FINAL NOTICE

To:

Amigo Loans Ltd

1. ACTION

1.1
For the reasons given in this Final Notice, the Financial Conduct Authority (“the

Authority”) hereby publishes a statement pursuant to section 205 of the Financial

Services and Markets Act 2000 (“the Act”) that Amigo Loans Ltd (“Amigo”)

contravened regulatory requirements.

1.2
The serious failings in this case warrant a substantial financial penalty. Amigo has

provided verifiable evidence that the payment of such a penalty would cause the

firm serious financial hardship. On 23 May 2022, the High Court approved Amigo’s

scheme of arrangement, which aims to provide redress to customers who were

mis-sold loans and who raised a complaint. The Authority is satisfied that the

imposition of a penalty could jeopardise Amigo’s ability to meet its commitments

under the scheme of arrangement. Had it not been for Amigo’s financial position,

the Authority would have imposed a financial penalty of £72,900,000.

2. SUMMARY OF REASONS

2.1.
Amigo’s sole business during the Relevant Period was as a guarantor lender. It

has strategically positioned itself in the market as a finance provider to consumers

2

who cannot access finance from traditional lenders, due to their circumstances or

credit history.

2.2.
Firms offering consumer credit are required to make a reasonable assessment of

not just whether a customer can repay a proposed loan, but also whether the

customer can do so affordably.

2.3.
Between 1 November 2018 and 31 March 2020 (the Relevant Period), Amigo

breached Principle 2 (skill, care and diligence), Principle 3 (management and

control) and Principle 6 (customers’ interests) of the Authority’s Principles for

Businesses (the “Principles”).

2.4.
By failing to appropriately consider regulatory requirements, recognise emerging

trends and adapt its lending approach accordingly, Amigo failed to ensure that its

systems and controls adequately assessed and monitored customer affordability.

2.5.
Amigo also failed to identify issues with its systems and controls through its own

governance and oversight. Amigo was overly focused on profitability and “getting

loans out the door”. As a result, it failed to give appropriate consideration to the

interests of customers and the risk that its business model, if not carefully

controlled, could lead to widespread lending that was unaffordable to customers.

2.6.
The risk of harm affected both borrowers and guarantors. Amigo did not have

appropriate processes in place to ensure it adequately assessed borrower and

guarantor circumstances prior to lending, to ensure that lending was affordable

and in compliance with its regulatory obligations. Guarantors were entitled to rely

upon Amigo’s assessment that a loan was affordable for a borrower and undertook

to make payments only where the borrower was not able to do so. However,

Amigo’s assessment did not make sure that the borrower would be able to repay

the loan affordably without it impacting their wider financial situation. As a result,

there was an increased risk that guarantors would have to step in and make

payments.

2.7.
By November 2018, Amigo business model involved a heavily automated

approach to lending, with increasing reliance placed on pre-programmed IT

system decision making. Amigo had gradually increased automation and

decreased the involvement of human agents leading up to the Relevant Period.

This allowed Amigo to grow its business by increasing its lending volumes and the

3

speed at which it could process applications. It was also intended to increase the

consistency of outcomes for customer applications.

2.8.
Lending decisions were primarily driven by the parameters within Amigo’s IT

system. The reduction in human contact as part of Amigo’s lending process had

potential benefits for customers as it meant they could easily make applications

with eligibility quickly and consistently assessed. There were also operational

advantages for Amigo as a large part of the underwriting process became

automated.

2.9.
But there were also significant risks inherent with this approach. The first was that

decision making was, to a significant degree, dependent on IT system logic. It

was therefore vital that the system incorporated appropriate parameters, triggers

and controls around affordability and creditworthiness to prevent Amigo lending

to customers in circumstances where it was potentially unaffordable for the

customer. Although there were controls of this type in place, they were insufficient

and design issues meant that the system approved (subject to final agent sign-

off) loan applications in circumstances where there were indicators that the

borrower could not afford the loan or where there was insufficient evidence to

conclude that the loan was affordable. This unaffordable lending caused harm to

both borrowers and guarantors. 1 in 4 of Amigo guarantors were asked to step in

and make payments to assist struggling borrowers at some point during the

lifecycle of the loan although payments by guarantors accounted for 10% of total

loan payments.

2.10.
Secondly, although some of the controls built into the system raised flags for

further review by human agents, there remained a risk, which crystallised in many

cases, of unaffordable lending if the agent did not sufficiently consider information

provided by the customer or adequately probe the information they were given

before approving the lending. This risk was heightened in circumstances where

the reward system for pay-out agents was heavily weighted towards loans paid

out.

2.11.
Strong governance and oversight of both the IT system configuration and the role

played by human agents, was critical to mitigate those risks. It required senior

management to fully consider the needs of Amigo’s customer base, the relevant

regulatory requirements under CONC and the operation and evolution of Amigo’s

business model. Amigo’s governance and oversight mechanisms failed to identify

significant weakness in its approach to lending, exposing consumers to a

significant risk of large-scale unaffordable lending. In particular:

(1)
Amigo did not adequately consider regulatory requirements around

affordability, focusing too narrowly on credit risk

2.12.
Historically Amigo viewed affordability of a loan as including a strong element of

choice for customers. The theory being they could cut back on certain expenses

which could be adjusted at the discretion of the customer in order to make a

proposed loan affordable. Amigo’s assessment focused too heavily on the

customer’s credit risk. On 1 November 2018 the Authority introduced clarified

consumer credit rules which included a clearer articulation of the expectation that

firms would look beyond the customer’s ability to repay, to their ability to repay

affordably. Amigo considered its position at this time and concluded that it was in

compliance with the requirements. As a result of this mistaken belief, it made very

limited changes to its lending parameters and processes, continuing to focus on

credit risk and the prioritisation of commercial results at the expense of customer

outcomes.

2.13.
Amigo also considered the low proportion and number of upheld FOS decisions in

the preceding years as an indication that its assessment of affordability was

adequate. However, Amigo failed to ensure that its approach to affordability

evolved in line with changes in regulation and market approach. Its horizon

scanning failed to recognise emerging trends and adapt its lending approach to

ensure that it was lending affordably. This was compounded subsequently by

Amigo’s failure to act appropriately on the findings of a number of internal and

external reviews from mid-2019 which had identified weaknesses in its approach

to the assessment of affordability and creditworthiness. Amigo also failed to take

on board lessons arising from root cause analysis of the irresponsible lending

complaints against the firm. This was in part due to inadequate root cause analysis

but also resulted from a failure to apply the learnings Amigo did identify from

complaints to its lending approach.

2.14.
This was in part due to a lack of visibility of key risks and emerging issues at

senior management level. Issue management was re-active and there were no

risk indicators that could be used to support risk-based decisions. Amigo failed to

give sufficient consideration to affordability in its MI, which would have informed

5

its oversight. The MI metrics it did have in place were geared towards measuring

credit risk and did not identify customer detriment.

(2)
Amigo’s assessment of customer affordability was inadequate

2.15.
Affordability was assessed based on Amigo’s Budget calculator. However, the

information gathered, and the verification of that information was not sufficient

to provide assurance as to the customer’s affordability and creditworthiness for

the following reasons:

a. Amigo unduly relied on information provided by customers in the Budget.

The Budget indicated to customers whether they had entered the required

income and expenditure to be approved for a loan and allowed them to

adjust the entered income and expenditure accordingly.

b. Amigo did not adequately consider the sustainability of income or

expenditure in its assessment beyond a tick box requiring customers to

confirm that nothing was due to impact the income they received or cause

their spending to increase.

c. Amigo’s approach to income verification did not provide reasonable

assurance as to the customer’s income and their ability to afford the loan.

d. Amigo’s approach to top up lending was inappropriate and risked getting

customers into a spiral of increasing debt. Although customers in arrears

could not apply for a top up, Amigo failed to pay due regard to borrowers

who had recently been in arrears or who had taken repeated top up loans.

It encouraged them to apply for additional lending by text message or

email without having sufficient regard for the potential impact of increased

lending on their financial circumstances.

e. Amigo’s assessment of expenditure was inadequate. Amigo failed to assess

the full range of non-discretionary expenditure or consider the composition

of a borrower’s household. Amigo also failed to ensure that the data it used

to verify expenditure was kept up to date.

(3)
Amigo failed to ensure that its controls were effective around

creditworthiness

2.16.
Amigo did not establish and implement effective policies and procedures to enable

it to carry out creditworthiness assessments and ensure adequate risk

management systems for the following reasons:

6

a. Amigo’s pay and reward system for pay-out agents was heavily weighted

towards loans paid out without adequate controls, which risked

incentivising the wrong behaviours.

b. Responsibility for ensuring that affordability was adequately assessed was

not clear.

c. Concerns raised by Internal Audit and Compliance from mid-2019 were

not addressed. These concerns related to:

i.
A lack of effective risk management framework.

ii.
Too narrow a definition of potential customer vulnerability.

iii.
Poor retention of documentary evidence in relation to the

assessment of customer affordability where required by Amigo’s

system; and

iv.
A lack of horizon scanning and compliance advice by the

Compliance function.

d. Amigo failed to effectively monitor its lending to ensure that it was

affordable for its customers through inadequate quality assurance.

(4)
Amigo did not adequately consider the profile of its customer base which

increased the risk of potential harm to customers

2.17.
A significant proportion of Amigo’s customers were potentially vulnerable. Amigo

had a responsibility to ensure that vulnerable customers were adequately

protected from the risk of harm. This placed additional responsibility on Amigo

beyond the responsibility to ensure it was lending affordably. Amigo failed to

identify potentially vulnerable customers at the application stage because it relied

on its lending system to pick up indicators of vulnerability in the data available to

it. However, the system was designed with an unreasonably narrow definition of

vulnerability, limited to customers on incapacity benefit for a mental health issue,

which Amigo considered might affect the customer’s understanding or

management of the loan agreement.

2.18.
The narrow definition of vulnerability in Amigo’s system made it very difficult for

Amigo to assess and identify possible signs of vulnerability prior to issuing loans.

Amigo was aware that it’s lending model would likely attract more vulnerable

customers compared to high street lenders, for example a significant proportion

of its customers were in receipt of benefits income. However, Amigo only identified

1.11% of customers as vulnerable at the point of lending during the Relevant

Period. This was despite Amigo having access to information in the customer’s

7

credit file which could have been flagged by its system to indicate that the

customer was in financial difficulty. This information, although available, was not

considered by Amigo as an indicator of potential vulnerability.

2.19.
The Authority considers the failings by Amigo to be serious. Amigo’s failure to

conduct an effective creditworthiness assessment created a high risk of customer

harm. Guarantors should have been able to rely on Amigo’s assessment that the

borrower could afford the loan. Amigo’s Budget assessment did not provide

sufficient assurance that a customer could afford to repay their loan. The Authority

considers it is likely that there was widespread harm as a result of this. The

seriousness of this is compounded by the fact that it is likely that a reasonably

significant proportion of Amigo’s customers were vulnerable.

2.20.
Amigo assessed its risk appetite in light of emerging COVID pandemic and decided

to stop lending in March 2020; although a very small amount of lending was

permitted to be advanced to key workers until November 2020, over which period

approximately 20 loans were written.

2.21.
Amigo did not maintain adequate records and on some occasions during the

Authority’s Enforcement investigation was unable to provide adequate responses

to questions by the Authority due to a lack of sufficient records historically. This

included negligently, automatically deleting the email accounts of relevant staff

that had left the firm after the Authority’s investigation commenced, despite the

Board acknowledging the Authority’s instructions regarding the importance of

retaining documents for the Authority’s investigation. This was considered to be

an aggravating factor for the purposes of the penalty calculation.

2.22.
On 23 May 2022, the High Court sanctioned a scheme of arrangement (the

"Scheme") which aims to provide redress to Amigo's customers. The majority of

eligible customers who voted on Amigo's proposals voted in favour of the Scheme.

Amigo has provided verifiable evidence of its financial position and the Authority

is satisfied that the imposition of a financial penalty would threaten Amigo's

solvency and its obligations under the Scheme. Were it not for Amigo's financial

position
the
Authority
would
have
imposed
a
financial
penalty
of

£72,900,000. The Authority has therefore decided to reduce the penalty to £nil.

3. DEFINITIONS

3.1.
The definitions below are used in this Notice:

“the Act” means the Financial Services and Markets Act 2000;

“Amigo” means Amigo Loans Ltd;

“the Authority” means the Financial Conduct Authority;

“Board” means the Board of Directors of Amigo;

“the Budget” means the income and expenditure section of Amigo’s lending

application form;

“CMC” means Claims Management Company;

“Committees” means Amigo’s Audit Committee, Nomination Committee,

Remuneration Committee and Risk Committee;

“CONC” means the Consumer Credit sourcebook, part of the Authority’s

Handbook;

“CXC” means Customer Experience Coach;

“Default” means where a payment is not made within the terms of a credit

agreement and a Notice of Default is issued. Amigo sent a Notice of Default to

both the borrower and guarantor on the 15th day in arrears.

“DMP” means Debt Management Plan;

“the FOS” means the Financial Ombudsman Service;

“IVA” means Individual Voluntary Arrangement;

“KRI” means key risk indicator;

“MI” means management information;

“ONS” means the Office for National Statistics;

“Pay-out agents” means underwriting agents as referred to by Amigo;

“QA” means Quality Assurance;

“the Principles” means the Authority’s Principles for Businesses, within the

Authority’s Handbook;

“RCA” means root cause analysis;

“the Relevant Period” means 1 November 2018 to 31 March 2020;

“RMI” means the cost of servicing an Amigo loan;

“top-up lending” means additional lending offered to existing Amigo customers;

“the Scheme” means Amigo’s scheme of arrangement approved by the High Court

on 23 May 2022; and

“the Tribunal” means the Upper Tribunal (Tax and Chancery Chamber);

4. FACTS AND MATTERS

4.1.
At the time of the Relevant Period, Amigo provided only guarantor loans and was

a non-prime lender. It had strategically positioned itself in the market as a lender

providing finance to consumers who due to their financial circumstances or credit

history, cannot borrow from traditional lenders. Amigo’s marketing promotes it as

a company which makes borrowing possible and affordable for people who would

otherwise be excluded from mainstream financing, offering borrowers the

opportunity to rebuild their credit scores through its guarantor lending

proposition.

4.2.
Amigo had ambitious targets for growth following its stock market listing. By the

start of the Relevant Period, Amigo’s business model was increasingly automated

with loan applications submitted online and lending decisions made according to

parameters set within a sophisticated IT system. There had been a gradual shift

away from a process whereby agents would speak to all borrowers and carry out

the affordability assessments manually.

4.3.
Guarantor lending introduces a second individual into a lending relationship,

typically a family member or friend with a stronger credit profile than the

borrower. The guarantor then guarantees that if the borrower is unable to make

a payment under the loan, the guarantor becomes liable and must make that

payment on the borrower’s behalf. Both borrowers and guarantors needed to pass

Amigo’s affordability checks. Amigo’s borrowers and guarantors generally had a

worse credit profile than the UK average.

4.4.
Amigo typically offered loans from £2,000 to £10,000 over terms between 12 and

60 months, with a fixed Annual Percentage Rate of 49.9%. Amigo assessed

affordability through its Budget calculator with its loans being deemed affordable

for customers who had a monthly net disposable income (which was above a

minimum buffer amount). Amigo specified the costs categories to be included in

its application form and included the monthly cost of the loan and a buffer in the

Budget calculation.

4.5.
In addition to issuing new loans to new customers, Amigo offered additional

lending to existing customers. This was referred to as top-up lending. If a

borrower took out a top-up loan, this would replace the original loan so there

would only ever be one loan outstanding per customer.

4.6.
A significant amount of Amigo’s lending was top-up lending, with 41% of loans

issued in the Relevant Period being top-up loans. Top-up lending was part of

Amigo’s lending strategy and was inherent in the design of its business model. It

was a driver for the growth of the business and was actively promoted to

borrowers. However, at the beginning of 2019, top up lending became a focus of

customer complaints and was also considered in some decisions by the FOS. In

July 2019, with respect to a selected number of cases, Amigo changed its eligibility

criteria to top up, increasing the number of payments that had to be made on a

loan to 6 before a borrower would be eligible. Prior to that, the majority of

customers were required to have made 3 monthly payments.

4.7.
Amigo’s underwriting process was predominantly task and system driven. This

meant that the system would make decisions according to parameters set and

underwriting agents, referred to by Amigo as pay-out agents, would be allocated

tasks where the borrower or guarantor profile was outside the parameters. This

created a risk if the system did not have the right parameters, triggers and

controls around affordability and creditworthiness, that customers could be

deemed to be affordable, in circumstances where in fact Amigo should not have

lent. There was a further risk that even where flags or triggers were appropriately

raised, that agents might not deal with issues appropriately.

4.8.
Strong governance and oversight were critical to mitigate those risks. The

Authority’s requirements around consumer credit, under CONC and as outlined in

the Authority’s guidance, noted that proportionate and appropriate affordability

assessments were key to avoiding harm in this area, and firms should ensure they

were making responsible assessments of the sustainability of borrowing by taking

into account the particular circumstances.

4.9.
Without robust oversight of the system and the tasks completed by agents, there

was a serious risk that issues with Amigo’s approach to lending would not be

picked up. In addition to this, without proper horizon scanning around the

assessment of affordability and creditworthiness, there was a clear risk that the

parameters in the system, the tasks raised, and the approach taken by agents

would not necessarily be sufficient to ensure lending was affordable.

Interventions by the Authority

4.10.
On 25 November 2019, following a review of guarantor lending, in which Amigo

participated, the Authority wrote to Amigo setting out a number of findings and

points to be actioned. The findings included concerns about the increase in

guarantors having to make payments which suggested a fall in the standard of

lending decisions being made, and lack of affordability for customers. The

Authority requested that Amigo review the increase in the number of guarantors

having to make payments, identifying any issues relating to affordability checks

carried out and asked the firm to put in place a plan to reduce the proportion of

guarantors required to make payments.

4.11.
Despite this, there was no review of the processes by which Amigo assessed

affordability. Amigo relied instead on MI around the guarantor payments in the

first 6 months of a loan remaining fairly constant as an indicator that its

affordability assessments were adequate. Amigo considered that arrears after 6

months would have indicated a change in the borrower’s circumstances.

4.12.
From the beginning of 2020, the Authority increased its interaction with Amigo

initially concerning its creditworthiness and forbearance policies, and then also its

complaints handling processes.

4.13.
Following further correspondence and site visits, the Authority wrote to Amigo to

set out 3 broad categories of concern:

1.
The first related to potential for unfair customer outcomes as a result of

Amigo’s affordability assessments and ensuring that income and

expenditure assessments were reasonable and properly verified.

2.
The second related to various aspects of Amigo’s complaints handling,

including concerns that Amigo had changed its approach to complaint

handing and that it might not be handling complaints in line with the

Authority’s DISP rules, which require that firms must take reasonable

steps to identify and remedy any recurring or systemic issues; and

3.
The third related to the provision of information to guarantors and

concerns that guarantors were not receiving appropriate information

about their potential obligations.

4.14.
The Authority conducted a review of 15 loan files and provided feedback on its

findings to Amigo in April 2020. The Authority found that in 11 out of 15 loans

(73%) there had been inadequate affordability assessments on at least one of the

parties. 40% of these customers subsequently experienced an adverse event in

their repayment history. In several instances, the files showed that Amigo did not

adhere to its own policies and procedures, leading to customer harm. The file

reviews also flagged a potential lack of appropriate oversight and governance.

The review also identified issues with the firm’s interpretation of borrower credit

files.

4.15.
It was noted that Amigo’s focus was on credit risk, with policies and procedures

set around reducing the likelihood of a customer failing to pay within the terms of

their credit agreement and a Notice of Default being issued as a result (i.e., the

risk of default) with inadequate consideration of whether the customer could

afford the loan. Further, the Authority found that several areas of Amigo's

affordability policies were weak, and the governance and oversight of these areas

was poor. For example, when completing the online application, the customer was

notified when their expenditure exceeded their income or if an expenditure item

was lower than the ONS average. This allowed customers an opportunity to

manipulate the application to secure credit. Also, the Authority was informed that

there had been no due diligence carried out by Amigo on the data it relied on to

verify customer income.

4.16.
Amigo stopped lending in March 2020, although a very small amount of lending

continued to be advanced to key workers until November 2020. Amigo was

referred to Enforcement and formally placed under investigation by the Authority

in May 2020.

4.17.
As part of its investigation into Amigo, the Authority completed a file review of a

further 37 customer files. These files demonstrated significant failings by Amigo

to adhere to the rules in CONC and to adequately consider customer interests.

The files highlighted the poor outcomes that its customers experienced as a result

of Amigo’s failings. The Authority found key failings in relation to:

a.
Inadequate verification to provide evidence of the customer’s current

income as required by CONC 5.2A.15R.

b.
Inadequate verification of expenditure meaning that Amigo did not have

a reasonable estimate of a customer’s non-discretionary expenditure as

required by CONC 5.2A.17R.

c.
Lack of consideration by Amigo of customer credit history and indicators

of financial difficulty; and

d.
Marketing of top-up loans to borrowers who had been in arrears or had

taken several top-ups in a short period of time. This may not have been

in the borrower’s best interests and exposed guarantors to an increased

chance of needing to make payments.

4.18.
A summary of the Authority’s findings is detailed at Annex B together with 3

detailed customer file examples highlighting typical issues and failings.

Vulnerable Customers

4.19.
A significant proportion of Amigo’s customer base during the Relevant Period had

low financial resilience and many displayed aspects of vulnerability. Amigo had a

responsibility to ensure that vulnerable customers were adequately protected.

Prior to the Relevant Period, the Authority defined vulnerability as something that

could be temporary, sporadic or permanent in nature noting that many people in

vulnerable situations would not diagnose themselves as vulnerable. The Authority

accepts that defining vulnerability can be difficult, but Amigo’s view was too

narrow. Amigo’s borrowers were individuals who did not always have access to

traditional forms of credit due to their credit history. Amigo’s borrowers were

often reliant on benefit income and around a quarter of loans made were for the

purpose of debt consolidation. Accordingly, a significant proportion of its

customers would have been financially vulnerable but were not flagged as such

by Amigo on its system.

4.20.
On 6 March 2019, the Authority issued a Portfolio Strategy Letter to firms

providing high-cost lending products. Notably, the letter stated that customers

who use high-cost credit products tend to share some key characteristics – for

example, they tend to have poor credit histories and low financial resilience and

many of them are also likely to be vulnerable. It noted that there was a risk of

considerable harm given these characteristics, lack of appropriate affordability

checks and poor treatment of customers who were behind on payments (in

arrears) or who had failed to pay within the time required under the credit

agreement and a Notice of Default had been issued (in default). The letter

reminded firms of their obligations to treat customers fairly and appropriately.

4.21.
Despite this, Amigo did not flag the high likelihood of potential vulnerability in its

customer base during the Relevant Period. Amigo primarily identified vulnerable

customers through contact after a lending decision had already been made,

through calls between a customer and a Collections agent. Amigo was aware that

its lending model would likely attract more vulnerable customers compared to

high street lenders, for example a significant proportion of its customers were in

receipt of benefits income. However, Amigo only identified 1.11% of customers

as vulnerable at the point of lending during the Relevant Period. The Authority

expects firms to pay attention to indicators of potential vulnerability. Amigo did

not do this. Amigo relied primarily on its lending system to pick up indicators of

vulnerability in the data available to it. However, the system was designed with

an unreasonably narrow definition of vulnerability. The only information that

would prompt Amigo to make any further enquiry from a vulnerability perspective

was if a customer declared receipt of incapacity benefit. On checking, only if the

incapacity benefit was received for a mental health issue which might affect the

customer’s understanding or management of the loan agreement, would the

agent flag the customer as vulnerable.

4.22.
The narrow definition of vulnerability in Amigo’s system made it very difficult for

Amigo to assess and identify possible signs of vulnerability prior to issuing loans.

Amigo has subsequently acknowledged that it should have taken a wider view of

vulnerability in respect of its customer base.

4.23.
In July 2019, Amigo’s narrow view of vulnerability was raised in a report by its

outsourced Internal Audit function and was rated ‘high’, with several

recommendations for improvement. This report noted that vulnerable customers’

circumstances may mean that they are significantly less able to represent their

own interests and more likely to suffer harm than the average consumer. The

report identified the risk that if indicators of vulnerability were not identified by

Amigo, customers in vulnerable circumstances may not have been adequately

protected or treated fairly.

4.24.
Amigo implemented additional training to ensure (i) consistency in defining

vulnerability and (ii) treating vulnerable customers fairly. However, this did not

address the narrow definition of vulnerability in Amigo’s lending system and would

have been more likely to identify customers as vulnerable after a lending decision

had been made on the basis that it was primarily through customer calls with the

Collections team that vulnerability was identified.

4.25.
Amigo’s Compliance function also expressed concern that some customers were

not being recognised as vulnerable who should be. It was also proposed that

Amigo flag as potentially vulnerable all customers in receipt of benefits income in

line with recent FOS decisions at that time.

4.26.
No further action was taken on that point and no further triggers beyond the

receipt of benefit relating to disability were included in Amigo’s system to flag

potential vulnerability.

4.27.
Information from the customer’s credit file was not considered for potential

indicators of vulnerability during the Relevant Period. Amigo had access to

information in the credit file which could have been flagged by its system to

indicate that the customer was in financial difficulty. That information should have

been considered as part of the assessment of affordability as it would have been

relevant to whether repayment was affordable for the customer. Where customers

are in financial difficulty, they could also be potentially vulnerable. Resilience is a

key driver of vulnerability and includes low or erratic income, over indebtedness

and low savings as factors. These indicators of potential vulnerability could have

been seen from the information Amigo held but were not considered.

4.28.
In 67.5% (25 out of 37) of loan files reviewed by the Authority, the Authority

assessed that there were indicators of potential vulnerability on the customer’s

credit file. These were indicators which would have suggested the customer may

have been in financial difficulty and susceptible to detriment. Despite this, none

of the customers were identified as potentially vulnerable by Amigo.

4.29.
For example, in one of the files, the borrower’s credit file showed a 126% balance

to limit ratio, 3 accounts in Default (including over-the-limit credit cards). This

information would suggest that the customer was possibly vulnerable due to over-

indebtedness, but they were not identified as such by Amigo and as such no

further assessment or contact took place prior to lending.

4.30.
Customers applied for guarantor loans through Amigo’s website. The application

form contained a series of questions requiring customers to input personal details

and details of their income and expenditure. The income and expenditure section

of the application form was referred to as the “Budget”.

4.31.
The underwriting process was predominantly task and system driven. The level of

automation meant that agents would focus on specific tasks raised by the system

and typically call customers with queries on their application or deal with questions

from customers rather than talk through the entire application.

4.32.
The parameters and triggers in the system around affordability and

creditworthiness were critical to ensuring that customers were able to afford the

money they applied to borrow. It was also important that where tasks were raised

by the system, Amigo’s agents probed sufficiently to ensure the customer could

afford to borrow. Together, these issues contributed to Amigo’s failure to ask for

more information about the customer’s financial position before lending.

Loan application form

4.33.
Amigo’s loan application form required borrowers to specify the purpose of the

loan and customers to provide details of their income and expenditure on specified

items in the Budget and to declare that these were complete and true.

4.34.
The application form did not ask customers about their household circumstances

until after the end of the Relevant Period. This meant that during the Relevant

Period, Amigo did not gather information about the composition of a customer’s

household and therefore its assessment of expenditure and customer

circumstances did not consider this. Amigo did gather information about the

number of children in the household in certain circumstances, but only used this

information in relation to the assessment of food costs and not other household

expenditure.

4.35.
Customers were able to modify the details they had entered in their Budget until

they submitted it. A banner at the bottom of the Budget page would show red if

the Budget meant the loan being applied for was unaffordable, turning to green

when the customer entered values which meant the loan would be affordable.

This meant customers were alerted to their affordability status and were able to

manipulate their income and expenditure entries, before submitting the Budget.

Amigo was aware that if used incorrectly this feature could lead to customers

undervaluing their expenditure items to get a loan. However, Amigo did not

remove this feature from the Budget until April 2020, after the Relevant Period.

4.36.
The Budget asked customers to provide details of their income and expenditure.

In relation to income, this included their employment status, types of income and

corresponding amounts. Types of income included earned income, overtime, self-

employment income and benefits income. Notably, overtime income was included

in Amigo’s assessment of affordability. However, there were no checks to

determine whether overtime income was regular or only occasional, which

presented a risk that irregular income was used to secure credit.

4.37.
Expenditure was divided into 2 categories: credit file expenses and household

outgoings. Credit file expenses were pre-populated from the credit agency check.

If a monthly repayment appeared on the credit search, this would be imported

into the application form. However, for defaulted credit accounts, Amigo

calculated the repayment amount for the Budget. This showed the amount Amigo

expected to be repaid and not necessarily the amount that customers would

actually pay against defaulted balances. Customers could also edit the figures if

they chose to. This feature, alongside the indicator of affordability built into the

Budget calculator, increased the risk that customers would underrepresent their

expenditure. This was particularly acute given that many of Amigo’s customers

were vulnerable due to financial stress.

4.38.
Household outgoings were requested from the customer for rent, food, gas,

electricity, water, travel, home insurance, media, childcare and clothing. These

were the only outgoings that Amigo identified as non-discretionary expenses.

4.39.
The questions on the application form in relation to expenditure did not create a

full or clear picture of the customer’s financial position in terms of affordability.

The form did not request details of all items of non-discretionary expenditure,

such as those required to meet essential living expenses and other expenditure

which is hard to reduce to give a basic quality of life, like healthcare costs, car

maintenance costs or household essentials beyond food.

4.40.
The application form also did not ask any questions about future known events

with a view to assessing reasonably foreseeable reductions in income or increases

in expenditure. This meant that Amigo could not accurately assess whether

repayments would be affordable over the life of the agreement.

4.41.
Amigo’s assessment of borrower affordability was based solely on assessment of

the Budget until November 2019, when Amigo introduced rules into its system to

identify any credit behaviours in the past 6 months which increased borrower

affordability risk. Customers were able to amend the Budget during the application

process and, as explored below, there was only very limited verification. Given

the issues set out above, there was a clear risk that Amigo could not rely on the

Budget to assess affordability.

The decision engine

4.42.
Once a customer filled in their details and Budget in the application form, Amigo’s

system would assess the application against lending parameters.

4.43.
The first stage of checks determined whether customers met basic lending criteria.

This was to confirm that:

a.
The borrower was a UK resident.

b.
The borrower was aged 18 to 75; and

c.
The borrower was not bankrupt, subject to an IVA or DMP.

During the Relevant Period, no minimum income was required for customers-

but the loan still had to be considered affordable.

4.44.
Failure to meet the eligibility criteria resulted in borrowers being declined without

proceeding to any affordability assessments.

4.45.
Once the customer Budget was submitted, Amigo’s system would compare the

data provided by customers on their application form against credit file

information and data sourced from third parties. For guarantors, the system would

look in detail at their credit file to assess creditworthiness. However, there were

no such checks for borrowers until November 2019. After this date, Amigo

introduced 8 “RAG rules” to identify any recent credit behaviours which increased

borrower affordability risk. Amigo’s system reviewed information from the

borrower’s credit file to consider factors including accounts in debt management,

arrears and limit utilisation on new accounts in order to categorise them as red,

amber or green risk. The intention was that red customers would be automatically

declined without an affordability assessment, amber customers would be subject

to a more detailed assessment to understand the customer’s situation in relation

to debt consolidation, consider whether additional verification was required and

establish whether Amigo was lending proportionally. In practice, both amber and

red customers were declined. Green customers followed the existing affordability

assessment based on the Budget.

Affordability assessment

4.46.
Amigo considered that a loan would be affordable for a customer if they had a net

disposable income once the expenditure named on their Budget, the RMI and a

buffer were deducted, represented by the following formula.

Income > (Expenditure (being non-discretionary expenditure and existing credit

commitments) + RMI + Buffer)

4.47.
Amigo’s affordability assessment was the same for borrowers and guarantors,

except for the level of buffer. The purpose of the buffer was to address

unexpected and unforeseeable expenses.

4.48.
If a customer submitted a Budget where the income was not sufficient to meet

the above formula, the lending was considered unaffordable to the customer and

the application would be declined.

4.49.
Alternatively, the application was flagged to an agent to contact the customer.

The aim of the agent’s contact was to see if there was a way for Amigo to still

lend to the customer whether that was through a loan for a lesser amount or for

a longer term.

Verification of income

4.50.
From 1 November 2018, Amigo introduced 5 categories of higher risk customers

where enhanced income verification checks were required to ensure the customer

could afford the loan. Amigo initially anticipated that these changes would affect

only between 5 – 7% of its customers for the enhanced check and mean that for

higher risk customers, the total income declared by the customer was verified.

However, for lower-risk customers (approximately 90% of Amigo’s loan book),

Amigo only validated whether the amount of income stated by the customer was

enough to cover their expenditure (which was non-discretionary expenditure plus

credit commitments), the RMI and the buffer.

4.51.
For the majority of the Relevant Period, Amigo sought to verify customers’

declared income using 1 of 3 methods for customers it assessed at lower risk:

a.
An income confidence score of 4 or above using credit reference agency

data.

b.
Income matrices; or

c.
Document proof.

4.52.
These methodologies were applied sequentially. If a customer’s income generated

a score of 4 or above using credit reference agency data, the loan moved forward

for approval. However, if it fell below 4, then the system would move on to check

the stated income against the income matrices. If that did not provide what Amigo

considered sufficient assurance, a task was created for an agent to request

documentary proof of income from the customer. If after the verification exercise,

insufficient income was verified for a customer to pass the affordability

assessment, a task was raised for an agent to call the customer. During that call

an agent could review the Budget with the customer in order to make the lending

affordable. This could involve the agent offering a lower loan amount or a longer

term and /or attempting to explore with the customer whether they could make

lifestyle changes to lower their expenses for the Budget to make the lending

affordable. If this call could not be completed, the application was declined.

4.53.
Where the customer was identified as higher risk, the process for verifying their

income was different. If their full income could not be verified using a score of 4

or higher, the customer had to be verified by document proof. It therefore skipped

the income matrices step.

Credit reference agency data checks

4.54.
Amigo used a credit reference agency data tool which analysed a customer’s

current account turnover data over a 12-month period and could provide

independent evidence of income. However, the threshold set by Amigo for

customers to pass this check was too low to provide independent evidence of

income.

4.55.
The check used generates a confidence figure in the level of stated income (the

confidence figure) and an indication of whether the income figure checked was an

under or over-statement as compared to the customer’s current account income

(the under/over statement). The confidence figure factor looks at a customer’s

application salary and current account and considers the age, source and volatility

of the income into the account as compared to the figure being checked. A

confidence figure score of 8 or 9 provides assurance of sole income.

4.56.
Where a customer had a confidence figure score of 7, this could include joint

income. The software used by Amigo did not exclude joint income.

4.57.
The Authority considers that it was unreasonable for Amigo to rely on a score of

7 or below as verification of income because such confidence figure scores can

include joint income. A score of 7 could only provide independent verification of

income if a firm undertakes further checks to ensure that partner income is

available and to take account of partner expenditure.

4.58.
Amigo did not use credit reference agency data checks appropriately in that it

relied on confidence figure scores that the Authority considers provided a low level

of assurance to justify lending to customers. Amigo verified income using a

confidence figure score of 4 or above. A confidence figure score of 4 would be

triggered where there were factors which undermined the confidence figure score

such as a large mortgage or a financial associate with a larger income. This could

indicate over-indebtedness or dependence on another person’s income. The

majority of customers with a confidence figure score of between 4 and 7 were

approved for loans without any interaction with agents to test the reliability of the

data that had been provided. This applied to 49.72% (108,832) of borrowers and

47.1% (103,295) guarantors for loans issued during the Relevant Period.

4.59.
The under/over statement check generated a red (high), amber (medium) or

green (low) rating to show whether the income figure checked was understated

(green or low) or overstated (amber/medium or red/high) when compared to the

customer’s current account data. This would have provided some assurance about

the reliability of the income declared by the customer and their ability to afford

the loan. However, the value of this check was undermined by the fact that Amigo

checked a lower figure than the customer’s income on the application form, just

the amount needed to cover outgoings, for the majority of its customers (see

paragraph 4.50 above). As a result, the under/over statement would be more

likely to generate a green or amber score but that showed sufficient income to

meet the costs in Amigo’s budget calculator.

4.60.
Where a customer had a red-rated (high) under/over statement score, this would

mean the income checked was overstated as compared to the customer’s current

account and it was Amigo’s practice in those circumstances to decline the

customer. Due to poor document and knowledge retention (due to lack of

adequate records historically), Amigo has been unable to confirm whether this

was the case throughout the Relevant Period. However, in 15 out of 37 loans

(40.5%) in the Authority’s file review customers had a red / high under/over

statement score but Amigo lent money instead of declining the customer. There

was a clear risk of customer harm in these instances because Amigo’s assessment

that these customers could afford to borrow was based on its Budget, which it

knew was likely to include an overstated income.

Income Matrices (for lower risk customers only)

4.61.
When the customer had a credit reference agency data check confidence figure

score of 1, 2 or 3 (indicating significant volatility in their accounts), employment

income would next be compared against ‘income matrices’. As noted in paragraph

4.53, income matrices were only used for low-risk customers. Amigo purchased

the income matrices from a third party, and they showed average salaries for

various job titles. These average national salaries were by role only and did not

allow for regional divergence.

4.62.
The Authority considers that it was unreasonable for Amigo to rely on income

matrices because they provided only nationalised salary figures for specific job

types, and they were not tailored to the borrower or guarantor. Income matrices

also relied on the accuracy of the job role entered by the customer from a pre-

populated list, and Amigo had no way of identifying if this had been entered

correctly or if the categories available were appropriate for the borrower or

guarantor in question.

4.63.
Amigo’s systems compared the amount of income declared by the customer

against the average figure contained for that role in the income matrices. If the

figure declared by the customer was more than 20% above the average figure,

the income was deemed not to have been verified by the income matrix. Any

amount below 120% of the average, nationalised salary figure was deemed

verified.

Benefit matrix

4.64.
Non-employment income (i.e., certain benefits and pensions (both state and

private) were compared to stated values against the ‘benefits matrix’. Like the

income matrix (see paragraph 4.62), the benefit matrix placed reliance on the

accuracy of information provided by the customer regarding the type and amount

of benefit income they received.

4.65.
Benefit income was counted as available funds for affordability purposes

irrespective of the purpose for which the benefit was given, provided it fell within

the expected threshold for that benefit. Except for child benefit, there was no

consideration given, in the application form or the affordability assessment, to

whether a benefit was intended for a specific purpose, except in circumstances

where the customer included child benefit income but no childcare cost, where

Amigo’s system would raise an agent task to explore this further. This would mean

that customers in receipt of benefits which also had associated costs would be

able to count the income from these benefits but not the associated expense

except for the circumstance outlined at paragraph 4.34 where the comparator for

food costs was increased where a customer disclosed child benefit income. This

could lead to overstated income and therefore lending being deemed affordable

by Amigo when it should not have been, resulting in irresponsible lending unless

they added it to their expenditure (e.g., travel costs).

Document proof

4.66.
If income was still unverified, pay-out agents were prompted to request document

proof from customers, for example, payslips or bank statements. Amigo specified

requirements for the documentary evidence in its Logic Manuals. These required

documents to be provided to prove income within the last 100 days.

4.67.
Once provided, agents would review the documents to confirm whether the

declared income could be verified for the purpose of affordability. If the

documents provided did not support the necessary amount of income, the

application should have been declined. However, Amigo’s policies allowed for

customers to self-certify income in some cases through a process called ‘manager

reconciliation’ where the underwriting manager reviewed the self-certification.

Concerns with Amigo continuing to take this approach were raised by its

Compliance function in July 2019, but it was not until February 2020 that Amigo

stopped accepting manager reconciliation as a way to verify income.

4.68.
Amigo’s outsourced Internal Audit function’s report on affordability in March 2019

made a finding that documentary evidence retention by Amigo to assess

affordability was insufficient. This was noted again in December 2019.

4.69.
This issue was a consistent feature of the review of customer files by the

Authority. In 10 out of 37 loan files, customers were recorded as having been

income-verified by document proof. This related to 7 borrowers and 4 guarantors.

In 3 out of 11 files no document proof was recorded on the file. The Authority

could not review whether income verification was adequate in those cases.

Top up loans

4.70.
Amigo pre-populated previous income details for further lending within 6 months

of the last income verification, although it gave the customer the option to change

the amount. In addition, the customer was presented with a tick box that asked

them to confirm that their income remained the same since their previous

application. The Authority considers that this approach was inappropriate because

Amigo did not carry out updated income verification checks at this time, in

circumstances where the customer’s income was stated to be the same or higher

than a previous application. This was of particular concern as a large proportion

of Amigo’s customers were financially vulnerable and may have fluctuating levels

of income and expenditure.

4.71.
Amigo sent borrowers a text or email to inform them when they became eligible

for a top up. Where a borrower had recently been in arrears or taken several top

ups with Amigo within a short period of time, the Authority considers that such

marketing by Amigo may not have paid due regard to the interests of those

customers and could have resulted in potential detriment.

4.72.
Analysis of Amigo’s lending over the Relevant Period by the Authority shows that

where borrowers topped up within 6 months of their original loan, they were more

likely to go into arrears or Default on the loan. Amigo made changes in July 2019

to its eligibility criteria for top-ups, increasing the number of payments that had

to be made to 6 before a customer would be eligible to top up. Amigo accepts that

it did not do enough to understand why repeat lending was happening or to

understand the potential detriment. An example of this can be seen in the case

study of Customer 2 in Annex B. That borrower took a total of 7 loans from

Amigo and although he was clearly in financial difficulty from the fifth loan, Amigo

continued to lend, contributing to the borrower’s spiralling debt.

Verification of expenditure

4.73.
Once the customer Budget was submitted, Amigo’s system would compare the

expenditure data provided by customers on their application forms against credit

file information and data sourced from third parties, including data from the Office

for National Statistics (“ONS”) relating to average household expenditure. The

calibration of tasks to be flagged was a key element of Amigo’s controls in relation

to the assessment of expenditure. Inconsistencies identified by Amigo’s system

would either:

a.
in relation household outgoings (verified by comparing the declared

amounts against ONS national averages), require a customer to select a

reason for that inconsistency from a pre-populated drop-down menu

which would typically be approved by the system without interrogation by

an agent; or

b.
in relation to credit file information where the customer had reduced any

pre-populated amount, raise a ‘flag’, requiring an agent to contact the

customer to address those issues.

4.74.
If no tasks were flagged, agents would not be required to speak to borrowers as

part of the affordability assessment.

4.75.
The Authority considers that Amigo did not take reasonable steps to determine

the amount, or make a reasonable estimate of a customer’s non-discretionary

expenditure for the following reasons:

a.
Save as noted at paragraph 4.34 above, Amigo did not consider the

composition of the customer’s household when checking customer

expenditure. This meant that the expenditure recorded by the customer

could have been too low, but this would not have been flagged by Amigo’s

system because it only considered expenditure per adult in an average

household. This was the case even when Amigo knew the household was

larger because the customer had declared they were in receipt of child

benefit income.

b.
In relation to the cost of food, if a customer entered zero for this cost,

Amigo’s system assumed the customer’s food cost would be £88 per

month. This figure was well below the ONS figure of £133.90 per month.

c.
Amigo failed to update the ONS data annually from at least 2016 until

2020 with the resulting cumulative effect of inflation meaning that the

comparative data relied upon by Amigo’s systems was undervalued. As

noted at paragraph 4.73, Amigo’s system would require an explanation to

be selected by a customer if their expenditure was below 80% of the ONS

average. The oversight by Amigo in updating its ONS data meant that an

explanation was not required until the customer data was 74% to 76%

below ONS, a considerable decrease on Amigo’s own policy.

4.76.
Documentary proof was rarely required for the purpose of verifying expenditure

from the beginning of the Relevant Period until at least 5 February 2020 and was

not sought for any loans issued during the Relevant Period.

4.77.
Amigo referred to its underwriters as pay-out agents. Pay-out agents were an

important element of the lending process. They were responsible for dealing with

tasks raised by Amigo’s system in respect of a loan application. They also reviewed

all loan applications before loans were paid out. Pay-out agents had the final say

on whether to approve a customer’s explanation of an affordability ‘flag’ or to re-

budget the relevant income or expenditure items.

4.78.
The pay-out agent role was entry-level with a high turnover of staff. Agents

received training on joining the firm that lasted around 3 weeks, this was known

as the Academy. This was supplemented by e-learning training for ongoing

general training needs. Training was also provided in certain key areas, for

example, how agents should examine documentary proof requested from

customers when assessing affordability.

4.79.
A buddy system existed for pay-out agents after they had completed the

Academy. Coaching (by coaches known as CXCs) and mentoring was available to

new agents. CXCs acted as Deputy Team Managers and were expected to dedicate

the majority of their time to training and coaching. In addition, Team Managers

were also expected to dedicate a significant proportion of their time to training

and coaching. That being said, agents still complained there was insufficient time

with coaches and that coaching did not focus on improvement but instead on

auditing mistakes. This should have raised concerns that pay-out agents needed

a greater level of support and guidance in their role.

4.80.
When Amigo’s decision engine could not approve (or underwrite) the loan from

an affordability perspective, pay-out agents would be prompted to contact a

borrower and/or guarantor. Applications were assigned to pay-out agents on a

round-robin basis and an agent was expected to deal with all tasks on an

application. The agent contacted the customer to address all issues/tasks raised

individually and if satisfied, approved the loan from an affordability perspective.

4.81.
Amigo’s tasks were recorded in a detailed and regularly updated “Logic Manual”.

This collated the procedures and considerations relevant to tasks generated by

the system. While the Logic Manuals provided details of the tasks that Amigo’s

systems might flag, these were not available to pay-out agents. Any changes

would be communicated by the Head of Department to the managers of each of

the underwriting/pay-out teams who would deliver training to their team of

agents.

4.82.
Pay-out agents were solely reliant on the details provided on the agent-facing

page of the application form, including whichever questions or directions were

provided on the screen as prompts to the agent for each specific tasks at the

relevant time during the Relevant Period. Agents were given limited guidance by

these on-screen details. Often, the prompts did not clearly direct agents on what

to do where they had concerns about the customer’s explanation. The wording

prompted agents to test information provided by a customer to find a way to lend,

rather than test the customer’s affordability.

4.83.
During the Relevant Period, there were a total of 39 income or expenditure tasks

that could be raised. These tasks raised by the system were added or removed in

an inconsistent manner across the Relevant Period. Some tasks disappeared and

reappeared from month to month. For example, the task "£0 for travel but does

travel sometimes” would be triggered only if the customer input zero for travel

and this task was not in place in April 2019, June 2019, August 2019 and October

2019 but was in place for the rest of the Relevant Period.

4.84.
The guidance in the Logic Manuals on these tasks only required an agent to probe

whether the figure was “probable, believable, realistic and sustainable”. For

example, the task "childcare expense contradicts their income” raised the

question "People in your area normally spend £X on childcare. We need to

understand why you pay less than that.” There was nothing further on this point

in the Logic Manuals. There was limited guidance as to the sorts of questions an

agent should be asking or relevant factors for consideration. This required a

significant level of judgement from the pay-out agent which the policies and

processes within Amigo’s system did not support them to make.

4.85.
Amigo viewed pay-out agents as less skilled than collections agents. Amigo’s

rationale for this was that the pay-out agent role was task-based (i.e. responsive

to known tasks identified by Amigo’s logic) whereas the collections role had to

adapt to unknown issues raised by customers and was therefore more difficult,

requiring additional skill and knowledge.

4.86.
Issues in relation to the narrow, task-based focus of agents and associated lack

of probing by agents of information provided by customers was raised by

Compliance in September 2019. On review of 10 customer files, Compliance found

that in 8 cases, documents revealed information which was not adequately

queried and, as a result, the loan was granted on the basis of incomplete

information which could have led to a different lending outcome had the full

picture been known.

4.87.
This is consistent with the Authority’s file review which found an unacceptable lack

of questioning by agents of information provided by customers. This was

particularly serious in instances where the information provided by the customer

clearly contradicted the loan application form. For instance, in customer file 3

(detailed in Annex B), the borrower mentioned in a call that they had 3 children

but only 2 were included on their loan application form.

4.88.
In addition to this, the level of automation of Amigo’s system meant that Amigo

was less able to evidence why a loan was considered to be affordable. A third-

party review commissioned by Amigo found that this was due to a reduction in

the number of questions that agents were asking about anomalies in customer

affordability assessments.

4.89.
Amigo pay-out agents approved an average total of around 12,800 loans per

month. The average size of the team responsible for approving this many loans

during the Relevant Period was around 10 people per team working within 4 main

pay-out teams. These figures take no account of the number of declines following

pay-out agent interaction. In order to deal with this number of applications, the

pay-out agents were task-focused, responding to triggers in isolation as dictated

by the IT system, and did not generally look at all the circumstances of

applications. This was not something they were trained to do or would have had

time to do. The need for enhanced skills at agent level was raised by Amigo’s

Compliance function in September 2019. Despite this, there was limited change

in the number of pay-out agents during the Relevant Period meaning that this

issue persisted throughout.

4.90.
The Authority considers that the way pay-out agents were rewarded during the

Relevant Period was unlikely to drive the right behaviours. Pay-out agents

received pay that could vary considerably based on performance. This did appear

to change in the Relevant Period, however until around mid-2019 a large part of

agent pay was dependent on how they ranked against targets and their peers on

a rolling 3-month basis. A significant part of the ranking related to volume,

meaning that pay-out agent’s variable pay was heavily weighted towards ‘loans

paid out’.

Impact on guarantors

4.91.
Guarantors were entitled to rely on Amigo’s assessment that a loan was affordable

for a borrower and undertook to make payments only where the borrower was

not able to. Amigo’s promotional material echoed this and stated that it would

“make sure the repayments are affordable” for the borrower. Due to the failings

outlined above at paragraphs 4.30 to 4.90, Amigo did not do this and accordingly

exposed guarantors to a much greater risk of having to make payments on the

borrower’s loan than they may have envisaged.

30

4.92.
Guarantors were always telephoned as part of the lending process prior to

payment. The pre-pay-out call would largely follow a script and its purpose was

to assess that the prospective guarantor understood their obligations as a

guarantor and was comfortable for the loan to be paid out. In November 2019,

the Authority, as part of its Guarantor Understanding Multi Firm Work, raised

concerns that Amigo may not be providing an adequate explanation to the

potential guarantor to allow them to understand the key associated risks to be

able to make an informed decision on being a guarantor. Changes to address this

issue were not made until July 2020, in co-ordination with the rest of the industry.

4.93.
The Authority raised concerns with the wider industry, including Amigo, that firms

did not adequately consider the necessary level and extent of information

provided to the guarantor. The Authority highlighted that:

a.
An explanation of whether the borrower can afford the loan including

information about the borrower’s monthly disposable income.

b.
This could include an explanation of the borrower’s credit score, missed

payments by the borrower on other debts in the last 12 months or recent

CCJs.

Amigo was concerned that providing this information to the guarantor could lead

to data protection issues.

4.94.
The absence of that information and the inadequacy of Amigo’s affordability

assessment meant that guarantors were exposed to a higher risk of making

payments than they expected.

4.95.
Amigo’s checks around creditworthiness were much more focused on the

guarantor than the borrower. A longer-term view of the guarantor’s credit file was

taken by Amigo as compared to the limited, short-term focus on borrowers

because Amigo knew that borrowers had poor credit files.

4.96.
Increasing the pool of potential guarantors was a key element of Amigo’s business

growth. It did this through its decision engine which assessed in detail, through

sequential questions, whether a potential guarantor met Amigo’s creditworthiness

criteria. Amigo knew that borrowers had poor credit files and only assessed them

on the Budget to determine if they could afford a loan.

4.97.
The disparity in approach by Amigo suggests to the Authority that Amigo was

much more concerned about the financial position of the guarantor than the

borrower when making the decision to lend. It was only if the guarantor defaulted

that Amigo stood to lose money. This mindset fed into Amigo’s approach to

lending and resulted in 1 in 4 guarantors being asked to make a payment for a

borrower during the loan term in relation to loans issued during the Relevant

Period. Although payments by guarantors accounted for 10% of total loan

payments.

Complaints and root cause analysis

4.98.
During the Relevant Period, Amigo saw a significant increase in complaints,

particularly about the affordability of its loans. In April 2019, the FOS noted that

it had seen complaints against Amigo triple, from what had been a low base. By

mid-2019, the volume of complaints against Amigo had increased substantially.

The increase in complaints and the increase in published decisions by the FOS

against Amigo after 31 May 2019 was accompanied by an increase in CMC

involvement in complaints, and a backlog of complaints. By the middle of 2020,

CMC cases made up the majority of new cases that Amigo was receiving.

4.99.
Amigo pointed to its low historic volumes of complaints and the low proportion

and number of upheld FOS decisions in the preceding years to suggest there was

no historic issue with irresponsible lending and its assessment of affordability was

adequate. However, as noted in the Authority’s thematic review of Complaint

Handling TR14/18, quality metrics are an important element of complaints

monitoring management information. The Authority’s Consultation Paper CP14/30

on Improving Complaints handling outlined that the reasons, or root causes of

complaints, as well as the volume of complaints, are important metrics for

analysing fair treatment of customers.

4.100. For that reason, it is always important to also look at the reasons for complaints

and carry out root cause analysis to understand if an issue flagged in one

complaint indicates a systemic problem.

4.101. Although a low proportion and number of FOS decisions had been upheld against

Amigo before May 2019, a clear theme around insufficient checks being completed

by Amigo could be seen in the complaints received. By the start of the Relevant

Period, Amigo had received a small number of FOS decisions which contained

comments on Amigo’s affordability assessment processes. These decisions should

have drawn Amigo’s attention to deficiencies in its affordability assessment

processes, particularly around its income and expenditure checks.

4.102. In June 2019, Amigo’s Internal Audit function issued a report on its first in-depth

internal audit review of complaints handling. The report was classified as “critical”

and identified serious weaknesses in both the design and operation of controls

with Amigo’s complaints handling. There were issues around complaints logging

and complaints numbers being underreported and a lack of audit trail of

complaints decisions. The report also noted that Amigo had no complaint handling

QA framework and practices contrary to the Authority’s requirements for

complaints handling in DISP.

4.103. The report found that Amigo’s system did not allow for root causes of complaints

to be captured. This meant that Amigo could not produce MI with the result that

senior management did not have an accurate view of causes for complaints and

inadequate assurance that root causes had been addressed.

4.104. Findings from FOS decisions were also not being effectively fed back as learning

into the business while Amigo sought advice and assurance from third parties on

the potential implications for its business. In June 2019, Amigo’s Compliance

function noted that root cause analysis for themes from FOS decisions was not

adequate, and identification of themes across complaints had come from the FOS

rather than internally. That report also recommended that Amigo’s approach to

lending and its general approach to the verification of customers circumstances

should be examined.

4.105. In July 2019, Amigo set up an RCA team. However, the approach to root cause

analysis and the work done by the RCA team was not adequate. Consequently,

the lessons learned were limited and they were not fed back into the business as

quickly as required. These issues were not addressed because senior management

considered that they were not systemic as they did not affect a particular cohort

of borrower. However, the issues revealed clear failings across Amigo’s systems

and controls which should have been addressed.

4.106. By late 2019, complaints volumes were still increasing, particularly for

irresponsible lending. On 1 November 2019, Amigo stopped processing complaints

until it could determine new affordability rules which could be applied back to

complaint cases. At this point, Amigo decided to re-evaluate its approach to

lending, while it considered the FOS decisions, but it continued to lend.

4.107. In December 2019, Amigo’s Risk Committee noted that it would be more

appropriate for the complaints policy to be used to revise the lending policy, rather

than vice versa. Amigo also accepted at that point that sometimes there was little

or no documentary evidence to confirm the information on which it relied in order

to make the lending decision.

4.108. In January 2020, Amigo adopted a different approach to complaints handling. This

approach diverged further from FOS decisions and was not aligned with customer

outcomes. Complaint handlers believed the new approach was to reject all

irresponsible lending complaints in breach of DISP 1.4.1R (2) which required firms

to investigate complaints competently, diligently, impartially and fairly, obtaining

additional evidence as necessary. The irresponsible lending complaints uphold

rate reduced to 2% following a shift in approach to defend complaints where

Amigo’s Budget calculator had assessed the loan as affordable to the customer

and the customer had provided evidence that the information had been

incomplete or incorrect. This was a clear failing by Amigo to pay proper regard to

the interests of its customers and to treat them fairly.

Governance and oversight

4.109. Overall responsibility for affordability within Amigo was unclear. Neither the Board

nor any of the Committees were expressly responsible for affordability or

creditworthiness. Historically, the Compliance function had responsibility for these

issues and there was an intention to move the focus on customer outcomes into

Operations, but this did not occur. Responsibility for credit risk was clearly defined

and sat with the Data Science Team throughout.

4.110. There were significant management and leadership changes within the business

across the Relevant Period. The various changes in management marked

significant shifts in approach towards affordability and customer focus. In mid-

2019, Amigo was seeking to change its culture having identified that the previous

culture and past practices could have encouraged bad agent behaviours and poor

customer outcomes. However, irresponsible lending complaints continued to rise

(see paragraph 4.98 above).

4.111. Amigo’s approach to affordability was reactive, whether as a result of FOS

decisions or the Authority’s interventions. However, Amigo did make changes to

its top up lending approach before the Authority intervened. In certain instances,

proactive management of affordability issues could have ensured lessons were

both learned and applied to lending practices sooner.

4.112. Amigo carried out a review of its systems and processes prior to the introduction

of the affordability rules in November 2018. This review was narrow and did not

adequately consider Amigo’s affordability processes ahead of the clarification of

the CONC rules on 1 November 2018. As mentioned in paragraph 4.50 above,

Amigo made minimal changes to its affordability assessment (and related

processes) at that time as it believed it was already compliant.

4.113. The Compliance function also failed to complete its planned review of affordability,

which was intended, among other things, to review irresponsible lending

complaints and review training and guidance available to agents concerning

customer affordability. The Board did not chase for the outstanding work product.

4.114. A significant internal review of affordability was completed in September 2019.

This review led to the introduction of the RAG rules in November 2019 that, as

mentioned in paragraph 4.45, considered indicators on the customer’s credit file

to suggest an increased affordability risk in the prior 6 months. This was the first

time Amigo considered the borrower’s recent credit history. This change was

reactive and driven by FOS decisions on complaints.

4.115. Even at this time, Amigo’s new lending parameters were significantly influenced

by commercial factors. The rules adopted did not reflect the recommendations of

its external auditors who suggested more evidence could be gathered in support

of lending decisions around income and the investigation of expenses. The

external auditors also specifically advised that Amigo should assess a potential

customer’s bad debt performance over 12 months rather than 6. Management did

not explain the reasons for this divergence; instead, it noted the RAG rules would

“likely need to be adjusted over time”. They did not focus on customer outcomes

or the risk of customer harm.

4.116. Amigo commissioned several external reviews though none of these reviewed the

parameters within its lending system. The reviews identified that:

a.
affordability assessments were inaccurate and did not take adequate

consideration of expenditure or income.

b.
red flags about indebtedness were not being adequately investigated; and

c.
Amigo’s horizon scanning failed to recognise emerging trends in relation

to affordability issues.

4.117. Each of these issues is serious and pointed to clear failings in Amigo’s systems

and controls in relation to creditworthiness and affordability assessments. There

was insufficient action taken by senior management in relation to these findings

to mitigate the risk of potential harm to customers prior to the end of the Relevant

4.118. Amigo’s MI on affordability was deficient. Amigo used the guarantor payment rate

as one of the indicators that loans were affordable. This looked at the payments

by guarantors in a given month, as a percentage of all loan payments. However,

it did not consider circumstances where the borrower took a top up loan. Where

a guarantor paid off the remaining balance of the loan, this appears to have been

treated as a single payment. This narrow view did not provide enough insight into

affordability for senior management to understand whether lending was affordable

for the borrower.

4.119. This issue was made worse by the lack of MI around root cause analysis of

complaints which could have identified key issues around affordability

assessments. As noted at paragraph 4.103, there was no MI on the root causes

of complaints, which meant that senior management was not able to see trends

or address them.

4.120. No specific KRI was developed in relation to affordability. After receiving the

Affordability Review in September 2019, the Board asked management to review

the MI used to monitor affordability and develop a KRI looking specifically at

affordability in addition to Amigo’s existing metrics. Amigo’s MI metrics were too

narrowly focused on credit risk and did not adequately consider customer

detriment. Affordability risk was not prioritised either by the Risk function, or input

was not provided to Risk from other parts of the business.

Amigo’s policies and procedures

4.121. Amigo’s policies and procedures in relation to affordability were not adequate. No

specific affordability policy was created either in readiness for the CONC rules

introduced on 1 November 2018, or during the Relevant Period. A draft

36

Affordability & Creditworthiness Policy was prepared by 29 August 2019;

however, this was never approved.

4.122. Amigo’s overarching framework for lending was its Responsible Lending Policy.

This required separate creditworthiness assessments of borrowers and

guarantors, taking account of credit risk (for guarantors) and affordability (for

borrowers and guarantors). The policy provided a framework for creditworthiness

assessment reflecting the CONC rules. However, the detailed implementation of

that framework into practices and procedures for the assessment of affordability

and creditworthiness at a customer level was inadequate.

4.123. There were clear issues with how Amigo translated and implemented its

Responsible Lending Policy into practice. Specifically:

a.
in relation to sustainability, Amigo’s process did not appropriately consider

the sustainability of income or expenditure. No direct questions about

future events were put to customers and Amigo only considered

sustainability through a confirmation box at the end of its online Budget.

b.
Amigo’s policy required it to consider (i) the potential adverse impact on

the customer’s financial situation of any loan commitments and (ii) the

customer’s ability to make repayments as they fell due when carrying out

the creditworthiness assessment. Amigo’s consideration of the borrower’s

financial circumstances was very limited until November 2019 and even

then, only the recent adverse credit information was considered.

4.124. The Responsible Lending Policy did not set out the specific processes or criteria

for the creditworthiness and affordability assessment. These were set out in the

Underwriting Logic Manual. As mentioned in paragraph 4.81 above, the Logic

Manuals provided details of specific tasks and some instructions, but they did not

contain the criteria or parameters used for the creditworthiness and affordability

assessment. These were hard-coded as parameters into the IT system’s logic by

Amigo’s engineering department – these have not been shared, nor have they

been independently tested by any third party. From a governance perspective, it

is unclear to the Authority who had oversight of this and was checking that the

system (a) operated in line with the Logic Manuals and (b) operated in accordance

with the Responsible Lending Policy.

4.125. Amigo’s policies were to be reviewed annually with material changes to be

approved and policies checked to ensure that the business was managed in

accordance with those policies. Amigo failed to follow its own process and conduct

annual reviews of its policies and reviews which evidence its procedures were

compliant with its policies.

3 lines of defence

4.126. Amigo did not apply an appropriate 3 lines of defence risk management framework

to its business. As a result, it failed to put in place effective monitoring and

controls capable of managing and mitigating risk of customer harm due to

inadequate assessment of affordability.

4.127. The 3 lines of defence risk management framework is a model for governing the

lines of responsibility for risk management and control activities. The model

stipulates that the first line of defence in terms of risk management is the

commercial business, in Amigo’s case this became the QA teams in Operations.

The second line of defence is usually a committee or business area that is not

engaged in commercial sales and instead helps develop and monitor the first line

of defence risk controls. In this case, it was the Compliance function. The third

line of defence sits above the other 2 lines and should provide independent review,

challenge and assurance on the effectiveness of both. This was an outsourced

Internal Audit function.

First and Second lines of defence

4.128. Until at least July 2019, in respect of monitoring, Amigo operated a 3 lines of

defence framework with a hybrid of the first and second line of defence. The QA

activities that were first line activities were conducted by the second line. This

blurred the distinction between these lines of defence so there was no

independent checking by the second line of defence. In practice, Amigo’s quality

assurance exercise was essentially call listening to check whether (i) internal

policies had been followed or (ii) a customer’s experience on the call was poor.

Although the QA form contained questions looking at whether the affordability of

the loan to the customer had been appropriately assessed, in practice the review

was deficient.

4.129. Concerns were raised in June 2019 about the lack of an effective risk management

framework. This also highlighted a lack of deep dive thematic reviews. A new 3

lines of defence structure was proposed with a new Compliance Quality Assurance

team as an independent line of monitoring with a significantly increased remit.

38

Senior management approved the proposal and accepted that there had been a

lack of investment in the second line defence capabilities which had hindered

objective challenge of decision making.

4.130. Internal Audit also reported in October 2019 on the proposals to restructure the

Compliance function. The report noted serious flaws in Amigo’s Compliance

function and supported Amigo’s proposals for change. The report noted:

a.
deficiencies in knowledge and skill in the function and the need for

external compliance specialist hires to address skill gaps within the

department; and

b.
that there was a clear lack of compliance advice and horizon scanning

responsibility in the function.

4.131. Although Amigo took steps to improve its three 3 lines of defence model, it

remained unsophisticated and was not embedded within the business by the end

of the Relevant Period with issues highlighted by Internal Audit around risk

identification, analysis and monitoring resulting in potentially sub-optimal

management of risk and the lack of a clear escalation process outside normal

Board activities.

Document management and knowledge retention

4.132. Under SYSC 9.1.1R, firms are required to maintain orderly records to enable the

Authority to monitor the firm’s compliance with regulatory requirements. This

obligation is particularly important in the context of an ongoing Enforcement

investigation by the Authority.

4.133. The Notice of Appointment of Investigators sent to Amigo at the outset of the

investigation in May 2020 highlighted the importance of preserving documents

relevant to the Authority’s investigation. The importance of this requirement was

acknowledged by Amigo’s senior management at the outset of the investigation

by the Authority. Despite this, Amigo did not maintain adequate records and on

some occasions was unable to provide adequate responses to questions by the

Authority (due to a lack of adequate records historically).

4.134. In addition to this, Amigo failed to retain staff email accounts after key staff

members had left the firm. This is a serious failing. This was after Amigo had been

notified of the Authority’s investigation and the suspected breaches. Although the

Authority has no reason to believe that this action by Amigo was deliberate, the

oversight was unacceptable and hampered the Authority’s ability to investigate

certain matters.

4.135. Amigo’s poor retention of documents has been considered as an aggravating

factor for the calculation of penalty (see paragraph 6.14).

Customer redress

4.136. On 23 May 2022 the High Court approved Amigo’s scheme of arrangement (the

“Scheme”) to provide redress to customers who were mis-sold loans and who

raised a claim by 26 November 2022. The Authority made clear to Amigo that it

needed to provide the best possible outcome for customers in seeking to limit its

liabilities through the Scheme. This would need to include the maximum amount

of funding possible to meet compensation claims by customers. Amigo’s Scheme

is intended to provide creditors with appropriate redress and more than they

would receive if the firm was to go into administration.

5. FAILINGS

5.1.
The regulatory provisions relevant to this Notice are referred to in Annex A.

5.2.
Principle 2 requires a firm to conduct its business with due skill, care and diligence.

5.3.
Amigo failed to conduct its business with due skill, care and diligence, in that it:

a)
Failed to act appropriately on the findings of a number of internal and

external reviews which had identified weaknesses in its approach to the

assessment of affordability and creditworthiness; and

b) Failed to retain documents and adequately record its historic processes

relating to its approach to affordability and creditworthiness. This included

negligently deleting the email accounts of relevant staff after the

Authority’s investigation commenced. This was despite the Board

acknowledging the Authority’s instructions regarding the importance of

retaining documents for the Authority’s investigation. As a consequence,

(1) Was unable to fully comply with information requirements issued

by the Authority; and

(2) Failed to retain corporate knowledge relating to important

elements of its historic business, systems and controls.

5.4.
Principle 3 requires a firm to take reasonable care to organise and control its

affairs responsibly and effectively, with adequate risk management systems.

5.5.
Amigo failed to take reasonable care to organise and control its affairs responsibly

and effectively because it failed to:

a)
Establish clear lines of responsibility and oversight for ensuring that

affordability was adequately assessed prior to lending.

b) Adequately record, monitor or review regulatory or operational risks

arising from its approach to affordability.

c)
Carry out adequate root cause analysis on affordability complaints, and

therefore failed to take reasonable steps to ensure that in handing

complaints it identified and remedied any recurring or systemic problems,

as required by DISP 1.3.3R.

d) Produce clear MI relating to the causes of affordability complaints, such

that its senior management could not effectively identify recurring or

systemic problems.

e)
Failed to ensure that its pay-out agents were dealing appropriately with

flags and triggers raised by the automated system.

5.6.
Principle 6 requires that a firm pay due regard to the interests of its customers

and treat them fairly.

5.7.
Amigo failed to undertake reasonable creditworthiness assessments and have

proper regard to the outcome of those assessments in respect of affordability risk,

as required by CONC 5.2A.5R. It also failed to determine, or reasonably estimate,

the income and expenditure of borrowers, as required by CONC 5.2A.15R and

5.2A.17R. In particular:

a)
Amigo unduly relied on information provided by customers in the Budget.

The Budget indicated to customers whether they had entered the required

income and expenditure to be approved for a loan and allowed them to

adjust the entered income and expenditure accordingly.

b) Amigo’s income verification processes were insufficient in that they

accepted unreasonably low credit reference agency scores as evidence of

income for the assessment of affordability of lending to a customer. They

also relied on nationalised average salary data as a basis for approving

large volumes of loans.

c)
Amigo failed to assess the full range of non-discretionary expenditure or

adequately consider the composition of a borrower’s household. It did not

seek documentary proof of expenditure from customers regardless of the

circumstances, and its only verification method relied on out-of-date ONS

data.

d) Amigo failed throughout the Relevant Period to gather information about

future known events which might have impacted the borrower’s ability to

make repayments.

5.8.
Amigo failed to pay regard to the interests of its guarantors by representing to

them on its website that when assessing borrowers, it would “make sure the

repayments are affordable”. It failed to do this, and 1 in 4 guarantors were asked

to make a payment for a borrower in relation to loans issued during the Relevant

Period.

5.9.
Amigo failed to pay due regard to borrowers who had recently been in arrears or

who had taken repeated top up loans. Although customers in arrears could not

apply for a top up, Amigo encouraged customers not in arrears to apply for

additional lending by text message or email often within a month or two of the

previous lending without having sufficient regard for the potential impact of

increased lending on their financial circumstances.

5.10.
Amigo’s culture sometimes prioritised commercial results at the expense of

customer outcomes. Its pay and reward system for pay-out agents was heavily

weighted towards loans paid out, which risked incentivising the wrong behaviours.

5.11.
Amigo failed to identify potentially vulnerable customers at the application stage

because it had an unreasonably narrow definition of vulnerability within its

system, limited to customers on incapacity benefit for a mental health issue, which

Amigo considered might affect the customer’s understanding or management of

the loan agreement. The issues with the automated system’s identification of

vulnerability were particularly serious given approximately 50% of Amigo’s

borrowers were approved for lending without any human interaction. Human

interaction provided a further opportunity for vulnerability to be identified. In

circumstances where such a high proportion of customers did not interact with a

human agent, the onus on the ability of the automated system to identify

vulnerability was significantly increased meaning that Amigo had extremely

limited opportunity to identify obvious signs of vulnerability prior to agreeing to

lend.

5.12.
Amigo also failed to pay due regard to the interests of consumers and treat them

fairly in relation to complaints about its affordability process. In November 2019,

complaint handlers stopped upholding irresponsible lending complaints for

approximately 2 months while Amigo assessed the situation. Amigo therefore

failed to investigate affordability complaints competently, diligently and

impartially, as required by DISP 1.4.1R.

6. SANCTION

6.1.
The Authority’s policy for imposing a financial penalty is set out in Chapter 6 of

DEPP. In respect of conduct occurring on or after 6 March 2010, the Authority

applies a five-step framework to determine the appropriate level of financial

penalty. DEPP 6.5A sets out the details of the five-step framework that applies in

respect of financial penalties imposed on firms.

Step 1: disgorgement

6.2.
Pursuant to DEPP 6.5A.1G, at Step 1 the Authority seeks to deprive a firm of the

financial benefit derived directly from the breach where it is practicable to quantify

this.

6.3.
The Authority has not identified any financial benefit that Amigo derived directly

from its breach.

6.4.
The Step 1 figure is therefore £0.

Step 2: the seriousness of the breach

6.5.
Pursuant to DEPP 6.5A.2G, at Step 2 the Authority determines a figure that

reflects the seriousness of the breach. Where the amount of revenue generated

by a firm from a particular product line or business area is indicative of the harm

or potential harm that its breach may cause, that figure will be based on a

percentage of the firm’s revenue from the relevant products or business area.

6.6.
The Authority considers that the revenue generated by Amigo is indicative of the

harm or potential harm caused by its breach. The Authority has therefore

determined a figure based on a percentage of Amigo’s relevant revenue. Amigo’s

relevant revenue is the revenue derived by Amigo during the period of the breach.

The period of Amigo’s breach was from 1 November 2018 to 31 March 2020. The

Authority considers Amigo’s relevant revenue for this period to be £405,000,000.

6.7.
In deciding on the percentage of the relevant revenue that forms the basis of the

step 2 figure, the Authority considers the seriousness of the breach and chooses

a percentage between 0% and 20%. This range is divided into five fixed levels

which represent, on a sliding scale, the seriousness of the breach; the more

serious the breach, the higher the level. For penalties imposed on firms there are

the following five levels:

Level 1 – 0%

Level 2 – 5%

Level 3 – 10%

Level 5 – 20%

6.8.
In assessing the seriousness level, the Authority takes into account various factors

which reflect the impact and nature of the breach, and whether it was committed

deliberately or recklessly. The factors that the Authority considers to be relevant

to Amigo’s breaches are set out below:

The impact of the breaches

i.
There was loss or risk of loss caused to individual consumers caused by

Amigo’s breaches. Amigo failed to appropriately consider customer

interests and mitigate the risk of customer harm through lending that

would be unaffordable to the customer. (DEPP 6.5A. 2G(6)(c)).

ii.
The breaches potentially had an effect on particularly vulnerable people,

whether intentionally or otherwise. A significant proportion of Amigo’s

customer base had low financial resilience and many displayed aspects of

vulnerability. Amigo failed to ensure that it identified potentially

vulnerable customers. (DEPP 6.5A. 2G(6)(d)).

iii.
The breaches caused inconvenience or distress to consumers. Amigo’s

inadequate affordability assessment failed to ensure that the borrower

would be able to repay the loan affordably without it impacting their wider

financial situation, thereby causing inconvenience or distress to its

borrowers, many of whom were financially vulnerable and often reliant on

benefit income. In some cases, this was exacerbated by top up loans,

which had the effect of entrenching the borrower in ongoing debt. This in

turn led to an increased risk that guarantors would have to step in and

make payments.

Guarantors were reliant on Amigo’s assessment that a loan was affordable for a

borrower and undertook to make payments only where the borrower was not able

to. Amigo’s promotional material echoed this and stated that the firm would “make

sure the repayments are affordable” for the borrower. Amigo did not do this and

accordingly exposed guarantors to a much greater risk of having to make

payments on the borrower’s loan than they would have envisaged. (DEPP 6.5A.

2G(6)(e)).

The nature of the breaches

i.
The nature of the rules, requirements or provisions breached. The

requirements of CONC should be a fundamental consideration in

everything that those who are authorised to provide consumer credit do.

Amigo’s lack of compliance with the rules around creditworthiness falls

below the standard expected of the industry (DEPP 6.5A. 2G(7)(a)).

ii.
The frequency of the breaches. The breaches occurred throughout the

Relevant Period and affected a significant number of the customers that

Amigo dealt with (DEPP 6.5A. 2G(7)(b)).

iii. The breaches revealed serious or systemic weaknesses in the Amigo’s

procedures and in the management systems and internal controls relating

to the firm’s business (DEPP 6.5A. 2G(7)(c)).

iv. In committing the breaches, the firm failed to take adequate steps to

comply with FCA rules. The market and regulations concerning

affordability both evolved, but Amigo failed to keep pace. Amigo’s horizon

scanning failed to recognise emerging trends in this regard. This was

compounded by Amigo’s failure to take lessons from the root cause

analysis for the irresponsible lending complaints against the firm (DEPP

6.5A. 2G(7)(h)).

6.9.
DEPP 6.5A.2G(11) lists factors likely to be considered ‘level 4 or 5 factors’. Of

these, the Authority considers the following factors to be relevant to the impact

and the nature of the breaches:

i.
The breaches caused a risk of loss to individual consumers (DEPP

ii.
The breaches revealed serious or systemic weaknesses in the firm’s

procedures or in the management systems or internal controls relating to

all or part of the firm’s business (DEPP 6.5A.2G(11)(b)).

6.10.
DEPP 6.5A.2G(12) lists factors likely to be considered ‘level 1, 2 or 3 factors’. Of

these, the Authority considers the following factors to be relevant:

i.
The breaches were committed negligently or inadvertently (DEPP

6.11.
Taking all of these factors into account, the Authority considers the seriousness

of the breach to be level 4 and so the Step 2 figure is 15% of £405,000,000.

6.12.
The Step 2 figure is therefore £60,750,000.

Step 3: mitigating and aggravating factors

6.13.
Pursuant to DEPP 6.5A.3G, at Step 3 the Authority may increase or decrease the

amount of the financial penalty arrived at after Step 2, but not including any

amount to be disgorged as set out in Step 1, to take into account factors which

aggravate or mitigate the breach.

6.14.
The Authority considers that the following factors aggravate the breach:

i.
Amigo failed to respond effectively to the FCA’s published amended rules

and guidance in PS18/19, in November 2018, and related public

announcements for improvements in standards in relation to behaviour

constituting the breaches. Amigo failed to ensure that its approach to

affordability evolved in line with changes in regulation and market

approach. Its horizon scanning failed to recognise emerging trends and

adapt its lending approach to ensure that it was lending affordably. This

was compounded by Amigo’s failure to take lessons from the root cause

analysis for the irresponsible lending complaints against the firm (DEPP

6.5A.3 G (k) and (l)).

ii.
Amigo took too long to consider and adapt to the decisions against it by

the Financial Ombudsman Service during a key period of time in the

Relevant Period, with a negative impact on its customers (DEPP 6.5A.3 G

(b)).

iii.
Amigo failed to fully co-operate with the Authority’s investigation by its

inability to provide certain information and documents, as a result of poor

document retention practices and a lack of adequate records historically.

Further, e-mail accounts of relevant personnel were deleted after the

commencement of the Authority’s Enforcement action despite instructions

not to do so. (DEPP 6.5A.3 G (b)).

6.15.
Amigo cooperated with the Authority’s investigation by providing open admissions

of various failings relating to its historic lending practices (DEPP 6.5A.3 G (b)).

There are no other mitigating factors.

6.16.
Having taken into account these aggravating and mitigating factors, the Authority

considers that the Step 2 figure should be increased by 20%.

6.17.
The Step 3 figure is therefore £72,900,000.

Step 4: adjustment for deterrence

6.18.
Pursuant to DEPP 6.5A.4G, if the Authority considers the figure arrived at after

Step 3 is insufficient to deter the firm who committed the breach, or others, from

committing further or similar breaches, then the Authority may increase the

penalty.

6.19.
The Authority considers that the Step 3 figure of £72,900,000 represents a

sufficient deterrent to Amigo and others, and so has not increased the penalty at

Step 4.

6.20.
The Step 4 figure is therefore £72,900,000.

Serious financial hardship

6.21.
Pursuant to DEPP 6.5D.4G, the Authority will consider reducing the amount of a

penalty if a firm will suffer serious financial hardship as a result of having to pay

the entire penalty. In deciding whether it is appropriate to reduce the penalty, the

Authority will take into consideration the firm’s financial circumstances, including

whether the penalty would render the firm insolvent or threaten the firm’s

solvency. The Authority will also take into account its statutory objectives.

6.22.
On 23 May 2022, the High Court sanctioned a scheme of arrangement (the

"Scheme") which aims to provide redress to Amigo's customers. The majority of

eligible customers who voted on Amigo's proposals voted in favour of the Scheme.

Amigo has provided verifiable evidence of its financial position and the Authority

is satisfied that the imposition of a financial penalty would threaten Amigo's

solvency and its obligations under the Scheme. Were it not for Amigo's financial

position the Authority would have imposed a financial penalty of £72,900,000.

The Authority has therefore decided to reduce the penalty to £nil.

Step 5: settlement discount

6.23.
Pursuant to DEPP 6.5A.5G, if the Authority and the firm on whom a penalty is to

be imposed agree the amount of the financial penalty and other terms, DEPP 6.7

provides that the amount of the financial penalty which might otherwise have

been payable will be reduced to reflect the stage at which the Authority and the

firm reached agreement. The settlement discount does not apply to the

disgorgement of any benefit calculated at Step 1.

6.24.
No settlement discount applies as the proposed financial penalty is £nil.

6.25.
The Step 5 figure is therefore £nil.

6.26.
The Authority therefore has decided not to impose a financial penalty on Amigo

for breaching Principles 2, 3, and 6. But for its financial circumstances, the

Authority would have imposed a financial penalty on Amigo of £72,900,000.

Public censure

6.27.
The Authority’s position in relation to the imposition of a public censure is set out

in Chapter 6 of DEPP. DEPP sets out non-exhaustive factors that may be of

particular relevance in determining whether it is appropriate to issue a public

censure rather than a financial penalty. DEPP 6.4.2G(1) indicates that whether or

not deterrence may be effectively achieved by issuing a public censure may be a

relevant consideration.

6.28.
As explained in paragraph 6.22 above, the Authority has had regard to the need

to balance deterrence with the need to ensure that customers who are creditors

under the Scheme are not adversely affected by the imposition of a financial

penalty. This is consistent with the Authority’s approach in previous cases where

the imposition of a penalty would impact adversely on creditors. Instead, the

Authority has decided to issue a statement censuring Amigo pursuant to section

205 of the Act.

7. PROCEDURAL MATTERS

7.1.
This Notice is given to Amigo under section 205 and in accordance with the section

390 of the Act.

7.2.
The following statutory rights are important.

Decision maker

7.3.
The decision which gave rise to the obligation to give this Notice was made by the

Settlement Decision Makers.

7.4.
Sections 391(4), 391(6) and 391(7) of the Act apply to the publication of

information about the matter to which this notice relates. Under those provisions,

the Authority must publish such information about the matter to which this notice

relates as the Authority considers appropriate. The information may be published

in such manner as the Authority considers appropriate. However, the Authority

may not publish information if such publication would, in the opinion of the

Authority, be unfair to you or prejudicial to the interests of consumers or

detrimental to the stability of the UK financial system.

7.5.
The Authority intends to publish such information about the matter to which this

Final Notice relates as it considers appropriate.

Authority contacts

7.6.
For more information concerning this matter generally, contact Rory Neary at the

Authority (direct line: 020 7066 7972/email: Rory.Neary2@fca.org.uk).

Financial Conduct Authority, Enforcement and Market Oversight Division

ANNEX A

RELEVANT STATUTORY AND REGULATORY PROVISIONS

1.1.
The Authority’s statutory objectives, set out in section 1B(3) of the Act, include the

strategic objective to ensure that the relevant markets function well and the

operational consumer protection objective[s].

1.2.
Section 206(1) of the Act provides:

“If the Authority considers that an authorised person has contravened a

requirement imposed on him by or under this Act… it may impose on him a penalty,

in respect of the contravention, of such amount as it considers appropriate.”

RELEVANT REGULATORY PROVISIONS

Principles for Businesses

1.3.
The “Principles” are a general statement of the fundamental obligations of firms

under the regulatory system and are set out in the Authority’s Handbook. They

derive their authority from the Authority’s rule-making powers set out in the Act.

The relevant Principles, in force from 03/01/2018 throughout the Relevant Period,

are as follows.

1.4.
Principle 2 provides:

“A firm must conduct its business with due skill, care and diligence”.

1.5.
Principle 3 provides:

“A firm must take reasonable care to organise and control its affairs responsibly

and effectively, with adequate risk management systems”.

1.6.
Principle 6 provides:

“A firm must pay due regard to the interests of its customers and treat them fairly”.

Specialist Sourcebook: Consumer Credit Sourcebook (“CONC”)

1.7.
CONC 5.2A.4R, in force from 01/11/2018 throughout the Relevant Period,

regarding creditworthiness assessments, stated:

“A firm must undertake a reasonable assessment of the creditworthiness of

a customer before:

(1) entering into a regulated credit agreement; or

(2) significantly increasing the amount of credit provided under a regulated credit

agreement; or

(3) significantly
increasing
a credit
limit for running-account
credit under

a regulated credit agreement”.

1.8.
CONC 5.2A.5R, in force from 01/11/2018 throughout the relevant period, regarding

creditworthiness assessments, stated:

“The firm must not take a step in CONC 5.2A.4R(1) to (3) unless it can

demonstrate that it has, before doing so:

(1) undertaken
a creditworthiness
assessment and,
where
relevant,
the

assessment
under CONC
5.2A.31R(2) (guarantors)
in
accordance
with

the rules set out in this section; and

(2) had proper regard to the outcome of that assessment in respect of affordability

risk.”

1.9.
CONC 5.2A.6G, in place from 01/11/2018 throughout the Relevant Period, stated:

“If an increase in the amount of credit or in the credit limit is not itself significant

but would result in there having been, since the last creditworthiness assessment,

a cumulative increase that is significant, then a further creditworthiness

assessment is required. This may be the case, for example, where a number of

consecutive increases have been made over a period, none of which is significant

when considered in isolation but the aggregate sum of which is significant”.

1.10. CONC 5.2A.7R, in force from 01/11/2018 throughout the Relevant Period,

regarding creditworthiness assessments, stated:

“A firm must base its creditworthiness assessment on sufficient information:

(1) of which it is aware at the time the creditworthiness assessment is carried out.

(2) obtained, where appropriate, from the customer, and where necessary from

a credit reference agency, and

the information must enable the firm to carry out a reasonable creditworthiness

assessment.”

1.11. CONC 5.2A.8G, in place from 01/11/2018 throughout the Relevant Period, stated:

“CONC 5.2A.20R to CONC 5.2A.25G contain rules and guidance in relation to the

factors that should be taken into account in an individual case when deciding how

much information is sufficient for the purposes of the creditworthiness assessment,

what information it is appropriate and proportionate to obtain and assess, and

whether and how the accuracy of the information should be verified”.

1.12. CONC 5.2A.12R (2)(b), in force from 01/11/2018 throughout the Relevant Period,

regarding creditworthiness assessments, stated:

“The firm must consider the customer’s ability to make repayments under the

agreement:

(2) out of, or using, one or more of the following:

(b) income
from
savings
or
assets
jointly
held
by
the customer with

another person, income received by the customer jointly with another person or

income received by another person in so far as it is reasonable to expect such

income to be available to the customer to make repayments under the

agreement”.

1.13. CONC 5.2A.15R, in force from 01/11/2018 throughout the Relevant Period,

regarding customer’s income and expenditure, stated:

“(1) This rule applies unless:

(a) the firm can demonstrate that it is obvious in the circumstances of the

particular case that the customer is able to make repayments in accordance

with CONC 5.2A.12R, so as to make the actions described in (2) to (4)

disproportionate; or

(b) the customer has indicated clearly an intention to repay wholly using savings

or other assets (see CONC 5.2A.13R).

(2) The firm must take reasonable steps to determine the amount, or make a

reasonable estimate, of the customer’s current income.

(3) Where it is reasonably foreseeable that there is likely to be a reduction in

the customer’s income:

(a) during the term of the agreement; or

(b) in the case of an open-end agreement, during the likely duration of

the credit (see CONC 5.2A.26R),

which could have a material impact on affordability risk, the firm must take

reasonable steps to estimate the amount of that reduction.

(4) The firm must take account of the customer’s income it has determined or

estimated in accordance with (2) and (3).

(5) The firm may only take into account an expected future increase in

the customer’s income where the firm reasonably believes on the basis of

appropriate evidence that the increase is likely to happen during the term of the

agreement or, in the case of an open-end agreement, during the likely duration

of the credit.”

1.14. CONC 5.2A.16G, in place from 01/11/2018 throughout the Relevant Period, stated:

“(1) A firm that
proposes
to
rely
on
the
exception
in CONC

5.2A.15R(1)(a) should keep in mind that the burden would be on the firm to

demonstrate, if challenged, that the absence of a material affordability risk was

obvious such as to make the process of determination or estimation of

the customer’s income disproportionate.

(2) An estimate of the customer’s income may include a minimum amount or a

range, provided that any assumptions on which the estimate is based are

reasonable in the circumstances.

(3) For the purpose of considering the customer’s income under CONC

5.2A.15R, it is not generally sufficient to rely solely on a statement of current

income made by the customer without independent evidence (for example, in

the form of information supplied by a credit reference agency or documentation

of a third party supplied by the third party or by the customer).

(4) An example of where it may be reasonable to take into account an expected

future increase in income would be a loan to fund the provision of further or

higher education, provided that an appropriate assessment required by this

section is carried out. If, in such a case, the customer’s income does not

increase in line with expectations, the firm should consider deferring or limiting

the obligation to repay until the customer’s income has reached an appropriate

level.

(5) Income can include income other than salary and wages.”

1.15. CONC 5.2A.17R, in force from 01/11/2018 throughout the Relevant Period,

regarding customer’s income and expenditure, stated:

“(1) This rule:

(a) applies only where CONC 5.2A.15R also applies; and

(b) does not apply where the firm can demonstrate that it is obvious in the

circumstances of the particular case that the customer’s non-discretionary

expenditure is unlikely to have a material impact on affordability risk, so as

to make the actions described in (2) to (4) disproportionate.

(2) The firm must take reasonable steps to determine the amount, or make a

reasonable estimate, of the customer’s current non-discretionary expenditure.

(3) Where it is reasonably foreseeable that there is likely to be an increase in

the customer’s non-discretionary expenditure:

(a) during the term of the agreement; or

(b) in the case of an open-end agreement, during the likely duration of

the credit (see CONC 5.2A.26R),

which could have a material impact on affordability risk, the firm must take

reasonable steps to estimate the amount of that increase.

(4) The firm must take account of the customer’s non-discretionary expenditure

it has determined or estimated in accordance with (2) and (3).

(5) The firm may only take into account an expected future decrease in non-

discretionary expenditure where the firm reasonably believes on the basis of

appropriate evidence that the decrease is likely to happen during the term of the

agreement or, in the case of an open-end agreement, during the likely duration

of the credit.”

1.16. CONC 5.2A.19G, in place from 01/11/2018 throughout the Relevant Period, stated:

“(1) For the purpose of considering the customer’s non-discretionary expenditure

under CONC 5.2A.17R, the firm may take into account statistical data unless it

knows or has reasonable cause to suspect that the customer’s non-discretionary

expenditure is significantly higher than that described in the data or that the data

are unlikely to be reasonably representative of the customer’s situation.

(2) It is unlikely to be appropriate to place reliance on statistical data, for

example, where the firm is aware, or has reasonable cause to be aware from

information in its possession, that the composition of the customer’s household,

or the number of dependants that the customer has, or the level of

the customer’s existing indebtedness, differs significantly from that of the sample

of persons on which the statistical data were based”.

1.17. CONC 5.2A.33R, in force from 01/11/2018 throughout the Relevant Period,

regarding policies and procedures for creditworthiness assessments, stated:

“A firm must:

(1) establish, implement and maintain clear and effective policies and procedures:

(a) to enable it to carry out creditworthiness assessments or assessments

under CONC 5.2A.31R(2); and

(b) setting out the principal factors it will take into account in carrying

out creditworthiness assessments or assessments under CONC 5.2A.31R(2).

(2) set out the policies and procedures in (1) in writing, and (other than in the case

of a sole trader) have them approved by its governing body or senior personnel.

(3) assess and periodically review:

(a) the effectiveness of the policies and procedures in (1); and

(b) the firm’s compliance with those policies and procedures and with its

obligations under CONC 5.2A.

(4) in the light of (3), take appropriate measures to address any deficiencies in the

policies and procedures or in the firm’s compliance with its obligations.

(5) maintain a record, on paper or in electronic form, of each transaction where

a regulated credit agreement is entered into, or where there is a significant increase

in the amount of credit provided under a regulated credit agreement or a credit

limit for running-account credit under a regulated credit agreement, sufficient to

demonstrate that:

(a) a creditworthiness
assessment or
an
assessment
under CONC

5.2A.31R(2) was carried out where required; and

(b) the creditworthiness
assessment or
the
assessment
under CONC

5.2A.31R(2) was reasonable and was undertaken in accordance with CONC

and so, to enable the FCA to monitor the firm’s compliance with its obligations

under CONC 5.2A; and

(6) (other than in the case of a sole trader) establish, implement and maintain

robust governance arrangements and internal control mechanisms designed to

ensure the firm’s compliance with (1) to (5)”.

1.18. CONC 5.2A.34G, in place from 01/11/2018 throughout the Relevant Period, stated:

“Firms are
reminded
of
the guidance on
record-keeping
in SYSC

9.1.4G and 9.1.5G”.

Dispute Resolution: Complaints (DISP)

1.19. DISP 1.3.3, in force from 03/01/2018 throughout the Relevant Period, regarding

complaints handing procedures for respondents, stated:

“A respondent must put in place appropriate management controls and take

reasonable steps to ensure that in handling complaints it identifies and remedies

any recurring or systemic problems, for example, by:

(1) analysing the causes of individual complaints so as to identify root causes

common to types of complaint.

(2) considering whether such root causes may also affect other processes or

products, including those not directly complained of; and

(3) correcting, where reasonable to do so, such root causes.”

1.20. DISP 1.3.3BG, in place from 01/08/2016 throughout the Relevant Period stated:

“The processes that a firm or CBTL firm should have in place in order to comply

with DISP 1.3.3 R may include, taking into account the nature, scale and complexity

of the firm's or CBTL firm’s business including, in particular, the number

of complaints the firm or CBTL firm receives:

(1) the collection of management information on the causes of complaints and the

products
and
services complaints relate
to,
including
information

about complaints that are resolved by the firm by close of business on the

third business day following the day on which it is received.

(2) a process to identify the root causes of complaints (DISP 1.3.3 R (1)).

(3) a process to prioritise dealing with the root causes of complaints.

(4) a process to consider whether the root causes identified may affect other

processes or products (DISP 1.3.3 R (2)).

(5) a process for deciding whether root causes discovered should be corrected

and how this should be done (DISP 1.3.3 R (3)).

(6) regular reporting to the senior personnel where information on recurring or

systemic problems may be needed for them to play their part in identifying,

measuring, managing and controlling risks of regulatory concern; and

(7) keeping records of analysis and decisions taken by senior personnel in

response to management information on the root causes of complaints.”

1.21. DISP 1.3.6G, in place from 01/09/2011 to 31/03/2019, covering the beginning of

the Relevant Period stated:

“Where a firm identifies (from its complaints or otherwise) recurring or systemic

problems in its provision of, or failure to provide, a financial service, it should (in

accordance with Principle 6 (Customers' interests) and to the extent that it applies)

consider whether it ought to act with regard to the position of customers who may

have suffered detriment from, or been potentially disadvantaged by, such problems

but who have not complained and, if so, take appropriate and proportionate

measures to ensure that those customers are given appropriate redress or a proper

opportunity to obtain it. In particular, the firm should:

(1) ascertain the scope and severity of the consumer detriment that might have

arisen; and

(2) consider whether it is fair and reasonable for the firm to undertake proactively

a redress or remediation exercise, which may include contacting customers who

have not complained.”

1.22. DISP 1.3.6G, in place from 01/04/2019 beyond the end of the Relevant Period

“Where a firm identifies (from its complaints or otherwise) recurring or systemic

problems in its provision of, or failure to provide, a financial service or claims

management service, it should (in accordance with Principle 6 (Customers'

interests) and to the extent that it applies) consider whether it ought to act with

regard to the position of customers who may have suffered detriment from, or been

potentially disadvantaged by, such problems but who have not complained and, if

so,
take
appropriate
and
proportionate
measures
to
ensure
that

those customers are given appropriate redress or a proper opportunity to obtain it.

In particular, the firm should:

(1) ascertain the scope and severity of the consumer detriment that might have

arisen; and

(2) consider whether it is fair and reasonable for the firm to undertake proactively

a redress or remediation exercise, which may include contacting customers who

have not complained.”

1.23. DISP 1.4.1R, in force from 01/09/2011 throughout the Relevant Period, regarding

investigating, assessing and resolving complaints, stated:

“Once a complaint has been received by a respondent, it must:

(1) investigate the complaint competently, diligently and impartially, obtaining

additional information as necessary.

(2) assess fairly, consistently and promptly:

(a) the subject matter of the complaint.

(b) whether the complaint should be upheld.

(c) what remedial action or redress (or both) may be appropriate.

(d) if appropriate, whether it has reasonable grounds to be satisfied that

another respondent may be solely or jointly responsible for the matter alleged

in the complaint.

taking into account all relevant factors.

(3) offer redress or remedial action when it decides this is appropriate.

(4) explain to the complainant promptly and, in a way that is fair, clear and not

misleading, its assessment of the complaint, its decision on it, and any offer of

remedial action or redress; and

(5) comply promptly with any offer of remedial action or redress accepted by the

complainant”.

1.24. DISP 1.4.2G, in place from 01/04/2013, covering the start of the Relevant Period

to 26/09/2019 stated:

“Factors that may be relevant in the assessment of a complaint under DISP 1.4.1R

(2) include the following:

(1) all the evidence available and the particular circumstances of the complaint.

(2) similarities with other complaints received by the respondent.

(3) relevant guidance published
by
the FCA,
other
relevant
regulators,

the Financial Ombudsman Service or former schemes; and

(4) appropriate
analysis
of
decisions
by
the Financial
Ombudsman

Service concerning similar complaints received by the respondent (procedures for

which are described in DISP 1.3.2A G).”

1.25. DISP 1.4.2G, in place from 27/09/2019 beyond the end of the Relevant Period,

“Factors that may be relevant in the assessment of a complaint under DISP 1.4.1R

(2) include the following:

(1) all the evidence available and the particular circumstances of

the complaint.

(2) similarities with other complaints received by the respondent.

(3) relevant guidance published by the FCA, other relevant regulators,

the Financial Ombudsman Service or former schemes; and

(4) appropriate analysis of decisions by the Financial Ombudsman

Service concerning
similar complaints received
by

the respondent (procedures for which are described in DISP 1.3.2A G)”.

1.26. DISP 1.4.3G, in place from 01/11/2007 throughout the Relevant Period, stated:

“The respondent should aim to resolve complaints at the earliest possible

opportunity, minimising the number of unresolved complaints which need to be

referred to the Financial Ombudsman Service”.

Decision Procedure and Penalties Manual (DEPP)

1.27. Chapter 6 of DEPP, which forms part of the Authority’s Handbook, sets out the

Authority’s statement of policy with respect to the imposition and amount of

financial penalties under the Act.

The Enforcement Guide

1.28. The Enforcement Guide sets out the Authority’s approach to exercising its main

enforcement powers under the Act.

1.29. Chapter 7 of the Enforcement Guide sets out the Authority’s approach to exercising

its power to impose a financial a penalty.

ANNEX B

1.1
In addition to the 15 files discussed at Paragraph 4.14 above, the Authority

reviewed 37 customer files for loans underwritten during the Relevant Period. This

included the borrower and guarantor files for each loan. These were not selected

at random.

1.2
The Authority sets out the rules for firms in the consumer credit sector in the

CONC Consumer Credit Sourcebook. The overall aim is to ensure that firms

undertake adequate creditworthiness assessments to determine whether a

proposed loan is affordable. In addition, to make sure firms treat customers fairly

when offering credit to customers.

1.3
The rules require lenders to conduct an assessment on both a borrower and

guarantor and to ensure that the loan does not negatively impact their financial

situation. This assessment does not need to be identical, but it must be based on

sufficient information to prevent any adverse impact or financial detriment.

1.4
As a consumer credit lender, Amigo was required to adhere to the CONC rules.

However, the customer files reviewed by the Authority demonstrate significant

failings. The files also highlight the poor outcomes that its customers experienced

as a result of Amigo’s failings.

1.5
The detailed findings of 3 customer file reviews are set out below. These detailed

examples clearly illustrate some of the deficiencies the Authority identified in

Amigo’s approach to assessing creditworthiness and affordability.

1.6
The Authority found the following key failings in its review of customer files:

Income verification failings

1.7
CONC 5.2A.15R requires Firms to take reasonable steps to determine or estimate

the amount of a customer’s current income. The supporting guidance outlined in

CONC 5.2A.16G, confirms it is not sufficient for a firm to rely solely on an amount

declared by a customer without verifying this with independent evidence (e.g.,

CRA data, third party or customer documentation). Throughout the Relevant

Period, Amigo verified the income of borrower's and guarantor’s using either credit

reference agency data, its income/benefit matrix or document proof (see

paragraphs 4.54 to 4.69).

1.8
Amigo’s use of credit reference agency data checks was insufficient as its policy

permitted it to approve loans based on a score between 4 and 7, which provided

an inadequate level of assurance that borrowers and/or guarantors could afford

the proposed repayments under the loans advanced (see paragraphs 4.54 to

4.60). Of the 37 customer files reviewed, Amigo approved the loan with a credit

reference agency data income confidence score below 7 for 7 of the 37 borrowers

(18.9%) and 4 of the 37 guarantors (10.8%). This impacted 11 files in total

(29.7%).

1.9
Where Amigo was unable to verify income using credit reference agency data, it

used one of two additional methods, namely its income/benefit matrix or

document proof (see paragraphs 4.61 to 4.69). The income/benefit matrix was

significantly flawed, it was a crude tool with estimated income bands for different

job types. As such it was not tailored to the borrower or guarantor. It also placed

reliance on the accuracy of the information provided by the customer regarding

their job role and benefit entitlement. In the event that this information was

incorrect, Amigo had no method of identifying this. The Authority considers this

to be an example of Amigo’s non-compliance with CONC 5.2A15R.

1.10
Where Amigo sought to verify a customer’s income via document proof, its policy

required it to obtain three months' worth of data. However, of the 11 customer

files that were verified via this method, only two of the documents provided were

over a three-month duration. This confirms Amigo failed to follow its own policy

when verifying income. The Authority considers that this failing is particularly

detrimental where a customer is self-employed or in receipt of inconsistent

income. This impacted 1 of the 37 files.

Inadequate expenditure verification

1.11
CONC 5.2A17R provides that Firms must take reasonable steps to determine or

estimate a customer’s non-discretionary expenditure. CONC 5.2A18G confirms

this includes other expenditure required to give a basic quality of life. However,

Amigo’s assessment was not sufficient to meet this requirement as it did not

explicitly account for items, which in the majority of cases would be considered a

necessity.

1.12
Amigo told the Authority that it verified a customer’s expenditure using ONS

statistical data. However, for all of the 37 files reviewed, the ONS data used was

significantly out of date as Amigo applied statistical data from 2013 to loans

approved throughout 2018 and 2019.

1.13
CONC 5.2A19G confirms it is inappropriate for a firm to rely on statistical data

when it is aware of, or in possession of information which suggests that a

customer’s expenses may be higher. For example, where a Firm has information

on the customer's number of dependents. The customer files demonstrate that

Amigo did not take this guidance into consideration. Of the 37 files reviewed,

there were 12 where Amigo was aware of the number of dependents for either

the borrower and/or guarantor. For these customers, Amigo only adjusted its

assessment against ONS data for the cost of food and did not adjust its standard

ONS data for any other household costs to take account of the number of

dependents.

Consideration of customer credit history and financial difficulty

1.14
In line with CONC 5.2A7R, Amigo was required to base its creditworthiness

assessment on sufficient information. This includes information obtained from the

customer, and where necessary a credit reference agency. Although Amigo did

obtain its customer’s credit reports, in the majority of files reviewed, Amigo

merely used this to determine the amounts paid towards credit items.

1.15
CONC 5.2A22G encourages Firms to consider whether a customer is in, has been

in or is likely to experience financial difficulty. A credit report is a key method of

identifying this, as it will include common indicators such as accounts over the

limit, in significant arrears, in default or sometimes with a debt collection agency.

Despite this, there is limited evidence that Amigo used credit reports in this

manner. The Authority identified 25/37 customer files with indicators that Amigo

should have explored, but in all of these instances, the customer was not probed

to determine the impact of this. Of these customers 22/25 (88%) subsequently

cycled into arrears or defaulted on the Amigo loan.

Top-up lending

1.16
During the Relevant Period, Amigo sent borrowers a text or email to inform them

when they became eligible for a top up. The Authority considers that this may not

have been in a borrower’s best interests, where they had recently been in arrears

or taken several top ups within a short period of time. There were 7 customer files

where this occurred.

Customer file 1

1.17
Borrower [A] had 2 dependents when she applied for the loan in July 2019. At the

time, she was working part time and receiving benefit income. The guarantor was

Borrower [A]’s parent, Guarantor [B], who was unemployed. Guarantor [B] was

also receiving benefit income. The loan reviewed was the second loan in a chain

of 2 top up loans that Borrower [A] took out with Amigo.

1.18
The first loan was taken out 4 months prior to Borrower [A]’s second loan from

Amigo reviewed by the Authority in customer file 1. Although Borrower [A] missed

a payment on their first loan during the four months that they had it, Amigo failed

to exercise caution before agreeing any further borrowing. Instead, Amigo

prompted Borrower[A] to apply for the top up loan, as it sent them a text

informing them that they were eligible, the same day Borrower [A] applied for the

top up. The Authority considers it is likely Amigo’s text was contributing factor in

Borrower [A]’s decision to apply for the top-up.

1.19
Borrower [A] stated during a call regarding their application, that the purpose of

the loan was to cover the arrears on her car insurance, council tax and water bills.

However, it does not appear that Amigo considered this as an indication of

financial difficulty. Or that Borrower [A] was likely vulnerable at the time of the

application. This is supported by the information on Borrower [A]’s credit file

which showed they were also in arrears on an advance against income account.

In addition, the fact that Borrower [A] told Amigo that the arrears on their bills

were a result of recent separation from an abusive ex-partner. This was a further,

clear indicator of potential vulnerability that was not acted on by Amigo.

1.20
Amigo also made several errors when assessing the affordability of Borrower [A]

and Guarantor [B] as its agents did not question all the expenses which were

below ONS figures or recorded as £0 with the rationale that a third party pays.

Amigo’s policy states a task would have been raised in the system for any

‘unrealistic costs’. However, it only questioned Borrower [A] on their travel

expenditure. Between Borrower [A] and Guarantor [B] there were expenses which

the Authority considers should have been questioned further, some of which were

more than 50% below the national average. However, this did not occur. As a

result, there was no meaningful expenditure verification check.

1.21
Amigo deemed Borrower [A]’s original Budget to be unaffordable. However, rather

than declining the application, Amigo re-budgeted the loan by lowering the overall

amount from £4,000 to £3,750. Outside of asking Borrower [A] whether this

would still be enough to cover her bills, Amigo’s agents did not probe and question

Borrower [A] sufficiently about this.

1.22
The loan fell into arrears on its second payment in August 2019. Borrower [A] told

Amigo that this was because they had been off work and receiving statutory sick

pay. Amigo agreed with Borrower [A] an arrangement to pay the arrears back

over the next three months. However, this arrangement failed. Amigo placed

Borrower [A] on 4 further arrangements, this time to pay a reduced amount of

£16 per week (£69.33 per month) rather than their usual payment of £182.96.

However, these arrangements also failed. In fact, Borrower [A] only ever made 2

reduced payments towards the loan throughout its entire duration. There were no

payments from the guarantor.

1.23
Amigo issued a Notice of Default in October 2019 when payments required under

the credit agreement had not been made. The account was then over 3 payments

behind and passed to a specialist team within Amigo in December 2019 which

sought to agree an arrangement with Borrower [A] to pay off the outstanding

balance. This was just five months after the loan was agreed. Despite this action,

Amigo were unable to obtain any further payments. At the time the Authority

reviewed the file, there was an outstanding balance of £4,765.27 on the loan.

1.24
The fact that neither Borrower [A] nor their guarantor ever paid the full

contractual amount towards the loan suggests this may not have been affordable.

This is supported by the fact that Borrower [A] failed their forbearance

arrangements, even when this was for a reduced amount of £16 per week (£69.33

per month). Amigo conducted a credit search on Borrower [A] in June 2020, which

showed a worsened financial position. For example, Borrower [A] had an increased

number of delinquent and defaulted accounts.

Customer file 2

1.25
Borrower [E] took out this loan in February 2019, for the purpose of “business”.

At the time, Borrower [E] was employed full-time with a declared income of

£8,500. Guarantor [F] was employed full time with a monthly income of £6,800.

1.26
In total, Borrower [E] took out 7 loans with Amigo between February 2018 and

October 2019. Of the 7 loans, 4 of them were for an amount of £10,000. This

includes the loan reviewed which was loan number 5. The Authority considers that

Amigo may not have acted in Borrower [E]’s best interests as it appears to have

encouraged Borrower [E]’s top ups on at least 3 occasions, by sending an SMS

and/or email to inform Borrower [E] that they were eligible in the days before

they took out the top up loan.

1.27
When assessing Borrower [E]’s income, Amigo also placed reliance on a credit

reference agency income confidence score of 5. In the absence of any other form

of verification, the Authority considers that this would have been insufficient.

Particularly as there is a possibility that this included other non-employment

income or was based on unreliable data. However, it does not appear that Amigo

considered this possibility and as such the figure used could be incorrect.

1.28
Amigo also failed to consider Borrower [E]’s pattern of borrowing when conducting

its affordability assessment. The fact that this was the third loan that they had

taken out over a period of five 5 months, and that each loan was for an amount

of £10,000 should have alerted Amigo that Borrower [E] may have been

experiencing financial difficulty. Or that their declared disposable income of

£3,539.63 may have been incorrect. Although each loan was for an amount of

£10,000, the majority of these funds were used to pay off the previous loan and

Borrower [E] only would have received around £500. Yet, Amigo’s agents failed

to question Borrower [E] about this.

1.29
There were also other indicators that Borrower [E] was in financial difficulty at the

point of application. Borrower [E]’s credit report showed advances against income

accounts and an increasing number of new credit accounts. Again, this was

inconsistent with the information that Borrower [E] provided to Amigo in terms of

the amount of his disposable income. However, it does not appear that this was

considered.

1.30
The file confirms some of Borrower [E]’s declared expenditure was inconsistent

with the amounts shown on their credit file. The amounts recorded for their

advance against income repayments on the Budget, was lower than the amount

shown on Borrower [E]’s credit file. It appears Amigo did not identify this, as it

did not record an explanation on file. As a result, Borrower [E]’s expenditure may

have been understated by at least £520.

1.31
There were also flaws in Amigo’s assessment of Guarantor [F]’s affordability as it

is not clear how it verified his income. Additionally, although Guarantor [F] told

Amigo that they paid their spouse £1,000 per month towards household expenses,

this was not recorded on the Budget or verified. As such, Guarantor [F]’s

expenditure has been understated by at least £1,000.

1.32
As Amigo also failed to question Borrower [E] and Guarantor [F] on their high

disposable incomes compared to the loan amount in line with its own policy. The

Authority considers that its assessment was flawed.

1.33
Borrower [E] made 2 contractual payments, before this loan was refinanced into

loan 6, in April 2019. There were no payments from the guarantor. Although there

were no arrears on the loan during its 2-month duration, Borrower [E] did fall into

arrears on their subsequent loans. On loan 6 they fell into arrears on the first

payment, before clearing this the following month. That loan was also topped up

into Borrower [E]’s last loan (loan 7) which also entered arrears.

1.34
Borrower [E] requested additional borrowing in May 2020, however, by this point

Amigo identified the indicators that Borrower [E] may be experiencing financial

difficulty and declined to lend. The Authority considers that these indications were

present on the loan reviewed (the fifth loan). Amigo should have identified this

earlier, and at least before the subsequent top up loans were taken out.

Customer file 3

1.35
Borrower [C] took out a new loan with Amigo in March 2019, to consolidate

existing debts. At the time, Borrower [C] had 3 dependents (including a 4-month-

old baby) and was working part time.

1.36
The guarantor was Borrower [C]’s sibling, Guarantor [D], whose income was

£1,700 per month. They both worked for the same employer. Borrower [C]’s loan

was for an amount of £750. Although they stated that this would be used to

consolidate existing debts, their credit file confirms that they only had 1 credit

item at the time. This was a hire purchase agreement that she had taken out a

month before with a balance just under £3,000. Amigo did not question Borrower

[C] about this.

1.37
Amigo overlooked information that Borrower [C] provided regarding their income

and future employment. During a call from the application, Amigo questioned

Borrower [C] about their income, as their bank statement showed a lower income

than the amount than recorded on their Budget. In response, Borrower [C] told

Amigo that they had “three children one who's four months” which meant they

has a “substantial amount of time off work”. Borrower[C] also stated they were

hoping to move to a “salary-based job” within the next two weeks. This should

have raised concerns about the sustainability of Borrower [C]’s income. However,

Amigo’s agents did not discuss this further with Borrower [C] to understand their

expected change in circumstance or determine the extent to which their income

was varied in nature. Amigo only confirmed that the weekly income of £84 (as

per the bank statement), was the minimum that Borrower [C] would receive.

1.38
Amigo also failed to follow its own policy, as it verified Borrower [C]’s income with

a bank statement which showed just one weekly payment from their employer.

However, according to the policy that was in place at the time, Amigo was required

to obtain bank statements over a period of three months.

1.39
The Authority considers that Amigo also failed to record the information that

Borrower [C] provided regarding their dependants during a telephone call.

Although Borrower[C] told an Amigo agent that they had three children, it

incorrectly recorded the number of dependents as two. As a result, any

assessment that Amigo conducted on her expenditure figures (e.g., food) would

have based on incorrect data.

1.40
There were also deficiencies in the affordability assessment of Guarantor [D].

They declared several items of expenditure as £0 providing justification such as

‘lives at home’ or ‘other’. This includes individual expenditure such as ‘Phone /

Mobile / Internet’ and ‘Clothing’. However, there is nothing on the file to

demonstrate that Guarantor [D] was questioned about this.

1.41
The loan fell into arrears on its first payment. Although Borrower [C] cleared the

arrears that same month, the failure to make the first payment in line with the

agreement, suggests the loan may not have been affordable for them. In total,

Borrower [C] missed [14] payments throughout the loan’s entire duration. This

includes a period where Borrower [C] made no payment for 6 months from March

until August 2020 when Borrower [C] cleared the loan in full. The file does not

confirm Borrower[C]’s financial position at the time they cleared the loan, and it

is unclear whether they used their own funds to do so. The Guarantor [D] was

never asked to make a payment.

1.42
The Authority recognises that the issues Borrower [C] experienced in March 2020

when they stopped making payments, were likely due to the pandemic and a

change in their circumstances as they told Amigo that they were unemployed. It

also came to light that Borrower [C] was vulnerable at the time, as they suffered

with a mental health condition. However, Borrower [C]’s file demonstrates that

Amigo did not do enough to determine the nature and frequency of Borrower [C]’s

income at the outset of the consumer credit relationship, and their erratic

payment behaviour suggests Borrower [C] struggled with payments. Had Amigo

explored Borrower [C]’s comments around their employment, and obtained

sufficient information on their expenditure, it is likely that it would have identified

that Borrower [C] was likely to encounter issues maintaining repayments.