The greatest threat to the independence of the US Federal Reserve does not come from President Donald Trump’s attacks or from a Supreme Court ruling that might expand presidential authority. It comes instead from the Fed’s own longerterm evolution from lender of last resort to lender of immediate resort. Without a clear distinction between temporary liquidity support and protection for insolvent institutions, the Fed’s independence becomes cover for ad hoc bailouts, and monetary policy turns into a hostage of weak institutions and regulators’ reluctance to admit supervisory failure.
With each successive crisis over the past decade and a half – from the global financial crisis of 2007–08, to the COVID-19 shock of 2020, to the turmoil among mid-sized US banks in 2023 – the Fed has steadily expanded the scope, scale and permanence of its interventions. What began as emergency liquidity support designed to calm panic has increasingly become a standing feature of financial-market management.
When every disruption is said to generate dangerous “spillovers”, and every balance-sheet wobble prompts intervention, the distinction between containing panic and propping up failing institutions collapses. Along with it goes the discipline that keeps moral hazard in check. At that point, central-bank independence no longer enforces restraint; it merely shields open-ended emergency measures from scrutiny.
The benchmark for central-bank restraint was set more than a century ago by Walter Bagehot: lend early and freely, but only to solvent institutions, against good collateral, and at a penalty rate. Under this elegant framework, the central bank supplies liquidity, the fiscal authority provides capital, and markets impose accountability. Viable institutions are protected from liquidity panics, while insolvent ones are restructured or shut down.
That framework endured as long as the boundaries between regulated banks and the rest of the financial system were relatively clear, and the line between liquidity and solvency was easier to draw. Modern crises have made the latter distinction harder to sustain, as sharp asset-price declines can quickly undermine institutions that appear sound on paper.
At the same time, non-bank entities have increasingly performed bank-like functions without comparable oversight. In response, the Fed extended its reach in 2008 and again in 2020, broadening collateral standards and creating new lending facilities on the fly. By the time the pandemic struck, interventions once considered extraordinary had become routine. While each step may have been defensible in isolation, together they pushed the Fed beyond the limits that once preserved its legitimacy.
Illiquidity is a short-term funding problem. Insolvency reflects deeper, long-term balance-sheet weaknesses that can be addressed only through new equity, mergers or orderly resolution. The central challenge for policymakers is to determine whether an institution is solvent but temporarily illiquid, or insolvent and therefore in need of restructuring. If regulators cannot reliably draw this distinction for banks, despite having granular supervisory data, they certainly cannot do so for non-banks, where transparency is far more limited.
The banking turmoil of 2023 underscored the risks created by this mission creep. My co-authors and I estimated that hundreds of banks faced large mark-to-market losses on long-duration assets, operated with thin capital buffers, and relied heavily on uninsured deposits. Yet instead of calling for recapitalisation or restructuring, the episode was widely framed as a liquidity crisis. New facilities effectively extended support to roughly US$9 trillion in uninsured deposits, vastly expanding the financial safety net.
By acting as a lender of immediate resort, the Fed may have stabilised markets in the short run. But it also left underlying incentives unchanged, setting the stage for future crises and putting its independence under growing strain. Higher interest rates, though necessary to rein in inflation, exposed widespread interest-rate risk across the banking system. This left the Fed in a bind: raise rates aggressively and risk breaking the weakest banks, or hold back and allow inflation to persist. Financial fragility thus became an undeclared ceiling on monetary tightening.
The Fed’s dual role as both bank supervisor and monetary authority magnifies this conflict. Admitting supervisory failures or hidden solvency problems is politically costly. That creates a strong incentive to characterise balance-sheet weaknesses as liquidity shortages, reinforcing a bias toward intervention that runs counter to the very purpose of central-bank independence.
The solution is not to abandon Bagehot’s framework but to update it. The Fed should set clear conditions for when emergency facilities can be activated, publish transparent eligibility rules that restrict lending to solvent institutions, impose meaningful penalty rates and haircuts, and disclose how each facility was used once it is wound down.
Most importantly, the Fed should confine itself to providing liquidity and leave solvency problems to markets and fiscal authorities. Without that separation, the central bank will continue to drift toward industrial policy, undermining both its legitimacy and its independence.
Some may argue that these safeguards already exist. But recent interventions have lacked the one constraint capable of limiting moral hazard. Banks that receive liquidity support should be required to raise equity commensurate with that support within a defined window, or face restructuring or consolidation. If markets are unwilling to provide capital, the institution is not illiquid – it is insolvent.
Central-bank independence must rest on sound governance, transparency and accountability. That includes acknowledging supervisory failures when solvency problems are misdiagnosed as liquidity ones, and explaining how those failures will be addressed rather than masking them through intervention. A central bank that cannot refuse intervention during a crisis cannot be expected to hold its ground when tightening monetary policy.
To protect its credibility, the Fed must resist the temptation to treat every problem as systemic. Otherwise, the financial system will remain fragile by design, and the Fed’s independence will erode each time it rescues another institution that should have been allowed to fail.
Amit Seru is Professor of Finance at the Stanford Graduate School of Business and a senior fellow at the Hoover Institution
Project Syndicate