Speech Reform of the Payments System

It is a pleasure to be here today to talk about the Reserve Bank's role in the payments system. As many of you will be aware, the Reserve Bank's Payments System Board met a couple of weeks ago and a number of decisions taken at that meeting were announced last Thursday. The timing of this conference is therefore ideal and I thank Visa for providing this opportunity to explain the Bank's recent decisions and to discuss the reform process more generally.

In my remarks this morning I would like to address four topics. The first of these is the Reserve Bank's general approach to the reform process. The second is the reforms to the credit card system. The third is the proposed standards for the EFTPOS and Visa Debit systems. And the fourth is the Bank's response to the arrangements between American Express and Diners Club on the one hand, and a number of banks on the other.

1.  The Reform Process

The Reserve Bank's payments system powers are exercised by the Payments System Board which typically meets quarterly, but more often if required. The Board is charged, amongst other things, with promoting efficiency and competition in the payments system and controlling risk in the financial system. The Board was set up by Parliament following recommendations by the Wallis Inquiry into the Financial System and came into existence on 1 July 1998.

Given the Board's mandate we have, not surprisingly, spent a lot of time considering efficiency and competition in the Australian payments system. As a result, I think it is fair to say that the Australian payments system has been subject to more scrutiny and analysis than almost any other payments system in the world.

In undertaking our analysis, we have used a number of benchmarks against which to judge both efficiency and competition. In essence, these benchmarks boil down to the following proposition: an efficient and competitive payments system is one in which five conditions are met, namely that:

  1. relative prices reflect the relative resource costs;
  2. merchants are free to choose the price they charge for accepting payment instruments and are free to choose which instruments they accept;
  3. prices are transparent;
  4. restrictions on access are limited to those strictly necessary for the safe operation of the system; and
  5. there is competition within and between individual payment systems.

While these conditions have formed the starting point for much of the Bank's analysis, we recognise that they need not hold in all cases, and at all times. The fact that prices do not reflect costs, for example, does not always mean that the system is inefficient or that competition is not working. Similarly, in some situations, restrictions on behaviour can actually promote, not detract from, efficiency. The Bank's general approach, however, has been to identify arrangements where these conditions are not met and then to place those arrangements under its analytical microscope to see if they are indeed in the public interest. In some cases, it has found that they are, but too often, it has found that they are not.

Throughout this process we have not been motivated by some ideological commitment to a textbook definition of efficiency. Neither have we had any preconceived notions that some payment systems are intrinsically ‘good’, and others are intrinsically ‘bad’. Rather, we have considered each system on its merits and attempted to assess, at a very practical level, whether the arrangements in that system promote efficiency and competition in the payments system as a whole. We have also adopted an evolutionary rather than revolutionary approach to reform, being mindful of the potential for large changes in the regulatory environment over a short period of time to significantly disrupt long-standing business arrangements.

In conducting our work, an important consideration has been the observation that, in many cases, retail payment systems are highly substitutable with one another. When you go to the supermarket, or pay a phone bill, you are faced with an array of payment choices. There are a variety of factors that influence the payment method chosen, but one important factor is price. Most of us have an instinctive feel for which payment method offers us the best deal, although we might not do the type of fine-pencilled calculations that economists sometimes assume that we do.

The fact that payment instruments are often substitutes has meant that the Bank has not just considered individual payment methods on a stand-alone basis, but, importantly, has also considered the payments system as a whole. After all, a change in arrangements applying to one method of payment clearly has effects on other methods, and these need to be taken into account in considering any proposed reform. Reflecting this, in our public documents you will typically see numerous references to the overall payments system, in addition to references to the particular method of payment being discussed. It is simply not possible to analyse the individual methods of payment that make up the overall payments system in isolation; they need to be analysed as a package.

We have also taken a medium-term focus to our work. We have been concerned not just with the way things currently operate, but also with how they are likely to operate in the future given the incentives facing cardholders, merchants and financial institutions. It would be incorrect to assume that just because something operates in a particular way today, that it will operate that way tomorrow. Experience in Australia and overseas suggests while the relative use of various payment methods does not change abruptly from one day to the next, payments patterns can change dramatically over a number of years in response to prices and technology. We obviously need to keep this in mind.

The final general point that I would like to make is that the reform process has been slower than many people would have hoped for, including ourselves. This reflects a number of factors. The first is that when we started this process, relatively little was known about interchange fees, nor had much been written about the economics of these fees and the nature of competition in the payments system. So there was a learning period for all involved. We have conducted a lot of analysis, read many submissions, and held numerous meetings in order that we understand the forces at play. Even so, we sometimes hear arguments that the Bank should do yet another study on this, or that, before any further action is taken. Where there are good grounds to do so, and reasonable prospects of success, we are keen to extend our analysis, but we are also mindful of the practical limits of what can reasonably be done.

The second reason for the relatively drawn out process is that when the Government gave the Reserve Bank its payments system powers it envisaged a co-regulatory approach, with the Bank only acting when the scope for voluntary reform had effectively been exhausted. As a result, we have spent considerable time attempting to achieve industry consensus on the need for reform. In many cases this has proved frustratingly difficult, particularly given that any individual reform has both potential winners and losers. Not surprisingly perhaps, those who see themselves as losers from a particular reform have not always embraced the process, making voluntary change very difficult in some cases. A third, and related factor, is that the reform process has been subject to legal challenges. This inevitably has also slowed things down.

This need to first seek voluntary industry reform and the legal challenges have led reforms in the various card-based payment systems to become decoupled from one another. This was certainly not our intention. We had hoped that the various reforms would become effective concurrently, given that the reforms are very much a package. But, given the constraints, this has simply proved impossible. We hope that the proposed measures announced last week, together with the timetable we have set out over coming years, will lead to a more co-ordinated outcome in the future.

2. Credit Card Reforms

So much for the generalities. I would now like to turn to the specifics. The first of these is the credit card reforms.

You might recall that one of the factors that initially ignited our interest in the credit card system was the observation that from a cardholder's perspective, credit card transactions are typically priced much more attractively than EFTPOS transactions. This appeared to be somewhat paradoxical, given that the EFTPOS system has substantially lower costs of operation than the credit card system.

In most other markets, competition ensures that if a good can be produced more cheaply than another, that good will be offered to consumers at the lower price. However, as we have discovered, this need not hold in the payments system. Too often the low-cost system seems to be offered to consumers at a higher price than the high-cost system. This has not just been the case in Australia, but around the world as well.

In the case of credit cards, many people face per-transaction prices that are effectively negative; that is they get paid to use the card! At current interest rates, the interest-free credit on many cards is worth a little more than half a per cent of the amount spent, and rewards are typically worth at least a further 0.6 per cent. In contrast, users of the EFTPOS system face, at best, a zero per-transaction price, and sometimes a positive price – typically in the 25 to 60 cents range.

A major reason for the lower-cost system being offered to cardholders at the higher price is the existence of interchange fees. Prior to the reforms, when I spent $100 on my credit card, my bank received, on average, a payment from the merchant's bank of around $1. In contrast, if I'd spent that same $100 on my EFTPOS card, my bank would have had to pay the merchant's bank around 20 cents. Not surprisingly, financial institutions have promoted the product where they received $1 rather than the product where they paid 20 cents, even after taking account of the difference in their costs.

This divergence in prices and costs clearly fell foul of the first condition that I discussed earlier: that relative prices reflect relative costs. After a long process of analysis and consultation we formed the view that this divergence was not in the public interest and set about reform of both credit cards and EFTPOS. For the variety of reasons touched on above, the EFTPOS process has turned out to be slower than that for credit cards.

These credit card reforms have cut the average interchange fee by around 40 basis points. So now when I spend $100 on my credit card, my bank gets, on average, around 55 cents, rather than around $1. This fall in interchange fees was quickly passed through to merchant service fees. The average fee is now around 1 per cent, compared with nearly 1½ per cent in early 2003 and around 1¾ per cent in the late 1990s.

At the time of the credit card reforms, some people were sceptical that the lower interchange fees would flow into lower merchant service fees. But they did, confirming that in a competitive environment when banks are faced with lower costs, they will reflect those lower costs in the prices they charge. We estimate that the reduction in merchant service fees saved merchants and their customers around $500 million in 2004. And we expect that over time, the annual savings will grow further.

A second area where the credit card system ran up against the conditions I outlined earlier was the restriction that prevented merchants from charging for credit card transactions – the so called no-surcharge rule. This restriction stymied normal competitive forces, preventing merchants from passing onto consumers their higher costs associated with accepting credit cards.

Since the removal of the rule we have seen charging in a wide range of industries, although the vast majority of merchants still do not charge. We often hear about the high-profile examples of Qantas and Telstra, but charging has also become quite common amongst small firms operating in very competitive low-margin businesses – computer shops, removalists and carpet layers for example. In some cases the ability to levy a charge has actually led merchants to introduce credit cards as a payment option. Previously these cards were simply too expensive, so they were not offered. But once merchants were able to pass the costs onto users of credit cards they were able to provide their customers with greater choice.

Over time I think we will see more charging as community acceptance of the practice increases. For more than 20 years the credit card schemes prevented the practice and most of us thought that this was the normal order of things. Not surprisingly, given this experience, many people have resisted the idea of charging. But attitudes are changing, and this is a welcome development.

A third area where the credit card schemes fell foul of the conditions I mentioned earlier is the transparency of pricing. Prior to the Reserve Bank's investigation, the level of interchange fees and merchant service fees were sometimes seen as close to state secrets. Now the level of interchange fees and the average level of merchant service fees are both published. This has helped improved understanding of the credit card market and given merchants better information when negotiating with their banks.

The credit card schemes also fell foul of the fourth condition – namely not having unnecessary restrictions on access. Here the Bank has opened things up by establishing a more liberal access regime. We are already seeing some positive effects here with the Australian Prudential Regulation Authority (APRA) having licensed one new Specialised Credit Card Institution, and a number of other organisations – both on the issuing and acquiring side – currently having discussions with APRA.

One access issue that has arisen recently is that successful entry as an acquirer in the credit card system is linked to successful entry as an acquirer in the EFTPOS system, where access arrangements remain overly restrictive. As you may know, the Australian Payments Clearing Association (APCA) has been drafting a new access regime. While the current proposal is an improvement on the existing arrangements, the Bank has recently written to APCA expressing some concerns. These relate to whether the regime provides sufficient certainty on the cost and timing of entry and whether volume requirements are necessary for new entrants. If these concerns are not addressed in a timely manner the Payments System Board will consider issuing a draft access regime for public comment as the first step towards putting in place more appropriate arrangements.

These various reforms have not led to the death of credit cards as some had predicted. Growth in spending on credit cards has fallen from the heady pace of the late 1990s, but remains robust at just above 10 per cent over the past year. Despite the cut in interchange fees to an average around 0.55 per cent, issuers are still able to offer interest-free credit and rewards whose value averages around 0.6–0.7 per cent. Issuers are able to do this partly due to the revenue they earn from annual fees and the interest earnings from those credit card users who do not pay their bill by the due date. For many people, credit cards remain one of the very few services that they are actually paid to use!

One claim that we sometimes hear is that cardholders have not benefited from the reforms: annual fees have been increased; reward points have been cut; charges have been imposed; and merchants have not passed on their lower costs in the form of lower prices.

This is not an assessment with which I would agree.

We are confident that the merchants' lower costs are flowing through into lower prices, given the competitive environment in which most merchants operate. Our estimate is that the cost savings will, over time, mean that the Consumer Price Index will be 0.1 to 0.2 percentage points lower than would have otherwise been the case. This may sound small, but it represents a significant benefit to consumers.

It is, of course, true that the credit card reforms have not affected everyone equally. Those that benefited most from the previous arrangements were those who used credit cards heavily and paid off the balance before the due date. More often than not these people tended to be those on high incomes. The reforms have undoubtedly made credit cards relatively less attractive for these people.

It is easy to forget, however, that the benefit that these people were receiving, and are still receiving, is paid for by someone else; in particular, those who predominantly use cash or debit cards to make their payments. These people were effectively paying higher prices for their goods and services than would otherwise have been the case, to pay for the subsidies to credit card users. Not only is this inefficient, but it means that people on lower incomes were often effectively subsidising those on higher incomes.

There is one final issue on credit cards that I would like to address before I move on to debit cards. And that is that under the current credit card standard there is a small difference in the interchange fees in the three schemes. For example, for electronic transactions, Visa has a fee of around 48 basis points and MasterCard a fee of around 51 basis points. This difference partly reflects differences in authorisation and processing costs across issuers as well as differences in the average cost of the interest-free period due to issuers having different portfolios of cardholders.

The issue here is whether a scheme with a higher interchange fee has a competitive advantage in attracting new issuers by virtue of the higher fee. If this were so, it would be another example where the scheme with the higher price gains at the expense of schemes with lower prices.

Last week the Bank announced that it was seeking submissions as to whether the same benchmark interchange fee should apply in all credit card schemes. This could be done in a number of ways, including taking an average of the costs across schemes, or using the cost data from the lowest-cost scheme. No doubt there are other possibilities as well. The Bank is seeking views on this issue by 8 April 2005, including views on the appropriate timing of any changes.

3. EFTPOS and Visa Debit Proposed Standards

I would now like to spend a few minutes talking about the Consultation Document released last week which included draft standards for the Visa Debit and EFTPOS systems.

Before I address the detail I would like to draw your attention to two points in particular. The first is that the draft standards are exactly that – drafts. We are seeking submissions from interested parties on these standards as well as the Bank's reasons for proposing these standards as set out in the Document. Submissions are due by 29 April 2005 and will be placed on our website to promote as open a process as possible. All those making submissions will be offered the opportunity to discuss their submissions in person with Bank staff.

The second point that I would like to make is that no final decisions will be made by the Payments System Board regarding the draft standards until after the outcome of the current court case, initiated by a group of merchants challenging the designation of the EFTPOS system, is known. Furthermore, if warranted, we will have a further round of consultation after the court case.

Now to the standards themselves. The interchange standards for EFTPOS and Visa Debit have the same basic format as that used for the credit card standard – namely they nominate eligible costs that are used to calculate a benchmark and the average of interchange fees must be at, or below, this benchmark. In the EFTPOS system the eligible costs consist of the switching and processing costs of acquirers, while for the Visa Debit system they consist of the authorisation and processing costs of issuers. Further, we are proposing that interchange fees in both systems be flat fees.

At current interchange fees, there is a difference of around 60 cents in the income to the issuer in the Visa Debit and EFTPOS systems whenever a cardholder makes an average transaction of around $80. Based on information available to us we estimate that, if implemented, the proposed standards would see a benchmark interchange fee of around 5 cents in the EFTPOS system paid to the acquirer, and a benchmark fee of around 15 cents paid to the issuer in the Visa Debit system. The effect of the standards would be to reduce the difference in the average interchange fees in the two systems from the current 60 cents to around 20 cents. We envisage that the same interchange fee would apply to any other scheme-based debit card introduced in Australia.

These proposed changes are likely to lead, in time, to more attractive pricing of EFTPOS to cardholders and a greater willingness of financial institutions to promote the EFTPOS system. If the current arrangements were to continue there would be a considerable risk that, over time, financial institutions would find it in their interests to steer customers away from EFTPOS to either credit cards or other types of debit cards. The end result would be an increase in merchants' overall payment costs. While merchants will likely face lower costs for accepting EFTPOS in the absence of reform, this is likely to be more than offset by cardholders switching to other forms of payment that are more expensive for merchants. The end result would not only be higher merchant costs, but also a higher overall price level for goods and services in Australia. Such an outcome would not be in the public interest.

It is important to stress that these reforms are not motivated by a desire to encourage the use of the EFTPOS system per se. Rather, our view all through this process has been that the Australian payments system will operate more efficiently if the various payment methods compete with one another on their own merits, rather than on the basis of interchange fees that are subject to limited competition. At the moment, payment systems do not compete solely on their merits. In particular, the nature of the interchange fees mean that the EFTPOS system is at a distinct disadvantage!

I might also note that we previously announced that we would review the credit card standard in 2007. As part of that review we will now also look at any debit card standards in place with a view to seeing whether the various standards should be put on a more consistent basis. As you are no doubt aware, the credit card standard and the proposed debit card standards are based on different eligible costs. This reflects, in large part, the Bank's measured approach to payments system reform, and the fact that the starting points were so far apart. Through the reform process we have sought to move things in the right direction, mindful of the adjustment costs, rather than move in one step to a fully consistent approach across all payment systems. Notwithstanding this, we do see considerable merit in a consistent approach being adopted through time. One such approach would be for a zero interchange fee to apply in all systems except where an interchange fee was required by the ‘balancing’ or other arguments discussed in the Consultation Document. There are no doubt other approaches as well. As is our usual practice, we will consult widely as part of this planned review.

The third proposed standard released last week would remove the current requirement that merchants accepting Visa credit cards also accept Visa Debit cards. In our opinion this requirement limits normal competitive forces. If merchants accept Visa credit cards they are unable to decline acceptance of Visa Debit cards even if they judge it in their business interests to do so. In addition, merchants are required to pay the same fee for Visa Debit transactions as they do for Visa credit card transactions, despite the Visa Debit product not offering an interest-free period. The proposed standard effectively removes these restrictions on merchants. It also formally applies the no-surcharge standard to the Visa Debit scheme.

The experience in the United States shows what can happen if the acceptance of scheme-based debit cards is tied to the acceptance of credit cards, and if the issuers of these cards are paid large interchange fees. In the United States, use of scheme-based debit cards, which are signature based, and thus prone to fraud, have grown much more quickly than the alternative, and more secure, PIN-based system. A primary reason for this is that issuers receive more revenue when their cardholders use the scheme-based card and the merchants are forced to accept the card as a condition of accepting credit cards. In the absence of reform, the same could well happen in Australia.

4. American Express and Diners Club

The fourth topic is our response to calls to regulate American Express and Diners Club, and in particular to regulate the payments between these institutions and their partner banks.

As you may be aware, last week the Bank made two announcements regarding these arrangements. The first was that it did not see a strong case for regulating the payments between American Express and Diners Club and their bank partners and that it would not be designating these schemes at this time. The second was that American Express and Diners Club had both agreed to the Bank's request that they remove clauses in their merchant contracts that effectively prevent merchants from expressing a preference for one type of card over another. American Express and Diners Club also agreed to the publication of their average merchant service fees, something that they had not done in the past.

In understanding the rationale for these decisions, it is useful to start by thinking about the existing four-party schemes operated by Bankcard, MasterCard, and Visa. In those schemes, the interchange fees are centrally set and are not subject to the normal forces of competition. In addition, the direction of causation clearly runs from interchange fees to merchant service fees. This was demonstrated when merchant service fees fell by around the same amount as the fall in interchange fees following the credit card reforms. The main reason for this is that there is significant competition amongst financial institutions for the acquiring of Bankcard, MasterCard and Visa transactions. The lower interchange fees meant lower costs for acquirers, and competition amongst acquirers meant that these lower costs flowed through to lower charges for merchants. As I noted earlier, in this case competition clearly worked!

While the arrangements that American Express has with a couple of banks look similar to those in the traditional four-party schemes, there are some important differences. One is that the size and nature of the payments to the partner banks are determined bilaterally through negotiation. Another, perhaps more important difference is that in these arrangements the direction of causation runs from merchant service fees to interchange fees, not the other way around. The reason for this is that American Express is the sole acquirer for transactions on its cards. Unlike the traditional four-party schemes there is no competition between acquirers for transactions using the American Express card. The same is true for Diners Club.

Given this lack of competition, our judgment is that any lowering of the fees paid to the bank partners through regulation would have little effect on merchant service fees charged by either American Express or Diners Club. To repeat, this is the opposite of the experience in the Bankcard, MasterCard and Visa schemes.

In assessing the case for regulation, we also considered the incentives facing bank issuers of American Express cards. In our view these incentives could only be changed through regulation, if the regulation was much more extensive than that currently applying to the Bankcard, MasterCard and Visa schemes. One possibility would be to regulate the totality of payments between American Express and its bank partners, including marketing and other related payments. On competitive neutrality grounds such regulation may have then also been required in the Bankcard, MasterCard and Visa schemes. In our view, there is not currently a strong case that such an extension in the reach of regulation is in the public interest.

Given these considerations, the Bank's opinion is that, at this stage, regulating the payments to the bank partners would do little to improve the overall efficiency of the payments system. Notwithstanding this, we had been concerned about restrictions imposed by American Express and Diners Club that prevented merchants from steering the cardholder to another form of payment. As I discussed at the outset, such restrictions on behaviour typically do not promote efficiency.

Both American Express and Diners Club have agreed to remove the relevant clauses from their standard merchant service agreements. Merchants will now have the option of asking a customer who offers a Diners Club or American Express card whether they would be happy to pay with another card that has a lower cost to the merchant. For many merchants this may be preferable to charging or not taking American Express and Diners Club cards at all.

The overall market will work better if merchants use this option where they feel that other cards are offering them better value for money. Just as we have said that we would like to see merchants charge for credit card use where they see it as in their interest, so too would we like to see merchants steering customers to cards that offer them better value for money. Again, to repeat a key theme of the reforms, the various payment systems should compete on their merits, not on the basis of interchange payments between financial institutions or by restrictions on merchants.


That brings me to the end of my list of four topics.

As you all know the reform process has been going on for a number of years and we are not yet at the end of the road. As you also know, that road has been a bumpy one, and I suspect that there are still more bumps to come. But at least, as we have travelled along that road together, the Bank and the payments industry have come to know and understand one another a little better. I hope that going forward we can build on that understanding to work co-operatively on reforms that are in the long-term public good.

Thank you.