US Bank Capital Relief Plan Leaves SVB’s Core Weakness in Place

US banking regulators are moving ahead with a capital relief package that could free tens of billions of dollars for the country’s largest lenders. Yet buried inside the same proposal is a pointed reminder that the core vulnerability exposed by Silicon Valley Bank (SVB) in 2023 has not been resolved: unrealized losses on securities portfolios can still undermine confidence faster than regulation can adapt.

What the new capital proposal actually does

In March, federal regulators unveiled a broad overhaul of capital requirements — the loss-absorbing buffers banks are required to hold against future stress. Headlines focused on one theme: deregulation and relief for Wall Street. The proposal would reduce required capital for the largest firms by nearly 5%, effectively releasing capacity that can support more lending, share buybacks, and dividends.

The Federal Reserve estimated that about $20 billion in capital could be freed up at the eight largest US banks alone. Michael Barr, then serving as the Fed’s Vice Chair for Supervision, suggested in his own assessment that the total could ultimately reach $60 billion once all related adjustments are included.

On the surface, that sounds like a straightforward loosening of post-crisis constraints. But capital ratios are only part of how markets judge bank strength. What matters in practice is not merely how much capital is reported, but whether investors and depositors believe the reported figures accurately reflect underlying risk — including embedded losses that have not yet been realized.

The quiet carve-out: unrealized losses for large regionals

When you move past the headline numbers and into the technical details, a notable exception appears. While the broader package reduces capital requirements, it simultaneously tightens one critical area for a specific slice of the industry: large regional banks.

Under the proposal, certain large regionals will be required to include unrealized losses on some securities in the capital metrics they report to regulators. That change is directly linked to the 2023 collapse of SVB, which had legally excluded such losses from its regulatory capital calculations.

This is effectively a regulatory admission. For years, mid-sized institutions were allowed to treat unrealized losses on long-duration securities as an accounting footnote, not a capital issue. Now, regulators are saying that at least for the regional cohort large enough to pose systemic concerns, that treatment was flawed.

The policy change acknowledges that paper losses can swiftly become real when confidence erodes. By forcing these banks to recognize unrealized losses in capital, supervisors are trying to reduce the gap between how healthy a bank looks on paper and how vulnerable it is in a rising-rate environment.

Unrealized losses and the SVB collapse

To understand why this matters, it helps to clarify what “unrealized losses” mean in practice. Consider a ten-year government bond purchased at $100 when rates are low. If interest rates later rise sharply, new bonds offer better yields. The older bond’s market value might fall to $80, even though the holder has not sold it and has not suffered a cash loss — yet.

That $20 gap is an unrealized loss. On many regulatory metrics used before 2023, banks could ignore that decline in value when reporting capital, especially if they classified the securities as “held to maturity.” For mid-sized banks, those paper losses effectively disappeared from key regulatory ratios, even when the cumulative exposure was substantial.

SVB’s failure showed how dangerous that disconnect can be. The bank’s portfolio consisted largely of long-term, ostensibly safe securities that had lost significant value as interest rates climbed. In early March 2023, SVB announced a $1.8 billion loss on the sale of securities, directly linked to those previously hidden unrealized losses, plus a plan to raise $2 billion in fresh capital.

The market reaction was swift and brutal. SVB’s stock price fell around 60% the next day. Uninsured depositors began pulling funds at scale: $42 billion exited the bank in a matter of hours, with another $100 billion queued for withdrawal by the following morning.

The key point for risk professionals and investors is that the losses were not new. They had been building quietly as rates moved higher. But as long as they remained unrealized and excluded from regulatory capital, most supervisors, investors, and depositors did not fully grasp the magnitude of interest rate risk sitting on the balance sheet. Once those losses became visible and crystallized through asset sales, confidence collapsed.

Banks that were already required to reflect unrealized securities losses in their capital calculations tended to manage interest rate risk more tightly. The post-mortem from officials, including Barr, has been clear: allowing institutions to obscure large embedded losses delayed corrective action until markets forced the issue.

Why regional banks still face tighter treatment

The new rule package retains that lesson for large regional banks, even as it reduces capital overall. The requirement to recognize unrealized losses will raise capital needs for those institutions by an estimated 3.1%. Yet once all elements of the broader proposal are applied, their total required capital is still projected to fall by about 5.2%.

By contrast, banks with less than $100 billion in assets will not face the same unrealized-loss requirement, and their capital ratios are expected to decline even more. The regulatory message is that the problem of hidden duration risk is most acute at a particular scale — large enough to trigger broader contagion but not large enough to already fall under the strictest global standards applicable to the biggest Wall Street firms.

This carve-out suggests that Washington regards SVB’s downfall primarily as a failure of the prior rule set. The bank had operated within the legal framework, opting out of including unrealized losses in its capital figures — an option made broadly available to institutions of its size. Regulators now appear determined not to repeat that specific configuration for similar banks.

Barr, who departed his vice chair role earlier this year rather than face removal under the Trump administration but remains on the Fed’s board, has openly questioned the broader direction of the relief package. In a formal dissent, he warned that capital requirements are being “significantly reduced,” that liquidity rules may also be relaxed, and that Fed supervisory staffing has been cut by more than 30%. His core argument is that banking ultimately rests on trust, and thinning out safeguards risks undermining that foundation.

Trust, transparency, and systemic risk

Barr’s emphasis on trust is not rhetorical. A bank can appear stable right up until the moment stakeholders lose faith in its numbers. Once that disbelief sets in, access to wholesale funding can dry up, uninsured depositors can run, and asset sales can lock in losses that had previously been theoretical.

Supporters of the new framework counter that the original 2023 Basel-inspired capital proposal was excessively heavy-handed. Critics at the time argued it risked over-calibrating capital upward in a way that could push risk-taking outside the regulated banking system, into less transparent channels. Fed Governor Michelle Bowman has taken the view that capital will remain robust under the revised framework and that the new design better matches capital requirements with actual risk profiles.

Even within this looser construct, however, the unrealized-loss requirement for large regional banks remains intact. That persistence is telling. If regulators believed duration risk and depositor confidence were no longer significant issues, the incentive to maintain a costly, targeted provision would be weaker. These rules impose real constraints on balance sheet strategy; they are not maintained casually.

For market participants, the signal is that regulators still view hidden interest rate exposure on securities books as a live risk channel — one capable of translating relatively mundane valuation swings into acute liquidity crises. Transparency about unrealized losses is being used as a tool to narrow the gap between how banks present their solvency and how markets perceive it.

Implications for investors and crypto markets

For financial professionals, particularly those active in credit, equities, or rates, the new US capital proposal should be read as a dual message. On one hand, banks — especially the largest — are being given more room on capital, freeing up tens of billions of dollars. On the other, regulators are not backing away from the specific lesson of SVB: embedded losses on long-duration assets matter, no matter how accounting rules categorize them.

For crypto-native investors, the stakes are also familiar. Stress in the regional banking system in 2023 intersected with stablecoin liquidity, fiat on-ramps, and broader risk sentiment across digital assets. When trust in traditional banking falters, it can both disrupt core infrastructure for crypto businesses and change the macro environment that drives flows into and out of digital assets.

The ongoing requirement for large regionals to recognize unrealized losses suggests that policymakers still see a non-trivial risk with interest rate shocks cascading through bank balance sheets. That has implications for how quickly regulators might respond to future stress and how markets might react to signs of renewed balance sheet pressure.

Ultimately, the proposal is not simply a story of deregulation. It is a recalibration that blends capital relief with a targeted safeguard born directly from the SVB episode. For investors, the key takeaway is that while headline capital ratios may improve, the underlying sensitivity to interest rate moves — and the potential for confidence-driven shocks — remains central to assessing US bank risk.