Electronic money has been around for about 10 years, but it did not become as popular as initially hoped in the 1990s. However, in 2000, Suica by East Japan Railway Company and Edy by BitWallet, Inc. began to spread as more common electronic money, and electronic money began to be used in earnest. The purpose of this chapter is to discuss, from both theoretical and empirical perspectives, the mechanisms and systems necessary for the further diffusion of the new e-money that began to emerge in the beginning of 2000s, drawing on the experiences of the 1990s.
4.1 Introduction
Electronic money has been around for about 10 years,1 but it did not become as popular as initially hoped in the 1990s. However, in 2000, Suica by East Japan Railway Company and Edy by BitWallet, Inc. began to spread as more common electronic money, and electronic money began to be used in earnest. The purpose of this chapter is to discuss, from both theoretical and empirical perspectives, the mechanisms and systems necessary for the further diffusion of the new e-money that began to emerge in the beginning of 2000s, drawing on the experiences of the 1990s.
The structure of this chapter is as follows. In Sect. 4.2, we survey the distribution of e-money, focusing on Suica and Edy, which have recently been expanding rapidly. Section 4.3 discusses the mechanisms for promoting e-money circulation from the perspective of e-money's transaction cost advantage, technological future, and scalability. Section 4.4 theoretically analyzes how e-money will be selected from multiple payment methods such as cash and credit cards, based on the actual situation up to that point. Here, we show that the usage ranges of e-money, cash, and credit cards are determined by the differences in their respective transaction costs. We also show that in practice, the use of e-money depends not only on the cost difference but also on the number of consumers per shop. In Sect. 4.5, I examine the implications of e-money for monetary policy by examining how the cash currency issued by the central bank would eventually increase in terms of monetary aggregates under the presence of e-money. I also show that the setting of the deposit reserve ratio for electroncic money is important in this process. Section 4.6 provides a brief summary of the overall results.
Advertisement
4.2 The Actual State of Distribution of Electronic Money
Today, Suica and Edy have become popular as mediums that serve as prepaid electronic wallets (e-money). Of course, their scale is still small, but with the entry of major transportation-related companies such as JR, JAL (Japan Airlines), and ANA (All Nippon Airways), their widespread use is gaining momentum.
The electronic money in circulation in Japan is not limited to Edy and Suica, but since these are the two that are spreading on a national scale, we will focus on them.2
As of the end of April 2005, 10.2 million Edy cards had been issued by BitWallet, which can be used at1,400 “am/pm” convenience stores and 20,000 other stores nationwide, including department stores, pharmacies, bookstores, hotels, airports and amusement facilities.3 Edy cards can be purchased for 315 yen at convenience stores and other locations, and up to 50,000 yen can be deposited at terminals at partner stores. There are also more than 500 Edy-enabled websites, making it easier to shop online. Member stores are also offering discounts to encourage the use of Edy. Among other things, the alliance with ANA Mileage Club allows users to exchange 10,000 miles for 10,000 yen worth of Edy, earning one mile for every 200 yen spent using Edy, or one mile for every 100 yen spent with credit cards such as VISA and Mastercard. You can earn one mile for every 100 yen you spend. The same concept can be applied to Mizuho Mileage Club. The other mainstay of widespread use, cooperation with mobile phones, is also under way. Edy is included in NTT DoCoMo's Osaifu-Keitai (mobile wallet), and is set to be included in au and Vodafone from autumn 2005.4
Suica, on the other hand, originated from the contactless payment system (automatic ticket gate) at stations introduced by JR East, and electronic money functions were added in March 2004. A total of 12.01 million Suica cards had been issued as of the end of April 2005, and more than 6.85 million shopping-enabled Suica cards have been issued. Suica cards are sold from vending machines at JR stations for 2,000 yen each (including a 500 yen deposit), with a maximum deposit of 20,000 yen.
Advertisement
As of April 2005, the number of stores where Suica can be used is said to have exceeded 1,000 at 802 stations. From 2005, Suica will be available for payment at 2,800 convenience stores, including New Days (300 stores) and FamilyMart stores in the Tokyo metropolitan area and Sendai area. It is also said that Suica will be available on private railways and buses in the Kanto area in 2006. However, compared to Edy, its use outside of train stations is limited and the number of stores where it can be used is still small.
As a benefit of using Suica, when you charge Suica on the View Card, a credit card issued by East Japan Railway Company, you can earn three times the amount of Thanks Points that you normally earn based on the amount you spend on your credit card. While Edy is tied up with ANA, Suica is tied up with JAL, and 10,000 miles can be exchanged for 10,000 yen worth of Suica as JAL mileage points (with a limit of 20,000 miles per year). The company also has a partnership with Mizuho Mileage Club. Although it is behind Edy in terms of mobile phone integration, it will be included in DoCoMo's Osaifu-Keitai mobile phone system as Mobile Suica from January 2006.
Both Edy and Suica are basically highly anonymous, but this also means that they can be used by a third party if they are lost. It is thought that the upper limit on the amount of money that can be loaded is set with this risk of loss in mind. However, in the case of commuter pass-type Suica, the user's name is registered, and the use of the lost Suica is stopped and the balance is reissued. Osaifu-Keitai (mobile wallet) also has a function to lock the mobile phone and prevent a third party from using it if it is registered in advance. In this way, it has become possible to maintain anonymity regarding the use of e-money and, to some extent, reduce the risk of loss. However, there is still a risk of losing the e-money if the card containing the e-money is physically damaged or submerged in water.
4.3 Mechanisms to Promote the Distribution of Electronic Money
In general, when a new product is introduced, the extent to which the product is accepted in the market is a matter of substitutability with existing products. In the case of electronic money, cash, prepaid cards, debit cards, and credit cards are considered to be substitutes. The question is how e-money can replace cash and various types of cards that are already in circulation.
The difference between e-money and a credit card is that you can settle your payment immediately. It can be used by people under 18 years old who cannot have a credit card. The difference with debit cards is that they do not require identification number and are faster to settle, but there is a limit to the amount that can be charged. E-money is designed to be used for small payments, and is intended to be used for that purpose. While prepaid cards are generally disposable, e-money can be recharged and used over and over again. Also, since e-money uses IC cards, it can record a large amount of data. This makes it possible to accumulate and use data such as points attached to various types of consumption and personal consumption history.
From the merchant's point of view, this reduces cash-holding balances, which reduces handling costs and reduces the risk of theft and other holdings. It is known that settlement during busy times can be shortened by about 25%. Another advantage is that while there is a risk of receiving counterfeit currency in cash settlements, there is no counterfeiting yet with electronic money, and it is safer than cash.
If these differences were considered sufficiently differentiating, e-money would be accepted as a means of payment and settlement, as it is for cash and various cards. But the advantages mentioned above were already there from the beginning of the first introduction of e-money in the 1980s and 1990s, and at that time it was not distributed nor was it promoted, thus other reasons must explain its success since the beginning of the year 2000.
According to BitWallet, (1) electronic money has become contactless, speeding up the exchange of information and making it possible to integrate it into mobile phones and other devices, as well as making it possible to use payment terminals at low cost. (2) By collaborating with transportation-related companies that have networks, the use of e-money has expanded from regional money to national money, which in turn has led to the participation of businesses, an increase in customers, and a reduction in costs. (3) The development of infrastructure (e.g., JR ticket gate systems) to accommodate the use of e-money.
Although it is beyond the scope of this chapter to analyze the technical aspects of (1) and (3), it can be said that the infrastructure for the distribution of e-money was promoted in cooperation with other network-related businesses. (2) is discussed below, but it can be assumed that cooperation with transportation-related companies was a major factor in transforming what had been a closed regional e-money system into a nationwide e-money system.5
Until now, cash recharging could be done easily at stations with Suica, but Edy was not so widespread, although recharging terminals were installed at convenience stores. There was also the problem of the cost of recharging using a PC, which required a separate IC card reader/writer. When e-money functions were installed on mobile phones, recharging became possible via i-mode, eliminating the need to install a recharge terminal in the first place and making it possible to recharge anytime, anywhere. Users, e-money issuers, and merchants alike will benefit greatly from this feature, as they can download various applications from the i-appli of i-mode, so there is no need to have them pre-installed on their mobile phones, and only the applications they use need to be loaded.
It is envisaged that the mainstream form of electronic money distribution in the future will be through mobile phones, and that electronic money will be exchanged between individuals, rather than being a mere means of payment, and that electronic money will be distributed in a form similar to current currency. In addition, by linking to mobile phones, it will be possible to purchase JR seats or concert tickets by making a reservation over the phone, and enter the venue using the mobile phone instead of the ticket as it is. In addition, if a contract is made to set a certain minimum amount of electronic money to be held, it can be automatically recharged at automatic ticket gates or through digital communication. In this way, the cost of recharging would be dramatically reduced.
I think it is safe to say that we are finally coming to the aspect of the use of electronic money on the Internet that was thought to be possible when electronic money was first introduced, and as a result, electronic money is being used more and more.
On the basis of these observations, let us theoretically derive the conditions for e-money to replace cash and other payment instruments.
4.4 The Problem of Choosing Between Multiple Payment Methods
According to Iwamura (1996), the segregation of settlement instruments can be represented as shown in Fig. 4.1.
A plot of low to high, use frequency of payment methods for micro, small, large, and super settlements. Credit cards are used for small transactions, and electronic for the rest.
Fig. 4.1
Conceptual distribution of payment methods
×
That is, settlements are for large amounts of money exceeding one billion yen, which include both inter-company settlements and inter-governmental transfers such as development assistance. Cash is rarely used for such settlements, but rather checks are exchanged between financial institutions or, today, almost exclusively, electronic payments between financial institutions (large electronic fund transfers, or EFTs). Smaller settlements, ranging from tens of thousands of yen to hundreds of millions of yen, are now mostly made by electronic fund transfer (small transfer EFTs). Smaller transactions, from a few thousand yen to tens of thousands of yen, are settled in cash or by credit card. Cash is legal tender and is accepted by all people, but credit cards are accepted only by a limited number of merchants. However, credit cards are convenient when traveling abroad because they allow people to make purchases of a certain amount of money without having to carry cash. Electronic money, which is the subject of this chapter's analysis, is generally expected to be used for very small payments, such as substituting for cash, especially small denominations. Looking at Fig. 4.1 from another perspective, it can be interpreted as follows. Settlements of tens of thousands of yen or more are already made using electronic media, such as credit cards and electronic payments, not by using cash, but by transferring information. This means that the use of cash is finally being limited by the introduction of electronic payment methods for very small settlements, which had been the last area of use for cash.
The problem of choosing multiple payment instruments began with the transaction cost currency demand model between deposits and cash of Baumol (1952) and Tobin (1956), and the problem of choosing multiple payment instruments including checks, credit cards, and other media was first discussed by Santomero (1979) and Whitesell (1989, 1992) and others, and was formulated as a general payment instrument selection problem in Santomero and Seater (1996). Although not multiple means of payment, Jovanovic (1982), Romer (1986), and Prescott (1987) formulated the Baumol–Tobin transaction cost currency demand function in the framework of general equilibrium theory.
Building on the accumulation of these studies, Shy and Tarkka (1998, 2002) present a convincing argument for the use of electronic money as a means of payment versus traditional payment methods such as cash, credit cards, and checks through a comparison of various costs, including transaction costs. The following is a summary of the discussion. In what follows, I will introduce the arguments of Shy and Tarkka (1998, 2002) and examine the conditions for and the scope of the use of electronic money. Electronic payments and payments by check, which appear in Fig. 4.1, are not considered in the following model because they are large in scale and cannot be substituted for electronic money.
4.4.1 Multiple Payment Models
We consider three types of payment instruments: electronic money (e-money), cash, and credit cards (abbreviated as “cards” in the following). We assume that there are four economic agents: (1) consumers or purchasers, (2) shops or merchants, (3) electronic money issuers, and (4) credit card issuers.
Let p (p > 0) be the value (price) paid in a transaction. Each shop is distributed according to the average transaction size p, which is called a shop of type p. The density function of this distribution of type p stores is denoted by m(p). This represents the number of shops of type p. Similarly for consumers, consumers with average price paid p are called type p consumers, and their density distribution is denoted by b(p).
We make the following assumptions about the distribution of merchants and consumers.
Assumption (1) b(p) and m(p) are differentiable continuous functions, and b(p) ≥ m(p) ≥ 0 for all p.
Assumption (2) b(p) is a decreasing function of p. The number of consumers according to the payment scale decreases with the amount.
Assumption (3) m(p) is a decreasing function of p. Shops based on the size of the transaction decrease with the settlement amount.
Assumption (4) The number of consumers per transaction per merchant decreases with the transaction value.
The assumption is that the number of customers of small retailers will be larger than the number of customers of large retailers.
Assumption (5) The consumer's density function b(p) is independent of the transaction costs of the payment instrument.
The assumption is that transaction costs affect the choice of payment instruments, but that transaction costs do not affect the sale or purchase of goods and services.
A detailed breakdown of the transaction costs for each economic agent is summarized in Table 4.1. The time cost T falls on the shopkeeper and the consumer. This is the cost of handling cash, such as calculation time and authentication cost. The theft and loss cost (λ) is incurred by the merchant, the consumer, and the e-money issuer. In the case of credit cards, theft insurance is included in the annual fee and is not a cost to the issuing entity. The cost of lost interest opportunity (i) is the cost of losing interest income by holding the money in cash. Here, the difference in holding period (v) is added to the interest opportunity cost in the case of a shop, a consumer, and an e-money issuer. For e-money, there is a technical breakage cost (γ) in the sense that if the magnetic data part is physically destroyed, it cannot be used. For credit card issuers, the authentication cost (VC) is the only cost. The authentication cost does not depend on the transaction amount.
Table 4.1
Transaction costs in the payment system
Cash transaction costs
Theft and loss costs (probability)(1)
Interest loss opportunity costs (interest rate)(2)
Technical damage costs
Authentification costs
Card user annual fee(3)
Card transaction fee (flat)
Card transation fee (Pay per use)
Payment terminal equipment instalation costs
Merchants
TM
λM
i
0
0
f0E
f1E
f2E
EE
Buyers/Consumers
TB
λB
vi
0
0
f0C
f1C
f2C
EC
E-money Issuers
0
λB
vi
γB
0
Credit card Issuers(4)
0
0
0
0
VC
Note (1) We assume that 0≤λ≤1.
Note (2) v is the relative rength of cash houlding period, compared with the merchants.
Note (3) Cards include both Electoronic cards (denotes E) and Credit cards (denotes C).
Note (4) Credit cards use the post paid system, there would be no opprtunity costs in interest rates. Both theft and loss costs and technical damage costs are covered by insurance (a part of annual fee), so that credit card issuers do not bear these costs.
Electronic money issuers and credit card issuers charge merchants an annual membership fee (franchise fee \({f}_{0}\)) and a transaction-based fee (divided into a fixed portion \({f}_{1}\) and a pay-as-you-go portion \({f}_{2}\)) for using their cards.6 In addition, electronic money and credit cards are subject to a communication equipment investment cost (E) for electronic payments.
The comparison of the various costs mentioned above by each economic entity will determine the extent of the use of payment instruments. For the consumer, using a credit card is the most cost-saving. However, merchants will only accept credit cards if the transaction costs are lower for credit cards than for cash. This condition can be expressed as follows.
With \(f_0^C = 0,\,\,f_1^C = V^C\) we now assume7 that if the transaction value p exceeds the lower bound given by (4.2), then credit card use outperforms other payment methods.
Next, we seek the extent to which e-money dominates. Merchants are more willing to use e-money than cash if there are no annual fees or transaction fees for e-money. The problem is that the boundaries of use are determined by the extent to which consumers are less willing to pay in cash than they are to pay in e-money at lower transaction costs.
The above results are summarized in Fig. 4.2. The classification of payment methods here is based on optimal boundaries. This result is consistent with the concept of the distinction of payment methods shown in Fig. 4.1.
A comparison line with payment methods indicate the condition under which merchants refuse credit card, and buyers refuse e-money.
Fig. 4.2
Equilibrium boundary of payment methods. Source Shy (2001), Fig. 8.7, p. 210
×
The following additional results are also obtained (Shy & Tarkka, 1998, Proposition 2).
Additional Result (1) A necessary condition for cash to remain is that the technical breakage cost of e-money is positive (γ B > 0).
Additional Result (2) A necessary condition for cash not to be eliminated by credit cards is that the authentication cost of credit cards is higher than the cash handling time cost (\(V^C > T^M\)).
So far, the optimal boundary between e-money, cash, and credit cards has been obtained by calculating the cost of each economic agent. However, this does not necessarily coincide with the socially optimal boundary. That is, the region where e-money is socially preferable to cash is when the total transaction cost of both merchants and consumers is lower for e-money than for cash.
Comparing this socially optimal boundary with the boundary in (4.3), if the relationship is as follows, it means that e-money is only underused in terms of the social optimum.8
Similarly, the area where credit cards are socially preferred to cash is when cash is higher than credit cards in terms of total transaction costs for both merchants and consumers.
Comparing this socially optimal boundary with the boundary in (4.2), it means that the credit card is only underused in terms of the social optimum if the relationship is as follows.
These results imply that e-money and credit cards are only underused in terms of the social optimum, while cash is overused in terms of the social optimum. This can be thought of as a kind of market failure, but as Fig. 4.3 shows, with respect to e-money, this is ultimately because the area of lower transaction costs is lower than the social optimum level when consumers pay with cash, and this is happening because the technological breakage cost of e-money is positive (\(\gamma^B > 0\)). If there is technological progress and the technological breakage cost of e-money declines, the area in which cash is used is expected to shrink.
A line with payment methods represents that e money and credit cards are only underused in terms of the social optimum.
Fig. 4.3
Discrepancy between transaction-cost-based boundary and socially optimal boundary
×
As for credit cards, in areas where the cost of credit card transactions is higher than that of cash, the level of credit card acceptance is higher than the socially optimal boundary, resulting in underuse of credit cards. As for credit cards, if the cost of authentication decreases, the credit card usage area will increase.
4.4.2 Competitive Conditions in the Market
So far, the boundaries of payment instruments have been determined basically only by comparing transaction costs. In the following, we consider the case of an oligopolistic market with a small number of credit card and e-money issuers, the case of a competitive market with a large number of issuers, and the case of a variable number of consumers based on the amount of transactions per shop.
The conclusion reached by Shy and Tarkka (1998) was that changes in the market structure would change who bears the burden of annual membership fees (connection and affiliation fees) and payment and communication equipment investment costs associated with card use. The following section presents the argument.
If the e-money issuer is an oligopoly, then the e-money issuer can collect all the surplus from the merchants. However, if the number of customers is small, the e-money issuer may not be able to collect any surplus. In this case, the e-money issuer will waive the annual card usage fee (connection fee) only for those merchants for whom the following equation holds (\(f_0^E = 0\)).
If a shop cannot satisfy this equation, it will either not be allowed to connect to e-money payments or will be required to connect for a fee. Thus, whether a shop is excluded from the e-money payment connection depends on the distribution of \(b(p)\) and \(m(p)\).
In the area where electronic money is used, \((0,\overline{p}]\), the maximum value of \(\overline{p}\), is determined to be the minimum value that satisfies Eqs. (4.9) and (4.3). In other words, \(\overline{p}\) is determined so that the following equation holds.
If e-money issuers are in competition, the annual fee for e-money card usage will be set to equal the cost of payment and communication equipment investment and will be borne by merchants. No other transaction fees are required. This means \(f_0^E = E^E ,\,\,\,f_1^E = f_2^E = 0\) that In this case, too, a shop will accept e-money only if the relationship in (4.9) is satisfied.
Shy and Tarkka's (1998) argument here is that whether the card issuer or the merchant bears the burden of annual fees and transaction fees for using the card indicates the state of competition in the e-money card market. An intuitive explanation for this result can be given as follows. The oligopolistic e-money issuers set their transaction fees to be slightly below the merchants’ cash transaction costs to maximize their surpluses, so they cannot afford to make the merchants pay annual fees, and the e-money issuers are forced to pay the annual fees. If merchants were required to pay an annual membership fee, the transaction costs of e-money would exceed those of cash, and merchants would withdraw from the use of e-money. On the other hand, competitive e-money issuers cannot set their transaction fees so as to recover all of the merchant's surplus, so the merchant will be left with some surplus, and as a result, the annual membership fee (connection fee) will be borne by the merchant.
Equation (4.10) implies that the e-money usage area is determined by the relative density of consumers and shops and the transaction cost ratio of e-money and cash. In the following, we show how the e-money usage area is divided into the e-money usage area and the cash usage area according to the \(b(p)/m(p)\) curve and the shape of the \(E^E /(T^M + \lambda^M p)\) curve.
In Fig. 4.4, the \(b(p)/m(p)\) curve is a monotonically decreasing function, and at the point of intersection with the \(E^E /(T^M + \lambda^M p)\) curve, the area is divided into the e-money use area and the cash use area. In Fig. 4.5, the \(b(p)/m(p)\) curve has a U-shape, so it intersects the \(E^E /(T^M + \lambda^M p)\) curve twice, and the pattern is that it changes from the e-money use area to the cash use area, then to the e-money use area again, and then back to the cash use area. This is a theoretical possibility, but it is difficult to imagine a situation in which the usage area switches in this way in reality. In Fig. 4.6, the \(b(p)/m(p)\) curve is consistently lower than the \(E^E /(T^M + \lambda^M p)\) curve, and the two curves do not intersect. This implies a case where the relative density of consumers and shops is so low that the incentive to use electronic money does not work. Specifically, it represents a situation where there is no need to use e-money, such as a souvenir shop in the countryside with a low population density, because few tourists visit the shop and the investment cost of payment and communication equipment is much higher than the time cost.
A pair of intersecting decreasing sloped graphs represent e money use below the point of intersection and cash use area above.
Fig. 4.4
A simple division between e-money and cash. Source Shy and Tarkka (1998), Fig. 3
A graph with an U shaped curve intersecting a decreasing sloped curve. The 2 points of intersection represent the switch between the e money use area and the cash use area.
Fig. 4.5
A complex division between e-money and cash. Source Shy and Tarkka (1998), Fig. 4
A graph with a pair of decreasing slopes. It represents a relative low density of consumers and shops and the the use of cash and no e-money.
Fig. 4.6
The case of no e-money. Source Shy and Tarkka (1998), Fig. 5
×
×
×
4.5 Implications for Monetary Policy
In this section, we will consider whether the emergence of e-money will reduce the effectiveness of monetary policy of the central bank that supplies money, and what restrictions should be imposed on issuers of e-money in order to maintain the effectiveness of monetary policy.
The currency aggregate (M1) by the narrow definition consists of cash (C) and deposit (D). Let us further consider the multiplier effect when electronic money (EM) is included in the model. As we saw in the previous section, there is substitutability between e-money and cash, and as the use of e-money expands, the use of cash will decline. If the use of cash declines, reserve deposits are also likely to decline. On the other hand, the decline in cash use could be recovered to some extent by imposing a reserve deposit requirement on e-money.
Under the current system, e-money issuers are not required to maintain accounts and reserve deposits at the central bank. However, under the Prepaid Voucher Regulation Law, an e-money issuer is obliged to (a) indicate on the prepaid voucher the amount of the voucher, the period of use, and other information, (b) deposit at least one-half of the unused balance of the advance payment accepted by the user,9and (c) prepare and preserve books and documents relating to the business of issuing prepaid vouchers. In addition, e-money is not included in the aggregate currency volume.10
Using the theory of currency multipliers, M1, the monetary base (MB), and required reserves (R) can be expressed as follows.
$$ M1=C+D $$
(4.11)
$$ MB = R + C + E $$
(4.12)
$$ R = r_D D + r_{EM} EM $$
(4.13)
where C is cash, D is deposits, EM is electronic money, E is excess reserves, rD is the deposit reserve ratio, and rEM is the reserve ratio for electronic money.
Using the currency multiplier m, the following relationship can be derived.
$$ M1 = mMB $$
(4.14)
Find the currency multiplier from Eqs. (4.11)–(4.14).
$$ \frac{M1}{{MB}} = \frac{C + D}{{r_D D + r_{EM} EM + C + E}} = m $$
(4.15)
Let us now see how the currency multiplier m varies with the e-money reserve ratio \(r_{EM}\) and the e-money EM.
$$ \frac{\partial m}{{\partial r_{EM} }} = - \frac{C + D}{{(r_D D + r_{EM} EM + C + E)^2 }} \cdot EM < 0 $$
(4.16)
$$ \frac{\partial m}{{\partial EM}} = - \frac{C + D}{{(r_D D + r_{EM} EM + C + E)^2 }} \cdot r_{EM} < 0 $$
(4.17)
In other words, we find that the currency multiplier decreases with the increase in the reserve ratio and e-money. However, its magnitude also depends on the deposit reserve ratio \(r_D\), excess reserve E, cash C, and deposits D.
Next, the setting of the reserve ratio for electronic money becomes important, but how should the cash reserve ratio for electronic money be determined? The answer to this question can be derived from the relationship between the volume of currency issuance and the interbank interest rate used in Iwamura (1996, p. 165). Let a be the rate of return on assets of the e-money issuer and rEM be the cash reserve ratio associated with the issuance of e-money. In this case, the revenue π earned on one unit of e-money can be expressed as follows
$$ \pi = (1 - r_{EM} )a $$
(4.18)
The cost of lost interest rate opportunity by keeping cash in reserve can be expressed as follows if the interbank interest rate is i
$$ {\text{Cos}} t = r_{EM} i $$
(4.19)
E-money issuers are expected to issue until the revenue from issuing e-money equals the cost of lost interest rate opportunities (we do not consider here the case of imperfect market competition among e-money issuers).
$$ {\text{Cos}} t = \pi \Rightarrow i = \frac{{1 - r_{EM} }}{{r_{EM} }}a $$
(4.20)
If we consider that the general equilibrium is working and the rate of return on assets a and the interbank interest rate i are equal, r EM = 0.5 (50%) from (4.20).11 Also, if i = 0, as in the case of the Japanese interbank interest rate after 2001, r EM = 1 (100%) is required regardless of a.
To ensure the central bank's controllability over the currency, the following measures could be taken: (1) the central bank could issue electronic money and include it in the currency aggregate; (2) the issuer of the electronic money could enter the framework of the central bank's reserve deposit system and accumulate somewhat higher reserve deposits; and (3) the excess liquidity created by the electronic money could be recovered by selling central bank assets.
At present, in recognition of the convenience of e-money and in order to take advantage of the results of research and development conducted by the private sector, no unnecessary regulations have been imposed, and the spread of e-money is being monitored. However, if e-money were to significantly reduce the currency multiplier effect, the measures described above might actually come into effect.
In terms of the actual scale of payments, cash payments of less than 3,000 yen are said to be 60 trillion yen per year, and even if only 10% of that amount, or 6 trillion yen, is shifted to electronic money payments, the scale of the retail business will be substantial.
However, this is only 0.27% of the 2,192 trillion yen that was settled annually by the Zengin system in 2003. Small-amount settlements using electronic money are so small as to be insignificant in terms of monetary policy, and the current judgment is that they do not have a major impact on the way monetary policy is conducted.
4.6 Conclusion
In this chapter, we have attempted to see what kind of analysis can be done from the standpoint of financial economics on the actual state of e-money distribution in recent years. The main conclusion is that e-money can be divided into e-money, cash, and credit cards in the order of transaction costs in Sect. 4.1. It is found that the boundary is determined by the technical breakage cost of e-money and the investment cost of payment and communication equipment for credit cards. With respect to the boundary of use, we show that the actual boundary is under-utilized for e-money and credit cards, and over-utilized for cash, compared to the socially optimal boundary. In the future, if technological progress is made and the cost of payment and communication facilities is significantly reduced, the use range of electronic money and credit cards will expand, and the use range of cash will shrink.
Furthermore, Sect. 4.2 implies that the use of e-money depends not only on the comparison of transaction costs, but also on the number of consumers per shop. In other words, the use of e-money is not only facilitated by the physical transaction costs, but also by the number of consumers who use it. This is an important point in the e-money business, where the emphasis is often on the technical aspects.
Finally, the impact on monetary policy is currently negligible in scale, and it is unlikely that e-money will be discussed as a subject of monetary policy or financial administration in the foreseeable future. However, it should be pointed out that if the amount of e-money issued expands and the currency multiplier declines significantly, the central bank may need to issue e-money or require e-money issuers to join a reserve deposit system and impose a high deposit reserve ratio.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits any noncommercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if you modified the licensed material. You do not have permission under this license to share adapted material derived from this chapter or parts of it.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
The number of Edy cards issued doubled from 5.1 million by the end of July 2004 to 10.2 million by the end of April 2005, nine months later. Judging from the pattern of the spread of goods in the past, this may be regarded as the beginning of a full-scale diffusion period.
JR East plans to offer Suica as a business with high future potential, using the technology for services such as building access control and personal authentication for PCs, and is not simply incorporating electronic money into its existing network business.
Although consumers may also be charged an annual fee for using the card, we ignore it here in comparison to the amount charged to merchants. And in fact, there are many cases where credit cards with no annual fee or electronic money are incorporated into credit cards so that no annual fee is charged for electronic money.
If credit cards were in a competitive equilibrium, if the cost of credit card authorization could be covered by fee income from merchants, the rest of the annual fee for card use and the pay-as-you-go portion of the handling volume would be waived.
Only when the unused balance as of the reference date (the last day of March or September of each year) exceeds the deposit standard amount (10 million yen).
An argument for including e-money in M1 is made in Berentsen (1997), but since M1 does not currently include e-money, we only consider the case where it is not included.
