VoxEU Column EU institutions EU policies

Refocusing the debate on risk sharing under a European Deposit Insurance Scheme

Recent months have seen many proposals for alternative designs for a European Deposit Insurance Scheme that would cater for concerns that such a scheme would lead to some banking sectors having to bear the costs of bank failures in other member states. This column, adding to VoxEU's Euro Area Reform debate, asks whether these concerns are well founded.

This column is a lead commentary in the VoxEU Debate "Euro Area Reform"

As the deadline for the June Euro Summit is approaching, expectations are rising on European leaders to see whether or not the Banking Union will finally be completed with its third Pillar: a European Deposit Insurance Scheme (EDIS). 

In recent months, we have seen many proposals (e.g. European Parliament 2016, Bénassy-Quéré et al. 2018, Schnabel and Véron 2018, Schoenmaker 2018) to break the deadlock by suggesting alternative designs for an EDIS, which would cater for the concerns that such a scheme would lead to some banking sectors having to bear the costs of bank failures in other member states.

But are those fears well-founded? In a recent paper, we seek to put such concerns into perspective (Carmassi et al. 2018). 

To do so, we follow a two-step approach. In the first step, we calculate the exposure of the EDIS to bank failures using banks’ estimated probabilities of default and loss given default.1 This allows us to measure the resilience of the EDIS. In a second step, we estimate risk-based contributions to the EDIS based on the European Banking Authority (EBA) Guidelines developed for the Deposit Guarantee Scheme Directive (DGSD). Comparing the EDIS’s exposure derived in the first step to banks’ contributions allows us to identify whether there is cross-subsidisation, in the sense of some banking system systematically contributing less than they would benefit from the EDIS under certain simulated loss scenarios. 

Would an EDIS be big enough in another financial crisis?

We look at how a fully mutualised EDIS, with a target size of 0.8% of covered deposit, would be affected in a crisis in the steady state. We simulate various shocks, among which is a scenario under which the 10% riskiest banks in our sample simultaneously fail.Our analysis considers losses from 5% to 25% of total assets for resolved banks (bail-in), which is an extremely conservative range if compared to the values observed in the last crisis.2 In addition, we also consider the same amount of losses simultaneously affecting all the failing banks.

We find that a fully-funded deposit insurance fund would be sufficient to cover losses in crises even more severe than the 2007-2009 Global Crisis. And this would also be true in case of simultaneous country-specific shocks. 

This is partly due to the substantive risk reduction that has already taken place in the banking system – notably the introduction of bail-in and the corresponding increase in banks’ loss absorbing capacity, the creation of a resolution framework, and the super-priority provided to covered deposits. These all contribute to reduce the likelihood of tapping into the EDIS. 

This doesn’t mean that the EDIS would not be relevant. On the contrary, its purpose is precisely to enhance depositors’ confidence, thereby avoiding risks of bank runs and contagion, and to serve as a credible line of defence in case of large shocks that would overwhelm national schemes. 

Would a fully-fledged EDIS lead to cross-border subsidisation?

The way banks’ contributions to the deposit insurance fund are designed is crucial: we calculate risk-based contributions according to different risk-adjustment factors and benchmarked at the banking union level, following a ‘polluter-pays’ principle.

Comparing these risk-based contributions to the EDIS exposure shows that there would be no unwarranted systematic cross-subsidisation across member states. While there are some cases in which the contributions of a banking system are lower than the amounts which would be received from the EDIS, this is only the case in extreme scenarios. 

Two elements are driving these results. 

  • First, the methodology used to calculate contributions can take into account several indicators linked to risk reduction in the banking system, such as the level of banks’ loss-absorbency, their ratio of non-performing loans, or the strength of national insolvency frameworks. We show in our analysis that the existing EBA methodology developed for DGS can be used for these purposes. 
  • Second, bank contributions are benchmarked on a banking union basis, and are not exclusively related to national peers. This makes it possible to take into account the relative riskiness of the banking sectors, and to reduce the risk of unwarranted cross-border subsidisation.

We have received the criticism that our proposed bank contributions would be pro-cyclical, as weaker banks would by design have to pay more into the deposit insurance fund. While this is a theoretical possibility, one has to keep in mind that the deposit insurance fund has to be built up over ten years. As a share of banks’ total assets, this represents an average contribution equal to 0.58%, roughly corresponding to yearly contributions equal to 0.06% of each bank’s total assets. We thus view this concern not as a first order issue for the design of the EDIS.3

Would a ‘mixed’ deposit insurance system better address concerns?

Some recent policy discussions have considered alternative approaches to the EDIS, notably a design under which national deposit guarantee schemes (as in the draft report by MEP Esther De Lange; see European Parliament 2016) or national compartments of EDIS (as in the proposal by Bénassy-Quéré et al. 2018, further investigated by Schnabel and Véron here on Vox) would intervene first, and the European deposit insurance fund would be a second line of defence. The rationale for these proposals would be that a mixed scheme would better align incentives and avoid risks of moral hazard.

Our empirical analysis shows that risks of cross-subsidisation under a fully-fledged EDIS are unfounded in the first place. Nevertheless, we analysed how a mixed deposit insurance system would fare. Under our assumptions, which are substantially in line with the draft report by MEP De Lange, such a system could consist of national deposit insurance schemes and a European deposit insurance fund, each with a target level of 0.4% of covered deposits. The European fund would intervene only after the depletion of at least one national fund. The key change with respect to a fully-fledged EDIS would be that the contributions paid by banks to reach the national 0.4% target level would be risk-based relative to their national benchmark, and not to their banking union peers. 

We find that cross-subsidisation would increase in such a mixed system relative to a fully-fledged EDIS, as shown in Table 1.4 The red cells indicate an increase in cross-subsidisation under the mixed system, and this would affect Belgium, Cyprus, Spain and Greece. 

Table 1 Cross-subsidisation: Fund exposure per euro contributed under a 'mixed' deposit insurance scheme and fully-fledged EDIS - 10% riskiest banks failing

Banks’ contributions based on DGS sliding scale method and DGS-baseline indicators, calibrated at the European level for the fully-fledged EDIS and for the European compartment in the "mixed" scheme, calibrated at the national level for the national compartments in the "mixed" scheme

Note: Under the fully-fledged EDIS, the EDIS exposure per euro contributed for each country is calculated as the sum of EDIS exposures to all banks within a country divided by the sum of contributions of all banks' of that country.  Under the mixed system, cross-subsidisation is measured as the ratio between the exposure of the European fund and the contributions paid to the European fund only.
Source: ECB staff calculations based on COREP and Bankscope data, 2015:Q4.

This is the direct consequence of a mixed deposit insurance scheme designed with fixed national target levels, and under which contributions to national compartments are benchmarked at the national level. Following this approach, banks are compared only to their national peers and banks’ contributions are bound by the risk-insensitive limit of a fixed share of covered deposits. Hence, having a mixed scheme for the EDIS with fixed target levels for national funds could in fact lead to more cross-subsidisation than a fully-fledged EDIS, rather than less.

It would of course be possible to have other designs for a mixed scheme. In our view, to fully take into account the relative riskiness of banks, the national target levels would in any case need to be flexible, i.e. allowed to vary with the relative riskiness of the banking systems with respect to the banking union benchmark.

However, the fact that national compartments have to be emptied first under a mixed scheme strongly diminishes the appeal of the EDIS in terms of ensuring a uniform level of confidence across the banking union, and maintains an element of bank-sovereign nexus. In that respect, we agree with Schoenmaker (2018) that preserving national compartments would go against the spirit of EDIS and that moral hazard is better addressed via other ways, in our case through well-designed risk-based contributions.


In conclusion, a fully-fledged EDIS would offer major benefits in terms of depositor protection while posing limited risks in terms of EDIS exposure. The EDIS would also play a key role in terms of confidence building, by avoiding risks of self-fulfilling prophecies on bank runs and ensuring that one euro saved in one country is worth the same as a euro saved in another. 

Most importantly, even a fully-fledged EDIS would not lead to some banking systems subsidising others, as well-designed risk-based contributions grounded on the polluter-pays principle and benchmarked at the Banking Union level would protect the EDIS from systematic cross-subsidisation. And this is even more the case than under a mixed system where a portion of the contributions is benchmarked at the national rather than at the banking union level. 

This finding strongly diminishes the case for having a mixed EDIS where national deposit guarantee schemes or national compartments are taking the first hit, as the key concern underpinning calls for such scheme seems unfounded. In addition, cross-subsidisation should be seen as a form of desirable risk-sharing in extremely adverse scenarios, as pooling resources ensures that national schemes are not overwhelmed in case of severe crises, thus reducing the bank-sovereign nexus. This is different from a systematic unwarranted cross-subsidisation.


Bénassy-Quéré, A, M Brunnermeier, H Enderlein, E Farhi, M Fratzscher, C Fuest, P-O Gourinchas, P Martin, J Pisani- Ferry, H Rey, I Schnabel, N Véron, B Weder di Mauro and J Zettelmeyer (2018), “Reconciling risk sharing with market discipline: A constructive approach to euro area reform”, CEPR Policy Insight No. 91.

Carmassi, J, S Dobkowitz, J Evrard, L Parisi, A Silva and M Wedow (2018), “Completing the Banking Union with a European Deposit Insurance Scheme: who is afraid of cross-subsidisation?”, ECB Occasional Paper Series No. 208.

European Parliament (2016), “Draft Report on the Proposal for a regulation of the European Parliament and of the Council amending Regulation (EU) 806/2014 in order to establish a European Deposit Insurance Scheme”, (COM(2015)0586 – C8-0371/2015 – 2015/0270(COD)), Committee on Economic and Monetary Affairs, rapporteur: Esther de Lange, November. 

Schnabel, I and N Véron (2018), “Breaking the stalemate on European deposit insurance”, 7 April.

Schoenmaker, D (2018), “Building a stable European Deposit Insurance Scheme”,, 17 April. (also published by Bruegel).


[1] For doing so, we are using an early warning model that accounts for different bank-specific, aggregate banking sector and macro-financial variables.

[2] As a term of comparison, the European Commission estimated average losses for 23 banks over the period of 2007-2010 to be 2.5% of total liabilities; the Financial Stability Board found that for G-SIBs losses as a fraction of total assets ranged from less than 1% to 4.7%, with most banks in a 2-4% range.

[3] However, this concern would be relevant in the case of a mixed scheme with national compartments, where ex-post contributions have to be called to replenish the national compartment.

[4] In the first case, cross-subsidisation is measured as the ratio between the exposure of the deposit insurance fund and the contributions paid to the deposit insurance fund in each country; in the second case, cross-subsidisation is measured as the ratio between the exposures of the European fund and the contributions paid to the European fund only.

2,310 Reads