Content uploaded by Jonas Hedman
Author content
All content in this area was uploaded by Jonas Hedman on Oct 03, 2017
Content may be subject to copyright.
Cashless Society: When will Merchants Stop Accepting
Cash in Sweden - A Research Model
Niklas Arvidsson1, Jonas Hedman2, and Björn Segendorf3
1KTH Royal Institute of Technology, Stockholm, Sweden
niklas.arvidsson@indek.kth.se
2Copenhagen Business School, Frederiksberg, Denmark jh.itm@cbs.dk
3Sveriges Riksbank, Stockholm, Sweden bjorn.segendorf@riksbank.se
Abstract. Over the past decades, we have witnessed changes into how
individual’s pay. In particular, there has been a drop in the use of cash as
payment instrument both in terms of value and frequency. Consequently, the
amount of outstanding cash is shrinking. For instance, in Sweden the level of
cash is around 1.5% of Gross Domestic Product. This might be a tipping point
for when cash is of practical use. In the paper, we present a research model that
explores when merchants will stop accepting cash.
Keywords: Cashless society, Merchants, Cash Adoption.
1 Introduction
Payments are essential in the exchange of money for goods and services between
sellers and buyers. The most used payment instruments in point of sales locations are
cash and payment cards. Over the past decades, we have witnessed changes into how
individual’s pay. In particular, there has been a drop in the use of cash as payment
instrument both in terms of value and frequency. Payment cards, such as charge,
credit, and debit, and more recently new payment instruments, such as mobile
payments and e-money, are replacing cash. These changes occur more or less in all
economies and across the globe, but are particular evident in the Nordic countries,
where you also can find a lively debate on the cashless society [3, 4, 8]. For instance,
in Sweden there is a cash rebellion “Kontantuppror”, where lobby groups in particular
representing the cash-in-transit service industry and older people, demand that banks
accept cash again (Note that most Swedish bank branch offices are cashless).
In parallel, payments are receiving increased attention from academic communities
and span several disciplines, including information systems [11, 20], consumer
research [10, 23, 25], marketing [17, 26], economics [6, 22], sociology [16],
management science [2, 3, 24], and banking and finance [12, 15]. This has resulted in
a variety of topics in the study of payments, including what money is [29], cost-
benefit analysis of cash and payment cards [7, 27], competition [9], social
implications [18], choice and spending behavior [25, 27], payment framework [5],
and adoption of mobile payments [1, 20, 31].
Despite the above, one aspect of payment research which has been largely ignored
is merchant acceptance of payments, i.e. why do merchants accept or don’t accept
specific payment instruments. There are some exceptions, including the adoption of
mobile payments by merchants [21] and the study of merchants point of sales data
[26]. One finding is that cash payments are much more expensive than card payments
[6] and we witness a “…movement toward greater use of electronic payment methods,
though gradual, is uniform and unmistakable, both across countries and over time”
[14, p. 936]. Schreft [28, p. 5] puts forward critic on existing research “…is backward
looking. It tells us what payment instruments were chosen in the past may not be a
good indicator of what will be chosen [accepted] in the future”. In the realm of an
emerging cashless society, we are in particular interested in when merchants stop
accepting cash.
We assume that merchants are economic rationale in their decision making, i.e.
merchants will stop accepting cash when it becomes more expensive to manage cash
acceptance than the marginal profit on cash sales. It is important to note that Swedish
merchants are not legally bound1 to accept cash as a mean of payment but can decide
themselves which payment services to accept. However, we acknowledge the
existence of other factors influencing this choice, including the risk of being robbed.
Our work has the potential of contributing to the understanding of merchant’s role
towards a cashless society by developing a research model that explains the when
merchants will stop accepting cash at point of sales.
2 Background
The context of this study is Sweden, since it is among the countries in the world with
the lowest value in banknotes and coins in circulation compared with gross domestic
product.
2.1 Retail payments in Sweden
One measure of cash use is the value of outstanding cash – bills and coins – compared
with the gross domestic product (GDP), which varies between countries, as shown in
Table 1. This measure provides an estimate of how dependent a payment system on
cash, since many cash transactions are person to person (P2P) transactions and
therefore not registered in any official statistics. The numbers for Sweden show a
long-term downward trend when comparing outstanding cash to GDP. In 1950 this
1 The Riksbank law states that cash is legal tender in Sweden and should therefore be accepted,
but the freedom to enter contracts underpinning contractual and commercial law implies that
a payer and a payee can enter an agreement that sets the Riksbank law aside. It should be
noted that there are few court case rulings in this area and none in the highest court. This is
not the case in Denmark or Norway, where central bank laws have superiority over
contractual and commercial law.
3
number was nearly 10 percent but the last ten years it has been below three percent,
and 1,8 percent in 2015, as shown in Table 2. The most recent statistics show that the
number is around 1,5 percent in August, 2016. This long-term decline is, however,
often the result of a process where GDP is increasing faster than the outstanding value
of cash. A second, and quite extraordinary observation, is that the nominal
outstanding value of cash has been declining since its peak in 2007. This is – to our
knowledge – unique for Sweden and a strong indication of the transition towards a
more or less cashless society in Sweden. According to data from the Riksbank, the
decline is significant and fast also in 2016 where the nominal value of outstanding
cash decreased over 12 percent in the period from January to July.2
Table 1. Outstanding cash in selected countries 2014, some selected countries
Countries / regions
Cash-in-circulation as share of GDP (M0/GDP; %)
Malaysia
102,4
Chile
30,2
Bulgaria
12,2
Czech Republic
10,1
Euro-zone
9,7
Pakistan
8,8
USA*
7,1
UK
3,5
Denmark
3,0
Norway
2,03
Sweden
2,0
Sources: European Central Bank (ECB), Norges Bank, Sveriges Riksbank and
www.knoema.com. *The figures for the US is from the year 2010.
Table 2. Value of banknotes and coins in circulation (annual averages; banks’ holdings
excluded)
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Value in billion
SEK
96,5
97,0
96,7
96,5
95,5
90,7
86,8
84,4
78,2
74,9
Value as share
of GDP
3,1
2,9
2,9
2,9
2,7
2,5
2,4
2,2
2,0
1,8
Source: The Swedish Financial Market 2015, The Riksbank (www.riksbank.se/en/) and
Statistics Sweden
The statistics on the usage of payment instruments in Sweden show that cash is
rapidly declining. There several explanations for this. The first explanation is the
long-term increase in the use of card payments in Sweden. Card payment schemas
were launched in a greater scale during the 1990s and merchants as well as consumers
2 It should be mentioned that Sweden is currently replacing all banknotes and most coins with
new ones. The direct short-term impact of this on the value of currency in circulation is
ambiguous but it does not affect the strong long-term negative trend.
3 This number is based on Norway’s main-land GDP, i.e. excluding the off-shore oil sector.
have adopted card payments as the most important payment instrument in retail point
of sales (POS) locations. The number of card transactions at POS and the value of
these transaction have been growing steadily during the last ten years. In addition, the
number of ATM withdrawals and value of such withdrawals have been declining over
that last ten years (Table 3). Finally, survey undertaken by Sveriges Riksbank
indicates that the use of cash at the point of sale in terms of volume of payments has
fallen from close to 40 per cent to close to 20 per cent between 2010 and 2014. A
recent, however not yet published, study by the Riksbank indicates that the number is
down to around 15 per cent in 2016.
Table 3. Card transactions in payment terminals (POS) and ATM withdrawals
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
ATMs
No. of ATMs
2816
3085
3236
3319
3351
3566
3416
3237
3231
3285
Transactions
(millions)
313
320
295
269
241
221
210
183
167
154
Value (SEK
billion)
270
240
239
232
225
206
190
174
171
153
Payment card terminals
Terminals
(thousands)
184,6
187,3
194,8
217,8
203,1
209,6
198,4
195,8
197,0
183,8
Transactions
(millions)
1000
1154
1358
1490
1646
1799
2048
2329
2423
2501
Value (SEK
billion)
423
463
488
496
557
598
654
722
754
747
Source: The Swedish Financial Market 2015, The Riksbank (www.riksbank.se/en/)
A second explanation is the recent is the introduction of a mobile payment service
that enables real-time clearing and settlement and therefore provide similar
functionality as cash, i.e. the value in the transaction is transferred in real-time from
the payer to the payee. Much in analogy with the passing of a bill from one hand to
the other. This service – which is called Swish – was launched by banks in 2012 and
has become very popular for person-to-person payments. The growth was high in both
2014 and 2015. By the end of August 2016, the service was used by 4,7 million
Swedes and transactions worth 8,2 billion SEK were made during August 20164. This
service has, however, not yet been adopted at a large scale by merchants.
2.2 Merchants’ choice of payment services
Merchants in Sweden can decide to not accept cash but are at the same time not
allowed to issue a surcharge to consumers related to which payment service that are
used, which in essence means that merchants’ choice of ideal payment service
primarily is based on the direct costs and revenues related to each specific payment
service. But research also shows that other factors play an important role as well.
4 See www.getswish.se
5
A study by Loke [19] of which factors that determine a merchant’s decision to
participate in a card scheme identified the following factors as important explanatory
variables:
• Merchant’s background (including age, number of personal credit cards held5, and
use of computers6)
• Merchant’s business characteristics (including business sector, total value of
transactions per month, average value of transaction, profit margin, location of
business)
• Effects of other players’ decision via merchant’s perception (including merchant’s
perception of customers’ use of cards and competitors’ participation in the card
scheme)
The study (ibid) arrived at two main conclusions. The quantitative analysis related to
the factors above showed that the statistically significant explanatory variables were:
age of the merchant which had a negative relation to the probability of accepting card
payments; number of cards held by merchant (positive relation); business sector
(where surprisingly enough non-technical stores were more positive to accepting
cards); total value of transactions (positive relation); merchants’ perception of
customers’ use of cards (positive relation); and competitors’ acceptance of card
payments (positive relation). When discussing these results, the study concluded that
the demand from customers was the most important factor while the merchants’ wish
to boost sales related to acceptance of card payments was the second most important
factor.
Other studies [12, 13, 30] highlight the importance of different characteristics of
merchants, characteristics of payers or consumers and a number of other factors.
Regarding the characteristics and decision factors of merchants this includes:
industry, location, margins or profitability, type of products sold, type of customers,
price of payment services, amount of revenue connected to payment service, inter-
operability of the service, as well as other factors such as, for instance, risks,
employees’ opinions and work environment issues. Regarding the characteristics and
decision factors of payers this includes: socio-demographic characteristics (age,
gender, education, income), transaction frequency and value, speed and ease of use of
services, need for integrity, technology interest, trust in services and in the payment
system in general, costs of payment and banking services, as well as other factors.
3 Research Model
Based on the review of retail payments in Sweden our research model is therefore
rather straightforward. We ask ourselves if and under which circumstance merchants
will stop accepting cash and instead accept only card payments, and base the analysis
on business factors such as revenues and costs related to cash and revenues and costs
related to card payments. Our model is based on the following features:
5 A proxy for experience with cards
6 A proxy for familiarity with new technologies
Let θ be a set of payment methods that can be used at the point of sale. For
simplicity, assume that there is only two such methods a and b. One may think of a as
cash and b as cards. Each merchant chose whether to accept cash (a=1) and cards
(b=1). If he does not accept cash (cards) then a=0 (b=0).
Let the continuous function ri:θ→R+
denote the revenue merchant i makes depending on his choice of payment methods. In
particular, let ri((0,0)=0. Accepting no payment methods can be interpreted as a
decision to exit the market. Here, for simplicity, we will initially assume the specific
functional form ri(a,b)=ra
i*a+rb
i*b, i.e. the revenue generated from card payments
does not change if merchant also accept cash and vice versa.
Let πi denote the profit of the merchant, cj the fixed cost and vj the variable cost
for each merchant of accepting payment methods j=a, b. With variable cost, we mean
the cost as a function of the size of the revenue generated by the payment method in
question. We also assume that all merchants have access to the same technology for
receiving payments. In Sweden, this can be motivated by nearly all points of sale have
to use cash registers that are approved by the tax authority, i.e. the tax authority limits
the merchants’ choice to a narrow and clearly defined set of technologies.
Let xϵ(0,1) denote the revenue margin, i.e. the share of the revenue that exceeds the
cost of the good sold. The merchants profit maximization problem is thus to
maximize πi over θ.
Max πi(a,b)=a*ri
a(x- va) + b*rb
i(x- vb) - ca *a – cb *b
In the general case this means that the merchant will accept cash if ri
a(x- va) - ca ≥ 0
for b=0 or 1 or both. In the following we will restrict ourselves to the case where
merchants already accept cards (b=1). The first reason for this is that it allows us to
focus on the question of when does a merchant abandon cash and still remains on the
market. The second reason is that in Sweden cards are nearly universally accepted.
To answer our research question, we have to estimate x, va and ca which will allow us
to find the critical revenue threshold
ra* = ca/(x-va)
which the merchant accepts cash and below which he does not.
3.1 Selected survey items and points for discussion:
Data collection will be carried out by the three big trade associations, covering
retailing and restaurants. The unit of data collection is a store (location) in 2016.
• What is the total turnover incl. VAT in 2016 of your store_____?
Total value in cash_____? And share of total transactions (%)____?
Total value in card_____? And share of total transactions (%)____?
7
• What is the shares of transactions size related to value (%) 1-19 SEK____? 20-99
SEK____? 100-499 SEK___? And 500-higher SEK____?
• What is the distribution of your costumers in age in percent? Children (0-11 years
old)____Youth (12-17 years old)____ Adults (18-65 years old)____Retired (over
65 years old)____
• How large is your profit margin in percent? (We will use industry averages.)
• Do you accept cash? ____ (Yes / No)
Cost for Accepting Cash.
• What is the average hourly labour cost per hour per employee at your business
(We will use industry averages)? ____
• Estimate your total costs in 2016 that are related to cash handling?____
• What was the cost of cash due to, for instance incorrect change, theft, forged
cash, robberies, etc. that was not covered by insurance?____
• Try to estimate how much time per day the employees devote to count cash.
Provide an estimate of a daily average____hours and____minutes per day?
• Estimate the costs of cash storage (depreciation of safety vaults, fees to baks and
cash depots, etc.) ____SEK in storage fees 2016?
Substitution effect.
• How much did your company pay the bank / card company on average on top
of the fixed costs? Per transaction (on average) or fee + ______% of the amount
Background questions.
• Do you have employees that sometimes work alone in a store? ____(Yes/No)
• Do you consider stop accepting cash? ____(Yes/No)
• Would your sales decrease if you stopped accepting cash?____(Yes/No)
• Would your profits decrease if you stopped accepting cash?___(Yes/No)
• Which year do you think you will stop accepting cash? ____
• Has your business suffered from robberies during the last five (or ten)
years?____(Yes/No)
• Do you know if any other store that is close to your stores suffered from
robberies during the last five (or ten) years? ____________(Yes/No)
• Is it common for companies in your industry to pay under the
table?____(Yes/No)
References
1. Arvidsson, N.: Consumer attitudes on mobile payment services–results from a proof of
concept test. International Journal of Bank Marketing, 32, 2, 150-170 (2014)
2. Arvidsson, N.: Det kontantlösa samhället: Rapport från ett forskningsprojekt. (2013)
3. Arvidsson, N.: Proceedings third international cashless society roundtable (ICSR). 2013.
Stockholm
4. Carton, F. and Hedman, J.: Proceedings: Second internationael cashless society roundtable
(icsr). 2013, Financial Services Innovation Centre
5. Carton, F., Hedman, J., Damsgaard, J., Tan, K.-T., and McCarthy, J.: Framework for mobile
payments integration. Electronic Journal of Information Systems Evaluation, 15, 1, 14-25
(2012)
6. Garcia-Swartz, D., Hahn, R., and Layne-Farrar, A.: The move toward a cashless society: A
closer look at payment instrument economics. Review of Network Economics, 5, 2, 175-197
(2006)
7. Garcia-Swartz, D.D., Hahn, R.W., and Layne-Farrar, A.: The economics of a cashless
society: An analysis of the costs and benefits of payment instruments. AEI-Brookings Joint
Center for Regulatory Studies (2004)
8. Hedman, J.: Proceedings first international cashless society roundtable (icsr). 2012.
Copenhagen, Denmark, April 18 & 19: Copenhagen Finance IT Region
9. Hedman, J. and Henningsson, S.: The new normal: Market cooperation in the mobile
payments ecosystem. Electronic Commerce Research and Applications, 14, 5, 305-318
(2015)
10. Hirschman, E.C.: Differences in consumer purchase behavior by credit card payment
system. Journal of Consumer Research, 6, 1, 58-66 (1979)
11. Holmström, J. and Stalder, F.: Drifting technologies and multi-purpose networks: The case
of the swedish cashcard. Information and Organization, 11, 3, 187-206 (2001)
12. Humphrey, D.B.: Retail payments: New contributions, empirical results, and unanswered
questions. Journal of Banking & Finance, 34, 8, 1729-1737 (2010)
13. Humphrey, D.B., Kim, M., and Vale, B.: Realizing the gains from electronic payments:
Costs, pricing, and payment choice. Journal of Money Credit and Banking, 33, 2, 216-234
(2001)
14. Humphrey, D.B., Pulley, L.B., and Vesala, J.M.: Cash, paper, and electronic payments: A
cross-country analysis. Journal of Money, Credit and Banking, 28, 4, 914-939 (1996)
15. Kahn, C.M. and Roberds, W.: Why pay? An introduction to payments economics. Journal
of Financial Intermediation, 18, 1, 1-23 (2009)
16. Knights, D., Noble, F., Vurdubakis, T., and Willmott, H.: Electronic cash and the virtual
marketplace: Reflections on a revolution postponed. Organization, 14, 6, 747 (2007)
17. Lawson, R. and Todd, S.: Consumer preferences for payment methods: A segmentation
analysis. International Journal of Bank Marketing, 21, 2, 72-79 (2003)
18. Linné, T.: Digitala pengar. Nya villkor i det sociala livet. in Lund Dissertations in
Sociology. 2008, Lund University: Lund
19. Loke, Y.J.: Determinants of merchant participation in credit card payment schemes. Review
of Network Economics, 6, 4, (2007)
20. Mallat, N.: Exploring consumer adoption of mobile payments: A qualitative study. The
Journal of Strategic Information Systems, 16, 4, 413-432 (2007)
21. Mallat, N. and Tuunainen, V.K.: Exploring merchant adoption of mobile payment systems:
An empirical study. E-Service Journal, 6, 2, 24-57 (2008)
22. Penz, E., Meier-Pesti, K., and Kirchler, E.: “It's practical, but no more controllable”: Social
representations of the electronic purse in austria. Journal of Economic Psychology, 25, 6,
771-787 (2004)
23. Penz, E. and Sinkovics, R.R.: Triangulating consumers' perceptions of payment systems by
using social representations theory: A multi-method approach. Journal of Consumer
Behaviour, 12, 4, 293-306 (2013)
24. Priem, R.L.: A consumer perspective on value creation. Academy of Management Review,
32, 1, 219-235 (2007)
9
25. Raghubir, P.: An information processing review of the subjective value of money and
prices. Journal of Business Research, 59, 10, 1053-1062 (2006)
26. Raghubir, P. and Corfman, K.: When do price promotions affect pretrial brand evaluations?
Journal of Marketing Research, 211-222 (1999)
27. Runnemark, E., Hedman, J., and Xiao, X.: Do consumers pay more using debit cards than
cash? Electronic Commerce Research and Applications, 14, 5, 285-291 (2015)
28. Schreft, S.L.: How and why do consumers choose their payment methods? : DIANE
Publishing (2006)
29. Simmel, G.: The philosophy of money. Psychology Press (2004)
30. Stavins, J.: Effect of consumer characteristics on the use of payment instruments. New
England Economic Review, 3, 19-+ (2001)
31. Xin, H., Techatassanasoontorn, A.A., and Tan, F.B.: Antecedents of consumer trust in
mobile payment adoption. Journal of Computer Information System, 55, 4, 1-10 (2015)
