Bank regulation: the many pitfalls of bail-in bonds

The ‘too-big-to-fail’ framework is intended to coordinate the restructuring or liquidation of systemically important banks in the event of insolvency. However, the Swiss Financial Market Supervisory Authority (FINMA) refrained from subjecting Credit Suisse to a forced restructuring or even letting it go bankrupt. The main reasons for this were concerns that a restructuring would cause major disruption in the international financial markets and uncertainty as to whether a restructuring would be successful. This decision meant that bail-in bonds amounting to over CHF 50 billion were not activated in Credit Suisse’s loss buffer.

This inaction calls for reform of the regulatory framework in order to make the restructuring and bankruptcy procedures for global and systemically important banks more practicable. This poses major challenges for Switzerland for three main reasons. First, these restructuring and bankruptcy procedures affect the jurisdictions of various countries.

Second, the recovery or contingency plans cannot safely contain the contagion dynamics of a failing bank, as such failures usually take place in a fragile environment.

And, third, the complexity of contingency plans makes them difficult to implement in practice. Restructuring and bankruptcy procedures for large global banks that do not involve major upheavals in the international financial markets are necessary from a regulatory point of view but are very difficult to implement, especially for Switzerland, owing to these problems.

A conceptually simpler measure could therefore help to further stabilise the banking system: moving international standards in loss buffers away from bail-in bonds and towards equity. This article sheds light on the idea behind – and the many pitfalls of – bail-in bonds compared with equity financing of large banks in a small open economy.

The promising idea behind convertible bonds

The bank bail-out through conversion or reduction of creditor claims, the so-called ‘bail-in’, has become the standard method of bank rescue internationally. Bail-in bonds play an important role in this process. These are contingent convertible bonds, which are converted into equity if either the issuer’s own funds fall below a predetermined threshold and/or if there is state intervention, such as compulsory restructuring proceedings. This may also include bonds whose claims are fully forfeited upon a triggering event.

There is a conversion ranking that determines how bail-in bonds are converted to recapitalise a bank depending on the event and threshold. In the case of Credit Suisse, AT1-type bail-in bonds worth around CHF 16 billion were written off. However, over CHF 50 billion of further reported loss buffers, consisting of bail-in bonds, were not used for recapitalisation purposes. In order to activate these, FINMA would have had to initiate compulsory restructuring proceedings.

The general principle of bail-in bonds, as in the case of Credit Suisse, seems to have something magical about it; a bank can become solvent again in one fell swoop. This evokes associations with ‘Schrödinger’s cat’, which is alive and dead at the same time, so a bank in distress automatically becomes a ‘Schrödinger bank’ – both solvent and insolvent at the same time. What looks like magic at first glance is, at second glance, a problematic path that has already been addressed in the academic literature and has now gained new relevance. The disadvantages of bail-in bonds are illustrated by the example of Credit Suisse in the case of Switzerland.

The trigger problem

Equity automatically absorbs losses and thus prevents a company from going bankrupt. In contrast, bail-in bonds only absorb losses when the predetermined conversion trigger – such as an intervention by the financial market supervisory authority – is activated. This makes both the design and application of bail-in bonds complex. Loss absorption through equity, on the other hand, is simple; it is not subject to any conditions and does not require the triggering of restructuring or bankruptcy proceedings. Rather, loss absorption through equity reduces the probability of state intervention. Moreover, it is always true that a bank that has more equity capital also has a better chance of raising liquid funds because it is more creditworthy.

In addition, there may be uncertainty about the conversion ranking. Even if the write-off of AT1 bail-in bonds carried out by the authorities – and, at the same time, the positive purchase price obtained for Credit Suisse shares – is legally unobjectionable, uncertainty remains as to what exactly can be done with bail-in bonds and what sort of state intervention may be involved. This can amount to the risk premiums that banks have to pay for financing through bail-in bonds, as long as no general state guarantee is provided.

The problem of contagion

As FINMA itself confirmed, it was not sure whether Credit Suisse would have been able to regain investor confidence after a bail-in. For this reason, it did not arrange for the bulk of the bail-in bonds to be converted. It is therefore uncertain whether bail-in bonds – in addition to their theoretical property of being converted into equity in the event of a crisis – can actually be used as a restructuring tool on a large scale.

Furthermore, the conversion of bail-in bonds into shares means substantial losses of claims on the part of creditors, which can have dangerous contagion effects. The extent of these effects depends on who holds the bail-in bonds concerned – for example banks, insurance companies, investment firms or pension funds – and how heavily indebted these institutions themselves are. The conversion of bail-in bonds can place an already distressed creditor institution in further distress, which means that the bail-in bonds issued by it must also be converted, and so on.

The activation of the relevant loss buffers can thus virtually multiply the initial crisis. Given the fragile environment in the international banking sector, such contagion risks could not be ruled out in the event of Credit Suisse being restructured. If the same loss buffer had been available in the form of equity capital instead of bail-in bonds, Credit Suisse’s poor operating performance could have been absorbed with less risk.

The discretionary scope of bail-ins

Obviously, bail-in bonds also cause unintended political-economic interdependencies because regulators have a great deal of discretionary leeway as to whether waivers are triggered or not. If a regulator has information that the bail-in bonds are mainly held abroad, it may be more willing to trigger the conversion and initiate restructuring and bankruptcy proceedings. If a large number of the bail-in bonds are held domestically, however, the incentives for conversion will be less owing to the uncertain impact on creditor companies. Moreover, if the regulator does not even know who exactly will ultimately be hit by the debt waiver, it will be reluctant to trigger the massive losses in value as a result of its decisions.

The considerable discretion enjoyed by the state authorities concerning activation of the loss buffers is not only delicate in terms of regulatory policy. It also invites foreign authorities to influence financial market supervision, as their decisions can potentially cause large losses for foreign financial institutions and investors. Moreover, uncertainty about the conversion trigger adds to the risk premiums of bail-in bonds, as long as no general state guarantee is provided.

Bankruptcies in general do, of course, cause unsecured lenders to incur losses while under sovereign regulation. In the case of banks and bail-in bonds, however, the state has far greater leeway than in normal bankruptcy proceedings.

Conclusions

In future it would make sense to provide loss buffers not through bail-in bonds but, to an appropriate extent, through equity capital. This move would strengthen confidence in banks and reduce the probability of distress and, therefore, liquidity shortages as well. This is because a bank that has more equity and does not depend on the activation of loss buffers by a regulatory authority has a better chance of obtaining liquid funds long before a crisis occurs.

Since bail-in bonds constitute an international standard, an international solution is needed to replace bail-in bonds with equity. This cannot be achieved in one fell swoop but requires an appropriate transition period. In addition, the extent to which bail-in bonds should replace equity needs to be clarified. Since banks have to pay a risk premium when issuing bail-in bonds, as long as no general state guarantee is provided, the question is whether debt financing through bail-in bonds is actually much cheaper than equity financing, as is commonly assumed. Ultimately, this partly depends on how reforms of loss buffers would affect bank lending and thus the real economy. Academic research can help to resolve this question.

Even after this step is taken, fundamental debate about further weaknesses in the ‘too-big-to-fail’ concept would, of course, remain crucial.