Over the past three decades, there has been a series of famous anti-trust cases in the United States concerning the operation of credit card associations and their potential for anticompetitive behaviour. In Australia, while there was some early competition scrutiny of the Bankcard association, it was only with the 1997 Wallis Inquiry that the payment system in Australia was identified as a key area for policymakers. The Wallis report considered a potential role for the Australian Competition and Consumer Commission (ACCC) in determining the conditions of access to payment systems and also recommended that a Payment Systems Board be set up within the Reserve Bank of Australia (RBA). The Payment Systems (Regulation) Act gave the RBA broad powers to designate and regulate standards (including pricing terms) that arise in payment systems; and in 2001, the RBA chose to exercise those powers by first designating and then, in 2002, regulating the rules and practices under which the major credit card associations (Bankcard, MasterCard and Visa) operate.
The basis for competition concerns has been detailed in a series of reports by regulatory authorities (Cruickshank, 2000; European Commission, 2000; RBA/ACCC, 2000). In each case, investigations were triggered by natural suspicions that arise when otherwise competing banks cooperate through credit card associations. However, beyond simple allegations of price fixing and the abuse of market power, the RBA has outlined in detail its major concern: that the rules of card associations promote too much credit card use from a social perspective. The displacement of other forms of payment (including cash and debit cards), they argue, is raising the costs of transacting in the economy (RBA, 2002).
While their final policy statement focuses on a lack of competition as the ultimate cause of inefficiency in transacting, the RBA’s Consultation Document (RBA, 2001) is more specific. Three particular features of credit cards are seen to drive potential over-use. The first is the ‘no surcharge rule’ that prohibits merchants from passing through potentially higher costs of processing credit card transactions to those customers who want to use credit cards. Consequently, all of a merchant’s customers, including those who pay with cash, bear the costs of the credit card system. The second is the collectively set interchange fee. This fee is the payment made from a merchant’s bank (the acquirer) to a cardholder’s bank (the issuer) for settling a credit card transaction. Because the interchange fee is collectively set by the issuers and acquirers, and because merchants pass it on to all customers (and not just credit card users), it is argued that the banks can set a high interchange fee to capture more rents from merchants without a loss in credit card usage. Moreover, because the high interchange fee flows to card issuers who can encourage more card usage, those issuers set low (and perhaps negative) fees to cardholders.1 The end result is overuse
of credit cards.
The third aspect of card associations that concerns the RBA are “minimum entry standards that are intended to ensure the safety of the scheme, but have the effect of unduly limiting competition.” (RBA, 2002, p.7) In this case, however, the RBA does not articulate how a potential lack of competition may drive the over-use of credit cards. Indeed, it is usually presumed that any lack of competition may lead to higher cardholder and merchant fees, and reduce rather than promote credit card use.
The RBA’s final policy was directed at each of these three aspects of credit card associations. First, it proposed to eliminate association rules that prohibited merchants from imposing a surcharge on credit card users. Second, the RBA proposed a methodology that amounted to the direct regulation of the interchange fee. Finally, it proposed to enhance the ability of non-bank entities to sell credit card services, both to merchants, as acquirers and customers, as card issuers. The RBA regarded each of these policies as necessary to improve the efficiency of credit card operations in Australia. In the absence of such regulation, it was the RBA’s view that there was insufficient competitive pressure on the major banks that dominated card associations to generate levels of credit card usage that were socially efficient.
The purpose of this paper is to develop a simple model of payment systems designed to analyse the impact of the RBA reforms. Previous models of payment systems, while providing insight into potential inefficiencies of credit card associations, have not been directed solely at the impact of direct regulation; particularly as it will be done in Australia. For example, Rochet and Tirole (2002) and Schmalensee (2002) demonstrate how associations might set interchange fees to balance the incentives of issuing and acquiring banks to promote the use of credit cards. In each model, there are pressures towards both under- and over-use of credit cards. Under-use can arise as a result of imperfect competition in the issuing or acquiring segments, but this is counterbalanced by the use of the interchange fee to promote adoption by either cardholders or merchants. However, each of these models has as a central feature the notion that network effects constrain the adoption of a particular payment instrument and that the practices of card associations provide a socially desirable function in resolving coordination issues.
The RBA, however, is explicit in its rejection of network effects as being important in today’s environment, where there is wide-spread adoption of credit cards (RBA/ACCC, 2000). Certainly, it is possible that network effects may no longer be a constraint on adoption and hence, something with significant welfare consequences. For this reason, our model differs from all others in the literature by modeling a credit card association in an environment with no network effects. Moreover, as a welfare standard for analysing the consequences of the RBA’s policies, we consider its goal of
minimising the cost of transacting in the economy rather than some overall measure of producer and consumer surplus. As such, our analysis is conducted on the RBA’s terms.
Given this, the main theoretical contribution of this paper is to highlight the important role played by customers when retail prices are the same regardless of the payment instrument (i.e., cash or credit card) used. It is the customer who determines the choice of payment instrument for any specific transaction; a choice that may impact upon the payoffs and profits of other participants in a payment system. As we will demonstrate, this insight is sufficient to both give some weight to concerns about inefficiencies in credit card associations but also to isolate the key impacts of the policies that will be enacted in Australia.
The paper proceeds as follows. The next section sets up our model of a credit card association. Section 3 then considers the operation of an unregulated credit card association. We demonstrate that, even absent network effects and strategic reasons for merchant adoption, interchange fees might be set too high and that there might be over-use of credit cards. Importantly, however, we do not find that a lack of competition in the issuing and acquiring segments drives potential inefficiencies (in contrast to the RBA’s claim). Instead, it is a potential distortion in price signals arising out of the interaction between a privately set interchange fee and the no surcharge rule. Section 4 then considers the removal of restrictions on surcharging. We demonstrate that this is unlikely to generate price signals that will generate the socially efficient level of credit card use. Instead, merchants, who have some market power, will likely use the choice of payment instrument as a means of facilitating third-degree price discrimination. Consequently, there will too little card use. Section 5 then turns to consider the impact of regulating the interchange fee. There, we highlight the fact (explored more fully in Gans and King, 2003a) that once surcharging is possible, the level of the interchange fee has no real consequences and hence, regulation of it will be of no effect. However, if surcharging does not actually occur (say, for transaction cost reasons), we can provide a formula for the socially
efficient interchange fee. We note, however, that this fee differs markedly from that proposed by the RBA. A final section concludes.
Downlaod
PDF Ebook A Theoretical Analysis of Credit Card Reform in Australia
