Thursday, March 20, 2008
On the VISA IPO
Adam Levitin, Associate Professor of Law at Georgetown University, posts on the Credit Slips blog about this week's scheduled Visa Inc. initial public offering. This is must read stuff for payment geeks. He writes: "The Visa IPO, along with the 2006 MasterCard IPO and the end to MasterCard and Visa's dual-exclusivity rules...is setting the stage for a major reconfiguration of the payments world in the next decade. These changes could have far-reaching effects for consumers, merchants, and banks because of potential shifts in the way payment networks will compete with each other." The full post here; "The Visa IPO" posted by Adam Levitin; "Visa is scheduled to have its IPO later this week. The IPO could potentially be the largest in history. This market seems like terrible timing for an IPO, although for liquidity-strapped banks, the IPO could be a much-needed source of cash. The Visa IPO, along with the 2006 MasterCard IPO and the end to MasterCard and Visa's dual-exclusivity rules, which prohibited banks that issued MC/Visa cards from issuing Amex or Discover cards, is setting the stage for a major reconfiguration of the payments world in the next decade. These changes could have far-reaching effects for consumers, merchants, and banks because of potential shifts in the way payment networks will compete with each other. A major factor driving Visa’s IPO is antitrust liability. It is Visa's litigation exposure, including a not just the $2.1 billion antitrust settlement with American Express for the dual exclusivity rule, still unresolved liability to Discover for the same behavior, and, on a much larger scale, the potential antitrust liability connected to interchange and merchant restraint litigation and the possibility of legislative changes to interchange. Visa has a lot to fear from changes to interchange. Although Visa does not itself collect interchange revenue, much of its raison d’être is related to its role in coordinating the interchange rate and rules for its member banks. Take away this coordinating role, and Visa is basically a really big computer clearing transactions. That's its core business line. Anyone with sufficient capital could get into that game. (First Data Corp. tried to a few years ago, only to get pushed out of it by litigation from Visa.). Other than coordination, there just isn’t much value added to the clearing role for banks, merchants, or cardholders. Recent developments on interchange and merchant restraints don’t bode well for Visa. The European Union, Australia, Brazil, Colombia, Germany, Honduras, Hungary, Israel, Mexico, New Zealand, Norway, Poland, Portugal, Romania, Singapore, South Africa, Spain, Sweden, Switzerland and the United Kingdom have all moved against or begun to regulate interchange fees and the merchant restraints that shield interchange from normal market pressures. US regulators have been behind the curve on this, but there’s now interchange legislation pending in several states (none of its particularly aggressive—at worst prohibiting interchange on sales tax) and a bipartisan bill was recently proposed in the House that would create as special administrative law judge panel to oversee interchange rate arbitration and set rates if necessary. The Visa IPO creates a company with an unusual 3 stock-class capital structure that can only be explained as an antitrust shelter. Like MasterCard’s post-IPO structure, the Visa post-IPO capital structure is designed to permit banks to retain effective control over the company without holding a majority of shares and giving a veneer of independence to decisions about the setting of the interchange rate and network rules, in short plausible antitrust deniability. The structure also reduces banks' antitrust exposure going forward, albeit at the cost of some of Visa's revenues, but as a mutual-type member association, Visa was not designed to be a profit center in the first place. Visa’s IPO also raises the question of who needs Visa. Consumers and merchants don’t. Indeed, consumers and merchants have virtually no relationship with Visa itself; only banks do. (I say “virtually” because of the existence of small programs like Visa concierge for high-end cardholders.) The reason Visa exists in the first place is because of the state of federal banking regulations in the 1960s and 70s. These regulations prevented multi-state branch banking. This meant that if banks wanted to form a national, private payment system that cleared at par, they would need to form a multi-bank association. (The Fed has to clear checks from member banks at par, 12 U.S.C. § 360, which has led to virtually all checks clearing at par today.) First Bank of America attempted to accomplish this through licensing and franchising, but soon it converted to the free-standing multi-bank association that became Visa. The current interchange system, in which acquirer banks pay issuer banks on each transaction was one (but only one) of the possible ways of encouraging banks to join the system and allocating transaction costs within the system. (Issuers could just have easily paid acquirers or no revenue whatsoever. In any case, there was never a historical need for interchange fees to vary by type of merchant and merchant transaction volume, except to maximize on merchants’ price elasticities.) The historical state of federal banking regulation also explains the origins of many of the merchant restraint rules, such as honor-all-cards and no-surcharge rules—these were needed to insure par clearance of all member banks’ cards. Fast forward to today. Federal law no longer prohibits multi-state branch banking. We now have several banks with truly national presence: Bank of America, JPMorgan Chase, Citi, Washington Mutual, Wachovia, Wells Fargo, to name a few. We also have national monoline card issuers, like Capital One, whose card business is separate from depositary banking (yes, CapOne has gotten into the deposit business recently, but the monoline card business is still its core, and its new decoupled debit card shows that monoline is possible not just for credit, but for debit too.) These banks dominate the credit card industry and do not need national associations like Visa in order to ensure par clearance. Many of them could strike out on their own and be stand-alone networks like Discover and American Express. To be sure, smaller banks still benefit from multi-bank networks like Visa, and for all banks, international transactions present certain some of the par clearance problems that interstate transactions used to pose. But it is far from clear what JPMorgan or BoA gain from Visa—they end up subordinating their brand to Visa’s and functionally give a free leg up to their smaller competitors.
The end of dual exclusivity has created significant inter-network competition for issuers. (This has a negative impact on consumers and merchants, unfortunately, because the networks compete in part by offering issuers higher interchange.) The ability for banks to shift their card portfolios between networks has forced networks to be more responsive to issuers’ concerns, which may curtail the demand for new network possibilities. But it has also weakened bank loyalty to networks—networks have become simply suppliers of clearing services to issuers, and nothing more. And if a bank (or anyone else) can do the clearing more cheaply than the existing networks, we can expect to see a new clearing service develop. Freely tradable Visa shares provide a lot of flexibility to member banks. They can stick around in the association and keep doing business as usual, but with a potentially reduced antitrust exposure going forward (the precise allocation of liability for past behavior among Visa and its member banks is not explained in the S-1 registration statement). Alternatively, they can decrease their relationship with Visa by selling shares. This makes it much easier for a large bank to exit Visa without taking a loss; there’s now a market for the shares. The IPO also makes it more feasible for member banks to create their own competing networks with Visa. Just as Ford could buy a limited number of GM shares without antitrust concern, so too could BofA own Visa shares and start up its own network. And the enhanced exit possibilities also increase the potential value of Visa for those banks that would remain. If many of the large banks left Visa, the power of those that remained would be relatively larger. What would the world look like if banks started breaking off from the big 4 networks and becoming stand-alones? There's a whole post or two to write about that, but here's one thing to consider: the basis for competition in the payments field might switch. Payments networks need to balance demand from merchants, consumers, and banks. Currently the dominant strategy is to cater to banks, which means catering to consumers in the form of rewards that are extracted from merchant fees. But if a network isn't competing to sign up issuer banks, perhaps the incentives change. This might lead to the development of real value-added services for merchants (data mining, e.g.) or to more meaningful product differentiation (not just variations in rewards programs) for consumers. In short, shaking up the structure of the payments field might encourage payment companies to do a little more thinking outside the box. (Enough of this faux consultant-speak, though.) We should not expect to see changes in card networks happen immediately. For one thing, there is a 3-year lock-up of the stock classes received by the banks in the IPO. But over the next 5 years, I would not be surprised if we saw a major reshaping of the payment card clearance landscape.