The Dodd-Frank financial reform legislation required the Federal Reserve Board to regulate debit card merchant fees. Over the past year, the FED has taken comments, issued proposals, delayed the announcement of a final recommendation, and unsuccessfully sought to change or delay this difficult task. This week, the Board adopted a staff recommendation that is in the truest sense a compromise that like most successful compromises is unlikely to please any of the players involved. The result is also an example of administrative law making at its most creative, interpreting the statutory language to most effectively achieve Congressional intent while minimizing the economic risks that the Act's proponents likely did not understand or anticipate.
Section 920(a)(2) of the Act required the FED to limit debit card interchange -- the portion of merchant card acceptance fees that are paid to the card issuer -- to an amount that "shall be reasonable and proportional to the cost incurred by the issuer with respect to the transaction."
A fee meeting this standard, the Act asserted, should include "the incremental cost incurred by an issuer for the role of the issuer in the authorization, clearance, or settlement of a particular electronic debit transaction," but should not include "other costs incurred by any issuer which are not specific to a particular electronic debit transaction."
Early on the FED staff concluded that it should set the fee based on costs for a representative issuer and transaction. Basing fees on the actual costs of particular issuers would impose undue compliance burdens. The Board decided to set a fee cap based on the average per-transaction cost, excluding fraud losses, of the issuer at the 80th percentile based on a survey of the large banks covered by the statute. Each issuer would be permitted to receive interchange fees not exceeding the cap without demonstrating its actual per transaction costs.
Initially, the staff read the statute to permit the Board to take account only of variable per transaction costs. That led to the initially proposed fee cap of $.12 per transaction. Issuers objected that fees at that level would not come close to covering their actual costs of operating a debit card system. The staff responded to those concerns by creatively reinterpreting the statute to include a third type of cost that was neither an "incremental cost" of a debit transaction, nor a cost of the system that was "not specific to a particular transaction." This third type of cost, the staff reasoned, consisted of fixed costs of a debit program that are nonetheless specific to particular transactions. And since the statute did not explicitly require or prohibit including these costs in the regulated fee, the Board could exercise its discretion.
The staff concluded that prohibited costs of a debit system included corporate overhead (e.g., executive compensation, human resources, the issuer's branch network); establishing account relationships; general debit program costs (e.g., production and delivery); marketing; research and development; and network membership fees. Conversely, non-incremental costs that are specific to particular debit transactions -- and that could thus be included in the regulated fee at the Board's discretion -- included (1) network connectivity; (2) software and hardware for processing transactions; (3) operational labor; (4) network processing fees; (5) transaction monitoring costs; (6) reward programs; (7) handling cardholder inquires; and (8) non-sufficient funds handling.
Ultimately, the FED adopted a regulation including the first five of these costs, but not the last three, resulting in a cap of $.24 per transaction. In addition, the FED determined that transaction monitoring as a means of fraud protection also fell within the discretionary category of fixed costs attributable to specific transactions. Because losses vary with the size of the transaction, the regulation permits a fee of up to 5 basis points on top of the $.24 flat fee per transaction.
Finally with respect to fees, the legislation permitted an adjustment for investments in fraud prevention, and the Board included a 1 cent per transaction bonus for issuers that meet the FED's standards to help offset the costs of implementing activities that are effective at reducing fraud looses.
The legislation also required the FED to adopt regulations prohibiting network exclusivity so that merchants had choices with respect to the network on which to process a transaction. The FED required that each card provide access to at least two unaffiliated networks, but it did not require that each card link to multiple networks for both PIN and signature access.
The FED also excluded from the regulation three-party networks that did not have separate entities issuing cards and signing up merchants. American Express, for example, is not covered by the fee limit.
The fee regulation provisions are to go into effect on October 1, 2011, followed by the network exclusivity rules on April 1, 2012. You can read the Board's full report on the regulations here.