Can US banks have it all?

Uuriintuya Batsaikhan is a Consultant on the project ‘Financing the Green Transition in Europe’

The United States is planning a revision of rules on bank liquidity. US Treasury Secretary Scott Bessent said in early March that the Treasury is working on changes including a proposal for pre-positioning of collateral, under which banks would hold assets in advance with the Federal Reserve to secure access to liquidity. This would, according to Bessent, help tackle a phenomenon that leads banks to avoid borrowing from the Fed in case they are perceived as struggling.

In short, the Fed’s discount window – its lending facility for banks needing short-term liquidity – should be a normal part of the system rather than a last-resort tool, use of which immediately signals that a bank is in distress. So-called ‘discount window stigma’ was apparent in 2023 when banks including Silicon Valley Bank, which failed in March 2023, held Treasuries and agency mortgage-backed securities booked at their original purchase price that looked liquid on paper but could be turned into cash only by selling at lower prices at the time of turmoil, cutting into capital.

Recalling this episode, Bessent said that a framework that delivers liquidity only in stylised stress scenarios is not credible in practice.

The plan for pre-positioned collateral he outlined sounds like a pragmatic adjustment to how liquidity support ought to be organised. But the question is not only how much collateral banks should post, but what kind of collateral should qualify and on what terms.

Bessent argued that more recognition of pre-positioned, collateralised borrowing capacity in liquidity rules would reduce the need for banks to sit on large piles of safe assets, allowing more of those portfolios to support lending. In practice, this could mean that less-liquid or less-marketable assets are parked at the central bank, while high-quality liquid assets are used elsewhere on bank balance sheets.

The future Fed framework could be similar to that of the European Central Bank, which accepts as collateral not only government bonds but also credit claims, asset-backed securities and other private-sector claims. By and large, this has not been a source of instability.

If the collateral pool eligible for pre-positioning includes riskier assets, banks are effectively manufacturing liquidity from the very positions most likely to become illiquid in a stress event, which is exactly what the reform is supposed to prevent

But the ECB framework is broad because it must serve a monetary union spanning diverse financial systems, legal traditions and asset markets. A narrow collateral pool would have systematically excluded banks in countries where sovereign debt markets were thin. In other words, breadth was a condition of equal access and not a concession to risk appetite.

Read in isolation, Bessent’s plan might look like an attempt to put idle assets to work. But read together with a broader push by US regulators to loosen prudential requirements on bank capital, the picture is different. The Fed, the Office of the Comptroller of the Currency (a bureau of the US Treasury) and the Federal Deposit Insurance Corporation have jointly proposed reductions in capital requirements for large and small banks. US banks appear to want it all ways: the ability to pledge less-liquid collateral, easier access to central bank liquidity and lighter limits on their risk-taking. Bessent seems willing to accommodate those demands.

The US debate also tends to overlook the broader context of the evolution of central bank operational frameworks. The Fed now operates a supply-driven ample reserves framework. In this environment, the interbank rate is no longer set by supply and demand for reserves but is anchored by the interest rate on reserve balances (IORB) that the Fed pays directly to banks.

Since December 2025, the Fed has been conducting reserve management purchases of $40 billion per month in Treasury bills as routine operations to maintain adequate reserve levels. This represents a world in which continuing reliance on central bank liquidity is formally integrated into the monetary plumbing, with the Fed actively expanding its balance sheet on a rolling basis to keep the system functioning at its intended levels. In this world, what does a genuine stress scenario look like?

If a bank needs to go beyond pre-positioned collateral and tap emergency liquidity assistance when the system is already swimming in reserves, the stigma could be worse, not better.

If the collateral pool eligible for pre-positioning includes riskier assets, banks are effectively manufacturing liquidity from the very positions most likely to become illiquid in a stress event, which is exactly what the reform is supposed to prevent.

Pre-positioned collateral could make the liquidity framework more credible, and a less stigmatised discount window would be an improvement. But the reform package Bessent has outlined cannot be assessed in isolation from the broader direction of policy. If it widens the safety net while loosening the constraints that justify it, it risks amplifying liquidity risk, potentially threatening more bank failures.

The author thanks Guntram Wolff, Rebecca Christie, Francesco Papadia, Ulrich Bindseil and Nicolas Véron. This article is based on a Bruegel First Glance.