7. A Cashless Society: Facts and Issues

The Ministry of Economy, Trade and Industry's “Fin Tech Vision” defines the cashless payment ratio as the amount of cashless payments (payments made with credit cards, debit cards, and electronic money) as a percentage of private final consumption expenditure. In Japan, the ratio was 18.2% as of 2015, which is quite low, and the government set a numerical target (Key Performance Indicator: KPI) in its Future Investment Strategy 2017 to raise the ratio to around 40% over the next 10 years.

The figures and future numerical targets for the cashless payment ratio have been circulated through the mass media and other media, and there is a widespread perception that Japan is a cash-based society compared with other developed countries. However, we consider the progress of cashless society not only in terms of the quantitative aspect, such as the increase in the ratio of cashless payment methods used, but also in terms of qualitative improvements, such as the addition of new value that contributes to user convenience and the provision of payment methods at lower prices than in the past. Also, as Fuchida (2018) writes, the cashless payment ratio set as a KPI in the “Future Investment Strategy 2017” ignores traditional electronic payment methods such as bank transfers, automatic deductions, and account transfers, and it does not include inter-personal transfers between mobile phones or Payment Initiation Service Provider (PISP)-type payments are also not considered. In order to take these points into account, we proposed to add the share of bank transfers/account transfers in private final consumption expenditure.

However, it should be noted that transfers/account transfers include direct debits related to the use of credit cards and so on, thus simply adding them up may result in double counting. Here, we assume that all credit card payments are included in transfers/account transfers, and recalculate the ratio of cash and cashless payments, which were 8 versus 92% in 2015 data. This means that electronic payments accounted for 92% of the total, while cash payments accounted for less than 10%. At the same time, a recalculation of the same data for Germany, which was identified as having a lower cashless payment ratio than Japan, showed a ratio of 2.58 to 97.42%. According to this definition, Germany was even more cashless than Japan. Interestingly, the same recalculation for Singapore, which had the highest cashless payment ratio, showed that the ratio was 16.52 versus 83.48%, indicating that Singapore was lagging behind Japan and Germany in cashless payments. As for South Korea, which had a high cashless payment ratio of 90%, the ratio was 0.43 versus 99.57%, which means that the country had almost completely achieved cashless payments. This seems to be an overestimation of cashlessness, given the realities in South Korea. Furthermore, I think it is a matter of national accounts in China; the amount of credit card payments is more than twice as large as private final consumption expenditure. Whether this is because credit card payments include payments other than private final consumption expenditure or because private final consumption expenditure is underreported, cannot be determined for China, which is often mentioned as a country that is making progress in going cashless, strict international comparisons using statistics should be avoided (see Nakajima, 2017).

According to the Committee on Payments and Market Infrastructures (CPMI, 2016, 2017) in BIS, the cashless payment ratio in Japan appears to have increased steadily from 14.3% in 2011 to 18.2% in 2015. However, if we recalculate the ratio to include bank transfers and direct debits, the ratio remains almost unchanged at 8.48 versus 91.52% in 2011.

So what is the reality? To give an overview of the flow of funds in the household sector in Japan, including my personal experience, most salaries and other income are paid by bank transfer. From this, monthly expenditures are made, but the majority of housing, utilities, transportation and communications, and education are paid by direct debit. Durable consumer goods such as clothing, footwear, furniture, and household goods are also likely to be paid for by credit card or other means of payment rather than cash. The only areas of household consumption that are likely to be paid for in cash are food, education and entertainment, health care, and others. The total share of these items is estimated to be about 50% of total expenditures at most.

As will be discussed later, the choice of means of payment differs according to the range of payments, with credit card payments and bank transfers/direct debits being chosen for high-value expenditures exceeding 10,000 yen, and electronic money being used more frequently in recent years for payments of 1,000 yen or less. Taking these factors into account, there is still a possibility that cash will be used for payments in the range between 1,000 yen and 10,000 yen. It is not surprising to consider this to be about 10% of total household spending.

In summary, the government's statistic that cash spending accounts for 80% of private final consumption expenditure, including households, does not reflect reality, and although individual differences may exist, it is thought that cash spending is at most 10%, with the remaining 90% being paid electronically.

This chapter is not interested only in retail payments in Japan, but also in trends in private (household and corporate) payments in the Japanese economy as a whole and the financial system's response to these trends. In particular, when considering the financial system's response, it is desirable to consider the entire payment system, including bank transfers and account transfers as an issue.

According to Bank of Japan (2017), the ratio of cash balances in circulation to nominal GDP was 19.4% in 2015, the highest among the CPMI member countries. The Bank's explanation for this high cash balance is as follows. (1) The demand for cash as a means of storing value is quite high, reflecting the high level of security in the country, the low level of counterfeit currency, the high level of public confidence in banknotes, and the low opportunity cost of holding cash due to low interest rates. (2) Cash is generally accepted as a means of payment. This is because cash is valued for its various characteristics, such as compulsory acceptability, general acceptability, finality, and anonymity.

As is a common problem in economics, it is impossible to explain economic variables that change over time with constant institutions and characteristics. Please see Fig. 7.1. In 1994, the ratio was 9.26%, which is not very high by international standards. It is not likely that the institutions and characteristics used by the BOJ to explain the ratio in Japan doubled between 1994 and 2016. It is unlikely.

A scatter plot with an ascending trend describes the circulation of cash to G D P in ratio. The ratio doubles from 1994 to 2020.

Rather, the background here is probably that the Bank of Japan has continued monetary easing almost consistently and has increased the supply of currency (base money) at a pace faster than the rate of economic growth. I think there is a problem in linking this statistic to trends in the demand for currency by the private sector, including households, and the resulting slump in cashless payments.

The choice of payment method depends on the size of the payment, and the main payment method is selected according to the size of the payment, as was established in an accumulation of theoretical and empirical studies in this area to date: Kitamura (2005, 2010); a series of studies by the Bank of Canada (Bagnall et al., 2016; Kosse et al. 2017; Wakamori et al., 1998; Shy, 2001), and Garcia-Swartz et al. (2006a, 2006b). Specifically, cash or electronic money is used for small payments of 1,000 yen or less, cash or debit cards for payments of 1,000 to 10,000 yen, credit cards for payments of 10,000 yen to 50,000 yen, and bank transfers for payments of 50,000 yen or more. In terms of the scale of retail outlets, sales of 1,000 yen or less are mainly at convenience stores and station kiosks, sales of 1,000–10,000 yen are mainly at supermarkets and individual specialty retailers, purchases of 10,000–50,000 yen are at department stores. For purchases of more than 50,000 yen, it would be high-end consumer durables specialty stores and high-end service stores. Each retailer already offers a payment method that matches its own payment range. In other words, convenience stores and kiosks offer electronic money such as Suica, WAON and nanaco; supermarkets offer credit cards with low annual fees; and department stores offer membership cards with credit card functions that charge a modest annual fee and provide extensive customer service.

If the cashless society is to be further advanced, the remaining areas of cash settlement will be (1) substitution of electronic money and debit cards for payments of small changes of 1,000 yen or less, and (2) substitution of credit cards, debit cards, and electronic money for payments at supermarkets and specialty retailers of between 1,000 yen and 10,000 yen.

In fact, in this field as well, supermarket e-money has expanded its charge limit to 50,000 yen and is building a system capable of handling payments of 10,000 yen or less. In addition, Suica and PASMO, which are transportation-related e-money, are working to expand the range of settlement amounts by issuing higher-end cards with automatic balance recharging and credit functions.

As shown in Chap. 5 (originally in Kitamura et al., 2009), the frequency of use of small coins such as 1 yen and 5 yen is steadily decreasing with the advent of electronic money, and a coinless system for small coins is under way.

A series of empirical and simulation studies conducted by the Bank of Canada has shown that consumers have a preference for cash that cannot be explained by various costs. So what other factors might be at play here?

When looking at settlement costs, what is often overlooked are the points given to electronic money and credit cards. From a common sense perspective, this is a return for the savings in management, accounting, and transportation costs associated with cash when using electronic money or credit cards, and at the same time, it is part of a strategy to attract customers by giving them an incentive to continue using electronic money or credit cards. It is also thought to be part of a strategy to attract customers by giving them an incentive to continue using electronic money and credit cards.

On the other hand, credit cards also require retailers to pay a fee, so some offer discounts for cash payments. There are still some long-established sushi restaurants and traditional Japanese cuisine restaurants that only accept cash payment. It is still there.

In this way, the incentives for choosing a payment method vary and do not seem to have a structure that can be statistically lumped together. In addition, the idea that it is desirable for retailers to offer a variety of payment options to acquire customers is not necessarily an accepted idea.

Although bank transfers and account transfers are widely used, they should be available 24 h a day, 365 days a year for convenience, but this is not yet the case in Japan. The 24-h operation of ATMs is also common practice overseas, but this is also not available in many places in Japan.

In the past, banks were not allowed to issue credit cards, but their affiliates issued credit cards in cooperation with credit card brands (Visa, MasterCard, etc.). Even today, when banks are able to issue credit cards, they do so in cooperation with the major credit card brands. As a result, credit card operations and management decisions are based on the judgment of the credit card brands, and may not necessarily be in line with the realities of the Japanese retail industry. The number of credit cards issued in 2017 was 272.01 million, a penetration rate of about 2.6 cards per adult. The conditions for issuing (or obtaining) a credit card in Japan are considered to be looser than those in other countries, and the use of credit cards is increasing. The use of debit cards is limited due to the fact that the number of stores where they can be used is still small and the hours of availability are restricted.

Electronic money cannot be used internationally because it cannot be withdrawn in cash (only one-way charging of cash to electronic money) and its specifications are different from those of foreign electronic money. Cross-border payment is the issue of e-money and Central Bank Digital Currency (CBDC) to be determined in the near future. From the point of view of foreign travelers, there is also the inconvenience of not being able to use the e-money in Japan that they use in their home country.

As described above, the choice of payment method is not necessarily made under the same conditions, or on equal footing. As mentioned, there is a segregation of payment methods according to industry and payment scale, and it can be judged that cash payment is selected to some extent among them.

Fuchida (2017, 2018) and Nikkei MOOK (2018) enumerate the benefits of going cashless.

Of these advantages listed, (1), (2), (3), and (4) can be achieved if cash is completely abolished, but they do not apply as long as cash currency continues to be issued. As for (2), if the demand for cash declines and the amount in circulation decreases, the cost may be reduced accordingly. The extent to which the number of ATMs and the vehicles to transport cash will be reduced will depend on the amount of the decline in demand. As for (4), currencies are more contaminated by mold than usually thought, and the risk of a pandemic would decrease if the use of currencies declined. Japanese currency is updated relatively frequently, and I have not heard of any cases where it has served as a vector for epidemics.

One of the arguments for (5) is that it eliminates the need to exchange cash and the need to confirm and count amounts. In terms of finality of payment, there is no quicker means of payment than cash payment. Electronic payments, even with debit cards, are slower than hand-delivered cash. It is true, nonetheless, that the time and effort required to confirm and count cash can be reduced if electronic payments are used. The point in (6) is that travel time and waiting time can be eliminated. Even with electronic payments, it takes time to operate a computer terminal or mobile phone, and if the payment does not go well, the customer may have to go to a bank, so it is not possible to assume that alternative payment methods will not eliminate loss of time. (7) Also, as long as cash currency remains, there is a limit to the cost reduction. This is more likely to be brought about by the evolution of FinTech using artificial intelligence (AI) than by going cashless.

Points (8) and (9) have been cited as benefits, considered from an economy-wide perspective. The argument in (8) has been proposed by Rogoff (2017) and many other experts; but empirically, it is unclear to what extent the underground economy and crime would be reduced. Simply put, if there is a separate objective of tax evasion and crime in the underground economy, going cashless will only lead to flight to anonymous payment methods that are alternatives to cash. Public finance economists regularly measure the size of the underground economy, and Japan's underground economy is estimated to be less than 10% of GDP, which is considerably lower than that of other developed countries. The fact that there is little relationship between cash and crime is probably one of the reasons cash can be used with confidence. As for point (9), it has been widely confirmed that citizens of developing countries who had no contact with finance until now have been able to participate in many economic activities by acquiring payment functions through mobile phones. However, this is probably an issue that is not directly related to the cashless society as it happened, because the information and communication industry entered the payment business.

If we assume that cash will continue to exist, there will be additional costs to the existing cash management costs, such as infrastructure development and technology investment for cashless transactions. If we assume that cash will continue to exist, it would be correct to assume that there will be additional costs, such as infrastructure development and technological investment, to existing cash management costs. In this context, if cash management costs are reduced by progress in the shift to cashless transactions, and if there is still a surplus even after investing the extra funds in infrastructure and technology development for cashless transactions, then the shift to cashless transactions will continue. On the other hand, if cash management costs cannot be reduced to such an extent that the cost of additional capital investment becomes a burden, progress toward a cashless society may not be as rapid as it could be.

As we will discuss in Sect. 7.5 of this chapter, if we consider that the shift to a cashless society is a by-product of the evolution of AI and FinTech, particularly in financial institutions, it is difficult and too simplistic to merely compare cash management costs with capital investment costs for a cashless society.

When considering the advantages of going cashless, many argued that it would reduce the cost of maintaining cash. The disadvantages of going cashless are, of course, the inconvenience of suppressing the use of cash and the fact that the scale of theft is limited by the physical constraints of cash. With electronic money and virtual currency, theft on a large scale that is unthinkable with cash can, in principle, occur at once.

Some countries, such as Sweden, Denmark, the United Kingdom, and South Korea, have given some retailers the authority to refuse to accept payments in cash in order to promote cashless transactions. One can argue that such measures are necessary to some extent to promote a cashless society, but cash is the so-called last means of payment for those who are alienated from finance. Refusal of cash payment may be acceptable in clubs or schools where the members are fixed, but it is preferable not to allow refusal of cash payment in places where an unspecified number of people may carry out transactions.

To begin with, the government's “Strategy for the Revitalization of Japan, Revised 2014” stated that the government would “work to improve the convenience and efficiency of payments through the widespread use of cashless payments in light of the hosting of the 2020 Olympic and Paralympic Games in Tokyo and other events. In other words, the government's proposal to go cashless included the goal of making it more convenient for the rapidly growing number of anticipated foreign tourists.

In this light, we should not be so preoccupied with promoting a cashless society that we inconvenience not only domestic consumers but also foreign tourists in making payments. I myself have been refused cash payments in Denmark and the UK, but I managed to cope with the situation using a highly versatile credit card. In the trend of internationalization, the payment methods that are easy for foreign tourists to use are those that they use in their home countries (e.g., Alipay, Apple Pay, bank debit cards). In order to make these payment methods usable in Japan, it is necessary to create a mechanism to authenticate the payment methods in the Japanese payment system.

Japanese financial institutions are open to the idea of going cashless in the sense of replacing a portion of domestic cash payments with credit and debit card payments, or replacing cash by issuing privately issued digital currency using blockchain technology, but they are very cautious about allowing compatibility with electronic money and debit cards issued overseas. A delay in addressing this part of the problem could lead to the creation of a Galapagos-island-like cashless society specifically evolved only in Japan, which would indirectly exclude foreign travelers.

The other disadvantage is the security problem. On January 26, 2018, 0.5 billion cryptocurrency XEM (currency unit of NEM, equivalent to 58billion yen at market value) was stolen via on-line within ten minutes or so. At the end of March 2018, the entire amount 0.5 billion XEM was converted into other cryptocurrencies and ultimately into legal tender, and the culprit has yet to be caught. The theft of cryptocurrency has often occurred, but this was the first theft on this scale in history.¹

A robbery of this scale would not have happened in a cash-transport robbery. In other words, to transport 58 billion yen in cash at one time, a truck with a considerable transportation capacity, a considerable number of people, and a considerable amount of time are required, and it is almost impossible to do so without being noticed by a third party.

Coincheck, which was involved in the incident, was found to have a number of problems with its security management after the fact, and was given a business improvement order by the Financial Service Agency (FSA). The incident prompted a review of the FSA's response to the security management system for crypto-currency exchanges under the amended Funds Settlement Law, and on March 8, 2018, the FSA issued business improvement orders to seven exchanges, including Coincheck.²

In contrast, debit cards, credit cards, and e-money are properly managed, and banks and other financial institutions may take the view that their security measures are perfect, but as with crypto-currencies, security problems occur where controls are weak. Specifically, there have been frequent incidents of credit card fraud through breaches of credit card information at the retail level. There is also no end to the number of elderly people who become victims of fraud through the use of counterfeit cards and lax cash-card authentication.

It is also impossible to assume that central banks are safe. In fact, although it is unclear how often crises that threaten the security of central banks occur, it seems that attempts are regularly made to use internal email information obtained through unauthorized access to send malware by impersonation.

On February 4, 2016, after business hours, hackers used malware to infiltrate the Bangladesh Central Bank's systems and, using a hijacked account, sent fake money transfer instructions to the Federal Reserve Bank of New York via SWIFT, successfully transferring funds to bank accounts in the Philippines and other locations. The total amount affected by the fraudulent remittance was $101 million, of which about $81 million (about ¥9.2 billion) has not been recovered. It has been reported that such crimes involving central banks have also occurred in Vietnam, Ecuador, and other countries, and the central banks and private banks of each country are required to share information and strengthen security.

As we have already seen, more than 90% of household payments are made electronically, and an even higher percentage of businesses rely on electronic payments. The reality is that cash holdings have been whittled down to a minimum, and the trend toward a cashless society is unlikely to stop. As a result, it is only natural that criminals will shift their focus from cash to electronic payments through the manipulation of digital information. Even if the government promotes the shift to cashless transactions, it should take sufficient security measures, recognizing that the risk will be exposed to consumers and retailers, who are the least security-conscious, and that criminals will exploit this risk.