Blogs review: The Capital requirements directive (CRD4)
What’s at stake: The last ECOFIN has been the occasion of a fierce negotiation over the adaptation into EU legislation of the Basel Committee rules known as Basel III. The Capital Requirements Directive proposal (CRD4) was proposed in July 2011 by the European Commission and is expected to enter into force in January 2013, with gradual implementation until completion in 2019. Yet no agreement was reached at the last ECOFIN where important divisions inside Europe have been exposed, and more work is still expected from the Basel Committee on a number of elements of the final proposal. This is at the heart of fundamental shifts within Europe’s banking sector and potential political conflicts primarily between the UK and continental Europe.
The importance of CRD4
Karel Lannoo of the Center for European Policy Studies points that the EU is the only jurisdiction codifying Basel 3 in EU law for application and implementation in the national law of 30 states, whose banking system represent one-half of the world’s banking assets. Other jurisdictions leave this responsibility to the discretion of national supervisory authorities. The Commission’s proposal is hugely complex, composed of a Regulation consisting of 488 articles and 4 annexes and a Directive consisting of 154 articles and 1 annex – in total good for about 220,154 words!
Benoît Cœuré, member of the executive board of the ECB, said in a recent speech that the ECB regards the single rulebook as a major step in the establishment of a true financial union. We consider the introduction of a single rulebook in the EU as an important step towards establishing a single market for financial services and enhancing financial integration in Europe. One of the benefits of this rulebook would be a commonly agreed definition of regulatory capital. The consistent application of such a definition would make it easier to compare eligible capital instruments from bank to bank within the EU, thereby strengthening the confidence of market participants in the loss-absorbing capacity of banks’ capital and, more generally, in the resilience of the EU’s financial sector.
What is CRD4 building on?
CRD4 is composed of two legislative instruments: a directive and a regulation. The directive needs to be transposed by member states into national law, while the regulation is applicable directly. Its key elements include:
1. Capital rules
a. More and better capital
b. Higher risk weights
c. A conservation buffer
d. A countercyclical buffer
2. Liquidity rules
a. Short term stress liquidity ratio applicable in 2015
b. Longer term Net Stable Funding Ratio applicable in 2018
a. Leverage ration backstop (with a view to binding measure by 2018)
The main issues
In his Sunday Wrap-Up email, Erik Nielsen argues that there are three key issues:
· First, while appropriate for the longer term, the very idea agreed some 18 months ago by the Basel Committee on Banking Supervision to raise bank capital requirements, reduce leverage etc is inherently pro-cyclical in the current environment, and comes on top of an aggressive fiscal tightening implemented in many countries while the US has not even started considering how they’ll implement Basel III.
· The second issue relates to the definition of capital when turning the Basel III agreement into legislation. Possibly under pressure from Berlin and Paris, the European Commission has defined capital in a way that seems to benefit German banks with respect to “silent capital” and French banks with respect to insurance subsidiaries.
· The third issue relates to whether individual member states should be allowed to impose stricter rules, e.g. higher capital requirements, on their banks without prior approval from Brussels. Basel III suggests that countries should be encouraged to do so, and the UK is adamant that it wants to exercise this option; i.e. put stricter requirements on UK banks than required by Basel. But the European Commission sees the EU as one entity because of the single market, and argues therefore that individual national rules would distort competition.
The maximum harmonization approach
Martin Wolf makes the case against maximum harmonization in EU banking. The UK wants to preserve its discretion in raising capital requirements, on the grounds that the consequences of failing to do so would fall on its own economy and taxpayers: responsibility should align with consequences. The proposal, in contrast, seeks to limit such freedom. The idea that member states should be prevented from making their banking systems “too safe” goes under the rubric of “maximum harmonisation”. It is a weird notion. Why, in sum, should anybody complain about a “race to the top” in banking? One argument is that if a country were to insist on higher capital, lending by its banks might shrink within other member countries. Wolf argues that, while negative spillovers from higher standards are conceivable, it cannot be legitimate for the European regulators, who are not fiscally accountable, to force member states to bear risks that its nationally accountable regulators view as excessive. This would not be a problem if banks could be resolved without adverse economic, financial and fiscal consequences. But that is not the case.
Benoît Cœuré points out what the ECB proposed in its Opinion on this issue. The ECB proposes the possibility for national authorities, within the framework of a single EU rulebook, to adopt – with specific safeguards – stricter requirements in their respective Member States for macro-prudential reasons. This is necessary because Member States need to address country-specific financial stability concerns stemming from different structural features of their domestic financial systems. It is key that the possibility for Member States to apply these stricter requirements should be framed within the single EU rulebook framework. To this end, specific safeguards should be put in place. First, definitions should remain intact; only quantitative ratios and limits can be tightened. Second, the European Systemic Risk Board could play a coordinating role in assessing financial stability concerns and possible unintended consequences and spillovers to other Member States. Furthermore, the European Banking Agency (EBA) and the ESRB should publish regular updates on their websites of measures adopted by Member States and the underlying reasons for stricter requirements. Third, where financial stability concerns that triggered the application of more stringent measures cease to exist, the quantitative ratios and limits should return to a harmonized level set by the regulation.
Erik Nielsen writes in a recent research paper that the Commission is absolutely right on this one. The EU desperately needs one single bank supervisor and one single “underwriter” of the banking system, i.e. a common deposit guarantee and a common resolution mechanism. It is now glaringly clear that one cannot have a functioning monetary union without a “banking union”. His personal view is that if any of the ten EU members who are not eurozone members try to block such further integration, then the eurozone will need to create arrangements for themselves and leave non-EMU-EU members behind on a slower track.
The limits of CRD4: the reliance on banks’ risk management models
Jacopo Carmassi and Stefano Micossi argue that CDR4 fails to delink the calculation of capital requirements from banks’ own risk management models. In the Basel III Accord capital requirements are still calculated with reference to risk-weighted assets and even broader national discretion. In their evaluation of banks’ internal models, the UK’s Financial Services Authority shows that internal models produce widely divergent risk assessments of identical portfolios, making them unusable for any objective evaluation of risk exposures. The models used by banks in the calculation of risk-weighted assets suffer from fundamental technical flaws that have been exposed by a blooming literature but are ignored by Basel regulators. Among other things:
· The models are estimated from non-stationary time series and thus have weak predictive value for large changes in the state variables;
· They are almost by assumption unable to account for systemic risks, when expectations converge and correlations between portfolio performances rise dramatically; and,
· They actually create incentives for banks to concentrate their exposures in the tail of risk distributions in order to economize capital, thus raising the probability and the impact of rare catastrophic events actually happening.
Of demand and supply of safe collateral
Manmohan Singh points to the problem that recent regulatory efforts will require significant collateral on many fronts – Basel’s liquidity ratios, EU Solvency II and CRD IV, and moving OTC derivatives to CCPs. Unless there is a rebound in the pledgeable collateral market, the likely asymmetry in the demand and supply in this market may entail some difficult choices for the markets and the regulators. Markets may create cottage industries for upgrade collateral to meet regulatory needs, which may not all be transparent. Also regulators may, due to collateral constraints, see some facets of financial industry shrink. Alternately, it is also possible that regulatory efforts may have to prioritize proposals to bridge the collateral shortfall.
Colm McCarthy notes that, according to calculations from Karl Whelan, the 0.5% deficit/GDP rule envisaged in the ‘fiscal compact’ eventually yields a debt ratio at only 17% of GDP. This happens whether you start from zero or from Greece. The liquidity part does not (yet) specify a quantitative ratio of liquid to total assets but could turn out to require 20% or 25%. If balance sheets end up at something sensible like 150% of GDP (UK currently 400%), this implies say 35% of GDP needs to be available as high-quality liquidity. Aside from central bank money, this means short (< 5 yr) sovereign bonds. It can’t all be central bank money (or if it can, explain how monetary policy would work). If the debt ratio drops much below 50%, sovereign debt duration has to drop dangerously. There are reasons for not having too many sovereign bonds about. There are also reasons for not having too few.
Republishing and referencing
Bruegel considers itself a public good and takes no institutional standpoint. Anyone is free to republish and/or quote this post without prior consent. Please provide a full reference, clearly stating Bruegel and the relevant author as the source, and include a prominent hyperlink to the original post.