data

22  Part One  The Bounce Back Loan Scheme: an update
Figure 6
Key dates for Bounce Back Loan Scheme (the Scheme) extensions and repayment options
The Chancellor extended the original timetable for closing the Scheme several times and made some changes 
to rules about managing repayments
Date
Change
Instruction from
Scheme extensions 
24 September 2020
Scheme extended until 30 November 2020
Chancellor of the Exchequer
2 November 2020
Scheme extended until 31 January 2021
17 December 2020
Scheme extended until 31 March 2021
Selected COVID-19 support scheme announcements 
24 September 2020
Flexible repayment system known as ‘Pay As You Grow’ introduced, where 
borrowers could request: 
• an extension to their loan term from six to 10 years; 
• interest-only payments for six months; or 
• a repayment holiday for six months. 
Chancellor of the Exchequer
2 November 2020
Option for borrowers to ‘top-up’ their loans introduced.
9 November 2020 
The British Business Bank wrote to the Department for Business, Energy & 
Industrial Strategy (the Department) about a significant increase in exposure 
from the Scheme extensions and ‘top-up’ options. 
It also reminded the Department of its reservations raised on 2 May 2020 
about the pace of launching the Scheme hindering its ability to put in place 
robust controls and governance.
The British Business Bank
6 February 2021
Amendment of ‘Pay As You Grow’ allowing borrowers to take a six month 
payment holiday immediately if they so choose, rather than only after having 
made six full or interest-only payments.
3 March 2021
Recovery Loan Scheme (RLS) launched initially until 31 December 2021 and 
then the Chancellor announced its extension until 30 June 2022 within the 
Autumn Budget 2021. 
Chancellor of the Exchequer 
National COVID-19 restrictions
5 November 2020
Second national lockdown until 2 December 2020.
Prime Minister 
6 January 2021 
Third national lockdown until at least 8 March 2021. 
Source: National Audit Offi ce review of published documentation and British Business Bank documentation

The Bounce Back Loan Scheme: an update  Part One  23 
Impact of the extensions and repayment options
1.8	 The Department’s analysis showed that the Scheme extensions and new 
repayment options would increase the chance of business survival. They also 
have the benefit of increasing the chances that the loans will be repaid, reducing 
the taxpayers’ credit risk. At the same time, it increases the costs to borrowers 
and lenders. Borrowers will be paying back their loans over a longer period and 
accruing more interest as a result, albeit at a lower rate than comparable commercial 
borrowing. Lenders will need to administer the loans for longer, and their capital is 
tied-up in the loans earning relatively less interest than if it were deployed elsewhere. 
In addition, cheaper credit offered through the Bank of England’s Term Funding 
Scheme (TFS),6 available to regulated banks, was initially only available for six 
years, meaning lenders might have been reluctant to meet higher funding costs for 
longer‑term loans. Access to the Bank of England’s TFS was extended to 10 years 
to match the longer loan terms under ‘Pay As You Grow’. The Department did not 
quantify these costs and benefits.
1.9	 Both the extended timetable and new repayment options were covered by 
the same Ministerial Direction issued when the Department launched the Scheme 
in May 2020. The British Business Bank (the Bank) wrote to the Department in 
November 2020 reminding it of the concerns it raised in its Reservation Notice 
ahead of Scheme launch and highlighting a significant increase in risk exposure 
resulting from the Scheme extensions and ‘top-up’ options. Both the Bank and the 
Department noted that they could not make a robust estimation of the value for 
money for extending the Scheme. Nevertheless, the Department’s Accounting Officer 
concluded that a new Ministerial Direction was not necessary. The Department based 
this decision on the grounds that the Scheme parameters and associated risks had 
not changed materially. The Department considered the extension and changes 
affordable as HM Treasury agreed to provide budgetary cover for the Department’s 
‘central estimate’ of losses.
6	
The Bank of England’s Term Funding Scheme (TFS) is only open to banks and building societies that are participants 
of the Bank of England’s Sterling Monetary Framework and signed up to the Discount Window Facility. Not all lenders 
are able to take advantage of TFS.

24  Part One  The Bounce Back Loan Scheme: an update
Take-up of the new repayment options
1.10	 Take-up among borrowers of the offer to increase their loan or change 
repayment terms, as of 30 September 2021, was relatively low:
•	
‘Top-up’ loans: Of the 45% of borrowers who were eligible for a ‘top-up’ loan, 
7% of borrowers chose to increase their loan:
•	
the total amount of ‘top-up’ loans amount to £934 million and the average 
additional borrowing was £8,909.
•	
‘Pay As You Grow’ options: Borrowers on around 21% of loans are using one 
or more ‘Pay As You Grow’ repayment options (Figure 7):
•	
14% of loans have taken advantage to extend the duration of their loans;
•	
9% of loans took a repayment holiday; and
•	
3% of loans moved to interest-only payments.7
Loan repayments and potential losses
1.11	 The Bank has a reporting system (the Portal) which collects from lenders 
loan‑level data to administer guarantees in the event of borrower default. The data 
are not real-time and are dependent on lenders submitting accurate and timely 
data. Based on Portal data at 30 September, borrowers have repaid £2 billion 
worth of loans and have defaulted on £1.3 billion of loans. The Department has paid 
£19 million to lenders to settle guarantee claims. It took the Bank until July 2021 to 
have in place an automated system for lenders to report on repayment activities. 
Before that date the Bank collected these data using a manual process. The Bank 
audits lender submissions to the Portal and we have not conducted our own audit 
of these data. Based on Portal data, assuming 1,539,788 loans were borrowers that 
have drawn-down funds (Figure 7):
•	
73% of all loans have either made repayments as planned or not yet reached 
the date to start repayments (loan repayments start one year after their 
draw‑down date);
•	
21% of loans have taken advantage of one or more ‘Pay As You Grow’ options 
(see paragraph 1.10); and
•	
7% were in arrears on loans payments.
7	
Borrowers can use these options individually or in combination with each other, so the percentage of borrowers who 
took out each repayment option do not sum to 100%.

The Bounce Back Loan Scheme: an update  Part One  25 
Figure 7
Bounce Back Loan Scheme (the Scheme) repayments, 30 September 2021, UK
At 30 September, borrowers have repaid £2 billion worth of loans and have defaulted on £1.3 billion of loans
Value
Number 
of facilities
Percentage of 
total facilities (all 
Scheme loans or 
repayments due) 
(£bn)
(%)
Scheme repayment data
Made repayments 
as scheduled or not 
started repayments2
Data not available3
Data not available3
73
Repaid the loan in full
2.0
61,809
4
Loans defaulted
1.3
37,670
2
Guarantees settled
0.02
601
<1
Amount of arrears 
(one or more 
payments)4
0.1
106,196
7
Scheme option take-up
‘Pay As You Grow’ 
option (one or more)
11.0
318,456
21
Notes
1 
The total loans as of 30 September used in the table above are 1,539,788.
2 
Once the loans have been drawn down, borrowers have 12 months before they need to start making repayments. 
The category ‘Made repayments as scheduled or not started repayments‘ includes those borrowers who either are 
in the fi rst 12 months and so have not started repayments or have passed the 12 month deadline and have started 
to make repayments on time. It does not include borrowers who are using one of the ‘Pay As You Grow’ options.
3 
We took the 73% fi gure from a Bank management information report which drew on the British Business Bank’s 
(the Bank’s) Portal data at 30 September 2021. We requested a detailed breakdown of the underlying loans but it 
was not available as the Portal data had been updated with more recent fi gures.
4 
The ‘amount of arrears’ category includes the value of arrears rather than the total value for the loans which are 
in arrears. For example, if a borrower had drawn a loan worth £50,000 but was in arrears of £25,000, then the 
data represent the £25,000 rather than the full £50,000 loan value.
5 
The fi gures do not sum to 100% as some loans may be counted in more than one category. For example, 
the Department for Business, Energy & Industrial Strategy could have settled a guarantee on a defaulted loan. 
For full details of limitations in these data see Appendix One.
Source: National Audit Offi ce analysis of British Business Bank data

26  Part One  The Bounce Back Loan Scheme: an update
1.12	 The Department estimated in March 2021 that around 37% of loans, at a value 
of £17.2 billion, will not be repaid but its estimate is highly uncertain. Its loss estimate 
ranged between 31% and 48%, with a ‘central estimate’ of 37%. This represents 
both credit losses – among borrowers who want to repay but are unable to – and 
fraud losses – among those who either were not eligible for a loan or took out a 
loan with no intention of paying it back. Credit losses are the largest component. 
We cover estimates of fraud losses in detail in Part Two.
1.13	 The Department’s estimated loss range from March 2021 is subject to a high 
degree of uncertainty. The estimated probabilities of default on the underlying loans 
are uncertain as they are based on several assumptions, including that:
•	
it is usually possible to estimate the probability of loan repayment by looking at 
the borrower’s credit score. Under the Scheme rules, lenders were not required 
to assess credit scores and therefore the Department has no credit score data 
on the borrowers;
•	
as repayments only began in May 2021, the Department had limited repayment 
data at the time to understand how the loans are performing; and
•	
uncertain macroeconomic conditions will affect repayment rates. A survey by 
the Bank, for example, found that one third of applicants were concerned about 
their ability to repay.
The Department developed these estimates using an expected credit loss model, 
which is detailed further in Appendix One. The Office for Budget Responsibility 
(OBR) suggested in July 2021 that up to 45% of Scheme guarantees may be called, 
an increase from 30% since our last report.8
8	
Office for Budget Responsibility, Fiscal risks report, July 2021, paragraph 2.35. Available at: https://obr.uk/docs/
dlm_uploads/Fiscal_risks_report_July_2021.pdf

The Bounce Back Loan Scheme: an update  Part Two  27 
Part Two
Counter-fraud measures
2.1	
This part covers the Bounce Back Loan Scheme (the Scheme) fraud 
estimates. We also evaluate: the Department for Business, Energy & Industrial 
Strategy’s (the Department’s) strategy for tackling fraud; the Department’s 
counter‑fraud governance and resourcing; and the role of lenders. This section 
is primarily evaluative in nature, as the Department has set out how it will pursue 
Scheme fraud since we reported in October 2020.
2.2	 Departmental accounting officers are responsible for managing their 
organisation’s response to fraud and error. For individual projects and programmes, 
that responsibility is often taken on by a senior responsible officer. In the case of the 
Scheme, the Department is responsible and accountable for counter-fraud activity, 
and is responsible for the relationship with other governmental bodies, such as the 
National Crime Agency (NCA), the National Investigation Service (NATIS) and the 
Insolvency Service. The British Business Bank (the Bank) manages most of the 
remaining fraud risk on its behalf.
Our evaluative approach
2.3	 We consider a cost-effective fraud control environment to be one which 
leads to the lowest level of fraud and error after taking all feasible options; we 
discuss this further in our good practice guide on fraud and error.9 A counter-fraud 
strategy should set out, for a specific programme, the level of risk the programme 
is prepared to tolerate and how fraud will be prevented, anticipated, detected and 
monies recovered. This strategy should also include clear indications of governance 
and accountability structures and the resources required to deliver the strategy. 
In normal circumstances, we would expect the strategy to be in place before the 
programme commences.
9	
National Audit Office, Good practice guidance: Fraud and error, March 2021.

28  Part Two  The Bounce Back Loan Scheme: an update
2.4	 We assessed the counter-fraud activities surrounding the Scheme against good 
practice. Our evaluative framework is outlined in Appendix Four and considers the 
relevant areas for our report, including: strategy and resourcing; governance; and 
implementation, measurement, and evaluation of controls to detect and prevent fraud. 
We consulted the Government Counter Fraud Function, and drew on its Standards.10
Fraud estimates
2.5	 Before implementing a scheme public organisations should agree with 
partners their risk appetite, and document it. This will allow them to benchmark their 
measures of fraud, which must be estimated properly and reported appropriately. 
At Scheme launch, the Department recognised that prioritising payment speed by 
allowing potential borrowers to self-certify application details made the Scheme 
vulnerable to fraud; but it did not set its fraud risk appetite or estimate the 
level. As the Scheme progressed, the Scheme continued to rely on businesses 
self‑certifying their application details, even as the immediate need for finance 
lessened. In October 2020, the Department stated internally that the Scheme was 
“currently the leading public sector fraud risk”. In July 2021, the Bank assessed 
Scheme fraud risk as “very high”.
2.6	 The Department commissioned, through the Bank, PricewaterhouseCoopers 
LLP (PwC) in November 2020 to estimate the rate of occurrence of fraud in the 
Scheme. PwC did this by selecting – following a discussion with the Bank and the 
Department – a representative sample of loans and identifying those which had 
characteristics indicative of fraudulent activity. The Department used an external 
party as its in-house capacity was limited (see paragraph 2.14). PwC’s work classified 
the sample loans as those with ‘probable’,11 ‘possible’12 or ‘no suspected fraud’.13 
The Department told us the total cost of this work was £1 million.
10	 Government Counter Fraud Function, Government Functional Standard – GovS 013: Counter Fraud (publishing.
service.gov.uk). Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/
attachment_data/file/1014385/6.7628_CO_Govt-Functional-Std_GovS013-Counter-Fraud_v4.pdf
11	
Sampled loans were classified as ‘probable fraud’ where fraud risk indicators and/or lender information leads the 
Bank to suspect that, on the balance of probability, fraud or a breach of the Scheme rules has occurred.
12	 Sampled loans were classified as ‘possible fraud’ where fraud risk indicators were identified but lender investigations 
have not concluded, or where the lender has chosen not to perform any additional investigations.
13	 Sampled loans were classified as ‘no suspected fraud’ – where no fraud is suspected based on the information available.

The Bounce Back Loan Scheme: an update  Part Two  29 
2.7	 The Department used the results of the PwC work to estimate the level of fraud 
present in the Scheme at 11.15% or £4.9 billion at 31 March 2021. This is based on 
PwC’s finding that 119 of the 1,067 sampled loans (11.15%) had indicators of fraud 
occurrence.14 The work resulted in 71 out of 1,067 loans (6.65% of the sample) being 
classified as ‘probable fraud’, and 48 out of 1,067 loans (4.50% of the sample) as 
‘possible fraud’. The estimate excludes some types of ‘possible fraud’ – for example, 
where an applicant overstates its turnover and gets a larger loan either deliberately or 
inadvertently. It also only partially covers misuse of funds – where funds are not used 
to support the business (a Scheme rule).
2.8	 The PwC exercise has known limitations (see Appendix One, Figure 15) which 
mean that fraud losses may fall outside the estimated range, and the exercise 
focused on identifying fraud occurrence rather than fraud loss. A loss to the 
taxpayer would occur if a borrower defaults, and government must repay the lender 
under the terms of the guarantee. The Department has assumed 100% loss for all 
fraud occurrence types, which is likely to result in an overestimate of fraud losses 
in the Scheme. In some cases, the lender will be able to recover funds. The exercise 
also looked for indicators of ‘possible’ and ‘probable’ fraud occurrence. The 11.15% 
estimate combines ‘possible’ (~4.50%) and ‘probable’ (~6.65%) fraud and may 
therefore overstate the actual fraud occurrence rate. The combined estimate does 
not include all potential types of fraud, including suspected turnover inflation fraud, 
which may increase the fraud occurrence rate.
2.9	 PwC’s fraud estimate did not draw on loan repayment data because the 
first payments were not made until May 2021. PwC observe that not all fraud 
occurrences had been identified owing to limitations to the information available at 
the time of the sample review. As more data becomes available, we would expect the 
uncertainty around fraud and credit losses estimates to reduce. In October 2021, the 
Department’s advisers, PwC, revised down its estimated level of fraud from 11.15% 
to 7.5% of loans. It did this after gathering additional information from lenders on the 
48 loans within its sample which had previously been classified as in the ‘possible 
fraud’ category, some of which had entered repayment. As a result:
•	
41 loans moved from ‘possible fraud’ to ‘no fraud suspected’ after further 
discussions with lenders;
•	
seven loans moved from ‘possible fraud’ to ‘probable fraud’; and
•	
two loans moved from ‘no fraud suspected’ to ‘probable fraud’.
Overall, 80 out of 1,067 loans were classified as ‘probable fraud’ (7.50% of the 
sample), owing to the reclassification that no loans had ‘possible fraud’; and 987 of 
1,067 loans were classified as ‘no fraud suspected’. We can provide no assurance on 
this estimate as we were not able to audit the underlying loan information in the time 
available before we published this report.
14	 PwC set its sample size (1,067 loans) based on the Bank’s desired statistical confidence level of 95% and 
confidence interval of 3%. This means that, the Department, as a result of PwC’s work can state with 95% 
confidence that the estimated fraud occurrence is between 8.15% and 14.15%, with 11.15% as a central estimate.

30  Part Two  The Bounce Back Loan Scheme: an update
2.10	 The figures in both estimates are consistent with the Government Counter 
Fraud Function’s estimate of Scheme fraud, which it described as “significantly 
above the general estimates of public sector fraud levels of 0.5% to 5%” when 
we reported in October 2020. In addition, the Department’s revised estimate is in 
line with the Government Counter Fraud Function’s April 2021 initial estimate of 
Scheme fraud. This estimated that between 4% to 11% of loans were fraudulent, 
equal to £1.8 billion to £5 billion, although it also had a high level of uncertainty. 
The Department has not, against our good practice guidance, set out the level of 
fraud loss that it is prepared to tolerate, based on its expectation of the intended 
impact of its counter-fraud controls over time.
2.11	 The Department’s revised fraud estimates are similar to that of Her Majesty’s 
Revenue & Customs’ (HMRC’s) Coronavirus Job Retention Scheme (CJRS), 
which also provided financial support to businesses. At 31 March 2021, the 
CJRS total estimated level of fraud and error ranged from 7% to 12%, equal to 
£4.1 billion to £7.3 billion, with its most likely estimate at 8.7%, equal to £5.3 billion.15 
The estimate is also subject to a high degree of uncertainty and the schemes are 
not fully comparable as CJRS incorporated additional controls including a 72-hour 
window to perform validation checks on applicants’ details, an honesty declaration, 
whistleblowing routes, transaction monitoring and bank account checks.16
Counter-fraud strategy and resourcing
2.12	 There are several documents setting out components of a counter-fraud 
strategy, but none of these covers the Department’s long-term ambitions, objectives 
or measures of success. The Bank identified fraud risks in an internal fraud risk 
assessment in October 2020. It also updated its action plan for fraud risks, in a 
‘Post Event Action Plan’, several times between October 2020 and April 2021. 
The Department set out its prioritisation of counter-fraud activity and related 
resources, based on the estimated cost-effectiveness of different approaches, in a 
Ministerial Submission rather than a formal strategy document. These documents do 
not set out the government’s strategic objectives, the full scale of resources needed, 
or any financial metrics for the impact of its counter-fraud activity or individual controls 
or actions. These documents are also not accessible or transparent, as there are 
multiple live versions of the action plan and the Ministerial Submission has varied 
access rights among officials involved in delivering the Scheme.
15	 HM Revenue & Customs, Annual Reports and Accounts, 2020-21, Report by the Comptroller and Auditor General, 
4 November 2021, see Figure 17.
16	 Comptroller and Auditor General, Implementing employment support schemes in response to the COVID-19 
pandemic, Session 2019–2021, HC 862, National Audit Office, October 2020. See paragraph 3.20 and Figure 14. 
Available at: www.nao.org.uk/wp-content/uploads/2020/07/Implementing-employment-support-schemes-in-
response-to-the-COVID-19-pandemic.pdf

The Bounce Back Loan Scheme: an update  Part Two  31 
2.13	 Good practice requires the cost-effectiveness of counter-fraud measures to 
be assessed. The Department prioritised its activity within three groups: top‑tier, 
mid-tier and bottom-tier fraud cases. It defined top-tier fraud cases as those 
involving organised crime groups with sums of more than £100,000; mid-tier fraud 
cases as having some evidence that individuals acted dishonestly but not on a 
large scale; and bottom-tier cases as those where individuals might have received 
loans dishonestly, for example, by misstating their turnover on the application. 
The Department could not tell us how many cases were in each group, and is not 
tracking caseloads based on these groups. The Department prioritised its fraud 
response based on the return on investment from recovering fraudulent loans, 
focusing on organised crime.
2.14	 It has taken some time for the Department and the Bank to build their 
counter‑fraud capacity. At the start of the Scheme, the Department recognised that 
its counter-fraud function – two full-time staff – would require significant organisational 
change to cope with the high levels of estimated Scheme fraud. The Government 
Counter Fraud Function placed a senior civil servant into the Department to aid 
the counter fraud response. By January 2022, the Department plans to increase 
its counter‑fraud team to 10 full-time equivalent (FTE) staff. The Bank reallocated 
resources internally, and also used secondees and external parties until March 2021 
when it set up a dedicated financial crime team, aiming to recruit six FTE staff by 
the end of 2021. In October 2020, the Department proposed options to enhance its 
counter-fraud measures, costing an additional £32 million, or 0.07% of the total value 
of the loans. This cost estimate is not limited to staffing and is regardless of the scale 
of fraud. As with other government departments, the Department has low numbers 
of staff with fraud expertise; the Committee of Public Accounts noted in June 2021, 
for example, that 77% of government counter fraud professionals work in the 
Department for Work & Pensions or HMRC.
2.15	 Organisations should make resourcing decisions with an understanding of 
the cost and impact on fraud and error. Owing to the scale of possible fraud, and 
limited internal resources, the Department is prioritising pursuing organised crime, 
and it has started to investigate mid-tier fraud cases. The Department decided not 
to focus its investigative resource on borrowers who overstated turnover by less 
than 25%, provided there were no other fraud risk indicators. This is despite the 
Bank identifying turnover misstatement as the top risk in its July 2021 fraud risk 
assessment (Figure 8 overleaf). The Department expects lenders to pursue upfront 
recovery where it is feasible and cost-effective to do so, and to always seek recovery 
action where loans are not repaid. HM Treasury told us that it asked lenders to 
prioritise investigating those applications which appear deliberately intended to 
mislead the Scheme over those who overstated turnover by less than 25% because 
it could be a genuine error by the borrower, and it does not necessarily mean the 
loan will not be paid back. This prioritisation approach risks diminishing the deterrent 
effect of counter-fraud activity.

32  Part Two  The Bounce Back Loan Scheme: an update
Figure 8
The British Business Bank’s top 10 fraud risks for the Bounce Back Loan 
Scheme (the Scheme), from highest to lowest risk, July 2021
Turnover misstatement is the top fraud risk
Fraud risk
Risk description
1
Turnover misstatement 
Business knowingly submits an application declaring that its 
turnover was higher in 2019. 
2
Application from an 
inactive company
Existing companies which are inactive apply for a 
Bounce Back Loan.
3
Use of Scheme funds for 
personal use
Business uses a Bounce Back Loan for personal use rather than 
supporting the business.
4
Third-party fraud
A third party impersonates a genuine business to get a 
Bounce Back Loan.
5
Non-UK company
A non-UK based company self-certifies that it is UK based, 
which is against the Scheme rules.
6
Multiple applications from 
sole traders
A sole trader makes multiple applications to the Scheme by 
reporting business activities that are not genuine. 
7
Group companies not applying 
as a single group entity
Group companies applying for multiple Bounce Back Loans as 
different subsidiaries rather than a single group entity.
8
A Limited Company retains 
its loan by seeking a 
voluntary dissolution
A Limited Company has no intention to repay a Bounce Back 
Loan and retain the funds by seeking a voluntary dissolution. 
9
A Limited Company retains its 
loan by failing to file accounts
A Limited Company has no intention to repay a Bounce Back 
Loan and fails to file accounts so that its director is subject to a 
compulsory strike-off application.
10
Business applying for multiple 
COVID-19 support schemes 
that it is not eligible for
Business applies to multiple COVID-19 loan schemes and obtains 
a loan from more than one scheme that it is not eligible for.
Note
1 
There were a total of 36 fraud risks in the British Business Bank’s fraud risk assessment of 27 July 2021.
Source: National Audit Offi ce analysis of the British Business Bank’s fraud risk assessment

The Bounce Back Loan Scheme: an update  Part Two  33 
2.16	 Because of the Department’s limited in-house skills and capacity, it uses external 
agencies to investigate, prosecute and recover funds for top-tier and the ‘higher end’ of 
mid-tier fraud cases. External agencies the Department relies upon include: 
•	
law enforcement organisations, for example, the National Crime Agency (NCA) 
and the National Investigation Service (NATIS) to investigate top tier fraud 
cases;17,18 and 
•	
the Insolvency Service which has a statutory role to investigate and prosecute 
breaches of company legislation on the Department’s behalf, as one of its 
executive agencies. 
In October 2020, the Department noted to its minister that it could not increase 
in‑house activity further, given its own limited resources (see paragraph 2.14) and 
the scale of potential fraud it would have to deal with. In that submission it identified 
additional counter-fraud work it could undertake for example: with lenders; external 
parties; and by increasing its own capacity – which it thought would cost around 
£28 million where figures were available, leading to the request for an additional 
£32 million budget (see also paragraph 2.14). The Department’s submission 
to ministers also highlighted the limited capacity of public law enforcement 
organisations and the Insolvency Service. The Insolvency Service had, for 
example, indicated to the Department that it was not adequately resourced to 
investigate a high volume of fraud cases outside of its core enforcement work.
2.17	 The Department identified a “strong return on investment” for using NATIS. 
It stated, as a benefit that NATIS had recovered £8 for every £1 spent investigating 
another public sector scheme. NATIS indicated that it could scale-up to deploy up 
to 136 FTEs to investigate this Scheme at a cost of around £39 million over three 
years. The Department did not pursue this option. NATIS told us during the final 
stages of our review that the Department agreed funding of £6 million between 
2020-21 and 2021-22, with a provisional £6.5 million for 2022-23, subject to the 
Spending Review settlement.
2.18	 As of 26 October 2021, NATIS had received more than 2,100 intelligence 
reports about alleged fraud, both directly and through a Crimestoppers COVID 
fraud hotline, set up to allow individuals to report fraud during the pandemic. 
The Department has set NATIS a target to recover at least £6 million of fraudulent 
loans from organised crime over three years, which is equal to the amount the 
Department agreed NATIS could spend on its caseload for the Scheme. This meant, 
however, that NATIS was able to investigate a maximum of 50 cases per year 
after triaging a larger number of cases. The Department intended to seek support 
from regional police forces and the NCA to mitigate the need to expand NATIS’s 
caseload. As of October 2021, NATIS’s work had resulted in 43 arrests across 
33 investigations and more than £3 million of recoveries.
17	 The NCA investigates and prosecutes the most serious fraud cases such as those involving serious or 
organised crime.
18	 NATIS is a law enforcement body specialising in financial crime and fraud, mostly at a local authority level. NATIS 
will investigate and prosecute medium to high severity fraud cases, supported by the Crown Prosecution Service.

34  Part Two  The Bounce Back Loan Scheme: an update
Counter-fraud governance
2.19	 Good practice suggests that a clear governance structure should be in place 
to manage fraud risk, with a clearly defined remit and ability to hold other parts of 
the organisation to account for implementing the fraud and error strategy. There are 
three fraud forums which aim to set the strategic direction for counter-fraud work 
and coordinate activity. Figure 9 shows that these forums were established after the 
Scheme launch. The forums meet regularly and have wide membership, alongside 
UK Finance (a banking and finance representative), and law enforcement agencies 
– including NATIS and the NCA. The Department sets the strategic direction, while 
HM Treasury provides counter-fraud activity funding. The Government Counter 
Fraud Function advises on good practice and provides challenge on activity.
2.20	The Bank’s internal audit of fraud in July 2021 gave the Bank an overall 
“reasonable” level assurance in how it managed fraud risks in the Scheme. 
The internal audit found, however, slow progress both with establishing internal roles 
and responsibilities and oversight of the Scheme’s counter-fraud strategy and fraud 
risk reporting processes. While the Bank’s Executive Committee approved the first 
fraud risk assessment and ‘Post Event Action Plan’ in October 2020, its Committee 
did not, for example, require the Bank to update its fraud risk assessment on a 
regular basis. The Bank did not update its fraud risk assessment until July 2021, 
adding seven new fraud risks to make a total of 36 risks. These included, for 
example, risks around borrowers obtaining loans from different COVID-19 loan 
schemes, which is against Scheme rules. The Bank has updated its ‘Post Event 
Action Plan’ on a more regular basis than its fraud risk assessment (paragraph 2.12).
2.21	Both the Bank and Department’s progress with developing counter-fraud 
governance structures was limited until they both increased their fraud expertise 
(see paragraph 2.14), as they focused initially on Scheme set-up. The Bank set out 
a clear line of accountability for escalating risks to the Department, including a new 
Fraud and Financial Crime Risk Forum for the Bank from the end of September 
2021 to highlight risks and options to the Department. The Bank is also completing 
a clear process map for who approves decisions and when, although it has not set 
a completion date for this work.
Controls to detect and prevent fraud
2.22	Counter-fraud controls should prevent and detect known fraud and error risks 
in a timely manner to keep pace with emerging threats and measure their impact 
properly. At the outset of the Scheme, in May 2020, the Department relied on 
lenders’ controls – these obliged lenders to ensure that applicants passed anti-fraud, 
anti-money-laundering and ‘know your customer’ checks. We identified a further 
13 Scheme-specific controls to prevent and detect fraud which the Department, 
the Bank, HM Treasury, Cabinet Office and the Government Counter Fraud 
Function implemented after Scheme launch and up to June 2021 (Figure 10 on 
pages 36 to 38).

The Bounce Back Loan Scheme: an update  Part Two  35 
Figure 9
Counter-fraud governance forums for COVID-19 loan guarantee schemes, UK
The Department for Business, Energy & Industrial Strategy (the Department) leads on three counter-fraud forums, which aim to 
set the strategic direction for COVID-19 loan guarantee scheme counter-fraud work 
Forum
Purpose
Membership
Frequency 
Date of first meeting
Escalation to
COVID-19 
Counter-Fraud 
Forum 
Provide support to 
the Counter-Fraud 
Strategy Board
The British Business Bank 
(the Bank)
The Department
HM Treasury
Government Counter 
Fraud Function
Home Office
National Investigation 
Service (NATIS)
Weekly
10 June 2020
Counter Fraud 
Strategy Board
COVID-19 
Counter-Fraud 
Strategy Board
Set strategic 
direction for 
counter-fraud 
work and 
coordinate activity
The Bank
The Department
HM Treasury
Government Counter 
Fraud Function
Home Office
HM Revenue & Customs
National Crime Agency 
(NCA)
NATIS
Fortnightly
3 November 2020
COVID-19 Fraud 
Ministerial Board
Bank Fraud 
Collaboration 
Working Group
Forum for 
lenders to share 
knowledge and 
best practice 
The Bank
The Department
UK Finance – a trade 
association for the banking 
and finance industry
Government Counter 
Fraud Function
NATIS
Fortnightly
6 May 2020
Counter-Fraud 
Strategy Board
Note
1 
COVID-19 loan schemes include, for example, the Bounce Bank Loan Scheme and the Future Fund. 
Source: National Audit Offi ce analysis of British Business Bank and the Department for Business, Energy & Industrial Strategy documentation

36  Part Two  The Bounce Back Loan Scheme: an update
Figure 10
Actions to detect and prevent fraud since the launch of the Bounce Back Loan Scheme (the Scheme)
Some counter-fraud controls were in place at Scheme launch on 4 May 2020, but 11 of 14 measures took four months or more to introduce by which time more than 
80% of funds were lent
Counter- 
fraud 
actions
Details 
Date started 
Updated
Who 
responsible
Time since 
Scheme 
launch
Value of loans 
when action
taken (bn)1
Percentage 
of total 
loans issued, 
by value1
Fraud prevention
Lender 
checks 
Lenders are required to conduct anti-fraud, 
anti-money laundering and ‘know your customer’ 
checks on loan applications as part of the 
guarantee agreement from the outset – but 
applicants self-certified their details. The British 
Business Bank’s (the Bank) accreditation process 
for lenders and post-accreditation audits gives 
it assurance that lenders have the appropriate 
controls in place to complete these checks.
4 May 2020
Not applicable
British 
Business Bank
0 months 
0
0% 
Change in 
director flag 
Identifies previously registered companies (before 
1 March 2020) that are inactive and have been 
subject to a director change since the Scheme 
was launched. This is a strong indicator of fraud 
and identifies companies that have been acquired 
by organised crime groups with an intent to 
defraud the Scheme.
15 July 2020 
Not applicable 
Government 
Counter Fraud 
Function 
2 months 
32.4
69%
HM Revenue 
& Customs 
(HMRC) 
Bounce 
Back Loan 
Verification 
Scheme
Used to verify a business’s turnover. Owing to the 
tax cycle, the data will lag by one year.
From 1 January 2021 
Not applicable 
Government 
Counter Fraud 
Function; 
HMRC
8 months 
43.6
93%
‘Bulk 
objection’ 
process
A mechanism to prevent companies closing 
(also called dissolving – see paragraph 2.25) to 
avoid repaying Scheme loans. This is intended to 
prevent fraudsters receiving a loan, withdrawing 
funds and then closing the company with the 
intention of preventing lender recovery of funds.
3 March 2021
Not applicable
Government 
Counter Fraud 
Function; 
Companies 
House; The 
Department 
for Business, 
Energy & 
Industrial 
Strategy (the 
Department)
10 months
45.4
97%

The Bounce Back Loan Scheme: an update  Part Two  37 
Figure 10 continued
Actions to detect and prevent fraud since the launch of the Bounce Back Loan Scheme (the Scheme)
Counter- 
fraud 
actions
Details 
Date started 
Updated
Who 
responsible
Time since 
Scheme 
launch
Value of loans 
when action
taken (bn)1
Percentage 
of total 
loans issued, 
by value1
Fraud prevention
Turnover pilot
Work on a sample of 10,000 loans using HMRC 
data to investigate the occurrence of inflation of 
turnover. Government Counter Fraud Function 
analysis has revealed that 35% of customers 
(of those that could be matched) appeared on the 
basis of HMRC data to have inflated their turnover 
greater than or equal to 25% on the basis of 
turnover stated on the Scheme’s loan application. 
Due to the tax cycle, the data will lag by one year. 
June 2021
Not applicable
Government 
Counter Fraud 
Function
13 months
46.7
100%
Fraud detection
Mandatory 
duplicate 
checks 
All lenders use a fraud prevention database 
(provided by CIFAS) to check loan applications 
to identify and prevent multiple application fraud 
(seeking multiple loans from different lenders, 
which is not allowed under the Scheme).
26 June 2020 
Not applicable
British 
Business Bank
1 month 
28.7
61%
NATIS start 
work
The Department commissioned the National 
Investigation Service (NATIS), to investigate, 
disrupt and recover money from fraudulent 
activity in the Scheme.
21 September 2020 
Not applicable 
The 
Department
4 months 
38.3
82%
Fraud risk 
assessment 
Sets out who might defraud the Scheme, how 
they could do it, and its likelihood and impact.
Good fraud risk assessments have specific 
fraud risks, laid out in the ‘Actor, Action, 
Outcome’ formula. 
Government and industry best practice is to have 
such assessments in place before products are 
launched. As the Department set up the Scheme 
quickly, it relied on high-level internal and external 
assessments, to begin with.
8 October 2020
23 August 2021 
British 
Business Bank
5 months 
39.6
85%

38  Part Two  The Bounce Back Loan Scheme: an update
Figure 10 continued
Actions to detect and prevent fraud since the launch of the Bounce Back Loan Scheme (the Scheme)
Counter- 
fraud 
actions
Details 
Date started 
Updated
Who 
responsible
Time since 
Scheme 
launch
Value of loans 
when action
taken (bn)1
Percentage 
of total 
loans issued, 
by value1
Fraud detection
COVID-19 
fraud hotline
Run in conjunction with Crimestoppers, under 
which the public can report fraud in relation to the 
COVID-19 support schemes. 
The Cabinet Office triage intelligence from the 
fraud hotline and then provided to the Department 
or NATIS.
12 October 2020
Not applicable
Cabinet Office
5 months
40.1
86%
Post-Event 
Action Plan
Action plan setting out the activities to be 
undertaken to give the Department confidence in 
the level of fraud and error loss identified.
19 October 2020
Updated six 
times up until 
July 2021
British 
Business Bank
5 months
40.6
87%
PwC fraud 
estimation
The Department commissioned 
PricewaterhouseCoopers LLP (PwC) to review a 
sample of 1,067 cases for possible or probable 
fraud. This level does not include certain types of 
potential fraud, for example, turnover inflation and 
possible lender fraud, and only partially covers 
misuse of funds.
23 November 2020 
June 2021: 
findings 
completed 
British 
Business Bank
6 months 
42.7
91%
Government 
Counter 
Fraud 
Function 
analysis
The Government Counter Fraud Function is 
delivering a Fraud Analytics Programme on behalf 
of HM Treasury, the Department and the Bank 
to identify fraud and guarantee ineligibility in the 
Scheme. The Programme has eight phases which 
will analyse the Scheme’s dataset against other 
datasets.
1 March 2021: 
First report 
July 2021: 
Second report 
Government 
Counter Fraud 
Function
10 months 
45.4
97%
Fraud team 
introduced 
The Bank introduced a Financial Crime Team to 
take the lead on its counter-fraud response for 
the Scheme. 
7 April 2021
Not applicable
British 
Business Bank
11 months 
46.6
100%
Fraud recovery
Fraud 
recoveries 
document for 
lenders
Sets out how government should approach loan 
repayment recoveries once lenders claim on the 
government’s guarantee.
9 December 2020 
Not applicable 
HM Treasury 
7 months 
43.1
92%
Note
1 
The British Business Bank Portal data provided a weekly breakdown of the number and value of loans which were drawn down. The ‘value of loans when action taken’ and the ‘percentage of 
total loans issued by value’ columns were calculated using the value of loans data from the week which included the ‘Date started’.
Source: National Audit Offi ce analysis of the Department for Business, Energy & Industrial Strategy and the British Business Bank documentation, and British Business Bank Portal data

The Bounce Back Loan Scheme: an update  Part Two  39 
2.23	A duplicate application check to identify borrowers seeking multiple 
Scheme loans from different lenders was implemented in June after 61% of 
the loans, by value, had been made. It is better to prevent fraud than to seek 
to recover funds later. However, the 10 further actions implemented in or after 
September 2020 were aimed mainly at detecting fraudulent loans that had already 
taken place, rather than preventing fraudulent applications, to allow lenders to 
take action to demand repayment of these loans in full. When the Scheme was 
extended, government ruled out introducing additional counter-fraud measures 
on the basis that it would undermine the Scheme’s self-assessed nature and its 
objective to deliver funds quickly. This was despite the Bank having identified 
turnover misstatement as the top fraud risk in October 2020.
2.24	Actions to detect fraud included, for example, a fraud risk assessment to 
set out the likelihood of certain groups defrauding the Scheme, but the Bank 
only updated this once in 10 months, in contrast to its ‘Post Event Action Plan’ 
(see paragraph 2.12). Actions to prevent fraud included a pilot to check company 
turnover against existing data, but the Department did not introduce this until 
June 2021, after all the loans were made. Similarly, the Department implemented 
a duplicate application check in June 2020, a month into the Scheme, but by 
this time, most loans (61% of the total value) were made. The Bank only began 
formally testing the effectiveness of this control some 14 months after introduction. 
By this point, there were still some duplicate applications, which suggests that the 
control did not have a deterrent effect. The Department stated that introducing 
these checks from Scheme outset would have slowed loan delivery to businesses 
and delivering financial support quickly to smaller SMEs (small and medium-sized 
enterprises) was the main objective at the outset of the Scheme.
2.25	The Department told us that it did not begin to analyse fraud data until several 
months after Scheme launch because its focus was on getting loans to businesses 
quickly. The Government Counter Fraud Function identified possible Scheme fraud 
by matching Scheme loan data to other data sources. These ranged from business 
registration and tax data to law enforcement data. It completed the first two of eight 
planned pieces of fraud analysis between March and July 2021:
•	
The first phase (March 2021) of the Government Counter Fraud Function’s 
analysis of Scheme risks focused on six fraud risk areas for loans to limited 
companies, which represent the largest share of borrowers. It identified loans 
with a higher likelihood of fraud worth £492.8 million (1.1% of the total loan 
book), of which more than half, at a value of £297 million, were from borrowers 
with filed dormant accounts. This suggested that dormant companies were 
resurrected simply to claim loans.

40  Part Two  The Bounce Back Loan Scheme: an update
•	
The second phase focused on disqualified company directors with a 
Scheme loan; and a ‘dissolution objection process’, which prevents companies 
voluntarily closing (or dissolving) before the Department and lenders have 
had the opportunity to recover loans. Government Counter Fraud Function 
analysis found a total of 119 companies with a disqualified director took 
loans. However, the Department told us that follow-up work revealed these 
results were largely made up of directors who received loans prior to 
the disqualification period or after the disqualification period had ended. 
The Government Counter Fraud Function also found that Companies House 
had lodged 34,232 successful objections to 34,123 companies dissolving, 
totalling £1.2 billion of loans.
2.26	In June 2021, the Department shared with lenders the Government 
Counter Fraud Function findings to help them identify areas of possible fraudulent 
behaviour. This information included, for example, loans with potential fraud risks. 
The Department told us that it plans to take this further by analysing loan repayment 
data for fraud indicators. It plans to match the loans where the borrower has made 
no attempt to repay against the fraud risk areas identified from its analysis, but it 
has not set a date for completing this work. This will help the Department focus its 
counter-fraud resources in the areas of highest risk and increase its chances of 
recovering the loan.
Lenders’ counter-fraud activity
2.27	The nature of the Scheme places the main responsibility for managing mid 
and bottom tier fraud risks on lenders. These rules required lenders to conduct 
counter‑fraud checks before making loans, such as checking borrower details 
against a counter-fraud database and ‘know your customer’ checks. They also pursue 
recovery processes in line with their existing business-as-usual standards, under the 
terms of the Scheme’s guarantee agreement. Lenders were expected to make loan 
payments within 24 hours of application, or in some cases, within 48 hours if further 
counter-fraud checks were required.
2.28	To support lenders in managing fraud-related risks, the Bank established fraud 
prevention forums with the lenders and a wider group of stakeholders to share good 
practice and aid implementation of additional counter-fraud measures (Figure 9). 
In addition, from October 2020, the Bank, alongside the Department and lenders, 
began using a system allowing lenders to provide a monthly fraud report. This provides 
estimates of prevented loss, detected loss, errors and recoveries. Although this is not 
active fraud prevention, given it is conducted once the lenders upload the details into 
the Portal, it is aimed at identifying and minimising potential losses.

The Bounce Back Loan Scheme: an update  Part Two  41 
2.29	As of April 2021, lenders claim to have prevented £1.97 billion of fraudulent 
applications and detected £5.3 million of fraudulent loans. Detected fraud refers 
to loans made which lenders have subsequently judged as fraudulent. We did 
not audit these figures as: each lender reports the information on a different 
basis; some lenders might choose not to report suspected fraud, which makes 
comparisons subjective; and some lenders do not share their underlying data. 
The Bank told us that it does not audit these figures either. In December 2020, 
the Department advised HM Treasury that there was still a high level of residual 
fraud risk even after lenders had conducted their up-front checks. These figures will, 
therefore, not give a full picture of the scale of fraud, or that detected and prevented 
by lenders. In July 2021, the Financial Conduct Authority, the main regulator for 
lending, reminded Scheme lenders of their wider obligations for reporting fraudulent 
activity. It also highlighted that it was important “to keep resourcing levels under 
review to ensure instances of financial crime are identified and reported in a timely 
manner through the appropriate channels, and to assess, monitor and manage 
controls to identify fraud cases effectively in line with existing requirements”.
2.30	If a lender follows the Scheme rules and its internal business-as-usual 
practices when making a loan which is subsequently identified as fraudulent, it can 
reclaim the funds through the government guarantee.19 The Government Counter 
Fraud Function and Home Office raised concerns with the Department, which the 
Department recognised, that lenders were not required, or indeed incentivised, to 
seek full recovery of fraudulent loans. Recovering funds adds to lenders’ costs and 
so commercial incentives to recover could be weak. For example, lenders can claim 
on the guarantee “within a reasonable time period” following the first formal demand 
for repayment and claiming on the guarantee is not conditional on having completed 
the recoveries process. Once the guarantee is paid any funds the lender subsequently 
recovers should be paid back to the government. The Department does not know how 
much lenders are spending on counter-fraud activity. We cover the incentives around 
the recovery process in more detail in our October 2020 report.
2.31	 Lenders continue to collaborate with the Bank by participating in fraud 
prevention forums and voluntarily providing information on their counter-fraud 
activities, which go beyond those set out in Scheme rules. The Bank audits lenders for 
compliance of the agreement rules and checks as part of the guarantee claim process.
19	 ‘Business as usual’ processes should be compliant with regulation on lenders which the Financial Conduct Authority 
and Lending Standards Board Standards of Lending Practice for Business Customers (“Treatment of customers in 
financial difficulty”) set. Available at: www.lendingstandardsboard.org.uk/wp-content/uploads/2020/08/Standards-
of-Lending-Practice-for-business-customers-August-2020-Covid-update.pdf

42  Part Three  The Bounce Back Loan Scheme: an update 
Part Three
Other changes and activities under the Bounce 
Back Loan Scheme (the Scheme)
3.1	
This part provides a factual account of the agreed process for lenders to recover 
outstanding loans, the borrower experience and impact of the Scheme on the small 
and medium-sized enterprises (SMEs) lending market, the Department for Business, 
Energy & Industrial Strategy’s (the Department’s) plans for evaluating the Scheme; 
and the costs of administering the Scheme. It does not evaluate these areas as the 
outcome for businesses, lenders and the taxpayer will not be known for some time.
Recovery process
3.2	 Lenders are responsible for collecting capital and interest payments and 
recovering any overdue loan repayments from borrowers. The Scheme recovery 
principles define how the collection process should operate, including protocols 
for collecting overdue payment. When we reported on the Scheme in October 
2020, HM Treasury had agreed the recovery process principles with lenders, but 
not the details of how the process would operate. In December 2020, it agreed 
the specific operational details with lenders. Each lender is also able to follow their 
‘business as usual’ approach where it does not contravene the Scheme principles. 
Lenders’ ‘business-as-usual’ standards are also covered by market regulations, such 
as the Financial Conduct Authority’s Fair treatment of customers rules and the 
Consumer Credit Sourcebook depending on the size of the loan and business entity.
3.3	 The recovery process gives borrowers who have taken up the ‘Pay As You 
Grow’ option more flexible repayment options before lenders issue a formal demand 
for payment. Lenders must complete the recovery process if formal demands are 
not met but can claim on the guarantee before completing the process “within a 
reasonable time period” following the first formal demand date, or sooner, if lenders 
believe “no further payment is likely”. Any additional funds the lender ultimately 
recovers from the borrower, less reasonable recovery cost, must be returned 
to government. Most lenders that we surveyed said that the recovery process 
principles were sufficiently clear and that their internal business-as-usual recovery 
policies were appropriate. Given that each lender’s recovery process will differ, for

The Bounce Back Loan Scheme: an update  Part Three  43 
example, including when and how debt collection agencies may be used, 
borrowers’ experience on loan default will depend on their lender.20 Recovery 
process steps include:
•	
in the first instance, lenders can offer borrowers one or more ‘Pay As You Grow’ 
option up to three months before repayments are due.21 Lenders, who offer ‘Pay 
As You Grow’ are required to send a reminder to borrowers with outstanding 
payments from three, two and one month before the end of the 12-month 
standard repayment holiday;
•	
if a lender judges that a borrower might be helped beyond any ‘Pay As You 
Grow’ options, it can offer its standard forbearance options, in line with its 
business-as-usual processes, for example, by setting a repayment plan;
•	
where a lender makes a further judgement that neither ‘Pay As You Grow’ 
nor forbearance options will help, or if payments are at least 90 days 
overdue, it can issue a formal demand to borrowers for repayment of the 
full outstanding balance. Any missed payments or forbearance options are 
reported to credit reference agencies, which can impact a borrower’s ability 
to obtain further finance;
•	
a lender can apply its existing processes to pursue a borrower for repayment 
for 12 months after issuing a formal demand, with some additional Scheme 
protection for borrowers; it is not obliged to pursue repayments after this time. 
Lenders cannot, for example, seize a borrower’s primary residence or main 
vehicle, but borrowers can offer these assets at their discretion. Lenders can 
pursue a borrower for repayments by contacting borrowers directly or by opting 
to use debt collection agents, although the Scheme does not allow ‘doorstep 
visits’. The recovery principles state that lenders should assess whether to 
pursue repayments based on the probability of recovering the loan, the amount 
outstanding and a borrower’s potential assets. It can suspend such action 
where no future repayment is likely, or at the end of the recovery process;
•	
borrowers who are traced through the recovery process by lenders can then 
enter into a repayment agreement based on the affordability of the debt. 
Under the Scheme, borrowers can offer, at their own discretion, their personal 
assets, but this is not a requirement; and
•	
any further enforcement action is only used in the event of serious or organised 
fraud, or the borrower refusing to repay despite having assets available 
(Figure 11 on pages 44 and 45).
20	 The experience of a borrower who has defaulted on their Scheme loan may also differ depending on the legal entity 
status of the business, and whether it has also defaulted on other debts.
21	 Under the Scheme, borrowers were offered options to allow more time and flexibility to repay. Collectively these 
options were known as ‘Pay As You Grow’ and enabled businesses who have started repaying their loan to request: 
an extension to the loan duration from six to 10 years, at the same fixed 2.5% interest rate; interest-only payments 
for six months, available up to three times; or a repayment holiday for six months, available for use once.

44  Part Three  The Bounce Back Loan Scheme: an update 
Figure 11
The process for recovering overdue loans in the Bounce Back Loan Scheme as agreed between 
the government and lenders in December 2020
The recovery process gives borrowers more flexible repayment options before lenders issue a formal demand for payment
Note
1 
HM Treasury agreed the recovery process in December 2020. Some loans have not yet reached the later phases of the process, 
which begin 12 months after issuing a formal demand.
Source: National Audit Offi ce analysis of HM Treasury’s recovery process
Action
Definition 
When used
Borrower role
Lender role
Report to 
credit reference 
agencies 
Third 
parties 
Pre/post-arrears
Pay As You 
Grow options 
Offers more time 
and flexibility to 
borrowers: 
• Term extensions 
for 10 years.
• Interest-only 
payments for 
six months.
• Payment holiday 
for six months.
Up to three months 
prior to repayments 
commencing.
Requests a 
‘Pay As You 
Grow’ option by 
self-attesting 
they are facing 
repayment 
difficulties.
Expected, but not 
required, to offer 
‘Pay As You Grow’. 
Opted-in lenders 
send a reminder 
three, two and 
one month before 
the end of the 
12-month standard 
repayment holiday.
Not reported 
solely as a 
result of using 
‘Pay As You 
Grow’ options
Not 
applicable
Arrears and default
Forbearance 
Forbearance options 
in line with lenders’ 
usual processes, 
for example: 
• Payment plan to 
resolve arrears.
• More time to 
seek advice 
on options.
• Waiving interest 
accrued on loan.
Where ‘Pay As You 
Grow’ is not deemed 
sufficient or helpful 
for the borrower, 
for example, 
for vulnerable 
borrowers.
Inform the 
lender they 
are in financial 
difficulty.
Contact borrowers 
who miss 
payments and 
offer the ‘Pay As 
You Grow’ options.
Any forbearance 
options reported
Not 
applicable
Issue formal 
demand 
A formal demand 
to borrowers for 
repayment of the 
full outstanding 
balance. 
Where lender 
judges that ‘Pay 
As You Grow’ or 
forbearance will 
not help.
When payments 
are at least 90 
days overdue or if 
the borrower has, 
permanently ceased 
trading or evidence 
of fraud.
Contact the 
lender to 
discuss an 
affordable 
repayment plan 
if it anticipates 
difficulties in 
full repayment 
immediately.
Issue the demand 
and offer an 
affordable 
repayment plan.
No cross-defaults 
on other loans 
unless there 
is cessation of 
trading, death, 
unwillingness 
to engage.
Missed payments 
reported 
Not 
applicable 

The Bounce Back Loan Scheme: an update  Part Three  45 
Action
Definition 
When used
Borrower role
Lender role
Report to 
credit reference 
agencies 
Third 
parties 
Recovery
Pursue 
borrower for 
repayment
Contact borrower 
for repayment, 
assessing recovery 
probability, including 
legal form of 
borrower, amount 
outstanding and 
potential assets; 
based on existing 
processes.
After formal 
demand issued. 
Suspended 
where no future 
repayment is likely, 
or at the end of the 
recovery process. 
Respond 
to contacts 
from the 
lender.
Continue contact 
with borrower 
12 months after 
issuing the 
demand. 
Scheme intent 
is for no cross-
default, except 
some situations.
Missed 
payments 
reported
Debt 
collection 
agents 
optional
No doorstep 
visits
Repayment 
agreement
Borrowers 
traced through 
the recovery 
process enter 
into a repayment 
agreement based 
on the affordability 
of the debt.
Negotiated during 
the 12 months from 
formal demand. 
No time limit. 
Offer a 
declaration 
for private 
residence 
or vehicle 
against the 
recovery at 
their own 
discretion.
Try to repair 
a break in a 
repayment plan 
at least once.
Missed 
payments 
reported 
Not
applicable
Enforcement
Enforcement action 
includes petitioning 
the borrower 
for insolvency 
or starting 
administration 
proceedings.
Enforcement is not 
expected unless 
in the event of 
serious or organised 
fraud, or borrower 
refusing to pay but 
has assets.
Not 
applicable
Use their 
discretion to take 
enforcement 
action in 
extenuating 
circumstances.
Not applicable 
Not
applicable 
Make a claim 
on the loan 
guarantee
A guarantee can 
be applied to net 
(post-recovery) loss 
of principal only; 
it does not cover 
interest. A lender 
can claim costs for 
dealing with debt 
collection agencies.
No later than 
12 months after 
issuing a formal 
demand to the 
borrower – unless 
the borrower 
repays after the 
demand issued.
Not 
applicable
Use Guarantee 
Agreement for a 
partial write-off 
of the loan, once 
followed all other 
options above. 
Not applicable 
Not
applicable 
Post-guarantee 
claim
Once a Guarantee Claim has been made, 
the Lender is still obliged to seek further 
potential recoveries if the recoveries 
process has not yet been completed and a 
12-month period from the date of issue of 
the formal demand has not passed.
Not 
applicable
Summarise the 
amount due 
to government 
from any further 
recoveries after 
a claim
Not applicable 
Not 
applicable 

46  Part Three  The Bounce Back Loan Scheme: an update 
3.4	 The recovery process includes some additional protection for borrowers, 
although this may vary by lender depending on its standard approach. As loan 
repayments were only due from May 2021, there is currently little complaints data 
from, for example, the Financial Ombudsman surrounding fair treatment.
3.5	 As lenders can claim the government guarantees before pursuing borrowers 
for a full 12 months, there is a risk that the lender would rather claim on the guarantee 
than seek to recover outstanding loans. The terms of the guarantee, however, require 
lenders to take ‘reasonable steps’ to recover overdue payments. The British Business 
Bank (the Bank) audits lenders’ compliance with the guarantee terms. This means 
that if a lender made a claim on the loan guarantee, but afterwards an audit found 
that the claim was made fraudulently, in bad faith, or other than in compliance with 
the guarantee terms, the lender is required to reimburse government. The Bank has 
engaged RSM and KPMG to conduct lender audits on the Bank’s behalf; they began 
in September 2020.22 Audits take a risk-based approach, and review a sample of 
lenders’ Scheme facilities taken from the Bank’s Portal data.
Borrower experience
3.6	 Limited anecdotal evidence we gathered from business representative groups 
suggests that businesses have found the loans useful to address cashflow shortages 
during the pandemic. In March 2021, the Bank surveyed 1,700 borrowers, finding 
that about 70% used the funds for working capital and day-to-day expenses, and 
about 30% took out the loans for ‘financial security’. Neither the Department nor 
the Bank have routinely collected information on loan use.
3.7	 Some businesses, however, experienced delays in the processing of their 
loan applications, with more than a third of the 2,000 complaints to the Financial 
Ombudsman in August 2021 relating to a delayed administration of their loan 
application. The Federation of Small Businesses (FSB) told us that setting up a loan 
was more challenging for borrowers who ran their business through their personal 
accounts. The FSB also found that some of its members had difficulties obtaining 
a loan from their own bank and had to apply to other lenders, which lengthened 
the time it took to receive a loan.
22	 A third auditor (BDO) is also contracted to undertake one-off audits where, for example, both RSM and KPMG 
are conflicted.

The Bounce Back Loan Scheme: an update  Part Three  47 
Scheme lenders
3.8	 When the Scheme launched in May 2020, there were seven accredited lenders, 
consisting of five main UK banks and two other banks. In March 2021, when the 
Scheme closed, there were a total of 24 accredited lenders (as of 13 September 2021), 
six of which were non-bank lenders – one lower from October 2020.23 Industry 
representatives for non-banks reported that many alternative lenders faced delays 
in the Bank accrediting them to make loans under the Scheme. Large banks 
usually already had a pre-existing relationship with the Bank so had been through 
its accreditation checks. In contrast, alternative lenders could take several months 
to be accredited as the Bank conducted additional checks.
3.9	 Most of the loans (nearly 90% of the total value of loans) were provided by 
the seven main UK banks. These larger banks had traditionally made relatively few 
loans to micro-businesses, which means that they took market share from smaller 
finance providers. The remaining 10% of loans were provided by other UK banks 
and non‑bank lenders, such as peer-to-peer lenders (1% of loans by value or 
4,000 loans), respectively as of 13 September 2021 (Figure 12 overleaf).
3.10	 Many non-bank lenders found it difficult to obtain cheap credit to finance 
loans, a challenge that the Bank, HM Treasury and industry bodies for alternative 
lenders recognised. While the government offered 100% guarantees to the loans, 
it did not provide finance directly, and lenders used their own capital to make loans. 
Unlike established banks, the non-bank lenders did not have access to cheaper 
credit offered by the Bank of England through the Term Funding Scheme (TFS).24 
Analysis by Innovate Finance, a representative body for non-bank lenders, showed 
that in 2020 the main banks increased their market share from 35% to 56%. 
Alternative lenders’ total lending remained constant year-on-year.25
23	 Non-banks are institutions that are not registered with the Prudential Regulation Authority.
24	 HM Treasury offered non-bank lenders the ability to assign the guarantee under the loans as an option to benefit 
from cheaper funding.
25	 Innovate Finance, Innovate Finance releases new analysis of SME market share, 1 September 2021. Available at: 
www.innovatefinance.com/news/innovate-finance-releases-new-analysis-of-sme-market-share-revealing-split-
between-big-banks-vs-new-lenders-before-and-after-the-covid-crisis/

48  Part Three  The Bounce Back Loan Scheme: an update 
Scheme administrative costs
3.11	 The Bank’s Scheme administration costs were £17 million for 
2020‑21 against a forecast of £9 million when we reported on the Scheme 
in October 2020. In September 2021 the Bank forecast administration costs 
to be £67 million between 2020-21 and 2024-25. This was an increase to its 
October 2020 forecast of £32 million (Figure 13). This latest estimate represents 
53% of the overall administrative costs of the three COVID-19 business loan 
support schemes, compared with 43% in October 2020. These estimates do not 
include the Department and HM Treasury’s costs. HM Treasury agreed to cover 
“all direct and ‘reasonable and justifiable’ indirect operational expenditure” in 
relation to the Scheme.
Figure 12
Value of loans provided by the Bounce Back Loan Scheme by lender type, 
September 2021
Notes
1 
Figures as of 13 September 2021.
2 
The British Business Bank does not base its categories of bank on any specific criteria such as size or product category. 
3 
Main Banks are: Bank of Scotland, Barclays, Clydesdale Bank, HSBC, Lloyds, NatWest and Santander.
4 
Banks are: AIB Group, Arbuthnot Latham & Co, Bank of Ireland, Close Brothers, Co-operative Bank, Danske Bank, 
Investec Bank, Metro Bank, Paragon Bank, Starling Bank and TSB Bank.
5 
‘Other’ includes non-banks which are institutions not registered with the Prudential Regulation Authority. These 
include: Funding Circle, New Wave Capital, Tide Capital, Conister Finance & Leasing, Skipton Business Finance and 
GC Business Finance (previously BFS).
Source: National Audit Office analysis of British Business Bank data
Nearly 90% of the total value of all Bounce Back Loans were provided by the seven main UK banks
Main Banks
89.7%
Banks
10.0%
Other
0.3%

The Bounce Back Loan Scheme: an update  Part Three  49 
3.12	 The Bank said that the Scheme extensions and flexible repayment options 
have added extra costs, including more extensive lender audit work and additional 
counter-fraud analysis. Most of the costs for the next five years are for external 
operational support, including costs for IT and legal and professional advice. 
However, 11% of the costs are for consultancy to carry out the lender audits 
(Figure 14 overleaf). The Bank expects the cost of external operational support and 
lender audits across all three COVID-19 debt schemes to rise to £75 million by the 
end of 2024‑25, compared with £55 million in its October 2020 forecast.
Figure 13
Administrative costs of the Bounce Back Loan Scheme (the Scheme), 
2020-21 to 2024-25
Costs (£m)
The British Business Bank's (the Bank's) revised forecast administrative costs for the Scheme have 
increased since we reported on the Scheme in October 2020, because of the flexible repayment 
options and Scheme extensions
Notes
1 
Figures as of 13 September 2021.
2 
Total administrative costs for 2020-21 to 2024-25 at October 2020 were £32 million.
3 
Total administrative costs for 2020-21 to 2024-25 at September 2021 were £66.6 million. 
Source: National Audit Office analysis of British Business Bank data
2020-21
2021-22
2022-23
2023-24
2024-25
0
2
4
6
8
10
12
14
16
18
20
8.8
17.4
5.7
16.6
5.8
13.8
5.8
9.2
5.9
9.6
Financial year
October 2020 predictions
September 2021 predictions

50  Part Three  The Bounce Back Loan Scheme: an update 
Scheme evaluation
3.13	 Our previous report on the Scheme found that it did not have a business case 
and lacked, at its launch, clear objectives beyond the aim of fast financial support 
for smaller SMEs. The Bank has since commissioned an external evaluation of all 
three COVID-19 business support schemes, which intends to report in stages from 
2022 to 2024. The evaluation will seek to draw lessons from both an operational 
perspective (‘process evaluation’) and economic perspective (‘impact evaluation’).
3.14	 When we reported on the Scheme in October 2020, the Bank told us that it was 
in the process of developing metrics for measuring the performance of the loan support 
schemes. The Department and the Bank have now developed some metrics, as part 
of their evaluation plans, to measure Scheme impact. The Bank’s ‘process evaluation’ 
will report in 2022 and use stakeholder survey data and management information 
to examine the effectiveness of the Scheme’s design and delivery mechanisms. 
This includes, for example, the effectiveness of the Department’s implementation of 
risk management measures. The Bank intends to report its ‘impact evaluation’ three 
times between 2022 and 2024, and includes some metrics for measuring the success 
of the Scheme, including the impact on business survival rates, business turnover, 
employment levels and productivity. Measuring the Scheme’s impact, however, is 
challenging because there is a lack of reliable data to compare the recovery of 
businesses that used the Scheme with a similar group that did not.
Figure 14
Breakdown of forecast total administrative costs for the Bounce Back Loan 
Scheme (the Scheme), 2020-21 to 2024-25
Notes
1 
Figures as of 13 September 2021. 
2 
British Business Bank staff costs also includes interim staff.
3 
External operational support includes IT, legal advice, professional advice and other outsourced services.
4 
Other operational expenditure includes expected credit loss modelling, marketing and communications.
5 
Figures may not sum due to rounding.
Source: National Audit Office analysis of British Business Bank data
Most of the Scheme’s administrative costs over the next five years are for external operational 
support, including costs for IT and legal and professional advice
External
operational support
67%
British Business
Bank staff costs
13%
Lender audits
11%
Other operational expenses
10%

The Bounce Back Loan Scheme: an update  Appendix One  51 
Appendix One
Our approach
Scope
1	
This report follows on from our October 2020 investigation into the Bounce 
Back Loan Scheme (the Scheme), which gave an overview of the Scheme, the 
number and value of loans issued and the main Scheme risks.
The report provides a factual update on the:
•	
number of loans made by the time the Scheme closed, the changes to the 
Scheme, including take-up of flexible repayment options to further help 
businesses repay, the number of businesses that have begun to repay the 
loans, and default data. The data used for this analysis are described in 
Figure 15 overleaf;
•	
Department for Business, Energy & Industrial Strategy’s (the Department’s) 
estimated Scheme credit and fraud losses. This is the Department’s latest 
estimate available in its accounts, which includes data up to 31 March 2021. 
The Department refined its estimates in October 2021; and
•	
Scheme principles for recovering outstanding loans, anecdotal information on 
businesses’ experience of the Scheme, the British Business Bank’s (the Bank’s) 
forecast administrative costs for the Scheme, and the Department’s plans to 
evaluate Scheme performance.
The report evaluates:
•	
the government’s counter-fraud strategy for the Scheme, and its activities to 
date, using our Good practice guidance: Fraud and error as the evaluative 
framework (see Appendix Four);26 and
•	
the report does not evaluate the Scheme’s performance as the full repayment 
of loans will not be known for some time.
The report only focuses on the Scheme; it does not review other COVID-19 schemes 
such as the Coronavirus Business Interruption Loan Scheme, the Coronavirus Large 
Business Interruption Loan Scheme, and the Recovery Loan Scheme.
26	 National Audit Office, Good practice guidance: Fraud and error, March 2021. Available at: www.nao.org.uk/wp-
content/uploads/2021/03/010381-001-Fraud-and-Error-Accessible.pdf

52  Appendix One  The Bounce Back Loan Scheme: an update 
Figure 15
Quantitative datasets used in our report, March to September 2021
We drew on six different main datasets from three different sources: the British Business Bank (the Bank), the Department for Business, 
Energy & Industrial Strategy (the Department) and the Government Counter Fraud Function
Data
Source
Date
How we used the data – 
example figures
Any limitations in the data
Loan value and 
distribution data 
(including by month, 
business size and 
type, location, sector 
and lender) 
The Bank: individual 
loan data from 
lenders
13 September 2021
The Bounce Back Loan 
Scheme (the Scheme) had 
agreed 1.5 million loans worth 
£46.8 billion at the time 
it closed in March 2021.
Lenders vary in how they collate and 
report the information to the Bank. 
Figure includes borrowers who 
applied before the Scheme closed 
but received the funds afterwards.
Loan repayment data 
and take-up of flexible 
repayment options 
The Bank: individual 
loan data from 
lenders
30 September 2021
• 
Borrowers had defaulted on 
2% of all loans, representing 
about £1.3 billion.
• 
Around 21% of eligible 
borrowers took advantage 
of the ‘Pay As You Grow’ 
repayment options.
Lenders vary in how they collate and 
when they report the information to 
the Bank.
Fraud estimates
The Department 
(using 
Pricewaterhouse-
Coopers LLP 
analysis)
31 March 2021 
(updated in 
October 2021)
Estimate for the scale of fraudulent 
loans as 11% by number and 
£4.9 billion by value.
With new information, this estimate 
was refined to 7.5% in October 
2021 when loans previously 
classified as ‘possible’ fraud were 
reclassified as either ‘probable 
fraud’ or ‘no fraud suspected’.
The Department assumed in its 
estimate that any fraud leads to a 
total loss of the loan which is likely 
to overestimate losses as some 
funds may be recoverable.
For full limitations see the 
Department’s Annual Report and 
Accounts 2020-21 – section ‘The 
Report of the Comptroller and 
Auditor General to the House 
of Commons’.
Government Counter 
Fraud Function
April 2021
Estimated that between 4% and 
11% of loans were fraudulent, 
equal to £1.8 billion to £5 billion.
Used a moderate level of 
extrapolation and assumptions on 
false positives and did not cover all 
fraud types.
Fraud prevented 
and detected
The Bank: individual 
fraud loan data 
from lenders
20 April 2021
Lenders claim to have prevented 
£1.97 billion of fraudulent 
applications and detected 
£5.3 million of fraudulent loans.
We could not audit these figures 
as: each lender reports the 
information on a different basis; 
some lenders might choose not 
to report suspected fraud which 
makes comparisons subjective; and 
some lenders do not share their 
underlying data.
Loans not repaid
(credit losses)
The Department 
31 March 2021
Estimated 37% of loans will not 
be repaid at a value of £17 billion.
Based on assumptions in expected 
credit loss model: no credit 
score data for the borrowers; 
limited repayment data; and 
repayment rates rely on future 
macroeconomic conditions.
Scheme administrative 
costs (2020-21 to 
2024-25)
The Bank
September 2021
The Bank forecast the Scheme 
administration costs at £67 million 
by 2024-25.
We have not audited the 
Bank’s underlying operational 
expenditure data.
Note
1 
For further explanation about the loan dataset, please see the section below on ‘loan data’. 
Source: National Audit Offi ce analysis of documentation from the British Business Bank and the Department for Business, Energy & Industrial Strategy

The Bounce Back Loan Scheme: an update  Appendix One  53 
Methods
2	
In reviewing these areas, we drew on a variety of evidence sources provided 
by the HM Treasury, the Department and the Bank. We reviewed key documents in 
relation to the Scheme’s:
•	
extensions and flexible repayment options;
•	
implementation and loan distribution;
•	
credit and fraud loss estimates;
•	
loan recovery process and forbearance options;
•	
counter-fraud strategy and counter-fraud activities agreed with the 
lenders and law enforcement agencies; and
•	
evaluation process.
3	
We undertook a survey of 24 lenders under the Scheme. The survey covered 
the offer of ‘Pay As You Grow’ options, the clarity of the recovery process principles 
under the Scheme, how each lender implemented the recovery process in their 
organisation, the impact of the recovery process has had on their customers, how 
each lender identified and treated vulnerable customers, and whether they collect 
information on how borrowers have used the loan proceeds. Of these, we received 
15 responses, giving a response rate of 63%. The responses included five of the 
seven main UK banks; five of the 11 other UK banks; and five of the six ‘non-banks’, 
such as asset-backed lenders.
4	
We interviewed or corresponded with:
•	
HM Treasury, the Department and the Bank;
•	
the National Investigation Service (NATIS);
•	
business and industry representatives, such as the Confederation of 
British Industry, the Federation of Small Businesses, the Finance & Leasing 
Association and Innovate Finance; and
•	
other interested stakeholders such as the Financial Ombudsman Service 
and customer representative groups.

54  Appendix One  The Bounce Back Loan Scheme: an update 
Quantitative analysis
Datasets used in our analysis
5	
We conducted quantitative analysis on the Department’s data on loan 
disbursements and repayments, credit losses, fraudulent loans and administrative 
costs. These are set out in more detail in the subsequent subsections below. 
Overall, we drew on six main datasets from three different sources, including 
the Department, the Bank and the Government Counter Fraud Function. 
These datasets, their sources and how we used them are shown in the table 
on page 52 (Figure 15).
We analysed loan data
6	
We analysed Scheme loan data from the Bank. Lenders provide data to the 
Bank via a collections system it refers to as ‘the Portal’. It took the Bank until 
July 2021 to have in place an automated system for lenders to report on repayment 
activities. The Bank collected these data using a manual process before July 2021. 
The Bank uses these data to administer guarantees, and they are not real-time. 
The Portal contains data on borrowers’ loan repayments, arrears, ‘Pay As You Grow’ 
options, loan defaults and loan repayments in full. The Portal depends on lenders 
submitting accurate and timely data. The Bank audits lender submissions to the 
Portal; we have not done so. The Portal data’s are intended to be used to manage 
the guarantees rather than monitor loan performance.
7	
In our report we use the Bank’s lenders’ Portal data as our primary dataset 
and hence not all figures in our report reconcile with other publicly available figures – 
which are often based on data up to the financial year end in March 2021. Despite the 
Scheme closure on 31 March 2021, loan numbers and value may differ from those at 
financial year end as lenders could process outstanding applications after the Scheme 
closed. Borrowers continued to draw loans until September 2021 at a very low level. 
The Bank data used in our report have a cut-off date of 13 September 2021, on that 
date 1,540,916 loans had been drawn with a value of £46.76 billion.
8	
The Bank’s Portal data are also used for the repayment statistics, for example 
the number of loans defaulted or in arrears, and loan top-up data. All but one lender 
provided the repayment data via the Bank’s automated Portal; one lender’s data 
on ‘Pay As You Grow’ loans are added manually and is only included in the ‘Pay As 
You Grow’ options figures in this report (Figure 7 and paragraph 1.10). Lenders may 
change loan data in the Portal for several reasons including, for example, a borrower 
in arrears restarting loan payments. As a result, data may vary from month to month, 
in particular ‘in arrears’ and ‘loans defaulted’ information. These repayment data 
are as of 30 September 2021 when 1,539,788 loans had been drawn with a value 
of £46.72 billion.

The Bounce Back Loan Scheme: an update  Appendix One  55 
9	
HM Treasury published data on the number and value of loans approved 
through the Scheme. These figures are available on HM Treasury’s website and we 
do not use them in this report. The figures may differ to the Bank’s Portal data as 
HM Treasury collects summary-level data of loan applications and loan applicants, 
whereas the Bank has data on a loan-by-loan basis. We have explained these 
differences in our 2020 report on Bounce Back Loans. The latest publication from 
HM Treasury shows that as at 31 May 2021, 1,560,309 loans had been drawn down 
with a value of £47.36 billion. HM Treasury has stopped reporting on the Scheme 
since it closed.
10	
We have not audited the underlying loan-level data owing to confidentiality 
issues. The data contain sensitive personal and commercial details. We relied on the 
summary data provided by the Bank. The Bank, in turn, rely on information that the 
lenders provide.
We reviewed the Department’s expected credit loss model
11	
We reviewed the Department’s estimate of credit losses based on our financial 
audit of its 2020-21 Annual Report and Accounts. This looked in detail at the Bank’s 
estimated credit loss model. For each individual guarantee that the government 
has issued, the model estimates the probability that the borrower will default on 
the loan, the amount of the loan at default and any funds which might subsequently 
be recovered. This model is then used to estimate the value of credit losses in 
the Scheme.
12	
The Department’s credit loss estimate is uncertain as it is based on 
assumptions in its expected credit loss model: there is no credit score data for the 
borrowers as this was not a scheme requirement; it has limited repayment data; and 
repayment rates will be affected by future macroeconomic conditions which are 
themselves uncertain. More information is available in the notes to the Department’s 
Annual Report and Accounts.
We looked at the Department’s estimate of the scale of fraudulent loans
13	
We also reviewed the Department’s estimate of the rate of occurrence of 
fraudulent loans in the Scheme as of 31 March 2021. The Department commissioned 
PricewaterhouseCoopers LLP (PwC) to analyse a sample of loans for indicators 
of fraud. PwC reviewed 1,067 randomly selected loans as a statistically significant 
sample of the Scheme at 12 November 2020. It then judged which loans had ‘No 
fraud suspected based on information available’, and for those with ‘indicators 
of potential fraud occurrence’ it classified the loans as ‘probable’, or ‘possible’ 
fraud. This led to its most likely estimate of fraud of 11.15% of loans by number. 
The Department’s estimate excluded some types of possible fraud – for example, 
where an applicant overstates its turnover and gets a larger loan either deliberately 
or inadvertently.

56  Appendix One  The Bounce Back Loan Scheme: an update 
14	
The Department converted the loans by number figure into a value 
(£4.9 billion) by weighting the loans to account for those most at risk of fraud. 
As a result, one cannot simply apply 11.15% to the total outstanding loan 
balance of £46.7 billion to arrive at the value of estimated fraud in the Scheme. 
The Department’s calculations assumed that any fraud leads to a total loss of the 
loan, which is likely to overestimate losses as some funds should be recoverable. 
These fraud estimates are included in the Department’s Annual Report and 
Accounts for 2020-21.
15	
The Department’s advisers, PwC, updated its estimated level of fraud in 
October 2021, which reduced from 11.15% to 7.5%. It did this after gathering 
additional information from lenders, which led to the reclassification of all 48 sample 
loans in the ‘possible’ fraud category. A total of 41 loans moved from ‘possible’ fraud 
to ‘no fraud suspected’ after further discussions with lenders; seven loans moved 
from ‘possible’ to ‘probable’ fraud; and two loans moved from ‘no fraud suspected’ 
to ‘probable’ fraud. A total of 80 out of 1,067 loans had ‘probable’ fraud (7.50% of 
the sample), owing to the reclassification no loans had ‘possible’ fraud, and 987 of 
1,067 loans had ‘no fraud suspected’. We were not able to audit these revised 
estimates in the time available to us.
16	
We also reviewed lender self-reported data on the value and volume of fraud 
from the number of fraudulent applications that they declined, or fraud that they had 
identified. We could not audit these figures as: each lender reports the information 
on a different basis; some lenders might choose not to report suspected fraud, 
which makes comparisons subjective; and some lenders would not share their 
underlying data as they had concerns about data privacy. The Bank told us that 
it does not audit these figures.
We analysed the Bank’s Scheme administrative cost data
17	
The Bank’s preliminary assessment of the Scheme’s administrative costs 
is based on a summary spreadsheet of the Bank’s forecasted annual cost from 
2020-21 to 2024-25. It updates the Bank’s September 2020 administrative cost 
estimate for the Scheme. The assessment is based on the Bank’s estimates of all 
COVID-19 business loan support schemes. The Bank allocated the total costs to 
the relevant scheme, expense category and year. The allocation methods used vary 
depending on the cost items, and mainly use volume, value, or number of schemes. 
We have not audited the underlying operational expenditure data. We have verified 
the calculations in the spreadsheet and conducted limited checks on the allocation 
of the cost between the schemes.

The Bounce Back Loan Scheme: an update  Appendix Two  57 
Appendix Two
Roles and responsibilities for the Bounce Back Loan 
Scheme (the Scheme)
1	
HM Treasury developed the Scheme, and subsequent changes, with the 
Department for Business, Energy & Industrial Strategy (the Department) and the 
British Business Bank (the Bank), which the Department owns. HM Treasury and the 
Department monitored the Scheme and set its overarching terms. In consultation 
with HM Treasury and the Department, the Bank was responsible for Scheme 
implementation and administration.
2	
Commercial lenders (for example, banks, building societies and peer-to-peer 
lenders) were responsible for administering loans to businesses and collecting 
loan repayments. The loans have a fixed interest rate of 2.5% and a maximum 
length of 10 years, and businesses are expected to repay the debt in full. In the 
first year of the loan, no capital repayments were due, and the government paid 
the interest – making it interest-free for the borrower. The Bank oversees lender 
compliance with the Scheme’s terms, and it engaged third parties (KPMG and RSM) 
to conduct periodic audits. Figure 16 overleaf outlines the roles and responsibilities 
for the Scheme.
3	
As part of an application, businesses had to self-certify their application, 
including the turnover of the business, and that they had been impacted by 
COVID-19 and were able to repay the loans. The Scheme allowed lenders to verify 
application details as they “see fit”, but there were no requirements to do so. 
Lenders were required to conduct anti-fraud, anti-money-laundering and ‘know 
your customer’ checks on all loan applications.
4	
As lenders were not allowed to assess whether customers were able to 
repay the loans, government provided a 100% guarantee to the lenders: if the 
borrower defaulted on the loan, the lender could recover the funds from government 
through the guarantee. If lenders do not follow the Scheme principles, for example, 
if the borrower was ineligible for a loan, government is not obliged to honour the 
guarantee. The Scheme continued to rely on this self-certification approach between 
October 2020 and March 2021, when the Scheme was extended, even though the 
immediate need for quick access to finance during the pandemic lessened when the 
government eased national COVID-19 restrictions.

58  Appendix Two  The Bounce Back Loan Scheme: an update 
Figure 16
Roles and responsibilities for delivering the Bounce Back Loan Scheme (the Scheme), UK
The government utilises arm’s-length bodies and private organisations to deliver the Scheme
Stakeholder
Description
HM Treasury
HM Treasury initiated the Scheme and, in conjunction with the Department for Business, Energy & Industrial 
Strategy (the Department) and the British Business Bank (the Bank), set the policy and designed the Scheme. 
HM Treasury collected and published Scheme data until the Scheme closed in March. It also led the discussion 
with lenders on finalising the operational guidance in relation to the debt recovery process under the Scheme. 
Department for 
Business, Energy 
& Industrial 
Strategy
The Department is the Bank’s sole shareholder and has a wider policy remit for business and enterprise. 
As sole shareholder, its accounting officer is ultimately responsible for the three COVID-19-related business 
support schemes and issued the Written Direction to the Bank to pursue the Scheme. The Department is also 
responsible for the design of the debt recovery process, the strategy for tackling fraud, and its counter-fraud 
governance and resourcing.
British 
Business Bank
The Bank is the Department’s delivery partner for the loan schemes (via lenders). The Bank helped design the 
Scheme and is responsible for its administration and implementation, in consultation with HM Treasury and 
the Department. The Bank delivers the loans via a network of accredited lenders. Within its responsibilities is 
the implementation of the counter-fraud strategy designed by the Department in collaboration with the lenders 
and other anti-fraud organisations. It is also in charge of auditing the lenders through contracted third parties. 
The Bank also helps with monitoring and analysing the repayments data reported through the lenders’ Portal. 
Lenders
The Bank accredited a network of lenders to review applications and provide loans to borrowers. Lenders are 
responsible for conducting the required anti-fraud, anti-money laundering and ‘know your customer’ checks prior 
to loan approval. Lenders administer loan repayments and are responsible for pursuing borrowers to establish a 
recovery plan for missed repayments for up to 12 months after the issue of a formal demand. Lenders are required
to follow counter fraud activities under the financial service regulations and other legislation.
Other 
organisations 
involved in 
fraud-related 
checks
The Bank coordinates with lenders and fraud-related organisations to mitigate fraud, including:
• Government Counter Fraud Function – Based in the Cabinet Office, the Function and works closely with the 
Bank and lenders to tackle fraud.
• National Crime Agency – Investigates and prosecutes the most serious fraud cases involving serious or 
organised crime.
• National Investigation Service (NATIS) – NATIS is a law enforcement body specialising in financial crime and 
fraud, mostly at local authority level. NATIS will investigate and prosecute medium/high severity fraud cases 
supported by the Crown Prosecution Service.
• Insolvency Service – Investigates and prosecutes insolvency and corporate offences on the 
Department’s behalf. 
• Companies House – The UK’s registrar of companies; can share company data to enable better identification 
of fraudulent applications. 
• UK Finance – A trade association for the banking and finance industry. 
• Counter-fraud organisations – There are several nationally recognised organisations that allow lenders to 
check loan applications for fraud. Under the terms of the Scheme, each lender must use a reputable fraud 
bureau to screen for fraud at the application stage.
Note
1 
The three COVID-19-related business support schemes are the Coronavirus Business Interruption Loan Scheme, the Coronavirus Large Business 
Interruption Loan Scheme and the Bounce Back Loan Scheme.
Source: National Audit Offi ce review of HM Treasury, British Business Bank and the Department for Business, Energy & Industrial Strategy documentation

The Bounce Back Loan Scheme: an update  Appendix Three  59 
Appendix Three
Number and value of loans at September 2021
1	
The Bounce Back Loan Scheme represents 94% of the three COVID-19 
business support schemes by the number of loans (Figure 17). These three schemes 
provided a total of 1,638,834 loans worth £77.4 billion as of 13 September 2021.
Figure 17
Number and value of loans provided by the three COVID-19 business support 
schemes, September 2021, UK
The Bounce Back Loan Scheme is the largest of the three COVID-19-related schemes
Business loan 
support scheme1
Total number of 
loans provided2
Total value 
of loans2
Percentage of 
total facilities
Percentage of 
total value
(£bn)
(%)
(%)
Bounce Back 
Loan Scheme
1,540,916
46.8
94.0
60.4
Coronavirus 
Business 
Interruption 
Loan Scheme
97,198
25.8
5.9
33.3
Coronavirus 
Large Business 
Interruption 
Loan Scheme
720
4.9
0.0
6.4
Total 
1,638,834
77.4
–
–
Notes
1 
The Coronavirus Business Interruption Loan Scheme, Coronavirus Large Business Interruption Loan Scheme and 
Bounce Back Loan Scheme were launched on 23 March 2020, 20 April 2020 and 4 May 2020, respectively.
2 
Figures as of 13 September 2021. 
3 
Figures may not sum to 100% due to rounding. 
Source: National Audit Offi ce analysis of British Business Bank data

60  Appendix Four  The Bounce Back Loan Scheme: an update 
Appendix Four
Our fraud evaluative framework
1	
We reviewed the government’s counter-fraud activity for the Bounce Back Loan 
Scheme against good practice as set out in our March 2021 good practice guide 
on fraud and error.27 We also drew on the Standards set out by the Government 
Counter Fraud Function, whom we consulted in the course of our work.
2	
The main areas of good practice covered counter-fraud strategy; governance 
and resourcing; and implementation, measurement and evaluation of controls 
to detect and prevent fraud. Our evaluative criteria for these areas are set out 
in Figure 18.
27	 National Audit Office. Good practice guidance: Fraud and error, March 2021, page 13. Available at: www.nao.org.uk/
wp-content/uploads/2021/03/010381-001-Fraud-and-Error-Accessible.pdf

The Bounce Back Loan Scheme: an update  Appendix Four  61 
Figure 18
Our criteria for evaluating the Bounce Back Loan Scheme’s (the Scheme’s) 
counter-fraud activity
To evaluate the counter-fraud activity we looked at the strategy, governance, design, implementation, 
measurement and evaluation of the Scheme
Strategy 
There is a strategy for tackling fraud risk, based on robust evidence and analysis, leading to 
clear prioritisation.
The overall strategy for tackling fraud and error prioritises activities based on evidence of their 
cost-effectiveness.
The organisation’s fraud and error risk appetite is clearly agreed with partners and documented.
Where new risks and opportunities are identified, options for new controls are evaluated on a 
cost-benefit basis and introduced on a timely basis.
The fraud and error impact of all changes to policy and operations are considered. Where an increased 
risk of fraud and error is accepted as a trade-off with other policy objectives this is explicitly laid out. 
Governance 
A governance structure is in place over fraud and error risk, with a clearly defined remit and ability to 
hold other parts of the organisation to account for implementing the fraud and error strategy. 
There is timely and comprehensive reporting to those charged with governance over fraud and error risk.
There is a strong counter-fraud and error culture at all levels within the organisation. 
The organisation publishes and reports against targets for fraud and error, based on its expectation of 
the intended impact of its counter-fraud and error initiatives over time.
The organisation has set clear targets towards a cost-effective control environment.
Design 
The strategy has defined the key fraud and error risks. This should include the different types of fraud 
and error, where those risks enter the system and what causes the error, for example whether it is 
organisation error, customer error or fraud. 
Controls are designed to effectively prevent and detect known fraud and error risks. 
The expected cost and impact of each control is understood.
Implementation
Data are shared and matched across all government departments where appropriate to identify fraud. 
Individual resourcing decisions are made with an understanding of the cost and impact on fraud and error.
Measurement 
A measure of fraud and error is properly estimated.
The estimate is appropriately and regularly reported. 
Individual controls are properly measured, with key performance indicators for each control. 
Evaluation and feedback 
Controls are evaluated regularly to look at how risks are being tackled and to identify new and emerging 
risks, and this is integrated into the strategy. 
Source: National Audit Offi ce 

This report has been printed on Pro Digital 
Silk and contains material sourced from 
responsibly managed and sustainable 
forests certified in accordance with the 
FSC (Forest Stewardship Council).
The wood pulp is totally recyclable and 
acid-free. Our printers also have full ISO 14001 
environmental accreditation, which ensures 
that they have effective procedures in place to 
manage waste and practices that may affect 
the environment.

You have reached the end of this document
9
781786
044013
ISBN 978-1-78604-401-3
£10.00
Design and Production by NAO Communications Team 
DP Ref: 010814-001

File and source

File

NAO_Bounce-Back-Loan-Scheme-an-update_2021-12.pdf

Size

519,370 bytes

SHA-256

dfe9d273407e9b9cf5cf02ebfd8a4d91e0c03371ff48bcd0dced0011bd85ec14

Our copy

NAO_Bounce-Back-Loan-Scheme-an-update_2021-12.pdf

Original

www.nao.org.uk