V, MA, DFS, AXP: Court Finding Signals End of Visa/MasterCard Duopoly in Signature Debit
SEE LAST PAGE OF THIS REPORT Howard Mason
FOR IMPORTANT DISCLOSURES 203.901.1635
August 14, 2013
V, MA, DFS, AXP: Court Finding Signals End of Visa/MasterCard Duopoly in Signature Debit
- Today, there is a follow-up hearing to the Federal Court finding of July 31st that the Fed “clearly disregarded Congress’ statutory intent by inappropriately inflating all debit-card transaction fees by billions of dollars and failing to provide merchants with multiple unaffiliated networks for each debit-card transaction”. In due course, we expect:
- The debit interchange cap for regulated issuers (i.e. those with more than $10 billion in assets) to be reduced to 12 cents/transaction from the current 21 cents (in both cases excluding add-ons). These issuers will see a 40% decline in debit interchange revenue which contributes ~30% of fees on retail checking accounts.
- A requirement for unaffiliated networks for signature-authenticated debit transactions. This will weaken the competitive advantage Visa gains from strategic relationships with key issuers creating a share-shift opportunity for MasterCard and market-entry opportunity for Discover and American Express.
- In signature debit, a requirement for unaffiliated networks will transform the competitive landscape, and create a strategic incentive for issuers to break the Visa/MasterCard duopoly by working with Discover and American Express. We believe this explains the timing of WFC’s recent decision to work with American Express (albeit only on credit cards, for now).
- If Visa and MasterCard are the only two viable networks, and issuers have to make both available under Durbin, then issuers will lose the ability to discipline interchange (by threatening to switch to another network); the consequence will be competitive declines in interchange as the networks look to be the preferred routing of merchants, and hence deeper impairment of checking-account economics.
- In PIN-debit, issuers can discipline networks because there are more than two of them; hence an issuer has an alternative for Durbin compliance if a network too aggressively cuts interchange. We believe this is why competitive, as opposed to regulatory, declines in PIN-debit interchange have been relatively modest (e.g. from 33 cents/transaction to 29 cents for exempt issuers on STAR).
- The issue is moot if the merchant does not have a routing choice as is the case for MasterCard-branded debit cards with an exclusive PIN-debit network. However, because Visa has enabled its signature-debit network for PIN-authenticated transactions, merchants have a PIN routing choice on all Visa-branded debit cards.
- Given limited ability to cut interchange, networks have competed for merchant-routing by making large “incentive” payments. Visa has defrayed the cost with the Fixed Acquirer Network Access Fee (“FANF”) which is under investigation by the Justice Department.
On July 31st, Visa traded down 8% on a Court finding (Judge Richard Leon of the US District Court of Columbia) that the Fed “clearly disregarded Congress’ statutory intent by inappropriately inflating all debit-card transaction fees by billions of dollars and failing to provide merchants with multiple unaffiliated networks for each debit-card transaction”. Debit card transactions are those made on a card linked to your checking account as opposed to credit card transactions where the card is linked to a pre-approved line of credit. Today, Wednesday August 12th, there is a follow-up Court hearing.
Debit Interchange Cap Affects Unit Economics of Checking Accounts
The first impact of the Durbin Amendment was that on October 1st, 2011 a cap on debit “interchange” went into effect: interchange is a payment made by the “acquiring” bank representing a merchant in a card transaction to the “issuing” bank representing the cardholder and administering the payment account. This cap was set at 21 cents/transaction (plus add-ons of 0.05% of transaction value plus a conditional 1 cents for fraud prevention) to 12 cents/transaction plus the add-ons. As a result average debit interchange/transaction for regulated issuers (i.e. those with more than $10 billion in assets) fell to 24 cents including add-ons versus the prior level of 51 cents (see Appendix); and total debit interchange revenue for the banking system fell 24% to $15 billion in 2012 from over $20 billion in 2011 (see Exhibit 1).
Exhibit 1: Debit Interchange Fees and Purchase Volume
Source: Federal Reserve Board and Nilson July 2013 for debit purchase volume
Given Judge Leon’s finding that the Fed set the debit interchange cap too high, it will likely fall to 12 cents (plus add-ons) leading to a decline of nearly 40% in debit interchange revenue for regulated issuers. Such a decline will have an important impact on the economics of checking accounts. On average, debit interchange generates just over $50 per checking account (see Exhibit 2), second only to overdraft fees at just over $90, and 30% of total fees of $160. A 40% decline in debit interchange will therefore reduce total checking account fees by 12.5%
Exhibit 2: Average Fees per US Checking Account
% on average checking account balance of $5,600
Impact of Network Exclusivity on Competitive Dynamics
The second impact of the Durbin Amendment was that on April 1st, 2012 all debit cards were required to have an unaffiliated network for at least one type of transaction authentication; this requires some explanation. A cardholder may authenticate a debit card transactions in one of two ways: by a signature (in which case the transaction is processed over the same Visa or MasterCard infrastructure used to process credit card transactions); or by a personal identification number or “PIN” (in which case the transaction is processed over the same infrastructure provided by electronic funds transfer or “EFT” systems used to process ATM transactions). Examples of EFT systems include STAR (owned by First Data), NYCE (owned by Fidelity Information Services), Pulse (owned by Discover) and the two EFT systems owned by Visa and MasterCard: Interlink and Maestro respectively.
This network exclusivity provision of the Durbin Amendment impacted Visa because most of the cards from its top signature-debit issuers (see Exhibit 3) were PIN-enabled only for Interlink. For these debit cards, there were three options for meeting the Durbin requirement:
- Change signature network to MasterCard so that signature authentication would be provided by MasterCard and PIN authentication by the unaffiliated (because it is Visa-owned) Interlink network: This was an unattractive option for issuers because it would mean weakening a strategic relationship with Visa extending across both debit and credit cards, and including volume discounts.
- Add an unaffiliated PIN-debit network (such as STAR, Pulse, NYCE, or Maestro): This was an unattractive option for issuers because it would give merchants a “routing” choice over which network to use for authorization and clearance, and lead transactions to shift to the network with the lowest interchange. Furthermore, and this is important to the more general competitive dynamics of the payments business, it would create an incentive for the networks to lower interchange in order to influence the merchant routing decision: in short, there would be competition between networks on the card-face. Finally, it would disqualify issuers from the premium terms available from a network if it was the exclusive network on a card.
- Abandoning Interlink: This weakened the strategic relationship with Visa but not by as much as changing signature debit networks: at ~$100 billion, Interlink volumes were lower than the over $1 billion for Visa signature transactions.
In practice, many issuers abandoned Interlink so that, in the quarter after the Durbin network exclusivity provisions went into effect, Interlink’s debit volumes fell by 50% to $48 billion; and Visa’s overall debit volumes (across both signature- and PIN-authenticated transactions) grew only 1% in the 12-months through June 2012 versus 14% in the prior 12-months (see Exhibit 4).
Exhibit 4: Immediate Impact on Visa Debit Volume of Durbin Network Exclusivity Rules
Source: Visa Company Reports
Ironically, then, the Durbin Amendment meant that Visa’s strong position in signature-debit initially became a disadvantage in PIN-debit since Interlink cannot not be an exclusive PIN-debit network on a debit card using Visa signature-authentication. MasterCard was in a similar position vis-à-vis its Maestro PIN debit network, but Maestro carried substantially less purchase volume than Interlink. Shortly after Durbin took effect Visa addresses this with a remarkable strategic move: it enabled PIN-authenticated transactions to be handled over the VisaNet network (i.e. the network represented by the Visa logo on the front of the card and traditionally used for handling signature debit and credit transactions).
VisaNet has long been capable of handling PIN-authenticated transactions (such as a cash advance from a credit card at an ATM) but enabling it for PIN-authenticated debit transactions in April 2012 allowed Visa to compete for PIN-debit transactions on its signature (in both authentication and importance senses of the word) debit product even if the card did not carry the Interlink bug. The difference from the pre-Durbin era is that it now competed for these transactions with the PIN-debit network represented on the reverse of the card as opposed to having an exclusive position with Interlink.
In the process, of course, Visa created the situation issuers had sought to avoid: two networks competing on the card face for PIN-debit transactions. And, indeed, Visa undercut interchange on PIN-debit networks to attract transactions and some PIN-debit networks responded with price cuts of their own. Thus, average PIN-debit interchange/transaction on the STAR network for exempt issuers (i.e. those with less than $10 billion in assets and so not subject to the Durbin interchange cap) has fallen from 32 cents to 29 cents. While this is not a large decrease, it is significant in the context of the prior trajectory of increasing PIN-debit interchange (see Exhibit 5).
Exhibit 5: Before Durbin, Fees (largely interchange) on PIN-Debit was Converging Up to those on Signature Debit
At first glance, it may be surprising that interchange rates have not fallen further: after all, the networks are high fixed-cost and do not want to lose marginal transactions. Furthermore, cutting interchange adversely impact issuer economics not directly network economics since interchange is passed-through to the issuers while the networks get paid a “switch” fee (typically of 1-2 cents/transaction) for each transaction processed, or “switched” from acquiring to issuing bank, on the network. In practice, a PIN-debit network that was too aggressive in lowering interchange would likely find it difficult to retain issuer clients.
To provide additional inducements to merchants, and avoid upsetting issuer clients through interchange cuts, Visa also made “incentive” payments to merchants who committed to route PIN-debit transactions over VisaNet rather than the on-card PIN-debit network, just as Visa and other networks makes incentive payments to issuers who commit to using the network on their cards. The Fed reported that in 2012 these incentive payments shifted to merchants from issuers, although issuers still received the bulk of them: specifically, merchants received 31% of incentive payments made by Visa in 2012, up from 25% in 2011: in both years, the total of incentive payments across issuers and merchants was $1.2 billion.
Possible Impact of Recent Court Finding
On July 31st, a Federal Court ruled that the Federal Reserve Board had not gone far enough in its implementation of the network exclusivity provisions of the Durbin Amendment, and that congressional intent was that two unaffiliated networks be available for each debit transaction and not merely on each debit card. The distinction between these two standards is important.
A debit card carrying the Visa “logo” (on the front of the card) for signature-authenticated transactions and the STAR “bug” (on the reverse of the card) for PIN-authenticated transactions is compliant under the second standard (since there are two unaffiliated networks on the card) but not under the first standard (since there is only one network available for a signature-authenticated transaction). To bring the card into compliance with the first standard would mean adding MasterCard (or, perhaps, Discover and/or American Express) for signature-authenticated transactions so that the debit card carried at least two unaffiliated logos on the front. This would change the dynamics in the signature-debit business with networks competing not to be the exclusive network for an issuer but the preferred network of the merchants (or acquiring bank if the merchant delegated the routing decision to it). As a result, the strategic advantage Visa enjoys in the signature-debit business because of strategic relationships with leading issuers would be weakened creating the possibility of a share-shift to MasterCard which presently has only a 30% share of signature-debit volumes in the US (see Exhibit 6). As discussed below, an implementation of Durbin insisting that two unaffiliated networks be available for each signature-authenticated debit transactions also creates an entry opportunity for Discover and American Express.
Exhibit 6: US Signature Debit Volumes for Visa and MasterCard
Of course, the availability of more than one network on a card gives merchants a choice of where to route a signature-debit transaction and so creates network competition on the card face for signature-authenticated transactions, just as Visa’s enabling of VisaNet for PIN-debit created network competition on the card face for PIN-authenticated transactions. Under these circumstances, the tendency is for networks to engage in competitive reduction in interchange (so as acquire transactions from merchants who ultimately pay the interchange); the governor on this competition is the threat that issuers will cease to issue cards enabled for an overly aggressive network. Of course, this governor does not work if there are only two viable networks since issuer compliance with Durbin (as interpreted by Judge Leon) would require an issuer to enable both.
In PIN debit there are more than two viable networks so that an issuer can realistically shun a network that is overly aggressive with interchange cuts; in signature debit, however, the only two players to date have been Visa and MasterCard. The risk to issuers is that if each debit card must be authenticated for two signature-debit networks and if these are the only two networks then they will engage in destructive price competition (read: sharp cuts in interchange) in order to win the merchant routing decision. Without an alternative signature debit network, issuers will not have the ability to threaten a transfer of their business since Durbin requires them to use two networks.
The consequence is that price-stability for signature-debit interchange will likely depend on entry into the market of Discover and/or American Express.
Debit Interchange in $/transaction and as a % of transaction value
Source: Federal Reserve