What happened in Australia?

Joshua Gans —  8 December 2009

Joshua Gans is the foundation Professor of Management (Information Economics) at the Melbourne Business School, University of Melbourne.

What happens when you take a key price in an industry and cut it in half? For normal markets economists would expect that this would have a dramatic effect on quantity. That, however, was not the experience in Australia when the Reserve Bank of Australian (RBA) used new powers in 2003 to move Visa and MasterCard interchange fees from around 0.95 percent of the value of a transaction to just 0.5 percent. The evidence demonstrates that this change was virtually undetectable in any real variable to do with that industry.

To begin, let’s review the RBA reforms. First, in January 2003, it moved to eliminate the card association’s ‘no surcharge’ rule. Then in October 2003, the new interchange fee came into effect. As this submission outlined, that latter move had an immediate impact on merchants service charges with acquirers passing on the full extent of the interchange fee reduction to retailers. However, there was no impact on the value of credit card purchases, the level of credit card debt or the share of credit versus debit card transactions. Econometric analysis by Melbourne Business School’s Richard Hayes (also appended to that submission) confirmed this. Moreover, that analysis demonstrated that credit card usage did vary with other economic factors including underlying interest rates in the expected direction. Put simply, if we did not know the reforms were actually taking place, you would not be able to observe it in the data.

What was there no impact? There are a couple of possible explanations. First, it may be that the interchange fee was only one of a number of payments between acquirers and issuers and that, unobserved to analysts and the regulator, those payments adjusted to net out the regulated cap. Second, consistent with economic theory, when surcharging is permitted (and it did occur in Australia most notably for online air ticket purchases and phone payments), the interchange fee is neutral (as I will discuss in my next post). That is, the interchange fee reduction causes merchant fees to fall but issuer fees to rise (or loyalty schemes to be curtailed) but otherwise does not impact on the consumer’s choice of payment instrument. However, even if that were the case, it is surprising that there was not some period of adjustment.

The RBA continues to regulate interchange fees but has signaled that it is unlikely to adjust them further. When it comes down to it, by capping the fee the industry has survived without disruption and the RBA has ensured that rising interchange fees and associated problems as has occurred in the US will be avoided. That said, it may interest non-Australians to learn that the previous interchange fee was set in the late 1970s and was never changed despite that dramatic changes in the industry over the next two and a half decades. If there was any country without a credit card anti-trust problem it was probably Australia.

The Australian experience tells us that interventions to regulate interchange fees are probably not as important as ones that might deal with other card association rules or generic competition. But it also tells us that such interventions are unlikely to have dramatic consequences for the industry on the choice of payment instruments.

11 responses to What happened in Australia?


    But it doesn’t need surcharging. I can tell you that there are plenty of retail outlets not accepting cards. That is enough to get the price signals working. You have a price at a merchant who accepts only cash and one who accepts cards. The differential is the POS premium for using cards over cash.


    Joshua: I get all of this. I think the problem is the evidence (I believe I have seen this evidence) that prices stayed the same and surcharging, though available, was little used. My comments and Todd’s comments were based on confusion over the seeming effect that usage stayed the same but prices went up. I understand what you’re saying, but I don’t know where the evidence of the price reduction at POS comes from. (And sorry about the posting order)


    Geoffrey, no it is neutrality which is what my next post is about. Of course, I had wanted to post tht first!

    Neutrality means that even though consumers have less direct inducement to use cards, the premium they have to pay (via outlet or surcharge) for using cards at the point of sale goes down by the same amount. So it is not inelasticity if the full price to consumers has not changed. That appears to be what has happened in Australia. When the prices change, demand shows itself to be elastic.

    Anyhow this is explained in the next post.


    So to echo Todd, doesn’t that still leave a real quandary? The price of using a credit card goes up, there is no change to the price of goods purchased, yet credit card usage stays the same? Isn’t that practically the definition of inelastic demand? Or am I missing something?


    There was a one to one pass through of the interchange fee reduction to the merchant charges. My understanding is that there has been no impact on bank profits.

    Credit card demand and usage is not inelastic. The empirical analysis demonstrated that interest rate changes changed usage.

    Bob Chakravorti 8 December 2009 at 3:21 pm

    Do you have any ideas on the change in merchant adoption of credit cards resulting from the lowering of interchange fees? For example in Spain, reductions in interchange fees resulted in greater adoption by merchants which allowed consumers greater ability to use their cards.

    If I recall correctly, before the regulation there were only two types of interchange fees–electronic and non-electronic. Has that changed? In the US, we have a menu of interchange fees depending on various factors such as type of merchant, type of card, card-present or not, etc. Does something like this exist in Australia today?

    Finally, do you have an idea about bank profits? Did they stay the same? If they fell, any long-term implications on future system innovations?


    I’ve seen data suggesting increases in annual fees and the like–is that data inaccurate? But the absence of a change in payment choice suggests either little of this or considerable inelasticity.


    Cardholder fees are not a real variable but a price. There was no accurate data to look at the offsetting price movements. What was looked at was the impact on the choice of payment instrument. You can’t find the change there suggesting there was no trade-off.


    You note that the “change was virtually undetectable in any real variable to do with that industry,” but as Todd Zywicki points out in his post following yours, and as you allude to here, there was a change in fees charged to cardholders, right? It is certainly interesting that consumer demand was insufficiently elastic for this to affect card usage, but was there a net effect, or did the increase in costs to consumers exactly equal the decrease in costs to merchants? I would assume it did, unless there were other efficiencies realized or inefficiencies encountered.
    It seems important to view the issue as essentially a trade-off between consumers and merchants (a dynamic, by the way, that is directly opposed to the merchants’ characterization of their own position as serving consumer interests), and the question remains whether any overall efficiency interest is served by a policy favoring merchants in such a trade-off.

Trackbacks and Pingbacks:

  1. This is what was supposed to happen : Core Economics - January 14, 2010

    […] become less negative because banks have less incentive to induce the to use more cards. That said, as I have noted elsewhere, the merchants are less resistant to that card use and it looks like it has balanced out. So the […]

  2. Visa’s point of sale practices : Core Economics - January 8, 2010

    […] merchants from offering different prices to consumers based on payment instrument. That rule was abandoned in Australia and we see many instances where it bites. In that situation, the strong theoretical prediction is […]