VoxEU Column EU policies Global governance International Finance

A short guide to the EBA’s recapitalisation results

In a bid to restore stability and confidence in the markets, the European Banking Authority (EBA) has recommended a plan to raise the required capital buffers of major European banks by summer 2012. This column, by economists at the EBA, describes how the capital targets have been calculated and outlines the main drivers of bank shortfalls with respect to these targets.

On 26 October 2011, the European Council agreed a comprehensive package aimed at addressing the deterioration of macroeconomic and market conditions. The package, which aims at restoring stability and confidence in the markets, includes a recapitalisation plan for major EU banks as proposed by the European Banking Authority (EBA). These measures are also in line with the European Systemic Risk Board (2011) Press Statement on the need for coordinated efforts to strengthen banks’ capital as well as for a transparent and consistent valuation of sovereign exposures.1

On 8 December, the EBA (2011a) adopted a recommendation that national supervisory authorities (NSAs) should require banks to strengthen their capital positions by building up an “exceptional and temporary capital buffer” such that the Core Tier 1 capital ratio (i.e. the ratio of Core Tier 1 capital over risk-weighted assets) reaches a level of 9% by the end of June 2012, after a prudent valuation of sovereign debt exposures to reflect market prices as at the end of September. As clarified by the EBA, the buffer is explicitly not designed to cover losses in sovereign exposures, but to provide a reassurance to markets about the banks’ ability to withstand a range of shocks and still maintain adequate capital.

Based on the public information disclosed by the EBA, this column describes how the capital needs have been computed and illustrates the main drivers of banks’ shortfalls with respect to the target capital level.

How the capital needs are quantified

According to the methodological note published by the EBA (2011b), a capital need may materialise because:

  • A bank needs to reach a Core Tier 1 ratio of at least 9%, and it is currently below that threshold.
  • A bank has to value in a conservative fashion the sovereign exposures towards European Economic Area (EEA) countries in their portfolios.

The (potential) valuation losses are to be covered by building up the so-called sovereign buffer. This buffer is, in turn, computed in different ways depending on the portfolio where the sovereign exposures are booked:

  • Prudential filters on EEA sovereign exposures held in the available-for-sale (AFS) portfolio are removed, i.e., banks are required to build a buffer of Core Tier 1 capital against prudential filters.2
  • EEA debt sovereign exposures in the held-to-maturity (HTM) and loans and receivable (L&R) portfolios are valued in a conservative fashion.

Banks have been required to build a buffer equal to the difference between the book value of these assets and their revalued amount3.

In addition, for the recapitalisation exercise, banks have been requested to adhere to the CRD3 (the European Directive implementing Basel 2.5 in the EU) for the calculation of the Core Tier 1 ratio. This implies that the changes in the trading book and securitisation treatment are fully incorporated in the requirement.

Recapitalisation exercise, not a stress test

The EBA has made it clear that the recapitalisation exercise is not a proper stress test, since no macroeconomic scenario is included in the simulation and the sensitivity analysis is limited to sovereign risk. On the other hand, the choice of the 9% Core Tier 1 ratio target – which is far higher than the 5% announced by the US Agencies for the 2012 stress test and used by the EBA itself during the 2011 EU-wide stress test – along with the inclusion of the CRD3 rules ensure that other risk drivers are (implicitly) taken into account in the quantification of banks’ capital needs.

Overall results

The overall results have been disclosed and widely analysed. The capital shortfall for the 71 banks included in the sample is about €115 billion. This amount includes €30 billion backstop measures provided under the EU/IMF programme for Greece.

Figures 1 and 2 show the amount of the capital shortfall by country and by bank, excluding the six Greek banks.

Figure 1. Shortfall (breakdown by country), € billion

Figure 2. Shortfall (including sovereign buffer) – breakdown by bank

Not surprisingly, Spanish, Italian, Portuguese, and Belgian banks are heavily affected by the recapitalisation package, given the size of their exposures to sovereigns under stress. However, German and French banks also show a material capital shortfall. A more in-depth analysis of the data helps understand this puzzle.

On the drivers of the shortfall

In the aftermath of the publication of the final recapitalisation figures, the attention has been mostly devoted to the scrutiny of the aggregate results, while little (if any) attention has been paid on the decomposition of the impact of the different determinants of the capital shortfall.

Here we untangle the three main drivers of the shortfall:

  • The 9% capital target.
  • The sovereign buffer.
  • The application of the European Directive implementing Basel 2.5 in the EU (so-called CRD3 rules).

Figure 3 presents this decomposition. A first key message is that, on average, the three drivers provide the same contribution to the strengthened buffer requirements requested by the EBA. This is consistent with the idea that, even without being a fully-fledged stress test, the recapitalisation exercise aims at capturing risks beyond sovereign exposures.

Figure 3. Contribution to the strengthened capital requirements

Figure 4 goes further and shows the contribution of the different accounting portfolios to the quantification of the sovereign buffer. The results are interesting, with the removal of the prudential filters covering the largest share of the sovereign buffer.

Figure 4. Contribution to the sovereign buffer

This is consistent with the allocation of sovereign exposures across accounting portfolios (Figure 5). On average, sovereign debt holdings of EU banks in the EBA sample are booked for 43% in the HTM portfolio, 41% in the AFS, 12% in the held for-trading (HFT) portfolio, and the remaining 4% in the “fair value option” portfolio.

Figure 5. Allocation of sovereign exposures by accounting portfolio

Overemphasis on sovereign component

Taken at face value, these results suggest that the excessive emphasis on the sovereign component of the overall shortfall is not well grounded. Clearly, aggregate results may hide differences across countries and a more granular investigation is needed. Figure 6 reports the country-by-country contribution of the three drivers.

Figure 6. Contribution to the strengthened capital requirements – by country

Results are more diversified, with banks from distressed sovereigns more affected by the sovereign buffer – about 70% for Belgium (70% of which is due to the removal of the prudential filters) and Cyprus, and 50% for Portugal and Italy (for the latter, 70% of which is due to the removal of the prudential filters). Spanish banks are most affected by the 9% threshold, while German, French, and Dutch banks by the CRD3 rules.


The recapitalisation package proposed by the EBA and agreed at the Summit of the Heads of State and Government of the EU requires a significant effort from banks, in light of the exceptional conditions in the financial markets in Europe. As highlighted in a joint statement by the Presidency of the ECOFIN Council and the EBA (2011), “the EU-wide recapitalisation exercise is an important element in strengthening European banks’ position in the current environment characterised by heightened systemic risk arising from the sovereign debt crisis. The increased resilience of the banking sector through higher capital levels should support banks in maintaining the ability of lending to the real economy in the EU”.

We show that the burden is spread across countries and is driven by different factors. On average, the higher target Core Tier 1 ratio, the conservative valuation of sovereign exposures, and the CRD3 rules contribute one third each to the increased capital buffer requirements. This should provide reassurance that banks from countries perceived as weaker are not unduly penalised (even though they do bear a substantial share of the burden) and that other risks – particularly credit and market risks – are adequately captured by the buffer requirement.

Authors’ note: The views expressed are those of the authors and do not necessarily reflect those of the European Banking Authority and its Members


IMF (2011), Global Financial Stability Report, September.
EBA (2011a), “Recommendation on the creation and supervisory oversight of temporary capital buffers to restore market confidence”, EBA/REC/2011/1.
EBA (2011b), “Capital buffers for addressing market concerns over sovereign exposures”, Methodological Note.
Ecofin Presidency and EBA (2011), “Joint statement by the Presidency of the ECOFIN Council and the EBA”.
ESRB (2011), Press Release, 21 September.

1 In the September Global Financial Stability Review, the IMF had identified potential losses for EU banks of about €200 billion as the consequence of marking to market sovereign exposures towards Greece, Ireland, Portugal, Spain, Italy, and Belgium (IMF 2011).
2 Prudential filters are the instrument introduced in some jurisdictions in order to avoid that market price volatility directly affects banks’ capital positions for assets that, according to accounting rules, are to be marked-to-market. Accounting valuation gains and losses are therefore “filtered” out for prudential purposes. When the prudential filters are not applied, valuation gains and losses are already reflected inthe capital levels. Under Basel 3, prudential filters will be dismantled.
3 Banks have been allowed to offset valuation gains and losses on bonds, not for loans. Since, net valuation gains have been capped to zero (i.e., the buffer cannot be negative), the impact of not applying any cap on the bonds is, on average, negligible.

1,680 Reads