Mark Hallerberg and Christopher Gandrud explain the logic of politically important financial institutions (PIFIs).
Over the weekend of 28 February to 1 March, the Austrian government refused to help Heta Asset Resolution fill a capital hole discovered by the financial regulator. The refusal was an about-turn following almost 10 years and billions of euros of public assistance to Heta and to Hypo Alpe Adria – the bank that Heta was created to help restructure.
Public sector bailouts of nationally or globally systemically important financial institutions (SIFIs) can be justified. Without intervention, the financial sector of a given country might collapse and bring the whole economy down. For this reason, Basel III now requires SIFIs to meet higher capital requirements than other banks. In the United States, Dodd-Frank obliges such banks to design contingency plans in case they get into trouble. In Europe, the roughly 130 largest banks that are systemically important face supervision by the European Central Bank’s Single Supervisory Mechanism (SSM), while the Single Resolution Mechanism (SRM) prescribes when to assist a failing bank and under what terms.
But the decision on whether to let a bank fail rarely depends alone on whether it is a national or global SIFI. Politicians also care about the local effects of a bank failure. Where banking regulation is sub-national and matches electoral boundaries, a given bank might be inconsequential for the health of the overall national economy, but critical to politicians making the decision on how to proceed with a given troubled bank. In such cases, politically important financial institutions (PIFIs) might receive benefits they otherwise would not get.
The PIFI logic drove the series of bank assistance packages in Germany. Despite a popular perception of Germany as an adamant opponent of bank bailouts, considerable public support was given to German banks during the euro-area crisis. In fact, only three very small banks were initially allowed to fail. The structure of German federalism and politicians’ banking sector competencies explain this apparent contradiction. German banks, such as WestLB and HSH Nordbank, that were not systemically important from a national perspective were nonetheless prevented from failing at the onset of the recent crisis because they were important to Länder-level politicians.
We are seeing the same dynamic unfolding in Austria with Hypo Alpe Adria. The Austrian state of Carinthia part-owned the lender, provided it with generous guarantees, and used it for pet projects of the party in power, namely Jörg Haider’s FPÖ. A German Landesbank, Bayern LB, bought the bank in 2007. When it got into substantial trouble during the 2008-09 financial crisis, Hypo Alpe Adria remained politically important and it received public financial support from both the state and national governments. It was ultimately nationalised in 2009.
The story, however, has a new twist. The 28 February-1 March revaluation of the remaining assets in Hypo’s bad bank Heta Asset Resolution means that the national government is theoretically on the hook for an additional €7.6 billion. This time, the national government baulked at supporting the institution, which has become a political liability. A true bail-in would mean that the Landesbank based in Munich would be expected to pay something like €2 billion. A series of lawsuits has followed.
Of course, some in other EU member states might experience schadenfreude over the banking problems in northern ‘surplus’ countries. But they should be careful. While this case involved Carinthian and Bavarian state governments, one can imagine a parallel with the European Union. A national government is concerned about one of its banks and takes measures to help it. The bank gets into even worse trouble, and the euro-area single supervisor orders that it be resolved. At the beginning of the process, the SRM expects a bail-in of shareholders. One can imagine a scenario in which a true bail-in would hurt significant shareholders, and hence politicians, in a neighbouring country. Would they simply pay up, or would there be the same sort of legal fights one now sees in Austria and Bavaria, with any true resolution still some years away?
Also, note that the Hypo Alpe Adria funding gap is now bigger than it should have been precisely because the bank was a domestic PIFI. This meant that its lending decisions suffered from moral hazard and it was given public support to enable it to hobble along for much longer than it should have. Once again, the parallel to the European level is clear. If a troubled bank is not part of the original 130 under the SSM, it could be that European authorities only learn about its problems after much delay and when the overall losses are much bigger precisely because it was not wound up earlier.
This text was first published on the Bruegel Blog on 6 March 2015.
 As we detail in: Deo, Sahil, Christian Franz, Christopher Gandrud, and Mark Hallerberg (2015) ‘Preventing German Banks Failures: Federalism and decisions to save troubled banks’, Politische Vierteljahresschrift 2/2015, forthcoming.