Article content
How has the feedback loop between banks and sovereigns changed since the crisis years?
One aim of Banking Union is to weaken the feedback loop between banks and their sovereigns so that increases in banks’ credit risk would no longer be reflected in sovereign credit risk and, conversely, banks’ financing costs would no longer be driven by their sovereign's creditworthiness. Currently, for banks and sovereigns alike, credit risk insurance costs much less than during the crisis. Although the bank-sovereign nexus has weakened, the feedback loop cannot be considered to be broken.
The bank-sovereign nexus cannot be broken by monetary policy
The feedback loop between banks and their sovereigns caused the financial crisis to escalate into a sovereign debt crisis in the euro area. In some countries, the problems arose from a major growth in bank lending, as well as from poor risk management. In these countries, the central government had to provide substantial financial assistance in order to prevent a collapse of the banking sector that would have shaken the whole financial system. In countries where the root cause of the problems was excessive government indebtedness, domestic banks ultimately ensured their sovereign’s access to financing. In both cases the outcome was identical: both banks and the sovereign ended up in significant distress, and external financial assistance was required to solve the problem.
As the global financial crisis escalated following the collapse of Lehman Brothers, euro area banks also saw their cost of hedging credit risk increase significantly. Soon thereafter, the price of credit risk insurance began to increase tangibly also for the respective sovereigns (Charts 1 and 2).
For stressed economies, such as Spain and Italy, the pricing of credit risk embarked on a markedly downward path at the end of July 2012, following the – epochal, as it later turned out – speech by the ECB's President Mario Draghi, in which he said that the ECB would do within its mandate whatever it took to preserve the euro. The message was further reinforced in September 2012 when the ECB announced the Outright Monetary Transactions (OMT) programme which created the possibility for the Eurosystem to purchase, if necessary, debt securities issued by countries hit by market turbulence.The conduct of purchases for monetary policy reasons under the OMT programme requires conditionality which aims at ensuring the debt sustainability of the targeted sovereigns. Thus far it has not been necessary for the ECB to conduct purchases under this monetary policy programme. However, the Eurosystem's intervention was crucial in helping decrease the price of credit risk both for sovereigns hit by the crisis and for banks located in these countries. Despite this, the OMT programme as such does not address the fateful feedback loop between banks and their sovereigns.
Banking Union aimed at weakening feedback loop
It was decided in summer 2012 that Banking Union would be established in the euro area. Over the short term, its creation was aimed at stabilising the financial markets, whereas the longer-term objective was to facilitate efficient financial intermediation and help loosen the bank-sovereign nexus in the euro area.
Back then, the euro area countries agreed upon single banking supervision and crisis resolution mechanisms, but the third pillar of Banking Union, a shared deposit guarantee scheme, was still left outside the scope of joint decision-making.For more information on the shared deposit guarantee scheme, see Construction of a risk-based European Deposit Insurance Scheme. The Single Supervisory Mechanism is responsible for the direct supervision of the largest and most significant banks in the euro area, while the Single Resolution Mechanism is tasked with the restructuring or, if necessary, the orderly liquidation at minimum cost of banks which are no longer financially sound.
In 2014, the ECB conducted a comprehensive assessment of the condition of euro area banks in cooperation with national supervisory authorities. Following the assessment, banks for which a capital shortfall had been observed had to raise additional capital. A precondition for any bank to enter the Banking Union was that it had ensured its viability. The results of the comprehensive assessment were disclosed in October 2014, and the Single Supervisory Mechanism began operating at the beginning of November.
The Bank Recovery and Resolution Directive (BRRD), which has been implemented at EU level, plays a central role in breaking the feedback loop between banks and their sovereigns by defining the shared tools available for handling distressed banks, as well as the related powers. Since the beginning of 2016, the bail-in rules embedded in the resolution mechanism limit the costs caused to taxpayers in the event of a bank collapse. Under these rules, the bail-in tool (debt write-down) is activated in situations of bank distress. In this way, the escalation of a bank’s problems should be less heavily reflected in sovereign credit risk than before.
The third pillar of Banking Union, a European deposit guarantee scheme, is currently under discussion. The lack of common deposit guarantee is one element maintaining the feedback loop between banks and their sovereigns. This feedback loop is further reinforced by the fact that banks in many countries still have large holdings of bonds issued by their sovereigns.
Crisis tools had their first real test
In the course of 2015 and especially at the beginning of 2016, the price of hedging credit risk associated with subordinated debt increased on the back of investors anticipating the entry into force of the bail-in rules (Charts 3 and 4). At the beginning of 2016, interbank differences in the cost of hedging credit risk also increased. At the same time, the price of credit risk associated with Spanish, Italian and French sovereign debt remained relatively stable, which seems to point to a weakening of the feedback loop.
The new crisis tools underwent their first real test in June 2017 as some Spanish and Italian banks began to encounter problems. In Spain, the control over Banco Popular was transferred by an ECB decision to the Single Resolution Board (SRB). In June, the ECB announced that two Italian banks, Veneto Banca and Banca Popolare di Vicenza, were failing or likely to fail. The SRB considered that these banks were not large enough to cause systemic risk in the financial system. If the conditions of the resolution procedure are not met, the resolution is conducted under national law.
Although in Spain the financial responsibility associated with the resolution is borne by investors and in Italy still in part by taxpayers, in both countries the new resolution procedure led to a decrease in CDS spreads for senior and subordinated debt alike.
Has the feedback loop weakened?
The feedback loop between banks and their sovereigns can be examined by analysing how changes in the price of sovereign credit risk correlate with changes in the price of bank credit risk over different periods of time. The evolution of prices is affected by several factors, and the correlation between different elements does not reflect a causality between them. In addition to a simple correlation, the link between changes in the prices of sovereign and bank credit risk can be examined by means of a principal component analysis in which the price path is adjusted for the impact of common components.
The outcome of the analysis was that, adjusted for common components, the correlation between banks and their sovereigns (for instance Italy and Spain) seemed to have decreased tangibly after the launch of Banking Union in November 2014 (Chart 5). Following the entry into force of the bail-in rules on 1 January 2016, the direct impact on the correlation seems to have been moderate, but the impact may have materialised with a delay in the aftermath of the first resolution cases in which the new model was applied.
Examining correlations between the prices of hedging sovereign and bank credit risk is only one, highly simplified, way of analysing the evolution of the feedback loop. The correlation analysis presented here does not imply any causality between the observations discussed. Moreover, the weakening of correlations does not necessarily mean that the feedback loop cannot strengthen again in the future. Reforms undertaken in the euro area have loosened the feedback loop in the banks-to-sovereigns direction, but only marginally in the opposite direction. A more complete breaking of the feedback loop remains conditional on the completion of Banking Union and cross-border banking in the euro area.