Bank Restructuring on the Cheap

The proposal to allow a regulator of a sinking financial institution to force a swift restructuring similar to a pre-packaged bankruptcy is clever. It could help avoid the massive loss of value that usually occurs when a financial institution goes into bankruptcy. Since financial institutions rely on the confidence of many creditors and counterparties to operate effectively, their going concern value can evaporate very quickly when they hit trouble and that confidence dissolves. In addition, their assets can decline sharply in price if it appears that they may be dumped on the market as part of a bankruptcy process.

The core insight behind the proposal seems to be that the loss of value in bankruptcy is so large that in most cases everyone would be better off with a forced division of the losses, even if done on a very ad hoc and rough basis, as long as it allowed the firm to continue operating with restored confidence from its customers, creditors, and counterparties.

If we buy that premise, then this would seem to be a useful tool to add to a regulator’s arsenal as a last resort. I stress the “last resort,” because the problems of using this tool are significant enough that it would be better to avoid getting into the circumstance where it is required. One way to avoid that situation might be through the use of contingent capital, which can serve a similar rescue function. (Contingent capital is essentially a debt instrument which includes an agreement under which it would automatically convert into equity under specified conditions when a financial institution becomes troubled.)

The downsides of the “bail-in,” as the authors call it, revolve around the tremendous uncertainty introduced by allowing a regulator to make a series of extremely important, binding decisions very quickly. First, it is not hard to create hypothetical examples where one or more class of investors would actually be worse off in reality, even if the regulator falsely believed that all classes would be better off with the bail-in. This could occur if the regulator misjudges the true value of the firm on an ongoing basis and/or mis-estimates the losses from the alternative of a normal bankruptcy resolution. For example, sometimes a financial institution has become so bad that it is better off being put out of its misery instead of being given the ability to lose still more money by forcing creditors to become shareholders. A regulator might not realise this, or might be under political constraints that force it to ignore the truth. (Imagine how hard it would have been for regulators to shut down Citigroup even if it had been the consummate money-losing machine that it has sometimes been portrayed as being.) It is true that regulators or governments could make the same political calculation and mount a rescue with taxpayer money, but such a rescue would be a lot more tempting if the funds come from creditors instead of voters.

Similarly, political constraints might cause one class of shareholders or creditors to be favoured so strongly in the restructuring that it takes only a small estimation error for a disfavoured class to be worse off than under bankruptcy. Remember the charges of political pressure that accompanied the Chrysler restructuring when some creditors felt that the unions were being given a strongly favoured status.

Second, such a bail-in could be too easy for regulators. They could come to see it as an action to be taken earlier in the process, rather than waiting until private options were exhausted. Given the technical difficulties of getting the bail-in right and the market’s reasonable dislike of arbitrary government decisions, it would be best not to drift into such a situation.

Third, there is a reason that prepackaged bankruptcies take so long. It is not just that negotiations take time; it is also a difficult and time-consuming process to make a good estimate of the real value of a troubled firm. Remember the necessity of essentially renegotiating the Bear Stearns rescue a few weeks later since the work over the critical weekend had necessarily been rushed. It’s not a perfect analogy, but it still illustrates that haste has its costs.

Fourth, the knowledge that regulators can exercise their bail-in option could make investments in financial institutions significantly less attractive. One would be going in knowing that if the institution became sufficiently troubled, the regulator could basically decide over a weekend what your share of the losses would be. The authors are probably right that a well-designed process might not be that frightening, but this would certainly be a key issue to examine further.

Even given these concerns, the authors’ core insight is a powerful one and the bail-in alternative is a tool that should be explored carefully.