Platform Competition in TwoSided Markets: The case of payment networks

1
Platform Competition in Two-Sided Markets The
case of payment networks Sujit Chakravorti and
Roberto Roson discussed by Wilko Bolt
Conference on The Economics of Payments, March
31 - April 1, 2004, Atlanta
2
Content
Theoretic analysis of competing payment networks,
where networks offer differentiated products in
terms of benefits to consumers and merchants. As
pricing arrangements for payment cards have
attracted controversy and triggered antitrust
scrutiny, understanding the competitiveness and
dynamics of card payment systems is a key policy
issue.
Punch line Competition unambiguously increases
consumer and merchant welfare, suggesting that
policymakers should promote competition among
networks.
3
1. Social Welfare
Duopoly leads to lower consumer and merchant
fees. How would this change if more and more
networks would enter the market? (as n?8)
In other words, what about social optimal
prices? Could you not characterize those prices
in your model under a uniform distribution? What
if social prices lead to a loss-making business
for card networks? (see e.g. Armstrong, 2004)
4
2. Pricing and elasticities
Demand elasticities are key determinants for
platform pricing. But what are their sizes? Can
they be estimated empirically with some accuracy?
There is not a lot of relevant payment data.
Concerning the optimal pricing structure,
consumer and merchant fees are proportionally
related to elasticities. Perhaps one would expect
an inverse relation as in Armstrong (2004). What
is the correct interpretation of the pricing
rule?
5
3. Network competition and efficiency
Card payment systems typically have large set-up
costs. Duplication of these fixed costs does not
seem socially desirable, which may lead to
competitive bottlenecks. The ability of a
potential entrant to access the existing network
becomes important and triggers the question of
access pricing and raising rival costs. Can
your model deal with this issue?
Network competition induces a trade off between
static and dynamic efficiency low end-user
prices may undermine investments in new network
technologies.

6
4. The importance of multihoming
There is a delicate distinction between
membership and usage multihoming (Rochet and
Tirole, 2004). Rysman (2004) presents evidence
that cardholders multihome in membership rather
than in usage. What would happen in the model if
cardholders multihome in membership and merchants
in usage? Would this lead to indeterminacy of
prices? Still, this happens in real life!
Concerning different modes of payment (cash,
debit card, charge card, credit card)
singlehoming seems strange, it also depends on
the size of the transaction. In Holland,
policymakers would like to induce consumers to
multihome!

7
(No Transcript)