A bank-merger moratorium isn’t just bad policy. It’s illegal.

As 2021 ended, several policymakers were calling for a moratorium on bank mergers above a certain size — uniformly, when the resulting firm’s assets would exceed $100 billion. Such a moratorium would be extremely poor policy. But of most immediate importance, it would be flatly illegal.

If effective, such a moratorium would have far-reaching effects. There are today 32 U.S. banking organizations that hold over $100 billion in assets. As a practical matter, a moratorium would have no consequence for a bank that holds more than 10% of nationwide deposits in view of the statutory ban on interstate bank acquisitions for banks at that deposit level. But a $100 billion-asset restriction would bar the remaining of these organizations (which today is nearly all of these banking organizations) from any bank acquisitions, no matter how small. In addition, six banking organizations hold assets between $75 billion and $100 billion; they would also be barred from significant acquisitions. Another five banking organizations have assets of between $50 billion and $75 billion, and they would be barred from any “mergers of equals.”

Such a moratorium would raise serious policy concerns, as it would prevent small, midsize and regional banks from fighting to remain competitive, as scale in banking has become all the more important given rising costs of digitization, cybersecurity and regulatory compliance. But the policy demerits of such a proposal should not overshadow the fact that its adoption would be legally void.

To wit, Section 3 of the Bank Holding Company Act was amended in 1970 to provide: “In the event of the failure of the [Federal Reserve] Board to act on any application for approval under this section within the ninety-one-day period which begins on the date of submission to the Board of the complete record on that application, the application shall be deemed to have been granted.”

Notably, the 91-day rule was adopted because the Fed had been “unduly slow in rendering decisions on applications made.” Thus, Congress took the extraordinary step of not just requiring the central bank to act within 90 days but deeming applications to be approved automatically if it failed to do so.

As one court observed in explaining this remarkable remedy, “it may well be a lesser evil as a matter of public policy for a nonmeritorious application to be deemed approved by law than to risk allowing a meritorious application to be delayed by federal bureaucracy for more than 91 days.” Thus, Congress’s reasoning in adopting this provision is precisely antithetical to the reasoning of moratorium proponents, who believe that as a matter of public policy it would be a lesser evil for innumerable meritorious applications to be delayed than for a nonmeritorious application to be approved.

In any event, as a matter of law, a decision by the Fed to stop considering applications actually would result in each pending application being approved at the 91-day mark.

Of course, a relevant question is when the 91-day period begins to run — that is, when the record is complete. Could the Fed decide that the record is not complete until the applicant answered an unending series of immaterial questions, or calculated pi to an endpoint, or explained why people find "The Big Bang Theory" funny, or the like?

The courts proved wise to that game in a series of decisions in the 1970s and early 1980s. As one stated, “[the 91-day rule] would have no meaning if the [Fed] had the power to determine when the time had begun to run.” In 1984, the Fed amended its Regulation Y to incorporate recent case law and acknowledged that the record need only include information material to the board’s decision.

Moreover, the courts have been clear that the 91-day period begins once the bank has done its work (filing the application with the required documentation) and the public comment period has closed, not when the Fed or its staff are finished with their work. As one court explained, “[S]taff reports and recommendations of the (Board) are no[t] part of the ‘complete record’ within the statutory contemplation. ... (T)he Congressional intent here was to define ‘complete record’ as meaning the facts upon which the agency would pass, not the internal documents relating to the process of that passing.” Thus, the Fed could not justify a moratorium on a desire to revisit its standards for reviewing merger applications.

The story with regard to the Bank Merger Act is similar to the Bank Holding Company Act, albeit with a longer gestation period. Thus, the Riegle Community Development and Regulatory Improvement Act, enacted in 1994, provides: “Each Federal banking agency shall take final action on any application to the agency before the end of the 1-year period beginning on the date on which a completed application is received by the agency.”

There is no legislative history or case law construing this provision, but Congress is legally presumed to be aware of similar statutory language and to have utilized different language purposefully. Thus, the one-year period must begin on a date that is earlier than the 91-day period in the BHCA because the Section 4807 initial date is the date on which the application is completed as opposed to the Section 3(b) date on which the record on the application is complete.

Of course, as all the law and policy is debated, banks continue to lose market share to fintechs, payday lenders and mega-techs that are not subject to capital requirements, liquidity requirements, constant examination and ever-growing compliance and community reinvestment obligations. Such firms must revel in the thought that their banking competitors could be barred from banding together to absorb the costs from which they are immune — laughing all the way to the nonbank.

Fortunately, the law is otherwise.