Regulators have made great progress over the past decade in designing mechanisms to deal with a failed systemically important financial institution (SIFI) in a way that avoids both a bailout and the risk of sparking or fanning a broader panic.[1] This progress is limited, however, to what happens once a SIFI is in resolution. The lack of adequate mechanisms to ensure a SIFI is placed in resolution in a timely manner lingers as a major weakness in these schemes. In a new paper, Resolution Triggers for Systemically Important Financial Firms ('Resolution Triggers'), I focus on the risk of delay in triggering resolution.[2]

Delay is pernicious because losses are likely to worsen at firms with razor-thin or negative capital; weak firms have an incentive to ‘gamble for resurrection,’ taking imprudent risks to climb back to solvency, confident that the costs of bad outcomes will be borne by creditors or taxpayers.[3] Timely resolution can arrest this process.

A bias toward delaying resolution, however, infects all key actors, including regulators. This has a variety of causes, but two are worth mentioning here. First, regulatory measures of (in)solvency famously lag real-time economic developments; and second, there may be implicit or explicit political pressure not to intervene until things really fall apart. As Paul Tucker has observed, ‘if faced with uncertain long-term benefits but an immediate risk of unpopularity, a policy maker might incline toward delaying action….’

In Resolution Triggers I propose different ways of trying to mitigate these factors, including (inter alia) exploring the use of market measures of insolvency. In this post, I want to home in one particular mechanism that can help mitigate the problem of delay: the ‘total loss absorbing capacity’ (TLAC) rules that apply to large bank holding companies, requiring them to maintain long-term debt and equity above prescribed minimums. The advantage of TLAC is that it can absorb losses without triggering panicked runs—a major risk generated by haircuts on short-term debt. TLAC plays a central role in efforts to facilitate resolution without forcing regulators to face a choice between bailouts and panic.

TLAC’s effectiveness rests first and foremost on whether the overall requirements are set at adequate levels. A related but distinct question, I argue, ties directly back to the issue of trigger timing: namely, for a given aggregate amount of TLAC, how much should be in the form of long-term debt, and how much in equity? (Currently the proportion is roughly one-third debt and two-thirds common equity.)

Equity has its advantages. It can absorb losses without implicating difficult triggering questions that arise with debt. Moreover, a higher proportion of equity can mitigate perverse risk-shifting incentives on the part of shareholders and firm managers, as captured by the notion of ‘gambling for resurrection,’ cited above. A thicker equity cushion operates like the deductible on an insurance policy, making a firm’s residual claimants more sensitive to the downside risks of aggressive strategies.[4]

Cutting in the other direction, some observers believe that higher debt-to-equity ratios impose discipline on firm managers, helping solve a potential agency problem by limiting the free cash flow firm managers have to hide underperformance or ‘benefit taking.’ Other observers, however, have critiqued this view as misplaced in the context of banks. From this latter standpoint, TLAC shouldn’t include long-term debt at all.

Even if one rejects the view that debt solves an agency problem vis-à-vis shareholders, there is another factor, overlooked in the literature to date, that counsels against eliminating long-term debt as a component of TLAC—and that may even support higher levels than we currently see. (I do not attempt to quantify the correct level here.) Specifically, long-term debt may help mitigate or counteract the delay-inducing factors cited above—regulatory capital’s lag vis-à-vis real economic developments, and political pressure not to rock the boat.

Long-term debt’s potential role in mitigating the bad effects of capital lag is straightforward: a SIFI should be put into resolution when it is at or near the point of insolvency, and capital lag means that a firm may be (deeply) insolvent by the time it hits a capital-based resolution tripwire. If we relied solely on equity to absorb losses in such a case, regulators would again face the bailout-or-panic dilemma. But if there is sufficient long-term debt, there will be credible loss-bearing capacity beyond the point of (real) insolvency.

A possible reply, of course, is that if timeliness is an issue, why not just mandate that TLAC be all equity but set a much higher threshold for triggering resolution? Financial journalist Matt Levine trenchantly makes the point in critiquing the current requirement that roughly one-third of TLAC be in the form of long-term debt:

Why do it this way? One answer is …: You want to ‘ensure that there are sufficient resources available in resolution.’ The idea is that, if banks run out of capital entirely—if they’re technically ‘insolvent’—then you can make them solvent again by poofing TLAC debt into equity. If they have no TLAC debt, only equity, then when they become insolvent you can’t do that. This strikes me as an extreme bit of formalism—why not make the requirement all equity, and seize the bank when it loses two-thirds of its equity?

What strikes Levine as formalism can be better understood, I argue, as a way to force regulators’ hand and weaken political and institutional inertia. Put simply, pulling the resolution trigger on a firm when its regulatory measures of capital indicate that it is fundamentally solvent is not a politically stable solution to the problem of timeliness. The remaining (ostensible) capital buffer against insolvency will weaken everyone’s sense of urgency and strengthen banks’ persuasiveness when they argue against specific triggering decisions. Even if one could credibly remove all regulatory discretion and make the higher equity-based trigger automatic, it would be highly vulnerable to bank lobbying and legislative rollback. The fallout of SIFI failure and capital as a lagging indicator are complicated ideas; government seizure of an ostensibly solvent firm is straightforward and presumably distasteful to many elected representatives, at least in the United States. The resolution of a firm that—by regulatory metrics—is insolvent is less likely to invite backlash.


Timeliness is essential if the resolution mechanisms designed to avoid the bailout-or-panic dilemma are to serve their purpose. Among the virtues of a long-term debt requirement for SIFIs,[5] one may count its utility in counteracting the pernicious effects of delay in triggering resolution, and in mitigating perverse incentives that create a bias toward delay. The mandate that a substantial percentage of TLAC be in the form of long-term debt is one thing regulatory reformers got right.


[1] I have described these mechanisms at greater length here and here.

[2] I also argue (inter alia) that there is a risk of triggering resolution too early if liquidity is used as a trigger. See Resolution Triggers, Part III.A.

[3] The most salient historical illustration of this dynamic, of course, is the U.S. Savings & Loan crisis of the 1980s. See Resolution Triggers Part III.B.1.b.

[4] See Richard Scott Carnell, ‘A Partial Antidote to Perverse Incentives: The FDIC Improvement Act of 1991’, (1993) 12 Ann Rev Banking L 317, 320  (‘[c]apital is like the deductible on an insurance policy: the higher the deductible, the greater the incentive to avoid loss. As capital falls, an institution’s incentives to avoid loss fall with it.’).

[5] I posit other virtues of the long-term debt requirement in Credible Losers.

John Crawford is a Professor of Law at UC Hastings College of the Law San Francisco