Reform of the EFTPOS and Visa Debit Systems in Australia:
Final Reforms and Regulation Impact Statement – April 2006
8. Evaluation

8.1 Parties affected

In assessing these various options, the Bank considered the likely impact on users and providers of payment services, including:

  • the institutions that are participants in the EFTPOS, scheme debit and credit card systems;
  • potential new participants in the payments system;
  • merchants that accept EFTPOS and scheme debit cards;
  • EFTPOS and scheme debit cardholders; and
  • the community as a whole.

Most banks, building societies and credit unions issue EFTPOS cards; in total there are around 25 million accounts that can be accessed by using an EFTPOS card. Visa Debit cards are issued mainly by building societies and credit unions and some banks, with around 4 million Visa Debit cards on issue. On the other side of the market, eight institutions account for almost all acquiring services.

The following discussion outlines the likely effects of the various options.

8.2 Option I: Implementation of Standards and Access Regime

Interchange Standards

Implementation of the two proposed interchange Standards would narrow the difference in interchange fees in the EFTPOS and Visa Debit systems noticeably. Reduction in the EFTPOS interchange fee would also narrow the difference in interchange fees in the EFTPOS and four-party credit card systems.

In the first instance, this realignment of interchange fees is highly likely to lead to merchants paying lower merchant service fees on Visa Debit transactions, and higher fees (or receiving lower or no rebates) on EFTPOS transactions. Despite the higher fees on EFTPOS transactions, merchants are unlikely to reduce their acceptance of EFTPOS as a means of payment. As discussed in Section 3, merchants are reluctant to refuse acceptance of widely used means of payment, and EFTPOS transactions will still attract lower fees than either Visa Debit or credit card transactions. The net effect (in the short run) is therefore likely to be higher overall merchant costs for accepting debit cards: at current transaction levels, and assuming that the change in interchange fees is fully passed through by acquirers, costs to merchants would rise by around $100 million per year.

This outcome, however, assumes that payment patterns are not affected by the reforms. In the Bank's view this assumption is unrealistic. Over time, the Bank expects that overall merchant costs of accepting card-based payments will be lower under the reforms than would have otherwise been the case. This reflects the fact that payment patterns by cardholders are likely to be different as a result of the reforms.

In particular, in the absence of a change in current interchange arrangements, issuers will have a strong incentive to promote scheme debit at the expense of EFTPOS, both through pricing to cardholders and other means of promotion. This reflects the fact that, under current arrangements, issuers of Visa Debit cards receive around 40 cents in interchange revenue on the average Visa Debit transaction, whereas if the same transaction is made through the EFTPOS system, they have to pay 20 cents. As a result of these issuer incentives, the Bank expects that scheme debit would grow at the expense of the EFTPOS system. Currently, those issuers who issue both Visa Debit and EFTPOS cards actively promote the use of Visa Debit over EFTPOS. A further consideration is that, at some point in the future, banks will need to invest in technical upgrades to the EFTPOS system. If an alternative debit card system is available which yields significantly more revenue to issuers, this upgrading may not occur.

To date, most issuers have not offered Visa Debit cards, despite the clear advantage (in terms of interchange fees) in doing so. In the Bank's judgement, this situation is likely to change if it were clear that current arrangements were to continue. While it is difficult to determine how extensive and rapid any switch away from EFTPOS to scheme debit might be, the possible effects on merchant service fees can be illustrated by making some assumptions. In particular, if it is assumed that scheme debit transactions were to grow to 50 per cent of the 1.2 billion debit card transactions reported in 2005, rather than the current share of around 15 per cent, merchants' costs would be around $250 million per annum higher than is currently the case. If instead, all debit transactions shifted to scheme debit (which, over time, is not an improbable outcome), the increase in merchants' costs would be over $600 million per annum (at current total debit card transaction levels). In time, these higher costs would flow into higher general prices of goods and services, which would affect the community at large.

The reforms will have a number of effects on consumers. In the short term, any increase in merchants' costs is likely to flow through into higher prices for goods and services paid by all consumers. However, over the longer term, as payments behavior adjusts, consumers will face lower prices than they would have in the absence of the reforms. For EFTPOS cardholders, the reforms are likely to see more attractive pricing than would otherwise have been the case. A number of financial institutions have already moved to zero charges for all or most electronic payments, including EFTPOS. Lower interchange fees in the EFTPOS system make it more likely that such pricing arrangements will continue. For other institutions, fee-free limits may be increased for EFTPOS transactions and/or lower per transaction charges may apply. Conversely, the reduction in interchange fees for Visa Debit transactions may mean that Visa Debit cardholders experience less attractive pricing, either in the form of account-keeping fees, or transactions charges.

The Bank also expects that the reforms will lead to some substitution away from credit cards and towards the EFTPOS system as a result of the change in pricing to cardholders. There may also be some substitution away from Visa Debit to credit cards (see Section 9).

The changes in interchange fees, and the subsequent changes in payment patterns, are likely to lead to lower overall resource costs than would otherwise have been the case. The most comprehensive assessment of the costs of making transactions through the different systems remains the Joint Study. While strict comparisons of the different systems are difficult due to the allocation of fixed costs, the available evidence is that the cost of processing a typical transaction through the EFTPOS system is lower than through the credit card and scheme debit card systems. This reflects a variety of factors, including lower fraud and processing costs. As above, the possible effects on resource costs of implementing the reforms can be illustrated by making some assumptions. In particular, if it is again assumed that scheme debit transactions would otherwise grow to 50 per cent of the current debit card transactions and that resource costs involved in an EFTPOS transaction were 20 cents lower than through the other systems, total resource costs would be $85 million lower under the reforms than otherwise. If the difference in resource costs were higher, say 30 cents per transaction, and scheme debit transactions would otherwise account for all debit transactions, the impact on overall resource costs would be around $300 million.

As part of its assessment, the Bank recognised that the extra resource costs in the Visa Debit system are partly explained by the extra functionality to cardholders of the Visa Debit card. However, it did not see this as an argument for maintaining the current configuration of interchange fees. If cardholders value this extra functionality (and the value they place on it is greater than the cost of providing it), then they should be prepared to pay more for the use of Visa Debit than they are prepared to pay for use of the EFTPOS system.

EFTPOS Access Regime

Implementation of the proposed Access Regime would place a cap on the price paid by a new entrant to establish a direct connection and limit the scope for existing participants in the EFTPOS system to use negotiations over interchange fees in a way that could act as a barrier to entry to the system.

These proposed reforms, together with the EFTPOS Access Code, would benefit potential participants in the EFTPOS system by making it less expensive and less difficult to enter. New participants are likely to be mostly on the acquiring side, with the result that competition among acquirers would increase with merchants being offered cheaper acquiring services. To the extent that acquirers need to be able to provide debit card acquiring services if they are to provide credit card acquiring services, improving access to the EFTPOS system is also likely to improve competition in credit card acquiring.

Some existing participants on both the acquiring and issuing sides of the EFTPOS market might also be expected to take advantage of the reforms to establish additional direct connections.

Liberalised access conditions would increase competition faced by existing participants in the EFTPOS system. The cap on the price that existing participants may charge for establishing a new direct connection would mean that existing participants in the EFTPOS system, other than the low - cost provider, would not recover their full costs of providing the connection. Over time, this may encourage participants to explore ways of lowering their costs of providing a connection.

The honour all cards Standard

Implementation of the Standard removing the honour all cards rule would allow merchants to make separate decisions to accept scheme debit cards and credit cards.

Merchants would be able to assess the benefits and costs of accepting scheme debit cards rather than being required to accept them as a consequence of deciding to accept scheme credit cards. This could be expected to result in a more efficient allocation of resources, which ultimately should benefit merchants' customers more broadly.

There is also a possibility that abolition of the rule may lead to lower interchange fees in the Visa Debit system. Both merchants and the schemes have an interest in cardholders being able to pay with a wide range of cards, including scheme debit cards. If merchants were to decide not to accept scheme debit cards, the schemes may decide to lower interchange fees as part of a process of ensuring continued acceptance. If the rule were not removed, this possibility would not arise.

If some merchants do not accept scheme debit cards, cardholders would bear some costs of having to identify which merchants accepted scheme debit cards. Removal of the rule may also mean that issuers and card schemes find it more difficult to launch new card products, since they would need to convince merchants that they should accept new products on their merits.

Implementation of this Standard would impose some costs on both issuers and acquirers. These costs include the costs of reissuing cards outside the normal timetable and the costs of any changes to systems that are required to ensure the cards are electronically identifiable. These costs are discussed in more detail below.

The proposed Standard abolishing the honour all cards rule also formally abolishes the no surcharge rule in the Visa Debit system. This would allow merchants to impose a surcharge on Visa Debit transactions if they so chose.

Compliance costs

The implementation of Option I (as described in Section 10 below) would impose some compliance costs on participants in the payments system. Using the nine categories of costs set out by the Department of Industry, Tourism and Resources for the purposes of their Business Costs Calculator, there are three types of compliance costs that might arise:

  • Record-keeping costs. These are the costs involved in collecting data to calculate the relevant interchange benchmarks and the access charge benchmark. There will be few additional data collection costs associated with the Visa Debit interchange Standard, since it primarily uses data already collected for the purposes of the credit card interchange Standard. In contrast, in the EFTPOS system, u p to six acquirers will incur costs of providing data on the cost of processing and switching transactions once every three years. These data are routinely available from institutions' cost accounting systems and the costs of extracting the data and providing them to the Reserve Bank are likely to be small. The relevant institutions have recently demonstrated their capacity to provide these data in a timely fashion. Similarly, Access Providers will, once every four years, incur costs of collecting data on the costs of providing direct connections. Again, the costs are likely to be small.
  • Notification costs. There are potentially two types of notification costs. First there is the cost of confirming compliance with the Standards to the Reserve Bank once a year. Second there is the cost of notifying merchants of their rights under the regulations. The Standard provides more than a year for this to occur, during which period acquirers would, in the normal course, have several communications with merchants. This requirement should therefore not require additional communication with merchants. In addition, acquirers will incur the costs of providing information to merchants to allow electronic identification of Visa Debit cards. These data are already available. Scheme participants will need to establish arrangements to make them available to merchants that request them.
  • Purchase costs. This is any costs of changes in the visual appearance of Visa Debit cards. These costs will be small. The Standard provides for cards to be issued on institutions' normal replacement cycles and therefore does not impose significant costs on financial institutions in addition to their usual costs of issuing cards.

While it is difficult to measure each of these various costs with accuracy, the Bank's judgement is that they are small and are outweighed significantly by the benefits of the reform package.

Most of these compliance costs would fall on acquirers, none of which are currently small businesses.

8.3 Option II: Implementation of elements of the current proposal

The second broad option considered by the Bank was to implement only parts of the proposed package.

(a) Impose Access Regime only

One possibility considered was to address only the issue of access to the EFTPOS system on the grounds that, typically, the starting point for ensuring strong competition in any market is to ensure that there are no unnecessary barriers to entry.

While this option would address the issues discussed above regarding entry, it would leave the important issue of relative interchange fees unaffected. Improving access would not cause a change in the current configuration of interchange fees, and thus the incentives discussed above would remain. The long-run costs to merchants of debit card payments would be higher than under Option I to the extent that a shift to scheme debit would be encouraged. Resource costs would also rise.

Compliance costs incurred by participants in the EFTPOS and Visa Debit systems would be lower than under Option I. The notification costs and purchase costs would not be incurred.

(b) Implement all reforms except the EFTPOS interchange Standard

A second possibility considered was to improve access to the EFTPOS system for new participants, reduce interchange fees in the scheme debit systems and prohibit the honour all cards rule, but leave the arrangements for determining interchange fees in the EFTPOS system unchanged. This possibility was canvassed by a number of merchants during the consultation process.

This option would narrow the difference in interchange fees in the EFTPOS and Visa Debit systems to around 35 cents on the average transaction, compared with 20 cents under Option I and the current difference of 60 cents. The effect would be to increase the incentive for issuers to promote the use of scheme debit over EFTPOS (compared with Option I) and thus increase the likely usage of scheme debit by cardholders. The long-run result is likely to be higher merchant costs and higher resource costs than if the full package of reforms were implemented.

This Option would nevertheless narrow the difference in interchange fees more than would Option II(a) and, from that perspective, is preferable to Option II(a).

Compliance costs would be less than for Option I, since EFTPOS acquirers would not need to provide cost data to the Reserve Bank.

(c) Implement all reforms except the honour all cards Standard

A third possibility considered was to implement the entire package with the exception of the requirement that the honour all cards rule be removed. This option was canvassed by Visa and Visa Debit issuers. They argued that, if Visa Debit interchange fees were reduced and merchants were able to surcharge for Visa Debit transactions, few merchants would elect not to accept Visa Debit cards, and any benefit from allowing them to do so would be small. Further, they argued that retaining the honour all cards rule would prevent customer confusion as to where various cards could be used.

Under this option, merchants would be unable to negotiate explicitly over the terms on which they accepted scheme debit cards. This could have a number of possible effects. First, it might reduce the pressure on acquirers to pass on the fall in scheme debit interchange fees to merchants, though the fall in interchange fees in the credit card systems suggests that competition between acquirers is likely to see a rapid pass through. Second, it might reduce the competitive pressure on the margin between the interchange fee and the merchant service fee charged by acquirers to merchants for scheme debit transactions. An inability of merchants to bargain specifically over this margin, against the background of the possibility that they would refuse to accept scheme debit cards, could reduce competitive pressure on acquirers of scheme debit transactions. And third, there would be no pressure for the schemes to reduce the interchange fee to ensure continued merchant acceptance.

The compliance costs of Option II(c) would be little different from those of Option I, since the implementation of the honour all cards Standard has been designed to minimise transition costs.

8.4 Option III: No regulatory action

The analysis of doing nothing is essentially the reverse of Option I.

Under this option the existing impediments to entry to the EFTPOS system would be likely to remain. New participants would need to continue to negotiate bilateral direct connections and would be likely to experience the difficulties described in Section 3.

Relativities between interchange fees in the EFTPOS and scheme debit systems would also remain, and thus so too would the incentives faced by issuers and cardholders favouring scheme debit over EFTPOS. In the Bank's view this is likely to lead to scheme debit growing more quickly than EFTPOS, and the possibility that the EFTPOS system might eventually cease to exist. As a result, merchants' costs would rise.

Merchants would continue to be required to accept scheme debit cards on the same terms as credit cards, with scheme debit continuing to have a competitive advantage over EFTPOS. Card schemes would be able to use this advantage in the introduction of new card products that would have a guaranteed merchant-acceptance base. Cardholders would continue to find their scheme debit cards accepted wherever the scheme's credit cards were accepted.

The main advantage of this option is that it would not impose any compliance burden on various parties in the payments system and would avoid the compliance costs discussed in Section 8.2.