Seven Truths About Regulating Interchange

Robert Stillman —  8 December 2009

Robert Stillman is a Vice President in the European Competition Practice of Charles River Associates

Interchange fees on payment cards are obviously a hot topic in the United States, but also in Europe and in many other countries around the world.  The report on interchange fees released last month by the US Government Accounting Office (GAO) notes that more than 30 countries have intervened or are considering intervening in the payment card industry.

Australia is one of the countries where there has been significant government intervention.  Most notably, the Reserve Bank of Australia (RBA) implemented a regulation in October 2003 that had the effect of reducing the average interchange fee on credit card transactions by approximately 50%.

From research that I have done on the effects of the regulations in Australia and from my reading of the ever-expanding literature on the economics of payment cards, I think there are a number of propositions about payment cards that are well supported by economic theory and the available evidence and that should constrain any debate about payment card issues.  I will be interested in the extent to which other participants in this blog symposium agree or disagree:

1.  A reduction in interchange fees will lead to reductions in merchant discount fees.

This proposition is almost trivial.  If this were not the case, it would be difficult to understand why merchants complain so vociferously about interchange fees.

2.  If interchange fees are reduced, it is highly unlikely that the resulting decline in merchant discount fees would be quickly and fully passed on to consumers in the form of lower prices and/or higher service levels.

The fact that merchants lobby vigorously for reductions (or even the elimination) of interchange fees must imply that merchants expect their profits to increase if interchange fees are reduced.  This is not to say that merchants will not pass-through some portion of any reduction in merchant discount fees; even a monopolist would find it profitable to lower prices if its input costs fell.  But I believe it is implausible that an expectation of higher margins is not an important part of the merchants’ interest in lower interchange fees.  Put differently, if it were really the case that merchants would quickly and completely pass-through any reduction in merchant discount fees, it seems unlikely that merchants would be fighting as hard as they have been fighting for lower interchange fees.

Note that, contrary to the claim that is sometimes made, it is not the case as a matter of economic theory that competition will necessarily ensure quick and complete pass-through of changes in input costs.  Even in a textbook model of perfect competition, an increase or decrease in input costs will not be passed through fully to consumer prices unless the industry has constant marginal costs.  In real-world industries of oligopoly and differentiated products, the economics of pass-through are even more complicated.

3.  A reduction in interchange fees will lead to increases in cardholder fees and/or reductions in card benefits (e.g. the value of reward programs).

I would be surprised if there was any argument about this proposition.  The evidence from Australia showed seemingly unambiguously that the reductions in interchange fees mandated by the RBA resulted in material increases in cardholder fees and material reductions in the value of reward programs.

4.  The government-mandated reduction in interchange fees in Australia did not appear to have any significant effect on the number of cardholders or card use.

In our paper summarizing the evidence from Australia, we pointed to possible factors that may have been operating in Australia at around the time of the RBA regulations and which may have obscured effects on card numbers and usage that otherwise would have been observed.  Even so, the evidence from Australia on card numbers and card use tends to support the view that, at least in a country such as Australia where card use is relatively “mature”, the demand to hold and use payment cards appears to be fairly inelastic.

5.  A reduction in interchange fees will reduce issuer profits.

Issuing banks tend to be opposed to regulatory efforts to limit interchange fees, which implies that issuing banks believe that such regulation would reduce their profits.  In theory, the opposition of issuing banks to interchange fee regulation could be due primarily to a concern that any significant reduction in interchange fees might trigger a “death spiral” (or, less colorfully, “negative network effects”).  However, if instead the demand to hold and use cards is relatively inelastic – which is what the Australian data tend to suggest is the case, at least for countries where card use is mature – then the opposition of issuer banks is less likely to be based on concerns about negative network effects and more likely to be based on concerns that a reduction in interchange fees will reduce issuer margins.

6.  The net effect of a regulatory reduction in interchange fees on the welfare of final consumers depends mainly on relative pass-through rates

The net effect of a reduction in interchange fees on the welfare of final consumers depends ultimately and mainly on relative pass-through rates – (1) the extent to which reductions in revenue from interchange fees on the issuing side are passed through to card holders in the form of higher cardholder fees and reduced card benefits relative to (2) the extent to which reductions in interchange fees on the acquiring side are passed through first to merchants in the form of reduced merchant discount fees and then on to consumers in the form of lower consumer prices and/or higher levels of service.

7.  If high interchange fees are the result of competition among schemes, the allegation of collusion seems misplaced.

As explained in the GAO report (and emphasized also by the RBA), interchange fees seem to be the result of competition among schemes.  In an environment in which merchant acceptance is not very sensitive to the level of merchant discount fees, competition among four-party schemes is likely to lead to higher interchange fees as the schemes use higher interchange fees as a tool to persuade issuers to issue and promote the usage of their particular scheme’s cards.  Against this backdrop, the suggestion by merchant organizations that they are the victims of collusion is curious.  If merchants are the “victims” of anything, it seems more accurate to describe them as the victims of competition in a two-sided market in which enough consumers on the other side of the market use only one card (“single homing”) that most merchants decide to accept all cards (“multi-homing”).  The same pressures for relatively high merchant discount fees would arise under these conditions even if all schemes were three-party schemes and there were no interchange fees.