“Too big to fail” became a household phrase during the 2007–08 Global Financial Crisis. When giant financial institutions teetered, the government reached for ad hoc bailouts, fearing contagion and a broader economic collapse. Congress responded with the Dodd-Frank Act, which promised that next time would be different.
The Dodd-Frank Act created a new resolution regime with two provisions to prevent taxpayer-funded rescues. Title I of the act required large financial firms to draft “living wills” – resolution plans mapping out how they could fail without harming financial stability. Title II then gave the FDIC “orderly liquidation authority” to wind down a failing holding company, backed by temporary emergency funding, in case proceeding under the bankruptcy code would be detrimental to financial stability. Together, these two provisions were supposed to make even the largest financial institutions “safe to fail.”
Fifteen years later, they remain unused. When Silicon Valley Bank (SVB) collapsed in 2023, regulators instead chose to guarantee uninsured deposits and launch a new emergency lending facility. In Switzerland, regulators abandoned their resolution regime and bailed out Credit Suisse. The world saw that when panic spreads, the resolution playbook gets tossed aside.
Why? Mike Tyson once observed, “Everyone has a plan until they get punched in the mouth.” It turns out that every financial regulator has a plan until they encounter a crisis. In a new working paper, I highlight two fundamental flaws – one substantive, one procedural – that explain why the post-crisis resolution framework has failed to live up to its promise. And if nothing changes, regulators will reflexively reach back into the bailout bag the next time a panic occurs.
A Microprudential Tool for a Macroprudential Problem
First, consider the substantive difference between microprudential and macroprudential regulation. The literature observes that “microprudential regulation focuses on financial institutions, while macroprudential regulation emphasizes the financial system as a whole.” The former seeks to ensure the viability of an individual institution, whereas the latter aims to bolster the resilience of the entire financial system. The implicit assumption, which bears stating explicitly, is that improving the stability of individual financial institutions will not necessarily improve the stability of the financial system as a whole.
Living wills and orderly liquidation are designed to handle the distress of a single firm. They are microprudential tools aimed at idiosyncratic risk. If one large institution falters because of a unique business model or a bad bet, resolution could contain the damage. But true crises rarely unfold neatly. Systemic events are driven by uncertainty and contagion, not by one firm’s mistakes. Confidence evaporates across the sector. In those moments, firm-by-firm solutions are inadequate. Policymakers need macroprudential tools that provide reassurance to the entire system.
This does not mean, however, that the resolution regime is useless. Some failures are idiosyncratic, and using a microprudential tool would make sense. Examples discussed in the working paper include Continental Illinois in 1984 and Long-Term Capital Management in 1998. Those episodes illustrate the niche where living wills and orderly liquidation might succeed – if regulators are willing to use them. But that leads us to the procedural flaw.
Regulators Won’t Experiment in a Panic
Regulators are unlikely to use an untested tool in the midst of a bank run, even if the run is contained to a particular institution. In 2023, SVB looked unique – more than 90 percent uninsured deposits, heavy exposure to interest-rate-sensitive assets. On paper, it was a great candidate for the resolution regime. But in real time, it was impossible to know whether SVB’s troubles were truly isolated, and politically connected depositors were lobbying for protection. In such an uncertain environment, no regulator wants to discover, mid-panic, that an untested tool can’t stop contagion. Faced with economic uncertainty and political pressure, regulators did what they always do: bail out first, rationalize later.
The same story played out in Switzerland. Credit Suisse had a resolution plan on file, but when confidence vanished, regulators engineered a shotgun merger with UBS instead. The revealed preference is clear: When the stakes are high, resolution frameworks sit on the shelf.
If the resolution regime is to matter, it needs credibility. That requires practice. Right now, the only time the tools might be used is during existential panics – precisely when regulators are least willing to try them. Congress should expand the regime to cover “almost-too-big-to-fail” firms. Lowering the threshold would let regulators use the resolution toolkit in less fraught situations. Demonstrating competence in lower-stakes cases would build the expertise and political will to use resolution when it really counts.
Conclusion
The Dodd-Frank Act was designed to end bailouts. It hasn’t. Living wills and orderly liquidation have proven too narrow in design and too daunting in practice. When crisis strikes, regulators still prefer familiar bailouts to untested tools. Unless lawmakers change course, the resolution framework will remain a costly exercise. The next panic will again be met not with orderly wind-downs, but with emergency rescues. And Mike Tyson’s truth will continue to haunt regulators.
This post comes to us from Professor Jeffery Zhang at the University of Michigan Law School. It is based on his working paper, “Too Scared to Use: Living Wills and Orderly Liquidation of Too-Big-to-Fail Financial Institutions,” available here.