Regulating systemically important financial institutions

The financial system is populated by a wide variety of financial institutions. Of these, banks are the best-known examples, but insurers and asset managers, among others, also play an important role. However, what if financial institutions become so important that they can drag the system down in their fall? These are called systemically important financial institutions, or SIFIs for short. Their failure or even malfunction can lead to such contagion to other institutions that the whole economy incurs severe damage. This dissertation deals with the regulation of SIFIs in the EU and the US. In it, I examine which rules apply, expose any shortcomings and propose policy alternatives where necessary. The analysis is divided into the life of the SIFI: designation, regulation and supervision, and resolution.

The SIFI is born when it is designated as such by a financial regulator. Although the legal frameworks for designation are sometimes extensive and sophisticated, they focus too much on banks, especially the largest banks. This entails a risk of a false negative: some institutions may wrongly escape SIFI status because they are not large banks. I therefore propose a new designation system based on how institutions, regardless of whether they are banks or not, can cause one or more forms of contagion.

During its lifetime, the SIFI is subject to a host of additional regulation and strict supervision. The focus thereof is mainly on avoiding the failure of the SIFI, lest a chain of failing institutions (mainly banks) endanger the system. There is less attention for other forms of contagion, although the system is equally at risk when the SIFI stops providing cash because it fears illiquidity or when it triggers fire sales in the financial markets. I therefore propose to make the regulatory picture more complete with more comprehensive rules, but at the same time to apply to a concrete SIFI only those rules that can help combat the relevant form of contagion already identified in the designation phase.

Finally, the death of the SIFI occurs: it is insolvent and/or illiquid. It will often enter resolution - a quick, administrative process (no longer than a weekend) rather than longer ordinary insolvency proceedings before a court. However, this is not always the case, even though the systemic impact of a failing SIFI is obvious. Moreover, little consideration is given to the systemic impact when the SIFI has not yet failed and while it is being resolved. I therefore propose to establish an automatic link between SIFI status and resolution, and to adapt the applicable recovery and resolution procedures to account for those systemic consequences.