Sitting on a gold mine: Getting what’s owed to pawnbroking customers

Abstract

In 2018 the Financial Conduct Authority identified that pawnbroking customers are not always collecting the ‘surplus’ money owed to them. Surplus money is generated when a customer defaults on their loan and their pawned item is sold at auction for more than what is owed by the customer. Although there is a process for notifying customers when a surplus is generated, collection rates are low. In the current paper we share the results of a first intervention designed to address this. We partnered with one of the UK’s largest pawnbroking lenders to trial a novel behavioural design approach. A first intervention, a new reminder letter for customers, is reported in this paper and found to increase surplus collection rates significantly. A second experiment, focused more on interventions on pawnbroker store processes, is currently in the field.

Full text

Financial Conduct Authority
59
Occasional Paper
20 January 2021
Sitting on a gold mine: Getting what’s
owed to pawnbroking customers
Paul Adams, Chris Burke, Alex Chesterfield, Bhavini
Parmar, Laura Smart and Anna Whicher

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 1
The FCA occasional papers
The Financial Conduct Authority (FCA) is committed to encouraging debate on all aspects
of financial regulation and to creating rigorous evidence to support its decision-making.
To facilitate this, we publish a series of occasional papers, extending across economics
and other disciplines.
The main factor in accepting papers is that they should make substantial contributions to
knowledge and understanding of financial regulation. If you want to contribute to this
series or comment on these papers, please contact Karen Croxson
(karen.croxson@fca.org.uk)
Disclaimer
Occasional papers contribute to our work by providing rigorous research results and
stimulating debate. While they may not necessarily represent our position, they are one
source of evidence that we may use while discharging its functions and to inform its
views. We endeavour to ensure that research outputs are correct, through checks
including independent referee reports, but the nature of such research and choice of
research methods is a matter for the authors using their expert judgement. To the extent
that Occasional Papers contain any errors or omissions, they should be attributed to the
individual authors, rather than to the FCA.
Authors
Paul Adams (Contributed whilst at the Financial Conduct Authority)
Chris Burke (Financial Conduct Authority)
Alex Chesterfield (Contributed whilst at the Financial Conduct Authority)
Bhavini Parmar (Financial Conduct Authority)
Laura Smart (Contributed whilst at the Financial Conduct Authority)
Anna Whicher (Cardiff Metropolitan University).
Acknowledgements
We would like to thank the staff and customers of several stores who spent their valuable
time discussing with us, Andrew Jones (FCA), Richard Dorsett (Univeristy of
Westminster) for cluster RCT power calculations, Michael Sanders (King’s College
London) for helpful methodological discussions and Marieke Bos (Stockholm School of
Economics) for academic review.
All our publications are available to download from www.fca.org.uk. If you would like to
receive this paper in an alternative format, please call 020 7066 9644 or email
publications_graphics @fca.org.uk or write to Editorial and Digital Department, Financial
Conduct Authority, 12 Endeavour Square, London E20 1JN.
FCA occasional papers in financial
regulation

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 2
1. Executive Summary 3
2. Introduction 6
3. Context 8
Market context 8
Our behavioural design approach 9
User research 15
Data analysis 16
4. Hypotheses, intervention design and experimental method 20
Drivers of low surplus collection 20
Intervention Design 22
Method (Surplus Reminder Letter) 24
5. Results 26
Surplus Reminder Letter 26
Surplus Flag 27
6. Conclusions 28
Table of Contents

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 3
Abstract
In 2018 the Financial Conduct Authority identified that pawnbroking customers
are not always collecting the ‘surplus’ money owed to them. Surplus money is
generated when a customer defaults on their loan and their pawned item is sold
at auction for more than what is owed by the customer. Although there is a
process for notifying customers when a surplus is generated, collection rates are
low. In the current paper we share the results of a first intervention designed to
address this. We partnered with one of the UK’s largest pawnbroking lenders to
trial a novel behavioural design approach. A first intervention, a new reminder
letter for customers, is reported in this paper and found to increase surplus
collection rates significantly. A second experiment, focused more on
interventions on pawnbroker store processes, is currently in the field.
Introduction
Pawnbroking is one of the oldest forms of lending in the world and is still used by around
350,000 people in the UK per year, many of whom are vulnerable and have low financial
resilience (FCA, 2018). In 2018, we identified around £1 million per year that was owed
to pawnbroking customers, but was not claimed. Using a novel behavioural design
approach to inform a randomised controlled trial, we investigated the reasons why this
occurred and tested interventions to increase collection rates. Our first intervention, a
behaviourally informed letter, almost doubled collection rates within 30 days – one of the
most successful letter-based interventions previously tested by us. This and a second
experiment currently in the field, aim to inform regulatory policy in this area and help
people collect money that is owed to them.
Problem
Pawnbroking customers can borrow money by pledging an item belonging to them at
pawnbroking shops, and reclaim the item when the loan plus interest is repaid. However,
when a customer does not pay back their loan within an agreed time, the pawnbroker is
entitled to sell the item to recover their costs. A little-known fact is that if the item sells
for more than the total that was owed to the pawnbroker, the excess revenue, or ‘surplus’
needs to be returned to the customer. In 2018 we reviewed the pawnbroking sector and
found that surplus return rates were relatively low, with some firms reporting less than
half of the outstanding surpluses were being collected by customers (FCA, 2018).
Because pawnbroking loans are a form of secured credit extensive credit checks are not
required or widespread in the market, meaning customers can benefit from rapidly
accessing credit when they need to. After initial registration, pawnbroking customers may
change their address or contact details without these being updated on systems and still
access credit quickly since the loans are secured. A downside of this is that it can be difficult
1. Executive Summary

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 4
for firms to return surplus to customers when it is incurred. We investigated this problem
with a view to increasing surplus collection rates and reducing the harm to consumers in
this sector, currently estimated to be in the region of £1m per year. For the firm that we
partnered with, the average uncollected surplus per customer was £70.
What we did
Using a novel behavioural design approach, we conducted user research, customer and
staff interviews and data analysis to map out the customer journey and discover the factors
that might be influencing low surplus collection rates in one large pawnbroking firm with
stores distributed throughout England and Scotland
1
. We utilised the ‘double diamond’
model from the design literature (Banathy, 2013). This starts with a bottom-up, divergent
thinking exploration process (discovering what the problem is through user research, user
experience and data analysis). It then leads to a detailed definition of the problem, and
culminates in a convergent thinking approach to focus in on the development and delivery
of solutions. While previous comprehensive studies of pawnbroking customers have taken
a survey approach (Collard and Hayes, 2010), our approach differs in providing detailed
insights on pawnbroking customers, their motivations and actions. This process helped us
to design 2 interventions that were tested in randomised controlled trials – a reminder
letter to address low surplus collection rates on the consumer side and a “surplus flag” on
the software system of the stores that would alert staff that repeat consumers were owed
a surplus on the supplier side.
Findings
We found that the behaviourally designed reminder letter, delivered to customers 2 weeks
after incurring a surplus due to their item being sold, almost doubled surplus collection
rates within 30 days. Moreover, the letter decreased the total amount of surplus money
that remained uncollected from 79% in the control group to 66% in the treated group
within 30 days. Due to challenges relating to Covid-19, it was necessary to delay the
surplus flag trial. The latter is in the field currently and the results from this will be
published in a future paper.
Implications
The characteristics of pawnbroking consumers, nature of the consumer journey and the
surplus process meant that traditional methods of nudging customers to collect surplus
(e.g. by contacting them via SMS) would likely have been ineffective. Our behavioural
design thinking approach, combining user experience research, data analysis and
behavioural insights, led us to consider alternative interventions to those that would have
been considered in traditional top-down experimental designs.
1
We do not identify the firm here due to ongoing trials. Readers can view the FCA
pawnbroking sector (FCA, 2018) review for market collection rates and wider contextual
information.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 5
The insights developed through our design approach helped us design a reminder letter
that had stronger effects than previously recorded for letter-based interventions,
highlighting the importance of understanding user needs and experiences. Additionally,
our surplus flag intervention takes into consideration pawnbroking processes on the firm-
side and is currently in the field with results to be reported in a follow-up paper. Although
not all firms in the sector will be able to implement a surplus flag system, redesigning
reminder letters is an effective method of reducing harm to consumers at little extra cost
to the firms themselves.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 6
Pawnbroking is one of the oldest forms of lending in the world (Caskey, 1994) and has
been part of the regulated UK credit market for over a century. Pawnbrokers offer cash
loans to customers after they leave something valuable (known as a pawn or a pledge) as
security for the loan. Anything of value that can be re-sold can be used as a pledge, but
jewellery is the most popular form of security used in the UK (Collard & Hayes, 2010).
Pawnbrokers lend money quickly to customers but at a higher rate of interest than high
street banks, although considerably lower than payday loans and other forms of high cost
short term credit.
Collard and Hayes (2010) give the most recent and complete overview of the UK
pawnbroking market. In it they find pawnbroking customers are generally very satisfied
with the service they receive, with satisfaction ratings of up to 95% (Collard & Hayes,
2010). The speed at which cash can be borrowed, the customer service and the convenient
locations of pawnbrokers on the high street are features that UK customers rate highly.
They also find that pawnbroking customers may have limited access to other forms of
credit and are typically on low incomes (less than £300 per week in 2010). Levels of home
ownership in the pawnbroking customer base is low and almost half of all customers rent
their homes from a local authority or housing association. Agarwal and Bos (2019) also
show that more than 70% of consumers don’t have access to mainstream credit when
taking out their pawn loan. The FCA’s sector review (FCA, 2018) reiterated these positive
features of the pawnbroking market, highlighting business models that focused heavily on
good customer relationships and offered considerable flexibility and personalisation of
credit to suit individual needs. However, one concern that did arise from the sector review
was that some customers were not receiving money that was owed to them if a pledge
was sold for more than the customer owed in the case of a default.
When customers default on their pawnbroking loans, the pawnbroker can sell the
customer’s pledge to recover their costs. If the pledge is of sufficient value, the pledge
must be sold at a public auction and any money that is raised above the outstanding debt
is known as a surplus. Legally, this surplus money from the sale of the pledge must be
made available to the customer and a standardised letter is sent to them to notify them
they can collect this money from the pawnbroker. Any unpaid surplus is held in a separate
account and the pawnbroking firm has no incentive to keep this money (it is held as
dormant funds and is not treated as firm profits), but our sector review found that surplus
collection rates varied widely among firms in this market, and some firms had returned
less than half of this surplus money to customers.
Although the surplus notification letters are standardised according to UK regulations, the
ways in which surplus could be returned to customers is not, and the harm to consumers
in non-collected surpluses is estimated to be in the region of £1m per year. There are a
number of possible reasons for this, many of which could be seen as side-effects of the
positive aspects of pawnbroking credit highlighted above. Because pawnbroking loans are
secured against a valuable item, it is not necessary for firms to conduct extensive credit
checks. Although customer identification is initially verified for anti-money laundering and
2. Introduction

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 7
prevention of crime purposes, pawnbroking customers are more likely to rent and move
often, meaning contact information could be out of date. Although average surplus
amounts are relatively low, they may represent significant sums to consumers on low
incomes. A key part of our mission is to protect consumers, especially if they are
vulnerable, so any measures to increase the amount of surplus returned to consumers
would be a positive step.
In order to try to increase surplus collection rates we employed a novel behavioural design
approach to create and test behavioural interventions in this market. In partnership with
one of the largest pawnbroking firms in the UK, we extensively mapped out the complex
consumer journey for a typical customer, conducted interviews with staff and customers
and combined these insights with analytical data work to holistically design an intervention
which was subsequently tested in a randomised controlled trial (RCT). Specifically, the
double diamond design approach (Discover, Design, Develop & Deliver) allowed us to
generate hypotheses and co-design effective, feasible interventions that would not
necessarily have been discovered using traditional insights from the behavioural science
literature. Furthermore, this bottom-up collaborative approach meant that these
interventions minimised costs and made sense from the perspective of customers and on-
the-ground staff, as opposed to a more traditional top-down approach imposed by
regulators or authorities that may be far-removed from the day-to-day processes and
consumer journeys.
We identified 2 main causes of consumers not collecting surpluses. Staff and customer
interviews highlighted that consumers had a poor understanding of the pawnbroking
process, found it difficult to comprehend that they could be owed money after entering
into a credit agreement, and were unlikely to pay attention to the standardised surplus
notification letter. As such our first intervention attempted to address these comprehension
and inattention issues by designing a new reminder letter that was sent to customers two
weeks after their pledged item(s) were sold at auction. Following on from this, our data
analysis demonstrated that a large number of people are repeat customers that use
pawnbroking stores to access a number of different products (such as cheque-cashing,
buy-back or foreign exchange services). A large number of consumers that were owed
surplus (~38%) had visited stores since without collecting their money, indicating they
were unaware of the outstanding surplus entitlement. To address this fact, we designed a
second, supply-side intervention consisting of a ‘surplus flag’ - a notification on store IT
systems to alert front line staff to pay any outstanding surplus to customers when they
visit the store.
We organise this paper as follows. The third section provides a diagnosis of the problem,
description of the behavioural design thinking approach, results of diagnostic data work,
staff and consumer research, the consumer journey map and hypotheses. The fourth
section details our intervention and trial design and the workshops where we generated
these interventions. The fifth section details our results and conclusions from the trial.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 8
Market context
It is estimated that the pawnbroking market in the UK consists of approximately 290 firms
that generate £125m of revenue per year, with pawnbrokers lending out approximately
£300m to customers in 2018 (Figure 1; FCA, 2018). Although there is a large disparity
between the smallest and the largest firms in the market, the average UK pawnbroking
firm employs 23 members of staff 4 stores and enters into 5400 pawnbroking agreements
per year. Across the whole market the average loan-to-value (LTV) is 50%, and the annual
percentage rate (APR) interest on loans is 120%. Customer typically borrow around £300
per agreement with an average maximum loan of £2000. The average firm turns over
£1.5m per year with a profit of £100,000.
Figure 1. Overview of the UK pawnbroking market (Financial Conduct Authority,
2018).
While pawnbroking loans are classed as a form of high cost short term (HCST) credit, the
interest charged is significantly less than other forms of HCST credit such as payday loans,
home collected credit and rent-to-own schemes. However, in recent years pawnbrokers
have diversified into offering other financial services, including foreign exchange, cheque
cashing, payday lending and rental purchase. This diversification is reflected in the fact
that only 43% of the average pawnbroker’s turnover in 2010 came from pawnbroking
agreements (Collard and Kempson, 2003).
A 2010 review commissioned by the National Pawnbrokers Association took an in-depth
look at the characteristics of the UK pawnbroking market (Collard and Heyes, 2010). They
3. Context

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 9
found that pawnbroking customers tend to be women with families (64% female) and most
are aged between 20-49, coinciding with the demographic that is most likely to use credit
products due to the financial pressures of setting up a home and family (Kempson, 2002).
The same review highlighted aspects that may suggest pawnbroking customers are more
vulnerable than the average consumer: only 20% of surveyed customers owned their own
home, with 48% living in accommodation rented from housing authorities or local
authorities. In addition, 53% lived in households with nobody in employment and 70%
reported household incomes of £300 per week or less.
The Collard and Heyes (2010) survey also highlighted several key insights on pawnbroking
customer’s access to other forms of credit, which could be a key driver in their use of
pawnbroking services. They found that consumers were less likely to have a current
account or basic bank account relative to the wider UK population. Of the customers that
did have access to a bank account (basic or current), a third reported having an overdraft
facility and 24% reported that they were currently overdrawn. 73% stated that they had
borrowed money recently from other sources, with approximately a third borrowing from
friends or family and a third borrowing from payday lenders or home collected credit firms.
It should be noted that in general pawnbrokers make most money from returning
customers who rollover their loans, explaining pawnbroker investment in maintaining
consumer relationships. It is not in the pawnbroker’s interests to see customers defaulting
and losing their pledges.
Previous behavioural research on the pawnbroking industry is relatively sparse, reflecting
the relatively low use (3% of the population) of pawnbroking loans in UK society. Despite
its low take-up rate, 95% of users of this form of credit rate their experience positively
and report high levels of trust, often resulting in repeated use of this service. Pawnbroking
loans offer customers relative affordability, speed (customers can receive their cash within
10 minutes or less) and access to loans without credit checks. In the US pawnbroking
market, it has been suggested that there are two main behavioural factors driving
customers’ decisions to take out pawnbroking loans (Carter & Skiba, 2012). Firstly,
customers might be conscious of the fact that they suffer from self-control problems when
it comes to paying back debts. This leads them to seek commitment mechanisms (i.e. the
pledging of a sentimental asset, such as a wedding ring) to ensure repayment as the
emotional cost of not retrieving such items outweighs the benefit of defaulting on the loan.
This was evidenced by different repayment rates for loans that were collateralised with
sentimental, rather than non-sentimental pledges of similar values. A study by Bos, Le Coq
and van Santen (2017) also found that customers that are aware of their self-control
limitations choose low loan-to-value arrangements to make it costly to default (Bos, Le
Coq and van Santen, 2017). Secondly, it has been proposed that customers show different
levels of loss aversion with respect to different items (e.g. the subjective feeling of losing
a sentimental item may be greater than losing a non-sentimental item even if they have
the same objective monetary value). As such, reclaiming an item subject to high loss
aversion allows customers to avoid the extra subjective, emotional costs they would incur
if they had defaulted on the loan.
Our behavioural design approach
We approached the problem of low surplus collection rates using an innovative ‘behavioural
design’ approach, blending behavioural insights with data analysis and a design approach

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 10
that starts from an analysis of user needs. This approach consisted of four phases: Discover
(exploration and analysis of user needs), Define (articulating the challenge from the user
perspective), Develop (jointly creating and testing solutions with users) and Deliver
(refining and upscaling solutions with users). Our behavioural design approach can be seen
in Figure 2.
This ‘double diamond’ model was first developed by the British Design Council in 2005
(British Design Council, 2015, Van Essen et al., 2016). At its core this model proposes an
exploration stage of divergent thinking to gain a wider and deeper understanding of a
challenge and then taking focused action in a convergent thinking stage. This happens
twice in the model process – firstly to define the challenge and problem and secondly to
create focused solutions. Critically, the process is non-linear (as opposed to traditional
design processes) and iterative, resulting in an agile approach that has been successfully
used in a number of interventions, from reducing violence in hospital accident and
emergency departments (British Design Council, 2011) to improving cyclists’ safety in the
Netherlands (Van Essen et al., 2016).
Fig 2. An illustration of our behavioural design approach
Our application of these principles differed from previous approaches to developing and
trialling interventions based on intuition or behavioural theory. By looking at the problem
of low surplus collection rates from the ground-up, a richer qualitative picture of the
challenge and a better diagnosis of the problem can be obtained. Combining these insights

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 11
with quantitative data enhances this further – especially in helping to discount
interventions which would assumedly work given previous research. For example, in
previous field trials, using text/SMS messaging has proven to be an effective way of
reminding people of an issue or nudging them into action (Adams et al., 2018). However,
after we conducted staff interviews as part of the “Discover” part of the design process,
we found that less than 20% of customers provided mobile phone numbers, and even then
a large number of these were incorrect or out of date. We were then able to move onto
other intervention ideas which were more likely to work. Our research investigated four
main questions:
1) Why might customers not claim a surplus they are entitled to?
2) What are the current processes and customer journeys for identifying, notifying,
claiming and receiving surplus?
3) What are customers’ needs?
4) What is driving the significant variation in levels of payout by store?
The initial phase of the “Discover” part of our research started with a scoping workshop to
map the customer journey from our perspective and the firm’s, followed by 7 days of user
research in stores across England where the actual experience of customers and store staff
could be assessed. A cross-section of stores were sampled (Table 1) to include stores with
higher/lower than average surplus collection rates, higher/lower incidences of surpluses,
and other variables (London/non-London, urban/suburban and store size). For the
randomised controlled trial all the stores from the firm were eligible (i.e. the RCT was not
confined to the stores sampled at this stage).
Table 1. Store characteristics for user research
>Average % surplus
collected
<Average % surplus
collected
Other variables
>Average number of
surpluses
Store A (90 / 52%)
Store B (77 / 44%)
Store D (234 / 9%)
Store F (51 / 13%)
London / non-
London
Urban /
suburban
Store size
<Average number of
surpluses
Store C (30 / 63%)
Store F (51 / 13%)
Total staff and
Customers
Staff: 19
Customers (not necessarily all in store): 12
A typical customer journey proceeds as follows (a more detailed user journey map can be
seen in figures 3A and 3B). In the loan application and repayment process (figure 3A), a
customer arrives at a pawnbroking store and decides to pawn an item. The pawnbroker
makes a valuation of the item and offers a loan to the customer (typically 50% of the value
of the pledged item) for a short period (often 6 months) at an agreed interest rate (typically
120-160% APR). Pledged items are securely stored and the customer leaves the store with
their loan. If the customer is a repeat customer and their details are already stored in the
system, the whole process can be completed in just 10 minutes. Over the course of the
loan term, customers can make repayments on an ad-hoc basis, according to a
prearranged payment plan, or simply make the full repayment at the agreed date.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 12
In the situation where the customer has not repaid the full amount (prin cipal loaned
amount plus interest) the pawnbroker can offer a discretionary grace period, or a renewal
of the contract. If the full amount is not paid by the end of the grace period, the pawnbroker
can recover their costs by selling the customer’s item. At all stages of the process
pawnbrokers will attempt to communicate with customers by letter or telephone to inform
them of the process, giving customers an opportunity to redeem the item by repaying in
full until quite late stages of the process.
If the customer’s item was worth more than £75, it enters the auction process (figure 3B).
If at auction the item sells at a price above the principal and interest owed by the customer,
then the difference, or ‘surplus’ amount is owed to the customer. If any partial repayments
have been made by the customer during the loan term, then these should also be returned,
after the pawnbroker has recovered their costs. If a surplus is owed to a customer, the
pawnbroker sends a standardised auction surplus notification letter to the customer
alerting them to their surplus and that they can collect it in store.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 13
Figure 3A. Consumer journey from pawning an asset to auction process

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 14
Figure 3B. Consumer journey after the auction process

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 15
User research
At the stores, we carried out interviews with staff and customers, observed transactions,
read example letters and Management Information materials, explored the IT system and
communications (e.g. posters, standard customer letters) and went through the user
journey map with participants. We paid customers a small sum for their involvement in
the research.
We conducted interviews with area managers (n=5), store managers (n=6), store
employees (n=8) and customers in and out of store (n=12), observing consumer
transactions and examining store-level data on surplus collections. Fieldnotes were drafted
by one user researcher and supplemented and verified by the second user researcher. The
fieldnotes were analysed by creating a master journey map illustrating the typical customer
and staff journeys, highlighting differences between stores and practices. A series of pen
portraits of typical and atypical customers were also developed. In parallel, we conducted
analysis of store data on surplus pay-outs (see data analysis section).
Customer interviews showed that the main reason for using pawnbrokers was that it was
a safe, convenient and fast form of short-term credit (to fill short-term cash needs) without
any credit checks. Customers said they used pawnbroking because of its relative
affordability (compared with payday loans and use of unauthorised overdraft facilities),
speed - with the average time to take out a pawnbroking loan being 10 minutes - and
access to loans without credit checks.
“It takes me 30 seconds to explain the [pawnbroking] process.”
Store Assistant
The user research highlighted several drivers that could be driving low surplus collection
rates. All the issues could be construed as side-effects of the positive and beneficial aspects
of pawnbroking loans, namely the speed at which credit is available, the lack of credit
checks and limited personal information required to deliver the credit (since the loan is
collateralised). The personal nature of pawnbroker-consumer interactions was also
reflected in a high number of repeat customers.
“I’ve been using the store for 5 years – I live really locally and have known
the store manager for 13 years.”
Customer interview
For example, a customer might know a particular member of staff and prefer to engage
with them but that staff member may not be aware that the customer is owed a surplus
(whereas a new or unknown staff member may follow a more formal process, such as
accessing the customer’s account and then become aware that the customer has incurred
a surplus).
“If I’m doing the [surplus] letters I’ll think I’ll see them soon but you
never do. Or you see them taking out more gold!”
Store manager

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 16
The behavioural design approach of combining a scoping workshop, user research and data
analysis allowed us to recreate a detailed consumer journey map and identify the potential
pressure points that are driving low surplus collection rates in the process.
Data analysis
In parallel to the user research, we submitted a data request to the firm we were working
with that covered general demographic data of pawnbroking customers, their loan
information, whether they had incurred a surplus and whether that surplus was collected
in the period 2016-2018. This rich dataset enhanced our qualitative user research,
allowing us to determine quantitative drivers of low surplus collection rates. We split the
data analysis into three sections; General demographics of pawnbroking customers,
general demographics of those customers incurring surplus, and factors associated with
low rates of surplus collection. This latter part of the data analysis focused on
understanding what (if any) variables were associated with non-collection of outstanding
surpluses, the demographic variables of these customers and their behaviour when it
comes to accessing pawnbroking services (i.e. repeat custom and store visit data). It
should be noted that our data came from just one firm in the UK market, and may not be
fully representative of the market overall.
General demographics of pawnbroking customers
Corroborating previous research (Collard & Hayes, 2010), there were more female
pawnbroking customers than men in all age brackets apart from those under 25 years,
who represented the smallest number of customers overall. The majority of customers
were between 25 and 55 years old age (Figure 3).
Figure 3. Customer breakdown by age and gender
52% of all pawnbroking customers were classed as repeat pawnbroking customers (defined
as taking out at least one pawnbroking loan at the firm prior to 2016). Additionally, 16%
of customers had used the firm’s pawnbroking shops to access other non-pawnbroking
services such as cheque cashing or buy back services. For all customers, the median loan
amount was £110 (mean amount=£255). Between 2016 and 2018, 82% of all loans were
0
5,000
10,000
15,000
20,000
25,000
Under 25 25-35 36-44 45-55 Over 55
Customer Age & Gender
Male Female

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 17
either paid in full or renewed, and approximately 10% of loans resulted in a customer’s
item being sold at auction (figure 3A). Overall, only 2% of loans resulted in a surplus being
generated for the customer.
General demographics of pawnbroking customers that incur surplus
A more detailed look at the characteristics of consumers that incurred surplus between
2016 and 2018 (our population of interest for this study) revealed that only a minority of
pawnbroking contracts result in surplus being owed. For those customers that incurred
surplus, the median loan amount was £200 (mean loan amount = £397.82) and the
distribution of loan amounts is positively skewed. Loan values varied widely at the regional
level (Figure 4), with mean loan values ranging from £293 (Merseyside) to £531 (South
West London). Part payments were made on 13% of all loans that incurred surplus over
the time period, and the mean part payment was £153 (median part payment = £155).
The overall surplus collection rate across the period was 18.1%. Surplus collection rates
differed by geographical region, from 10% in Scotland to 28% in Merseyside (figure 5).
The mean surplus amount incurred per item was £94.05 (median surplus value = £9.26).
Approximately 80% of customers that incurred surplus were repeat pawnbroking
customers (compared to 52% of all customers). In addition, approximately 6% of
customers that incurred a surplus had more than one loan (or had pawned multiple items
at the same time), which may suggest that customers who incur surplus may be more
frequent or experienced users of pawnbroking services. In addition, customers lived a
mean distance of around 4 miles from the store where they took their loan out, with a
minimum distance of 0 miles and a maximum distance of around 380 miles.
Figure 4. Mean loan amounts at the regional level. Map colours represent the
geographical regions and not any quantitative information.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 18
Figure 5. Surplus collection rates at the regional level. Map colours represent the
geographical regions and not any quantitative information.
We next analysed the frequency of store visits for those customers that incurred surplus.
In order to do this, we examined all customers who incurred a surplus (e.g. their forfeited
item was sold at auction for more than the amount owed) over 2017 (n=5,256). Note that
these customers had opened a pawnbroking loan agreement between January 2016 and
April 2017, reflecting the lag between the eventual auction of their items after defaulting
or any renewal or extension agreements between the store and customer. Of the 5,256
customers that incurred surplus over 2017, 1,066 collected their surplus within a year of
their items selling at auction (representing a 20.2% collection rate over the analysis period,
which is not dissimilar to the surplus collection rate across the whole data collection period
of 18.1%). However, of the customers who did not collect their surplus (n=4,190), 1,582
customers (38% of those that incurred a surplus) had visited a store within 1 year of being
notified that they were owed surplus, but did not collect it. In fact, 50% of these non-
collecting customers visited a store within 50 days of incurring a surplus, increasing to
75% within 110 days.
Factors associated with low surplus collection
To inform us of the potential drivers of surplus (non) collection, we considered those
customers who incurred surpluses over the dataset time period and investigated if there
were statistically significant differences between customers who did and didn’t collect the
money that was owed to them. There were slight differences in the mean and median loan
values for collected and uncollected surpluses (Table 2) and initial loan value was
significantly predictive of surplus collection (logistic regression, z= 6.38, p<0.001).

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 19
Table 2. Loan amounts associated with collected/ uncollected surpluses
Surplus
Status
Unique
Customers
Mean Loan
(£)
Median Loan
(£)
Minimum Loan
(£)
Maximum Loan
(£)
All
7811
397.81
200
75.04
9000
Uncollected
6474
379.59
200
75.04
9000
Collected
1337
480.04
250
76.00
8699
As expected, there were large differences in the average collected and uncollected surplus
amounts (Table 3) and larger surpluses were significantly more likely to be collected than
small surpluses (logistic regression, z= 13.31, p<0.001).
Table 3. Surplus amounts associated with collected/ uncollected surpluses
Surplus
Status
Unique
Surpluses
Mean Surplus
(£)
Median
Surplus (£)
Minimum
Surplus (£)
Maximum
Surplus (£)
All
10036
94.05
9.26
0.02
11195.00
Uncollected
8215
70.02
3.84
0.02
9496.80
Collected
1821
202.45
100.00
0.09
11195.00
Further analysis on the differences between uncollected and collected surpluses revealed
that the mean distance between individual customers and their local store was greater for
those customers with uncollected surpluses (6.6km as opposed to 5.0km for collected
surpluses). In addition, repeat customers are more likely to collect a surplus that is owed
to them, with surplus collection rates at 20% for repeat customers compared to 13% for
new customers (logistic regression, z= 6.44, p<0.001). Similarly, customers that had
made part payments throughout the loan period were four times more likely to collect a
surplus that is owed to them than those customers who did not make any part payments
(logistic regression, z=31.8 p<0.001). Those customers that had opted in to be contacted
by SMS or telephone were not significantly more likely to collect surplus (logistic
regression, z= 1.54 p=0.124). In summary, the following factors were associated with low
surplus collection were discerned from the data:
• low initial loan amounts
• low surplus amounts
• increased distance between the customer and their store
• new customers and customers that hadn’t made any partial payments over the
course of the loan term

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 20
Our interventions were based on hypotheses generated from the combination of user
research, data analysis and the synthesis workshop with the firm where key qualitative
and quantitative insights from the research were reviewed. There were 13 participants in
the synthesis workshop including 6 firm employees across functions including
management, marketing, training and store/area managers. The workshop focused on 4
exercises: reviewing the user research and pen portraits, data analysis and detailed
customer journey map; refining and validating the customer journey map and insights;
developing and prioritising hypotheses; and generating and prioritising ideas for each
hypothesis. These ideas and interventions were prioritised according to feasibility and
estimated impact.
Drivers of low surplus collection
The drivers are divided into 2 sections: communications and processes. These were further
distilled into five overarching themes (D1-D5) that our intervention was designed to
address:
D.1 Frequency and quality of communication with customers
A customer may only receive one letter prior to their pledge being sent to auction. The
appearance and language of the letter is legalistic and not easy to understand at first
glance. One letter is sent to customers advising them they are owed money – this is also
legalistic and dense (and its format is prescribed by legislation). Phone calls are time
consuming for staff to make and the store may not have the customer’s phone number. In
addition, it is difficult to know whether the letters are reaching customers’ addresses or
indeed even being opened and read, and phone calls are often unanswered.
Very often customers do not provide (correct) contact details such as phone numbers and
email addresses or they may change their address but not update the pawnbroker. Or they
may not provide it at all – there is variation within and across stores on what details are
collected and checked. This could result in letters being sent to a past address or not being
able to make reminder calls, for example.
“Calls are not mandatory. We do it out of the goodness of our own hearts...If
you’re busy you can’t do them... When we do get through, sometimes people say
they haven’t got our letter – they’ve moved and haven’t updated their address
with us. Regular customers always tell us if they’ve moved.”
Staff member
“Today I made 20 ‘courtesy calls’ to let people know their contracts are expiring
and their items will be sold at auction. I only got through to 6 or 7 people.”
Store Assistant
4. Hypotheses, intervention design and
experimental method

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 21
D2. Lack of customer understanding/ knowledge of both the surplus and the
auction process
There is a great deal of information to take in when pledging an item of value. There may
be limited understanding among customers of the circumstances under which a surplus
(including part-payments) might be paid. More specifically, the word ‘surplus’ may also not
be clear for people in this context. There may also be a lack of understanding or awareness
of the entire process; being owed a surplus is entirely counter to peoples’ mental models
of borrowing money from a firm. The length of time of the whole process means that some
people might simply forget about the good(s) or the money and adapt to life without it.
“I don’t read the small print. I would read the main points. I usually sign without
reading... [After explaining about the surplus] I wouldn’t have thought that would
be possible… I’ve always read it that you would just lose the items. I assumed that
was how they make their money.”
Customer interview
D3. Auction process is very manual and time-consuming for staff
The auction process requires manual cross-checking of multiple documents then inputting
the relevant information into the computer system by the store manager or employees of
a certain grade. Staff reported the process as the most effortful/onerous part of the whole
customer journey. Staff were also frequently pulled away from back office tasks like this
to serve customers. This could potentially lead to errors and letters not being sent or the
incorrect amounts/information being sent.
“The hardest part is processing everything for auction. Got to make sure everything
is correct – weight, hallmark, description. Only I do this. If I’m not disturbed to
serve it takes me about one to two hours...”
Store Manager
D4. No staff reward or incentives for increasing collection rates.
Staff are highly motivated to achieve certain targets and KPIs through various rewards.
Rewards are both financial – e.g. individual bonuses or team holidays – as well as social –
e.g. praise, respect and status. Despite the surplus issue being an increasing challenge,
there are no rewards or incentives for staff or stores with higher collection rates.
“We have KPIs for pawnbroking, cash loan conversion rates, retail, foreign currency
exchange, buy-backs, and purchase....We have an internal audit system where 1
star is what you want and 5 stars is the lowest/worst.” Area Manager
of store with surplus collection rate below average
D5. Limited attention, salience and/or knowledge of the problem and, more
generally, surplus rates, at store, regional and Head Office level.
There was limited or no awareness attributed to the scale of the challenge, particularly
among store staff. Knowledge of the surplus process was generally lower compared to the
rest of the customer journey – for example, many staff did not know part-payments can
make up the surplus. There is a widely held (false) belief in stores that all customers
collected their surplus. Surplus collection rates are not covered in the firm MI e.g. mystery
shopping, audits, area manager checklists or KPIs. There is no way to quickly see at-a-

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 22
glance how many and which customers are owed surplus or who is owed surplus at an
individual level.
“Everyone comes back to collect their money – why wouldn’t you?” Store
Manager of store with surplus collection rate below average
Intervention Design
To capture emerging ideas, throughout each session, participants were encouraged to note
down ideas and place them on the section of the journey map to which they corresponded.
In addition, there was a structured ideas generation session following the hypotheses
generated by the user research. The groups generated more than 50 ideas to address the
various hypotheses. These were refined into a list of 37 distinct concepts and categorised
according to higher or lower impact and higher or lower feasibility (Table A1)
After the synthesis workshop, user research and data analysis we focused on interventions
that could be tested scientifically in randomised controlled trials (RCTs). Our intervention,
designed to address surplus non-collection issues on the consumer side, consisted of a
surplus reminder letter designed using behavioural science techniques. This letter was sent
14 days after a customer incurred a surplus after their item was sold at auction, in addition
to the standard legal letter that is sent.
The second intervention, designed to address issues on the firm-side, consisted of a
modification to the store software system that keeps track of customer’s relationships with
the pawnbroker, including outstanding loans and whether the customer has incurred a
surplus when their pledged item has been sold. If a customer visits a store for any purpose
that requires their account to be accessed (not limited to pawnbroking services), a surplus
flag pop-up appears to alert the staff member if the customer has an uncollected surplus.
Intervention 1: Surplus Reminder Letter to Customers
As part of the “Develop” phase of the design process, we designed 4 different variants of
a surplus reminder letter that took into account all of the insights generated in the Define
and Discover phases. We focused on conveying the message that a customer was owed
surplus using: 1) A (basic) concise letter 2) A visually direct letter 3) A letter with a visual
portrayal of why the customer was owed money, and 4) A letter that portrayed why the
customer was owed money in a non-visual way. On the reverse side of the four letters was
a table specifying the items that the customer had incurred surplus on, and a map to locate
their nearest store. In addition, we tested a number of different envelopes for the letters.
We visited a store and received feedback from 5 customers and 5 staff members to decide
which letter would be most effective to utilise in the larger scale field trial. This in-store
testing followed the same format for every customer and staff member. We first explained
the objective of the testing and described a hypothetical situation for the customer (that
they had pawned an item but haven’t been able to repay their loan so they have lost the
item. A couple of weeks later, they collect their post). We then presented the customer or
staff member with a selection of 3 envelopes and asked which one they were inclined to
open and why. Next, the customer or staff member was shown one letter and asked to
read it (with letters presented in random order for each customer and staff member). We

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 23
asked if they understand the contents, the pros and cons of the letter, and how it could be
improved. The test subject was then shown all of the letters and asked to repeat the
previous step.
We found that the most effective combination was a blue envelope, which attracted the
attention of the customers, and a visually direct letter. Notably, none of the participants
looked at the reverse side of the letter so this intervention was removed from the final
version that was used in the trial. An example of the letter used in the field trial can be
seen in Figure 6.
Figure 6. Final letter design for testing in the randomised controlled trial.
Intervention 2: Surplus Flag to Alert Staff Members
Our analysis of store visit data as part of the design process suggested that many
customers who have outstanding surplus amounts on their accounts subsequently visit the
store without collecting them. Customers also visited stores to access a variety of other
services (such as cheque cashing or foreign exchange) and sometimes these services
require a staff member to consult the customer’s account on the system.
The analysis showed that a majority of pawnbroking customers were repeat users of this
form of credit. Indeed, 80.1% of items that incurred surplus were pawned by repeat
customers, and 52% of all the customers in our dataset were repeat customers, a finding

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 24
that is supported in the literature documenting pawnbroking loan use (Collard & Hayes,
2010). Our analysis of customers’ visit data to pawnbroking stores suggested that 37.8%
of customers who are owed a surplus visit a store within a year of receiving a surplus
notification letter, but do not collect it. The median delay from surplus notification to next
visit is 50 days, suggesting that most customers would have received a surplus notification
letter prior to their next store visit.
We therefore proposed that an automatic surplus flag to alert staff members that a
customer has an uncollected surplus could increase collection rates from 20.2% to 50.4%
(assuming that the surplus flag is 100% effective in facilitating staff members to process
the surplus collection). Although it is technically possible to see if a surplus is owed to a
returning customer in the customer database, staff would previously be required to click
into the database to look for it manually. Because this trial is still in the field due to the
impact of Covid-19, we will publish these results in a follow-on research paper.
Method (Surplus Reminder Letter)
To test the effectiveness of the new reminder letter, we conducted a randomised controlled
trial with randomization occurring at the customer level. One day post-auction, we received
an anonymised list of all unique customers IDs that had incurred a surplus every 14 days.
We then randomly allocated unique customer IDs to the treatment group (receive the
reminder letter 14 days post-auction) or control group (no change from normal process).
As such, the sample population for the trial was those customers who incurred surplus in
an auction, as determined by the store computer database centrally. Prior to allocation,
the customer IDs were checked against previous treatment/control group allocations to
ensure that customers who repeatedly incurred surplus over the trial period were assigned
to the same group.
A central hypothesis of the research was that the initial surplus notification letter was
ineffective in encouraging most customers to collect surplus (relating to driver D1. as
identified in the user research). Of those customers that do collect, the majority do so
within a short time-frame (see data analysis section above). We therefore sent a reminder
letter to customers within 14 days of it being incurred.
Because the items placed in auctions each month are selected on a random basis (i.e. not
determined by geographic location or value) we did not assign customers to treatment or
control groups based on any stratification variables.
There was one categorical outcome variable: whether or not the surplus has been paid out
to the customer within one month of receiving the letter and one continuous outcome
variable: monetary amount collected. Our primary hypotheses were that the reminder
letter would: 1) Increase the number of surpluses being collected by customers within one
month of receiving the letter, and 2) Increase the monetary amount collected by customers
within one month of receiving the letter. In addition, our secondary hypotheses were that
the reminder letter 3) Have an interacting effect with the amount owed to consumers
(aggregated) on increasing collection, and 4) Have a larger impact on collection rates for
repeat customers. Multiple logistic/ linear regressions were used to assess the impact of
the intervention on the outcome variables.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 25
For a parallel arm design, approximately 210 participants were required per group to detect
an effect size of 10%, from the low baseline of ~10% surplus collection within 30 days. To
detect an effect size of 15%, approximately 100 participants were required per group.
These sample size calculations were based on α=0.05 and β=0.2). These samples are
estimated based on an increase in the rate of surpluses paid out, irrespective of the fact
that some customers incur multiple surpluses in the same auction.

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 26
Surplus Reminder Letter
Over the course of 2 months 569 customers who incurred a surplus entered into the trial.
Customers were randomly allocated to control and treatment groups. 277 customers
received a behaviourally designed reminder letter informing them of the outstanding
surplus (treated group). 292 customers were assigned to the control group. We estimated
effect sizes to be in the 5-15% range based on the firms previous in-house research and
the results of our previous disclosure-based RCTs. Treatment and control groups were
balanced, with no significant differences in the previously identified drivers of surplus
collection such as loan value, surplus value, distance to the store and the number of new
and repeat customers in each arm (Table 4).
Table 4. Drivers of surplus collection rates were not significantly different
between control and treatment groups.
Group
Mean
loan
value
(£)
95% Loan
value
confidence
intervals (£)
Mean
auction
surplus
value (£)
95% Auction
surplus
confidence
intervals (£)
Mean
store-
customer
distance
(km)
95 %
Distance
confidence
intervals
Repeat/
new
customer
Control
383.65
326.58,
440.71
83.42
39.83,
127.01
4.8
3.5, 6.1
236/56
Treatment
408.63
339.95,
477.31
96.90
66.20,
127.60
7.6
4.3, 10.8
217/60
The behaviourally-designed reminder letter had a significant impact, with 22% of people
collecting their surplus in the treatment group compared to 11% in the control group,
within 30 days of incurring a surplus (Figure 7). This result supported our first hypothesis
that the reminder letter would increase surplus collection, even when controlling for surplus
amount, loan amount and distance to the store, all of which were determined to affect
surplus collection rates from our initial data analysis exercise (multiple logistic regression,
z=3.403, p<0.001, table A2). A linear probability model confirmed this result (t=3.465,
p<0.001). In both models, only auction surplus amount came close to significance in
explaining collection rates (p=0.09 and p=0.06 respectively). A regression controlling for
geographic region confirmed that the sample was evenly distributed across geographic
region (z=0.003 p=0.99). Our regression analysis showed that a customer is almost twice
as likely to collect their surplus having received our letter than not. For context, a previous
FCA RCT showed that a letter trial looking at increasing redress rates had a much smaller
effect size of maximum 4 percentage points, when the letters were simplified with salient
messaging (Adams & Hunt, 2013).
Moreover, the percentage total value of surpluses ‘left on the table’ was reduced from 79%
in the control group to 66% within 30 days in the group that received the reminder letter,
5. Results

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 27
supporting our second hypothesis that more money would be returned to customers
compared to the control group (Figure 8). In testing our secondary hypotheses (that the
reminder letter would have a greater impact on repeat customers and for higher surplus
amounts) we found there was no effect (multiple logistic regressions, z=-0.176 p=0.86
and z=0.749 p=0.45 respectively).
Figure 7. Surplus collection rates for the control group and the treatment
group (error bars denote 95% confidence intervals)
Figure 8. Surplus amount returned to customers for the control group and
the treatment group
Surplus Flag
Due to Covid-19, the surplus flag trial had to be halted. We will provide the results of
this trial in a further report once the trial is concluded.
0%
5%
10%
15%
20%
25%
30%
Control Treated
Surplus collection rate (within 30 days)
0
2000
4000
6000
8000
10000
Control Treated
Total collected surplus (£)

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 28
When pawnbroking customers default on their loan, the item that they pledged as collateral
can be sold to recover the loss to the pawnbroker. However, if the item sells for more than
what the customer owed, this surplus money should be returned to them. However, our
analysis in 2018 found that many customers had not collected this surplus money and the
harm to consumers may be in the region of £1m a year. Academic reviews suggest that
many pawnbroking customers may be classed as vulnerable, so we are working with a
large pawnbroking company, looked at designing a remedy to drive up collection rates for
these people.
Using a novel design approach combined with extensive data analysis and user research,
we mapped the full consumer pathway and discovered that a key problem that was driving
non-collection was a lack of understanding of the surplus process on the consumer side.
Although statutory letters are sent out to consumers when they are owed a surplus, these
letters might be difficult to understand or could be ignored by consumers.
The behavioural design approach helped us to create several reminder letters, combining
new messaging and visual techniques to explain to consumers that they were owed money
and could come to the store to collect it. User testing allowed us to narrow these letters
down to one that would be sent out 14 days after a customer’s item was sold at auction
and a surplus was generated. We tested the effectiveness of this letter in a randomised
controlled trial (RCT), where half of the customers received the statutory letter and the
new reminder letter, and the other half just received the statutory letter.
Our results indicate that the reminder letter caused a statistically significant increase in
collection rates within 30 days and reduced the amount of surplus money being held by
the pawnbroker. Encouragingly, this effect was independent of the surplus amount and
whether the customer was a new or repeat customer, 2 drivers that were previously
identified to affect collection rates. This simple intervention may reduce harm to vulnerable
consumers in this market and could be implemented cheaply by firms of any size.
As mentioned previously, we also designed and are testing an intervention on the firm
side: the introduction of a flag that alerts staff to a customer’s surplus. The introduction of
public health interventions during the Covid-19 pandemic affected the timelines for the
surplus flag field trial, since the intervention relied on random return visits to stores that
were closed for a period in the Spring and early Summer of 2020. We commit to publishing
the findings of this trial in a future research note.
6. Conclusions

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 29
Annex 1.
Table A1. Ideas, hypotheses and feasibility/impact assessment
Idea
Hyotheses
Impact/
Feasibility
Text message reminders
Comms (D1, D2)
High impact,
low feasibility
Automatic flags on Neo when customer comes
in
Comms (D1, D2, D5)
High impact,
high feasibility
Take bank details and automatically refund
surplus into account
Comms (D1, D2)
High impact,
low feasibility
Phone customers outside of work hours
Comms (D1, D2)
High impact,
low feasibility
Send letters in customers' own languages
Comms (D1, D2)
High impact,
low feasibility
Origin of surplus within training (e-learning,
classroom, monthly doc)
Limited firm attention
/ salience (D5)
Low impact,
high feasibility
Phone calls after auction when surplus owed
Comms (D1, D2)
High impact,
high feasibility
Rename surplus: "refund", "cashback"
Comms (D1, D2)
High impact,
high feasibility
Mandatory fields for comms preferences
(languages and or channels)
Volume and quality of
customer data (D3)
Low impact,
high feasibility
Store/area manager driving culture and
behaviour e.g. including surplus in their store
checklists / making it a focus
Limited firm attention
/ salience (D5)
Low impact,
high feasibility
Store certificates/award for best improvement
in collection rates
Incentives and limited
firm attention /
salience (D4, D5)
High impact,
high feasibility
Simplify auction settlement/central settlement
(e.g. creating one sheet showing all sold
items)
Incentives (D4)
Low impact,
low feasibility
Include surplus in audit process
Incentives and limited
firm attention /
salience (D4, D5)
Low impact,
high feasibility
Engaging message on surplus letter envelope:
"You are owed money"
Comms (D1, D2)
High impact,
high feasibility
Make surplus collection a KPI
Incentives and limited
firm attention /
salience (D4, D5)
High impact,
high feasibility
Store surplus collection rate
ranking/competition
Incentives and limited,
H&T attention /
salience (D4, D5)
Low impact,
high feasibility
Compliance focus - make this issue part of
compliance strategy which gets C-suite
attention
Limited firm attention
/ salience (D5)
High impact,
high feasibility

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 30
Head office marketing - website stats, Power
BI, posters (MI)
Limited firm attention
/ salience (D5)
Low impact,
high feasibility
Note unsuccessful contact with customers on
system
Volume and quality of
customer data (D3)
Low impact,
high feasibility
Automatic auction settlement
Incentives (D4)
Low impact,
low feasibility
Centralise letter production
Incentives (D4)
High impact,
low feasibility
Letter envelopes to include details of sender
Comms (D1)
Low impact,
high feasibility
Put info about surplus on website
Comms (D1, D2)
Low impact,
high feasibility
Provide map of consumer journey incl. surplus
Comms (D1, D2)
Low impact,
high feasibility
Tell customers about surplus when they take
out the loan
Comms (D1, D2)
Low impact,
high feasibility
Re-print surplus letters for non-collectors until
surplus collected (make it the default)
Comms (D1, D2)
Low impact,
high feasibility
Check customer's history every visit
Comms (D1, D2)
Low impact,
high feasibility
Tell customers about surplus in auction letter
Comms (D1, D2)
Low impact,
high feasibility
Emphasise best practice of keeping letters that
are returned to sender
Comms (D1, D2)
Low impact,
high feasibility
Put info about surplus on till receipts
Comms (D1, D2)
High impact,
high feasibility
Signs in store about process and surplus
Comms (D1, D2)
High impact,
high feasibility
Combine multiple surpluses on letters
Comms (D1, D2)
High impact,
high feasibility
Allow customers to get refunds of part
payments during loan
Comms (D1, D2)
High impact,
low feasibility
Word of mouth education/community groups
as messengers
Comms (D1, D2)
Low impact,
low feasibility
Send process survey to customer
Comms (D1, D2)
Low impact,
low feasibility
Focus on customers who intended to sell but
pawned instead
Comms (D1, D2)
Low impact,
low feasibility
Have an online account to check progress and
collect surplus
Comms (D1, D2)
Low impact,
low feasibility

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 31
Annex 2.
Table A2. Regression table
Dependent variable:
Collected
logistic
OLS
(1)
(2)
Treatment
0.808***
0.110***
(0.237)
(0.032)
Auction surplus value
0.0005*
0.0001*
(0.0003)
(0.00005)
Distance between customer and store
-0.026
-0.001
(0.019)
(0.001)
Constant
-1.959***
0.116***
(0.202)
(0.023)
Observations
550
550
R2
0.031
Adjusted R2
0.025
Log Likelihood
-242.536
Akaike Inf. Crit.
493.072
Residual Std. Error
0.372 (df = 546)
F Statistic
5.767*** (df = 3; 546)
Note:
*p<0.1; **p<0.05; ***p<0.01

Sitting on a gold mine: Getting what’s owed to pawnbroking customers
20 January 2021 32
References
Adams, P., & Hunt, S. (2013). Encouraging consumers to claim redress: evidence from a
field trial. FCA Occasional Paper, (2).
Adams, P., Kelly, D., Grubb, M., & Osborne, M. (2018). Time to act: A field experiment
on overdraft alerts. FCA Occasional Paper, (40).
Agarwal, S., & Bos, M. (2019). Rationality in the consumer credit market: choosing
between alternative and mainstream credit. In Handbook of US Consumer Economics
(pp. 121-139). Academic Press.
Banathy, B. H. (2013). Designing social systems in a changing world. Springer Science &
Business Media.
Bos, M., Le Coq, C., & van Santen, P. (2017). Scarcity and Consumers’ Credit Choice.
Swedish House of Finance Research Paper, (16-19).
British Design Council. (2011). Reducing Violence and Aggression in A & E: Through a
Better Experience. Design Council.
British Design Council (2015). What is the framework for innovation? Design Council’s
evolved Double Diamond. Retrieved from https://www.designcouncil.org.uk/news-
opinion/what-framework-innovation-design-councils-evolved-double-diamond
Carter, S. P., & Skiba, P. M. (2012). Pawnshops, behavioral economics, and self-
regulation. Rev. Banking & Fin. L., 32, 193.
Caskey, J. P. (1994). Fringe banking: Check-cashing outlets, pawnshops, and the poor.
Russell Sage Foundation.
Collard, S., & Hayes, D. (2010). Pawnbroking Customers in 2010: A Survey–A Report to
the National Pawnbrokers Association.
Edwards, L., & Lomax, W. (2017). Financial credit and social discredit: the pawnbroking
dilemma. Journal of Financial Services Marketing, 22(2), 77-84.
Financial Conduct Authority (2018, July 10). Pawnbroking sector review. Retrieved from
https://www.fca.org.uk/publications/multi-firm-reviews/pawnbroking-sector-review
Van Essen, A., Hermsen, S., & Renes, R. J. (2016). Developing a theory-driven method
to design for behaviour change: two case studies. Proceedings of DRS 2016, 2016(4),
1323-1337.