Completed Dissertation Project

MasterCard and Visa vs. the

European Commission

Nicola Guiso

1103276

Analyse a recent competition law case in the USA or EU. Why was the case initiated?

What were its effects? What are its implication for competition policy?

EC831 Project Nicola Guiso

Introduction

This paper will focus on the topic of interchange fees within payment systems, which has been the

centre of a legal case between the European Commission, MasterCard and Visa. Interchange fees are

a side payment going from an acquiring bank, which deals with merchants, to an issuing bank, which

provides services to cardholders. To be more pragmatic, when we plug a credit or debit card in a

POS terminal, our bank will make a payment to the acquirer equal to the cost of the purchase minus

a set percentage of such expenditure, which is the interchange fee. The acquirer will then transfer

the amount of money received from the issuer to the merchant, minus the merchant service charge

(MSC) which, among other things, is based on the level of the interchange fee. Hence although

initially the interchange fee is charged to acquiring banks, it is actually merchants that bear the cost

through the MSC.

The constantly increasing usage of cards to substitute cash makes interchange fees one of the "most

frequently paid prices of modern economies"1, and yet consumers are widely unaware of them.

Interchange fees have been, in the last decade, greatly examined by regulators and competition

authorities, due to their nature and settings, which are believed to harm consumers under certain

circumstances.

I will begin my analysis by firstly analysing part of the recent economic and legal literature on the

topic of interchange fees. In this section, I will introduce what interchange fees are and what their

technical role within a payment system is, provide information related to their historical background

and briefly explain the two-sided nature of the payment market. In addition I will set out the specific

legal arguments the European Commission is basing its case around. In the main body of this paper, I

will link the main economic models on the topic of interchange fees to the specific case of

MasterCard and Visa. This part of my work will require me to analyse the research of authors such as

Baxter, Tirole, Rochet, and finally Vickers. After that I will move onto the actions that competition

1 Vickers (2005)


authorities can take to eliminate or contain the anti-competitive aspects of interchange fees, with

the aim to understand whether regulation and competition policy can increase efficiency for this

market.

In the conclusive paragraph of this paper I will attempt to answer questions such as:

Did the Commission utilise fair and efficient measures in its investigation against MasterCard

and Visa Interchange Fees?

Are the commitments taken by the two firms enough to reach social efficiency within the

payment system?

In there still room for market failure and negative externalities within the payment market?

And if so, can regulation improve the current situation?

Is there need for a full elimination of Interchange Fees?

Open vs. Closed Payment Systems and The Role of Interchange Fees

Payment Systems can be of two types, closed or open. Closed systems, also called three parties, such

as American Express simply consists of a network owner, a consumer and a merchant.

Figure 1; Source: Rysman and Wright (2012)

On the other hand, open systems like MasterCard and Visa, also called four parties since they do not

account for the network owner, include an issuing and an acquiring bank, plus a consumer and a

merchant.


Figure 2; Source: Rysman and Wright (2012)

As the figure one and two show interchange fees only exist within open payment systems. The

reason for this is that in closed payment systems the network performs both issuing and acquiring

activities, and sets cardholders and merchants fees as to include an implicit side payment. In open

payment systems external issuers and acquires join the scheme, deal individually with cardholders

and merchants, and also set their own charges for end-users. Interchange fees which in this case are

a separated and explicit payment, can either be 'bilateral', if stipulated between individual banks, or

multilateral, when set by the network itself. The two types of fee play exactly the same role within

the payment system. The only difference between them is that multilateral interchange fees are

imposed only if the issuer and acquirer taking part in the transaction do not reach a bilateral

agreement on how much they have to pay each other for the interdependent services they give to

end users2.

In accordance to this view Rysman and Wright sustain in their research that as well as being a

substantial part of the cost faced by acquiring banks, which if raised would end up harming

merchants, interchange fees, whether bilateral or multilateral, also represent a vital flow of revenue

2 Lauwers and Draffan, Slaughter and May (2012)


for issuing banks. Hence, if they did not exist issuers would not be able to offer the same benefits

and incentives to cardholders3 which, as I will explain in this research, indirectly increase benefits

and profits for merchants.

The European Commission however, considers the multilateral nature of interchange fees capable of

harming merchants and ultimately consumers. The Commission in fact, compares interchange fees

to a minimum price imposition on merchants for accepting cards, and took the decision to prohibit

those applied by MasterCard4.

This belief is contradicted by Christoph Baert, MasterCard's Regulatory Affairs Counsellor, who

stated " MasterCard is a global payment scheme with thousands of licensees in Europe, acquiring

millions of merchants and issuing cards to an even greater number of cardholders. When a

cardholder goes to a merchant displaying the MasterCard logo on its door, the cardholder knows

that he can pay with his MasterCard, and the merchant knows that it will be paid (even if it is

eventually a fraudulent transaction or if the cardholder is insolvent). For a scheme to work with so

many parties involved, and where it is practically unrealistic to have bilateral agreements between

all acquiring and issuing banks, it is indispensable to establish fallback terms of dealings"5. Without

assuming that this side of the discussion is more correct than the other, we still have to consider that,

after the actions that MasterCard and Visa proposed to take to be compliant with competition laws,

interchange fees do still exist. Additionally, it has to be noted that economics teaches us that in any

market or company involving two sectors which act as individual entities, a transfer price needs to

exists in order to avoid a negative externality called 'double marginalisation', which I will discuss

further later in this paper. In the next section I will explain the particular economic nature of the

payment market which, due to the presence of two kinds of end-users, behaves differently from

most others.

3 Rysman and Wright (2012) 4 European commission, IP/071959 (2007) 5 Christoph Baert (2012)


A Brief Analysis of Two-Sided Markets

The payment market is said to be two-sided, due to the existence of two groups of end users whom

the platform provides services and products to. The actions of such groups are interdependent, and

the will of each group to maximise their personal welfare rather than the social one generates

negative externalities. Therefore the market or, as it is sometimes called, platform, provides a

"meeting point" for end users where those externalities can be resolved, so that they can interact

and overcome the otherwise very high transaction costs. A perfect example of a two-sided platform

is a heterosexual night club. A night club in fact serves two kinds of end users, men and women, and

provides an actual meeting point where they get to interact. The club aim is to attract both men and

women in the right proportion , and can use differential pricing to get the right balance. As it can be

imagined a club almost completely populated by one sex would probably not be too attractive for

the other, which is why certain clubs charge a lower price to the sex in short supply. If this was the

case for women for example, their entrance price might be lower than that for man, or it might even

be zero. Additionally the club may offer them some extra benefits, such as queue-jumping. One

could obviously argue the unfairness of this price structure for men, although if women, whose

demand for nights out in this example is more elastic, faced a higher entrance price for clubs they

could decide to change venue, or potentially stop clubbing completely. Men on the other hand,

would face a lower entrance price for clubs but have also less opportunities to meet women on a

night out. Therefore the decrease in entrance price for men does not necessarily increase the

benefits they enjoy because of the different levels of elasticity of demand6. A great definition of this

phenomena is given by Rochet and Tirole who said:

6 David S. Evans and Richard Schmalensee (2007)


" A market is two-sided if the platform can affect the volume of transactions by charging more to one side of the

market and reducing the price paid by the other side by an equal amount; in other words, the price structure

matters, and platforms must design it so as to bring both sides on board.7”

In the payment market, card schemes are the platforms which enable cardholders and purchasers to

interact. As I will explain more in detail later on in this paper, the absence of a network would in fact

create huge transaction costs, and cause issuers and acquirers to aim at maximising their individual

profits, without thinking that their interaction and cooperation is essential for the system to work.

Just like in the nightclub example, it is extremely important that merchants and cardholders join the

payment system in the right proportions. Suppose only one person in the whole community carried

a card, but all merchants within the market could accept it. The individual cardholder would enjoy a

great deal of benefits, but the merchants would be paying acquirers simply to allow one person to

pay by card. Similarly, if all individuals in a market carried cards, but only one merchant in the whole

community could accept them, cardholders would be paying issuing fees only to have the advantage

to pay electronically in one shop. The payment network however, facilitates interaction and

cooperation among issuers and acquirers, enabling them to lower joint costs and offer better terms

to merchants and cardholders which, by increasing the overall amount of card transactions, benefits

the whole society.

Brief Historical Background on Interchange Fees

Interchange fees were first introduced by Bank of America (BoA) in 1971, and applied to its

BankAmericard system of credit cards, the predecessor of Visa. The system was created in 1958, it

dealt with both sides of the payment system, issuing and acquiring, and set all the fees related to the

system: annual fees, late fees, interest rates, and finally, cardholder and merchant charges. Bank of

America had therefore the power to set both the price level and determine its structure. In 1966

BoA started to allow other banks to join the system, and choose their own fees for merchants and

7 Jean-Charles Rochet and Jean Tirole (2006)


cardholders. The individual banks were therefore free to choose their preferred price structure

when the merchant and cardholder involved with the transaction were all among their members.

However as the size of the system increased, more and more transactions involved a cardholder and

a merchant belonging to different banks, which according to Bank of America's rules meant that the

acquirer had to pass the entire merchant fee to the issuing bank. This caused an increase in the

incentives acquiring banks had to lie about the fees charged to their merchants and, more

importantly, a strong lack of incentives to sign up merchants, since the revenue gained by acquirers

for card transactions was often zero. In order to put an end to this situation, which was negatively

affecting the whole system, an interchange fee was introduced. The fee was initially set at 1.95

percent of the value of the transaction. Individual acquirers could therefore set the merchant

discount higher in order to cover the cost of the interchange fee, which had become a source of

revenue for the issuing side of the payment system.

Up until 1979, only a few payment systems , such as MasterCard and Visa, adopted Interchange Fees

as a transfer price from acquires to issuers. During that year however, interchange fees became

subject of legal litigation for the first time. The National Bancard Corporation (Nabanco), claimed to

be damaged by the practices carried out by Visa member banks which, by setting the level of

interchange fees charged to merchants, engaged in illegal price-fixing. In 1986 however, the case

ended in favour of Visa, with the commission acknowledging the potential efficiency benefits

brought by interchange fees. The court declared that:

"Another justification for evaluating the [interchange fee] under the rule of reason is because it is a potentially

efficiency creating agreement among members of a joint enterprise. There are two possible sources of revenue

in the Visa system: the cardholders and the merchants. As a practical matter, the card-issuing and merchant-

signing members have a mutually dependent relationship. If the revenue produced by the cardholders is

insufficient to cover the card-issuers’ costs, the service will be cut back or eliminated. The result would be a

decline in card use and a concomitant reduction in merchant-signing banks’ revenues. In short, the cardholder

cannot use his card unless the merchant accepts it and the merchant cannot accept the card unless the


cardholder uses one. Hence, the [interchange fee] accompanies “the coordination of other productive or

distributive efforts of the parties” that is “capable of increasing the integration’s efficiency and no broader than

required for that purpose.”8

After the Nabanco case, only a few academics and policy makers focused on the topic of interchange

fees. In 1983 William Baxter published a groundbreaking paper on this topic which will be discussed

more in detail later on in this research. The interest towards the topic of interchange fees started to

increase exponentially among academics and regulators at the end of the 90's, due to the growing

importance of cards as a mean on payment in a huge number of countries. The higher level of card

usage is indeed positively correlated with the share of merchant costs taken up by the merchant

discount, which caused a high number of retailers organisations to seek for regulations that could

protect them against the constant increase of interchange fees. This led to a number of legal

complaints, such that of EuroCommerce in 1977, which brought Visa to lower its Interchange Fees,

and the two cases that I will now analyse in this project9.

Explanation of the Legal Arguments Behind the Case

The initial cause of both cases of the European Commission against MasterCard and Visa was a

possible breach of the article 101 TFEU reported below.

Article 101, the ex Article 81 TEC, states that:

1. The following shall be prohibited as incompatible with the internal market: all agreements between

undertakings, decisions by associations of undertakings and concerted practices which may affect trade between

Member States and which have as their object or effect the prevention, restriction or distortion of competition

within the internal market, and in particular those which:

(a) directly or indirectly fix purchase or selling prices or any other trading conditions;

(b) limit or control production, markets, technical development, or investment;

8 National Bancard Corp v. Visa U.S.A., (1986) 9 David Evans, Richard Schmalensee (2005)


(c) share markets or sources of supply;

(d) apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a

competitive disadvantage;

(e) make the conclusion of contracts subject to acceptance by the other parties of supplementary obligations

which, by their nature or according to commercial usage, have no connection with the subject of such contracts.

2. Any agreements or decisions prohibited pursuant to this Article shall be automatically void.

3. The provisions of paragraph 1 may, however, be declared inapplicable in the case of:

- any agreement or category of agreements between undertakings,

- any decision or category of decisions by associations of undertakings,

- any concerted practice or category of concerted practices,

which contributes to improving the production or distribution of goods or to promoting technical or economic

progress, while allowing consumers a fair share of the resulting benefit, and which does not:

(a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these

objectives;

(b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the

products in question.10

Article 101 deals with horizontal and vertical agreements. Such agreements, both involving multiple

and different parties of a market, have a very different effects on the competitiveness of a market.

Horizontal agreements consist in agreements among competitors, where information related to

products, price level, strategies, and technology are shared. Such agreements can lead to price-fixing,

therefore limiting competition within a market and reducing social welfare. For these reasons, other

than in some rare cases (i.e. cooperative agreements in R&D), they are prohibited. Vertical

10 European Commission, Official Journal 115 , 09/05/2008 P. 0088 - 0089


agreements, on the other hand, involve cooperation among firms which preside at different stages

of the production process. An example of this can be a merchant who offers to publicise a particular

product in his shop in exchange for lower retail prices. Such agreements often enhance competition,

whilst only pose a threat towards it if the parties involve enjoy particularly high levels of market

power.

As one can imagine, attempting to regulate such different kind of agreements with a unique

legislation is unlikely to be an efficient strategy. For this reason in 1999 the European Commission

decided to introduce two additional features to the article in order to make it more consistent with

the economics behind it. The first is a 'block exemption' from the constraints of the article if a

vertical agreement was found not to involve any company with a market share superior to 30

percent. The second one is a so called 'black list' of clauses and hard-core practices which are

believed to be able to fix the price level of a market, directly or indirectly. If a vertical agreement was

found to have among its different parts any of these hard-core practices no exemption could have

been granted to the agents involved, even if their market share was drastically smaller than 30

percent11.

MasterCard and Visa were both prosecuted by the European Commission with the accuse that the

level of interchange fees set by the two payment systems was in fact a decision taken by

associations of undertakings, due to the commonality of interests of their members banks. The two

titans of the payment market however disputed and protested against such statement saying that,

since both companies became publicly traded, their main interest is the welfare of their

shareholders and not its licensees. As well as that MasterCard and Visa attempted to gain an

exemption under the article 101(3) stating that interchange fees brought strong efficiency

improvements to the payment market such as higher consumer benefits and technological progress,

carried out using the revenue stream that interchange fees generate. Failure by MasterCard and Visa

11Motta (2007)


to provide evidence to support this claim caused the companies to have to agree to take the

measures listed in the following two sections of this paper, which are supposed to get the level of

interchange fees closer to the socially optimal one.

The Commission Case vs. MasterCard

After an initial investigation into MasterCard's network rules in 1999, the European Commission

opened a case into MasterCard's cross- border Interchange Fees in 2002. The reason which caused

the start of the investigation was the possibility that interchange fees were set at a level that would

inflate merchant costs, which would ultimately lead to higher prices for consumers12. For this reason

in 2003 the Commission sent a statement of objection to MasterCard, where the company was given

six months to prove that the positive efficiencies provided by interchange fees outweigh their

negative effects13. For failing to provide the evidence the Commission was looking for, MasterCard

interchange fees were declared not exempted from Article 101(3) of the EC Treaty, unless the

company agreed to take a number of regulatory measures ordered by the commission. One of these

undertakings was the reduction of its interchange fees level to 0.30% per credit card transaction,

and 0.20% per debit card transaction. Before the investigation MasterCard's interchange fees for

Maestro debit cards and MasterCard credit cards ranged from 0.40 to 1.9 percent of the sale value,

depending on the card used. As well as that, MasterCard agreed to take a number of measures

aimed at increasing the level of transparency of its scheme, in order to enhance the level of

competition and allow cardholders and merchants to make more informed decisions14.

12 European Commission, MEMO/07/590 (2007) 13 European Commission, MEMO/06/260 (2006) 14 European Commission, IP/09/515 (2009)


The Commission Case vs. Visa

The European commission formally initiated investigating Visa's interchange fees before 2002. As

well as that, the commission was looking into some of Visa's network rules such as the "honour all

cards rule" and the "no surcharge rule". The former obliges merchants to accept all cards issued with

a Visa logo, whilst the second one stops merchants from charging consumers extra if paying with

Visa branded cards. As I will discuss in the penultimate section of this paper, which focuses on the

regulation of interchange fees, both rules strongly affect public policy. In 2002 the commission took

the decision to exempt the interchange fees proposed by Visa, and the "honour all cards rule" until

2007, due to a number of measures that Visa decided to adopt. Interchange fees were reduced from

1.1 percent to 0.7 percent or lower, and the company released a higher amount of information

related to interchange fees in order to increase their level of transparency. At the end of 2007

however Visa would have had to prove that its interchange fees and rules were in compliance with

the Article 101(3) in order for the exemption to hold up15. Visa was therefore sent another

statement of objection in 2009, once again on the basis that the current levels of its interchange fees

inflated merchants' cost of accepting cards. Visa therefore used the tourist test to lower the

Interchange Fees' cap to 0.20% of the transaction cost, whilst taking further actions to increase

transparency in this market. According to the Commission, Visa's commitments resolved its concerns

with relation to the interchange fees applied to debit cards, and ended this part of the investigation.

The commission however is still pursuing an ongoing investigation with relation to the interchange

fees applied to Visa's credit cards and deferred debit cards16.

It's now time, after introducing all the basic elements needed to understand the case, to focus on

the economics side of payment systems and interchange fees. Using the earlier mentioned

researches and publications I will attempt to determine whether interchange fees constitute a

market failure, and if so whether government regulation is the best way to resolve it.

15 European Commission, Memo/08/170 (2008) 16 European Commission, Memo10/224 (2010)


Determining the socially optimal MIF level

The European Commission started investigating MasterCard and Visa's interchange fees on the bases

that their multilateral aspect would inflate costs for end-users of the payment process, and provided

an opportunity for price fixing without necessarily increasing innovation and development within the

payment system. This motivations could be considered correct, particularly in view of the fact that

MasterCard was not able to provide the required evidence that interchange fees' advantages

outweigh their negative and anti-competitive effects on this market. Many academics who have

focused their research around the topic of interchange fees have however found that the non-

existence of such fees could also create a negative externality and therefore prevent the payment

market from reaching its socially optimal status. Such externality, which was mentioned a few

paragraphs earlier, is called 'Double Marginalisation' and as Shmalensee points out in his research, it

can be resolved using a transfer price, which in the case of payment systems is an interchange fee17.

A double marginalisation problem arises when two entities within the same market, or even within

the same company, which gain benefits from working interdependently, act like two separated

entities and aim at maximising their own individual profit rather than the total one. In the case of

the payment market, a double marginalisation problem will cause issuers and acquirers to set their

price to end-users too high. Issuers therefore will experience a loss of customers since the higher

price charged would result in higher annual fees, interest rates, or less benefits. Consequently

demand for card transactions would decrease and affect merchants negatively, as well as driving the

market away from the socially optimal amount of card transactions. An interchange fee, paid by

acquirers to issuers, allows the latter to lower their costs, and offer better terms to cardholders,

generating a higher level of card transactions. Consistently with this view, Tirole defines Interchange

Fees as a "reallocation of cost between two categories of end-users"18.

17 Richard Shmalensee (2002) 18 Tirole (2011)


Of course, transfer pricing is a two way street. It is true that the interchange fee increases costs for

acquirers and necessarily merchants but, if set at the right level, it also generates a number of

benefits which can outweigh such costs. On the other hand if the interchange fee was set too high,

demand for card acceptance by merchants would decrease, due to a lower benefits-to-cost ratio,

resulting in a lower level of card transaction and a loss of revenues for issuers.

At this point it can therefore be concluded that economic principles and theory strongly contrast the

view of the European Commission with regards to interchange fees, its accusation to MasterCard

and Visa of price fixing, and the request to eliminate interchange fees made to MasterCard. The

economic literature with regards to this topic seems in fact to point at a direction where interchange

fees, if set at the right level, can generate a number of benefits for the payment market and the

community. What is relevant therefore in order to understand and solve the issues caused by

interchange fees is whether the business of payment cards can expand as a whole by transferring

more revenues in one direction or the other19.

In the following paragraph I will discuss Baxter's model of interchange fee. The model makes some

strong and possibly unrealistic assumptions. Competition is perfect there is a complete absence of

fixed costs and asset prices. Regardless of that though, the model provides some very important

insights, and points out that the existence of a transfer price is essential for the system to work in

the most efficient way possible.

19 Rysman and Wright (2012)


The Baxter Model of Interchange Fee and the tourist test

The first model of interchange fees which was able to determine a level closer to the socially optimal

one was created by William Baxter in 1983, and was based on a simple demand and supply concept.

Demand and supply for transactional services are in fact both joint, since there would be no point in

consumers demanding cards if merchants were not demanding card-acceptance solutions, nor in

issuing cards to consumers when acquirers are not signing merchants, and vice versa. The

interdependency of demand and supply causes the existence of joint costs faced by end-users, as

well as the banks involved with

the transaction.

Figure four graphically shows

this scenario. Line dm and dp

respectively represent the

demand curves for merchants

and cardholders, which, if

added up vertically, generate d,

the aggregate demand for

transactional services. On the

supply side, Cm and Cp are the

cost curves, faced by the merchant and cardholder's banks respectively, which are added vertically

to obtain the aggregate supply curve, C.

As it can be seen from the graph equilibrium occurs at point e, where the number of electronic

transactions taking place is q*, and the total supply cost incurred to deliver it is p*. Such cost is

covered by acquirers and issuers, which will pay an amount equal to q*a and q*d, respectively.

Subsequently the cost incurred by banks will be recouped through merchant fees (q*c) and

cardholders fees (q*b ). Note that the cost incurred by issuers, q*d, is not covered by the

Figure 3; Source: Baxter (1983)


contribution of the cardholders, q*b, which is smaller by the distance bd. On the other hand, the

price paid by merchants to their banks exceeds the banks' cost by ac, which is equal to bd. Hence,

the merchant banks will have to make a side payment of ac = bd to the purchasers banks in order to

equalise the cost created whilst supporting bank transactions.

The model develops in the following way:

firstly, the purchaser receives the goods

and services bought from the merchant.

The acquirer pays the merchant the sale

price, minus the merchant service charge,

q*c. Subsequently the issuer reimburses

the acquirer for the price paid for the transaction, minus the difference [q*c - q*a]. Finally the

cardholder's account gets credited for the cost of the purchase, plus an additional premium of q*b.

The acquirer therefore earns revenue q*a, whilst the issuer earns a revenue of [(q*b + q*c)-q*a]=

[q*e-q*a]= q*d , which is the cost incurred by the issuer in the first place20.

As I have already mentioned in the previous sections of this paper, the reason for the need to

equalise the cost incurred to provide transactional services is that, even though merchants do not

directly deal with issuers, the benefits that the latter provide to cardholders increase the level of

electronic transaction within the market, enhancing the convenience benefits and profits for

merchants, and also acquirers. Therefore, in a perfectly competitive market the increase in merchant

costs that is caused by the interchange fees should be balanced out by a number of benefits that

card-accepting merchants enjoy. Additionally by offering the possibility to pay by card in their shop,

merchants can exert a higher level of attractiveness on customers, which allows them to increase

sale prices and extract a bigger portion of the consumer surplus from customers.

20 Baxter(1983)

Figure 4


When deciding whether to accept cards a merchant should consider:

The next best alternative to card payment? Cash or nothing?

The net benefits to his business when cards are used to substitute cash

His willingness to pay for transactional services once he takes into consideration the fact

that accepting cards attracts a higher number of customers21

Suppose a customer has the possibility to pay by either card or cash. If the last two considerations

above do not outweigh each other, then the merchant should be indifferent to what mean of

payment the consumer chooses. Under such circumstances the cost he incurs by accepting a card

payment will in fact not be higher than the benefits created by the transaction. In this case, it can be

said that the interchange fee set by the network passes the "tourist test". The tourist test was

invented by Rochet and Tirole in 2006. It relates to a situation where a random customer (the tourist)

enters a shop unaware of their card acceptance policy, and exhibiting clear signs that he/she has

enough cash in his pocket to make the purchase. In this scenario, if the merchant has no incentive to

deny a card payment neither ex-ante (before the payment), or to regret allowing it ex-post (after the

transaction has taken place) , the interchange fee will be at a level that does not exceed the

convenience benefits earned by the merchant through card acceptance22. Any extra charge related

to a card transaction will be passed on to the consumer who now faces all the benefits and all the

costs of card acceptance, and will be led by the interchange fee to take the efficient decision as to

what mean of payment to use. In other words by setting the level of interchange fee equal to the

convenience benefits of merchants, consumers will internalise the externality imposed on

merchants. As well as that, interchange fees will not be set at a level that allows banks to take

advantage of the fact that merchants might feel obliged to accept cards even when the cost of doing

so exceed the benefits.

21 Tirole (2011) 22 Rochet and Tirole (2007)


Therefore the tourist test is based on the model created by Baxter where there is perfect

competition among issuers and acquirers, and the setting the level of interchange fee is merely a

price-restructuring practice rather than an exercise of market power. The level of interchange fee

calculated using such method coincide with the Baxter interchange fee which is socially optimal.

Provided the assumptions made by the two models hold, a card transaction is the efficient choice

only if:

[Benefits for cardholder + Benefits for merchant]≥[ Cost to cardholder + cost to merchant].

The question which remains unanswered is what the benefits gained by merchants through card-

acceptance actually are, and how can their value be translated in a financial charge like the

interchange fee. In one of his papers Tirole outlines the convenience benefits of merchants who

accept cards very clearly:

Table 1; source: Tirole (2011)


The first column of the table refers to the second bullet point listed above, the direct benefits gained

by merchants who allow for card payments. If we name the cost of card acceptance for the

merchant Cm, and the benefits S, an interchange fee ≤ [S - Cm] will leave the merchant completely

indifferent as to the method of payment chosen by the customer, and hence, it will pass the tourist

test. The right hand column on the table however, introduces a variable A which accounts for the

increased attractiveness of a card-accepting merchant. When factoring A into the equation a

merchant will be willing to accept a higher interchange fee, since the additional service provided to

his customers enables him to charge higher prices and extract a bigger portion of the consumer

surplus. The merchant will therefore accept an interchange fee ≤ [S +A - Cm], which will leave him

indifferent to the mean of payment ex-ante, but wishing the customer had opted for cash ex-post.

For example, consider a merchant who is dealing with a customer that carries an American Express

card. The merchant is aware that Amex often charges fees as high as four percent of the value of a

transaction, but also knows that American Express cardholders are often well-off and are likely to

spend a considerable sum of money in his shop. In order to accommodate the needs of the customer,

and earn the substantial mark-up of the possible transaction, the merchant is likely to accept the

cost imposed by the network, regardless of the level23.

The choice as to where to place the interchange fee in the interval [S - Cm; S +A - Cm ] is in the hands

of the network, and will depend on two factors. First of all, how many payment networks a

cardholder is member of. Suppose a cardholder can use an Masterard card or a Visa card to make a

purchase. If MasterCard' interchange fee ≥ Visa’s interchange fee, the merchant will decide to simply

turn down MasterCard branded cards, putting pressure on the scheme to decrease its interchange

fee level. This phenomenon is called "multi-homing".

The second factor is how informed consumers are about the shops' card-acceptance policy. Suppose

in fact, that a shop only deals with tourists who, unaware of local shops' card policy, carry cash on

them. The possibility to substitute cards with cash will put downward pressure on the interchange

23 Tirole (2011)


fee, which will be closer or equal to [S - Cm]. On the other hand if the shop only deals with usual

customers, who are aware whether a shop allows for card payments, they might not bring cash with

them as a substitute for cards. The network can then afford to charge a higher interchange fee,

closer or equal to [S +A - Cm], as the merchant will not want to lose the acquired attractiveness that

accepting cards confers, nor miss a sale. Avoiding missed sales are is one of the most important

merchant advantages out of the list above. A merchant who refuses to accept cards in fact, shuts his

door to a multitude of customers; those who do not carry cash, due to having forgotten it or not

being able to reach an ATM, and those who do not momentarily have enough liquidity in their

account to make the purchase but rely on the interest free period offered by credit cards. As well as

that, for businesses like take-away restaurants or florists, refusing card payments means turning

down all customers who would like to make the purchase and pay for it over the phone. Finally, not

accepting cards completely shuts down any web-based channel of sale which, due to the constantly

increasing importance of online-shopping, could create huge profit losses for a merchant24.

As I mentioned earlier on in this paragraph, merchants might feel compelled to accept cards, in

which case benefits generated by card-acceptance, such as the ability to avoid missed sales, might

be seen more like something that puts pressure on merchants to take cards. If they were to turn

electronic payments down, they would run the risk to lose customers and succumb to their

competitors who might decide to allow them. This is the base of a very important argument, the

"must take cards" argument, made by John Vickers, which I will discuss in the next section.

24 Tirole (2011)


Vickers "Must Take Cards" Argument and the negative externalities cause by interchange fees

In his study of interchange fees and the payment market Vickers considers the compelling feeling

that merchants might experience when deciding whether or not to accept payment cards. Whilst is

true that accepting cards confers a higher level of attractiveness to a merchant, as well as a number

of benefits, turning cards down might result in missed sales, loss of customers which might choose

to go to a card-accepting shop, and ultimately loss of profits. Like in the examples given in the

previous section, sometimes consumers might not be able to pay in any way other than

electronically. Furthermore, some purchasers might simply not want to pay by cash to make the

most out of the reward schemes offered by issuers, which sometimes award points to customers

who use their cards more frequently. Merchants might then feel obliged to accept cards even when

cost of doing so is higher than the benefits gained, which creates a "must take" feature of payment

cards that can be further enhanced by a competition externality. In a market where, as it often

happens in real life, competition is less perfect on the issuing side than on the acquiring side, issuers

might not give back 100 percent of the revenue gained through interchange fees to cardholders, and

will be able to increase their profit by doing so. Issuing banks could then, set the level of interchange

fees as to maximise their own profit, and exploit the "must-take" characteristic of cards to push for

card usage, over and beyond the level required to reach social optimality. As well as that, issuers

could use the pressure merchants experience when it comes to card-acceptance to increase the

revenue gained through interest payments on credit cards outstanding debts. If issuers are aware

that, for the reasons outlined above, merchants will accept cards even then it does not necessarily

make sense from a financial point of view, they might offer better convenience benefits for

cardholders, in order to distort the way in which means of payment are chosen. This distortion can

cause an over usage of cards, which are highly costly for merchants, but also highly profitable for

issuing banks25.

25 Vickers (2005)


The argument made by Vickers does state that the way in which interchange fees are set, and also

the asymmetrical competition of payment markets, leave room for collusion and cartelisation,

particularly on the issuing side. The supporters of this theory back this claim up by saying that

MasterCard and Visa did in fact start as associations of undertakings, serving banks which also

owned the two schemes. Disregarding the fact that both MasterCard and Visa are now publicly

traded companies, which act in the interest of their shareholders and not licensed banks,

interchange fees apply to a market which serves two kinds of end-users, meaning that when the cost

for one side goes up, it automatically goes down for the other. In other words, regardless of the level

at which interchange fees are set, even if that was zero, the only thing they affect is the price

structure and not its level. Therefore the two-sided nature of the payment market makes the

cartelisation claim invalid. Economic theory of two-sided markets in fact teaches us that, in payment

systems, charging one side more than the other, according to their elasticity of demand and

marginal evaluation for transactional services, is the only way to attract both groups of end-users on

the two-sided platform provided by the payment network, so that they can make the most of the

benefits which arise from their interdependent relationship.

As well as that, in accordance to the conclusion of the Nabanco case, interchange fees do represent

a source of revenues for issuers, which can be used not only to benefit cardholders with additional

benefits and lower fees, but also to carry out R&D. Payment systems have in fact in shown a huge

development curve in the last few years, with banks bringing out products such as online and mobile

banking, e-wallets and many more. Hence, if seen under this light, interchange fees could be eligible

for an exemption under the article 101(3), which states that even agreements taken by associations

of undertakings can be allowed if they generate progress and development within their market.


Regulation of Interchange fees

There are two possible market failure that interchange fees can cause within the payment market.

First of all, their multilateral nature can limit competition by stopping individual banks to stipulate

bilateral agreements between themselves. Second of all, the level at which they are set, if not

socially optimal, could distort competition, for instance by imposing unnecessarily high costs on

acquirers, and therefore merchants.

There are three ways in which these issues could be tackled. The first one is to adopt a Laissez-faire

approach, leaving market forces to try and reach an equilibrate and efficient state within the

payment market. The remaining two are government regulation, and intervention by competition

authorities, which can sometimes overlap in both procedures and outcomes. For this reason, and for

the nature of this paper which investigates an antitrust case and its effects of on public policy, I will

focus on the actions that competition authorities can take to resolve the possible market failure

caused by interchange fees.

Suppose that the European Commission has very strong evidence that the current way in which

interchange fees are set can harm and distort competition, and hence negatively affect consumers.

The first conceivable solution to this problem is to eliminate multilateral interchange fees and allow

individual banks to stipulate bilateral agreements between them. Whilst, on one hand, this could be

seen a plausible way to enhance the level of competition within the payment market and resolve the

issue, there have also been many critics to it. It is easy to imagine the huge time and financial costs

that an individual bank would have to face if it had to stipulate bilateral agreements with each and

every single other bank involved with payment systems such as MasterCard and Visa, which include

thousands of licensees. Not only this would shrink the size of the industry, at least in the short run,

but it would also create an issue for cardholders if they were to deal with a merchant whose

acquirer has not yet made any sort of agreement with their issuing bank. The elimination of

multilateral interchange fees in favour of bilateral agreement could also generate a free-riding and


hold-up problem. Suppose that all banks associated with Visa or MasterCard were let free to

stipulate bilateral agreements between them. If one of the banks was to set its interchange fee level

really high compared to that of the other banks, it could oblige merchants to accept that by

enforcing the "honour all cards" rule. In fact, If merchants were to decide not to accept that one

bank's interchange fee level, and refuse any card which carries their brand, they would have to turn

down all other Visa or MasterCard branded card. This would cause huge profit losses for merchants,

and a much lower level of card transaction, which would harm social welfare. Finally, many

researchers have proved that there is no direct link between the level of competition within a

market and the height of interchange fees. Weiner and Wright for instance, analysed cross-sectional

data for 20 countries, and found that interchange fees were set at pretty much the same level in

countries where Visa or MasterCard were the dominating platform, as in those were the two

networks shared equal portions of the market share26.

Another way in which the European commission and other competition authorities can deal with the

market failure created by interchange fees is by setting or capping their level. This strategy has been

implemented many times including during the case which I'm analysing, where Visa was forced to

reduce its level of interchange fees according to the tourist test.

Some people would affirm that the distortion will remain and affect the payment market as long as

interchange fees are higher than zero. However, research has proved that, due to the existence of

network externalities, both merchants and cardholders can benefit from a positive interchange fee,

if set at the right level. Figure 9 shows that the benefits arsing from a positive interchange fee

increase, for both merchants and cardholders, with its level up to the point Qmax. After that

threshold, if the level of interchange fee keeps increasing, both groups of end-users will be penalised

since merchants might decided not to accept card payments, stopping cardholders from using cards

as much as they used to.

26 Weiner and Wright (2005)


Therefore the issue lays on

determining the level of

interchange fee which would

allow payment markets to

reach a socially optimal status. Although the vast amount of literature and research that has been

conducted on this topic, a level of interchange fees which can guarantee equilibrium in the payment

market has not yet been found. As well as that, there is no proof that the public sector could set

interchange fees at a level which would be better for the community than the one determined by

the private sector. Interchange fees in fact, aim at reaching two conflicting objectives. The first one

is to send the right price signal to cardholders in order to get them to use card. The second one is to

send the right price signal to merchants, in order to get them to accept cards. However, in order to

achieve such targets, the cost faced by the end-users of a card transaction must not exceed the

benefits that they enjoy. The trade-off between these objectives is imposed by the difference in the

benefits that cardholders and merchants gain by trading electronically. If interchange fees were to

be set equal to the benefits enjoyed by cardholders that might impose a negative externality on

merchants, who would then face a very high cost and might not be able to recoup it by extracting a

bigger part of the consumer surplus from customers. On the other hand, if interchange fees where

set equal to the convenience benefits gained by merchants, issuers could argue that this might not

cover the marginal cost they incur to provide cardholders with enough benefits to incentivise card

usage, which as I've said many times enhances the gains of card-accepting merchants also.

Surcharging could solve this determination problem by introducing the neutrality effect. Gans and

King, whom provided the most general version of the neutrality phenomena, said that merchants

would earn the same net mark-up on a sale, regardless of the method of payment used, if they were

allowed to surcharge customers who pay by card an amount equal to the merchant service charge,

Figure 5


minus any convenience benefit for card-acceptance27. This way the level of interchange fees, which

through the merchant service charge inflates merchants costs, would be completely irrelevant, since

it would be offset by the surcharge applied. To quite Vickers: "The interchange fee would not matter for

practical purposes, and no one would have reason to care about its level. What one paid on the swings one

would receive back on the roundabout"28.

Therefore, another decision that competition authorities could take to eliminate the distortion

which interchange fees might create, is to ban the "no-surcharging" rule. In reality merchants'

behaviour with regards to surcharging is quite despaired . Surcharging is already allowed in a

multitude of countries including Australia, Czech Republic, Denmark, Ireland, Hungary, Netherlands,

New Zealand and the U.K., and yet in none of this countries this practice is widespread among

merchants. A reason for this is that surcharging can be seen as very significant from a customer

point of view, whilst an increase in consumer price might not be as evident. However, where

surcharging is not seen as an expensive add-on by consumers, merchants can sometimes surcharge a

price much higher than the merchant service charge, and rely on the fact that, if a purchaser is

unaware about the surcharging policy, it will be harder for him to terminate the transaction or use

another mean of payment once at the cashier. Hence, it can be stated that often, merchants who

can legally surcharge, do not behave according to the neutrality theory. This makes them not-

indifferent to the level of interchange fees. In reality however, the limited evidence available on this

matter shows that in the above mentioned countries, the parties involved within payment systems

do not seem to care about interchange fees. 29

27 Gans, J. and S. King. 2003 28 29 Rochet and Tirole (2007)


Conclusion

To conclude my research I can now attempt to answer to the questions presented in the

introduction of this paper.

Did the Commission utilise fair and efficient measures in its investigation against MasterCard and

Visa Interchange Fees?

I personally believe that the European Commission is correct in giving so much attention to

interchange fees and payment markets. Cards are used to substitute cash more and more often

every day, and it is important to make sure that such an important market works in the most

efficient way possible in order to preserve consumers. That said, I think the judgement of some

regulators and agents involved with this case might be clouded by the original status of associations

like MasterCard and Visa, which, before going public, were originally owned by banks. The fear that

MasterCard and Visa might still try to maximise the profit of their licensed banks through a higher

interchange fee, could in fact push regulators to try to implement anything that would lower

thecurrent level of interchange fees, regardless of the uncertainty to whether this can actually

increase social welfare.

Are the commitments taken by the two firms enough to reach social efficiency within the payment

market?

After surveying the literature regarding interchange fees I can say that we do not have enough

information to answer this questions. Both firms' interchange fees have currently been set in

accordance to the level suggested by the tourist set which, although being a powerful and reliable

benchmark, still leaves room for questions when competition in the market is not perfect like Baxter

assumed. However I believe that by agreeing to release more information about payment systems

and interchange fees, MasterCard and Visa have taken a big step forward towards increasing


efficiency and competition within the payment market, allowing cardholders and merchants to take

more informed decisions.

In there still room for market failure and negative externalities within the payment market? And if

so, can regulation improve the current situation?

Economic theory points out the need for the existence of interchange fees as it does for the

possibility that the same instrument might generate distortions within the payment market. Whilst

this might be interpreted as a sign for the necessity to regulate payment markets, the difficulties in

estimating the size and the impact that these distortions can have, as well as the asymmetries in the

way issuers and acquirer compete, make it difficult to provide a valid alternative to interchange fees.

Economic theory in fact, does not back up any claim that the public sector could either provide a

better balancing tool for open payment systems, nor that it could set interchange fees at a level

more optimal than the privately determined one. Therefore, whilst it is true that interchange fees

can have a distortive effect, intervention by public and competition authorities might not be the best

way to increase efficiency within payment markets. As far as I'm concerned I believe that a

substantial amount of R&D should be conducted before taking any action. Technological

development of payment markets has in fact drifted towards more mobile and cloud-based payment

systems, which drastically lower the costs incurred by the parties involved with an electronic

transaction, and in some cases do not require the use of interchange fees. Ernest Doku, in an

interview to the Guardian about the soon-to-com-out payment application "Zapp" said: "The days of

cash and debit cards could be numbered. The ease of managing everything – your diary to online payments –

from your Smartphone with mobile payment technology, could see the wallet become a thing of the past."30 If

the telecom expert from uSwitch.com is correct, any form regulation which applies to interchange

fees might not be needed anymore very shortly, and the cost and risks involved with it might turn

out to be unnecessary.

30 Harriet Meyer, the Guardian (2014)


Is there need for a full elimination of Interchange Fees?

In my opinion eliminating interchange fees would be a huge mistake which could badly affect social

welfare. The lower cost for merchants would be counteracted by a higher cost for cardholders whom,

depending on their elasticity of demand for transactional services, might reduce the usage of credit

or debit cards. Revenues for acquirers would then drop, and card-accepting merchants might have

to pay a higher price to enjoy the same level of convenience benefits. As well as that there are other

costs and negative externality that the elimination of interchange fees would impose. These include,

for instance, the cost of stipulating bilateral interchange fees, and the possibility of free-riding which

some banks might exploit.

Overall, the current situation of the payment market, which shows constant growth and a high level

of innovation, does not grant for the implementation of regulatory measures which, as stated in this

research, might or might not improve the current situation of this sector. For this reason the

implementation of a Laissez-faire method could be very beneficial. Contemporarily, more research

should be conducted on topic such as surcharging, the relation between the benefits and costs of

end-users of the payment process, and competition among issuers and acquirers, in order to reduce

the uncertainty related to this sector. For the time being, I believe that the tourist test, which makes

cardholders internalise the externality imposed on merchants, is the most reliable benchmark for

the level of interchange fee, and should be used to make sure that the exploitation of market power

does not drive the level of interchange fee too high or low.


Bibliography

William F. Baxter. (1893). Bank Interchange of Transactional Paper: Legal and Economic

Perspectives. Journal of Law and Economics. 26 (3), 541-588.

Christoph Baert. (2012). Cross-border Interchange Fees: Why the General Court Got it

Wrong in the MasterCard v. Commission Case. European Payment Council Newsletter. (Issue

16). Available at:

http://www.europeanpaymentscouncil.eu/article.cfm?articles_uuid=567705BA-5056-B741-

DB26318F233469A0. (Accessed: 5/12/2013).

European Commission (2007) 'IP/07/1959'. Available at: http://europa.eu/rapid/press-

release_IP-07-1959_en.htm. (Accessed: 5/12/2013).

European Commission (2007) 'MEMO/07/590'. Available at: http://europa.eu/rapid/press-

release_MEMO-07-590_en.htm. (Accessed: 9/12/2013).

European Commission (2006), 'MEMO/06/260' . Available at: http://europa.eu/rapid/press-

release_MEMO-06-260_en.htm?locale=fr. (Accessed: 9/12/2013).

European commission (2007), 'IP/07/1959'. Available at: http://europa.eu/rapid/press-


European Commission (2009), 'IP/09/515'. Available at: http://europa.eu/rapid/press-


European Commission (2008), 'Memo/08/170'. Available at: http://europa.eu/rapid/press-

release_MEMO-08-170_en.htm?locale=en. (Accessed on 10/12/2012).

European Commission (2010), 'Memo10/224'. Available at: http://europa.eu/rapid/press-

release_MEMO-10-224_en.htm. (Accessed on 10/12/2012).

European Commission Official Journal 115 , 09/05/2008 P. 0088 - 0089. Available at:

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:12008E101:EN:HTML


Harriet Meyer.(2014). Zapp app to enable millions more shoppers to pay by Smartphone.

The Guardian. Available at: http://www.theguardian.com/business/2014/jan/15/zapp-app-

millions-shoppers-pay-smartphone. (Accessed: 07/04/2014)

Jean-Charles Rochet and Jean Tirole. (2007). Must-Take Cards and the Tourist Test. DNB

Working Papers. 127 (1).

Jean Tirole. (2011). Payment Card Regulation and the Use of Economic Analysis in Antitrust.

CPI Journal . 4 (1).

Joshua S. Gans and Stephen P. King. (2003). The Neutrality of Interchange Fees in Payment

Systems. The B.E. Journal of Economic Analysis & Policy. 3 (1).

Marc Rysman and Julian Wright.(2012). The Economics of Payment Cards. Working papers

series.

Motta Massimo.(2007). Competition policy: theory and practice.Cambridge University Press.

National Bancard Corporation v. Visa U.S.A. Inc. (1982) 'Defendant trial brief' Southern

District of Florida.

Richard Schmalensee and David S. Evans . (2005). The Economics of Interchange Fees and

their Regulation: An Overview. MIT Sloan Working Paper. 454805 (5).

Richard Schmalensee and David S. Evans . (2007). Industrial Organization of Markets with

Two-Sided Platforms. Competition Policy International. 3 (1).

Richard Schmalensee. (2002). Payment Systems and Interchange Fees. Journal of Industrial

Economics. 50 (2).

Slaughter and May (2012) 'The General Court confirms the Commission decision prohibiting

Interchange Fees applied by MasterCard ', EU Competition and Regulatory, (22), pp. 1.

Available at:http://www.slaughterandmay.com/media/1982329/eu-competition-and-

regulatory-newsletter-25-may-31-may-2012.pdf. (Accessed: 5/12/2013).

Stuart E. Weiner and Julian Wright. (2005). Interchange Fees in Various Countries:

Developments and Determinants. Payments System Research Working Paper. 5 (1).


John Vickers. (2005). Public policy and the invisible price : competition law, regulation, and

the interchange fee. Proceedings – Payments System Research Conferences.

Completed Dissertation Project

Documents

Transcript of Completed Dissertation Project