A Reinsurance Approach to the Single Resolution Mechanism
Daniel Gros recently published an interesting column contrasting recent banking crisis in Ireland and the US state of Nevada concluding that an integrated banking system saved Nevada after a local real estate boom turned to bust whereas Ireland had to deal with a similar crisis on its own. In Nevada transfers from the Federal authorities amounted to about 10% of Nevada’s GDP, but the state didn’t need a bailout whereas the government of Ireland had to ask for a loan package from the European Commission and the IMF as a result of the cost incurred while saving the banking system. Gros went on to propose a full banking union for Europe warning that the alternative is a return to nationally segmented financial markets.
We agree that the banking union is an important missing piece in the institutional setup of the eurozone. This has also been recognized by the European authorities and both the single supervisory mechanism and the single resolution mechanism are important parts of the blueprint for a deep and genuine economic and monetary union recently published by the European Commission. However, so far politicians have not even been able to agree on the design of the first pillar of the banking union – the single supervisory mechanism.
When designing the banking union in Europe, important differences between a fully federal system such as the one in the US and a union of nation states will have to be kept in mind as Europe is unlikely to move towards a full federation in the foreseeable future. For example, while there is no separate financial markets regulator in Nevada, national regulators will continue to play an important role in the eurozone even after the single supervisory mechanism is established. This has important implications not only for the design of the single supervisory mechanism, but also the two other pillars. In fact, attempting to design one part separately from the other parts might be one of the key reasons that currently prevent an agreement on the banking union from being reached.
The Single Supervisory Mechanism in the euro zone to be different from that in the US
In a recent speech Mario Draghi admitted that while all banks established in the participating Member States would in principle fall within the remit of the single supervisor, the intensity of centralised supervision would differ depending on the systemic importance of a particular institution.
„The intensity of centralised supervision will be the highest for those institutions deemed systemically important. But also for these institutions, close cooperation with national authorities on daily activities will remain. The role and responsibilities of national supervisors will increase when going down the dimensional scale, to banks of predominantly national or local relevance.„ (Draghi, 2012)
In other words, the single supervisor cannot be expected to carry the main regulatory responsibility for all of the more than 6000 financial institutions in the euro area, but just for a few hundred systemically important institutions. Furthermore, even in case of supervising the systemically important institutions the national regulators will still play an important role. This setup can help avoid incentive problems similar to the ones encountered in the US when mortgages were securitised and the risk no longer held by those institutions that had originated the loans. However, it also has important implications for the joint deposit guarantee scheme and the single resolution mechanism, which are two other key parts of the banking union design. In particular, it is unreasonable to require a central financial backstop to cover all losses while the national regulators still remain to a large extent responsible for the supervision of national or local banks.
The Single Resolution Mechanism needs to be aligned with the Single Supervisory Mechanism
To align the supervisory mechanism with the resolution mechanism one could draw some lessons from reinsurance. There are two main types of reinsurance: proportional and non-proportional. Under proporational reinsurance the reinsurer covers a certain percentage of claims arising from the policies underwritten by the insurer that are subject to the reinsurance contract. Under non-proportional reinsurance, in particular, excess of loss resinsurance, the reinsurer covers losses that exceed a certain amount either on aggregate or per policy (or per event) basis. In both cases the original insurer retains a part of the risk.
The parallel we see with the joint deposit guarantee scheme and the single resolution mechanism for banks in the eurozone is as follows: while national governments would still keep a part of the risk of the collapse of a particular bank or the banking system, the centralized financial backstop in the euro zone would act as a reinsurer preventing the losses from a major bank failure in a country from compromising the solvency of the sovereign. At the same time the national authorities would still have the incentives to provide the national supervisors with the necessary resources and ensure their independence.
In particular, the mechanism could function as follows:
- in case of the systemically important banks for which the responsibility of supervision would largely be transfered to the single supervisor, the centralized financial backstop would provide „proportional reinsurance”, i.e., cover a substantial portion of the cost of resolution of a troubled systemic bank
- in case of the national and local banks for which the responsibility of supervision would still largely rest with the national regulators, the centralized financial backstop would provide „non-proportional reinsurance”, i.e., cover just the excess of loss, for example, costs of banking resolution in a nation that within a year exceed a certain percentage of GDP
An approach that would take a holistic view of the banking union and try to design all the elements of the banking union together might have a higher chance of succeeding. In particular, viewing the central financial backstop in the banking union as a reinsurer rather than a mechanism of collectivizing all losses might help reconcile some of the opposing views that currently prevent an agreement on the banking union from being reached.
Draghi, Mario, „Rationale and principles for Financial Union”, Speech at the 22nd Frankfurt European Banking Congress, 23 November 2012
European Commission, „A blueprint for a deep and genuine economic and monetary union. Launching a European Debate”, 28 November 2012
Gros, Daniel, „Banking union: Ireland vs Nevada, an illustration of the importance of an integrated banking system”, VoxEU.org, 27 November 2012
2 Responses to “A Reinsurance Approach to the Single Resolution Mechanism”
Excellent article. I fully agree that bank supervision and resolution is a key missing element of the EZ setup. Perhaps the leading question is whether this element (and others like fiscal union features) can be added retroactively, or whether starting the EZ without them was a fatal flaw that will end it its collapse.
In one respect, your argument may be even stronger than you claim. You write "For example, while there is no separate financial markets regulator in Nevada." Actually, I think that is not quite true. Nevada (like most states) does have a financial markets regulator, in this case the Nevada Financial Institutions Division http://www.fid.state.nv.us/ . Its mission statement reads as follows:
"The mission of the Financial Institutions Division is to maintain a financial institutions system for the citizens of Nevada that is safe and sound, protects consumers and defends the overall public interest, and promotes economic development through the efficient, effective and equitable licensing, examination and supervision of depository, fiduciary, and non-depository financial institutions."
A common criticism of US banking supervision is that there are too many agencies, not well enough coordinated. They include not only the Fed, the FDIC, and the Comptroller, but also all the state supervisory agencies. If I remember correctly, in the 1980s, failures of state-level regulation (Ohio and Maryland stick in my memory) were a big problem, but as far as I know the state agencies did not play a big role in regulatory failures in 2008. If there have been reforms of the federal/state supervisory relationship that have eliminated failures, that would be relevant to the EZ case. I confess, I haven't looked into this, but it might be interesting to follow up.
I would think that the most important thing to avoid would be any situation where banks and other FIs could "shop around" among agencies for the one that was most permissive. That could be avoided by clear-cut rules, say, based on size, that define who regulates whom. Draghi's formulation, "The role and responsibilities of national supervisors will increase when going down the dimensional scale, to banks of predominantly national or local relevance," strikes me as a little too elastic and might leave too much room for shopping around.
Sorry for the late comment, but this is a pet issue of mine and I'm just now reading this.
You and Gros are spot on except perhaps that you should be stating the case more strongly. The real heart of the Euro crisis has always been the combination of a federal currency with state banking systems. The intractable problem is capital flight from individual states that drain their state banking systems of cash.
Others have been telling a bungled story of European national central banks purportedly not being able to print euros. They can and they have printed plenty, mainly through refinancing, plus some emergency liquidity assistance. But those euros don't stay within their state banking systems. They flee mainly to German banks.
There's no quick fix to this major design flaw. Even if Europe were given to making hasty reforms, it would still take decades for a European banking system to develop.