RDP 2009-02: Competition Between Payment Systems 5. Conclusions

In this paper we have developed a model of competition between payment platforms which avoids many of the limiting assumptions commonly made in analyses of such competition to date. Specifically, our model allows for heterogeneity in consumer and merchant benefits from the use of platforms, flat rather than per-transaction pricing by platforms to consumers, and fully endogenous choices by both consumers and merchants about how many cards to hold/accept.^[31] These features have a number of potentially far-reaching consequences. In particular, they allow for different consumers to face different effective per-transaction prices for the use of a given platform's card – something which is not the case when platforms' pricing to both sides is purely per-transaction, or when (as in the CR model) consumers are prohibited by fiat from multi-homing. This represents a fundamental change – and one that better matches reality in a number of settings of interest, such as the case of competing credit card schemes.

In developing our ECR model we have also shown how the card subscription decisions of consumers and card acceptance decisions of merchants may be represented geometrically. This yields frameworks which allow us to more readily examine and understand the behaviour of, and incentives facing, merchants, consumers and platforms. We have also used the model to concretely illustrate a new potential source of non-uniqueness in consumer and merchant market outcomes, for given platform fees, which is distinct from the usual ‘chicken and egg’ phenomenon. Unlike the latter, which may arise even for a single-payment system, this source of potential non-uniqueness relates to the extent of consumer multi-homing/merchant steering, and so can only occur in the context of competing payment systems.

In the sequel to this paper (Gardner and Stone 2009a) we use the model developed here to explore more fully, via simulation analysis, the likely effects of competition on platforms' pricing strategies. In particular, we investigate how these effects may be influenced by allowing fully endogenous multi-homing by both consumers and merchants, rather than having only one side of the market allowed to hold/accept multiple cards.

Footnote

It also incorporates the ‘derived demand’ aspect of payments markets (although it does not allow for business stealing considerations). [31]