A look back: what Eurozone “risk sharing” actually meant

The following guest post is from Marcello Minenna, the head of Quantitative Analysis and Financial Innovation at Consob, the Italian securities regulator. The views expressed here are his personal opinions and do not necessarily reflect the views of Consob.

A recurrent debate since the eurocrisis has been whether stability tools should share risk among Member States or, instead, segregate risks within individual countries. As the eurozone discusses — and postpones — genuine risk-sharing measures, such as European deposit insurance, it’s important to look back at when risks were shared, and who actually benefited.

Risk sharing when needed

The common narrative is that rescue programs have helped deeply troubled countries avoid sovereign bankruptcy or widespread bank failures. But, by avoiding extreme outcomes, these programs also protected the banks of the core countries — Germany and France, in particular — that had accumulated huge exposures to the periphery before the crisis. At the time, risk sharing (however unpleasant) was the best available option for the governments of the core countries. It saved them from intervening (at the expense of their taxpayers) to prop up their own national banking systems.

“Risk sharing,” since the crisis, has always been a “twin bailout.” One bailout to banks on the periphery, which in turn offered another bailout to banks in the core. According to the Bank for International Settlements (BIS), in 2010 the total exposure of French and German financial institutions to Greece and its banks amounted to $120bn, more than ten times that of Italian and Spanish banks. In May that year, Eurozone governments began disbursing €52.9bn of bilateral loans to Greece; France’s share of this risk-sharing measure only slightly exceeded Italy’s. And by mid 2011, French and (less abruptly) German banks had reduced their own exposure to Greece and Greek banks by $35bn.

Then it was Ireland’s turn: in September 2010 claims of German and French banks to counterparties in Ireland exceeded $200bn. Those of Spanish and Italian investors were less than $30bn. A bailout came with the shared financial backing of the European countries. German banks reduced their private exposure by $58.7bn by 2011. Again, Germany’s share of the bailout was far less.

A few months later, Portugal’s bailout shows a similar story. (Note that Spanish investors since 2014 have sharply increased their exposure to Portugal, which today has essentially become its neighbour’s financial colony).

The same is true in Spain a little later. In the event of a default by Spanish lenders, German and French banks – exposed to $140bn and $128bn respectively – would have suffered large impairments. Not surprisingly Moody’s had revised the ratings outlook of several German institutions from stable to negative. In December 2012, the European Stability Mechanism (ESM) provided €41bn for the indirect recapitalisation of Spanish banks: funds that, in part, served to repay to German counterparties the loans generously granted before the crisis. Meanwhile countries with a marginal exposure to the Spanish financial sector were called to play their part: a €14.4bn bill for Italy, for example.

Risk segregation as the rule

During the emergency years of the crisis, eurozone countries found selective ways to share risk. But since then — as core banks have reduced their exposure to the periphery — the main decisions of the Euro-bureaucracy to counter the crisis have been in favor of risk segregation. The list includes:

Provisions on burden sharing and bail-in;
Pressing banks to dispose of non-performing loans (with consequent fire sales to vulture funds and high impairments for banks);
The systematic postponement of the pan-European deposit insurance scheme (which was supposed to have been the third pillar of the banking union), as well as proposals to make it conditional on risk reduction;
Continuous pressure on the banking systems of the periphery that, because of tight constraints on asset quality and provisioning, keep credit to the non-financial private sector at subdued levels;
Recent guidelines on full provisioning of new banks’ non-performing loans;
Discussions on the risk-weighting of banks’ sovereign exposures and on the introduction of concentration limits to these same exposures;
Proposals to introduce automatic mechanisms for the management of sovereign debt crises;
The attempt to turn the ESM into a European monetary fund with the new role of watchdog of the national budgetary policies.

But the two interventions which most contributed to the risk segregation within the periphery were Long Term Refinancing Operations (LTROs) and quantitative easing (QE).

Through the LTROs, the ECB lent the banking system more than €1trn in the form of central bank reserves, which can be used only to settle inter-bank liabilities. The idea was to help banks in the periphery cope with the sharp contraction of inter-bank credit. They did just that. A large part of what they borrowed, however, absorbed the excess supply of peripheral government debt that French and German banks were selling off, and settled commercial liabilities owed to those same banks. The other main use was to respond to the collapse of domestic deposits, which were departing for northern Eurozone destinations.

Nobody questions that QE ensured a stable and massive demand for government bonds. Yet it also created important anomalies (still mirrored by the negative yields on short- and medium-term German government bonds) and re-ignited the capital flight from peripheral to core countries. The allocation of bond purchases has favored countries (like Germany and France) with a negligible deflation at the expense of others (like Spain and Italy) much more affected by the drop in the general level of prices.

At the same time, only a marginal part of the risks inherent in purchased bonds is shared across all central banks participating into the euro system. Indeed, the large majority of bond purchases are conducted directly by the national central banks, with funds borrowed from the ECB. Each national central bank remains exposed to the insolvency risk of its own domestic government.

The key point for both programs: core countries’ exposure to the periphery has been dramatically reduced.

Measuring risk segregation

The size of all of this deleveraging can be measured by using BIS data on the consolidated position of foreign banks to counterparties residing in Italy, Greece, Spain, Portugal and Ireland.

After having accumulated a whopping credit towards the periphery in the period 2000-2008, these banks have dismantled 64 per cent of their exposures in the following decade. Indeed, at its peak (June 2008) the total exposure of the Franco-German banking system to the periphery exceeded $1.9trn; in June 2012 it had already fallen to $800bn and in the following five years, it decreased further, reaching $680bn at the end of 2017.

In terms of direct exposures, on the eve of the crisis, Germany was leading in Spain ($315.5bn), Ireland ($240.7bn) and Portugal ($52bn), while France in Italy ($553.4bn) and Greece ($86.1bn). But in reality, a good chunk of French investment in Southern Europe was channeling German savings.

The colossal disinvestment from the periphery – over $1.2trn – evokes a comparison with QE data: in July 2018 the peripheral countries (excluding Greece) were beneficiaries of security purchases for €667bn euros, just over half of the disinvestment carried out by the banks of the core countries.

What true, consistent risk sharing could look like

Since the crisis, European institutions have consistently prioritised segregating risks within the peripheral countries. The strategy was officially intended to prevent cross-country contagion. In practice, segregation has made the most vulnerable countries even less stable. Exceptions to this rule have been presented as extraordinary assistance in favor of individual peripheral States. In practice, they have been “twin bailouts”: the rescue of a specific peripheral country protected private banks in core countries that had built up strong exposures to the periphery in the years before the crisis. And backing all of the “white knight” European bailout funds were member governments, which means that many governments offered financial aid programs to countries to which their own private sector was negligibly exposed.

These episodes of selective risk sharing were not enough to make the Eurozone crisis-proof. Rather, as the ECB contemplates the end of quantitative easing, we still see market uncertainties on the compactness of the economic and monetary union.

The only antidote to this climate of mutual mistrust is to authentically inform European policies on risk sharing in both the private and public sectors. Hence, the importance of completing the banking union with the European deposit insurance scheme and opening to feasible proposals for the mutualisation of sovereign risks, such as the one I developed with Dosi, Roventini and Violi, which provides for a European Stability Mechanism supranational guarantee on public debts of all member States. This guarantee, paid at fair market conditions and conditioned upon a set of constraints to discourage moral hazard, would be a balanced solution to restore credibility to a battered periphery.