The Moral Hazard Paradox of Financial Safety Nets

Financial panics are pernicious, but they can be countered with government guarantees of panic-prone debt. In the wake of the crisis, however, Congress has stripped regulators of this sort of guarantee power, motivated in large part by concerns that such powers could exacerbate moral hazard. In a new article, The Moral Hazard Paradox of Financial Safety Nets, I suggest that the moral hazard impact of guarantee authorities in the current system is ambiguous – indeed, it is plausible that guarantee authorities could reduce the (net) cost of moral hazard arising from expectations of government intervention. This supports the view that freestanding regulatory guarantee authorities should be reestablished. The issue is especially important because, despite a number of sensible and necessary reforms since 2008, the financial system remains vulnerable to panic.

First, a brief review: As damaging as the runs and panics of 2008 were to the financial system and the real economy, it could have been much worse: the system could have imploded, but did not, thanks largely to regulatory intervention. Many observers believe the moment the system came closest to imploding was during the run on prime money market funds in the wake of Lehman Brothers’ failure – at this juncture, “[i]t was overwhelmingly obvious that [financial regulators] were staring into the abyss.”

The system did not fall apart because regulators halted the run, and they did so by providing a temporary guarantee of all money market fund accounts – essentially insuring trillions of dollars of private debt. The Treasury Department determined it had authority to establish the program through a creative interpretation of its power to deploy something called the Exchange Stabilization Fund. (The guarantee program never paid out a dime and collected more than $1 billion in premiums.)

Shortly thereafter, Congress eliminated Treasury’s authority to extend similar guarantees in future crises. Likewise, almost two years later, Congress eliminated the freestanding authority of the FDIC to establish the type of broad guarantee program it had set up for banks and bank holding companies in late 2008, a program that played a critical role in stabilizing the system.

Why would Congress take away the very tools that regulators used to hold the system together? Again, the most (apparently) compelling justification is moral hazard: market participants, knowing regulators can save them in a crisis, will fail to take adequate precautions against losses. In the moral hazard narrative, by tying regulators’ hands, we force market actors to protect themselves, leading to a more stable, fair, efficient system over the long run.

Bailouts thus involve trade-offs: saving the system today could weaken the system in the long-run by distorting incentives. Some observers believe bailouts are worth the cost; others don’t. Most, however, agree that a trade-off exists.

My article supports the view that the elimination of these guarantee powers as freestanding authorities was a mistake by suggesting that the principal cost of providing regulators with these authorities may not be a (net) cost at all. Indeed, the lack of guarantee powers – “strong firefighting authorities” – could plausibly increase moral hazard costs in the current system by making targeted bailouts of weak firms more likely. As Tim Geithner notes in his memoirs, “[s]trong firefighting authorities actually make it easier to let firms fail; when you know you have the ability to prevent fires from spreading out of control, you can afford to let them burn for awhile.”

A central argument of the article is that regulators, though stripped of guarantee authorities, retain a number of weaker tools to effect targeted bailouts, where “bailout” is defined as “steps the government takes either to ensure the continued existence and operation of an insolvent firm, or to ensure that if the institution ceases to exist in its erstwhile form, at least one class of creditor does not suffer losses that it otherwise would in the absence of intervention.” Some of the tools regulators might use to save a firm or its creditors formally prohibit bailouts, but these prohibitions are likely weak constraints in practice due to the difficulty of distinguishing insolvency from illiquidity in a crisis. (Charles Goodhart calls the notion that we can reliably make such a distinction a “myth.”)

What are some of these weaker tools?

  • The FDIC can, in concert with the Fed and Treasury, invoke the systemic risk exception to save uninsured creditors of a commercial bank in resolution.
  • The Fed can engage in emergency lending to non-banks in “unusual and exigent circumstances.” While Dodd-Frank required that such loans be extended only to solvent institutions as part of a program of broad-based eligibility, I point to several crisis-era programs with general eligibility requirements that were ultimately used by only one or two firms. And it is plausible that the Fed could establish such a program very quickly, particularly if it used one of the programs from the recent crisis as a template.
  • The Treasury Department can lend to a firm in resolution under Title II of Dodd-Frank (the Orderly Liquidation Authority) if, for example, the short-term creditors of the firm’s subsidiaries refuse to roll over their debt. Such lending requires, inter alia, a determination of solvency, but again, the value of the firm’s assets may be opaque and its solvency (to quote Goodhart once more) subject to “multiple equilibria.”
  • Regulators may use moral suasion to facilitate private bailouts. If, for example, Barclays had been able to buy Lehman in September 2008, part of the deal would have involved a bailout from a consortium of Wall Street banks, who had agreed to finance up to $30 billion of bad assets that Lehman would have left behind in the merger.

In some circumstances, regulators would prefer not to use these tools, and to let a weak firm fail and its creditors suffer losses. The problem, of course, is that a large firm’s failure could spark a panic. If regulators had guarantee authorities – i.e., a strong tool to contain a panic if one started – they would plausibly be more willing to accept a slight risk of a panic arising from a large firm’s failure. If the firm failed and sparked a panic, regulators could contain it; if it failed without sparking a panic, it would have a salutary effect on the system by clearing away the “underbrush” of weaker firms and setting a sobering example for other market actors.

Because regulators lack a strong tool to contain a panic, however, they will be less likely to accept this risk, and more likely to use one of the weaker tools described above to save a weak firm. If this is the case, it is possible that stripping regulators of guarantee authorities makes bailouts of weak firms likelier, exacerbating moral hazard.

This argument raises a number of potential questions that I try to address in the article’s penultimate section, two of which I will adumbrate here. First, some may question the relevance of regulators’ formal authorities, believing that regulators and/or Congress will find a way to save the day if worse comes to worst. While it is impossible to disprove this, I believe the risk that regulators will not find a plausible legal grounding for extending guarantees absent congressional action, and that Congress may not act in time to prevent extraordinary damage, is enough to motivate regulators to avoid situations in which congressional action or extralegal interventions may be necessary to save the system. (And, of course, avoiding these situations will often involve relying on the weaker bailout tools to save individual firms and their creditors.)

Second, some may ask why we don’t eliminate the weaker tools as well as the guarantee authorities; wouldn’t that eliminate moral hazard? Perhaps, but the history of our financial system in the era before these weaker tools existed suggests that it would do little to prevent damaging crises. As Geithner writes in his memoirs, “[t]aking away the fire department’s equipment certainly ensures that the equipment won’t be used, but it isn’t much of a strategy for reducing fire damage.” Regulation should aim to limit the net costs of a vibrant financial system. Addressing moral hazard is important as a means to achieving this goal, but it should not constitute an end in itself.

The article is currently in the editing process; I would welcome any thoughts or criticisms, either in the comments or by email.

The preceding post comes to us from John Crawford, Associate Professor of Law at UC Hastings College of the Law. It is based on his paper entitled “The Moral Hazard Paradox of Financial Safety Nets” and available here.