To what extent is the use/retention of physical means of payment necessary for the Dutch economy?

To what extent is the use/retention of

physical means of payment necessary for the

Dutch economy?

Bachelor thesis

Name: Jorg Melchers Student number: 11011483 Specialization: Economics Academic year: 2017-2018

Assigned supervisor: Daniel Dimitrov Number of words: 7413

2

Statement of originality

This document is written by Jorg Melchers who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

3

Content:

Abstract ... 4

Introduction ... 5

Literature review ... 6

Methodology and Empirical analysis ... 9

Discussion and Conclusion ... 19

4

Abstract

In line with global trends on digitization, the usage of digital payment instruments in the Netherlands is increasing. This is because digital payment instruments are easy to use, less costly, have a higher transaction speed and a lower risk of economic crimes compared too cash. In a cashless situation, banks can set negative interest rates, which can reduce the adverse effects of a crisis. A positive side effect for the economy is that people spend more when they pay with debit or credit cards instead of cash. Next, the Baumol-Tobin model of money demand indicates that the demand for cash will increase,since the cost of cash will increase due to the reduction in the number of ATMs. This is illogical, since the data shows that digital payment instruments will replace cash, especially when the costs of cash increase. The available data of payments in the Netherlands indicates that debit and credit card payments have an increasing share in the total number of transactions and their value. The observations correspond with the earlier findings stated in the literature review. It seems feasible that cash will disappear in the future. There are no signs that the Dutch economy without cash will not be able to function. The main finding of this thesis is that the use/retention of physical means of payment are not necessary for the Dutch economy.

5

Introduction

In recent years it is becoming clear that people use less cash in payment transactions. Based on the figures for the annual payment transactions, there is a shift toward a cashless society. This leads to several costs and benefits (Garcia-Swartz et al., 2006). Thereby, different forms of money, specifically digital versus cash, have different impacts on the working process of an organization and its

customers (O’Neill et al., 2017). There are also companies and institutes which don’t accept specific payment instruments. An example of this is that in 8 municipalities in the Netherlands, cash is no longer accepted as a payment mechanism. They have a so-called 'pin-only' policy (Haegens & Bolwijn, 2018). It is interesting to know why this is the case. Besides this, several studies have shown that people will spend less money when they use cash compared to digital money. This can have a major influence on the Dutch economy.

Nowadays, the number of cash payments is decreasing, where the number of digital payments is strongly increasing. Since 2015, Dutch citizens used their debit card more than cash for their payment transactions (Dutch Banking Association, 2017). This is made even clearer by a number of observed trends. Concluding from data from 2010 to 2017, the number of debit card payments increases, while the number of ATMs, the number of cash withdrawals at ATMs and the total value of these withdrawals decreases (Dutch Banking Association, 2017).

As the use of digital payment instruments show an increasing trend, this digitization of payment instruments should have advantages. It is necessary to investigate all the pros and cons of this digitization. This paper is doing so by answering the following question: To what extent is the use/retention of physical means of payment necessary for the Dutch economy?

I expect that the use of electronic means on payment transactions will grow further and is good for the economy as a whole. However, the use of electronic means on payment transactions cannot fully replace cash. I expect that people want to keep a certain form of privacy, which is lost when cash in no longer accepted as a payment mechanism. In addition, I think that the

macroeconomic model used in this paper will confirm that cash is necessary for an efficient, safe and reliable payment system. Because of this, my hypothesis is as follows: The use/retention of physical means of payment is necessary for the Dutch economy to be safe and reliable.

The research question will be answered through a literature review. This literature review shows the advantages and disadvantages of cash. A review of all these pros and cons of cash is been listed in a table at the end of the literature review. Thereby, the Baumol-Tobin model, a

macroeconomic model, will be consulted to see if it matches with the mentioned literature. Finally, several facts, based on available data from payment transactions in the Netherlands, will be listed. This data is based on documents published by the Dutch Banking Association. This concerns data about the use of the payment instruments from 2005 till now.

6

Literature review

It’s important to have an efficient, safe and reliable payment system. The digitization of the payment instruments can contribute to this. In order to create and maintain such an payment system, it was decided to set up the MOB (Maatschappelijk Overleg Betalingsverkeer). Their participants are 13 social organizations, which are representatives of various providers and users in the payment system. In addition, the Ministry of Economic Affairs and the Ministry of Finance are observers of the MOB (De Nederlandsche Bank, MOB). The MOB advices, since they cannot enforce this, all the Dutch municipalities to accept cash. They argue that it is undesirable to refuse cash payments in situations where, for a similar product/service, there are no alternative providers which will accept cash (De Nederlandsche Bank, MOB). The MOB makes it clear that it considers the continued existence of cash as a legal tender, since payment instruments must be usable for everyone. This is also noted by the Dutch Banking Association in the “Visie op Betalen 2018-2021”. Digital payment instruments are considered to be too complicated for a number of citizens who, for various reasons, do not want or are unable to use technical means. This is visible in Sweden, where most bank branch offices are cashless. This has led to an cash rebellion. This involves several lobby groups, that mainly represent the money transport sector and the elderly, which demand that banks accept cash again (Arvidsson et al., 2016).

Despite the advice of the MOB, there are several companies and institutes which do not accept cash. They have a so-called 'pin-only' policy. The main reason for this is the cost reduction that occurs with digital payment methods. Using a cost-benefit framework, Garcia-Swartz et al. (2006) found that the shift to a cashless society appears to be a beneficial one. They constructed two case studies, showing that transaction size has influence on the net social marginal cost of all

payment instruments. First of all is shown that cash usage decreases as transaction size increases. This corresponds with the findings at supermarkets, which show that digital payments at large transaction sizes are considerably less costly compared with cash. Retailers lose 25 to 27 euro cents per cash payment, which is more than the 15 to 19 euro cents that it loses per digital payment (Haegens & Bolwijn, 2018). In addition, digital payment instruments make unmanned payment terminals possible, which reduces labor costs for the retailers. Garcia-Swartz et al. (2006) conclude that there are also parties, market traders for example, who benefit the most from cash payments. Their advantages, however, do not outweigh the negative consequences for the consumers.

Money can be used as a payment method, a unit of account or a savings unit. Dewi (2010) argues that cash is easily recognized and accepted, easy to transfer, difficult to duplicate, anonymous and easily movable. Digital payment instruments must meet these conditions in order to fully replace cash. One of these payment instruments, digital money, can be seen as any means of payment that’s stored in a purely digital form. The bitcoin is an example of such digital money. Gann et al. (2015) argue that digital money makes transactions faster, cheaper and more widespread. They explain this by the fact that digital money does not need the services of a bank or financial service provider to fulfill payment transactions. Digital money is therefore the most advanced form of payment option, since the debit card and the credit card still need this services.

However, as opposed to digital money, other digital payment instruments do not fully meet the aforementioned requirements. As cash decreases and the payment instruments become digital, there appear major social issues around digital security and privacy (Gann et al., 2015). The Dutch Banking Association emphasize, in there “Visie op Betalen 2018-2021”, that a full digital payment system can have major negative consequences in the event of a cyberthreat, a power failure or banking crisis. In these cases, it will become impossible to fulfill payments. Thereby, the big advantage to cash is the anonymity of transactions. Most digital payment instruments do not have this anonymity, since every transaction is registered. As mentioned before, digital money is an exception on this. In contrast to debit card and credit card transactions, digital money allow users to remain anonymous (Berentsen, 1998). This could facilitate illegal activities. Armey, Lipow and Webb (2014) refuse this by providing evidence that shows that cashless transactions leads to less crime. In their paper Armey et al. (2014) show, according to data of 71 countries, that there is a negative,

7 significant, and robust relationship between greater access to electronic financial transactions and lower incidence of economic crime.

The research of Armey et al. (2014) is in line with that of König (2017), in which is stated that criminal activities, such as money laundering, tax fraud and financing terrorism, are less frequent when there is less cash. Furthermore, the German government used this as an argument to announce its plan to impose a ceiling on cash payments. The same applies to the European Central Bank, who announced the end of production and issuance of the 500-Euro-Bill (König, 2017). However, these last two statements are criticized by König (2017). He believes that other interests have led to the decisions of the German government and the European Central Bank. König (2017) argues that if cash is fully replaced by digital payment instruments, no one is able to convert his money into cash. In that case, banks are allowed to successfully introduce negative interest rates on private savings. When the bank applies a negative interest rate, it encourages people to spend their money. Where König (2017) considered this as negative, since it limits the freedom rights of citizens, Haegens and Bolwijn (2018) think it could be positive. They argue that it can be used as a powerful weapon to prevent future crises. Thereafter, König (2017) concludes, based on his findings, that the temptation for the government to impose new taxes will increase when it gains easier access to the property of its citizens. Despite the fact that this will increase the government revenues, cash ceilings will not help but rather harm society (König, 2017).

As mentioned earlier, digital money is the most advanced form of a digital payment

instrument. Despite the fact that Dewi (2010) indicates that anonymity is a condition that has to be met for digital payment instruments, digital money is often criticized because of its contribution to the anonymity of the payment system. König (2017) stated that the true motive behind limitations for cash is the desire to obtain more control and access to personal data. Digital money is an exception to this, with the rise of the bitcoin as an example. Sahoo (2017) investigated the bitcoin and concluded that it has the potential to be stable in the future. This means that it will be easily accepted and increase the faith in the cryptocurrency technology and its usability, so it will become an alternative of cash.

Another remarkable effect of digital payment instruments is that it has influence on people’s spending behavior. Prelec and Simester (2001) conducted 2 studies in which they show that the willingness-to-pay increases when people use a credit card instead of cash. The credit card is one of the most used payment methods to study the change of spending behavior. Based on 2 field experiments, they find a potentially large credit card premium. Despite this, the effect is not always present. This can be explained by the fact that a large credit card premium is less likely when people know the market value of the good they want to buy. Their results were used for follow-up studies. Using a bid game, Runnemark et al. (2015) found that average bids are higher when participants get their debit card to bid instead of cash. In both treatments the participants gets payed in cash, but the payment options with which they could make their bid differed. Thereafter, using another treatment group, they also found that average bids depends on the payment method. If the participants were paid by means of account payments instead of cash, there bids were lower.

Following the aforementioned studies of the influence of digital payment methods on people’s spending behavior, Raghubir and Srivastava (2008) conduct research into the psychological effect of digital payment instruments. They argue that the change of the spending behavior is caused by a difference in transparency. Digital money is less transparent than cash and thereby the outflow of it feels less vivid. When the outflow feels less vivid, the pain of paying reduces (Raghubir and Srivastava, 2008). In addition, Raghubir and Srivastava (2008) explain that credit card and debit card payments differ from each other. This is in line with the study of Runnemark et al. (2015), in which they look at debit card payments instead of credit card payments. In the Netherlands, the number of debit card transactions has increased by more than 60 percentage points since 2010, to more than 3.6 billion debit card transactions in 2016 (Dutch Banking Association, 2017). The number of debit transactions increases, while the number of payments with the credit card remains at the same (low) level. As a result, their results are representative for the Dutch economy. According to Raghubir and Srivastava (2008), there are two reasons for the fact that credit cards payments alleviate the pain of

8 paying. Firstly, the payment is not simultaneous with the consumption, which takes place later. Secondly, credit cards make it possible to combine purchases into one payment. This ensures that this payment cannot be attributed to a specific consumption.

In Sweden the level of cash is around 1.5% of Gross Domestic Product (Arvidsson et al., 2016). Sweden is one of the first countries in which this percentage is that low. Although this percentage decreases in most countries, it is not as strikingly low as it is in Sweden. As a result, Arvidsson et al. (2016) developed a model by which they want to investigate when merchants in Sweden will stop accepting cash. It should be mentioned that merchants in Sweden, according to the Swedish law, can determine themselves which payment options they accept. Loke (2007) found out that this decision to accept credit cards depends on the personal background, the type of business and the total value of sales of the merchant. It is plausible to assume that this also applies for the debit card, since Loke (2007) focused on card acceptance. The findings of Loke (2007) imply that credit card acceptance in payments is influenced by the age of the merchant. The older the merchant, the smaller the probability of credit card acceptance in payments will be. Next, Loke (2007) found that credit card acceptance in payments is higher in the retail trade sector compared with the food and beverages sector. Finally, the results of Loke (2007) implies that a merchant takes into account the needs of the consumer. The merchant will accept credit cards in payments when he believes that its favored by his customers. Above stated arguments certainly differs from the

arguments used by Arvidsson et al. (2016) and Garcia-Swartz et al. (2006), who consider the cost difference between cash and digital payment instruments as the main reason for merchants to accept cash less often.

Another interesting phenomenon, which is described by Tee and Ong (2016), is that any policy that promotes cashless payment will not affect the economy immediately. Their study is based on findings of 5 EU countries, making the results representative for the Netherlands. Tee and Ong (2016) describe the digitization of payments instruments and find that the adoption of one type of cashless payment has effect on other types of cashless payment. This applies to both the short and the long term. However, when one looks at the effect of cashless payments on economic growth, this can only be observed in the long run. The research by Tee and Ong (2016) makes it clear that if the Netherlands really strives for a cashless society this will happen gradually and not abruptly.

Finally, the main pros and cons of cash, discussed in this literature review, are stated in table 1. This has been viewed from the general perspective, since an advantage for someone can be a disadvantage for someone else.

Pros Cons

Available for everyone High costs

Anonymous Relative low transaction speed

Reliable and generally accepted Banks can’t set negative interest rates in times

of crisis

Includes pain of paying People will spend less than with digital money

Digital crime is lower/ less harmful Contributes to economic crime

9

Methodology and Empirical analysis

This section looks at the influence of the digitization of payments instruments on the Dutch

economy. Hereby, the Baumol-Tobin model is used, to investigate the influence of this digitization on the money demand. This model will show step by step what effect the digitization of payment instruments has on the money demand in the Netherlands.

According to Baumol (1952), Y euros will be payed to an individual at the beginning of a period, considered as his income. The individual can choose to withdraw this money and spend it in a steady stream over the period or he can deposit it at the bank and earn interest (i). If the individual withdraws the full amount Y at the beginning of the period and will use it to spend it in a steady stream over the period, his average money holding will be equal to Y/2. It is assumed that the length of the period is equal to 1. But when the individual will only withdraw half of Y at the beginning of the period, spends it in a steady stream till half of the period and then withdraw the second part of Y, his average money holding will be equal to Y/4. In general, the average money holding can be

described as Y/2N, where N is the number of withdrawals during the period (Skaar and Torp, 2017). In addition, the Baumol-Tobin model contains the cost per transaction, which will increase with the number of withdrawals at a ATM and is displayed by A. It must be mentioned that the original Baumol-Tobin model was designed when there was only cash. If you combine all the information given, the total cost of money management (Cc) is equal to:

CC = NA +

Y𝑖

2𝑁 (1)

Taking the derivative from (1) with respect to N and set it to zero will lead to: CC’N = 0 = A –

Y𝑖 2𝑁2

Solving this for N, the optimal number of withdrawals (Nc*) can be displayed as (2): A = _2𝑁Y𝑖₂ 2N2_{A = Yi} N2 ₌ Y𝑖 2𝐴 NC* = ( Y𝑖 2𝐴 ) 1/2 ₍₂₎

These optimal number of withdrawals can be used to calculate the minimum total costs for money (CC*), which is shown as (3): CC* = ( Y𝑖 2𝐴 ) 1/2_{A +} Y𝑖 2( Y𝑖 2𝐴 )1/2 CC* = (2AYi)1/2 (3)

Furthermore, using the fact that average money holdings are equal to Y/2N, the money demand (D) looks like (4):

D(A,Y,i) = Y

2(𝑌𝑖

10

D(A,Y,i) = ( AY_2𝑖 )1/2 ₍₄₎

First of all, the Baumol-Tobin model indicates that the costs per transaction and the income have a positive effect on the money demand. On the other hand, the interest has a negative effect on the money demand. Digital payment instruments are an alternative for cash. This is not included in the model, but it does have influence.

Hereafter will be shown that the number of ATMs in the Netherlands is decreasing.

According to the Baumol-Tobin model the number of cash withdrawals should decrease, since A will rise as a result. This should be accompanied by a rising money demand. However, this is not the case with cash. The value of these transactions namely also decrease. This can be explained by the influence of the other two factors (income and interest rate) on the money demand function and/or the fact that the Baumol-Tobin model only considers cash and does not take into account

alternatives. If we only take into account the transactions at the counter, see further in this section on page 16 and 17, a decline in the value of the cash transactions and the amount of these

transactions can already be seen. Besides this, the total amount of transactions is rising. The Baumol-Tobin model can still be a useful model to analyze the money demand. It just should be extended with the effect of alternative payment instruments.

In this extended version the total cost of money management (Cc) can only be seen as the

total cost of cash management. The money demand depends not only on these costs, so (4) is no longer correct. Cash demand is influenced by the digital payment instruments. The costs of these digital payment instruments are identified by means of CD. Despite the fact that CD is not specified in

the same way as (1), the literature review shows that it is expected that it will be lower than CC.

Thereby, it is in line with expectations that the costs of cash will rise because cash payments are not always possible anymore, with the 8 ‘pin-only’ municipalities as an example. If CC rises, NC* will

decrease. According to (4), the demand for cash should increase. However, this is illogical, since alternative payment options would replace cash if the costs of cash will increase. This again indicates that the original Baumol-Tobin model does not take into account alternative payment options other than cash.

Nowadays, people are not dependent on cash to make payments. The money that one possesses to meet expenses no longer consists solely of (2), (3) and (4). Based on the information from the literature review, the combination of the increasing number of alternatives used and the increasing costs of cash results in the decreasing gradient of cash demanded. As long as CD is lower

than CC, the number of debit and credit card transactions will increase. This will lead to a fall in cash

demand, as people mainly will use the cheapest payment option. This means that (1), (2), (3) and (4) needs to change. All 4 comparisons are influenced by the usage of debit and credit cards and this influence has to be included to restore the usability of the model. However, these renewed comparisons are not reflected in this paper. The difficulty of renewing these comparisons ensures that doing it can be a paper in itself. This is proved by Kamada (2017), who in his paper expands the Baumol-Tobin model for digital money.

Now some background information will be given about payment transactions in the Netherlands. This information comes from the factsheets “Betalen aan de kassa 2017” and “payments” from the Dutch Banking Association.

Between 2010 and 2017, the total number of debit card payments has shown a rising trend, which is shown in figure 1. In contrast to this, the total number of cash payments between 2010 and 2017 decreased, which is also shown in figure 1. The period between 2012 and 2013 is an exception to this. Hereby, cash is usually used for payments of small amounts. In other words, the higher the amount, the more often this is paid by a debit card.

11 Figure 1: Total number of payments (Dutch Banking Association, 2017) Figure 2:Number of payments in relation to all payments

(Dutch Banking Association, 2017)

Figure 2 shows that the number of cash payments as a percentage of the total number of payments decreases. This is in line with the increasing number of debit card payments as a percentage of the total number of payments. As can be seen from these figures, since 2015 debit cards are used more often than cash. Incidentally, the number of credit card payments as a

percentage of the total number of payments remains at the same (low) level. Since 2010, the number of cash payments decreases on average with 3,61 percentage points per year. If we increase the length of these trends, it will be expected that cash will disappear in 2029. However, (yet unknown) future payment options can have impact on this trend. These could ensure that cash will disappear even earlier from the Dutch payment system.

Based on the figures 1 and 2, it is possible to look in more detail at the total number of payments. The Dutch Banking Association provides data, by means of an excel sheet, about the retail payment traffic. This includes detailed figures of credit and debit card payments at the counter, which are shown in figure 3 and 4. Despite the fact that the total number of credit card payments remains the same, judging from payment transactions in the Netherlands (see figure 1), the number of cross-border credit card payments at the counter shows an increasing trend. Cross-border transactions concern payments in which 1 of the parties involved is not a Dutch account holder. It also seems that the increase per year is taking place in even larger steps. However, there is a decrease from 2008 and 2009, which is probably caused by the financial crisis at the time. The number of domestic credit card payments at the counter is relative stable. From 2015 till 2017 it increases. This is mainly because of the remotely credit card payments that were introduced in 2015. This means that neither the cardholder nor the credit card is physically present at the time of the transaction. In particular, the number of internet sales payed with a credit card increased since then. As a result, a grey line has been added to figure 3, without the remotely payments. This line shows that the increased value is due to the remotely payments. The number of payments without

remotely payments even decreased. This seems to be due to the fact that remotely purchases partly replaced the "normal" credit card purchases, where the credit card or the cardholder is physically present. 0 1 2 3 4 5 2010 2011 2012 2013 2014 2015 2016 2017

Total number of payments (in

billions)

Cash Debit card Credit card

0% 20% 40% 60% 80% 2010 2011 2012 2013 2014 2015 2016 2017

Ratio per type of payment in

relation to all payments

12 Figure 3: Number of credit card payments at the counter (in thousands) (Dutch Banking Association, Retailbetalingsverkeer)

Where the cross-border number of credit card payments at the counter increases, it’s the domestic number of debit card payments that show an increasing trend (see figure 4). Thereby, in terms of figures, the growth of domestic debit card payments is many times larger than the growth of the cross-border credit card payments. This applies to all years, from 2005 to 2017. Nevertheless, the number of cross-border debit card payments also seems to be increasing, but this stands in no relation with the domestic increase. It can be seen from these figures that debit card payments often take place domestically, while credit card payments are more cross-border. Besides this, the

increasing trend of debit card payments at the counter looks more gradual. This data underlined that debit card payments are a much larger part of all Dutch payments compared with credit card

payments, which corresponds to the literature study, where this is mentioned.

Figure 4: Number of debit card payments at the counter (in thousands) (Dutch Banking Association, Retailbetalingsverkeer)

Until now, the figures indicate that the total number of credit and debit card payments at the counter increases and the cash payments decreases. Figures provided by the Dutch Banking

Association, in their factsheet “Betalen aan de kassa 2017”, indicate that the most debit card payments are done in the supermarkets and in the hospitality sector. At vending machines, street

0 20.000 40.000 60.000 80.000 100.000 120.000 140.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Number of credit card payments at the counter (in

thousands)

Domestic Cross-border Domestic minus remotely

0 500.000 1.000.000 1.500.000 2.000.000 2.500.000 3.000.000 3.500.000 4.000.000 4.500.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Number of debit card payments at the counter (in

thousands)

13 sales and services the amount of debit card payments has increased the most compared with 2016. It is mentioned that this is an increase of about 7 percentage points.

The Dutch Banking Association provide, in their factsheet “Betalen aan de kassa 2017”, also data about the spending behavior of the Dutch citizens. This information is given in figures 5 and 6. They divide the number of payment methods into 4 classes: Cash, Debit card payments (with contact), Debit card payments without contact and Credit card payments. The difference between the two debit card payments is that it is not necessary with the contactless option to enter a pin code to fulfill the payment. It is shown that men generally pay less with cash than women. This is partly explained by the fact that debit card growth in percentage point compared to 2016 has increased more for man than women. Furthermore, the number of cash payments increases as the age rises. This trend is only visible from the age group 19-24. On the contrary, there occurred a 4 percentage point growth of cash usage in the age group 12-18 compared to 2016. The cause of the high cash percentage in this age group might be explained by the fact that parents give their children pocket money, which is mainly in the form of cash. As mentioned earlier, small amounts are usually paid in cash. This can also be an explanation of the high percentage of cash in this age group, since pocket money and payments made by children are usually small amounts.

In figure 5 is indicated, by means of the red line, that the number of debit card payments is increasing among all age groups compared to 2016. This also includes the groups 12-18 and 75+, in which payments nowadays are mainly fulfilled with cash. People in the age group 19-24 make the least use of cash to meet payments. In this age group, debit card payments without contact are as large as debit card payments with contact. From the age group 19-24 a downward trend of these contactless debit card payments can be observed.

Besides this, it can be seen that cash usage decreases as the level of education increases, which is shown in figure 7. These education levels are divided into 5 levels. People are classified on the basis of their highest level of education completed. The red line in figure 7 indicates that the usage of debit cards compared to 2016 is increasing as education level increases. According to the level of education, the number of debit card payments (with contact) can be considered stable, while the number of contactless debit card payments is increasing as education level increases. In the basisschool group it is clear that the number of debit card payments is less compared to the other groups. However, this is explained by the earlier proven age effect. In the basisschool group there are only children. As previously shown, they pay more with cash than with debit and credit cards.

Figure 5: Spending behavior of the Dutch citizens judging from age (De Nederlandsche Bank, betalen aan de kassa 2017)

0% 20% 40% 60% 80% 100% 120% 12-18 19-24 25-34 35-44 45-54 55-64 65-74 75+

Spending behavior judging from age

14 Figure 6: Spending behavior of Dutch men and woman (De Nederlandsche Bank, betalen aan de kassa 2017)

Figure 7: Spending behavior of the Dutch citizens judging from education (De Nederlandsche Bank, betalen aan de kassa 2017) As mentioned before, digital payment instruments are an alternative for cash. This can be seen, among other things, through the figures concerning the Dutch ATMs. The World Bank has data about the number of ATMs per 100.000 adults, which is shown by figure 8.

Figure 8: Number of ATMs per 100.000 adults within the Netherlands (The World Bank, Automated teller machines (ATMs) (per 100,000 adults)

Man

Cash Debit (contact)

Debit (contactless) Credit

Woman

Cash Debit (contact)

Debit (contactless) Credit

0% 20% 40% 60% 80% 100% 120%

Basisschool VMBO MBO/HAVO/VWO HBO WO-docteraal

Spending behavior judging from education

Cash Debit (contact) Debit (contactless) Credit Growth

0 10 20 30 40 50 60 70 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total number of ATMs (per 100.000 adults)

15 From 2008 we see a decreasing trend of ATMs per 100.000 adults. This indicates higher costs per transaction, since the ‘shoe-leather’ costs will increase. These are the costs of getting cash at an ATM. When there are less ATMs, these costs increase. It also indicates, seen from the Baumol-Tobin model, that the number of cash transactions will decrease.

If we study this data further we see more trends. One of the indicators of money demand is the number of cash withdrawals at ATMs. As can be seen in figure 9, the number of cash withdrawals at ATMs in the Netherlands is decreasing. In addition, figure 10 shows the total value of cash

withdrawals at ATMs. This shows also a decreasing trend. This gives the impression that the demand for cash is actually falling.

Figure 9: The total number of cash withdrawals at ATMs within the Netherlands (in thousands) (De Nederlandsche Bank, Retailbetalingsverkeer)

Figure 10: The total value of cash withdrawals at ATMs within the Netherlands (in millions) (De Nederlandsche Bank, Retailbetalingsverkeer)

However, it is possible to explain this phenomenon in a different way. This is done by not only taking the cash transactions into account, but by studying all money transactions at the counter. This is done on the basis of figure 11 and figure 12. In figure 11, all money transactions at the counter (and within the Netherlands) are given. The total cash transactions consists of all cash withdrawals at ATMs with a Dutch card, cash withdrawals by Dutch citizens at the cash desk (or in another way) of their own bank, minus the cash deposits of euros to their bank. These three variables have been taken because the remaining part can be regarded as the total cash used for spending. In figure 11 is shown that the total cash transactions decrease over time. This decrease is smaller than the increase

0 100.000 200.000 300.000 400.000 500.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total number of cash withdrawals at ATMs (in

thousands)

Cash withdrawals at ATM's

0 10.000 20.000 30.000 40.000 50.000 60.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total value of cash withdrawals at ATMs (in millions)

16 of credit and debit card payments, which led to an overall increase of money transactions. In

addition, the credit and debit card curve gradually gets closer to the total transaction curve as time goes on. Figure 11 observes the digitization of payment instruments and gives the impression that money demand is increasing and cash demand is decreasing.

What is immediately noticeable is that Figure 11 differs from Figure 1. The Dutch Banking Association reveals figures, by mean of the factsheet "Betalen aan de kassa 2017", which indicates that the number of debit card payments has been greater than the number of cash payments since 2015. This does not seem to be the case in Figure 11. This implies that the total cash transactions is not a good indicator of the amount of cash used for payments. This is probably because the total cash transactions are not year-related. Nevertheless, these figures contain a lot of information about the demand for cash and are therefore usable.

Figure 11: Number of money transactions within the Netherlands (in thousands) (Dutch Banking Association, Retailbetalingsverkeer) Moreover, it’s also necessary to look at the total value of transactions. The value of all money transactions at the counter shows a less increasing trend as the number of transactions. This does not correspond with the expectations regarding the literature review, in which was stated that people will spend more using digital payment instruments. Especially since most debit payments take place in supermarkets, which has a large share in the number of counter payments. Figure 12 makes it clear that the increase of the value of credit and debit card payments just compensates the decrease of the value of cash transactions. According to figure 12, the money demand remains relative stable. Looking from this perspective, digital payment instruments are just an replacement of cash to facilitate transactions. Nevertheless, since 2015, there is an increase of the total cash

transactions. The reason for this is that the value of cash deposits of euros to the banks is decreasing a lot since 2015. This indicates that people keep their money or use it to spend more. Since we do not see an increase in the number of cash transactions in figure 1, it seems like people keep their money instead of using it for extra transactions.

0 500.000 1.000.000 1.500.000 2.000.000 2.500.000 3.000.000 3.500.000 4.000.000 4.500.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total number of transactions (in thousands)

17

Figure 12: Value of money transactions within the Netherlands (in millions) (Dutch Banking Association, Retailbetalingsverkeer) However, figures 11 and 12 only look at the debit and credit card payments at the counter. The digitization of payments instruments make it possible to perform payment transactions from a distance. Some of these payment methods are included in figures 13 and 14. The digital payment options that are added are: Cash transfers via telephone, internet, smartphone (app) or the tablet (app), which are available since 2013 and IDEAL payments, which are available since 2008. Figure 13 is not much different from figure 11. The only difference is that the digital payment instruments are a bigger part of the total transactions and the two curves became steeper afterwards.

Figure 13: Number of money transactions within the Netherlands (in thousands) (Dutch Banking Association, Retailbetalingsverkeer) In contrast to the number of transactions, the value of transactions differ a lot when the remote payments described above are added. Since 2012, figure 14 show an increase in the total value of transactions, which is caused by the increase in the number of digital payments instruments used. It corresponds with the aforementioned studies in the literature review, that stated that people will spend more using digital payment instruments compared with cash. This data makes it clear that the digitization of payment options both ensures more money transactions and an increasing value of these transactions.

0 20.000 40.000 60.000 80.000 100.000 120.000 140.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total value of transactions (in millions)

Total cash transactions Credit & Debit card payments Total transactions

0 1.000.000 2.000.000 3.000.000 4.000.000 5.000.000 6.000.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total number of transactions (in thousands)

18 Figure 14: Number of money transactions within the Netherlands (in thousands) (Dutch Banking Association, Retailbetalingsverkeer)

0 50.000 100.000 150.000 200.000 250.000 300.000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Total value of transactions (in millions)

19

Discussion and Conclusion

In this paper is investigated to what extent the use/retention of physical means of payment is necessary for the Dutch economy. First of all, the pros and cons of cash are stated. This showed that the costs of cash are higher than the costs of digital payment instruments. It is however, available for everyone and its existence is encouraged by the Dutch Banking Association and the MOB. They emphasize that the continued existence of cash is necessary to enable everyone to participate in the payment system. Thereby, the anonymity of payments is ensured by using cash. However, there are a number of papers to the disadvantage of cash. The arguments they used are among others the relative low transaction speed of cash and the high number of economic crime that’s influenced by cash. By contrast, digital payments instruments will contribute to the amount of digital crime. Subsequent, some papers found evidence of a changing spending behavior by the use of digital payment instruments. This means that cash led to fewer expenses compared with digital payment instruments.

In the empirical part of this paper it’s analyzed which trends are observed

actually/statistically and if these match the aforementioned theory. The Baumol-Tobin model of money demand is used, to see if our predictions about the money demand correspond with reality. According to this model, the costs of cash increased, which would lead to an increase of cash demanded. It seemed clear that this did not apply to cash. Both the value of the cash transactions and the number of cash transactions themselves decreased. The fact that this does not correspond to the Baumol-Tobin model is probably due to the fact that this is a dated model that does not take account of alternatives to cash, which do have effect on the demand for cash. The usage of the alternatives is increasing, most likely due to the lower costs compared to cash. If cash usage should be encouraged, its costs will have to be reduced.

This paper shows an overall picture and therefore does not deeply investigate specific topics. However, it can be used for follow-up studies. This primarily refers to the influence of digital money on the Dutch economy, which has been discussed, but has not been included in the data. Regarding the theory, it is expected that it will contribute to reducing the cash share as a part of the total transactions. In addition, digital money is a good example of the digitization of payment options, which continues to increase. Eventually more alternatives will arise that meet (a part of) the specific advantages of cash. This will threaten the existence of cash.

Returning to the research question, both digital payment instruments and cash are criticized. However, an increase in the usage of digital payment instruments and an decrease of cash payments is shown. In addition, this increase is observable in all the various classes of citizens. As long as this trend continues, cash will eventually disappear. This is supported by research in other EU countries, which concluded that a cashless society will arise gradually and not abruptly. Most important is that there are no signs that an economy without cash will not be able to function. The arguments in favor of cash are worthless as long as people keep using less cash. So the use/retention of physical means of payment is not necessary for the Dutch economy. This differs from my expectations beforehand.

20

References

Armey, Lipow, & Webb. (2014). The impact of electronic financial payments on crime. Information Economics and Policy, 29, 46-57.

Arvidsson, Niklas, Hedman, J., & Segendorf, B. (2017). Cashless society: When will merchants stop accepting cash in Sweden - a research model. Enterprise Applications, Markets And Services In The Finance Industry, Financecom 2016, 105-113.

Baumol, W. (1952). The transaction demand for cash an inventory theoretic approach. The Quarterly Journal of Economics, 66(4), 545-556.

Berentsen, A. (1998). Monetary Policy Implications of Digital Money. Kyklos, 51(1), 89-118. CBS. (2018) Aantal werkenden verder toegenomen. Retrieved from

https://www.cbs.nl/nl-nl/nieuws/2018/20/aantal-werkenden-verder-toegenomen De Nederlandsche Bank. (2016). Bereikbaarheidsmonitor. Retrieved from

https://www.dnb.nl/binaries/Bereikbaarheidsmonitor%202016_tcm46-349693.pdf De Nederlandsche Bank. (2018). Visie op betalen 2018-2021. Retrieved from:

https://www.dnb.nl/binaries/DNB_Visie-op-Betalen_tcm46-376439.pdf?2018060510 De Nederlandsche Bank. Betalen aan de kassa 2017. Retrieved from

https://www.dnb.nl/binaries/Factsheet%20Betalen%20aan%20de%20kassa%202017_tcm46-375218.pdf

De Nederlandsche Bank. Maatschappelijk Overleg Betalingsverkeer (MOB). Retrieved from https://www.dnb.nl/betalingsverkeer/overige-taken/maatschappelijk-overleg-betalingsverkeer/index.jsp

De Nederlandsche Bank. Retailbetalingsverkeer (Kwartaal). Retrieved from

https://statistiek.dnb.nl/downloads/index.aspx#/?fq=%7Ctheme%3Abetalingsverkeer&tags.li mited=0

Dewi, Lusiana Citra. (2010). An Introduction To Digital Cash World. ComTech, 1(2), 493-508. Dutch Banking Association. (2017). Fact Sheet Payments. Retrieved from

https://www.nvb.nl/feiten-cijfers/2082/betalingsverkeer.html European Council. (1992). Treaty on European Union. Luxembourg: OPOCE.

Gann, Dm, Dodgson, Mark, Wladawsky - Berger, I, Sultan, N, & George, Gerard. (2015). Managing Digital Money. Academy of Management Journal, 58(2), 325-333.

Garcia-Swartz, Daniel & W. Hahn, Robert & Layne-Farrar, Anne. (2006). The Move Toward a Cashless Society: A Closer Look at Payment Instrument Economics. Review of Network Economics. 5. 175-198. 10.2139/ssrn.641441.

Haegens, K., Bolwijn, M. & Lenskens, S. (2018 March 7). Pinnen? Ja graag/Nee liever niet. De Volkskrant, pp. A8, A9.

Kamada, K. (2017). "The Transactions Demand for Paper and Digital Currencies," IMES

Discussion Paper Series 17-E-06, Institute for Monetary and Economic Studies, Bank of Japan. König, J. (2017). Bargeld als Garant für Freiheit und Eigentum. List-Forum Für Wirtschafts- Und

Finanzpolitik, 42(4), 341-364.

Loke, Y. (2007). Determinants of Merchant Participation in Credit Card Payment Schemes. Review of Network Economics, 6(4), pp. 474-494. Retrieved 1 Jun. 2018, from doi:10.2202/1446-9022.1130

O’Neill, J., Dhareshwar, A., & Muralidhar, S. (2017). Working Digital Money into a Cash Economy: The Collaborative Work of Loan Payment. Computer Supported Cooperative Work (CSCW), 26(4), 733-768.

Pilbeam, K. (2013). International Finance. fourth edition, Basingstoke (UK): Palgrave MacMillan. Pp. 72-89.

Prelec, D., & Simester, D. (2001). Always Leave Home Without It: A Further Investigation of the Credit-Card Effect on Willingness to Pay. Marketing Letters, 12(1), 5-12.

21 and Form on Spending Behavior. Journal of Experimental Psychology: Applied, 14(3), 213-225.

Runnemark, Hedman, & Xiao. (2015). Do consumers pay more using debit cards than cash? Electronic Commerce Research and Applications, 14(5), 285-291.

Sahoo, Pradipta Kumar. (2017). Bitcoin as digital money: Its growth and future sustainability. Theoretical and Applied Economics, (4), 53-64.

Skaar, M., Torp, K., & Molnar, Peter. (2017). Money Demand and Quality of Institutions ; - A Baumol- Tobin Approach to Money Demand.

Tee, H., & Ong, H. (2016). Cashless payment and economic growth. Financial Innovation, 2(1), 1-9. The World Bank. Automated teller machines (ATMs) (per 100,000 adults). Retrieved from: