Regional Banks Should Share Back Offices, Not Owners

Scale should be something a regional bank can buy without being bought. Fifth Third’s proposed all-stock acquisition of Comerica, valued at $10.9 billion when announced on October 6, would create a bank with about $288 billion in assets.^[1] It still requires shareholder and regulatory approvals. The Office of the Comptroller of the Currency^[2] withdrew its 2024 merger-policy statement in May. Washington has eased the approval problem, not the cost problem. I favor technology pacts that share compliance systems while each bank keeps its own board and lending decisions.

FDIC-insured commercial banks and savings institutions fell from 8,975 at year-end 2004 to 4,487 at year-end 2024, a 50% decline.^[3][4] The last five-year interval was slower than either of the preceding two. These are institution counts, not a measure of competition in any particular town. The FDIC’s “2020 Community Banking Study” found voluntary mergers were the primary cause of the decline from 2012 to 2019 and judged small banks’ compliance function “likely a factor contributing to scale economies.”^[5] The burden differs sharply by size. In a 2015 survey of 2014 operations analyzed by the St. Louis Fed, compliance averaged 8.7% of noninterest expense at banks below $100 million in assets, against 2.9% at those with $1 billion to $10 billion.^[6]

FDIC-insured institutions, year-end 2004 to 2024

Regional banks also face regulatory steps. The 2018 banking law raised the automatic threshold for enhanced prudential standards to $250 billion for bank holding companies while retaining Fed discretion from $100 billion.^[8] The 2019 tailoring rules impose capital planning and liquidity stress tests from $100 billion, with stricter requirements at $250 billion or other risk thresholds.^[9] That encourages either staying below a tier or growing far enough beyond it to spread the work. Comerica’s reported $77.4 billion in assets at September 30 was well below $100 billion;^[10] the rules alone cannot explain this sale. Size brings another advantage. After Silicon Valley Bank failed in 2023, the Fed recorded substantial deposit inflows at the largest banks.^[11] When depositors expect a rescue, size attracts funding while weakening their incentive to monitor risk.^[12] That is a subsidy to size and a moral hazard, not a reward for better lending.

Banks already share technology by renting it. The FDIC found that 94% of community banks in the 2019 survey it studied used outside digital-banking providers.^{[13, p. 6-7]} A merger can put one management team in charge of investment and failures, while a consortium can turn either into a negotiation. Shared systems also leave capital requirements and depositor fears intact. I would not preserve a weak bank just to preserve the institution count. But renting a vendor’s service and co-owning a system are different bargains: joint owners can set the development budget and priorities, rather than accepting a supplier’s pace. The pact has to beat renting on cost and control, not merely keep the old signs above the doors.

The Fed, FDIC and OCC joined 2018 guidance on sharing anti-money-laundering resources, aimed chiefly at community-focused banks with simpler operations and lower money-laundering risk.^[14] It allows pooling people and technology while keeping each bank responsible for compliance. That is a starting point for smaller banks, not a ready-made alternative to the Fifth Third deal. The three agencies should apply this model to mortgage compliance and regulatory reporting, explicitly addressing regional-bank consortia. Members should keep their own boards, prices and credit decisions; supervisors should examine both the shared service and each bank’s use of it. Start with limited functions, measure the savings and keep the option to leave. I would not set a target for how many banks America ought to have. A business refused credit by one should still have another to ask.