ArticlePDF Available

Abstract and Figures

It has been known for some time that the specific payment mechanism used by consumers when making purchases has a significant effect on spending behaviour. This seems to be particularly true of credit cards: customers using credit cards spend more than those paying by cash or cheques in spending situations that are otherwise identical in every other respect. 1 That this should be the case is also evidenced 2 by the willingness of retailers to pay credit card companies some 3% of their revenues on credit card purchases. This customer behaviour is at odds with standard economic theory, which argues that the method of payment should have no effect on spending, so consumers seem to be indulging in "irrational" behaviour. But recent research discussed in this paper indicates that differences in the method of payment have psychological side-effects that influence consumer behaviour in predictable ways. Understanding these psychological side-effects has a number of important implications for financial services companies. Firstly, there is the benefit of increased understanding of customer behaviour, which should allow greater accuracy in modelling and prediction of spending patterns across different customer groups. This knowledge would assist in planning, pricing and the targeting of product offerings. More significantly, however, knowledge of consumer psychology gives financial institutions a direct means of influencing customer behaviour through the manipulation of individual features of payment mechanisms. This could be achieved by either changing product designs or using marketing to change the perception of product features in ways that would stimulate changes in behaviour. The obvious benefits of such understanding are increased revenues through higher customer spending, and increased market share through judicious targeting of psychological aspects of spending behaviour. A further benefit, however, lies in risk management: an understanding of the psychology may assist in designing product features that reduce overspending, and thus limit both adverse selection and the risk inherent in offering credit to those with low levels of financial self-control. Credit card users are particularly prone to such failures of self-control, and may well be prepared to pay a premium for credit structures that help them to stay in charge. Such products may also benefit banks and insurers in terms of increased brand loyalty 3 . A comprehensive understanding of spending psychology should thus allow financial services companies to improve planning and pricing accuracy, and to encourage spending whilst simultaneously limiting the likelihood of default. Irrational spending The first factor that helps to explain why credit cards boost spending is to do with the psychophysics of how the brain subjectively perceives external stimuli. In this case the stimulus of concern is money, and the psychophysics of value suggests that an expense is perceived as being far larger on its own than when it is part of a much bigger payment. So adding a £30 expense to a £637 credit card bill will make the £30 appear smaller than it would seem on its own. Grouping transactions in credit card bills therefore reduces the perceived size of individual expenditures, resulting in increased spending. 4 A second issue is that, according to the standard
Content may be subject to copyright.
SEARCHING FOR THE RIGHT QUESTIONS
22 Volume 2 Issue 6 Argent www.thefsforum.co.uk
It has been known for some time that the specific payment
mechanism used by consumers when making purchases
has a significant effect on spending behaviour. This seems
to be particularly true of credit cards: customers using
credit cards spend more than those paying by cash or
cheques in spending situations that are otherwise identical
in every other respect.1That this should be the case is also
evidenced2by the willingness of retailers to pay credit card
companies some 3% of their revenues on credit card
purchases. This customer behaviour is at odds with
standard economic theory, which argues that the method
of payment should have no effect on spending, so
consumers seem to be indulging in "irrational" behaviour.
But recent research discussed in this paper indicates that
differences in the method of payment have psychological
side-effects that influence consumer behaviour in
predictable ways.
Understanding these psychological side-effects has a
number of important implications for financial services
companies. Firstly, there is the benefit of increased
understanding of customer behaviour, which should allow
greater accuracy in modelling and prediction of spending
patterns across different customer groups. This
knowledge would assist in planning, pricing and the
targeting of product offerings.
More significantly, however, knowledge of consumer
psychology gives financial institutions a direct means of
influencing customer behaviour through the manipulation
of individual features of payment mechanisms. This could
be achieved by either changing product designs or using
marketing to change the perception of product features in
ways that would stimulate changes in behaviour. The
obvious benefits of such understanding are increased
revenues through higher customer spending, and
increased market share through judicious targeting of
psychological aspects of spending behaviour.
A further benefit, however, lies in risk management: an
understanding of the psychology may assist in designing
product features that reduce overspending, and thus limit
both adverse selection and the risk inherent in offering
credit to those with low levels of financial self-control.
Credit card users are particularly prone to such failures of
self-control, and may well be prepared to pay a premium
for credit structures that help them to stay in charge.
Such products may also benefit banks and insurers in
terms of increased brand loyalty3.
A comprehensive understanding of spending
psychology should thus allow financial services
companies to improve planning and pricing accuracy,
and to encourage spending whilst simultaneously
limiting the likelihood of default.
Irrational spending
The first factor that helps to explain why credit cards boost
spending is to do with the psychophysics of how the brain
subjectively perceives external stimuli. In this case the
stimulus of concern is money, and the psychophysics of
value suggests that an expense is perceived as being far
larger on its own than when it is part of a much bigger
payment. So adding a £30 expense to a £637 credit card
bill will make the £30 appear smaller than it would seem
on its own. Grouping transactions in credit card bills
therefore reduces the perceived size of individual
expenditures, resulting in increased spending.4
A second issue is that, according to the standard
The realities of spending
It is obvious that customers often behave irrationally, but the
psychological factors that affect spending behaviour are many and
complex. Greg Davies discusses some of the latest thinking in this
area from Cambridge and Warwick, and shows how marketers might
use these insights to improve product design and presentation.
Customer behaviour
SEARCHING FOR THE RIGHT QUESTIONS
www.thefsforum.co.uk Argent Volume 2 Issue 6 23
economic model of decision making, a rational consumer
should purchase goods or services if the utility from
consumption is greater than the present value of the
payments associated with that consumption. So the
longer that payments are delayed, the better the
consumption experience for the consumer, as the cost of
consumption is cheaper. The means of payment should
have no effect on the likelihood of purchase.
However, recent studies5provide numerous counter-
examples:
A holiday costing £1200 is arranged six months in advance.
There are two payment options:
1. Six monthly instalments of £200 during the six months
before the vacation.
2. Six monthly payments of £200 in the six months after
the holiday.
According to the standard model, everyone should opt
rationally for option 2, as it carries an implicit interest bonus
of about £20. However, when Prelec and Loewenstein
posed this question to ninety-one individuals, six in ten
opted for the earlier payments. The academic literature of
choice behaviour is full of similar examples, indicating that
people do not, in general, act in line with the theory. More
important than this realization, though, is an understanding
of why it should be the case.
Mental accounting
Possibly the most important concept to come out of recent
research into the cognitive psychology of financial
behaviour is the recognition that people intuitively organize
their finances into "mental accounts". Mental accounting
may be seen as analogous to the accounting and
budgeting systems used by companies – a "set of
cognitive operations used by individuals and households to
organize, evaluate, and keep track of financial activities", as
Thaler6put it. This helps to reduce the complexity of
everyday financial decisions and assist in making trade-offs
between different types of spending. Although the theory
of mental accounting covers a very rich set of human
behaviour, this paper focuses on how people combine the
costs and benefits of individual transactions when making
spending decisions.
At the transaction level, consumers tend to “open” an
account mentally for each transaction as it occurs, and
to base decisions about whether to purchase or not on
an evaluation of the perceived benefits of consumption
and the costs of payment in this account. The
advantage of this approach is that it enables decisions
to be evaluated in isolation from other transactions,
reducing the cognitive load on the decision maker, and
making the assessment easier.
However, this simplicity comes at a cost: the decision
becomes mentally disassociated from other decisions
which might more appropriately be evaluated together.
These accounts are mental constructs, which means
that the perceived costs and benefits are evaluated in
the context of the mental account, and are themselves
open to psychological biases.
To make the concept more tangible, consider the
following pair of questions7:
1. Imagine that you have decided to see a play and have
paid the admission price of £10 per ticket. As you enter
the theatre, you discover that you have lost the ticket.
The seat was not marked, and the ticket cannot be
recovered.
Would you pay £10 for another ticket?
2. Imagine that you have decided to see a play where
admission is £10 per ticket. As you enter the theatre, you
discover that you have lost a £10 note.
Would you still pay £10 for a ticket for the play?
When this question was administered to two groups of
subjects8, less than half answered yes to the first question,
whereas nearly 90% answered yes to the second. These
responses are problematic for the standard model of
behaviour: in both cases total wealth and the future costs
and benefits of the decision are identical. However, there
is clearly a significant psychological difference between
the two choices. Viewed from a mental accounting
perspective, the choices make a great deal more sense.
In the first question, the price of the second ticket is
added to the open "theatre account", in which the first
ticket has already been booked. For many people, this is
evidently an unacceptably high cost for the play. In the
second case, the money lost is not perceived as being
related to the play, and the "theatre account" is evaluated
in isolation of this amount – it will not influence the
decision to see the play, except insofar as it makes the
person feel slightly less wealthy.
Prelec and Loewenstein9use mental accounting to
provide both a theory of the effect of payment
mechanisms on spending and evidence supporting it, by
introducing the concepts of prospective accounting and
coupling. Within each "transaction account", the
psychological evaluation of the transaction can be broken
down into the perceived benefits of consumption and the
perceived costs of paying. Unlike the standard economic
theory, these two components are affected both by each
other and by the timing of the consumption acts and
payments. Specifically, within each transaction account,
thinking about the cost of a purchase at the time of
consumption can undermine the pleasure derived from it;
thinking about the benefits of a purchase can blunt the
pain of making payments.
The psychological evaluation of the
transaction can be broken down into
the perceived benefits of consumption
and the perceived costs of paying.
SEARCHING FOR THE RIGHT QUESTIONS
24 Volume 2 Issue 6 Argent www.thefsforum.co.uk
Prospective accounting
This framework can explain the holiday example given
above. People tend to evaluate both payment and
consumption prospectively – that is, at any given point
they are far more concerned with future outcomes than
past occurrences. Thus, they prefer to prepay for the
holiday because:
• At the moment of payment, the “pain of payment” is
buffered by the thought of the future holiday to come.
During the holiday itself, the payments have already
been made so, the holiday can be enjoyed fully without
thinking about the costs to come – it seems "free".
Paying for the holiday after the event, on the other
hand, means that during the holiday the pleasure will be
reduced by the thought of the unpaid bills that will be
waiting afterwards. In addition, when actually making the
payments there is no future benefit to buffer the pain – it
feels as if the payment is for "nothing".
Given this framework, people would in general prefer to
prepay for consumption items than pay after the event,
and consumption spending should thus be higher when
payments precede consumption. This will not always be
the case, however, because the effect is offset by the
financial benefits of delaying payment and receiving
interest on the outstanding amount. In some cases, the
impact of time discounting will be strong enough to over-
ride the benefits of prepayment. This has the interesting
implication that the net desire for prepayment may be
weaker during periods of high interest rates.
The strength of the prospective accounting effect also
differs according to the nature of the purchase. In
particular, the effect is stronger for consumption that is
intense and short-lived rather than longer term.
Consumer durables, for example, provide long-term
consumption benefits so, even if the payments are
spread out over an extended period, the pain of these
payments may always be offset by the benefits of use
that will continue well into the future10.
Coupling
Consumers thus have a psychological preference for
prepayment, which is offset to some degree by time
discounting. Bringing forward payment schedules tends
to raise customers’ evaluations of the net benefits of a
transaction, and hence causes an increase in overall
expenditure. However, it is not yet clear why this should
result in the observed effect of credit cards increasing
expenditure. Credit cards, after all, are about post-
payment. The explanation lies in another psychological
effect – coupling.
Coupling refers to the degree to which consumption
and payment are bound together psychologically. For
any purchase, the thought of future payments will
attenuate the pleasure of consumption. However, the
degree of attenuation will depend on how readily these
future payments can be brought to mind (or come to
mind unbidden!). The more the payments can be
psychologically “decoupled” from the consumption, the
less they will reduce enjoyment and the better the
overall assessment of the value of the transaction.
This decoupling may be achieved in a number of
ways, and credit cards come into their own here as a
means of reducing the “pain of payment”:
An increased time period between purchase and
payment will tend to reduce coupling. If the
payment is viewed as being far in the distance at
the time of consumption, it will have less impact on
perceived consumption benefits.
Combining many purchases into a single payment
will also reduce coupling. When settling a credit card
bill, several individual purchases are combined into a
single payment. At the point of consumption, then,
there is no single clear payment that is attributable
specifically to that consumption act. An immediate
payment for a specific purchase, on the other hand,
creates very tight coupling, and the payment event is
more readily brought to mind, reducing the pleasure
to be had from consumption.
The greater the diversity of consumption types
associated with any payment, the greater the
decoupling. If a payment covers a wide range of
different goods and services, it becomes even more
difficult psychologically to partition the payment
between them.
• Lower salience and vividness of the payment lead
to lower coupling. Cash payments have high
psychological salience, as they involve handing
over a visible amount of cash, driving home the fact
that the benefits are being paid for. Other means of
payment have lower salience – cheques do not
have quite the vividness of cash payments, but do
require the consumer to write out the amount
physically, which serves to imprint this
psychologically. Credit cards produce lower
coupling, as they merely require a signature on a
slip of paper – the imprinting of the payment
amount as a memory trace is thus much weaker11.
All of these factors reduce the degree to which the
thought of the payment attenuates the benefits of
consumption, leading to a higher evaluation of the
transaction, and thus increased propensity to spend.
There is, however, a countervailing factor: most of these
elements that reduce coupling will also reduce the
degree to which the thought of consumption buffers the
pain of paying.
So, whilst the enjoyment of consumption is not
reduced much by the payments, the pain of payment is
not reduced much by the thought of consumption either.
This reduces the overall evaluation of the transaction and
thus lowers the likelihood of purchase. An example can
SEARCHING FOR THE RIGHT QUESTIONS
www.thefsforum.co.uk Argent Volume 2 Issue 6 25
be seen in Figure 1. The upper diagram shows the
temporal coupling relationship for a cash purchase. Both
the simultaneous timing and the one-to-one relationship
cause high coupling.
The middle diagram demonstrates that paying by
charge card reduces the degree to which the payment
attenuates the pleasure of consumption: the payment is
separated from the associated consumption events by
both timing and the fact that many items are bundled
into a single bill, making it difficult to attribute the
payment to any one of them.
The lower section shows the situation for a credit card
that is not paid off automatically each month12. The
payment and consumption are now decoupled in both
directions. The benefits are not reduced much by the
thought of the payments, but neither are the payments
buffered by the thought of associated consumption.
When making the payment, it is difficult to associate it
with any one remembered benefit. In addition, because
the consumption events precede payment, the buffering
effect on the payments is lower than the effect of the
payments in attenuating consumption – the relationship
between the two is asymmetrical.
Thus, the ideal payment mechanism to encourage
spending is one that has high decoupling between
consumption and payment (so that the benefits of
consumption are not greatly mitigated by the pain of
paying), and low decoupling in the direction from payments
to consumption (so that the pain of paying is buffered to a
large extent by the associated consumption).
To discourage spending, or perhaps encourage only
responsible spending, requires a product that does the
opposite. Credit cards cause extensive decoupling in both
directions, but a greater understanding of the sources of
coupling should allow for product designs to emphasize
coupling in one direction over the other.
Coupling provides an answer to why credit card debt
is considered so aversive by many people13. In this
framework, payments are particularly painful if:
The associated consumption has already
happened at the time of payment.
• The payment cannot easily be associated with any
consumption benefits (decoupling).
Credit cards display both of these features,
magnifying the regret of overspending – and yet the
decoupling of payments from consumption encourages
spending and consumption at the time of purchase,
which is when the decision is made.
Retrospective evaluations
Dilip Soman offers an alternative theory13 for the way in
which payment mechanisms affect spending behaviour.
This theory also involves mental accounting, but at the
higher level of spending budgets rather than transaction
level accounting. To assist in the complicated process
of keeping track of daily, monthly or annual expenditure,
people group transactions together into higher-level
accounts.
Most people probably have a rough idea of what they
spend monthly on, say, eating out, and also of whether
their current spending is above or below this regular
budget. Some individuals use this reference as a simple
guideline to control spending, whereas others might
FIGURE 1: TEMPORAL RELATIONSHIP BETWEEN PAYMENTS AND PLEASURE FROM CONSUMPTION
pleasure from
consumption
associated
payment
pleasure from
consumption
associated
payment
pleasure from
consumption
associated
payment
Cash purchase
High coupling
Charge Card
Low attenuation
(reduction of
consumption-
enjoyment by
payment)
Credit card
Low attenuation
(reduction of
consumption and
low buffering)
Adapted from Prelec and Loewenstein 1998: See note 1
SEARCHING FOR THE RIGHT QUESTIONS
26 Volume 2 Issue 6 Argent www.thefsforum.co.uk
maintain explicit limits against certain categories of
spending, or even total spending.
In general, though, most people use their memories
of recent spending, and some sort of reference value
(analogous to a budget limit), to guide their current
expenditure, cutting back when they feel they are
overspending relative to this amount, and treating
themselves when they feel they have been spending
less than usual.
This comparison is not made using an exact running
total of expenditure, but one based on the memory of
recent spending. Research indicates that consumers’
recall of past expenses is imperfect, at best: individuals
may be unaware of the exact price paid; they may find
that maintaining an accurate total of all spending is
cognitively difficult; they may allow small expenses to slip
through as they are not thought worth tracking; and they
might fail to deduct mentally many expenses from the
consumer’s actual wealth at the time of purchase (instead
waiting until these are charged to their current account
balance at the end of the month).
Soman suggests that the assessment of these
personal budget limits is made by a retrospective
evaluation at the time of a purchase opportunity.
Payment mechanisms that result in low recall and low
aversive impact of past expenditures cause consumers
to underestimate their recent spending and, as a result,
to increase current spending. Aspects of the payment
mechanism have a direct influence on memory and
aversive impact in much the same way as coupling:
• The greater the salience and vividness of the past
payment, the stronger the memory trace it leaves.
Credit card payments are low in both salience and
vividness, thus causing an underestimate of past
spending, and an increased propensity to current
spending.
The disassociation of payments from benefits
results in a weaker aversive impact.
Note that these effects, unlike those of decoupling, work
only in one direction – in leading individuals to make lower
estimates of their past spending, payment mechanisms
like credit cards will tend to increase the likelihood of further
expenditure. Although the two theories are not mutually
exclusive and may operate simultaneously, Soman offers
the suggestion that coupling will have more of an impact
on larger one-off purchases, whilst retrospective evaluation
will be more important for smaller regular purchases.
Product design
Applying these theories to data on customers’ spending
behaviour, demographic characteristics and purchasing
environment will allow financial services companies to
predict and manage consumer behaviour as never before.
Few financial institutions currently have such a close
understanding of how consumers use their products or
how these products fit into their customers’ lifestyles. Yet,
it is precisely this understanding that is required to guide
the development of existing products and the innovation of
entirely new customer propositions and product
categories. Employing the ideas covered in this article
should enable providers to develop attractive products that
work with customers rather than against them, by assisting
them to maintain adequate control over their spending and
reduce, where appropriate, the pain of paying15. In addition
this knowledge will allow institutions to reach better pricing
decisions and make better predictions of customer
responses to marketing and product design changes.
The psychological levers that might be used to
influence customer behaviour include:
Presentation at moment of purchase:
o Designing credit card slips to increase salience by
requiring the amount to be written on the slip by the
customer as confirmation.
o Emphasizing the remaining credit limit at each
transaction, to enhance payment coupling with
current consumption and self-control.
o Presenting the customer with the outstanding
credit balance to enhance self-control.
Statement design:
o Emphasizing consumption benefits to increase
coupling (ie, pointing out what customers got for
the bill they are about to pay).
o Grouping of similar items to increase coupling and
buffer the pain of payment.
o Sending customized alerts through e-mail or text
when pre-set spending limits are reached.
o Allowing customers to place sub-limits on particular
expenditure categories, to complement their natural
system of mental accounting.
• Moment of payment:
o Offering shorter periods between the presentation
of credit card bills, to increase coupling and recall,
and to assist self-control.
o Allowing automatic transfers of credit card
payments to the reduce the pain of paying.
Rethinking the future
The financial services industry has become accustomed to
constant transformation through technological innovation.
However, with increasingly detailed knowledge of
consumer psychology emerging from such fields as the
cognitive, neuro- and behavioural sciences, companies will
need to focus increasingly on product designs that
complement the way in which consumers think about their
finances, rather than just increasingly sophisticated delivery
of standard products. To position themselves to benefit
from these advances, institutions need to undertake
research programmes into existing patterns of customer
behaviour and explore how financial services might be
adapted to those patterns.
Ultimately, this should lead to the development of an
SEARCHING FOR THE RIGHT QUESTIONS
www.thefsforum.co.uk Argent Volume 2 Issue 6 27
entirely new approach to financial services provision.
The current reality is that most retail finance
propositions are guided by historical accident rather
than the needs of customers. Ideally this means a
clean slate redesign of the entire customer proposition
with three objectives in mind. Firstly, simplifying and
clarifying financial decisions in a way that
complements natural psychological processes.
Secondly, reducing the amount of time required by
customers to manage their finances. Yet, thirdly,
achieving this whilst simultaneously improving
consumers’ control over their financial position. The
ideas in this article give examples of what is possible in
only one of many currently active areas of research.
Institutions need to recognize the work that is required
before the psychological understanding of financial
behaviour can be readily transformed into market
share and economic value. Recent research means
that many of the processes governing financial
behaviour are understood far better now than ever
before, and this knowledge has been translated into
powerful behavioural models. Understanding what
really drives consumers is the holy grail of retail finance,
and the fields of behavioural finance and cognitive
psychology are starting to put it within our grasp.
Greg Davies researches at Warwick University’s Decision
Technology Group.
1See, for example, Feinberg, RA (1986): "Credit cards as
spending facilitating stimuli: a conditioning interpretation".
Journal of Consumer Research 13 348–356; Prelec, D and
Loewenstein GF (1998): "The red and the black: mental
accounting of savings and debt." Marketing Science 17(1)
4–28; Soman, D (2001) "Effects of payment mechanism on
spending behavior: the role of rehearsal and immediacy of
payments." Journal of Consumer Research 27 460–474.
2In part, at least. Many would surely still argue that the
merchant fees are tolerated because trade would otherwise
be lost to rivals who accept cards. Ed
3And also in terms of their relationships with the government
and the regulators, and in general public relations. The
current investigation of the credit card market by the Treasury
Sub-Committee is a case in point. Ed
4Psychophysics is used to good effect in this context by car
manufacturers. The added extras do not seem expensive
when combined with the cost of the car, but would cause you
to think twice were you to have to pay for them separately.
[But this can be taken too far, as BMW found in the 1980s,
when customers eventually balked at the realization that even
the radio was an "optional extra" on a £25k car. Ed.]
5This example is adapted from Prelec and Loewenstein, op cit.
6Thaler, RH (1999). "Mental accounting matters." Journal of
Behavioral Decision Making 12 183–206.
7Taken from Kahneman, D and Tversky, A (Eds) Choices,
values, and frames Cambridge CUP 2000. Daniel Kahneman
and Amos Tversky were pioneers in the field of behavioural
finance, in recognition of which Kahneman was last year
awarded the Nobel Prize for Economics. [Tversky had died in
1996. Ed]
8That is, one group was asked the first question and the
second (matched) group was asked the second.
9op cit.
10 Prospective accounting also helps to explain the popularity of
"fixed-fee pricing", of which prominent examples are flat-rate
Internet and telephone charges, prix fixe dinners, "all-
inclusive" package holidays, and fixed rather than per-use
health club fees. This flat-rate bias cannot be accommodated
within standard consumption theories.
11 The coupling effect with credit cards is thus stronger in those
restaurants that require the purchaser to add the tip to the
credit card slip and write in the combined total.
12 If the full amount is settled each month, the situation is the
same as that for the charge card.
13 Apart, that is, from the high interest rates! A survey by Prelec
and Loewenstein op cit finds that people feel a refund from
credit card bills to be more satisfying than even an equivalent
cancellation of parking tickets!
14 Soman, D (2001) "Effects of payment mechanism on
spending behavior: the role of rehearsal and immediacy of
payments." Journal of Consumer Research 27 460–474.
15 It should be pointed out that the pain of paying does have an
important role to play, in that it helps people to exert control
over their spending. "The ideal payment arrangements, for
rich and poor alike, will be those that facilitate rational
spending while mitigating the pain of paying; that create the
illusion of free benefits without sacrificing accountability."
(Prelec and Loewenstein op cit).
Cause Example Effect on spending propensity
Psychophysics Bundling transactions Increase
Time discounting Prepayment Decrease
Prospective Prepayment Increase
accounting
Decoupling Decoupling payments Decrease
from consumption
Decoupling Decoupling consumption Decrease
from payments
Retrospective evaluation Low recall of spending Increase
Low aversive Increase
impact of spending
FIGURE 2: SUMMARY OF PSYCHOLOGICAL INFLUENCES ON SPENDING BEHAVIOUR
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.