Silicon Valley Bank’s failure exposed many familiar weaknesses: uninsured deposits, interest-rate risk, a concentrated depositor base, and weak interest-rate hedging. But the more revealing lesson lies one layer deeper. The supervisory failure was not simply that regulators missed red flags at a fast-growing bank. The size-tiered supervisory system itself created a hidden grace period precisely as the bank’s risk profile worsened.

In a recent paper, I argue that the underappreciated weakness of size-tiered supervision is its creation of hidden grace periods during supervisory transitions. That weakness matters because recent policy debates over supervisory tailoring, the practice of applying stricter standards as banks become larger, more complex, or more systemically important, often ask the wrong question.[1] They ask whether tiered supervision is good or bad in the abstract. It is neither. A differentiated system is sensible: Banks are not all alike, and regulators should not supervise a $30 billion bank as if it were a global systemically important institution. The harder question is what happens during the transition from one supervisory tier to another.

Tiered supervision does more than sort banks into categories based on size. It also determines when new standards apply, when new rating systems begin, how many examiners are assigned, and how quickly identified weaknesses translate into binding constraints. Those transition choices are easy to treat as mere implementation issues, but they define the interval between recognition and enforcement.

That interval can become a hidden grace period. A bank crosses a threshold and is formally on its way to a more demanding supervisory category. Yet the practical consequences arrive only gradually. Enhanced prudential standards phase in over several quarters. Existing supervisory ratings may carry over. Initial large-bank ratings may be delayed. Serious findings can be framed as matters to remediate over time rather than conditions requiring immediate constraint. On paper, supervision has tightened. In practice, the bank still enjoys much of the tolerance associated with its past tier.

This is dangerous even in ordinary times. But it is most dangerous when the bank is growing rapidly and already deteriorating. Rapid growth is often treated as evidence of commercial success. Yet supervisors should also treat it as a warning that the supervisory clock may be running too slowly. A fast-growing bank can change its funding mix, take on new interest-rate duration risk, and increase its operational complexity far faster than a threshold-based supervisory regime can fully recalibrate. Recent empirical work reinforces the point: Failing banks often deteriorate in plain sight, and rapid asset growth is itself a powerful medium-term risk signal.[2] If governance, liquidity, or interest-rate risk is already worsening, a long transition does not merely preserve flexibility. It gives management time to compound fragility.

SVB illustrates the mechanism with unusual clarity. As it moved from the Federal Reserve’s regional-bank program into the large-bank portfolio, the system formally ratcheted up expectations. Yet key constraints did not bind at once. Some liquidity requirements were phased in well after the bank crossed the relevant size threshold. Its initial large-bank ratings were delayed. Supervisors identified weaknesses but downgraded slowly and escalated unevenly. The result was not an absence of oversight but a long supervisory runway during which concerns were recognized, documented, and debated, but not translated quickly enough into binding constraint.

That distinction is critical. The standard story of supervisory failure emphasizes missed warning signs. SVB suggests a different failure mode: Warning signs were seen, but the architecture of transition stretched the time between seeing the problem and forcing change. In that setting, delay is not neutral. It changes incentives. Management can continue growing, relying on brittle funding, and postponing painful adjustments while the system signals that full large-bank discipline is still arriving rather than already here.

This is why the policy issue is narrower, and more easily managed, than the broader debate over supervisory tailoring suggests. The question is not whether supervisors should differentiate among banks. They likely should. The question is how to prevent tier transitions from operating as structured delay. A sensible regime can still create perverse timing effects if the path into stricter oversight is too slow, too opaque, or too discretionary.

The answer, therefore, is not simply “more regulation.” It is also a clearer approach to escalation during transitions. My paper does not offer a detailed reform blueprint. It makes a narrower point: Supervisory runway length, transition governance, and transparency should be treated as design parameters rather than afterthoughts. That means asking how to shorten supervisory runways, how to coordinate escalation across supervisory portfolios, and how to tie observable growth and funding signals more directly to supervisory triggers. It also means making clear that supervisors should act more forcefully when prudential concerns arise, even during transition periods, and that waiver or classification decisions that prolong delay should face stronger contemporaneous justification and review.

That approach would preserve the core logic of tiering while focusing attention on the transition problem my paper identifies. Tiered supervision is supposed to allocate attention more intelligently. It should not allow a weakening bank to continue operating under supervisory tolerance that no longer fits its condition.

ENDNOTES

[1] United States Senate Committee on Banking, Housing, and Urban Affairs. Recent Bank Failures and the Federal Regulatory Response. Hearing, 118th Cong. (2023), https://www.banking.senate.gov/hearings/recent-bank-failures-and-the-federal-regulatory-response; United States House Committee on Financial Services. The Federal Regulators’ Response to Recent Bank Failures. Hearing, 118th Cong. (2023), https://financialservices.house.gov/calendar/eventsingle.aspx?EventID=408671; United States Government Accountability Office. Bank Supervision: More Timely Escalation of Supervisory Action Needed. Report GAO-24-106974 (2024), https://www.gao.gov/products/gao-24-106974.

[2] See Correia, S. A., S. Luck, and E. Verner. 2024. Failing Banks. NBER Working Paper No. 32907. Cambridge (MA): National Bureau of Economic Research; Available from: http://www.nber.org/papers/w32907.

Pedro M. Batista is a lecturer in commercial, corporate, and banking law at the University of Leeds School of Law and a research fellow at New York University School of Law. This post is based on his recent article, “Escalation without enforcement: hidden grace periods in tiered bank supervision,” published in the Journal of Banking Regulation and available here.