The FDIC, which institutionally combines deposit insurance, supervision and resolution in the US, was established in 1933, but has undergone multiple reforms and changes over the past 88 years. So, it is not surprising
that the EU reforms in bank resolution over the past decade, including BRRD and SRM are not the final word. On Friday, I participated in an important and timely workshop, organised by the Banca d’Italia, focusing specifically on how to deal with
failing small and medium-sized banks. I was asked to discuss three papers focusing on the gaps in the current framework and possible solutions (two of them are here and here).
One major gap is that currently resolution options are only available for banks where such resolution is in the public interest (focusing mostly on financial stability concerns).
All other banks have to be sent into national insolvency proceedings, which vary quite a lot across member countries, some of them court-based and others of administrative nature. This, however, is not only inefficient but also leaves gaps, such as when the
regulator declares a bank failing or likely to fail, but the court finds the bank still solvent and it can thus not enter insolvency proceedings (one example was the Latvian ABLV bank where the ECB’s declaration that it was failing was followed by the
assessment of the Single Resolution Board that a resolution procedure was not in the public interest. While the Latvian parent shareholders decided to liquidate the bank voluntarily, the Luxembourg subsidiary was initially not declared insolvent). Further,
such loopholes give space for political inference into the process, pressure for taxpayer support on the national level, thus leading us back to the bank-sovereign linkage that the banking union was supposed to eliminate.
Second, and as Andrea Enria also pointed out in his keynote speech, “significant
differences in national legal regimes for the liquidation of banks imply divergences from the European supervisory framework; they generate level playing field concerns that might impair banking market integration and they may stand in the way of a smooth
exit from the market for the weakest players.”
A third issue is that there are restrictions on the extent to which deposit insurance schemes can contribute
to such a liquidation. Experiences in Germany and Italy have shown that a strong involvement of deposit insurance schemes in resolution frameworks (either formally or informally) can be helpful for quick and efficient intervention and resolution of failing
banks. On a broader level, Luc Laeven and I documented in a cross-country study, published in 2008, higher bank stability in countries where deposit insurers are in charge of bank resolution
and lower bank stability in countries with court-led bank insolvency frameworks. A stronger role for deposit insurers in the resolution process can thus be useful.
solutions have been suggested. One, would be to get completely rid of the public interest assessment and allow resolution tools for any bank. While dropping the public interest assessment might be a step too far, it could be used for a different purpose, such
as determining whether or not state aid is allowed. In any case, extending the possibility to use resolution tools such as purchase and assumption (P&A) to most if not all banks, is certainly an important proposal. The experience of the US in the
use of such tools has been promising, especially for smaller banks.
One controversial issue is whether the resolution framework should be further centralised to
the Euro area level. On the one hand, this would allow reducing political interference in banking further and it would allow for more effective resolution as there are more good bank candidates for a P&A operation in the Euro area than in any given country.
On the other hand, there are still significant cross-country differences in legal frameworks and market structure across countries that might make a completely supra-national approach inefficient and slow. Industry-based deposit insurance schemes and
institutional protection schemes (as, for example, in Germany) complicate things further in terms of funding and resolution. A SSM-like solution, where resolution (i.e., tools beyond insolvency) for most banks is on the national level, but under the umbrella
of the SSM, and with the largest and cross-border banks directly subject to SRB resolution, might thus be preferable to a completely centralised solution, at least in the medium-term. Further legal convergence on bank insolvency is needed, however, for
such a step.
A final institutional question concerns the structure of a future European Deposit Insurance Scheme (EDIS), which by now almost all observers see
as necessity for the completion of the banking union. One idea would be to go all the way towards the US/Canadian model of a European Deposit Insurance Corporation (EDIC), which combines resolution and deposit insurance and maybe even a role in bank
supervision. Such a centralised approach, however, might shock with the cross-country differences mentioned above. It might also require treaty changes (to my understanding and I am happy to be contradicted, the decision to house the SSM at the ECB was to
avoid such treaty changes), which is politically always tricky.
In summary, short-term reforms must focus on making the system, especially for mid-sized
banks, more efficient. As laid out in one of the three papers, this involves further harmonisation of bank resolution/insolvency frameworks across the EU and facilitating adequate liquidation funding,
including through deposit insurance schemes This is critically important on the background of Covid-related bank fragility, to be expected later this year and in 2022, but also on the background of the need for restructuring and consolidation of the
European banking system. While industry observers and insiders often dismiss the calls by regulators and academics alike for more cross-border mergers, only these will ultimately result in a true European banking market. A truly European financial safety net
is a necessary though not sufficient condition for this.