A proposal for ending the privileges for sovereign exposures in banking regulation

Jochen Andritzky, Niklas Gadatsch, Tobias Körner, Alexander Schäfer, Isabel Schnabel

04 March 2016



The close interrelationship between sovereigns and banks is seen as a major reason for the severity of the Eurozone Crisis. Reducing the risk of bank bailouts endangering sovereign solvency has been one of the central aims of the European Banking Union (European Council 2012). However, the reverse channel – sovereign debt crises endangering banks’ solvency – has yet to be addressed. Severing the sovereign-bank nexus in both directions is needed to further strengthen the stability of the banking system. It would also facilitate sovereign debt restructurings and thus underpin the credibility of the no-bailout clause (GCEE 2015).

Yet, the sovereign-bank nexus appears larger than before. Banks’ sovereign exposures increased in the years after the Global Crisis (Acharya and Steffen 2015, ESRB 2015). A comparison of bank-by-bank data provided by the European Banking Authority (EBA) shows that the sovereign exposures of large banks did not change substantially between end-2013 and mid-2015. Banks in most EZ countries still invest more than their own funds in sovereign debt (EBA 2015; see Figure 1, left panel). In addition, banks’ sovereign debt holdings exhibit a strong home bias, particularly in the southern European countries and in Germany. While the associated risks as indicated by credit ratings vary substantially between countries (Figure 1, right panel), current banking regulation treats all these assets as free of risk.

Figure 1. Sovereign exposures of banks in selected Euorzone member states

Benefits of a regulatory reform

Current banking regulation in Europe features a considerable number of provisions privileging sovereign exposures vis-a-vis exposures to private debtors. For example, exposures to private debtors must not exceed 25% of a bank’s eligible capital; this limit does not apply to sovereign borrowers. In addition, there are no capital requirements for sovereign exposures in domestic currency. Finally, sovereign exposures are considered entirely safe and liquid in the new liquidity regulation.

The removal of these privileges would further contribute to loosening the sovereign-bank nexus. Several policy options have been discussed recently by regulatory and advisory bodies (e.g. ESRB 2015, Deutsche Bundesbank 2015). In particular, the discussion on directly limiting exposures of EU banks to sovereign debtors has gained momentum (see European Commission 2015). A reform of the regulatory treatment of sovereign exposures should serve three objectives:

  • Reduce concentration risks on bank balance sheets in order to prevent the insolvency of a member state from bringing about the insolvency of a bank;
  • Increase banks' loss-absorption capacity in order to be able to better cope with a sovereign default; and
  • Reduce price distortions in order to mitigate incentives for sovereigns to borrow excessively and for banks to lend preferentially to governments.

A reform proposal

The German Council of Economic Experts (GCEE) has developed a proposal for removing privileges for sovereign exposures resting on two key elements (GCEE 2015): risk-adjusted large exposure limits, and risk-adequate regulatory capital requirements. Large exposure limits are crucial to reduce the sovereign-bank nexus by limiting concentration risks. In addition, regulatory capital requirements increase banks’ loss absorption capacity and reduce price distortions. Capital requirements are to be based on the Basel sovereign risk weights. These are lower than the risk weights for corporate borrowers (Table 1). It should be noted that the Basel III leverage ratio, expected in 2019, already implies a regulatory capital requirement for sovereign exposures. For example, given a risk-weighted capital requirement of 8%, a leverage ratio of 3% would imply a fixed risk weight of 37.5%.

The large exposure limits are to vary with the sovereign’s default risk because concentration risks in bank balance sheets primarily represent a threat to financial stability in the event of a significant threat to a sovereign’s solvency. The default risk could be determined by country ratings or alternative indicators that are not prone to manipulation. Different levels of government should be viewed as a single unit if default risks are strongly correlated or a separate treatment enables regulatory arbitrage.

For the countries with the lowest credit standing, the GCEE proposes the same large exposure limit as that for corporate exposures (25% of eligible capital). The limit gradually increases to up to 100% of eligible capital for countries with the best credit standing, using relative distances between risk categories as in the Basel sovereign exposure risk weights (Table 1).

Table 1. Proposed large exposure limits and risk weights for sovereign exposures

Mitigating procyclicality

An important lesson from the Global Crisis is that regulatory measures should not induce adjustment reactions that deepen a crisis. For instance, in the presence of large exposure limits, banks could be forced to divest sovereign exposures if their capital base contracts in a crisis. This procyclical effect could in turn increase governments’ financing costs and inhibit countercyclical fiscal policy. In addition, there could be abrupt credit rating adjustments in a crisis triggering sales of sovereign exposures to comply with exposure limits.

To dampen this effect large exposure limits should be based on long-term averages of sovereign ratings and own funds which would smooth any portfolio adjustments required under the regulation.

Risk-weighted capital requirements have conceptually the same procyclical effect for sovereign exposures as they have for private ones. This problem should therefore be solved within the existing macroprudential toolkit, i.e. via time-varying buffers.

Quantitative impact of the reform

An analysis of banks’ balance sheet data from the EBA transparency exercise allows estimating the quantitative impact of the reform measures based on a bank sample of large banks, which represent about 70% of total EU bank assets (see Figure 2).

The current snapshot of banks’ sovereign exposures suggests that about €565 billion of sovereign debt exceeds the risk-adjusted large exposure limit based on rating averages as proposed by the GCEE. This amount is significantly smaller than the €1,118 billion of excess exposures under a 25% fixed large exposure limit, but is still substantial. Italy and Spain as well as German state-owned banks account for the majority of the excess exposures.

The overall excess exposures represent at most 10% of the respective country’s outstanding sovereign debt (see the red dots in Figure 2 corresponding to model 4). If the new rules were to be phased in over, say, ten years, it is likely that banks could achieve compliance by allowing their exposures to mature rather than disposing them actively. An adjustment path could be specified for the phase-in period, starting, for example, from three times the final limits, which are then gradually reduced. Capital requirements could be introduced with a grandfathering clause so that only new exposures would be subject to the requirements. The cut-off date would have to be in the past in order to avoid an incentive to stockpile still privileged sovereign exposures.

Figure 2. Sovereign exposures exceeding large exposure limits

Given current exposures, additional capital requirements for EU banks in the EBA sample would amount to €35 billion, which is approximately 2% of total own funds. This volume seems manageable compared to previous capital increases, such as in the run-up to the comprehensive assessment, which amounted to around €50 billion of common equity tier 1 (ECB 2014). The low volume implies, however, that only a small amount of loss absorption capacity would be created. This reinforces the point that exposure limits, rather than capital requirements alone, are key to severing the sovereign-bank nexus.


Abolishing regulatory privileges for sovereign exposures of banks is an important further step towards severing the sovereign-bank nexus. To this end, we propose introducing risk-adjusted large exposure limits for sovereign exposures as the key element to protect banks from sovereign risk. In addition, capital requirements should be introduced by activating existing Basel risk weights for sovereigns.

Our proposed exposure limit is simple and transparent; it accounts for the importance of safe and liquid bonds in banking and addresses the problem of procyclicality. Since capital requirements for banks will create relatively little loss-absorption capacity, large exposure limits are crucial. In addition, risk-adequate capital requirements for sovereign exposures help to reduce price distortions vis-a-vis private borrowers.

A uniform regulation of sovereign exposures at global level (i.e. in the Basel Committee) would be desirable. However, aiming first for an agreement at the EU level would enable its introduction in the near future. As the problem has its roots in the nature of the monetary union, namely the combination of largely sovereign states and a central monetary policy, along with the resulting incentive problems, introducing the regulation in the Eurozone only would also be conceivable. The advantages from the stabilisation of the Eurozone are likely to outweigh the supposed competitive disadvantages to banks from stricter regulation. The reform should be phased in now, not after the next crisis.

Authors’ note: This column is based on the German Council of Economic Experts’ Annual Economic Report 2015/16, chapter 1.IV: Paths to more stability in Europe.


Acharya V and S Steffen (2015), “The “Greatest" Carry Trade Ever? Understanding Eurozone Bank Risks”, Journal of Financial Economics 115 (2), 215 - 236.

Deutsche Bundesbank (2015), Annual Report 2014, Frankfurt /Main.

EBA (2015), “2015 EU-wide transparency exercise”, European Banking Authority, London.

ECB (2014), “Aggregate report on the comprehensive assessment”, European Central Bank, Frankfurt/Main.

ESRB (2015), “ESRB report on the regulatory treatment of sovereign exposures”, European Systemic Risk Board, Frankfurt /Main.

European Commission (2015), “Further Risk Reduction in the Banking Union”, Five Presidents’ Report Series, Issue 03/2015, European Political Strategy Centre, Brussels.

European Council (2012), “Euro area summit statement”, Brussels, 29 June 2012.

GCEE (2014), “More Confidence in Market Processes, Annual Economic Report 2014/15”, German Council of Economic Experts, Wiesbaden. Chapter V: The long road to more financial stability in Germany and Europe.

GCEE (2015), “Focus on Future Viability, Annual Economic Report 2015/16”, German Council of Economic Experts, Wiesbaden. Chapter I: Economic policy: Focus on future viability.



Topics:  Financial regulation and banking

Tags:  banking regulation, sovereign exposure, capital requirements

Secretary General, German Council of Economic Experts

Economist at the German Council of Economic Experts

Economist, German Council of Economic Experts

Economist, German Council of Economic Experts

Professor of Financial Economics, University of Bonn; Member of the German Council of Economic Experts; CEPR Research Fellow