This blog post coauthored with Anna Gelpern was published last Thursday by Bruegel, and this morning (in slightly modified version, below) by the Peterson Institute.
Update (January 2021): a lightly edited version was republished in CESifo Forum, also in PDF (downloadable here here if the previous link is broken).
Six years after starting a reform process to protect its banking system against financial crises, the European Union (EU) has reiterated its members’ commitment to “make further concrete progress on the Banking Union by the end of the year.” EU officials are right not to let COVID-19 derail necessary debates over this objective. But the reinvigorated discussion has become increasingly confused when it comes to dealing with failed banks. There is a danger that the EU could cite the experience with the US Federal Deposit Insurance Corporation (FDIC) to make its already fragmented regime even more fragmented. That would be a mistake. A closer look at the FDIC model highlights the value of a unitary process for resolving all deposit-taking banks, no matter how large or small.
The banking union is intended to pool the instruments of banking sector policy at the European level. Otherwise, bank failures can require expensive bailouts that wreak havoc on national budgets in a crisis, undermining the integrity of the euro area. The implementation of the banking union began in 2014, with the assumption of bank prudential supervision by the European Central Bank (ECB). The pandemic posed a major test, but ECB banking supervision has passed it so far by quickly granting banks leeway to absorb pandemic-related losses, while suspending their dividend distributions to preserve their capital.
EU bank failures need to be much better managed
Still, the bank crisis management framework remains a halfway house. Most of it is enshrined in the Bank Recovery and Resolution Directive (BRRD) of 2014. In a number of bank failures since that legislation went into effect, the BRRD has fallen short of its principal goal of forestalling taxpayer bailouts. Hence the growing consensus for more significant reform.
Several officials, including at the Bank of Italy, the German finance ministry, and the Single Resolution Board (SRB)—the Brussels-based EU agency that acts as a hub for BRRD implementation in the euro area—have proposed a new EU bank liquidation regime as a centerpiece of reform, with implicit or explicit references to the FDIC model. There is an inescapable irony to invoking the FDIC for that.
Established in the 1930s, the FDIC insures deposits, undertakes bank supervision, and oversees the apportioning of costs to creditors, investors, depositors, and others when a bank fails, a process known as bank “resolution.” Its success had helped inspire the process established by the BRRD in the first place. But unlike in the US, EU legislators decided that the BRRD resolution procedure would apply only to banks judged to implicate the public interest after a “public interest assessment” (PIA).
That assessment is guided by the vaguest of criteria. Whereas the FDIC is the sole resolution authority, regardless of the size or systemic importance of a bank, or whether it has a state or federal charter, the BRRD left it up to national regimes to resolve a failing bank that receives a negative PIA. The FDIC is also responsible for the as-yet untested Orderly Liquidation Authority for systemically important nonbank entities, including large bank holding companies. For banks, the FDIC has no equivalent to the BRRD’s PIA process that might allow it to hand over the failing institution to a less squeamish undertaker.
The incentive structure behind recent bank failure controversies reflects the sort of dysfunction of a multiplicity of potentially overlapping regimes.
Dealing with a failing bank is thankless business, especially if one lacks access to unlimited public resources to bail out the various stakeholders. The resolution process under BRRD severely restricts bailout options. Some national insolvency regimes are less stringent and leave the door open to generous bailouts. This process puts pressure on the authority in charge of making public interest assessments to make a negative PIA and keep the ailing bank out of EU resolution.
For example, in June 2017, the SRB gave a negative PIA on two mid-sized banks in the Veneto region of Italy, followed by their administrative liquidation under Italian law. The latter process was managed by the Bank of Italy, with generous financial support from the Italian government. This was in line with the letter of BRRD but at odds with its spirit: When presenting that legislation on June 6, 2012, European Commission President José Manuel Barroso stressed that it would “help protect our taxpayers (…) from the impact of any future bank failure.” It defies common sense to declare that a bank does not implicate the public interest only to have it benefit from more, not less, public financial support.
Unlike the SRB, the FDIC cannot wash its hands of a failing bank—there is no one else to handle the mess. The unitary structure lends itself to formal and informal public accountability and has led to continual reform and gradual improvement in FDIC’s practice over several cycles of bank failures that now span more than eight decades. For its part, the SRB not only has defended its decision on the two Veneto banks but also has elevated it to a point of general policy, with the SRB chair emphasizing that BRRD resolution was “for the few, not the many.” This position leaves many significant banks in the banking union beyond the reach of one of its key institutions, contrary to the expressed initial intent of BRRD legislators.
The absence of a common deposit insurance authority in the euro area compounds the regime arbitrage problem. As its name indicates, the FDIC manages deposit insurance for all banks in the United States. By contrast, the 21-country banking union (19 countries in the euro area, plus Bulgaria and Croatia) has national deposit insurance regimes (in some countries, several of them), national resolution authorities, national institutions in charge of insolvency processes, plus the SRB: countless cooks in the bank failure kitchen, where the United States has only one.
EU reformers should take the time to draw on the core strengths of the US FDIC model
We used the FDIC as a starting point for our analysis, further detailed in a paper published last year for the European Parliament. This blog post does not aim to address all dimensions of the technically complex matter. We nevertheless submit three suggestions for the EU reform debate.
First, policymakers should not rush for a piecemeal solution at a time when pandemic-related risks loom large. Completing the banking union before the pandemic was arguably the most important priority of the European commissioner for financial services. But now, the more immediate priority is to address the COVID-19 crisis, including by implementing the Next Generation EU blueprint for pooled borrowing by the EU and financial transfers to its neediest members. If recovery stalls and economic deterioration leads to bank failures or requires bank recapitalization, that will probably be before any significant banking union reform can be enacted, so they will have to be handled with the existing legislation anyway.
Second, EU reformers should consider the tradeoffs embedded in the design of the FDIC and its evolution over time, including stronger protection of all deposits, even uninsured ones, but also lesser implicit protection of other creditors—the FDIC’s track record establishes that its pledge not to bail these out is credible, at least for institutions up to a fairly significant size (Washington Mutual, resolved in 2008, had around $300 billion in assets).
Third, the EU should take away the right lesson from the FDIC’s history and pursue an integrated approach to the banking union: a unitary regime to handle all bank failures, amending and improving the BRRD resolution concept, encompassing reform of mandatory deposit insurance that would integrate it under the SRB. All things equal, a European system that would match the FDIC’s performance would entail a lower future fiscal impact of banking crises. To be sure, it would still entail financial risk-sharing through the deposit insurance and resolution mechanism and its necessary public backstop, but Next Generation EU will facilitate that by giving the EU financial firepower of its own.
The advocacy of a new EU bank liquidation regime, somewhere between EU resolution and national insolvency procedures, evokes the ill-starred Council of Pisa in 1409, which decided to elect a third Pope to solve the conflict of two competing Popes and in so doing exacerbated the Western Schism. The solution came several years later at the Council of Constance, where all three papal claimants resigned and gave way to a single newly elected Pope. European reformers should go directly for the Council of Constance approach, drawing on the core substantive strengths of the FDIC model. If that requires more time for careful debate and preparation, it will be time well spent.