The technological advances in the field of near-field contactless communications (NFC) and the development of sophisticated mobile applications have enabled mobile phones to become a potential means of payment. In 2009, more than two-thirds of the population worldwide was equipped with a mobile phone. In developed countries, where the market for mobile telephony is mature, mobile payments are seen as a potential source of revenue by mobile operators, who are trying to diversify their services. In developing countries, where mobile telephony is also widely adopted, mobile payments are seen by the governments as an opportunity to improve access to banking services for the unbanked population. Overall, according to a study from Arthur D. Little, so-called “mobile payments” will represent a transaction volume of $250 billion in 2012. The growing importance of remittances also builds a case for the development of mobile payments. According to the World Bank, the market for remittances totaled $433 billion in 2008. Remittances are money transfers made by migrant workers, who send a part of their revenue to their family or their acquaintances in their country of origin. They are mainly person-to-person money transfers, from developed countries to emerging countries, which can be processed through mobile networks or through traditional financial institutions such as Western Union.
However, though mobile payments have attracted a lot of attention, they have so far developed slowly, except in a few countries. For instance, the success of NTT DoCoMo, which launched a contactless payment solution in Japan, is often cited as prime example of the potential upheaval that mobile phones can create in the payments landscape of developed countries. The launch of the M-Pesa service by Vodafone in Kenya, which reached a total of 7.5 million subscribers in 2009, has also generated hopes to reduce financial exclusion in developing countries. However, a close analysis of these cases reveals that success stories cannot be easily generalized. The failure of the solution developed by Postfinance in Switzerland and the slow development of M-Pesa in Tanzania demonstrate that several ingredients are needed for mobile payments to succeed: standardization, incentives for consumers and merchants to adopt, and incentives for banks and mobile network operators to invest in the technology. In developed countries, the absence of these factors is often cited to account for the slow development of mobile payments.
In this paper, we examine the case of mobile payments in developed countries, as incentives to adopt mobile payment solutions seem to be fairly different in developing countries, and therefore we believe it would require a separate analysis. We propose to study the cooperation models for the development of mobile payment solutions. Do the players (banks, MNOs, nonbanks) have incentives to cooperate? What could happen if they decide to compete to provide mobile payment solutions? First, we start by defining mobile payments. Then, we analyze the potential path of development for mobile payment solutions. Afterwards, we study the various cooperation models between the players that are involved in mobile payments. Finally, we address the regulatory issues.
Definition of Mobile Payments
Mobile payments are generally defined as the process of two parties exchanging money using a mobile device in return for goods and services. This definition has the advantage of being simple enough to capture a wide array of technological possibilities. For instance, the mobile device can either be a mobile phone, a computer, a PDA, or even a wireless sticker that can be attached to any object, such as a ring or a key. The transaction can either be remote (SMS-based for instance) or processed locally via contactless technologies such as Near Field Communication or RFDI. However, this definition may be potentially confusing, as the mobile device is not necessarily a means of payment. For instance, Vodafone in the United Kingdom has developed a payment solution that uses mobile phones to initiate and authenticate card payments. The purchases are charged directly to the payment cards of the users who have preregistered to the service. Hence, a broad definition needs to include the cases in which the payment process involves a mobile device that is not automatically used as a means of payment. Another problem is that the usual definition does not take into account the money transfers that can be processed through mobile devices, without any exchange of goods or services, such as person-to-person money transfers and remittances.
Therefore, in this paper, we decide to focus on “mobile money transfers,” which we broadly define as transfers of money between two parties involving a mobile device. Mobile money transfers can be either remote, in-store, prepaid, or postpaid through reverse billing. In this paper, we also choose to focus on the case of mobile money transfers in developed countries.
The Potential Path of Development for Mobile Payment Solutions
An historical perspective on the development of payment card systems in developed countries can be useful to better understand the factors of success for the adoption of electronic payment systems. Payment cards were first introduced in the 1950s by closed platforms, such as Diners Club or individual banks, which managed to develop an important acceptation network for their customers. Afterwards, some banks decided to create interbank joint ventures, which enabled them to agree on common standards and increase their acceptation networks. The most important joint ventures, which later became Visa (1975) and MasterCard (1979), have a global reach and are now used and accepted in many different parts of the world. Recently, in 2006, these companies transformed their organizational structures to become publicly traded. The example of payment card systems shows that at least two ingredients are essential to the development of electronic payment systems for mass markets. First, the players must cooperate for the development of common standards, or specify the conditions for interoperability. Building a joint venture can considerably reduce the costs of incompatibility between different standards. However, as we will argue in the next section of the paper, the cooperation for the development of mobile payment services seems to be much more complicated than in the case of payment cards. Second, the players must develop an important acceptation network. Usually, the economic literature stresses the “two-sided” nature of retail payment systems. Retail payment systems are indeed characterized by membership and usage externalities between two distinct groups of users, the consumers and the merchants. The more consumers adopt a payment instrument, the more the merchants will be willing to accept it, and vice versa. By taking advantage of network effects, joint ventures can increase the probability of success of each service provider who tries to affiliate consumers and merchants.
However, one should note that the problems for the adoption of mobile payment services are very different from the problems that the players faced for the diffusion of payment cards in developed countries. Consumers and merchants in developed countries are already well equipped with payment cards, a widely-used and accepted electronic payment solution. Hence, if they were to replace the use of payment cards, mobile payment services would have to provide sufficient additional value to be adopted by consumers and merchants. For instance, Ondrus et al. (2009) explain the failure of the mobile payment solution developed by PostFinance in Switzerland by the lack of value added to the existing payment card solution. On the merchant side, for in-store payments, the incentives to adopt mobile payment solutions depend on the costs of upgrading the existing payment terminal, the level of security, and the potential additional benefits generated by the service, such as mobile couponing. For remote payments, such as e-commerce payments, merchants may find it valuable to add an additional payment option to their website, as cyber merchants who offer multiple payment instruments to their consumers tend to convert more visitors into customers. On the consumer side, the mobile phones can potentially have a competitive edge over payment cards, as they can be used as an interactive device, in which the consumer can store information. However, consumers are often used to being delivered payment instruments for free, either in a bundle with their bank account, or by merchants who try to increase user stickiness. Hence, if the service is to be provided by a nonbank, the provider will have to find a way of recouping its costs of investment in infrastructure and security, which is not necessarily eased by the presence of low cost services. Another option for mobile payment service providers would be to target niche markets, in which payment cards are absent. Such markets could include person-to-person money transfers, or payment services at a low cost or a low risk for the unbanked or “underbanked.” The reasons why some people do not use existing electronic payment instruments are varied; for instance, if a consumer values privacy, he will not necessarily adopt mobile payments more easily than payment cards.
Cooperation Models for the Development of Mobile Payment Solutions
In this section, we study different cooperation models between the key players that could be involved in mobile payment solutions. We can view a mobile payment solution as based on three key inputs: i) a mobile phone, ii) a bank account, and iii) an acceptation network. Each of these inputs is to some extent controlled by a key player. Mobile network operators (MNOs) and mobile phone manufacturers have control over the design and distribution of mobile phones, as the former commercialize the phones at subsidized prices in their commercial agencies and own the SIM card while the latter produce the phones. Banks have control over their consumers’ accounts. And, finally, payment systems like Visa or Mastercard have control over large acceptation networks.
Though these key players have some control over the three key inputs of a mobile phone solution, we argue that they can be bypassed. First, a solution can be developed without the cooperation of MNOs and mobile phone manufacturers, as the payment application can be resident on a separate SD card, for instance. Another example is the payment solution developed by the start up Square, which has been launched by the former CEO of Twitter. This solution is based on a plastic device that plugs into the mobile headphone jack, hence, it is completely independent of MNOs or manufacturers. Second, the mobile payment solution could be based on the payment card of the consumers, in which case the provider does not need the cooperation of the banks to have access to the consumers’ bank accounts. For instance, Obopay allows consumers to add money to their Obopay account with their debit or credit cards, and then send money to relatives or merchants with their mobile. Though Obopay proposes its service to banks, it has been developed without their cooperation. Third, a mobile payment service provider could develop a solution without a large acceptation network (like Visa or Mastercard) if it decides to target a niche market. For instance, the provider could limit the acceptation of its payment solution to vending machines (like Mobilkom A1 in Austria) or to a few affiliated merchants (like Obopay, which targets mainly P2P transfers but proposes merchants to affiliate to the system at no fee).
Since each of the three key players (banks, mobile network operators or mobile phone manufacturers, and large acceptation networks) could be bypassed or not for the development of a mobile payment solution, we have a priori six different possible combinations. The table below gives some examples for five combinations. The last combination corresponds to a situation where the mobile payment provider owns a bank or is a bank but bypasses the MNOs and the acceptation network. We consider that this combination is not relevant, as banks have strong incentives to develop a mass market solution, which requires cooperation with the acceptation network.
These different models involve different degrees and forms of cooperation. Benefits of cooperation are cost-sharing and taking advantage of complementarities between different players. Banks have gained a great deal of experience in operating mass-market payment systems, which might be critical for a wide adoption of a mobile payment solution. They also have experience in risk and fraud management that other players, like MNOs, do not have. In contrast, MNOs have strong partnerships with mobile phone manufacturers which might help to develop payment-enabled mobile handsets. Costs of cooperation are twofold. First, partners will have to share not only costs but also revenues. For instance, in case of a cooperation between banks and MNOs, if mobile payment adds little value relative to existing payment solutions (e.g., payments by card), this may impede the constitution of a partnership. Second, like in any joint venture, there are costs of coordination. As they have different objectives and interests for the development of mobile payments, we think that the costs of cooperation between banks and MNOs are probably high. The literature on research joint ventures (RJVs) indeed suggests that asymmetries between members of an RJV make the RJV less likely to succeed.
Finally, the incentives of the different players to develop mobile payments differ. If they develop a mobile payment solution, MNOs would be new entrants in the payment industry. In contrast, banks are incumbents in the payment market, and therefore would view mobile payments as an improvement over other payment solutions that they commercialize (like payment cards). Therefore, banks would face a “replacement effect” for the development of mobile payments. Hence, their incentives to develop mobile payments might be lower than the incentives of MNOs, except that they could have high preemption incentives to protect their market share from an entry threat.
Some Regulatory Issues
Finally, the cooperation models for the development of mobile payments could be impacted by the regulation of nonbank players. Bradford et al. identify several risks to the presence of nonbanks in payment systems: operational risk, settlement risk, legal risk, reputational risk, and systemic risk. The provision of mobile payment solutions could increase the presence of nonbanks in retail payment systems, and thus, could require a regulatory intervention to limit the potential risks involved for the economy. So far, in developed countries, regulators have often tolerated the use of mobile phones for small transactions without requiring any banking license. However, this situation could be called into question if the volume of mobile payments were to become important. In Europe, the payment service directive offers to the new entrants the possibility to adopt the status of “payment service provider,” which means that mobile payment service providers will be supervised by the relevant national regulatory authority. In France, for instance, the national supervisor (the CECEI) stated precisely that MNOs must apply for the right to enter the payments market, and it specified minimum capital requirements. If the MNOs were to issue electronic money, they would fall into the regulations of electronic money institutions. Some economic areas and countries such as Europe, Japan, and the Philippines have decided to design a specific status for e-money institutions. In other countries, there are still many regulatory loopholes that could slow down the adoption of mobile payments, either because the consumers may not trust nonbanks for payments, or because some companies may decide not to run the risk of investing in technologies that do not comply with the regulatory rules (compliance risk).
In this paper, we have argued that there are three key inputs for the development of mobile money transfers: bank accounts, mobile phones and large acceptation networks, with each of these inputs being somehow controlled by a key player (banks, MNOs or mobile phone manufacturers, and payment systems). A mobile money transfer solution could either involve cooperation with or bypass of these key players. Hence, we define five different cooperation models and we give some examples for each of these models.
However, the competition that could emerge within each model and between the various models remains to be seen. One could be tempted to argue that the light model is easier to implement to target niche markets, but that prospects for mass adoption of mobile payments are higher with the full integration model.
Finally, we focused on developed countries. However, many interesting questions are raised by the development of mobile payments in developing countries, which would deserve further attention. An interesting question would be to examine under which conditions mobile payments can really improve access to banking services in countries where a large share of the population is unbanked.