The FDIC’s cleanest design still broke in March 2023

The slogan with an asterisk

The FDIC has one claim that is actually true: since it opened in 1934, no depositor has lost a penny of insured deposits.

On its own, that is a misleading way to describe March 2023, when three regional bank failures forced regulators to set aside the insurance limit the agency exists to enforce.

Of the government corporations worth comparing, the FDIC comes closest to the textbook version Congress imagined in the 1945 Government Corporation Control Act. It is funded by risk-based premiums on the industry it insures, takes no annual appropriations, and repaid its original Treasury seed capital instead of living on open-ended soft financing.

That design held for ninety years. Then three banks failed inside a few weeks, and the gap between “the FDIC works” and “the FDIC decides who actually gets protected” became hard to ignore.

The design that almost was the point

Congress created the FDIC in the Banking Act of 1933 after more than 9,000 banks failed between the 1929 crash and 1933, wiping out roughly $1.3 billion in depositor losses, by the FDIC’s own institutional history. The Treasury advanced about $289 million to start it. The agency later repaid that debt in full.

Since then it has funded itself through bank premiums and earnings on Treasury securities. The Minneapolis Fed has noted that it receives no congressional appropriations, which puts it closer than most government corporations to the self-sustaining entity the statute nominally requires.

That structure held through the savings-and-loan crisis and 2008. Then, on March 10, 2023, Silicon Valley Bank failed, followed within days by Signature Bank and, weeks later, First Republic. Respectively, these were the second, third, and fourth largest bank failures in U.S. history.

The Government Accountability Office (GAO) later found that the Federal Reserve and the FDIC had flagged concerns at SVB and Signature as far back as 2018, and that neither escalated those concerns into formal enforcement in time to force a fix. The FDIC estimated the failures would cost the Deposit Insurance Fund about $18.7 billion. Most of that cost was not ordinary insured deposits. It was the decision to protect balances above the $250,000 cap.

The limit that got overridden

Federal law lets Treasury invoke a systemic-risk exception, on the joint recommendation of the FDIC and the Fed, to cover deposits beyond the statutory cap when a failure threatens the wider system. GAO reviewed the invocation for SVB and Signature and concluded it likely slowed the panic: deposit outflows from smaller banks eased within a week.

The mechanism worked. It exists only because the ordinary limit (the number depositors are told to rely on, and the number the premium structure is priced around) was not enough once a run could move at the speed of a group chat instead of a line outside a branch.

The FDIC’s inspector general did not wave this through. A December 2024 review found the agency’s readiness to resolve large regional failures was “not sufficiently mature” for a consistently efficient crisis response, citing gaps in staffing, technology, and coordination. The OIG issued eleven recommendations. A parallel GAO review found the FDIC lacked a centralized system for tracking supervisory recommendations across examiners, so emerging risks could fall through cracks nobody was specifically watching.

The morning after

The policy response did not tighten that machinery. Travis Hill, installed as acting chairman on January 20, 2025, moved the other way. Within weeks he rescinded the FDIC’s 2024 statement of policy on bank mergers and withdrew four other pending proposals, what one law-firm analysis called a regulatory purge.

The stated aim was faster, more predictable merger review. The practical effect was that the same agency whose watchdog had just flagged readiness and coordination failures spent the next chapter loosening review of the consolidation that produces large, complex institutions whose failures are hardest to manage. The Senate confirmed Hill as the 23rd chairman in December 2025; he was sworn in the following January, after the direction was already set.

This is not the Reconstruction Finance Corporation’s cronyism, and it is not TVA’s multi-decade debt spiral. It is quieter. A self-funded insurance program with a hard dollar cap still depends on a discretionary political judgment about who gets covered beyond that cap when the alternative looks worse. The FDIC did not need a conservatorship. It needed an emergency exception to its own rules within days of the crisis starting.

Why the design is the story

Deposit insurance is not a bad idea because this happened. Public power is not a bad idea because TVA’s debt ceiling gets raised. The point is narrower. The closest thing to a clean government-corporation design still needed an emergency override in its first real stress test in fifteen years, and the response since has been to loosen the oversight watchdogs said was not ready.

The line between “self-funded insurance” and “implicit federal guarantee” gets redrawn the moment enough uninsured money is at stake. The premium is priced to a cap. The political incentive, once a run is visible, is to ignore the cap. In 2023 that incentive won, at a cost charged back to the banks that survived, which is still a transfer, just not a direct appropriation.

The question worth asking of any similar design is not whether it looks sound on paper. It is what happens the first week it gets tested at the speed of a withdrawal.

https://stacker.news/items/1588394