Europaudvalget 2016
KOM (2016) 0853
Offentligt
1715591_0001.png
EUROPEAN
COMMISSION
Brussels, 24.11.2016
SWD(2016) 377 final/2
CORRIGENDUM
This document replaces SWD(2016) 377 final of 23.11.2016.
Insertion of cross-references
COMMISSION STAFF WORKING DOCUMENT
IMPACT ASSESSMENT
Accompanying the document
Proposal amending: - Regulation (EU) No 575/2013 on prudential requirements for
credit institutions and investment firms; - Directive 2013/36/EU on access to the activity
of credit institutions and the prudential supervision of credit institutions and investment
firms; - Directive 2014/59/EU establishing a framework for the recovery and resolution
of credit institutions and investment firms; - Regulation (EU) No 806/2014 of the
European Parliament and of the Council of 15 July 2014 establishing uniform rules and
a uniform procedure for the resolution of credit institutions and certain investment
firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund
{COM(2016) 850 final}
{COM(2016) 851 final}
{COM(2016) 852 final}
{COM(2016) 853 final}
{COM(2016) 854 final}
{SWD(2016) 378 final}
EN
EN
 kom (2016) 0853 - Ingen titel
Table of contents
1.
2.
INTRODUCTION ....................................................................................................... 4
PROBLEM DEFINITION .......................................................................................... 8
2.1.
2.2.
2.3.
2.4.
2.5.
2.6.
2.7.
2.8.
2.9.
2.10.
3.
3.1.
3.2.
3.3.
3.4.
4.
4.1.
4.2.
4.3.
4.4.
4.5.
4.6.
4.7.
4.8.
4.9.
4.10.
5.
Excessive reliance on short-term funding ....................................................... 10
Excessive leverage........................................................................................... 11
Inadequate calibration of risk weights for exposures to SMEs ....................... 13
Weaknesses to the regulatory framework for loss absorption and
recapitalisation capacity .................................................................................. 15
Inappropriate level of capital requirements against trading activities ............. 17
Problems on remuneration rules ...................................................................... 18
Problems on insolvency ranking of unsecured bank debt instruments............ 20
Lack of effectiveness of the current rules on moratorium ............................... 21
Insufficient proportionality of the current rules .............................................. 24
Consequences from the baseline scenario ....................................................... 25
General, specific and operational objectives ................................................... 30
Consistency of the objectives with other EU policies ..................................... 33
Consistency of the objectives with fundamental rights ................................... 33
Subsidiarity ...................................................................................................... 34
On excessive reliance on short-term funding .................................................. 35
On excessive leverage ..................................................................................... 40
On inadequate calibration of risk weights on SME exposures ........................ 42
On weaknesses to the regulatory framework for loss absorption and
recapitalisation capacity .................................................................................. 45
On inappropriate level of capital requirements against trading
activities ........................................................................................................... 51
On problems on remuneration rules ................................................................ 60
On problems on insolvency ranking ................................................................ 64
On lack of effectiveness of the current rules on moratorium .......................... 67
On insufficient proportionality of the current rules ......................................... 69
The choice of the instrument ........................................................................... 71
OBJECTIVES ........................................................................................................... 30
POLICY OPTIONS AND ANALYSIS OF IMPACTS ............................................ 35
THE CUMULATIVE IMPACTS OF THE ENTIRE PACKAGE ........................... 72
5.1. Introduction ..................................................................................................... 72
5.2. Quantitative assessment of benefits and costs related to FRTB and the
LR .................................................................................................................... 73
2
 kom (2016) 0853 - Ingen titel
5.3. Impact of the preferred options on administrative costs ................................. 76
5.4. The impact on SMEs ....................................................................................... 77
5.5. Impact on third countries ................................................................................. 78
6.
MONITORING AND EVALUATION..................................................................... 78
GLOSSARY ...................................................................................................................... 81
ANNEX 1. PROCEDURAL ISSUES AND CONSULTATION OF
INTERESTED PARTIES.......................................................................................... 83
Possible impact of the CRR/CRD IV on financing of the economy ("CRR
consultation")................................................................................................... 83
Call for Evidence ....................................................................................................... 86
Targeted consultations ............................................................................................... 89
ANNEX 2. PARTIAL EVALUATION OF THE EXISTING POLICY
FRAMEWORK ......................................................................................................... 95
Annex 2.1. Evaluation of rules on remuneration....................................................... 96
Annex 2.2. Impact on the bank financing of the economy, including SMEs ............ 98
ANNEX 3. ASSESSMENT OF OTHER PROPOSED AMENDMENTS TO
CRR/CRD IV/BRRD .............................................................................................. 104
Annex 3.1. Calculation of derivative exposures in the counterparty credit
risk framework............................................................................................... 105
Annex 3.2. Disclosure ............................................................................................. 110
Annex 3.3. Supervisory reporting ........................................................................... 112
Annex 3.4. Pillar 2 additional capital ...................................................................... 114
Annex 3.5. Equity investments into funds .............................................................. 116
Annex 3.6. Bank financing of infrastructure projects ............................................. 118
Annex 3.7. Large exposure framework (alignment with Basel rules) ..................... 121
Annex 3.8. Exemptions on large exposures ............................................................ 124
Annex 3.9. Rules on exposures to CCPs ................................................................. 126
Annex 3.10. Contractual recognition of bail-in (article 55 BRRD) ........................ 129
Annex 3.11. Changes to MREL .............................................................................. 132
Annex 3.12. Application of IFRS 9 by the EU banks ............................................. 134
Annex 3.13. Comparative analysis of characteristics of EU G-SIIs and O-
SIIs ................................................................................................................. 137
Annex 3.14. Analysis of a leverage ratio requirement for different business
models and exposure types ............................................................................ 139
ANNEX 4. ESTIMATED IMPACT OF POLICY OPTIONS ........................................ 142
ANNEX 5. BACKGROUND TO CUMULATIVE IMPACT ASSESSMENT ............. 151
Annex 5.1. Estimation of costs of FRTB and LR using the QUEST model ........... 151
Annex 5.2. Estimation of benefits of FRTB and LR using the SYMBOL
model ............................................................................................................. 154
ANNEX 6. IMPLEMENTATION OF PROPOSED MEASURES ................................ 168
3
 kom (2016) 0853 - Ingen titel
1715591_0004.png
1.
INTRODUCTION
Financial crises, particularly when they involve the banking sector, can result in huge costs, both
in terms of direct fiscal costs and associated costs for the real economy. The 2007-2008 financial
crisis was a case in point. Between the years 2008 and 2014 EU governments used almost €2
trillion in State aid (an amount equal to almost 14% of the 2014 EU GDP) to rescue the financial
sector
1
. The losses to economic activity due to the crisis were also significant. Some estimates
2
show that the present value of cumulative output losses across the EU may amount to 50-100 %
of annual pre-crisis EU GDP (about €6-12.5 trillion), if not more. For the euro area alone, output
is now 20% below the level it would have achieved had the trend growth in the previous 15 years
continued after 2007. Furthermore, according to some estimates, the present value of the total
loss of output until 2030 would represent more than three times the whole economic output of the
euro area in 20083.
In response to the crisis the EU implemented a substantial reform of the financial services
regulatory framework in order to enhance the resilience of EU institutions (the term institution is
used to refer to both credit institutions (i.e. banks) and investment firms, as both are subject to the
requirements of the CRR and the CRD IV) and thus increase EU financial stability. Two
legislative initiatives targeted institutions, in particular:
Regulation (EU) No 575/2013, also known as the Capital Requirements Regulation (CRR)4
and Directive 2013/36/EU, also known as the fourth revision of the Capital Requirements
Directive (CRD IV)
5
enhanced prudential requirements for institutions by implementing
global standards adopted by the Basel Committee on Banking Supervision (BCBS)
6
in
December 2010;
Directive 2014/59/EU, also known as the Bank Recovery and Resolution Directive (BRRD)7
and Regulation (EU) No 806/2014
8
on the Single Resolution Mechanism introduced a new
recovery and resolution framework for dealing with institutions that are failing or likely to
fail, including a minimum requirement for own funds and eligible liabilities (MREL). The
main objectives of the Directive are to maintain financial stability and minimise losses for
society in general and tax payers in particular in case an institution fails.
See http://ec.europa.eu/competition/state_aid/scoreboard/financial_economic_crisis_aid_en.html.
Economic Review of the Financial Regulation Agenda, Commission Staff Working Document, 2014, p.
42.
Lecture by Vítor Constâncio, Vice-President of the ECB, at the Conference on “European Banking
Industry: what’s next?”, organised by the University of Navarra, Madrid, 7 July 2016
Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on
prudential requirements for credit institutions and investment firms and amending Regulation (EU) No
648/2012 (OJ L 321, 26.6.2013, p. 6)
Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the
activity of credit institutions and the prudential supervision of credit institutions and investment firms,
amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC (OJ L 176,
27.6.2013, p. 338).
Those standards are known as the Basel III framework or Basel III.
Directive 2014/59/EU of the European Parliament and of the Council of 15 May 2014 establishing a
framework for the recovery and resolution of credit institutions and investment firms and amending
Council Directive 82/891/EEC, and Directives 2001/24/EC, 2002/47/EC, 2004/25/EC, 2005/56/EC,
2007/36/EC, 2011/35/EU, 2012/30/EU and 2013/36/EU, and Regulations (EU) No 1093/2010 and (EU)
No 648/2012, of the European Parliament and of the Council (OJ L 173, 12.6.2014, p. 190)
Regulation (EU) No 806/2014 of the European Parliament and of the Council of 15 July 2014
establishing uniform rules and a uniform procedure for the resolution of credit institutions and certain
investment firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund
and amending Regulation (EU) No 1093/2010
4
1
2
3
4
5
6
7
8
 kom (2016) 0853 - Ingen titel
1715591_0005.png
The new EU regulatory framework has contributed to reinforcing financial stability, restoring
investor confidence and allowing institutions to play their fundamental role in supporting
economic recovery. The adoption of the Basel III framework at international level, and of the
legislative initiatives mentioned above at EU level, did not mark the end of the post-crisis reform.
Work continued on several elements which were left outstanding at the time. For example, while
Basel III introduced a requirement to calculate and disclose a leverage ratio, it did not introduce a
capital requirement based on that leverage ratio; that was to be introduced in 2018. Similarly,
although the BCBS had agreed on the necessity of introducing liquidity requirements, the Basel
III framework actually did not provide detailed rules for those requirements; those were
published later. Moreover, the BCBS has carried out a fundamental review of the trading book
framework to address the flaws of the existing rules unveiled by the financial crisis.
The BCBS was not the only international body involved in the post-crisis reform. Following a
call from G20 Leaders, the Financial Stability Board (FSB) in November 2015 issued standards
9
aimed at ensuring that global systemically important banks (G-SIBs) have sufficient loss-
absorbing capacity to be recapitalised in case they fail. This work has led to the introduction of
standards on total loss-absorbing capacity (TLAC).
At EU level, the Commission carried out various initiatives in order to assess whether the
existing prudential framework and the upcoming reviews of global standards were the most
adequate instruments to ensure that EU institutions would continue to provide the necessary
funding to the EU economy.
In particular, the Commission launched in July 2015 a public consultation on the possible impact
of the CRR and the CRD IV on bank financing of the EU economy with a particular focus on the
financing of micro, small and medium-sized enterprises (SMEs) and of infrastructure and in
September 2015 a Call for Evidence (CfE)
10
covering EU financial legislation as a whole. The
two initiatives sought empirical evidence and concrete feedback on i) rules affecting the ability of
the economy to finance itself and growth, ii) unnecessary regulatory burdens, iii) interactions,
inconsistencies and gaps in the rules, and iv) rules giving rise to unintended consequences. In
addition, the Commission carried out specific analysis on rules relating to remuneration
11
and on
the proportionality of the rules contained in the CRR and the CRD IV.
12
Finally, the Commission
contracted a study to assess the impact of CRR on the bank financing of the economy
13
.
All the initiatives mentioned above have provided clear evidence of the need to update and
complete the current rules in order i) to reduce further the risks in the banking sector and thereby
reduce the reliance on State aid and taxpayers' money in case of a crisis, and ii) to enhance the
9
10
11
12
13
November 2015 by the Financial Stability Board: http://www.fsb.org/wp-content/uploads/20151106-
TLAC-Press-Release.pdf
See
http://ec.europa.eu/finance/consultations/2015/long-term-finance/docs/consultation-
document_en.pdf
and
http://ec.europa.eu/finance/consultations/2015/financial-regulatory-framework-
review/docs/consultation-document_en.pdf.
Commission Report COM(2016)510 Report from the Commission to the European Parliament and the
Council of 28 July 2016 – Assessment of the remuneration rules under Directive 2013/36/EU and
Regulation (EU) No 575/2013.
The Call for Evidence was intended to cover the entire spectrum of the financial services regulation.
The impact assessment address issues limited to the areas of banking only. Other issues involving other
segments of the EU financial legislation will be dealt with separately.
Insert the link to the study
5
 kom (2016) 0853 - Ingen titel
1715591_0006.png
ability of institutions to channel adequate funding to the economy. More specifically, the
evidence that was collected demonstrates that the existing EU rules:
14
are not able to cover all risks that institutions face;
are not always sufficiently risk-sensitive and able to take into account adequately all relevant
risk drivers;
are too complex or too burdensome and create excessive compliance costs for smaller
institutions;
are not always formulated in a sufficiently clear way and can give rise, in places, to different
interpretations and applications; and
do not always support economic growth.
In order to enhance the resilience of EU institutions and thereby increase financial stability, this
impact assessment considers various options for incorporating the remaining elements of the
regulatory framework recently agreed by the BCBS and for enhancing legal certainty, especially
in the area of resolution. The options considered in this impact assessment aim at:
better addressing the long-term funding risk;
reducing excessive leverage;
increasing the loss absorption and recapitalisation capacity of global systemically important
institutions (G-SIIs);
better addressing market risks by increasing the risk sensitivity of the existing rules; and
increasing legal certainty and enhancing convergence among Member States (MS) in the area
of insolvency law and restructuring proceedings, particularly in the area of creditor hierarchy
and the use of the moratorium tool.
Many of the measures considered in this impact assessment are included in the roadmap
developed by the Commission in response to a request from the Council to complete the Banking
Union. These measures are seen as flanking measures in the context of the establishment of the
European Deposit Insurance Scheme (EDIS).
15
When contemplating the introduction of the above measures, a number of options contained in
this impact assessment explore the possibility of adjusting the calibration of some of the
new/revised Basel standards (e.g. the leverage ratio, the market risk rules) to reflect better the
specificities of EU institutions and the EU economy. The aim of those adjustments is to avoid
situations in which the strengthening of prudential requirements could lead to insufficient lending
to the economy.
Furthermore, some of the other options related to the above measures explore potential
adjustment aimed at mitigating potential disincentives for certain activities carried out by
institutions which are important for the efficient functioning of capital markets. This is necessary
because, in addition to their fundamental role of providing finance to the economy, institutions
are also important actors on capital markets, as issuers of or investors in securities and other
14
15
Respondents to the CfE also argued that rules agreed by the BCBS but not yet included in the CRR and
the CRD IV would have a disproportionate impact on certain activities and business models.
Communication from the Commission to the European Parliament, the Council, the European Central
Bank, the European Economic and Social Committee and the Committee of the Regions "Towards the
completion of the Banking Union", COM/2015/0587 final. See also the Council conclusions on a
roadmap to complete the Banking Union of 17/6/2016.
6
 kom (2016) 0853 - Ingen titel
1715591_0007.png
financial instruments (e.g. covered bonds, securitisations). They also play an important role in
facilitating the efficient functioning of those markets by providing essential services, such as
underwriting or market making. The Commission has already tabled a proposal
16
to increase the
efficiency and soundness of the EU securitisation market including measures to enhance the role
of institutions in this market, both as investors and as issuers. The abovementioned options
consider additional ways in which to foster the creation of a Capital Markets Union (CMU).
Specifically, they are looking at how the measures would need to be adjusted in order to:
avoid disproportionate capital requirements for trading book positions, including those
related to market making activities;
reduce the costs of issuing/holding certain instruments (covered bonds, high quality
securitisation instruments, sovereign debt instruments, derivatives for hedging purposes);
avoid an increase in the costs of providing services to clients for trades cleared by central
counterparties (CCPs).
Lastly, some of the options related to the abovementioned measures contemplate adjustments
aimed at preventing any potential unfavourable treatment for business areas which are
particularly important for cross-border trade. Specifically, they explore the possibility of
introducing a more risk-sensitive treatment for trade finance instruments within the contemplated
rules on the net stable funding ratio (NSFR) and on the leverage ratio.
In addition to the abovementioned measures aimed at enhancing the resilience of EU institutions,
this impact assessment also contemplates measures aimed at:
enhancing the risk-sensitivity of capital requirements for exposures to SMEs;
reducing the administrative costs linked to some rules in the area of remuneration (namely
those on deferral and pay-out in instruments); and
making the rules contained in the CRR and CRD IV more proportionate and hence less
burdensome for smaller and less complex institutions.
Finally, in addition to all the measures described above, which would constitute the main
building blocks of a potential proposal by the Commission and which are expected to have the
largest impact on EU institutions in case they would be introduced, this impact assessment also
considers the possibility of introducing several other measures. These measures are included in
the impact assessment for reasons of completeness, as their introduction is seen as largely
uncontroversial and straightforward and would generally have a limited impact. These measures
would:
implement a number of changes, most of them agreed at international level, to better specify
the technical aspects of certain existing rules (calculation of the exposure value of derivatives
in the counterparty credit risk framework, disclosure requirements, capital requirements for
equity investments in funds, large exposures limits, rules on exposures to CCPs, changes to
MREL, application of International Financial Reporting Standard 9 (IFRS 9));
clarify existing rules on the basis of the outcomes of the CfE or other consultations with
stakeholders (Pillar 2 requirements, exemptions from large exposures limits, supervisory
reporting, contractual recognition of bail-in); or
enhance the overall consistency of the treatment of investments in infrastructure projects
between the CRR and CRD IV on one side and Solvency II on the other. This would increase
16
See
http://ec.europa.eu/finance/securities/securitisation/index_en.htm
for more details.
7
 kom (2016) 0853 - Ingen titel
1715591_0008.png
the contribution of the banking sector to the goal of mobilising additional private finance in
the context of the Commission's
Investment Plan for Europe
17
.
For all the measures listed above the review at global or EU level has been already completed. In
addition, there is widespread acceptance among stakeholders about the need to introduce those
measures. In view of this, it is imperative to introduce these measures now in the interest of legal
certainty and for the creation of a robust financial sector. This would allow the EU to meet the
deadline agreed at global level for certain standards (e.g. on TLAC
18
, leverage ratio
19
, NSFR
20
)
and fulfil the timeline requested by the Council for the risk-reduction measures included in the
roadmap for the completion of the Banking Union
21
. Continuing work on risk-reduction in the
Banking Union remains a top priority for the Commission
22
.
In order to give institutions sufficient time to adapt to the new regulatory framework it is of the
outmost importance to provide them with the necessary legal certainty regarding the exact shape
of the new rules as quickly as possible. Institutions should also benefit without delay from the
alleviations of the compliance burden envisaged by some of the measures, especially given the
current economic context in the EU.
The list of preferred options responding to the problems analysed in the impact assessment as
well as the indicative list of legislative amendments is presented in annex 6.
This impact assessment does not include measures stemming from a strategic review of Basel III
on the methods used to calculate the risk-based capital requirements for credit risk and
operational risk as those changes are still under discussion at the BCBS level. The review is
expected to be completed by the end of 2016. The Commission will consider whether and how to
implement those measures once they are adopted at international level.
2.
PROBLEM DEFINITION
Given the limited time elapsed since its entry into force, a fully-fledged evaluation of the CRR,
the CRD IV and the BRRD could not be carried out yet. Nevertheless, the need of amending
these instruments in order either to introduce new provisions or to review the existing ones has
emerged as a result of the work carried out by the BCBS, obtaining evidence on the national
implementation of the Directives or as an outcome of specific consultations and studies, solicited
by the Commission (for more details see annexes 2 and 6).
The following issues have been identified in relation to the existing rules contained in the CRR
and the CRD IV:
they do not cover all risks that institutions face (e.g. the CRR currently does not
foresee specific capital requirements to limit the leverage of institutions or
17
18
19
20
21
22
Communication "An Investment Plan for Europe", COM(2014) 903 final
1 January 2019, agreed in November 2015 by the Financial Stability Board:
http://www.fsb.org/wp-
content/uploads/20151106-TLAC-Press-Release.pdf
1 January 2018, agreed in January 2016 by members of the Basel Committee’s oversight body, the
Group of Governors and Heads of Supervision (GHOS):
http://www.bis.org/press/p160111.htm
1 January 2018, agreed in October 2016 by members of the Basel Committee:
http://www.bis.org/bcbs/publ/d295.pdf
Council conclusions of 17 June 2016 on a roadmap to complete the Banking Union:
http://www.consilium.europa.eu/press-releases-pdf/2016/6/47244642837_en.pdf
State of the Union 2016,
page 28, priority 5.
8
 kom (2016) 0853 - Ingen titel
specific funding requirements to limit the maturity mismatches between assets
and liabilities);
they are not always sufficiently risk sensitive (e.g. one of the simpler approaches
used by institutions to calculate the size of their derivatives exposures does not
fully take into account the risk reduction benefits of netting agreements);
some of them are too complex or too burdensome for institutions (e.g. reporting
and calculation methods), create in some cases excessive compliance costs
(remuneration rules on deferral and pay-out in instruments), or may
disproportionately affect certain activities or business models (including for new
measures introduced to cover existing risks such as the leverage ratio);
some of them are not formulated in a sufficiently clear way and give rise to
different interpretations (e.g. there are different interpretations on the way in
which the capital requirements in the CRR and the institution-specific capital and
buffer requirements in the CRD IV interact).
As regards the BRRD existing rules, the following issues have been identified:
certain MREL eligibility criteria are loosely defined, leaving room for interpretation (e.g.
inclusion of large corporate deposits and certain types of structured notes);
the adoption of the FSB's TLAC standard for global systemically important banks (G-
SIBs) would create a misalignment with the existing MREL calibration conditions and a
parallel standard for G-SIBs in the EU;
regarding the insolvency ranking of unsecured debt, the requirement for subordination of
certain unsecured senior claims is missing from the current text;
Lack of clarity on supervisory reporting and public disclosure of items that meet MREL
eligibility criteria;
implementation issues with Article 55 BRRD on recognition of bail-in in third countries
which was too prescriptive and led to the withdrawal of EU banks from business
contracts with certain third countries;
on moratorium powers, BRRD already contains provisions allowing the suspension of
payment obligations but these have been implemented in very different ways at national
level and may not provide a sufficiently consistent application with respect to important
elements such as the scope, phase of application, trigger conditions and duration of the
suspension.
Sections 2.1 to 2.9 present the most important problems addressed by this impact assessment and
concern the following areas:
stable funding of institutions;
capital requirements for risk of excessive leverage;
capital requirements for exposures to SMEs;
remuneration;
insolvency ranking and moratorium in relation to the BRRD;
proportionality.
For areas for which the solution to the identified problem is seen as largely straightforward and
uncontroversial and as having limited impact, are presented in annex 3. They concern:
capital requirements for derivative exposures;
9
 kom (2016) 0853 - Ingen titel
1715591_0010.png
disclosure and supervisory reporting to primarily address proportionality issues;
institution-specific (Pillar 2) capital requirements;
equity investments into funds;
capital requirements for specialised lending exposures (infrastructure);
large exposure limits;
capital requirements for exposures CCPs;
contractual recognition of bail-in and changes to MREL;
application of IFRS 9 by EU institutions.
2.1.
Excessive reliance on short-term funding
When an institution takes decision regarding its balance sheet structure, it does not take into
account all the impacts of its choice on the rest of the economy. In addition, private incentives to
limit excessive reliance on unstable funding of core (often illiquid) assets are weak. Institutions
may have private incentives to expand their balance sheets, often very quickly, relying on
relatively cheap and abundant short-term wholesale funding. Rapid balance sheet growth
increases the likelihood that individual institutions will face funding problems in case of liquidity
shocks, and weakens their ability to respond to these shocks when they occur. As shown by the
examples below, this fragility can have systemic implications when institutions fail to internalise
the costs associated with large funding gaps. This can have negative consequences on financial
stability in case of economic shocks.
During the financial crisis, institutions made use of excessive amounts of short-term wholesale
funding to finance their long term activities. When short-term funding became unavailable,
institutions were either forced to request emergency liquidity assistance from central banks or
engage in 'fire sales' of assets, triggering a downward spiral in prices and eroding their liquidity
positions, with the ultimate consequence of driving a number of them into insolvency. Some
credit institutions also had to be bailed-out by their governments. For example Hypo Real Estate
Holding AG (HRE) had - through a subsidiary (Depfa Bank Plc) - funded its long term public
sector and infrastructure loans either on the interbank market or through other short-term
wholesale funding. Following the Lehman Brothers bankruptcy, it was unable to refinance itself
on the wholesale market and requested State support. Ultimately, the state guaranteed more than
€120 billion of HRE's liabilities and had to inject around €10 billion of capital to nationalise it
23
.
Similarly, Northern Rock faced in the second half of 2007 substantial outflows of wholesale
funds as maturing short-term loans and deposits used to fund its long-term assets were not
renewed. This combined with the inability to tap the securitisation and covered bond markets led
to a request for liquidity support from the Bank of England. The public announcement of this
request led to a run on Northern Rock. The full year net outflow of wholesale funding amounted
to £11.7 billion and by end-2007 a loan from the Bank of England amounted to approximately
£28.5 billion
24
. Ultimately, in 2008, Northern Rock was nationalised. In both cases these crisis
periods were preceded by years of extensive long-term assets growth without a similar increase in
stable funding sources.
23
Source: European Commission
24
See Song (2009):
Reflections on Northern Rock: The Bank Run that Heralded the Global Financial
Crisis,
Journal of Economic Perspectives, Volume 23, Number 1,Winter 2009, Pages 101–119
10
 kom (2016) 0853 - Ingen titel
1715591_0011.png
The CRR introduced a reporting requirement and a general requirement that long-term assets
have to be adequately met with a diversity of stable funding instruments (liabilities) under both
normal and stressed conditions. More detailed requirements to cover funding risk were not set at
that time at EU level given that the BCBS was still in the process of completing its work to
specify the NSFR requirement. Therefore, the current European regime does not provide an
adequate framework to ensure that institutions’ assets are sufficiently stably funded by their
liabilities. The BCBS completed its work and published the NSFR standard in October 2014. In
December 2015, the EBA submitted a report to the Commission on whether and how it would be
appropriate to ensure that institutions use stable sources of funding and on the impact of such a
requirement.
2.2.
Excessive leverage
The financial crisis has shown that institutions' leverage can increase to unsustainable levels and
have a pro-cyclical effect on the financial system. In the run up to the crisis, many investors,
including institutions, actively sought higher yields as high levels of available liquidity resulted
in risk premium falling to historically low levels. Low interest rates, combined with issues of
moral hazard, pushed them to search for higher returns, whether through an increase in leverage
or investment in more risky financial products. This caused a high level of financial fragility of
individual institutions as well as the financial system as a whole. When prices of financial assets
started to fall, institutions had to mark those assets to market thus recognising the losses incurred.
This in turn forced institutions to de-leverage by selling assets in order to minimise regulatory
capital requirements and meet margin calls from their counterparties. This prompted further
decreases in asset prices. In short, institutions’ leverage showed a pro-cyclical pattern: significant
increase of leverage in financial booms and strong de-leveraging in financial downturns
25
.
Equally important, it was observed that institutions that were severely affected by this mechanism
showed strong risk-based capital ratios before the crisis. This is due to the fact that risk-based
capital requirements tend to vary over the economic cycle: they decrease as borrowers'
creditworthiness improves during economic expansions and increase during economic downturns
as borrowers' creditworthiness deteriorates. The combination of incentives for higher leverage
before the crisis on one side and the irresponsiveness of regulatory capital requirements to the
build-up of risk at the macro level on the other side enabled institutions to grow their balance
sheets. While the countercyclical capital buffer introduced by the CRD IV aims at addressing this
pro-cyclicality to a certain extent, it is not considered sufficient as it leaves certain discretion in
setting the buffer rates.
Moreover, as shown by the recent crisis it is difficult to quantify systemic risk as well as to model
accurately the different types of risks, in particular at the micro-level (i.e. at the level of the single
institution). This makes risk-based capital measures less reliable and calls for the introduction of
a simpler and non-risk-sensitive back-stop measure. The misperception of risk may be
exacerbated by a strong industry-wide drive for profit, bonuses and moral hazard due to implicit
safety nets. Hence during favourable macro-economic conditions, institutions would be prone to
25
See, for example, Haldane, A (2015):
Multi-polar regulation, International Journal of Central Banking,
Volume 11(3); Kalemli-Ozcan, Sorensen and Yesiltas (2011):
Leverage across firms, banks, and
countries,
NBER working paper No. 17354; Altunbas, Manganelli and Marquez-Ibanez (2011):
Bank
risk during the financial crisis: Do business models matter?,
ECB Working Paper No. 1394; Beltratti
and Stulz (2012):
The credit crisis around the globe: Why did some bank perform better?,
Journal of
Financial Economics 105, 1-17; Blundell-Wignall and Roulet (2012):
Business models of banks,
leverage and the distance-to-default,
OECD Journal: Financial Market Trends 2012/2
11
 kom (2016) 0853 - Ingen titel
1715591_0012.png
engage in a rapid expansion of their balance sheets without due consideration about implications
for system-wide financial stability. As the ex-ante identification of systemic risks and formation
of asset-bubbles is a very complex exercise, the introduction of a 'hard' leverage ratio would also
help alleviate an excessive expansion of leverage.
Figures 1 and 2 provide an indication of how leverage has evolved for a selected number of credit
institutions in the years prior to the financial crisis compared to risk based capital requirements.
As can be seen the leverage of European credit institutions had increased roughly by half since
1995. Had the leverage ratio requirement been in place before the onset of the financial crisis
there would have been fewer failures during the crisis
26
.
Figure 1.
Total assets to total equity
Figure 2.
Risk weighted assets to Tier 1 capital
Sources: CGFS (2009)
Sources: CGFS (2009)
The leverage ratio framework was introduced in December 2010 by the BCBS in order to: i)
restrict the build-up of leverage in the banking sector (and hence avoid destabilising deleveraging
processes that can damage the broader financial system and the economy) and ii) reinforce the
risk-based requirements with a simple, non-risk based “backstop” measure. The framework did
not foresee an immediate introduction of a capital requirement based on the leverage ratio.
Instead, it set out an expectation that such requirement would enter into force in 2018. In the EU,
the leverage ratio was introduced in the prudential framework in 2013. In line with the BCBS
decision, it was not introduced as a capital requirement that institutions must meet. Rather, the
CRD IV included it in the Pillar 2 framework, while the CRR introduced requirements to
compute it, report it to supervisors and, from January 2015, to disclose it publicly. This has set
regulatory expectations for institutions which has already had a positive impact on the evolution
of the leverage ratio in the EU: the average level of the leverage ratio for Group 1 and Group 2
credit institutions
27
was above 5% and 4.5% respectively as of December 2015 (see figure 3).
Figure 3.
Evolution of the leverage ratio for Group 1 and Group 2 credit institutions
26
27
See Haldane, A. G., & Madouros, V. (2012).
The dog and the frisbee.
Federal Reserve Bank of Kansas
City’s 36th Economic Policy Symposium, p. 1–36.
Group 1 banks are banks with Tier 1 capital in excess of EUR 3 billion and internationally active. All
other banks are categorised as Group 2 banks.
12
 kom (2016) 0853 - Ingen titel
1715591_0013.png
Source:
CRD IV – CRR / Basel III monitoring exercise – results based on data as of 31 December 2015,
EBA, p. 19, Figure 4.
In January 2016, members of the Basel Committee’s oversight body, the Group of Governors and
Heads of Supervision (GHOS)
28
agreed on a Tier 1 definition of capital and a minimum level of
3% for the leverage ratio with the view of making it a Pillar 1 requirement by 1 January 2018.
This international agreement confirmed the market and industry expectations of a binding 3%
leverage ratio. However, only when imposed as a hard capital requirement which must be met at
all times the leverage ratio will be an effective measure requiring institutions to constantly
manage their balance sheet in a way that will prevent excessive de-leveraging during downturns.
A non-binding measure can simply not bring about the same prudential rigour. Furthermore,
given the scope for discretion allowed by the current measures for Member States and
supervisors in their application of the leverage ratio to institutions, the introduction of
harmonised minimum binding requirements across the EU is deemed beneficial in terms of
consistency, effectiveness and promoting coherence in the regulation of institutions as in
principle all would have to meet the 3% requirement.
2.3.
Inadequate calibration of risk weights for exposures to SMEs
SMEs are the backbone of the EU economy and an important source of employment and growth
for the EU economy. They remain largely reliant on bank lending (e.g., credit lines, leasing) to
finance their activities. In fact, other sources of financing, such as equity finance and debt
issuance (e.g. bonds), although available, are not as widely used, or are only used through special
public support schemes.
Following the financial crisis, bank lending to SMEs has suffered a significant drop in volumes,
from a peak of €95 billion in mid-2008 to approximately €54 billion in 2013/2014 and currently
hovers around €60 billion, which is still almost 20% below the level observed in 2003. Lending
to larger corporates, on the other hand, after reaching a higher peak before the crisis and after
experiencing a sharper drop thereafter, is roughly back to the volumes observed in 2003 – 2004
(see figure 4).
Figure 4.
New bank lending to SMEs and larger corporates (EUR million; three-month moving
average)
28
Available at
http://www.bis.org/press/p160111.htm.
13
 kom (2016) 0853 - Ingen titel
1715591_0014.png
Note: SME loans proxied by loans up to and including €1 million; loans to large corporates proxied by
loans over €1 million.
Source: ECB MFI interest rate statistics.
SME are more constrained in receiving external funding also because of their high sensitivity to
economic cycles and shocks, which due to their greater sectorial and geographical specialisation.
Moreover, the asymmetry of information which exists between SMEs and potential lenders, is
particularly acute, and further limits SMEs' ability to switch sources of funding quickly. This
disadvantage is reflected in higher interest rates on small loans when compared to large loans as
well as in other forms of credit constraints. A comparison of the average cost of loans in the EU
shows a significant gap between lending to SMEs and to large firms (see figure 5).
In addition, unlike large corporations, small companies have limited access to capital markets and
thus remain disproportionally reliant on banks. The smaller a firm, the more restricted the
spectrum of potential non-bank funding options (see table 1). Alternative sources of financing are
shown to be accessible only to larger firms, firms having high credit ratings, and firms located in
countries with better developed financial markets. Ensuring that SMEs have adequate access to
finance is therefore a main consideration when setting out policies.
Figure 5.
Yields of and spread between small and large loans, euro area
Source: ECB
14
 kom (2016) 0853 - Ingen titel
1715591_0015.png
Table 1.
Use of financing instruments by non-financial corporations (percentage averages out of
total sample over 2009-2014)
Sources: ECB and European Commission Survey on the access to finance of enterprises; European Central
Bank (2015c), Non-bank financing for euro area NFCs during the crisis, Box 6 in Economic Bulletin, Issue
4.
In the light of the overall increase in capital requirements and in order to avoid
disruptions to lending to SMEs in the aftermath of the financial crisis, Article 501 of the
CRR introduced a 24% discount to capital requirements for exposures to SMEs, the so-
called SME supporting factor, but also included a review clause and asked EBA to
provide a report on the issue by June 2016. EBA published the report in March 2016
29
. It
provided evidence that the capital requirements, including the SME supporting factor,
have overall been consistent with the riskiness of SMEs. The report also indicated that
the €1.5 million exposure cap for the application of the SME supporting factor was not
indicative of a change in riskiness of SMEs. This implies that SME exposures beyond
€1.5 million exposure threshold have been subject to too high minimum capital
requirements in comparison to other bank exposures classes and could have likely
resulted in insufficient lending to SMEs
30
. The issue is heightened by the current
environment of low economic growth and high unemployment and thus requires to be
promptly addressed.
The issue is also underpinned by the views expressed in the responses to the CRR consultation
and the Call for Evidence. Some stakeholders, particularly banks, claimed that the overall
increase in capital requirements had negatively affected their willingness to provide sustainable
financing to the economy. They also claimed that the systematic risk stemming from exposures to
SMEs was lower than for exposures to larger corporates, and asked that the SME supporting
factor should be at least maintained, if not expanded.
2.4.
Weaknesses to the regulatory framework for loss absorption and
recapitalisation capacity
The absence of adequate crisis management and resolution frameworks forced governments
around the world to rescue banks following the financial crisis. The subsequent impacts on public
finances as well as the undesirable incentive effects of socialising the costs of bank failures have
29
EBA report on SMEs,
March 2016
30
The main estimate of the transitional effect, taken from the study of May 2016 conducted by London
Economics using data for the period 1985-2014, shows that for a one percentage point increase in the
Total Capital Ratio the impact on lending by credit institutions in the EU is -0.8% over one year with
the implied impact over a three-year period being -1.5%.
15
 kom (2016) 0853 - Ingen titel
1715591_0016.png
underscored the need for a different approach. The G20 leaders have publicly committed not to
use public funds anymore to bail out banks
31
.
Significant steps have been taken in international fora and at the EU level in order to reduce the
systemic risks of failing banks, through – among others – effective resolution frameworks. A
cornerstone tool of a robust resolution framework is the “bail-in”: a system which consists of,
writing down debt or converting debt claims or other liabilities into equity according to a pre-
defined hierarchy. The tool can be used to internally recapitalise an institution that is failing or
likely to fail, so that its viability is restored. Therefore, shareholders and certain creditors, rather
than taxpayers, will have to bear the burden of an institution's failure.
In the EU, these objectives are already covered by the BRRD. The latter harmonises and
improves the tools for dealing with financial crises across the EU and requires all EU institutions
to meet a Minimum Requirement for own funds and Eligible Liabilities (MREL). The policy
objective of MREL is to ensure that institutions have a sufficient amount of bail in-able liabilities
to allow for smooth and quick absorption of losses and recapitalisation in resolution. After the
agreement of the BRRD in the EU, the Financial Stability Board (FSB) has developed, in
consultation with the Basel Committee on Banking Supervision (BCBS), a new international
standard for G-SIBs
32
. The standard is intended to end "too-big-to-fail" problem by ensuring the
adequacy of G-SIBs' total loss-absorbency capacity (TLAC), should they fail. Indeed, absent
sufficient amounts of readily bail-in-able liabilities, a failure of a G-SIB may either impose large
costs on the global financial system or necessitate fiscal intervention, which is to be avoided. The
possible systemic effects, in particular the possible large costs to other market players and the
economy at large through the contagious effects of interbank exposures, asset fire-sales and
uncertainty among holders of operating liabilities (e.g. derivative counterparties) are illustrated
by the Lehman Brothers case. The MREL requirement and TLAC share the objective of ensuring
that banks have sufficient loss absorption and recapitalisation capacity. TLAC addresses the
particular global systemic problems posed by G-SIBs worldwide, whereas MREL is part of an
EU framework to promote an orderly and feasible resolution or winding down for every bank.
The BRRD framework cannot protect the EU from contagion of the collapse of a third country G-
SIB. The particular global contagion risk and world-wide social costs of a G-SIB's failure by
contrast require a backstop
33
on the minimum requirement on loss absorption and recapitalisation
capacity to ensure that these G-SIBs hold a sufficient amount of bail in-able liabilities so that
they can absorb losses internally without worldwide societal implications or a fiscal intervention
in their favour. As the MREL was designed to be applicable for all types of institutions regardless
of the global systemic implications of their failure, there is no harmonised minimum requirement
but the requirement is to be tailored to each institution by the resolution authority.
31
32
33
1 January 2019, agreed in November 2015 by the Financial Stability Board:
http://www.fsb.org/wp-
content/uploads/20151106-TLAC-Press-Release.pdf
Basel Committee's methodology
for assessing and identifying global systemically important banks (G-
SIBs)
In the context of the development of the TLAC standard, the FSB conducted an analysis of the
historical losses and recapitalisation needs for 13 large banks that failed or received official support.
This report shows that losses and recapitalisation needs vary significantly across banks. Total losses
and recapitalisation needs in terms of total assets are mostly in the range of 4-6 percent, with outliers
around 9 percent. In terms of RWAs total losses and recapitalisation needs are mainly in the range of 5-
15 percent with outliers around 25 percent. Moreover the report concludes that the full extent of the
losses would have even been higher since a number of banks ceased to report separately either because
they failed or were taken over. The FSB used these results as an input for the TLAC standard, including
the calibration of a minimum requirement.
16
 kom (2016) 0853 - Ingen titel
1715591_0017.png
As the failure of a third country G-SIB would impose significant costs on the EU economy
through contagion effects, a global minimum standard is very much in the EU's interest. Other
jurisdictions have not implemented frameworks ensuring minimum requirements for bail-in-able
liabilities like the EU did in the BRRD. Even if this were the case, it would be difficult for the
EU to have confidence in the practical application of a framework comparable to the MREL
requirement in third countries absent a clearly quantified minimum standard. Finally, as third
country G-SIBs are by definition active worldwide and compete with EU banks, from a level
playing field perspective it is also desirable to hold them to a clearly quantified minimum
standard in order to avoid competitive disadvantages that could result from the unilateral
introduction of the EU's sound MREL requirement. However, the EU can only credibly expect
third countries to implement the TLAC standard if it holds its own G-SIBs to the same
requirements.
2.5.
Inappropriate level of capital requirements against trading activities
Financial instruments held by institutions for trading purposes (e.g. shares, bonds,
derivatives), are subject to the risk of movements in their market prices, which has a
daily impact on institutions' profits and losses. These market price movements can be
large and sudden; sudden large drops in market prices can damage the solvency position
of institutions. Because of the idiosyncrasy of this risk, the prudential framework
embedded in Council Directive 93/6/EEC
34
contains a specific regime for these financial
instruments (they are often referred to as trading book exposures), which is different
from that applicable to other types of exposures, such as loans (those are usually referred
to as banking book exposures).
During the financial crisis, the level of capital required against trading book exposures
proved insufficient to absorb losses. Trading book losses in EU institutions were very
substantial and some of those institutions had to be injected State aid and/or resolved as a
result (e.g. Dexia, Royal Bank of Scotland). This revealed a number of weaknesses in the
design of the prudential framework for the trading book, which needed to be addressed.
In 2009, a first set of reforms were finalised at international level (known as the 'Basel 2.5'
package of reforms) and transposed in the EU via Directive 2010/76/EU (CRD III).
35
These
reforms, subsequently retained under the CRR, sought as a main objective to increase the overall
market risk capital requirements to addresses the most pressing deficiencies of the standards on
market risk. However, the 2009 reform did not address the design flaws present in those
standards, such as:
a scope of application of the market risk capital requirements which is not
sufficiently clearly defined. This allows institutions to engage in regulatory
arbitrage, i.e. they can allocate some of their instruments to the regulatory book
that generates the lower capital requirements. As an example, prior to the crisis,
securitisation instruments were usually allocated to the trading book because of
the low volatility of the securitisation markets (leading to low capital
requirements under the market risk rules) even if there was no evidence of regular
trading in these instruments (implying that they had little chances to be traded);
34
35
Council Directive 93/6/EEC of 15 March 1993 on the capital adequacy of investment firms and credit
institutions (OJ
L 141, 11.6.1993, p. 1).
Directive 2010/76/EU of the European Parliament and of the Council of 24 November 2010 amending
Directives 2006/48/EC and 2006/49/EC as regards capital requirements for the trading book and for re-
securitisations, and the supervisory review of remuneration policies.
17
 kom (2016) 0853 - Ingen titel
1715591_0018.png
lack of risk-capture. Many features of market risk are not reflected in the capital
requirements. As a consequence, the amount of capital required for certain instruments is
not aligned with the real risks that institutions face for these instruments. As an example,
the risk of holding more illiquid instruments is not recognised since the current market
risk capital requirements assume that all trading positions can be extinguished within two
weeks;
high variation of modelling outcomes. Internal models used by institutions to calculate
capital requirements for market risk may generate very different estimates of the amount
of capital required for similar portfolios. A comparative study performed by the BCBS
36
across a sample of 15 large banks worldwide (half of them Europeans) with permission
to use internal model showed that, for the same hypothetical diversified portfolio of
trading assets, the bank with the highest capital requirements generated for this portfolio
had capital requirements that were roughly three times higher than those of the bank with
the lowest capital requirements.
Consequently, the BCBS initiated the fundamental review of the trading book (FRTB) to tackle
those flaws. This work was concluded in January 2016, following three public consultations in
May 2012
37
, October 2013
38
and December 2014
39
.
A more comprehensive overview of what went wrong during the 2007-2008 financial
crisis with the trading book framework and of why the Basel 2.5 reforms did not
sufficiently improve the capture of market risk was provided in the BCBS consultation
paper of May 2012.
2.6.
Problems on remuneration rules
CRD IV contains a number of detailed rules on how institutions should determine and
pay out variable remuneration of staff whose activities have a material impact on the
institutions’ risk profile.
Problem 1: Excessive compliance costs arising from the rules on deferral and pay-out in
instruments
Under CRD IV rules, institutions are not allowed to immediately pay out the full amount
of variable remuneration or to pay it entirely in cash. Instead, CRD IV requires that at
least 40% (or in some cases at least 60% of the variable remuneration) be paid out only
after a number of years
40
. It moreover requires that at least 50% of the variable
remuneration be paid out in instruments instead of cash
41
. These rules are applicable to
all institutions, regardless of their size and complexity, and to all identified staff,
regardless of the level of their variable remuneration.
Because of this broad scope of application, compliance with the above requirements
entails high costs outweighing prudential benefits in the following cases:
36
37
38
39
40
41
Available at
http://www.bis.org/publ/bcbs240.pdf.
Available at
http://www.bis.org/publ/bcbs219.htm.
Available at
http://www.bis.org/publ/bcbs265.htm.
Available at
http://www.bis.org/bcbs/publ/d305.htm.
“Deferral”, see Article 94(1)(m) of the CRD.
“Pay-out in instruments”, see Article 94(1)(l) and the second subparagraph of Article 94(1)(o) of the
CRD.
18
 kom (2016) 0853 - Ingen titel
1715591_0019.png
(i) small and non-complex institutions
42
(for instance local cooperative and savings
banks) need to make considerable investments in human resources (HR), information
technology (IT) and advisory services and are faced with difficulties in creating
instruments appropriate for remuneration purposes. According to EBA estimates
43
, the
average one-off costs for these institutions would range from €100 000 to €500 000 per
institution, and ongoing costs from €50 000 to €200 000.
(ii) other institutions also incur important costs resulting from the fact that they need to
apply the rules to all of their identified staff, which will often include a high number of
individuals with only non-material levels of variable remuneration. For instance,
according to EBA estimates
44
, a full compliance by large institutions with the above
requirements in respect of all staff, even that with non-material levels of variable
remuneration, would imply one-off costs ranging from €1 to 5 million, and ongoing costs
ranging from €400 000 to €1.5 million.
At the same time, the prudential benefits of applying the requirements on deferral and
pay-out in instruments in the above cases are low. If a staff member receives only a non-
material level of variable remuneration, then such variable remuneration is unlikely to
provide him/her with incentives to engage in excessively risky behaviour, which would
need correction through deferral and pay-out in instruments. Given that small and non-
complex institutions are typically not among the institutions paying the larger portions of
variable remuneration, and mostly pose lesser risks to financial stability, the prudential
benefit of deferral and pay-out in instruments in their case would be limited.
Problem 2: Excessive compliance costs arising from the requirement for listed
institutions to pay out part of the variable remuneration in shares
Under the CRD IV rules, listed institutions are always required to pay out part of the
variable remuneration in shares; on the other hand, non-listed institutions have the
possibility to use, in addition to or instead of shares, share-linked instruments (Article
94(1)(l)(i)).
Compliance with the pay-out in shares requirement entails important difficulties and
burdens for the approximately 200 institutions that are listed. They would need to either
create new shares or buy them on the market. Both are cumbersome procedures for the
institution. The creation of new shares would risk negatively affecting the shareholders
by diluting their voting rights. The purchase of shares could trigger speculation and force
the institution to pay a premium. Acquiring shares would moreover lead to reducing the
own funds of the institution.
Furthermore, staff remunerated in shares may not be able to sell them because of
problems of insider dealing which is criminally sanctioned, lowering the perceived value
of such remuneration for staff. Moreover, payment in shares in different countries can be
subject to legal, accounting or tax constrains. For example, some institutions with
42
By way of illustration, based on a sample of about 3,200 credit institutions in the EU extracted from the
SNL database, there are around 2,722 credit institutions with total assets of no more than €5bn,
compared to around 303 credit institutions with total assets between €5 and €30bn, and 156 credit
institutions with total assets above €30bn. At EU level, the around 2,722 credit institutions with total
assets below €5bn represent 5.12% of total assets of credit institutions in the sample (however, when
calculated at country level, this percentage differs significantly between Member States).
43
EBA
Opinion on proportionality
44
EBA
Opinion on proportionality
19
 kom (2016) 0853 - Ingen titel
1715591_0020.png
subsidiaries in non-EU jurisdictions (Russia, US) have signalled problems they encounter
with shares to remunerate staff in their non-EU subsidiaries. While these are arguably
significant difficulties and burdens, it is not possible to precisely quantify the absolute
costs resulting from them for listed institutions.
At the same time, an equally effective yet less difficult and burdensome alternative for
shares exists, namely share-linked instruments.
This means that, in the case of listed institutions, the requirement to pay out part of the
variable remuneration exclusively in shares entails unnecessary compliance costs
compared to other available alternatives with similar prudential benefits.
2.7.
Problems on insolvency ranking of unsecured bank debt instruments
One of the key objectives of the BRRD is to facilitate private sector loss absorbency in the event
of a bank crisis. To achieve this objective, all banks are required to meet a Minimum
Requirement for Own Funds and Eligible Liabilities (MREL) to ensure that sufficient financial
resources are available for write down or conversion into equity. Under the BRRD, MREL does
not generally require mandatory subordination of eligible instruments for MREL. This means, in
practical terms, that a liability eligible for MREL may rank in insolvency at the same level (pari
passu) with certain other liabilities which are not bail-inable in accordance with the BRRD (e.g.
operational liabilities, such as short-term inter-bank loans), or certain other liabilities which are
bail-inable, but could be excluded from bail-in on a discretionary basis (as allowed under the
BRRD) if the resolution authority can justify they are difficult to bail-in for reasons of
operational execution or systemic contagion risk (e.g. derivatives, structured notes). This could
lead to situations where bailed-in bondholders may claim they have been treated worse under
resolution than under a hypothetical insolvency. In such case, they would need to be compensated
by financial means of the resolution fund. To avoid this risk, resolution authorities may decide
that the MREL requirement should be met with instruments that rank in insolvency or resolution
below other liabilities that are either not bail-inable by law or difficult to bail-in (“subordination
requirement”). Harmonising the ranking of unsecured bank debt holders in insolvency and
resolution would provide the means to ensure an effective and transparent bail-in, especially in
cross-border cases and would provide certainty and clarity to investors and resolution authorities.
In addition to the MREL standard for which subordination of debt instruments could be required
by resolution authorities to the extent it is needed to facilitate the application of the bail-in tool in
a given case, the minimum TLAC requirement for G-SIBs, as clearly stated by the FSB Term
Sheet
45
, should be met using a certain amount of subordinated debt instruments.
The results of international negotiations and the consensus among Member States indicate that
the future EU TLAC standard applicable to G-SIIs will stay aligned with the FSB TLAC Term
Sheet as regards the subordination condition. This means that G-SIIs will have to satisfy the
TLAC level with instruments that are subordinated to other excluded TLAC instruments (e.g.
operational liabilities) with the aim to enhance the operational execution and robustness of bail-in
powers and to avoid legal uncertainty.
45
Total Loss-Absorbing Capacity (TLAC) Principles and Term Sheet,
FSB, 9 November 2015
20
 kom (2016) 0853 - Ingen titel
1715591_0021.png
The TLAC requirement to hold subordinated instruments combined with the potential
discretionary request by the resolution authority to meet MREL also with subordinated
instruments have driven some Member States to re-assess national insolvency ranking.
A number of Member States have amended (or are in the process of amending) the insolvency
ranking of certain banks’ creditors under their national insolvency law to operationalise the
possible application of the bail-in tool and to ensure that banks comply with the “subordination
requirement” of the international FSB standards on TLAC for G-SIBs.
As the national rules adopted so far diverge significantly, they can create competitive distortions
in the single market and complicate the operationalization of the bail-in tool, in particular for
cross-border banks. Moreover, the national approaches have very different effects on G-SIIs’
ability to address potential shortfalls in meeting TLAC standards. Under some approaches TLAC
shortfalls were addressed with immediate effect through statutory retroactive subordination of the
existing stock of unsecured senior debt, possibly without the issuance of new debt instruments,
meaning a limited additional cost of funding was incurred to become TLAC compliant. Under
other approaches banks would likely need to issue new debt, which meets the subordination
criterion, at a higher marginal cost than senior debt for the period running to and after the TLAC
compliance date. The effects ultimately depend on a bank’s shortfall of TLAC eligible
instruments and its liability structure, but two banks with comparable shortfalls and liability
structures could face significantly different treatment depending on the insolvency ranking of
unsecured debt in their respective jurisdictions. Additionally, the creditors of banks under such
divergent national insolvency regimes would be treated very differently when buying the claims
of banks falling under different national hierarchy of creditor regimes.
There is a broad agreement among stakeholders that having divergent approaches to the statutory
insolvency ranking of bank creditors provides uncertainty for issuers and investors alike and
makes more difficult the application of the bail-in tool for cross-border institutions. This
uncertainty could also result in competitive distortions in the sense that unsecured debt holders
could be treated differently in different jurisdictions and the costs to comply with the TLAC and
MREL requirement for banks may be different from jurisdiction to jurisdiction.
In its conclusions of 17 June 2016
46
, the Council invited the Commission to put forward a
proposal on a common approach to the bank creditors' hierarchy. During the meeting of the
experts of the European Parliament and of the Member States of 23 of June 2016, a large number
of Member States communicated that they were clearly in favour of harmonisation and endorsed
partially harmonised EU approach to subordination. They insisted, however, that any EU
approach should provide sufficient flexibility to take account of different bank business models
across the EU and reduce at minimum impacts on bank funding costs.
2.8.
Lack of effectiveness of the current rules on moratorium
A moratorium tool can be broadly defined as the power to temporarily suspend payments or
performance of obligations and / or temporarily prohibit contracting new obligations.
Use of moratorium in a supervisory/resolution context can be useful in several scenarios:
46
Council conclusions of 17 June 2016 on a roadmap to complete the Banking Union:
http://www.consilium.europa.eu/press-releases-pdf/2016/6/47244642837_en.pdf
21
 kom (2016) 0853 - Ingen titel
for liquidity stabilisation: in case of severe liquidity outflows, an institution could
have to sell assets at a discount (“fire sale”). This creates losses which are bound to
be borne by creditors and particularly those “left behind”. Even in the absence of fire
sales, there might be a first mover advantage that sparks a bank run: the first creditors
to redeem their claims would be repaid fully, while those who act late will face losses
(due, for example, to asset discounts in an insolvency procedure). A moratorium
could ensure a stabilization of the liquidity position and equal treatment of creditors
and foster financial stability by eliminating the first mover advantage. Potentially this
may, depending on the circumstances, also address the contagion issue;
to ensure stability in the pre-resolution phase: a moratorium can help ensure the
stability of an institution in the days leading up to resolution, provide ample time for
the resolution authority to conduct a prudent valuation and determine, for example,
the appropriate amounts for bail-in;
to restore the capital position of the institution: the use of a moratorium tool in a
supervisory context can be a useful tool to address temporary issues with respect to,
for example, the composition of a bank's capital;
to prevent increases in secured funding: an institution that is experiencing distress
may not be able to issue unsecured (term) debt. Such an institution would need to
attract secured funding, which may require that in order to provide safety to the new
secured creditors the secured claim might have to be over-collateralised. If over-
collateralised funding increases significantly, this effectively increases the loss rate
for unsecured creditors as well as possibly depositors / the DGS in case of default.
Stabilizing the liquidity situation through a moratorium could prevent such an effect,
while ensuring the equal treatment of creditors.
The issue of the harmonisation of moratorium tools was raised in the meetings of the Council Ad-
hoc Working Party on strengthening the Banking Union. In that context a questionnaire was
submitted to the Member States and the ECB, and the Dutch Presidency produced two non-
papers on the topic, mainly summarising Member States replies to the survey. The Council
conclusions of 17 June 2016 invited the Commission to conduct further work on whether and
how further harmonisation of the rules and application of moratorium tools can contribute to the
stabilisation of an institution in the period before, and possibly after, an intervention. Further to
that, DG FISMA carried out internal analysis and consultations – including a questionnaire –
with national experts to assess the most appropriate way forward.
An uneven playing field resulting from the identified differences listed below would lead to
detrimental consequences and could hamper the effectiveness of resolution tools in a cross-border
scenario. For example, the very different duration of the suspension from one Member State
would make it more difficult for a resolution authority to devise a consistent resolution strategy
cross-border. Also, it would in certain cases impair the effectiveness of the moratorium tool
altogether because it would create an incentive for creditors to move their investments in the
bank to countries where the duration of the suspension is shorter.
Similarly, the possibility allowed by certain national legislations to use the moratorium as an
early intervention tool, which would allow supervisors/resolution authorities to intervene more
effectively at an earlier stage when the specific situation of the bank requires, may be impaired
by the different approaches at national level in this respect. An effective application of the tool in
a cross-border scenario would be greatly reduced if supervisors / resolution authorities were not
in a position to apply the same tools across the board at the same time.
22
 kom (2016) 0853 - Ingen titel
1715591_0023.png
Existing provisions in CRD IV and BRRD already provide some basis for competent authorities
to exercise certain moratorium powers. In particular, Member States transposed provisions on
moratorium in very different manners
47
. This can negatively impact the practical application of a
moratorium and create an uneven level playing field. Therefore these provisions may be
improved and further harmonised to make existing tools more effective by enhancing legal clarity
and providing further certainty in a cross-border scenario.
All Member States have some type of moratorium tools available in their jurisdiction. Most
introduced these tools in their legislative framework as a result of the transposition of BRRD or
CRD IV
48
. The relevant national provisions however vary in terms of scope of the liabilities
covered (particularly with respect to covered deposits and payment systems), means of activation
(supervisory / resolution / both), and duration.
With respect to the scope, in several Member States moratorium powers extend also to
covered
deposits
(12 MS). In most Member States a moratorium intervention on covered deposits would
be considered as a pay-out event and would therefore trigger the application of the Deposit
Guarantee Scheme.
49
Payment obligations to
CCPs or payment settlement systems
are on the
other hand excluded from the scope of moratorium powers in most MS (9).
50
Marked differences in transposition can be encountered also with respect to the duration of the
payment suspension in case of moratorium. Most national legislations (16 MS) provide a
predetermined maximum duration. The duration can however range widely (from one working
day to twelve months). Some Member States have comparatively short durations of twenty days
or one month, while the most frequent indicated maximum duration is six months. Several
Member States indicated that an extension of the suspension period would be possible (although
these extensions are sometimes also subject to a predetermined maximum).
51
47
48
49
50
51
Information provided below on existing moratorium tools at national level was provided by Member
States' experts in response to a questionnaire circulated by the Dutch Presidency in the context of
technical meetings with Member States. These were followed-up by the Commission with direct
exchanges with the relevant MS on specific issues.
24 MSs responded to the questionnaire (EE, DK, BG, ES, FI, SE, HR, LU, FR, PT, SK, BE, PL, AT,
IE, EL, CZ, RO, LV, HU, LT, DE, UK, MT). The only ones who indicated that they do not have any
type of moratorium tool are DK and SE (while MT indicated that the concept of a moratorium tool does
not exist in national law but underlined the general scope of the powers of national competent
authorities). BE does not have a full-fledged moratorium in place but a similar tool to be used as an
extraordinary recovery measure.
The Deposit Guarantee Scheme Directive (Directive 2014/49/EU) provides the rules and principles for
the protection of covered deposits (deposits below 100.000 Euros). According to the Directive, in
presence of a pay-ut event – an event which indicates that the bank is not in a position to repay the
deposit for reasons connected to its financial circumstances activates the use of the DGS to protect such
deposits. EE, BG, FI, LU, SK, BE, AT, IE, EL, HU, LT, DE indicated that covered deposits fall in the
scope while ES, HR, PL, CZ, LV, FR and UK indicated that such deposits are not subject to
moratorium powers. PT indicated that while the national provision transposing Article 63 BRRD does
not apply to covered deposits, for other moratorium tools existing at national level an exemption of
such liabilities is not foreseen. Out of those MSs who include covered deposits in the scope, BG, LU,
PT, BE and AT indicated that this would constitute a pay-out event under DGSD transposition laws.
BG, FI, LU, SK, BE, AT, IE, LV, DE provided a positive answer to the question. EE, ES, HR, FR, PT,
PL, EL, CZ, HU, LT,UK and MT do not apply moratorium tool to payment obligations owed to CCP or
payment settlement systems
EE, BG, ES, FI, HR, LU, PT, SK, PL, AT, IE, EL, CZ, HU, LT, UK indicated that the suspension has a
maximum duration. The most common indicated maximum duration is six months (EE, LU, AT, IE,
CZ, LT). Others indicated one month (BG), 20 working days (EL), 90 days (HU) or a longer period of
12 months (SK). Finally, some referred to the very short duration indicated in Art. 69 BRRD (midnight
23
 kom (2016) 0853 - Ingen titel
1715591_0024.png
Moreover, with respect to the intervention phase for moratorium tools, legislative provisions at
national level do not appear consistent. While more consistency can be observed with regards to
moratorium powers applied under resolution, national legislative frameworks seem to follow
different approaches with respect to the use of such tools in the early intervention phase. In
several countries moratorium powers can be activated during the early intervention phase and in
this context the precautionary powers provided by the CRD IV framework can usually be
exercised. In other countries, however, a moratorium power seems to be considered eminently a
resolution-related tool and can only be activated once a bank is deemed to be failing or likely to
fail or put under resolution.
52
Finally, some of the consulted stakeholders highlighted possible means to improve this tool, such
as on the duration of the suspension and its scope.
2.9.
Insufficient proportionality of the current rules
It can be argued that the CRR and the CRD IV are already "proportionate" to a large extent,
insofar as they take into account the size, complexity and business model of institutions for
various purposes. The framework as a whole is formulated in a modular manner, such that
institutions must only apply those requirements which are relevant to the risks they incur.
Furthermore, the framework provides for specific exemptions and preferential treatments for
various purposes (e.g. own funds, liquidity, covered bonds), thus reflecting the relative
complexity and riskiness of institutions and the activities they undertake.
Nevertheless, several Member States and Members of the European Parliament have raised the
concern that the current EU regulatory framework does not sufficiently differentiate between the
very large systemic institutions and very small local institutions. Moreover a sizable number of
respondents to the CRR consultation and the Call for Evidence submitted that, in their view,
some of the prudential requirements in the CRR and CRD IV may impose a disproportionate
burden on smaller and less complex institutions.
Respondents to the Call for Evidence singled out complexity of rules, administrative burden and
compliance costs as the most pressing concern for smaller institutions. They argued that costs
resulting from complex prudential rules create a competitive advantange for larger institutions
insofar as these can benefit from economies of scale to allocate more resources to compliance
functions. In particular, respondents pointed to costs resulting from current CRR and CRD IV
requirements on:
52
of the following day - UK). It seems that most MSs were referring to the moratorium tool as per Article
63 BRRD (since Article 69 contains a precise duration of the suspension). However, responses to the
questionnaire were not very clear in this respect and of course the reading depends to an extent on how
MSs transposed the relevant BRRD provisions in national law.
12 MSs, namely EE, ES, FR, PT, BE, EL, LV, HU, LT, UK, MT provided positive answer to the
question and indicated that moratorium tools are or seem to be intended also as early intervention tools
(or in the case of LT, simply that the tool is not attached to any specific phase in the
supervision/resolution process). Out of these, FR and ES indicated that the power derives from the
transposition of CRD provisions. Other respondents, an particularly IE and CZ, gave more nuanced
answers, highlighting that the criteria to apply moratorium tools are different than those that justify
early intervention but that a moratorium could have effects also towards a bank that is subject to early
intervention measures.
24
 kom (2016) 0853 - Ingen titel
EU harmonised (Pillar 1) reporting, whose volume and frequency was regarded as
disproportionate for smaller institutions, as well as reporting required by supervisors
under Pillar 2 on an ad hoc basis, over and above Pillar 1 reporting;
disclosure of capital and liquidity requirements, which applies to all institutions in
largely the same fashion, as a result of which it was regarded as too detailed and
frequent for smaller institutions and of little practical use for institutions with no
publicly traded securities; and
the complexity and large volume of rules that respondents have to deal with and the
inability to keep up with all the changes in the legislation.
Section 4.9 below discusses various potential policy options to address undue burden on smaller
institutions resulting from reporting and disclosure requirements.
Whilst these measures address proportionality issues related to the size of a credit institution,
other measures proposed in the impact assessment address proportionality concerns related to
credit institutions' business model (e.g. the types of activities carried out).
More precisely, where relevant, each section in this impact assessment discusses specific policy
options and limited exceptions tailored to simpler or less risky business models or activities
undertaken by any institution, including for these purposes smaller institutions (see sections on
TLAC, lending to SMEs, trading book, leverage ratio, NSFR and remuneration).
This approach has been chosen taking into account the specificities of the banking sector in the
EU, where the market is highly polarised (i.e. there is a very large gap between the biggest and
the smallest banks) and the composition (i.e. size and type of business models of banks) of the
banking sector across Member States is highly different. The possibility of developing a 'lighter
regime' across the board for small/less complex EU credit institutions would be very complex
since solutions that could work for a certain type of credit institutions might not work for others.
Instead, the introduction of tailored measures for different metrics (e.g. TLAC, lending to SMEs,
trading book, leverage ratio, NSFR and remuneration) ensure a degree of flexibility able to cover
credit institutions with different sizes and business models in all Member States.
2.10. Consequences from the baseline scenario
Not dealing with the problems described above would have several broad potential consequences:
from the
safety
point of view they include mispricing of risk, inadequately
capitalised or funded individual institutions and too-big-to-fail institutions. All of
these would ultimately lead to a higher probability of financial crises in the future
and to higher economic and social costs of those crises, both in terms of foregone
output and unemployment;
from the point of view of
smaller institutions,
they include a continued high
level of administrative costs;
from the point of view of the services provided by institutions to the
EU
economy
- to the extent that the current regulatory framework imposes capital
requirements which are disproportionate to the actual risks faced by those
institutions - they include an insufficient supply of those services (e.g. lending to
SMEs or client clearing services).
Looking at the individual areas, more detailed consequences would likely materialise.
25
 kom (2016) 0853 - Ingen titel
1715591_0026.png
On stable funding of banks,
while the LCR ensures that banks will be able to withstand a severe
stress on a short-term basis it does not ensure that they will have a sustainable stable funding
structure on a longer-term horizon. General requirements on stable funding and market discipline
would likely mitigate some of risks related to insufficiently stable funding, but are unlikely to
prevent banks from relying on too-high amounts of short-term funding. Banks would therefore be
more prone to liquidity problems in situations where markets for short-term funding were
disrupted. This would likely lead to the failure of those banks and potentially even to a new
financial crisis.
On the loss absorption of systemically important institutions,
there would be no backstop on
the minimum loss absorption and recapitalisation capacity in G-SIIs, the level playing field
between G-SIIs could be difficult to assess and there would be no incentive for other jurisdictions
to impose a similar framework on third country G-SII. Each of these elements would impose
significant costs on the EU economy in case of failure of a G-SII.
On the leverage ratio,
not implementing a capital requirement based on this ratio would mean
that the risk of excessive leverage would continue to be monitored by supervisors during the
supervisory review process and institutions would have to calculate, report and disclose the
leverage ratio. However, the combination of market discipline and supervisory review would not
serve as an effective deterrent against excessive leverage of institutions compared to a binding
leverage requirement and thus risks to financial stability would remain. Furthermore, there would
be no backstop to risk-based capital requirements calculated using institutions' internal models
(as the existing backstops expire will expire at the end of 2017). Finally, the effects of economic
cycles would not be addressed properly as risk-based capital requirements alone are insufficient
to deal with this issue.
On market risks,
the weaknesses and design flaws of the current prudential framework for
trading book transactions will remain unaddressed. As the result, the allocation of capital
requirements across those transactions may still be inadequate as compared to the true risks faced
by the institutions. For certain transactions in the trading book, institutions subject to the CRR
would therefore not have sufficient amounts of capital to absorb the potential losses that may
arise from adverse changes to the market conditions for those transactions. Institutions with very
concentrated portfolios in those transactions would suffer significant losses, potentially requiring
State aid and/or be resolved as a result. Other transactions of the trading book may suffer from an
excess of capital requirements which would continue negatively affecting the market liquidity
and transactions costs.
On
the SME supporting factor,
leaving the existing rules unchanged would ensure the
continuity of the current regulatory framework with no new compliance burden.
Maintaining the status quo would be also in line with EBA's findings, which
demonstrated that the SME SF had been found to be consistent with actual systematic
riskiness of EU SMEs
53
, except for retail exposures of banks using the Internal Ratings-
Based (IRB) approach. This option would also address numerous calls from banks to the
CRR consultation
54
and the Call for Evidence for retaining the SME SF in the CRR. The
53
EBA report shows that this was indeed the case in Germany, France and Ireland, whereby additional
capital relief banks obtained from the SF was consistent with the systematic riskiness of SME loans,
except for retail exposures (i.e. less than 1 million euros) of banks using IRB approach. See paragraph
on Option 3 for further details.
54
http://ec.europa.eu/finance/consultations/2015/long-term-finance/index_en.htm
26
 kom (2016) 0853 - Ingen titel
stability of the regulatory framework would ensure the consistency in monitoring of the
use of the SME SF in accordance with Article 501(3).
Moreover, the EBA report provides evidence showing that additional capital reduction
for SME exposures above the current €1.5 million exposure threshold could still be
consistent with the riskiness of these exposures. Not providing further capital reduction
for SME exposures above €1.5 million would thus likely result in a sub-optimal level of
bank financing of these SMEs.
On
remuneration,
the application of the rules on deferral and pay-out in instruments to
small and non-complex institutions, as well as towards staff with low, non-material levels
of variable remuneration would trigger for the institutions concerned important
compliance costs and burdens. This would also translate into non-negligible supervisory
burden for competent authorities. At the same time, the prudential benefits of applying
those requirements to small and non-complex institutions and towards staff with non-
material levels of variable remuneration would be low.
Moreover, listed institutions would have to sustain important compliance difficulties resulting
from the requirement to use shares in fulfilment of the requirement under Article 94(1)(l)(i) of
the CRD IV, while the prudential benefit would not be any higher than in case of the use of
share-linked instruments.
On
insolvency ranking,
the current heterogeneity of approaches would lead to a confusing and
unclear situation for investors and create an uneven playing field for both banks and investors
which could be detrimental for the European debt market or even lead to regulatory arbitrage.
This fragmentation would likely lead in some countries, to a less liquid and more expensive
market for European TLAC eligible debt which could have a negative impact on banks’ funding
costs and their ability to roll-over debt. This could arise for instance in cases where creditors,
who have been statutorily subordinated by law, could be incentivised to limit their exposures to
that particular market potentially impacting liquidity and driving funding costs up. Along a
similar line, banks whose unsecured debt has been statutorily subordinated would be potentially
incentivised to move into riskier funding (e.g. derivatives, structured products) rather than roll-
over subordinated debt that is in excess of their TLAC holding.
With regards to transparency and clarity, it is expected that investors would be able and willing to
evaluate the insolvency laws of Member States with sizable capital markets, but might be
reluctant to do so for 28 different regimes. This could be to the detriment of Member States with
less developed capital markets.
Furthermore, the heterogeneity of approaches would increase the complexity for resolution
authorities to set the minimum requirements for bail-inable liabilities and might impede the
effectiveness of the bail-in tool, especially for cross-border groups.
Most Member States and stakeholder groups acknowledge these risks associated with divergent
national insolvency regimes and are clearly in favour of a partial harmonisation of creditor claims
on unsecured liabilities.
On
moratorium,
the diversity of national approaches to the implementation of the tool as well as
the lack of clarity of certain elements may reduce the effectiveness of this tool and result in
undesired consequences such as bank runs or reduction of liquidity in a supervisory/resolution
context.
27
 kom (2016) 0853 - Ingen titel
On
proportionality,
costs resulting from complex prudential rules and high administrative
burden would maintain the current competitive advantange of larger institutions insofar as these
can benefit from economies of scale to allocate more resources to compliance functions. Failure
to embed more proportionality in the prudential rules in an adequate fashion would result in
excessive compliance costs for institutions, an uneven playing field for smaller institutions and
barriers to entry for potential new market players.
The transmission mechanism is shown in the problem tree below.
28
 kom (2016) 0853 - Ingen titel
1715591_0029.png
Figure 6.
Problem tree
Problem drivers
Shortcomings in banks' risk management on
maturity matching between assets and liabilities
Lack of explicit pan-EU regulatory measures
mitigating the funding risks
Banks' incentive to increase leverage in order
to maximise their profits and return on equity
Lack of explicit pan-EU regulatory limits to
leverage of credit institutions
Risk-based capital requirements cannot fully
ensure that all risks are adequately captured
EUR 1.5 mio threshold for SME exposures is not
indicative of a change in riskiness of an SME
Lack of capacity in G-SIBs worldwide to
absorb losses from major financial crisis
without the taxpayer support
MREL does not impose a minimum
requirement on the loss absorbing and
recapitalisation capacity for SIFIs
G-SIBs worldwide are not subject to a clear
minimum standard
Rules are disproportionate for smaller and less
complex institutions
All institutions are required to apply the
requirements on deferral and pay-out in
instruments, and all listed institutions must
use shares
Lack of clarity with regard to the boundary
between the trading and the banking book
Lack of risk sensitivity to address certain
market risks (e.g. tail risk or liquidity risk)
Lack of consistency between standardised and
internal approaches for market risks
Lack of rigour in the model approval by
supervisors, of more consistent identification
and capitalisation of material risk factors
across banks
Heterogeneous approaches to the insolvency
ranking of senior bank debt as well as national
differences in the implementation of
moratorium tools
Problems
Excessively
reliance on short-
term wholesale
funding to finance
long term
activities
Excessive
leverage of credit
institutions
Consequences
Relatively too
high capital
requirements for
some SME
exposures
Risk of disorderly
failure of
systemically
important banks
Insufficient
proportionality of
the current rules
leading to
excessive
compliance
and/or
administrative
costs
Inappropriate
level of capital
requirements
against trading
activities
Un-level playing
field
Significant
risks to
financial
stability
Likely
continuing
taxpayer
support in
the failure of
'too-big-to-
fail'
institution
Risk to
sustainable
bank
financing of
the economy
29
 kom (2016) 0853 - Ingen titel
1715591_0030.png
3.
OBJECTIVES
3.1.
General, specific and operational objectives
There are three broad general objectives behind the initiative: contributing to
financial
stability, reducing the likelihood and the extent of taxpayers' support in bank
resolution
as well as contributing to
sustainable financing of the economy.
These can be broken down in the following, more
specific objectives:
enhance risk-capturing (incl. risk-sensitivity) of the prudential framework so that
it better reflects all the different risks embedded in the banking activity (S-1);
increase proportionality of rules that lead to unnecessary administrative burden
and compliance costs (S-2);
enhance the level playing field and reduce risk arbitrage opportunities (S-3);
enhance capacity of loss-absorption and recapitalisation of G-SIBs worldwide (S-
4);
enhance legal certainty and coherence (S-5).
Table 2.
Mapping of problems and objectives
Problems
Problem Drivers
Operational
Objectives
Introduce regulatory
measures ensuring a more
stable funding structure for
EU banks
Field
Specific Objectives
S-1
S-2
S-3
S-4
S-5
Funding risk
Shortcomings in banks' risk
management on maturity
matching between assets and
liabilities
Fragility of banks which
excessively use short-term
wholesale funding to finance
long term activities
Un-level playing field
Lack of explicit pan-EU
regulatory measures mitigating
the funding risks
Inadequacy of the current
regulatory framework to
address funding risks over the
long term and to ensure that
banks finance their long term
activities with stable sources of
funding
√
√
√
√
Develop and implement
appropriate
methodology/metrics to
measure the degree of
stability of liabilities and
of liquidity of assets over a
one-year horizon
√
√
√
√
30
 kom (2016) 0853 - Ingen titel
1715591_0031.png
Problems
Problem Drivers
Operational
Objectives
Field
Specific Objectives
S-1
S-2
S-3
S-4
S-5
Overextension of credit in
the economic upturn
resulting in excessive de-
leveraging spiral during the
economic downturn
Banks' incentive to increase
leverage in order to maximise
their return on equity without
taking into account externalities
Limitations in risk
measurement, information
asymmetries and inappropriate
responses to risk and changes
in economic conditions
Risk-based capital
requirements cannot fully
ensure that all risks are
adequately captured.
Lack of explicit regulatory
limits to leverage of credit
institutions
Excessive leverage
Generous discretionary
distributions during periods
of stress, when capital should
be conserved
A too favourable picture of
the financial robustness of
the financial institutions
leading to further leverage
and reduced the resilience of
the financial sector to future
shocks
Provide a backstop to the
risk sensitive capital
requirement
√
√
√
√
SME exposures
Banks' ability to provide
adequate funding to EU
SMEs could be hampered,
particularly by the most
capital-constrained banks
Capital requirements for SME
exposures beyond €1.5 mio
threshold do not reflect
sufficiently systematic risk
stemming from SMEs and
consequently are too high in
comparison to other exposures
classes
Lack of G-SIBs' capacity to
absorb losses from major
financial crisis without the
taxpayer support
Re-calibrate Risk Weights
for exposures to SME
loans so that they better
reflect risks of SME
exposures
√
√
√
√
Loss absorption and
recapitalisation capacity
Distress or disorderly failure
of a G-SIB anywhere in the
world would create
significant disruption to the
wider financial system and
economic activity
Introduce a minimum
requirement on the loss
absorbing and
recapitalisation capacity of
systemically important
institutions
Introduce a minimum
requirement on the loss
absorbing and
recapitalisation capacity of
systemically important
institutions
Establish a more objective
boundary between the
trading and the banking
book
Provide a more prudent
capture of “tail risk” and
capital adequacy;
Incorporate varying
liquidity horizons into the
revised SA and IMA
Make a standardised
approach more risk-
sensitive to serve as a
credible fall-back for, as
well as a floor to, the
Internal Models Approach
√
√
Un-level playing field
G-SIBs worldwide are not
subject to a clear minimum
standard
√
√
Possible bank failures
resulting from:
-inadequately captured risks
inherent to banks' trading
book resulting from:
-regulatory arbitrage between
banking and trading books
-high risk of a sudden and
severe impairment of market
liquidity across asset markets
Lack of consistency and
comparability among banks
using models to calculate
their capital requirements for
some trading book positions
Market Risk
Lack of clarity with regard to
the boundary between the
trading and the banking book
Lack of risk sensitivity of the
whole framework to addressee
certain risks (e.g. tail risk or
liquidity risk)
√
√
√
√
√
√
√
√
Lack of consistency between
standardised and internal
approaches
√
√
√
√
31
 kom (2016) 0853 - Ingen titel
1715591_0032.png
Problems
Problem Drivers
Operational
Objectives
Field
Specific Objectives
S-1
S-2
S-3
S-4
S-5
Lack of rigour in the model
approval by supervisors, of
more consistent identification
and capitalisation of material
risk factors across banks
Constraints on the capital-
reducing effects of hedging and
diversification
Fragmented framework for
insolvency ranking for
unsecured bank debt:
Possible competitive
distortions due to differences
in cost of funding and
different treatment for
investors, as well as investor
uncertainty and asymmetry
of information costs.
Probability that claims arise
due to a breach of the no-
creditor-worse off principle
differs from MS to MS.
Bank's ability to use
unsecured debt to meet
TLAC and MREL may also
differ. This could lead to
possible competitive
distortions in the EU debt
markets.
Introduce a revised internal
models-approach (IMA)
√
√
√
√
Insolvency ranking
Divergent approaches for
ranking unsecured debt holders
in insolvency creating debt
market fragmentation and
uneven playing field.
Different investor treatment
and cost of funding impact for
banks which need to issue
TLAC eligible instruments to
satisfy shortfalls.
Pari passu ranking of unsecured
bank debt with liabilities that
are more likely to be excluded
from bail-in for operational
reasons, increasing the risk of
legal challenge and likely to
hinder the operational
execution of bail-in.
Enhance clarity for
investors and issuers, by
partially harmonising the
hierarchy of unsecured
claims in insolvency.
Enable banks to meet
TLAC/MREL shortfalls in
due time, with a tailor-
made solution and more
clarity on costs, under
fairer competitive
conditions,
Enable banks to maintain
flexibility in adequately
choosing the funding mix.
Avoid competitive
distortions that result from
different treatment of
unsecured bank debt
holders under various
national insolvency laws.
Increase the robustness of
the bail-in tool.
√
√
√
Moratorium
Lack of a level playing field
in the application and
implementation of
moratorium tools
Potential lack of clarity with
respect to important issues
such as duration, intervention
phase, scope
Excessive compliance costs
arising from the rules on
deferral and pay-out in
instruments
Variety of approaches at
national level
Enhance consistency
across EU Member States
and improve clarity for
supervisors and resolutions
authorities as well as
creditors and provide an
effective tool to be used
when assessing banks'
liquidity in a
supervisory/resolution tool
√
√
Excessive compliance costs
arising from the requirement
for listed institutions to pay
out part of the variable
remuneration in shares
The existing CRD IV rules on
deferral and pay-out in
instruments are applicable to all
institutions, regardless of their
size or complexity, and to all of
their identified staff, regardless
of the level of their individual
variable remuneration
The existing CRD IV rules
require listed institutions to pay
out a part of the variable
remuneration in shares
Remuneration
Eliminate excessive costs
related to compliance with
the rules on deferral and
pay-out in instruments,
without posing risks to
financial stability
√
Eliminate excessive costs
for listed institutions
related to compliance with
the rules on payment in
shares, without posing
risks to financial stability
√
32
 kom (2016) 0853 - Ingen titel
1715591_0033.png
Problems
Problem Drivers
Operational
Objectives
Reduce administrative
burden and compliance
costs for smaller
institutions
Enhance the modular
approach of the CRR/CRD
IV to take into account risk
profile and complexity of
institutions and the
activitities they undertake
Maintain overall
consistency of the
prudential framework for
all institutions
Field
Specific Objectives
S-1
S-2
S-3
S-4
S-5
Disproportionate compliance
and administrative costs
Disclosure and reporting
requirements are burdensome
and disproportionate for
smaller institutions
Prudential requirements need to
take into account the risk
profile and complexity of
institutions and the activitities
they undertake
Proportionality
Overall framework too
complex and burdensome
√
√
√
√
3.2.
Consistency of the objectives with other EU policies
Four years after the European Heads of State and Governments agreed to create a Banking
Union, two pillars of the Banking Union – single supervision and resolution – are in place,
resting on the solid foundation of a single rulebook for all EU banks. While important progress
has been made, further steps are needed to complete the Banking Union.
The CRR/CRD IV review is part of this effort and the overall objective of this initiative, as
described above, are fully consistent and coherent with the EU's fundamental goals of promoting
financial stability, reducing the likelihood and the extent of taxpayers' support in bank
resolution
as well as contributing to a harmonious and
sustainable financing of economic
activity,
which is conducive to a high level of competitiveness and consumer protection (Article
169 TFEU).
These overall objectives are also in line with the objectives set by major EU initiatives such as
the Juncker investment plan (EFSI), a proposal for European Deposit Insurance Scheme (EDIS)
and its focus on risk reduction as well as with the objective of moving towards a Financial Union,
with the completion of the Economic and Monetary Union and the creation of a Capital Markets
Union. Some of the proposed provisions on leverage, liquidity and loss-absorbance capacity in
particular are also consistent with internationally agreed standards (Basel Committee and FSB) to
which the EU has actively contributed and committed to implement.
3.3.
Consistency of the objectives with fundamental rights
The EU is committed to high standards of protection of fundamental rights and is
signatory to a broad set of conventions on human rights. In this context, the proposed
measures as discussed above are not likely to have a direct impact on these rights, as
listed in the main UN conventions on human rights, the Charter of Fundamental Rights
of the European Union which is an integral part of the EU Treaties, and the European
Convention on Human Rights (ECHR).
33
 kom (2016) 0853 - Ingen titel
1715591_0034.png
3.4.
Subsidiarity
Following the liberalisation of international capital flows in the 1970s and 1980s, banks
have provided an increasing amount of cross-border services. To ensure that banking
regulation remains effective, regulators have developed internationally agreed principles
and standards that large cross-border banks have to respect irrespective of their location.
Those standards are developed by the Basel Committee on Banking Supervision (BCBS).
Several EU Member States and the European Commission take part in those discussions
and the Basel standards form the backbone of the prudential requirements set out in EU
banking legislation. Following the financial crisis, the BCBS fundamentally revised the
international standards leading to the Basel III regulatory framework, which sought to
improve banks' ability to absorb shocks, improve risk management and governance; and,
strengthen transparency and disclosures. These were incorporated into EU law by means
of the CRR and the CRD IV.
The prudential requirements for institutions are accordingly already dealt with at EU
level. The legal bases are Article 114 TFEU for the CRR, BRRD and SRMR, and Article
53(1) TFEU for the CRD IV.
The BCBS has since the adoption of the CRRIV and CRD IV finalised a number of
additional standards, including a binding leverage ratio; a NSFR requirement to ensure
that banks have adequate funding structures on a long-term horizon; and following a
fundamental review, revised capital requirements for the trading book.
The objectives pursued by these measures as discussed above can be better achieved at
EU level rather than by different national initiatives. National measures aimed at e.g.
reducing bank’s leverage, strengthening bank’s stable funding and trading book capital
requirements would not be as effective in ensuring financial stability as EU rules, given
the freedom of banks to establish and provide services in other Member States and the
resulting degree of cross-border service provision, capital flows and market integration.
On the contrary, national measures could distort competition and affect capital flows.
Moreover, adopting national measures would be legally challenging, given that the CRR
already regulates banking matters, including leverage requirements (reporting), liquidity
(LCR) and trading book requirements.
The amendment of existing CRR and CRDIV legal instruments is thus considered to be
the best alternative striking the right balance between the single rules for banks and
maintaining national flexibility, such as on some macro prudential measures, for
competent authorities to address risks to financial stability
55
. Therefore the amendments
would further promote a uniform application of banking regulatory standards, the
convergence of supervisory practices and ensure a level playing field throughout the EU
banking system (see annex 6 for the indicative list of parts of legislation to be amended).
These objectives cannot be sufficiently achieved by Member States alone. This is
particularly important in the banking sector where many banks operate across the EU
single market. Full cooperation and trust within the single supervisory mechanism (SSM)
but also within the colleges of supervisors and competent authorities outside the SSM is
essential for banks to be effectively supervised on a consolidated basis. National rules
would not achieve these objectives.
55
National flexibility, such as in the field of macro-prudential policy, has not been
reviewed and is out of scope of this impact assessment.
34
 kom (2016) 0853 - Ingen titel
1715591_0035.png
4.
POLICY OPTIONS AND ANALYSIS OF IMPACTS
4.1.
On excessive reliance on short-term funding
Policy options
1. No policy change
2. A single NSFR requirement as per Basel for all banks
3. A single NSFR requirement as per Basel with some adjustments for all banks
Option 1: No policy change
The Basel III framework, implemented through the CRR and the CRD IV, already comprises
minimum capital requirements and a liquidity requirement, the LCR. As mentioned in the
problem definition, capital requirements are useful to ensure the solvency of banks but they do
not capture the liquidity and maturity of off- and on-balance sheet items. Furthermore, the LCR
takes account of the liquidity of assets, liabilities and off-balance sheet items but focuses on a 30
days horizon in stressed conditions. As such, the LCR increases the resilience of banks in case of
severe short-term liquidity stresses but does not capture the risk of excessive maturity
mismatches on a longer term horizon. As a consequence, the LCR ensures that banks will be able
to withstand a severe stress on a short-term basis but does not ensure that banks will have a
sustainable stable funding structure on a longer term horizon. Banks would then continue to be
prone to funding risks and, if short-term bank funding dries-up, they will not be able to maintain
their funding structure on a longer term horizon, which could lead to a new banking crisis.
A fast-growing body of literature
56
has developed in the past few years, which assesses for a
sample of banks considered in various countries and time periods, whether the existence of a
stable funding requirement would have significantly diminished the number of failures relative to
what happened in the absence of such a requirement. E.g. the IMF working paper “Bank Funding
Structures and Risk: Evidence from the Global Financial Crisis” finds a significant impact of the
stable funding ratio: higher levels of the stable funding ratio decrease the probability that a bank
will subsequently fail.
The most recent EBA Basel III monitoring exercise report of 13 September 2016, based on a
different sample of EU banks than the sample of the EBA NSFR report, shows that during 2015
these banks in aggregate terms have already reduced their NSFR shortfall
57
. This is likely to be a
result of supervisory monitoring, market discipline, implementation of other prudential
requirements that help improving the NSFR and anticipation of EU implementation of
international rules. However, only when imposed as hard requirements which shall be met at all
times, stable funding requirements will be effective in preventing excessive maturity mismatches
between assets and liabilities and overreliance on short-term wholesale funding. This would
advocate for the introduction of a detailed stable funding requirement at EU level.
56
See for example, International Monetary Fund (IMF) - Francisco Vazquez and Pablo Federico: Bank
Funding Structures and Risk: Evidence from the Global Financial Crisis (2012); Huang and Ratnovski
(2011); Bologna (2011), Dagher and Kazimov (2013); Haman et al.; (2013), Lallour and Mio (2015);
Hahm et al. (2011).
CRD IV – CRR / Basel III monitoring exercise – results based on data as of 31 December 2015,
EBA,
p. 40, figure 19
35
57
 kom (2016) 0853 - Ingen titel
1715591_0036.png
Option 2: A single NSFR requirement as per Basel for all banks
A complementary binding detailed NSFR would ensure that banks adequatly fund their activities
with more stable sources of funding on an ongoing structural basis. It would provide an effective
requirement of more stable longer term funding sources for banks’ obligations on a one-year
horizon in normal and stressed conditions compared to the current situation where banks have to
ensure that long term obligations are adequately met with a diversity of stable funding
instruments without it being a detailed requirement. The advantage of this option would also be
the full compliance with the Basel NSFR for all banks established in the EU.
However, this approach may unduly penalize some activities or specific business models that are
not or not adequately recognised by the Basel NSFR framework. This could lead to difficult
adjustments for some banks that could have important unintended consequences on the European
economy.
Indeed, as of end-December 2014, 30% of the banks participating in the data collection for the
EBA Report on NSFR, representing 75% of total assets in the EU, did not meet the Basel NSFR
requirements. The stable funding shortfall
58
for these non-compliant banks was estimated at 595
billion EUR, representing 3,5% of the available stable funding amount for all the banks in the
sample. As the way to comply with the NSFR requires deep restructuration of the balance sheet’s
structure, the adjustments to the shortfall could be difficult to implement for non-compliant
banks. This NSFR shortfall could mean that non-compliant banks have to find additional stable
funding (equity, medium/long term bonds/loans or retail deposits), which would typically incur
some compliance costs
59
, or to restructure their activities (the exact amount would depend on the
RSF factor applied to the banks' assets) or to undertake a combination of both.
Table 3.
NSFR shortfalls
NSFR shortfall (as of
end-December 2014):
No. of
banks
Number of
compliant
banks
NSFR
NSFR
shortfall
(bn. Euro)
NSFR
shortfall
(%
available
funding)
3.5
3.2
Total banks in the
sample
Consolidated results
(removing identified
subsidiaries of banks
included in the
sample)
279
234
196
(70%)
169
(72%)
103.6
103.6
594.7
522.7
Source:
EBA report on the NSFR,
data as of end-December 2014
Option 3: A single NSFR requirement as per Basel with some adjustments for all banks
58
The NSFR funding shortfall corresponds to the difference between weighted assets and off-balance sheet
items after application of the corresponding required stable funding - RSF - factors (denominator) and
weighted liabilities after application of the corresponding available stable funding - ASF - factor
(numerator)
59
Under current market condition replacement of 3m debt with 5y debt could results in marginal costs of
around 30bps.
36
 kom (2016) 0853 - Ingen titel
Using the Basel framework as a basis for the definition of a European NSFR would ensure a level
playing field for European banks, especially for the ones undertaking cross-border activities.
Moreover, as the Basel NSFR has been subject to an extensive observation period and public
consultation (apart from the treatment of derivative transactions) and has been thoroughly
discussed, its calibration is broadly satisfactory.
However, the necessity to take specific account of some European specificities in order to ensure
that the NSFR does not hinder the financing of the European real economy would justify
adopting some adjustments to the Basel NSFR for the definition of the European NSFR.
This mirrors the feedbacks received from the industry through the NSFR targeted public
consultation and the call for evidence. The industry indeed widely accepts the introduction of the
NSFR which is deemed as being a useful complementary supervisory measure but criticizes the
miscalibration of some specific banking activities that could have an important impact on these
activities and on the real economy.
These adjustments to the European context are recommended by the EBA NSFR report and relate
mainly to specific treatments for:
pass-through models in general and covered bonds issuance in particular, whose
funding risk can be considered as low when assets and liabilities are matched
funded;
trade finance and factoring activities, whose short-term transactions are less likely
to be rolled-over than other type of loans to non-financial counterparties;
centralised regulated savings, whose scheme of transfer renders the client deposits
(liabilities) and claims on the state-controlled fund (assets) interdependent;
residential guaranteed loans, whose specific characteristics make them similar to
mortgage loans;
credit unions, whose statutory constraints on investment of their excess of
liquidity entail a funding risk similar to that of non-financial institutions for the
institution receiving the deposits.
These proposed specific treatments reflect the preferential treatment granted to these activities in
the European LCR compared to the Basel LCR. Such treatment was widely supported by
Member States during the expert group meeting and by the industry during the NSFR targeted
consultation.
Beyond these European specificities, the EBA NSFR report does not advocate for other
adjustments to the Basel NSFR for its implementation at EU level. However, the conclusions of
the EBA NSFR report should be taken with caution, mainly because of the limitation of data
underlined in the report (data only cover a single point in time (data as of end-December 2014)
representing 75% of total assets held by credit institutions in the EU; 40% of credit institutions in
the sample are from DE and IT).
The stringent treatment of derivative transactions in the Basel NSFR could have an important
impact on banks’ derivatives activities and on the access to some operations (e.g. hedging of
currency risk, interest risk, exposure to a commodity etc.) for end-users (e.g. corporates, pension
funds, public sector entities, insurance companies, retail banks etc.).
Additional data gathered from the EBA show that the stable funding requirement linked to
derivatives transactions for banks included in the sample of the EBA NSFR report amounts to
37
 kom (2016) 0853 - Ingen titel
1715591_0038.png
more than €615bn (data as of end-December 2014), with €260 billion due only to the 20% stable
funding requirement on gross derivatives liabilities.
The disproportionate impact the NSFR could have on derivatives activities and, consequently, on
European financial markets and on the European economy is one of the main concern expressed
quite unanimously by the industry, including end-users, through the call for evidence and the
NSFR targeted consultation.. The treatment of derivative transactions and of some interlinked
transactions (e.g. clearing activities) could be unduly and disproportionately impacted by the
introduction of the NSFR without having been subject to extensive quantitative impact studies
and public consultation. The additional requirement to hold 20% of stable funding against gross
derivatives liabilities is very widely seen as a rough measure that overestimates additional
funding risks related to the potential increase of derivative liabilities over a one year horizon. The
rules underpinning the calculation of NSFR derivative assets and liabilities and in particular the
asymmetric treatment between variation/ initial margins received and posted is also cited as
detrimental to derivatives markets. According to a first impact study of the industry, the treatment
of derivatives liabilities and variation and initial margins in the NSFR could lead to an additional
funding requirement of €750 billion for the entire world-wide industry (not limited to the
European industry).
On the basis of available data and of Member States’ opinions expressed during the expert group
meeting, it seems reasonable to slightly adjust the Basel treatment of derivatives, in particular the
20% RSF factor that applies to gross derivatives liabilities, not to hinder the good functioning of
EU financial markets and the provision of risk hedging tools to credit institutions and end-users,
including corporates, to ensure their financing as an objective of the Capital Market Union.
Furthermore, regarding short term transactions with financial institutions, a Sub-Committee of
the Economic and Financial Committee on EU Sovereign Debt Markets (ESDM) raised concerns
that the asymmetric treatment of short term (less than 6 months) transactions with financial
counterparties60 may further affect the market-making ability of financial institutions on EU
sovereign debt bonds.
During the expert group meeting, some Member States also raised the issue of the potential
impact of this asymmetry on sovereign debt’s market making.
According to the EBA NSFR report, the estimated impact of this asymmetric treatment in terms
of additional required stable funding is of more than €250 billion
61
for EU banks participating in
the sample of the EBA NSFR report.
Finally, the vast majority of respondents to the call for evidence and the NSFR targeted
consultation expressed concerns on this asymmetry that could be very detrimental to market
making activities and, as a consequence, to the liquidity of repo market and of the underlying
collateral. Repo markets are presented as essential for the smooth functioning of both banks’
liquidity management and market makers' inventory management. This treatment also raises
some concerns regarding the impact on the interbank markets, in particular for liquidity
60
The treatment of short term (less than 6 months) transactions with financial counterparties is asymmetric
as the funding, including repos, received from a financial counterparty is not recognised as a source of
stable funding (0% available stable funding - ASF) while the lending, including reverse repos, granted
to a financial counterparty is subject to a stable funding requirement (10% or 15% required stable
funding - RSF - depending on the quality of the underlying collateral for secured transactions).
61
Data as of 31 December 2014 on the sample of 279 banks (representing 75% of total assets in the EU)
included in the EBA NSFR report
38
 kom (2016) 0853 - Ingen titel
1715591_0039.png
management purposes. It may then affect the liquidity of interbank markets, of the securities
(including sovereign bonds) and undermine market-making activities, thereby contradicting the
objectives of the CMU. The estimated impact of this asymmetric treatment in terms of additional
required stable funding is of €300 billion
62
in Europe according to the industry.
The ESDM also expressed concerns on the 5% RSF factor which applies to Level 1 high quality
liquid assets - HQLA - as defined in the LCR, including sovereign bonds, and that would imply
that banks would need to hold ready available long-term unsecured funding in such percentage
regardless of the time during which they expect to hold such EU sovereign debt bonds. This
could potentially further incentivise credit institutions to deposit cash at central banks rather than
to act as primary dealers and provide liquidity in sovereign bond markets.
During the expert group meeting, several Member States favoured the alignment of the RSF
applied to HQLA Level 1 for the calculation of the NSFR with the haircut applied for the LCR
(0%) to ensure consistency between the LCR and NSFR.
The banking industry also expressed its concerns regarding the 5% RSF factor applied to Level 1
HQLA through the call for evidence and the NSFR targeted consultation. This RSF factor is
deemed as being too high and not consistent with the LCR that recognizes the full liquidity of
these assets even in time of severe stress.
On the basis of available data and of Member States’ opinions expressed during the expert group
meeting, it seems reasonable to bring limited changes to the treatment of both short-term
transactions with financial institutions, and of HQLA Level 1 not to hinder the good functioning
of EU financial and repo markets.
The possible minor changes of some limited Basel provisions to take into account the
specificities of the EU economy as well as to limit disproportionate and unjustified impact on
certain activities will apply to all banks.
The analysis performed in the EBA NSFR report does not show any correlation between the size
of the bank and its compliance with the NSFR or the impact of lending to the economy and
underlines the issue of the “too many to fail” which could impact financial stability if small banks
were exempted from the NSFR requirement. Therefore, the EBA report does not recommend
introducing a different stable funding requirement for small banks but recommends applying the
same requirement to all banks on individual and consolidated basis. Answering to a call for
advice of the Commission, the EBA issued a report on the assessment of the introduction of a
possible core funding ratio for banks having a low funding risk profile in the EU. The EBA
defined the core funding ratio as followed: (retail deposits + wholesale funding>1 year + equity
instruments)/ (total liabilities + equity instruments) and used the sample of the EBA NSFR
report. They do not support the introduction of a core funding ratio for a subset of European
banks because of the weaknesses of this metric and because of the significant and costly
reporting burden for supervisors triggered by the potential implementation of two different
metrics for different banks.
The Member States through the expert group meeting and the industry through the NSFR
targeted consultation supported the analysis of the EBA report not to introduce an alternative
funding requirement for a subset of European credit institutions, in particular due to the implicit
62
European Banking Federation estimate on a sample of 65 EU banks, February 2016.
39
 kom (2016) 0853 - Ingen titel
1715591_0040.png
proportionality of the NSFR which is simple to calculate for banks having simple funding
structures.
There is hence lack of support and evidence to introduce a differentiated NSFR requirement for
small banks. Simpler reporting and disclosure requirements could however be introduced for a
subset of European banks to alleviate the administrative costs related to the implementation of the
NSFR.
Comparison of policy options
The summary of the analysis of different options to achieve the specific objective of providing a
requirement promoting funding stability and limiting over-reliance on short-term funding at a
reasonable cost is presented in the table below. This analysis leads to the conclusion that option 3
is the preferred option.
Table 4.
Comparison of policy options against effectiveness and efficiency criteria
Objectives
Policy options
S-1
Option 1: No policy change
Option 2: A single NSFR
requirement as per Basel for all
banks
Option 3: A single NSFR
requirement as per Basel with some
adjustments for all banks
EFFECTIVENESS
Specific objectives
EFFICIENCY (cost-
effectiveness)
S-2
0
-
++
S-3
0
+
++
S-4
0
0
0
S-5
0
+
++
0
+
++
0
+
++
Table 5.
Comparison of the impact of policy options on stakeholders
Stakeholders
Companies
Policy options
Banks
and
households
Option 1: No policy change
Option 2: A single NSFR requirement as per Basel for all
banks
Option 3: A single NSFR requirement as per Basel with
some adjustments for all banks
Supervisors
0
+
+
0
+
++
0
+
++
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly
positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
4.2.
On excessive leverage
Policy options
1. No policy change
2. A single leverage ratio requirement as per Basel for all institutions
3. A leverage ratio requirement differentiated for business models or adjusted for exposure types
Option 1: No policy change
40
 kom (2016) 0853 - Ingen titel
As mentioned in the problem definition risk sensitive capital requirements do not dampen the
cyclical effects of economic upswings and are good as they deliver capital requirements
proportionate to risks but they may not always capture the risk fully. Under current Union law the
risk of excessive leverage is monitored by supervisors during the supervisory review process and
institutions have to calculate report and disclose the leverage ratio. The approach so far based on
supervisory monitoring, market discipline and anticipation of EU implementation of international
rules has led to a situation where most European banks currently have a leverage ratio of more
than 3%. However, only when imposed as a hard capital requirement which must be met at all
times the leverage ratio will be effective in requiring banks to constantly manage their balance
sheet in a way that will prevent distortive de-leveraging during economic downturns. A non-
binding capital measure can simply not bring about the same prudential rigour to prevent the
building up of excessive leverage.
Option 2: A single leverage ratio requirement as per Basel for all institutions
A binding requirement would add trust in the overall financial stability of the institutions
established in the EU. A complementary binding leverage ratio requirement of 3% of Tier1
capital could provide an effective backstop compared to the current situation in Union law where
banks have to calculate, report and disclose the leverage ratio subject to supervisory review. The
advantage of this option would be to have in the Union a common measure against the building
up of excessive leverage and as a hard backstop against model risk irrespective the type of
business. This option would ensure also full compliance with the Basel leverage ratio for all
banks established in the EU is so far as the international agreed calibration would apply.
Option 3: A leverage ratio requirement differentiated for business models or adjusted for
exposure types
The one size fits all leverage ratio under option 2 has relatively more impact on banks which
have business models with overall low risk sensitive capital requirements than banks with across
the board higher risk weighted assets. In particular when banks with low risk weighted business
models are subject to legal constraints on their business models such as public development
banks' lending to the public sector, the leverage ratio may have an undesirable adverse impact on
the availability or pricing of public sector lending. The leverage ratio requirement should
therefore be adjusted by excluding from the leverage ratio exposure measure public development
loans and pass-through promotional loans provided by public development banks set up by a
Member State, central or regional government or municipality.
Moreover, export credits, which are guaranteed by sovereigns or export credit agencies receive a
considerably lower risk weight. In these instances the leverage ratio would be constraining capital
requirement leading to higher capital charges. Since export credits are important for jobs and
growth, guaranteed export credits deserve to be excluded from the leverage ratio exposure
measure.
More detailed analysis of the possible adjustment of the leverage ratio to a business model or
exposure type is presented in annex 3.14.
Based on the analysis, the options 2 and 3 score better than the baseline option. Option 3 would
be more proportional and hence be beneficial for certain types of institutions but would arguably
have reduced benefits for investors.
Table 6.
Comparison of policy options against effectiveness and efficiency criteria
41
 kom (2016) 0853 - Ingen titel
1715591_0042.png
Objectives
EFFECTIVENESS
EFFICIENCY
(cost-
effectiveness)
S-5
0
++
+
0
≈
–
Policy option
Option 1: No policy change
Option 2: A single leverage ratio
requirement as per Basel for all
institutions
Option 3: A leverage ratio
requirement
differentiated
for
business models or adjusted for
exposure types
S-1
0
––
+
S-2
0
+
++
S-3
0
+
–
S-4
0
––
+
Table 7.
Comparison of the impact of policy options on stakeholders
Stakeholder
Companies
Institutions
Institutions
and
using SA
using IRB
Policy option
households
Option 1: No policy change
Option 2: A single leverage ratio
requirement as per Basel for all
institutions
Option 3: A leverage ratio requirement
differentiated for business models or
adjusted for exposure types
Regulators/
supervisors
0
+
+
0
–
+
0
≈
+
0
–
–
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly
positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
4.3.
On inadequate calibration of risk weights on SME exposures
Policy options
1. No policy change
2. Removal of the SME Supporting Factor
3. Introducing additional capital reduction for SME exposures above €1.5 million
Option 1 – No policy change
The continuity of the current regulatory framework would imply no compliance burden.
The current framework for SME exposures would largely be coherent with EBA
findings, which demonstrated that the SME SF had been found to be consistent with
actual systematic riskiness of EU SMEs
63
, except for retail exposures of banks using
Internal Rating-Based (IRB) approach (see also annex 2.2, which provides an overview
of key conclusions from the EBA report on SMEs relevant for the SME SF).
Table 8.
Exposures:
Approach:
63
Retail (<=€1 mio)
Corporate (<=€1.5 mio)
Corporate (>€1.5 mio)
EBA report on SMEs,
p.88
42
 kom (2016) 0853 - Ingen titel
1715591_0043.png
SA
IRB
SF of 24%
SF of 24%
SF of 24%
SF of 24%
N.A.
N.A.
This option would respond to numerous calls from banks and most supervisors and
ministries to the CRR consultation
64
and the Call for Evidence for retaining the SME
Supporting Factor in the CRR. Moreover, the stability of the regulatory framework
would facilitate EBA's continuing monitoring of the use of the SME Supporting. No
policy change in the use of the SME SF might also increase confidence on the
sustainability of the measure and contribute to its more extensive use by banks.
However, the application of the SME SF to SME exposures of up to €1.5 million would
not be consistent with the findings of the EBA report stating that the threshold does not
indicate a change in the riskiness of SME exposures (see annex 2.2). This means that
SME exposures above €1.5 million would not benefit from the SME SF. Given the
negative association between the level of capital requirements and bank financing of the
economy, too high capital requirements for SME exposures beyond €1.5 million is
expected to result in a suboptimal level of bank financing of these SMEs
65
.
Option 2: Alignment with the Basel rules
This option would make the capital calibration for SME exposures internationally consistent.
BCBS is currently reflecting on reducing capital charges for some SME exposures due their
lower systematic risk. The second BCBS consultative document on the review of the
Standardised Approach (SA) of 10 December 2015
66
includes a lower risk weight (85% instead
of 100%) for all exposures of SMEs falling in the corporate exposure class (i.e. above 1 million
euros) under the Basel SA. Capital charges applicable to retail exposures under the Basel SA
would probably remain unchanged (75% risk weight), implying that SF of 24% currently applied
for retail exposures of up to € 1 million under SA and all SME exposures under IRBA would be
removed from the current CRR framework.
Table 9.
Exposures:
Approach:
SA
Retail (<=€1
mio)
N.A.
Corporate (<=€1.5 mio)
Corporate (>1.5 € mio)
IRB
N.A.
Reducing baseline risk weight from 100% to 85% to all
corporate exposures if BCBS adopts the approach
currently consulted (equivalent
to a SF of 15%)
N.A.
N.A.
Alignment with the Basel rules would imply that the average reported capital ratios of banks
would diminish by up to 0.16% points on average
67
with a significantly varying impact between
64
65
http://ec.europa.eu/finance/consultations/2015/long-term-finance/index_en.htm
Main estimate of the transitional effect, derived in from the study of May 2016 conducted by London
Economics using data for the period 1985-2014, shows that for a one percentage point increase in the
Total Capital Ratio the impact on lending flows of banks in the EU is -0.8% over one year with the
implied impact over a three-year period being -1.5%.
66
http://www.bis.org/bcbs/publ/d347.htm
67
EBA report on SMEs,
March 2016, figure 37, page 69.
43
 kom (2016) 0853 - Ingen titel
1715591_0044.png
individual banks and Member States
68
(see annex 2.2). The decrease of capital ratios would at the
same time lead to compliance costs, which on average would be marginal, but could however be
significant for individual institutions depending on their capital position. Moreover, as observed
by the EBA
69
, the increased capital requirements could lead to a reduction in lending, primarily
by the most capital-constrained banks.
Except a few think tanks, respondents to the CRR consultation overall did not support alignment
with the Basel rules. Few supervisors and ministries to CRR consultation noted that the SME
supporting factor might distort the risk-based framework for capital requirements, but invited the
Commission not to change the current calibration of risk weights before more evidence on the
effectiveness of the SME SF is obtained. EBA in its report on SMEs also underlined that it might
be too early to draw conclusions on the effectiveness of SME SF, given the limitations of the data
available and the relatively recent introduction of the SME SF
70
.
Option 3: Introducing additional capital reduction for SME exposures above
€1.5
million
This option would imply maintaining the SF for exposures in its current form as
presented in option 1 (i.e. up to €1.5 million for SA and IRB banks) and complementing
it with a discount of 15% in capital charges for loans to SMEs above €1.5 million euros.
This option reflects calls from banks, corporate buyers and SMEs to consider further
extension of the SME supporting factor to cover more SME loans.
15% capital reduction for SME exposures above €1.5 million would be consistent with
the EBA findings, which state that "the
limit of €1.5 million for the amount owed set in
the Article 501 of the CRR does not seem to be indicative of any change in riskiness for
firms".
71
At the same time, the EBA analysis suggests that the systematic risk may
increase for SME exposures above €2.5 million
72
.
Based on the EBA analysis on the riskiness of SME exposures in France and Germany
over the whole economic cycle, a 15% reduction would likely remain prudentially sound
for the EU banks. Relatively low capital requirements currently observed for the retail
asset class of IRB banks would likely be outweighed by relatively more prudent
requirements in the corporate asset class for exposures above €1.5 million
73
.
15% capital reduction for SME exposures above €1.5 million would also be aligned with the
second BCBS consultative document on the review of the Standardised Approach (SA) of 10
December 2015
74
, which proposes 15% capital discount for all SME exposures falling in the
corporate exposure class (i.e. above 1 million euros) under the SA.
As compared to the current CRR framework, this option would provide additional capital relief
for banks and thus would provide incentives for banks to increase lending to the economy as a
68
69
EBA report on SMEs,
March 2016, figure 41, p. 73
EBA report on SMEs,
March 2016, figure 23, p. 55
70
EBA report on SMEs,
March 2016, p. 11
71
EBA report on SMEs,
March 2016, p. 92
72
EBA report on SMEs,
March 2016, figure 50, p. 93
73
EBA report on SMEs,
March 2016, figures 47-48, p. 90-91
74
http://www.bis.org/bcbs/publ/d347.htm
44
 kom (2016) 0853 - Ingen titel
1715591_0045.png
whole, and SMEs particularly. The most effect is likely to be seen in the most capital-constraint
banks.
Table 10.
Exposures: Retail (<=€1 mio)
Approach:
SA
SF of 24%
Corporate (<=€1.5 mio)
SF of 24%
Corporate (>€1.5 mio)
Reducing baseline risk
weight from 100% to 85%
(equivalent to SF of 15%)
SF of 15%
IRB
SF of 24%
SF of 24%
Comparison of policy options
Table 11.
Comparison of policy options against effectiveness and efficiency criteria
EFFECTIVENESS
EFFICIENCY
(cost-
effective-ness)
Objectives
Policy option
Option 1: No policy change
Option 2: Alignment with the Basel rules
Option 3: Introducing additional capital
reduction for SME exposures above €1.5
million
S-1
S-2
S-3
S-4
S-5
0
–
+
0
–
++
0
≈
≈
0
0
0
0
+
+
0
-
+
Table 12.
Comparison of the impact of policy options on stakeholders
Stakeholder
Companies
Banks using Banks using
and
SA
IRB
Policy option
households
Option 1: No policy change
Option 2: Alignment with the Basel rules
Option 3: Introducing additional capital
reduction for SME exposures above €1.5
million
Regulators/
supervisors
0
≈
≈
0
–
+
0
––
+
0
≈
+/≈
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++
strongly positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain;
n.a. not applicable
In a view of the analysis above, the option 3 would achieve the highest cost-effectiveness
and would make most key stakeholders better-off.
4.4.
On weaknesses to the regulatory framework for loss absorption and
recapitalisation capacity
Policy options
1. No policy change
45
 kom (2016) 0853 - Ingen titel
1715591_0046.png
2. Integrate TLAC standard in MREL for G-SIIs
3. Integrate TLAC standard in MREL for G-SIIs and O-SIIs
A potential option to implement TLAC for G-SIIs in parallel to existing MREL requirements is
disregarded as it would result in duplicate regulatory frameworks, inconsistencies and an
unnecessary regulatory burden.
Option 1: No policy change
Under this option, the BRRD would continue to apply in its current form. The BRRD already
creates a framework according to which MREL is set by the resolution authority, on a case-by-
case basis for all institutions.
In terms of benefits, this option will continue to materially reduce the risk that the failure of an
EU G-SII would destabilise the broader financial system in turn lowering the probability of a
financial crisis. In addition, it will continue to ensure a significantly reduced probability of
taxpayers' support in case of such failure and the amount of such support should it still be deemed
necessary (for example, when losses would be significantly higher than implicitly assumed in the
calibration of the requirement). Indeed, as a direct consequence of this option the capacity for
loss absorption and recapitalisation of EU G-SIIs will improve and as an indirect consequence,
incentives for creditors and shareholders to scrutinise the EU G-SII's risk-taking (ex-ante, rather
than just ex post absorbing the losses) will be enhanced. Furthermore, by removing the current
implicit subsidy for EU G-SIIs provided by governments, this option could help avoid the build-
up of excessive risk and leverage within those institutions and consequently the EU banking
system as a whole. It will also remove the competitive distortions created by the implicit
guarantee.
Nevertheless, this option also has several potential drawbacks. After the adoption of the BRRD,
the FSB has developed an international standard on adequate loss absorbing capacity for G-SIIs.
EU members of the FSB have committed to implementing those standards. If the EU now
decided not to implement them it would be seen as reneging on its commitment, irrespective of
the fact that the BRRD would still continue to apply. This could potentially lead other
jurisdictions that are home to G-SIIs to decide not to implement the standards either, while the
latter however may not dispose of a bank resolution framework like BRRD to make good for this.
If so, this would lead to an outcome which is not in the EU's interest: a less safe global financial
system. Moreover, G-SIIs being active worldwide, a level playing field among them is of crucial
interest for the EU. A situation where the EU held its G-SIIs to stringent MREL requirements
under BRRD and other jurisdictions did not implement the global standard and others did not
implement TLAC would imply competitive distortions.
Furthermore, while TLAC and MREL share the same policy objective - ensuring sufficient loss
absorption and recapitalisation capacity - the features of MREL and TLAC contain important
differences. Under this option there would be no minimum ('Pillar 1') requirement for loss
absorption and recapitalisation capacity for G-SIIs. While it could be argued that resolution
authorities could use the BRRD framework to impose an MREL that would match the amount of
loss absorption and recapitalisation capacity under the TLAC framework, different Member
States and resolution authorities could choose different approaches when implementing the
46
 kom (2016) 0853 - Ingen titel
1715591_0047.png
requirement, potentially creating an un-level playing field between the different EU G-SIIs. In
addition, without transposing the TLAC requirement in EU law, there would be no legal
guarantee that all G-SIIs would be subjected to it at all times as compliance would be dependent
on discretional decisions of national resolution authorities.
For example, each Member State could implement the eligibility criteria in a different manner
which could create situations where a given type of a liability could be used to meet the
requirement in one Member State but not in another. Since the BRRD does not provide any
specific treatment for an institution's holdings of MREL-eligible liabilities, Member States could
decide not to require that they be deducted or more generally implement inadequate
75
or different
approaches on how to deal with those holdings. In the former case the issue of potential
contagion effect of a bail-in of an EU G-SII's liabilities would remain unresolved; in the latter
case a plethora of different rules would unnecessarily increase the compliance burden of
institutions. Furthermore, the BRRD currently does not prevent an EU G-SII from using CET1
capital, which it uses to meet its MREL, to meet the combined buffer requirement in the CRD IV
(i.e. it does not prevent dual use of capital). This, inter alia, impedes effectiveness of capital
buffer requirement as a policy tool. Member States may decide to address this issue differently in
their national law, leading to further divergence of rules and also to insufficient amounts of own
funds available to absorb EU G-SIIs' losses.
Not implementing TLAC could also lead to a situation where EU G-SIIs would be perceived as
riskier (for example, if the resolution authority would not exercise its discretion to require the EU
G-SII to meet its MREL requirement partially with subordinated liabilities, investors might have
difficulties in establishing the insolvency ranking of those liabilities, and hence may shy away
from purchasing them or require higher compensation to do so) by markets compared to their
third-country peers subject to a TLAC requirement and therefore increase their funding costs,
which could in turn reduce their international competitiveness.
Option 2: Integrate TLAC standard in MREL rules for EU G-SIIs
Under this option, the MREL rules would be amended to integrate the TLAC standard for EU G-
SIIs. This would mean that, compared to option 1, option 2 would include the following
additional elements:
As of 1 January 2019 a Pillar 1 MREL would be set at the higher of either 16% of the risk-
weighted assets (RWAs) or 6% of the leverage ratio exposure measure (LREM). After 1
January 2022, the requirements would be increased to 18% and 6.75%, respectively;
the Pillar 1 requirement could be met only with i) own funds and ii) eligible liabilities that
would meet eligibility criteria that would be the same for all G-SIIs (as an exception, G-SIIs
would be allowed to use non-subordinated liabilities up to an amount equivalent to 2.5% of
RWAs (3.5% after 1 January 2022) to meet the requirement);
based on the resolution strategy of each G-SII, the Pillar 1 requirement could be
complemented with a firm-specific ('Pillar 2') additional requirement and with firm-specific
guidance;
a clearly spelt hierarchy of the different types of requirements (a G-SII would need to meet
first its Pillar 1 requirement, then its Pillar 2 requirement, then the combined buffer as
defined in the CRD IV and finally the guidance);
Member States could potentially require deductions of those holdings from MREL, but could not
require deductions for own funds items. The reason is that the latter deductions are laid down in directly
applicable Union law (the CRR), which cannot be changed by national law.
47
75
 kom (2016) 0853 - Ingen titel
1715591_0048.png
a GSII's holdings of eligible liabilities issued by another G-SII would need to be deducted
from the former's MREL or own funds.
All other institutions would remain subject to the current BRRD requirement for MREL. Some of
the elements listed above would however apply also to those institutions (e.g. the common set of
eligibility criteria for the 'Pillar 2' requirement with the exception of the subordination criterion,
which would not be mandatory for MREL required under BRRD, or the alignment of the
hierarchy of the different requirements).
Compared to option 1, this option would have several additional benefits. First and foremost, it
would deliver on the EU's commitment to implement the TLAC standards into Union law. This
would reinforce the expectation on other jurisdictions to follow suit and hence help ensuring that
the TLAC standard is complied with world-wide so that also third-countries would have in place
rules ensuring that their G-SIIs could be resolved in case of failure. Second, it would promote a
level playing field amongst G-SIIs, both in the EU and internationally. Third, it would solve the
issue of potential contagion effects stemming from holdings of eligible liabilities issued by G-
SIIs. Fourth, it would provide a higher degree of legal clarity on the regulatory framework
applicable in the EU because of the presence of a single set of rules applicable to all EU G-SIIs
(e.g. common eligibility criteria, common treatment of holdings of eligible liabilities and clearly
spelt rules on interactions between different types of requirements). Finally, given the partial
“subordination requirement” of the TLAC Term Sheet
76
, investors and resolution authorities
would benefit from enhanced clarity on the ranking of instruments issued by G-SIIs in insolvency
and in resolution, which, given the complexity and size of G-SIIs, is particularly desirable in their
case.
The marginal impact of this option compared to option 1 is difficult to estimate as an important
element for the cost and benefit estimation would be to have a view on the level at which MREL
will be set under option 1. The current BRRD requires resolution authorities to set an institution's
specific MREL requirement. The exact level of these requirements as well as the decision on
whether and, if so, extent to which the MREL eligible instruments need to be subordinated are
discretionary decisions which have not yet been taken by resolution authorities. Depending on a
series of assumptions on how resolution authorities would exercise their discretion, EBA
estimated the shortfall for G-SIIs under current MREL between € 87 bn and 720 bn. The
resolution authorities' decisions on requesting subordination or not will be a key driver of the
shortfalls under the current MREL rules.
Under option 2, TLAC is introduced into the MREL framework. The MREL framework would
be complemented with a minimum on the level and quality of bail-inable liabilities for EU G-
SIIs, whilst resolution authorities can still require more. Based on the existing requirements of the
RTS on the methodology for setting an MREL
77
, EU G-SIIs are unlikely to be required to meet
an overall MREL lower than the minimum that would be introduced by TLAC. However, the
mandatory subordination for the majority of the eligible debt can still have an impact. The extent
to which there would be an impact from this mandatory subordination, depends on whether and
to what extent the resolution authorities will require mandatory subordination for their G-SIIs –
76
Total Loss-Absorbing Capacity (TLAC) Principles and Term Sheet,
FSB, 9 November 2015
COMMISSION DELEGATED REGULATION (EU) 2016/1450 of 23 May 2016 supplementing
Directive 2014/59/EU of the European Parliament and of the Council with regard to regulatory
technical standards specifying the criteria relating to the methodology for setting the minimum
requirement for own funds and eligible liabilities
48
77
 kom (2016) 0853 - Ingen titel
1715591_0049.png
they have discretion to do so - under the current BRRD. The interaction between the shortfalls
estimated under the option 1 and the 2022 calibration of the TLAC Term Sheet are:
Table 13.
Shortfall under
option 1 (current
BRRD)
Additional shortfall
under option 2 (MREL
incl. TLAC minimum)
310 bn subordinated debt
(34 bn non subordinated
debt under option 1 is no
longer needed)
In €
MREL scenario 1:
- subordination of MREL instruments not
required,
- MREL set at twice the capital requirement
(including Pillar 2 and capital buffer
requirement
78
)
MREL scenario 2:
- subordination of MREL instruments
required
- MREL set at (i) twice the capital
requirement (including Pillar 2 and capital
buffer requirement) and (ii) 8% of total
assets)
34 bn non
subordinated debt
720 bn subordinated
debt
No additional shortfall
Given the transition period until 2022, it can be expected that G-SIIs would act upon their
shortfall by replacing their current stock of non-eligible senior debt, at maturity, with eligible
subordinated debt. This would increase the funding cost for these instruments. FSB estimates on
this increase range between 30 and 50 bps. Specifically for the EU, as an upper limit, the funding
cost increase should not be higher than the spread difference observed between senior debt and
subordinated tier II debt which is between 100 and 200 bps for EU G-SIIs. As a lower bound, the
German law subordinating all senior debt (thereby achieving the required subordination), resulted
in a minor increase in spreads bellow 30 bps. While in the short term it may be possible that the
type of subordination EU G-SII's apply for their eligible liabilities could play a role in
determining the impact on its funding costs (for example, it would appear that right now senior
bonds issued by a parent holding company that would be subject to bail-in in case of structural
subordination involve a lower risk premium than subordinated bonds issued by a parent which is
an operating company although the risk premium should, in principle, be the same), in the
medium to long term this impact is expected to fade. It is important to stress that the choice of the
strategy to achieve statutory subordination (cfr. section 4.7.) could also influence the overall
impact on the funding cost. Finally, the actual impact would vary depending on the current
amount, maturity profile, corporate structure, perceived strength and type of liabilities of each EU
G-SII. The more TLAC eligible liabilities an EU G-SII would already have, the less eligible
liabilities it would have to 'create' (e.g. by issuing subordinated debt instruments) and the lower
the impact on the funding costs (and vice versa). This impact could be partially offset by a
reduction in the funding costs of senior liabilities (since there would be more loss absorbing
capacity 'sitting' below senior liabilities, in case of insolvency or resolution of the G-SII the
likelihood of senior liabilities bearing losses would be lower and hence the risk premium required
from investors to buy them would be lower). However it is unlikely that this offset would be
complete.
78
Pillar 2 is assumed to be set at 2%. The capital buffer requirements assumed are a Capital Conservation
Buffer of 2.5% and an institution specific G-SII buffer.
49
 kom (2016) 0853 - Ingen titel
1715591_0050.png
Option 3: Integrate TLAC standard in MREL for G-SIIs and O-SIIs
Under this option, the approach contained in option 2 would be extended to EU O-SIIs in order to
ensure that those comply with the same minimum requirement as G-SIIs. While the failure of an
O-SII may not have the same impact as the failure of a G-SII at global level, it can have a
significant impact on the provision of critical functions at the local level (i.e. in the Member State
where the institution was designated as a O-SII). Therefore, bail-in is likely to be part of the
resolution strategy of a O-SII and sufficient amounts of MREL should underpin the resolution
strategy.
The main drawback of this option is that applying a regime designed for G-SIIs to O-SIIs may
have disproportionate effects on the latter. One issue is related to the “subordination
requirement”. EU O-SIIs that are not subsidiaries of G-SIIs may have a more restricted or even
no access to markets when it comes to issuing subordinated debt instruments. Around 10% of
such EU O-SII currently do not report any subordinated liabilities. This may mean that they may
be unable to issue sufficient amounts of these instruments to meet the MREL in case of shortfalls
and they may therefore be forced to issue more expensive instruments (shares). Assuming that
resolution authorities would not exercise their discretion to impose the subordination of eligible
liabilities under option 1, this would mean that option 3 could lead to an increase in the funding
costs for EU O-SIIs compared to option 1. For the same reason it would also likely lead to higher
increases for EU O-SIIs than would be the case for EU G-SIIs. It could be argued that imposing
the same requirement on both may level the playing field to the extent the O-SIIs would in
practice be subject to a lower requirement than the TLAC minimum. G-SIIs and O-SIIs are often
competing in local markets. However, this level-playing field argument does not hold when the
TLAC standard would be disproportionate for a smaller O-SII, thereby weighing on its ability to
compete with both G-SIIs and non-O-SIIs.
Another issue is related to applying TLAC minimal requirements to a heterogeneous group of
institutions such as the O-SIIs. As recommended by EBA
79
, the calibration of MREL should be
closely linked to, and justified by the institution’s resolution strategy. This resolution strategy
should depend on factors such as the business model, size, interconnectedness, legal structure,
and scope and complexity of activities
80
. The differences in characteristics between EU G-SIIs
and O-SIIs have been analysed in annex 2.13. Both in terms of size compared to GDP and in
terms of business model and activities there is a large heterogeneity between the different O-SIIs
both within and across member states.
Implementing the TLAC standards' harmonised minimum requirement, to such a heterogeneous
group of institutions could overestimate the required bail-in capacity which would have to be
met, for the most part, with subordinated liabilities. TLAC is calibrated to ensure that there is
market confidence that each G-SII has a minimum amount of loss-absorbing capacity that is
available to absorb losses and recapitalise the bank in resolution. As the resolution strategy for
EU O-SIIs might vary depending on their size, business model and critical functions, it can be
envisaged that the bail-in tool would not be suitable for certain O-SII or at least part of their
activities. In cases where for example only the retail activities would need to be continued post
resolution, it could be envisaged that the remaining activities would be liquidated. This would
imply that there is no need for a recapitalisation amount for these non-critical activities. Hence
setting a minimum MREL requirement at the levels required under the TLAC standard could
overestimate the MREL required in accordance with the resolution strategy. Therefore the
79
See
Interim Report on MREL
(2016)
80
BoE on Resolution Planning
50
 kom (2016) 0853 - Ingen titel
1715591_0051.png
framework where MREL is determined as a Pillar 2 requirement, based on the loss absorption
amount, the recapitalisation amount (determined by taking into account potential divestments and
other resolution actions under the preferred resolution strategy) and the DGS adjustment, is better
suited then a Pillar 1 requirement.
The impact of this option, in terms of shortfall of eligible instruments, compared to option 1
cannot be quantified precisely as current MREL should be set by the resolution authorities on a
case-by-case basis and at this point, the relevant decisions are not known.
Comparison of policy options
The summary of the analysis of different options to achieve the specific objective to enhance
capacity for loss-absorption and recapitalisation of G-SIFIs is presented in the table below.
Table 14.
Comparison of policy options against effectiveness and efficiency criteria
EFFICIENCY
EFFECTIVENESS
Specific objectives
Objectives
S-1
Policy option
S-2
S-3
S-4
S-5
(cost-effectiveness)
Option 1: No policy change
Option 2: Integrate TLAC
standard in MREL for G-SIIs
Option 3: Integrate TLAC
standard in MREL for G-SIIs and
O-SIIs
0
n.a.
n.a.
0
n.a.
n.a.
0
++
+
0
++
++
0
+
+
0
++
≈
Table 15.
Comparison of the impact of policy options on stakeholders
Stakeholder
Banks
Policy option
Bank debt- and
shareholders
Supervisors
Companies and
households
Option 1: No policy change
Option 2: Integrate TLAC standard
in MREL for EU G-SIIs
Option 3: Integrate TLAC standard
in MREL for EU G-SIIs and O-SIIs
0
+
+/
–
0
+
+
0
+
+
0
+/
–
+/
–
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++
strongly positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain;
n.a. not applicable
4.5.
On inappropriate level of capital requirements against trading
activities
Policy options
1. No policy change
51
 kom (2016) 0853 - Ingen titel
1715591_0052.png
2. Adopt the FRTB standards for all the institutions subject to the CRR
3. Adopt the FRTB standards with some adjustments to reflect European specificities and a revised regime for small
trading book businesses
Option 1: No policy change
This option would consist in keeping unchanged the existing prudential framework for market
risk, including the derogation for small trading books.
Under the status quo, institutions would suffer no costs related to the implementation of new
rules for trading book exposures. However, the weaknesses and design flaws of the current
prudential framework for these transactions would remain unaddressed. As a result, the allocation
of capital requirements across trading book transactions may still be inadequate as compared to
the true risks faced by the institutions. On the one hand, for certain transactions in the trading
book, institutions subject to the CRR would not have a sufficient amount of capital to absorb the
potential losses that may arise from adverse changes to the market conditions for those
transactions. Those losses could be particularly significant for institutions with very concentrated
portfolios in those transactions, which could potentially require State intervention or resolution of
those institutions as a result. On the other hand, certain transactions in the trading book may be
subject to capital requirements which are too high compared to their inherent risk. This could
translate in reduced liquidity and increased transactions costs for those transactions.
Option 2: Adopt the FRTB standards for all the institutions subject to the CRR
This option would consist in replacing the current framework for market risk capital requirements
by the new BCBS standards (i.e. the FRTB standards) for all the institutions subject to the CRR.
This would include the new standardised and internal-models approaches as well as new rules for
allocating positions to the trading book (i.e the "boundary" between banking and trading books).
The flexibility for institutions to choose between the internal models and the standardised
approaches would be retained, consistently with the new rules. CRR elements which are not
included in the FRTB, such as the derogation for small trading books, would be carved out from
the CRR.
The FRTB requirements would significantly improve the design of the prudential framework for
market risks which was welcomed by both the supervisory authorities and the banking industry
when the standards were developed:
more objective rules would be defined to allocate transactions to either the trading
or banking books, therefore reducing the risks of regulatory arbitrage whereby a
trading position would be subject to inappropriate banking book capital
requirements. Stricter limits would also be implemented to move transactions
between the two regulatory books;
capital requirements would be more risk-sensitive under the FRTB standards
which means that they would be more proportionate to the true market risk faced
by the institutions. A number of technical improvements have been developed to
make the measurement of market risk more risk-sensitive. First, the standardised
approach formulas have been fundamentally revised to better reflect
diversification and hedging effects. Second, some changes would affect both
52
 kom (2016) 0853 - Ingen titel
standardised and internal-models approaches: (i) replacement of the Value-at-
Risk by the Expected Shortfall risk measure to better capture potential extreme
losses; (ii) calibration to stress conditions in order to reduce the pro-cyclicality of
the capital requirements; (iii) the introduction of variable liquidity horizons to
reflect the liquidity of the transactions;
The FRTB standards would reduce the likelihood that institutions would develop
unrealistic and deliberately lenient modelling assumptions. First, the permission
to use the revised internal-models approach would be conditional to fulfilling new
quantitative criteria that measure the performance of models (P&L attribution and
back-testing). Second, the revised standardised approach has been designed as a
real backstop to the internal-models approach that can be applied at a more
granular level ("trading desk") in case of poor internal-models performances.
Finally, third, certain risk factors (non-modellable risk factors) or asset class
(securitisation) would be restricted to specific standardised treatment for the
calculation of their capital requirements.
However, institutions subject to the CRR would also incur additional compliance costs related to
the implementation of the FRTB standards, even the ones that intend to apply the standardised
approach of the FRTB. At present, the FRTB standards do not contain any proportionality
features which raise some questions about their appropriateness for the least sophisticated
institutions or the ones with small trading activities (although the BCBS is currently considering
the inclusion of an extra, simpler standardised approach for those institutions). Therefore,
implementing the FRTB for all banks as it currently stands could be contradictory with the
principle of proportionality and would therefore not address the concerns of the industry raised
via responses to the targeted consultation paper on this topic. In fact, the industry unanimously
recommended to include a simplified standardised approach for medium-sized institutions as well
as to maintain the derogation for institutions with small trading activities
Although the design of the prudential framework for market risks has been improved with the
FRTB standards, it could have a potential detrimental impact on the functioning of the EU
financial markets via an excessive level of capital required for certain product types that could
lead to increased prices, reduced trading volumes and restricted access to capital market for
certain actors of the economy. This concern was confirmed by a majority of respondents to the
Call for Evidence on the impact of the FRTB standards on market-making activities and market
liquidity who suggested to reconsider the calibration of the final Basel standards. Not only
current market-making activities could be negatively affected by the excessive level of capital
requirements of the FRTB standards but the CMU objectives, which aim to expand of capital
market access for corporates in the EU, are also jeopardised. Once CMU is implemented, banks
will play an essential role in providing liquidity in the trading of corporate securities. According
to the industry, such market-making function could be dis-incentivised if the capital requirements
for those products are too excessive.
Finally, implementing the FRTB as it currently stands could lead to some inconsistencies with
other parts of the CRR, in particular:
The FRTB standards does not propose any beneficial treatment for the capital
requirements of STS (simple, transparent and standardised) securitisations while
the Commission proposed to extend to trading book positions the beneficial
treatment for the capital requirements of STS securitisations in the banking book
which is currently under negotiation.
53
 kom (2016) 0853 - Ingen titel
1715591_0054.png
The FRTB establishes capital requirements for sovereigns which are higher than
in the current market risk framework. This may introduce a disparity between
capital requirements for this type of securities in the trading and in the banking
book, making them significantly higher in the former.
Option 3: Adopt FRTB standards with adjustments to the calibration and to reflect
European specificities and a revised regime for small trading book businesses
This option would consist in the implementation of the FRTB standards with calibration
adjustments to ensure that EU capital markets are not excessively affected by the introduction of
the FRTB standards. It also aims to take into account certain EU specificities and ensure
consistency with other parts of the CRR (e.g. STS securitisations and sovereign exposures) and
with the objectives of CMU. Moreover, this option would allow a revised derogation for small
trading book business to account for proportionality in the new regime.
The key mechanics of the FRTB framework would be maintained but its calibration
would be modified to address the concerns about the conservativeness of the FRTB
framework in general, as expressed by Member States during the CEGBPI group meeting
on 19
th
July 2016, by the responses of many EU institutions and banking associations to
the Call for Evidence but also during a number of physical meetings scheduled but the
Commission services on this topic since the beginning of the year.
So far, two limited data analyses about the capital impacts of the FRTB have been
performed based on mid-2015 data: (i) an impact analysis
81
from the Basel Committee
for a sample of banks worldwide, including European institutions; this sample was
mostly composed of banks with large trading books and (ii) an analysis from the Global
Association of Risk Professionals (GARP), initiated by the banking industry which
comprised 21 internationally active banks, 13 of which are designated G-SIBs, and 12 of
which are European institutions.. The GARP analysis concentrated on the largest market
dealers, which also participated in the Basel Committee analysis.
The samples in both analyses are sufficiently diversified to draw some broad conclusions
about the capital impacts of the FRTB framework. However, both analyses contain a
number of caveats:
-
The Basel Committee analysis does not take into account the final adjustments that
were made to the FRTB framework before it was adopted. It is not technically
possible to correct the results of the analysis to take into account the impacts of these
adjustments, the analysis would have to be reproduced again with the recalibrated
parameters;
The GARP analysis is based on a smaller sample of banks, with a high percentage of
participating banks being large market dealers;
The banks in both samples operate in different markets and jurisdictions and it is not
possible to isolate the impacts of the FRTB for European institutions only.
-
-
The global capital impact of the FRTB framework at bank level is broadly consistent in
the two analyses: median impact at bank level of +22% in the Basel Committee analysis
and, assuming full approval of bank internal models under FRTB, +20% in the GARP
81
http://www.bis.org/bcbs/publ/d352_note.pdf
54
 kom (2016) 0853 - Ingen titel
1715591_0055.png
analysis; weighted average impact at bank level of +40% in the Basel Committee
analysis and, assuming full approval of bank internal models under FRTB, non-weighted
average impact at bank level of +50% in the GARP analysis. In light of these results, it
seems that, even though the final adjustments made to the FRTB framework - taken into
account in the GARP analysis - lowered capital requirements, a significant increase can
be expected overall.
More recent and granular estimates by the EBA shows that the increase in capital
requirements resulting from the implementation of the FRTB is more pronounced for
those banks that expect to be granted the permission to use the internal model approach
as compared to those banks that will use only the standardised approach.
Table 16.
Capital requirements impacts of the FRTB framework at bank level split per
approach used
Percentage change from Current to
Revised at Bank level
Split per Approach used
Mean
Banks using internal model
approach fully or partially
Banks using the standardised
approach only
63%
183%
25th
Percentile
3%
47%
Median
36%
170%
75th
Percentile
94%
269%
Sample
Size
26
17
Source: EBA report on SACCR and FRTB implementation, November 2016.
Beyond these analyses of the overall impact of the introduction of the FRTB standards, the Basel
Committee investigated the capital impacts at a more granular level for different asset classes.
However, the analysis is limited to the banks using internal models and has not been performed
under the standardised approach. It is hence very difficult to understand on that basis whether
certain risk categories are more impacted than others due to the introduction of the FRTB in
general, given that the most significant impact overall is observed under the Standardised
approach.
All in all, the above analyses suggest that the overall calibration of the FRTB framework
could be too conservative and possibly undermine the good functioning of financial
markets in the EU by setting an excessive level of capital requirements. The following
recalibrations could be envisaged:
(i) A general recalibration as we are concerned that the general calibration of the FRTB will
significantly increase market risk capital requirements of EU banks. An overall multiplicative
factor equal to 65% would be applied to the own fund requirements for market risks, irrespective
of the approaches used to calculate it, to broadly offset the estimated average increase. This
treatment would be in line with the expectation of the Basel committee that the remaining
measures to complete the post crisis banking reforms should not result in a significant increase in
capital requirements. It would also address concerns from both Member States and the industry
about the potential significant increase in capital requirements for market risks that could
undermine the market-making activities of European institutions and more broadly the market
liquidity of the EU financial markets.
55
 kom (2016) 0853 - Ingen titel
1715591_0056.png
(ii) No targeted recalibrations per asset class. The analyses that have been performed so
far do not give a sufficient level of understanding of the impact at the level of the various
products in scope. Moreover, the relative impact of an adjustment at asset class level
would be much more difficult to assess, which risks undermining the horizontal
consistency of the framework. Both the overall multiplicative factor and the opportunity
to adjust calibrations at asset class level would be subject to revision 3 years after the
entry into force of the new standard in the EU.
In addition to the revision of the calibration of the FRTB framework, some adjustments
to the Basel standards would be proposed in order to reflect specificities of financial
markets in the EU and to ensure consistency with the capital requirements for banking
book transactions under the CRR.
Firstly, granular data on covered bonds have been received from the industry, which
argues that the calibration of these products under the standardised approach of the FRTB
(400bps shock) is too high for the European market, as highlighted by the historical
estimates shown below for different European jurisdictions. According to these
estimates, most European covered bonds markets experienced a maximum shock
between 50bps to 150bps in the period 2008-2016. Moreover, the risk weight assigned to
covered bonds seems high in comparison with other asset classes and does not reflect the
good performance of European covered bonds.
Figure 7. Historical estimates of covered bonds volatility
56
 kom (2016) 0853 - Ingen titel
1715591_0057.png
Source: EBA report on SACCR and FRTB implementation, November 2016.
These figures would support maintaining a beneficial treatment for covered bonds as it is
currently the case for market risks capital requirements in the CRR. This market has been
historically very important in the way certain European institutions obtain lower cost of funding
in order to grant mortgage loans for housing and non-residential property and the supervisory
authorities of those jurisdictions (e.g. Denmark) warned us about the potential detrimental impact
that would have a significant increase in capital requirements for covered bonds.
Secondly, we would introduce a beneficial treatment for STS securitisation exposures
comparable to the one established for banking book positions in the Commission proposal on
STS securitisation. First, the risk of arbitrage opportunities between the trading and banking
books would be reduced if the capital requirements between the two books are more aligned
(although the new boundary will make it more difficult to move an instrument between the two
books). More importantly, the beneficial treatment would improve the secondary market liquidity
of STS securitisation by keeping less costly inventories.
Thirdly, we would adjust the capital requirements for domestic (i.e. EU) sovereigns. Since the
FRTB introduces a new standardised capital charge (CSR) applicable to sovereigns, it would no
longer be possible to hold EU sovereigns in the trading book with only a single, relatively thin
capital charge, the one for interest rate risk, as it is now the case. Furthermore, the current
underlying principle in the CRR that capital requirements for EU sovereigns are unrelated to their
ratings would no longer be preserved, since the CSR is rating-dependent. Given the fact that the
Basel is performing a comprehensive review of the treatment of sovereign exposures, it would
seem reasonable to wait for its conclusions, and apply, in the meantime, a treatment for EU
sovereigns in terms of the CSR charge that is in line with the current framework provided in the
CRR.
The beneficial treatment offered to institutions with small trading book businesses
under Article 94 of CRR would be maintained for institutions with gross market
values of trading positions
82
, excluding FX and commodity trading positions for
which the treatment under derogation has no effect, below €50 mio and for which
these positions would not exceed 5% of total assets. Based on a data collection
performed by the EBA who tested different levels for recalibrating the absolute
threshold on a sample of EU institutions with small trading activities (277
institutions with gross fair valued assets and liabilities below EUR 500 million),
the level of EUR 50 million for the absolute threshold would ensure that the
beneficial treatment for small trading book businesses under Article 94 of CRR
would apply to 85% of the sample tested.
Table 17.
Gross market value of trading assets and liabilities (excluding FX & commodities) of
banks with small and medium-sized trading book businesses
82
As defined by all the fair-values assets plus all the fair-values assets held for trading. This definition is
more prescribed than the current definition of the size of trading activities under CRR 94 which offers
some discretion for banks to calculate it and lacks overall clarity.
57
 kom (2016) 0853 - Ingen titel
1715591_0058.png
Absolute threshold based on
the size of gross trading
assets and liabilities
(excluding FX and
commodities)
< 20 m€
20 m€ < and < 50 m€
50 m€ < and < 150 m€
150 m€ < and < 300 m€
300 m€ < and < 500 m€
Total
Number of
institutions
tested
Number of institutions
with relative size of gross
trading assets
below 5% (proposed
threshold)
223
13
16
5
2
259
227
17
21
8
4
277
Source: EBA report on SACCR and FRTB implementation, November 2016
In addition, to avoid the excessive burden associated with the introduction of the
revised standardised approach for market risk, for some institutions with medium-
sized trading book but limited trading activities, as defined by institutions with
gross market values of trading positions below EUR 300 million, and for which
trading positions would never exceed 10% of total assets, the current standardised
approach for market risk would be used. This treatment would apply to all trading
positions and also to FX and commodities positions held in the banking book and
subject to own fund requirements for market risks under CRR. Based on a slightly
broader sample of EU institutions with small trading activities (997 institutions
with gross fair valued assets and liabilities below EUR 500 million), at least 46
institutions will be targeted by this measure (a large majority of the remaining
ones would be eligible for the derogation of small trading book businesses).
Table 18.
Gross market value of trading assets and liabilities (including FX & commodities) of
banks with small and medium-sized trading book businesses
Absolute threshold based on
the size of gross trading
assets and liabilities
(including FX and
commodities)
< 20 m€
20 m€ < and < 50 m€
50 m€ < and < 150 m€
150 m€ < and < 300 m€
300 m€ < and < 500 m€
Total
Number of institutions
with relative size of gross
trading assets
below 10% (proposed
threshold)
920
15
31
15
9
990
Number of
institutions
tested
922
16
33
16
10
997
Source: EBA report on SACCR and FRTB implementation, November 2016
58
 kom (2016) 0853 - Ingen titel
1715591_0059.png
This option would make the FRTB standards proportionate to banks' involvement in the trading
business. It would also address respondents' calls in the Call for Evidence for making it less
difficult to apply the standardised approach.
Table 19.
Comparison of policy options against effectiveness and efficiency criteria
EFFECTIVENESS
EFFICIENCY
(cost-effecti-veness)
Objectives
S-1
S-2
S-3
Policy option
Option 1: No
policy change
Option 2: Adopt
the FRTB
standards for all
the institutions
subject to the
CRR
Option 3: Adopt
FRTB standards
with
some
adjustments
to
reflect European
specificities and a
revised
regime
for small trading
book businesses
0
0
0
0
++
––
+
+
+
(less risk
senstitive if
we
introduce
simpler
approaches
)
++
(This could
even improve
with respect to
the baseline, if
we decide to
increase the
scope of the
derogation + a
simplified
approach)
++
(With the
implementation of
STS in the Trading
book we further
reduce arbitrage
opportunities with
respect to the
banking book).
++
(With the
implementation of
STS in the Trading
book we further
reduce arbitrage
opportunities with
respect to the
banking book).
++
Table 20.
Comparison of the impact of policy options on stakeholders
Stakeholder
Policy option
Option 1: No policy change
Option 2: Adopt the FRTB standards for
all the institutions subject to the CRR
Option 3: Adopt FRTB standards with
some adjustments to reflect European
specificities and a revised regime for
small trading book businesses
Institutions
using SA
0
–
+
Institutions
using IMA
0
–
+
Companies
and
households
0
?
?
Regulators
/
supervisor
s
0
+
++
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly
positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
In light of the above analysis, Option 3 has the best overall score. Option 3 would
introduce appropriate proportional capital requirements for market risks under the CRR
taking into account some EU specificities that are not adequately reflected in the Basel
standards.
59
 kom (2016) 0853 - Ingen titel
1715591_0060.png
4.6.
On problems on remuneration rules
Policy options for problem 1: : Deferral and pay-out in instruments
1. No policy change
2. Allow Member States or supervisory authorities to exempt some institutions and staff from the rules on deferral and
pay-out in instruments
3. Exempt small and non-complex institutions and staff with low variable remuneration in other institutions from the
rules on deferral and pay-out in instruments, based on harmonised exemption criteria defined at EU level
Policy options for Problem 2: Payment in shares
1. No policy change
2. Allow listed institutions to use share-linked instruments in addition to or instead of shares in fulfilment of the
requirement under Article 94(1)(l)(i) CRD IV
Problem 1: Deferral and pay-out in instruments
Option 1: No policy change
Under this option, the text of CRD IV would remain unchanged and leave no possibility for
waiving the rules on deferral and pay-out in instruments in the specific circumstances where this
would nevertheless be justified.
This would significantly affect the efficiency of these rules with regard to certain institutions and
staff. A full application of the rules on deferral and pay-out in instruments to small and non-
complex institutions, as well as towards staff with low, non-material levels of variable
remuneration, would mean that these institutions have to sustain important compliance costs and
burdens. Moreover, a full application of the rules to all institutions and all staff is likely to
translate into non-negligible supervisory burden for competent authorities. At the same time, the
prudential benefits of applying those requirements to staff with non-material levels of variable
remuneration are low.
In conclusion, under the current situation, the CRD IV text does not address the objective (S-2) of
increasing the degree of proportionality in the application of the deferral and pay-out in
instruments rules, which are too cumbersome and costly in case of certain categories of
institutions and staff, without significant prudential benefits.
Option 2: Allow Member States or supervisory authorities to exempt some institutions and
staff from the rules on deferral and pay-out in instruments
By exempting some institutions and staff from the application of the requirements on deferral and
pay-out in instruments, this option would introduce a degree of proportionality, thereby meeting
objective S-2. It would moreover positively influence these institutions’ competitiveness, by
reducing their cost base.
However, the possibility for Member States or supervisory authorities to set their own exemption
criteria risks leading to a situation in which there are significant divergences in the way the rules
are applied in the different Member States. Institutions of similar size and with similar activities
and staff receiving similar levels of variable remuneration would be treated differently depending
on where they are located. This would allow for regulatory arbitrage opportunities and lead to
regulatory complexity and unwarranted compliance costs, in particular for institutions operating
60
 kom (2016) 0853 - Ingen titel
1715591_0061.png
cross-border. This would also be at odds with the broader objectives of the European single
rulebook, which is to a set of truly unified and directly applicable rules for all banks operating in
the EU
83
and, with respect to institutions supervised by the SSM, could affect the SSM's ability to
supervise banks efficiently and from a truly single perspective.
Given this concern, the overall benefits of Option 2 would not outweigh its costs, and thus the
efficiency of this option is assessed as negative.
Option 3: Exempt small and non-complex institutions and staff with low variable
remuneration from the rules on deferral and pay-out in instruments, based on harmonised
exemption criteria defined at EU level
By exempting small and non-complex institutions and staff with low variable remuneration, a
much needed and appropriate degree of proportionality in the application of the rules on deferral
and pay-out in instruments would be introduced, thereby meeting the objective S-2. This would
be without an impact on financial stability, as all the prudentially-relevant institutions will
continue to be captured by the rules.
84
Under this Option, there would be notable savings for institutions on the costs related to the full
application of the requirements on deferral and pay-out in instruments for all identified staff.
Based on EBA’s current estimates of on-going compliance costs, cost savings for “small
institutions” could be in the range of €50 000 to €200 000 yearly. In the case of other institutions,
the cost savings from exempting staff with low variable remuneration are more difficult to
estimate. Currently, according to EBA estimates, ongoing costs from the application to all staff in
the case of large institutions range from €400 000 to €1.5 million.
The benefits of cost savings and reduced burden would be secured without the risk of other
unintended consequences that can be associated with Option 2.
Indeed, institutions of similar size and with similar activities and staff receiving similar levels of
variable remuneration will in principle be subject to or exempted from the rules on deferral and
pay-out in instruments in the same way independently of where they are located. The harmonised
exemption criteria would reduce regulatory complexity and avoid unwarranted compliance costs,
in particular for institutions operating cross-border activities. They would promote further
integration in the EU market and contribute to the elimination of regulatory arbitrage
opportunities.
Therefore, the efficiency of Option 3 is assessed as positive.
83
The term Single Rulebook was coined in 2009 by the European Council in order to refer to the aim of a
unified regulatory framework for the EU financial sector that would complete the single market in
financial services (see
European Council conclusions, June 2009)..
The key objectives of the Single
Rulebook are to eliminate legislative differences among Member States; ensure the same level of
protection for consumers, and ensure a level playing field for banks across the EU.
84
For example, estimates show that a threshold of EUR 5 billion in total asset value for "small" institutions
would imply the exemption of institutions accounting for around 7% of the EU market size in terms of
total assets. In order to further ensure that in all the individual Member States all prudentially relevant
institutions are covered, it can be considered to combine the EU harmonised exemption criteria with a
possibility for supervisory authorities to adopt a stricter approach where they consider this prudentially
relevant.
61
 kom (2016) 0853 - Ingen titel
1715591_0062.png
Options 2 and 3
reflect the views expressed by the majority of stakeholders, including Member
States, supervisors and industry.
Option 3
is in line with the proposal put forward by EBA in its
Opinion on proportionality.
Table 21.
Comparison of policy options for
Problem 1
against effectiveness and efficiency
criteria
EFFECTIVENESS
Objectives
Policy option
Option 1: No policy change
Option 2: Allow Member States or supervisory authorities to
exempt some institutions and staff from the rules on deferral and
pay-out in instruments
Option 3: Exempt small and non-complex institutions and staff
with low variable remuneration from the rules on deferral and
pay-out in instruments, based on harmonised exemption criteria
defined at EU level
Objective S-2
0
+
EFFICIENCY
(cost-effectiveness)
0
–
+
+
Table 22.
Comparison of the impact of policy options on stakeholders for Problem 1
Stakeholders/
Options
Option 1: No policy change
Option 2: Allow Member States or supervisory
authorities to exempt some institutions and staff
from the rules on deferral and pay-out in
instruments
Option 3: Exempt small and non-complex
institutions and staff with low variable remuneration
from the rules on deferral and pay-out in
instruments, based on harmonised exemption
criteria defined at EU level
Regulators /
supervisory
authorities
0
Institutions /
Shareholders
Employees
Tax-payers/
consumers
0
0
0
+
+
+
?
+
++
+
≈
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly positive; +
positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
Problem 2: Payment in shares
Option 1: No policy change
Under this option, the text of CRD IV would remain unchanged and leave no possibility
for listed institutions to use share-linked instruments instead of or in addition to shares in
fulfilment of the requirement under Article 94(1)(l)(i) CRD IV. In the case of exclusive
use of shares, these institutions would have to sustain important compliance difficulties,
while the prudential benefit would not be any higher than in case of the use of share-
linked instruments.
Option 2: Allow listed institutions to use share-linked instruments in addition or
instead of shares in fulfilment of the requirement under Article 94(1)(l)(i) CRD IV
62
 kom (2016) 0853 - Ingen titel
1715591_0063.png
Under Option 2, listed institutions would no longer need to face the unnecessary
difficulties and burdens associated with the requirement to pay out part of the variable
remuneration of identified staff in shares. Institutions could create share-linked
instruments, without the need to purchase or create shares. As opposed to shares, share-
linked instruments moreover do not bring problems of insider dealing.
At the same time, if it is ensured that they closely track the value of the underlying shares, that
they have the same effect in terms of loss absorbency as shares and that they are presented in a
transparent way to staff, share-linked instruments can be equally successful in achieving the
prudential objectives of payment in shares (to limit the portion of variable remuneration paid in
cash, to align the interests of the staff with that of shareholders, and to align the level of variable
remuneration with the risk profile and long-term interests of the institution). In order to achieve
this equivalence of effectiveness with shares, share-linked instruments should be designed in
such a way that they closely track the value of the underlying shares, have the same effect in
terms of loss absorbency as shares and be presented in a transparent way to staff.
As Option 2 would allow achieving the same prudential outcome in a less burdensome
way, its efficiency is assessed as strongly positive.
Table 23.
Comparison of policy options for
Problem 2
against effectiveness and efficiency
criteria
EFFECTIVENE
SS
Objectives
Policy option
Option 1: No policy change
Option 2: Allow listed institutions to use share-
linked instruments in addition or instead of shares
in fulfilment of the requirement under Article
94(1)(l)(i) CRD IV
0
+
0
++
Objective S-2
EFFICIENCY
(cost-effectiveness)
Table 24.
Comparison of the impact of policy options on stakeholders for
Problem 2
Stakeholders
/
Options
Option 1: No policy change
Option 2: Allow listed institutions to use share-
linked instruments in addition or instead of shares in
fulfilment of the requirement under Article
94(1)(l)(i) CRD IV
0
0
0
0
Regulators /
supervisory
authorities
Institutions /
Shareholders
Employees
Tax-payers/
consumers
≈
++
+
≈
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly positive; +
positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
63
 kom (2016) 0853 - Ingen titel
4.7.
On problems on insolvency ranking
Option 1: No policy change
Under this option, MS would continue to have divergent approaches to the subordination of
creditor claims, creating an uneven playing field and increasing uncertainty for investors and
issuers alike, potentially rendering the cross-border application of the bail-in tool more difficult.
Option 2: Partially harmonise insolvency ranking for unsecured debt
Option 2a: Statutory subordination of all unsecured debt, retroactive application
Based on market observations made in markets which implemented such an approach,
the marginal increase in the cost of funding for subordinated debt is estimated to be
between 2 – 30 bps with a higher likelihood towards the lower range of that interval. As
indicated by industry stakeholders who contributed to our impact assessment exercise, it
is very difficult to accurately isolate the impact of a retroactive subordination of all
unsecured debt from other market developments (e.g. rating downgrade, Brexit, other
developments).
An immediate effect of applying Option 2a would be the immediate compliance with
TLAC (assuming GSIBs held sufficient senior unsecured liabilities that would become
subject to statutory subordination) at no additional funding cost because outstanding
contracts would continue under the previous issuing pricing conditions.
The short-term effect of a retroactive subordination approach would be a gradual increase
in the cost of funding assuming some subordinated instruments reach maturity in the
short-term and need to be rolled over as subordinated debt at an additional cost of
maximum 30bps.
In the medium-long-term banks would experience a significant increase in the cost of
funding because more debt gradually matures and must be rolled over as subordinated, at
the extra cost of funding (up to 30bps), irrespective of the TLAC needs.
Another very significant effect of this option is that there will no longer be a senior
unsecured debt category in practice; therefore banks would no longer be able to fund
themselves by issuing senior unsecured debt. The impact would extend also to investors,
especially mandated investors who cannot invest in subordinated instruments.
Option 2b: Creation of a non-preferred senior debt category
Since no jurisdiction implemented already such an approach and no issuance has been
done yet into the new senior "non-preferred" class it is not possible to base oneself on
market observations to estimate the impact on the cost of funding.
The impact on cost of funding has been estimated by making reference to the issuance of
similar Tier 3 instruments by an EU bank. The extrapolated effect is set between 20 – 50
bps with a higher likelihood towards the upper range of that interval (50bps). This impact
should be treated with caution since it is extrapolated based on a single issuance by a
single bank, which means it could be biased by market conditions in that particular
Member State. Another potential proxy for this estimate is the spread differential
between debt issued by a holding company compared to an operating company (HoldCo
vs Opco spread). Our limited data (comparison of HoldCo vs OpCo issuances for one
64
 kom (2016) 0853 - Ingen titel
1715591_0065.png
bank) shows an increase of 30 – 150bsp depending greatly on currencies and maturities.
Given the limited sample and the large variation, this method may not be representative
enough for this assessment.
The immediate and short-term effect of this option would be a sharp increase in the
marginal cost of funding (50 bps or more) for the senior "non-preferred" issuance aimed
to close the TLAC gap. The total cost of funding would depend on the size of the TLAC
shortfall but also currencies, maturities and general market conditions at that point in
time. Banks have between 2017 and 2019 to build the TLAC buffer, assuming January
2019 as the start of the TLAC compliance period and the hike in the cost of funding
under this approach is expected to be more pronounced in the first part of that time
interval as several banks might issue non-preferred instruments at similar times.
After 2019 and beyond it is expected that, once banks have satisfied the TLAC levels
with the non-preferred senior class, the marginal cost of issuing such instruments would
gradually decrease, as banks move to "cruise" mode and issue such instruments only to
replace the stock as it comes to maturity.
Option 2c: Statutory preferred status for all deposits vis-à-vis senior debt
This approach separates senior liabilities only through preference of deposits. Uninsured
deposits are preferred and rank higher to senior bonds, which effectively minimises (but
does not eliminate) the risk of them bearing losses in resolution or insolvency.
Other senior debt, including net uncollateralised derivative liabilities and structured notes
continue to rank pari passu with unsecured senior debt.
Table 25.
Comparison of policy options against effectiveness and efficiency criteria
EFFECTIVENESS
COST-EFFICIENCY
Objectives
Policy option
Legal
clarity
Availability
of TLAC
eligible/bail-
inable debt
– short-term
0
++
0
65
Availability
of TLAC
eligible/bail-
inable debt
– long-term
0
++
++
Short-term
Long-term
Option 1: No policy
change
Option 2(a):
Option 2(b):
0
++
++
0
+
–
0
––
++
 kom (2016) 0853 - Ingen titel
1715591_0066.png
Option 2(c)
+
0
0
0
0
Table 26.
Comparison of the impact of policy options on stakeholders
Stakeholder
Policy option
Institutions
with larger
TLAC
shortfall
0
++
–
0
Institutions
with smaller
TLAC
shortfall
0
+
–
0
Investors
Resolution
authorities
Option 1: No policy change
Option 2(a):
Option 2(b):
Option 2(c)
0
–
–
+
0
0
++
++
0
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly
positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
Option 1 is the least preferred option. It does neither have a positive impact on the
effectiveness of the bail-in tool, nor does it help banks to meet their TLAC target. On the
contrary, maintaining a heterogeneous framework in insolvency ranking may impede
resolution authorities’ ability to apply the bail-in tool, may decrease investor confidence
and create distortions of competition. A significant number of Member States and
industry stakeholders endorse an EU partially harmonised approach to the subordination
of unsecured debt because this would facilitate the resolution of cross-border institutions
while providing for more clarity for both banks and investors as well as a level playing
field in the EU debt markets. Options 2(a) and 2(b) have both advantages and
disadvantages when assessing the short and long-term effects.
Option 2(a) has the strong advantage of addressing TLAC shortfalls with immediate
effect and it may be advantageous in the immediate and very short term as it
accomplishes TLAC compliance at low additional cost of funding. This is because the
retroactive subordination of the outstanding stock of debt changes the order of preference
of ongoing contracts concluded at what was previously, a senior issuing price. A
significant disadvantage for the medium and long-term is the fact that banks would be
prohibited to issue senior unsecured debt for funding reasons other than TLAC
compliance, and would need to roll-over the stock of subordinated debt at an increased
cost of funding irrespective of how much they actually need to meet TLAC. In
conclusion, this option is rigid and a "one-size fits all" type of option, with very
significant effects on the debt market in the long run (e.g. the absence of senior
unsecured market, the crowding out of mandated investors who cannot buy subordinated
instruments).
However, Option 2(b) will enable banks to define and maintain the optimal funding mix
according to their business and funding model. This option would allow banks to issue
TLAC eligible senior non-preferred instruments up to the level of the TLAC shortfall.
Although potentially costly in the short-run when several banks may issue senior non-
preferred to close their TLAC shortfalls, the cost increase is deemed to dampen in the
long-run when banks would have filled-in their TLAC buffers and only need to issue in
this category to roll-over TLAC debt as its maturity period is below 1 year. The investor
base for senior non-preferred debt, which would be bail-inable only in resolution, is
66
 kom (2016) 0853 - Ingen titel
1715591_0067.png
expected to be similar to that for senior unsecured debt rather than that for subordinated
debt. This is because subordinated debt bears higher risks, being potentially subject to
write-down or conversion to equity also outside of resolution.
In conclusion, Option 2(b) would provide sufficient flexibility to take account of different bank
business models across the EU and reduce over time the impact on bank funding costs. It would
avoid the crowding out of investors with a mandate outside of subordinated instruments and
allow for appropriate calibration to ensure a level playing field in the market. This direction has
been endorsed by Member States and industry representatives in the expert group discussions as
it fulfils the acceptance criteria set forth namely: flexibility, taking account of banking business
and funding models, the possibility to adjust so that level playing field is ensured and confined
impact on funding costs.
4.8.
On lack of effectiveness of the current rules on moratorium
Policy options
1. No policy change
2. Further harmonise moratorium tools in the EU
Option 1: No policy change
This option would leave the current situation unchanged.
Option 2: Further harmonise moratorium tools in the EU
A further harmonisation of moratorium tools would require amendments to existing legislations,
and particularly CRR/CRD IV and/or BRRD. A further harmonised moratorium tool at EU level
could help smoothening the resolution process and guarantee that resolutions and supervisory
authorities can freeze the bank's liquidity for a short period of time to assess whether the bank
should be subject to early intervention measures, or should be declared failing or likely to fail, or
to more precisely quantify its assets and liabilities in the context of the valuation process or to
choose a certain resolution tool.
The objective of a further harmonisation of moratorium tools is to provide banks with a flexible
and effective instrument to prevent undesirable effects on liquidity in a supervisory or resolution
scenario.
At the same time it is of utmost importance to ensure that the execution of moratorium tools does
not affect important safeguards (such as the rights of depositors) and, apart from this, is carried
out in a proportionate manner. Also, it is important to ensure that the approach of
supervisors/resolution authorities is as harmonised as possible to facilitate the smooth
management of the resolution process in case of cross-border institutions.
The effectiveness of such a tool may vary substantially in intensity depending on how the
moratorium tool is structured. The main features of the moratorium tool currently being assessed
are:
-
specific and clear conditions of application. The moratorium could be used only if
necessary for specific purposes, namely:
67
 kom (2016) 0853 - Ingen titel
1715591_0068.png
-
o
decide on the existence of the conditions for early intervention measures
o
decide on the determination whether the bank is failing or likely to fail
o
assess the exact amount of assets and liabilities of the bank
o
decide on the application of a specific moratorium tool
short duration (minimum period needed for the purpose above and anyway not
longer than 5 days)
These features may effectively address the key concerns encountered by the Commission when
consulting stakeholders in the course of expert discussion on the topic.
In particular, the clear conditions of application would be beneficial to ensure clarity and avoid
rash reactions from creditors' of the bank, which may in turn limit the risk of liquidity outflows.
Also, the fact that such conditions are directly linked to specific steps in the pre-resolution and
resolution procedure should further contribute to ensure certainty by avoiding excessive
discretion for the supervisory/resolution authority.
Moreover, the short duration appears as a key factor in avoiding bank runs. The short duration
would be instrumental to the specific objective pursued by the regulator and would limit the
impact on creditors. At the same time, the proposed provision should provide clarity on the
maximum duration of the suspension, thereby limiting the possibility of diverging interpretations.
Also, it seems important to clarify the possibility to apply a moratorium tool in a pre-resolution
(early intervention) scenario. It seems important for supervisors/resolution authorities to be
provided with the possibility to use a moratorium tool where possible to reduce the impact of a
resolution or avoid it altogether. This however, as explained above, requires a consistent
approach in case of cross-border resolution and should keep into account the safeguards in favour
of creditors. In this light, the conditions of applications of the tool in an early intervention
scenario should be clear and specific to avoid diverging interpretations. Also, the suspension
should have a short duration to preserve creditors' prerogatives.
It is worth mentioning that transactions which cannot be suspended – such as those involving
CCPs – as well as covered deposits would be out of scope.
Table 27.
Comparison of policy options against effectiveness and efficiency criteria
Objectives
Policy option
Option 1: No policy change
Option 2: further harmonisation
Legal clarity
Level playing field
0
+
0
+
Table 28.
Comparison of the impact of policy options on stakeholders
Stakeholder
Institutions
Policy option
Option 1: No policy change
Supervisory
authorities
0
68
Resolution
authorities
0
Depositors
0
0
 kom (2016) 0853 - Ingen titel
1715591_0069.png
Option 2: further harmonisation
≈
+
+
?
Magnitude of impact as compared with the baseline scenario (the baseline is indicated as 0): ++ strongly
positive; + positive; – – strongly negative; – negative; ≈ marginal/neutral; ? uncertain; n.a. not applicable
4.9.
On insufficient proportionality of the current rules
Policy options
1. No policy change
2. Measures to reduce administrative burden and legal complexity for smaller credit institutions
3. Exemption of very small credit institutions from CRR/CRD IV
Option 1: No policy change
Under this option rules would remain unchanged. The CRR and CRD IV would simply continue
to impose a general duty on Member States and European authorities to apply the rules in a
proportionate manner (see Recital 46 of the CRR). Specifically, reporting would continue to be
subject to the general requirement that it must be "proportionate to the nature, scale and
complexity of the activities of the institutions" (see Art. 99(5) of the CRR). This formulation has
proven to be excessively high-level and, as a result, insufficient to deliver a meaningful
differentiation in the reporting required from smaller institutions.
With regard to disclosure, the requirements set out in the CRR would continue to apply to all
institutions without any distinction with regard to their size and complexity.
Option 2: Measures to reduce administrative burden and legal complexity for smaller
credit institutions
To address concerns related to excessive administrative burden, option 2 would set out a specific
reporting and disclosure framework for smaller institutions with reduced frequency and content.
More precisely, smaller credit institutions would be required to provide to the supervisors less
granular information by eliminating those reporting obligations that are not relevant for
supervisory purposes. Smaller credit institutions would also be subject to less frequent reporting
obligations (e.g. quarterly instead of monthly, biannually instead of quarterly). To this end, the
CRR and the CRD IV would be amended to give a mandate to the EBA to develop ad hoc
reporting for smaller credit institutions (see annex 3.3 for details). Furthermore, the CRR would
be amended to provide for differentiated disclosure requirements for small credit institutions. In
particular, similarly to reporting obligations, the number of information to be disclosed would be
reduced for smaller credit institutions, as well as the frequency of disclosure obligations (see
annex 3.2 for details).
Finally, to address the difficulties for smaller institutions resulting from the volume and
complexity of the current framework, the EBA would be mandated to develop an IT tool
to guide them through the rules which are relevant to their size and business model. This
IT tool could help smaller institutions to gain a better of understanding of the rules and
reduce compliance costs.
Option 3: Exemption of very small credit institutions from CRR/CRD IV
69
 kom (2016) 0853 - Ingen titel
1715591_0070.png
As an alternative to ad-hoc reporting and disclosure requirements, small credit institutions would
be exempted from the CRR and CRD IV. The result of this would be to devolve to Member
States full responsibility to regulate these firms from a prudential perspective.
However, exempting very small credit institutions would result in a fragmentation of the Single
Market undermining the level playing field. In fact, similar smaller credit institutions would be
treated differently depending on the Member State where they operate. Moreover, exempted
credit institutions would still being able to compete in the EU internal market with credit
institutions to which CRD IV/CRR are applicable in full. Eempting very small credit institutions
would also increase risks to financial stability since it wouldn't provide any EU instrument to
address the simultaneous failure of different small credit institutions, which may hamper the
stability of the whole EU banking sector. During discussions with Member States
85
they shared
this view and were also concerned that exempted credit institutions could be deprived from
access to the EU Deposit Insurance Scheme (i.e. EDIS)
86
, currently under discussion, and that,
once the EDIS was implemented, an exemption from the CRR/CRD IV would contradict the
stated objective of the EDIS proposal, in particular its risk-sharing premise. Moreover, such an
exemption was regarded as potentially hindering efforts in some Member States to restructure
their banking system through consolidation as a means to improve the solvency of weaker credit
institutions.
Whilst calling for more proportionality in the prudential framework, stakeholders replying to the
call for evidence did not advocate for very small credit institutions to be exempted from
CRR/CRD IV.
Comparison of policy options
The combination of tailored prudential requirements as described under each relevant section
(e.g. NSFR, leverage ratio etc.), specific reporting and disclosure requirements for smaller
institutions and the introduction of an IT tool is assessed to be sufficient to deliver an appropriate
balance between proportionality and consistency of prudential requirements for all institutions.
Smaller institutions would benefit from prudential requirements and limited exceptions tailored
to their business model, size, complexity and relative risk of the activities they undertake, but
would be subject to the same baseline prudential standards as their larger counterparts. This
would introduce in the EU banking system a degree of flexibility that would allow a significant
number of credit institutions to benefit from proportionality measures. As a matter of fact, the
high polarisation of the EU banking sector (i.e. there is a high difference between smaller and
bigger credit institutions in terms of assets) and the differences across Member States in the
composition of market operators (i.e. size and business model of credit institutions) would not
allow for setting a 'one size fits all' type of regulatory framework for smaller institutions. For this
reason, the possibility of exempting small credit institutions from the application from the CRD
IV and CRR appears less effective. A plain exclusion of certain banks would not take into
account differences in the composition of the EU banking sector in different Member States (no
risk-sensitive) and would not improve legal certainty since 28 regulatory regimes would be
applicable to credit institutions excluded from the CRD IV and CRR. This would also result in an
unlevelled playing field both among credit institutions exempted and between the latter and credit
institutions to which the CRD IV and CRR apply.
85
86
This option was discussed with Member States during the Expert Group on Banking Payment and
Insurance (EGBPI) meetings in June and July.
For more information on the EDIS see
http://ec.europa.eu/finance/general-policy/banking-
union/european-deposit-insurance-scheme/index_en.htm.
70
 kom (2016) 0853 - Ingen titel
1715591_0071.png
Other than the necessary measures to transpose the new provisions amending the CRD IV,
Member States would not be required to put in place new specific administrative procedures.
The summary of the analysis of different options to achieve the specific objective of providing a
more proportionate prudential framework is presented in the table below.
Table 29.
Comparison of policy options against effectiveness and efficiency criteria
EFFECTIVENESS
Objectives
Policy option
Option 1: No policy change
Option 2: ad hoc reporting and
disclosure requirements for small
institutions and IT tool (plus tailored
prudential requirements and limited
exemptions)
Option 3: exempt very small credit
institutions from CRR/CRD IV
EFFICIENCY
(cost-effectiveness)
Specific objectives
S-1
0
S-2
0
S-3
0
S-4
0
S-5
0
0
++
++
++
0
++
++
-
+
--
0
-
+
Table 30.
Comparison of the impact of policy options on stakeholders
Stakeholder
Policy option
Option 1: No policy change
Option 2: ad hoc reporting and disclosure requirements
for small institutions and IT tool (plus tailored prudential
requirements and limited exemptions)
Option 3: exempt very small credit institutions from
CRR/CRD IV
Small credit
institutions
Other credit
institutions
Supervisors
0
++
+
0
++
-
0
++
-
Based on the above analysis, option 2 scores better than the baseline option and option 3. Option
2 would, therefore, be more appropriate and beneficial for both financial stability protection and
economic growth promotion perspectives.
4.10. The choice of the instrument
The policy options retained in the sections above could be implemented by amending the CRR
and the CRD IV. The proposed measures indeed refer to or develop further already existing
provisions inbuilt in those legal instruments (liquidity, leverage, remuneration, proportionality). It
is therefore suggested that these measures be put forward as an amendment to the existing legal
instruments. The indicated list of proposed amendments is presented in annex 6.
As regards the new FSB agreed standard on total loss absorbance capacity it is suggested to
incorporate the bulk of the standard into the CRR, as only a regulation can achieve the necessary
uniform application, much in the same way as the pillar 1 primary capital requirements. Shaping
prudential requirements in the form of an amendment to the CRR would ensure that those
requirements will in fact be directly applicable to them. This would prevent Member States from
71
 kom (2016) 0853 - Ingen titel
1715591_0072.png
implementing diverging national requirements in an area where full harmonization is desirable in
order to prevent an un-level playing field. Minor fine-tuning of the current legal provisions
within the BRRD will however be necessary to make sure that TLAC and MREL requirements
are fully coherent and consistent with each other.
5.
T
HE CUMULATIVE IMPACTS OF THE ENTIRE PACKAGE
5.1.
Introduction
This section discusses the cumulative impact of additional capital and liquidity requirements for
the EU banking industry, in terms of their costs and benefits, compared to the baseline scenario.
The baseline scenario represents the cumulative impact in the absence of measures under
consideration but including the CRR and CRD IV measures currently in force. Specifically, this
cumulative impact consists of two parts:
(1) a quantitative assessment of benefits and costs related to the FRTB and the
leverage ratio, finding a modest reduction in expected losses in the banking
system and associated potential burden on public finances at a very small overall
costs to the economy;
(2) a qualitative assessment based on available empirical evidence of benefits and
costs related to the NSFR, which shows that banks disposing of a higher level of
stable funding tend to show a smaller decrease in lending to the real economy
during the financial crisis and that introducing the NSFR would not have a
significant impact on the supply of credit to the economy.
As indicated by Dewtripont and Hancock et al. (2016), capital and liquidity requirements have
different direct impacts on banks' balance sheets which often interact. The reaction of individual
banks can have an impact on aggregate economic activity - both positive (benefits) and negative
(costs). Literature suggests
87
, on the benefits side, that higher capital and liquidity ratios improve
resilience to shocks of both individual banks, and the financial system. Improved resilience, in
turn, lowers both the probability of a financial crisis and reduces the size of economic losses in
the event that a crisis occurs. The benefits, in this sense, are the expected losses that are avoided.
On the costs side, higher capital and liquidity requirements may increase bank funding costs
which could be passed on to end-users (i.e. banks could react by reducing the volume or
increasing the price of lending to households and non-financial firms). Changes to liquidity
requirements could impact interbank lending and maturity transformation, which also has an
impact on aggregate borrowing. Lower borrowing reduces aggregate consumption and
investment and, eventually, gross domestic product (GDP). Some authors
88
also point to the
effects of some of these measures on market liquidity in securities markets although the effects
are difficult to disentangle from other factors influencing market liquidity.
Overall, the net benefits of regulation can be thought of as the expected loss that is avoided in the
event that a crisis occurs (the benefit), which is offset by the opportunity cost of reduced
economic activity during non-crisis periods.
This analysis focusses on the impact from additional capital requirements related to the leverage
ratio requirements, market risk requirements (FRTB) and the NSFR, as calibrated by the Basel
Committee. This implies that the actual impact of the preferred options, which leads to a
87
For example: Miles et al (2013), de-Ramon et al (2012), BCBS (2010), de Bandt (2015), Brooke et al.
(2015), Elliot et al. (2016)
88
For example: Elliot et al. (2016)
72
 kom (2016) 0853 - Ingen titel
calibration better reflecting EU specificities, should provide the same benefits at lower costs. In
general, it needs to be stressed that, given the inherent complexity and special nature of banking
and given that many benefits and costs are dynamic in nature (often related to unobservable
incentives), there are limitations to the reliability and precision with which quantitative models
can comprehensively estimate the social benefits and costs. Nevertheless, these models are useful
to better understand the transmission mechanisms and the results generally give a good estimate
of the direction and order of magnitude of expected impacts.
5.2.
Quantitative assessment of benefits and costs related to FRTB and the
LR
In order to support the qualitative assessment and comparison of reform options carried out
above, the Commission services have attempted to quantify some of the costs and benefits that
could result from the proposals on the leverage ratio and on FRTB.
Benefits (further details in annex 5.1)
Benefits are measured as a decrease in the potential costs for society due to bank defaults and
recapitalisation needs. The analysis estimates the losses in excess of bank capital, as well as the
recapitalization needs of banks to allow them to continue operating on an on-going basis. The
effect of the various tools available in the various regulatory tools (i.e. bail-ins and resolutions
funds) aimed at mitigating the leftover losses and recapitalisation needs, is also taken into
account.
Banking losses are simulated using the SYMBOL model (Systemic Model of Banking Originated
Losses). SYMBOL simulates losses for individual banks using information from their balance
sheet data. The model also allows taking into account the safety-net that is available to absorb the
simulated shocks (capital, bail-in, resolution funds). The initial simulation output is the full
distribution of bank losses. These initial individual bank losses are then transformed into losses in
excess of capital and recapitalization needs, to be covered by the safety net, and the residual is
finally aggregated at EU level.
The simulations consider the case of a systemic crisis event, similar in severity to the one started
in 2008, and the conservative assumption is used that all simulated bank excess losses and
recapitalization needs that the safety net cannot cover would eventually fall on public finances.
The exercise uses post-2014 data for a sample of 183 banks covering 83% of the EU total assets.
A crisis comparable to the last global one is approximately placed on percentile 99.95 when
considering excess losses and recapitalization needs based on pre-crisis data.
As indicated above, the analysis ignores excess capital buffers that many banks currently hold,
partly in anticipation of future capital requirements; the analysis assumes that all banks hold just
enough capital to cover their 10.5% RWA minimum capital requirement (MCR), both before and
after the reforms. In reality there are banks which already hold an actual capital commensurate
with the new rules MCR or even above. For these banks the associated costs- and benefits would
not arise. However, considering currently existing additional capital buffers in the baseline may
lead to an underestimation of the benefits, since it is not certain that banks currently holding a
buffer will maintain it. Moreover, to the extent that the analysis focuses on the adjustment to the
new rules, looking at actual buffers may ignore some of the adjustment that has already taken
place. For that reason we use a scenario where banks start from the minimum of their current
capital level and the MCR incl. a limited buffer.
73
 kom (2016) 0853 - Ingen titel
1715591_0074.png
The baseline scenario represents the case where banks' capital equals 10.5% risk based capital
requirements (excl. policy measures). The policy scenario represents the case where banks'
capital is the highest of 10.5% risk based capital requirements (including policy measures) and
4%
89
of the leverage ratio exposure measure. Recapitalisation needs to take place at the highest of
8% of risk based capital requirements and 3% of the leverage ratio exposure measure.
As shown in table 1, the model estimates that the introduction of FRTB and the leverage ratio
reduces expected bank losses in a severe 2008-type crisis from €346.32bn to €313.49 (a 9.19%
reduction), and reduces the impact on public finances after taking into account bail-ins and
resolution funds from an already very low figure of €5.49bn to €2.87bn (a -47.85% reduction).
Table 31.
Overview of benefits
Baseline
Benefits -
Reduction in
Financial
needs
Financial
needs
after
capital is
used
Financial
needs
after
bail-in
90
Financial
needs
after
resolution
fund
Policy scenario
Financial
needs
after
capital is
used
Financial
needs
after
bail-in
Financial
needs
after
resolution
fund
Impact
Impact
on
Financial
needs
after
capital is
used
Impact
on
Financial
needs
after
bail-in
Impact on
Financial
needs after
resolution
fund
In % of EU
GDP
In € bn
2.52%
0.37%
0.04%
2.29%
0.25%
0.02%
-9.19%
-32.20%
-47.85%
346.32
50.68
5.49
314.49
34.36
2.87
Costs (further details in annex 5.2)
In relation to costs, the analysis has focused on estimating through the QUEST model. In general,
regulation induces banks to increase capital relative to debt (including deposits). This has two
opposing potential effects on funding costs. Shifting to bank capital and paying an equity
premium increases funding costs, while lowering the demand for deposits reduces the deposit
rate, which lowers funding cost. The latter effect is, however, usually small, and likely even
smaller in the current environment with effectively zero deposit rates. The first effect therefore
dominates.
89
90
It is assumed that banks hold a 1% buffer in excess of the minimum leverage requirements, this is to
keep consistency with the risk based capital requirement where not every euro of losses immediately
leads to a recapitalisation need.
On the use of bail-in, some assumptions had to be taken. The actual amounts of bail-inable debt are not
available from the data, therefore we assumed that the amount of available bail-inable debt equals the
double of the minimum capital requirements corresponding to a loss absorption amount and a
recapitalisation amount.
74
 kom (2016) 0853 - Ingen titel
1715591_0075.png
Optimising banks could try to shift the higher funding costs onto the non-financial private sector
in the form of higher loan rates. This could increase capital costs for firms which partly finance
their investment with loans which could have an impact on the size of their investments. Based
on the same assumptions as for the estimation of the benefits, the estimated impact on long term
GDP could range between -0.03% and -0.06% depending on whether an offset
91
is applied on the
cost of equity or not resulting from the improved capitalisation of banks. The impact on banks'
funding costs would range between 1.38 bps and 2.71 bps. These estimates do not include
potential adjustments considered in this impact assessment to counteract the negative impact of
the contemplated measures.
It is important to note that the costs and benefits that have been quantified are not comprehensive
and are dependent on underlying assumptions (how to separate banks' balance sheets, behavioural
responses of banks, required rates of returns for the different funding sources under different
scenarios, etc.). Moreover, important social benefits (including the reduction in the occurrence of
systemic crisis and reduction in possible contagion between banks as well as the impact of the
reform on avoiding conflicts of interest, misallocation of resources and facilitating supervision,
etc.) and costs (such as economies of scope and scale, impacts on liquidity of secondary markets
and legal costs) have not been quantified and modelled.Qualitative assessment of benefits and
costs related to the NSFR
Because stable funding requirements have only recently been defined and have not yet been
introduced in any jurisdiction, empirical results on the impact of stable funding requirements are
sparse, especially the ones focussing on marginal costs and benefits of a stable funding
requirement. Potential benefits of a stable funding requirement are the reduced likelihood of bank
failure caused by liquidity shocks and smaller contraction of banks’ lending in reaction to a
liquidity shock. It is important to keep in mind that, for stable funding requirements, the potential
costs of the regulation in 'business as usual times' is compared to the costs of liquidity shocks.
As indicated in Dewatripont and Hancock et al. (2016) and EBA (2015), existing literature on the
NSFR discusses its expected impact on banks' balance sheets via a lengthening of liabilities’
maturity and a shortening of assets’ maturity. Funding sources considered as less stable may also
be replaced with more stable funding sources through e.g. substituting short term wholesale
funding with retail funding. As a direct effect, banks' profitability could decrease due to the
increase in funding costs. At macroeconomic level, the theoretical effects of liquidity regulation,
as discussed in existing literature, are a consequence of the interplay of the LCR and the NSFR.
On the benefits side, these effects are a reduced cost of bank failures (Calomeris et al (2015),
lower probability of simulatenous bank failures (Perotti-Suarez (2011)), a banking system less
vulnerable to liquidity shocks (Goodhart (2011), EBA (2015), Farhi-Tirole (2012)) and a lower
contraction of bank lending following a liquidity shock (Acharya-Viswanathan (2010), EBA
(2015)). On the costs side, these effects are a possible greater impact on market prices of liquidity
shocks due to similar asset holdings and herding (Bonfim-Kim (2012), Allen et al (2012)), a
decrease in bank lending in ‘business as usual times’ (Acharya-Viswanathan (2010)), lower
overnight and wholesale funding which could reduce the effectiveness of monetary policy (Bech-
Keister (2013)) and lower discipline of banks by wholesale investors (Calomeris-Kahn (1991),
Diamond-Rajan (2001)).
While none of the literature empirically tests all costs and benefits of the NSFR in an integrated
way, some empirical evidences are available:
91
This 50% offset is based on the work of Miles et al (2013)
75
 kom (2016) 0853 - Ingen titel
1715591_0076.png
Some papers examined bank behaviour during the most recent financial crisis and provide
indirect indications on the potential benefits of stable funding requirements. Cornett et al (2011)
found that the contraction of bank lending during the financial crisis was significant and that US
banks that had extended more contingent credit lines and banks having a lower proportion of
stable funding sources reduced more significantly their lending than other banks. In the same
vein, Pessarossi and Vinas (2015) found that banks with lower funding risk profile and a lower
ratio of long term loans to long term funding and deposits provide more lending after the
interbank market freeze in 2007-2008.
Chiaramonte and Casu (2016) tested the relevance of both structural liquidity and capital ratios,
as defined in Basel III, on EU banks' probability of failure. Estimates from several versions of the
logistic probability model indicate that the likelihood of failure and distress decreases with
increased liquidity holdings. The results show that banks that ran into difficulty almost always
had low NSFR and capital requirements well above the statutory minimum.Stakeholder analysis
The retention of simplified approaches to calculate capital requirements would ensure continued
proportionality of the rules for smaller banks. Furthermore, the additional measures to increase
proportionality of some of the requirements (related to reporting, disclosure and remuneration)
should decrease the administrative and compliance burden for those banks.
Other stakeholders, such as banks' clients (e.g. consumers and businesses), investors in securities
issued by banks, and financial markets as a whole, would be affected by the initiative indirectly.
As indicated above, there proposed measures could lead to an increase in lending rates, but the
increase is not expected to be so marked that it would lead to a significant impact on banks'
clients access to loans. Furthermore, after banks adjust to the new rules, they would be better
placed to provide loans to their clients. The resolution-related measures contained in the proposal
would have an impact on investors in banks' securities issued by G-SIIs: they will need to
become more active in monitoring the amount of risk-taking of the G-SII. The proposed
measures to increase the transparency of banks should help them in that respect. The same is true
for markets more in general. Finally, the combined proposed measures would increase the safety
and soundness of the financial system which is in the interest of all stakeholders.
5.3.
Impact of the preferred options on administrative costs
Administrative costs
92
stemming from the implementation of the whole package of
preferred options will be reduced to banks as a whole, mainly due to more proportionate
requirements for supervisory reporting and disclosure (for more details see annexes 3.2
and 3.3), which will primarily concern smaller institutions.
Burdensome supervisory reporting was frequently mentioned in the call for evidence.
The EBA is undertaking the analysis on proportionality of these costs. The preferred
options would reduce recurring costs through the introduction of more proportionate
reporting and significantly reduced disclosure requirements.
92
Administrative costs are defined as the costs which stakeholders (e.g. companies, citizens or public
authorities) incur due to legal obligations to provide information. The administrative costs have two
components: the business-as-usual costs and administrative burdens. The business-as-usual costs are the
costs resulting from obligations to provide information which would be done by an entity even in the
absence of the legislation. At the same time, the administrative burdens are the costs which the entity
would not incur in the absence of legislation, i.e. which is borne solely because of a legal obligation.
76
 kom (2016) 0853 - Ingen titel
As regards disclosure by small banks, at this point defined as banks with total assets
below €1.2 billion, they would be required to disclose only a simple key metrics table
with capital, liquidity and leverage ratios once per year compared to the current detailed
annual disclosure requirements. At this stage, however, precise number of concerns
banks is not known as some of them can already be relieved from disclosure
requirements if their parent company decided to provide disclosures only on the
consolidated basis.
On reporting, small banks would be subject to a reduced frequency of reporting half year
instead of quarterly and the EBA will be mandated to make proposals to further reduce
the granularity of supervisory reporting templates for small banks. Moreover, an
independent study would identify additional ad-hoc supervisory reporting requirements
that supervisors are currently imposing in addition to the single rule book on supervisory
reporting.
In some areas opportunity gains in administrative burden will be achieved by scoping out
smaller banks: revised CRR requirements such as for the trading book and TLAC. No
incremental administrative costs are expected from setting binding NSFR and LR. In the
same way, no significant administrative costs will be incurred due to the extension of
SME Supporting Factor, even for small banks, and the measure has been widely
supported by the banking industry. Similarly, no significant administrative costs will be
incurred due to the further harmonisation of moratorium, as these tools will continue
operating based on the same procedures. The option to partially harmonise the hierarchy
of creditor claims for unsecured debt would not have an impact on the administrative cost
of banks.
Moreover, in parallel to the proposed measures and with a view to further reducing
administrative costs resulting from supervisory reporting, the EBA has already a mandate
for developing common IT solutions and has developed a single data point model and a
single XBRL taxonomy for the supervisory reporting package. These can be used by
banks and software developers to optimise the collection and transfer of supervisory
reporting data between the banks and supervisors as well as between supervisors and the
EBA at an aggregated level. Overall, institutions are not expected to incur significant IT
development costs from the implementation of the preferred options.
5.4.
The impact on SMEs
The proposed recalibration of the capital requirements for bank exposures to SMEs is expected to
have a positive effect on bank financing of SMEs. This would primarily affect those SMEs,
which currently have exposures beyond €1.5 million as these exposures currently do not benefit
from the SME Supporting Factor.
Other proposed options in the impact assessment, particularly those aimed at improving
resilience of banks to the future crisis, are expected to increase sustainability of bank lending to
SMEs.
Finally, measures aimed at reducing compliance costs for credit institutions, particular the
smaller and less complex institutions are expected to reduce borrowing costs for SMEs.
77
 kom (2016) 0853 - Ingen titel
1715591_0078.png
5.5.
Impact on third countries
Third countries will benefit from the proposed review with regard to three important elements.
On one side, the proposal will enhance the stability of EU financial markets thereby reducing the
likelihood and costs of potential negative spillovers for global financial markets. Moreover the
proposed amendments will contribute to increase the harmonization of the regulatory framework
across Member States thereby reducing substantially administrative costs for third countries
banks operating in the EU. Finally, since several amendments are intended to align the EU
legislation to most advanced internationally agreed standards, the proposal will contribute to
increase the level playing field between EU banks and banks established in third countries with
an even more significant reduction of compliance costs for doing business in the EU.
6.
MONITORING AND EVALUATION
It is expected that the proposed amendments will start entering into force in 2019. The
amendments are tightly inter-linked with other provisions of the CRD IV and CRR, which are
already in effect since 2014. The current proposals underscore the importance of timely and
appropriate changes of the rules in response to the markets events, the evolution of the EU
economy, in particular its financing mechanisms, the new institutional setup with Banking Union
in place and the EU commitments with international fora (FSB and Basel in particular).
The Basel Committee and EBA will continue to collect the necessary data for the monitoring of
leverage ratio and the new liquidity measures in order to allow for the future impact evaluation of
the new policy tools. Regular Supervisory Review and Evaluation (SREP) and stress testing
exercises will also help monitoring the impact of the new proposed measures upon affected credit
institutions and assessing the adequacy of the flexibility and proportionality provided for to cater
for the specificities of smaller credit institutions. Additionally, the Commission services will
continue to participate in the working group of the BCBS and the joint task force established by
the European Central Bank (ECB) and by EBA, that monitor the dynamics of institutions' own
funds and liquidity positions, globally and in the EU, respectively.
The set of indicators to monitor the progress of the results stemming from the implementation of
the preferred options are the following:
On NSFR:
Indicator
Target
Net Stable Funding Ratio (NSFR) for EU institutions
As of date of application, 99% of institutions taking part to the EBA Basel III
monitoring exercise meet the NSFR at 100% (65% of group 1 and 89% of
group 2 credit institutions meet the NSFR as of end-of December 2015)
Semi-annual the EBA Basel III monitoring reports
Source of data
On leverage ratio:
Indicator
Target
Leverage ratio (LR) for EU institutions
As of the date of application, 99% of group 1 and group 2 credit institutions
will have the leverage ratio of at least 3% (93,4% of group 1 institutions met
the target as of June 2015)
Semi-annual EBA Basel III monitoring reports
Source of data
On SMEs
Indicator
Financing gap to SMEs in the EU, i.e. difference between the need for
78
 kom (2016) 0853 - Ingen titel
1715591_0079.png
Target
Source of data
external funds and the availability of funds
As of two years after the date of application, <13% (last known figure – 13%
as of end 2014)
European Commission / European Central Bank SAFE Survey (data coverage
limited to the euro area)
On TLAC:
Indicator
Target
TLAC in EU G-SIIs
All EU G-SIBs meet the target (>16% of risk weighted assets (RWA) /6% of
the leverage ratio exposure measure (LREM) as of 2019, > 18% RWA/6.75%
LREM as of 2022)
Semi-annual EBA Basel III monitoring reports
Source of data
On trading book:
Indicator
RWA for market risks for EU institutions
Observed variability of risk-weighted assets of aggregated portfolios applying
the internal models approach.
- As of 2023, all EU institutions meet the own funds requirements for market
risks under the final calibration adopted in the EU.
- As of 2021, unjustifiable variability (i.e. variability not driven by differences
in underlying risks) of the outcomes of the internal models across EU
institutions is lower than the current variability
*
of the internal models across
EU institutions.
_______________
*
Target
Reference values for the "current variability" of value-at-risk (VaR) and incremental
risk charge (IRC) requirements should be those estimated by the latest EBA "Report on
variability of Risk Weighted Assets for Market Risk Portfolios", calculated for
aggregated portfolios, published before the entry into force of the new market risk
framework.
Source of data
Semi-annual EBA Basel III monitoring reports
EBA Report on variability of Risk Weighted Assets for Market Risk
Portfolios. New values should be calculated according to the same
methodology.
On remuneration:
Indicator
Target
Use of deferral and pay-out in instruments by institutions
99% of institutions that are not small and non-complex, in line with the CRD
requirements, defer at least 40% of variable remuneration over 3 to 5 years
and pay out at least 50% of variable remuneration in instruments with respect
to their identified staff with material levels of variable remuneration.
EBA remuneration benchmarking reports
Source of data
On proportionality:
Indicator
Target
Source of data
Reduced burden from supervisory reporting and disclosure
80% of smaller and less complex institutions report reduced burden
Survey to be developed and conducted by EBA by 2022
On insolvency ranking:
Indicator
Target
Complaints about competitive disadvantages due to different insolvency
rankings of unsecured bank debt
Commission receives no indications or complaints about competitive
disadvantages due to different insolvency rankings of unsecured bank debt
79
 kom (2016) 0853 - Ingen titel
1715591_0080.png
Source of data
after the harmonisation.
Stakeholder feedback
On moratorium:
Indicator
Target
Source of data
Status of banks' liquidity before and after moratorium is used
Absence of bank runs, no transfer of funds cross-border and smooth
functioning of procedures in supervisory/resolution context
SRB/NRAs/possibly survey
The evaluation of the impacts is expected to be conducted within five years after the date of the
application of the new measures. The methodology should be designed taking into account the
output of monitoring indicators.
Compliance and enforcement will be ensured on an ongoing basis including, where needed,
through infringement proceedings for lack of transposition or for incorrect transposition and/or
application of the legislative measures. Reporting of breaches of EU law can be channelled
through the European System of Financial Supervision (ESFS), including the national competent
authorities, EBA as well as through the ECB. EBA will also continue publishing its regular
reports of the CRD IV-CRR/Basel III monitoring exercise on the European banking system. This
exercise monitors the impact of the Basel III requirements (as implemented through the CRR and
the CRD) on EU institutions in particular as regards institutions' capital ratios (risk-based and
non-risk-based) and liquidity ratios (LCR, NSFR). It is run in parallel with the one conducted by
the BCBS.
80
 kom (2016) 0853 - Ingen titel
1715591_0081.png
G
LOSSARY
ASF
BCBS
BRRD
CCP
CfE
CRR
CRD IV
CSR
EBA
ECB
EDIS
EU
FSB
FRTB
GDP
G-SIB
G-SII
HQLA
IFRS
IRB
LCR
MREL
MS
MtM
MtMM
NSFR
O-SII
RSF
Available Stable Funding
Basel Committee on Banking Supervision
Bank Recovery and Resolution Directive
Central CounterParty
Call for Evidence
Capital Requirements Regulation
Capital Requirements Directive
Credit Spread Risk
European Banking Authority
European Central Bank
European Deposit Insurance Scheme
European Union
Financial Stability Board
Fundamental Review of the Trading Book
Gross Domestic Product
Global Systemically Important Bank
Global Systemically Important Institution
High Quality Liquid Assets
International Financial Reporting Standard
Internal-Ratings Based
Liquidity Coverage Ratio
Minimum Requirement on own funds and Eligible Liabilities
Member State
Mark-to-Market
Mark-to-Market Method
Net Stable Funding Ratio
Other Systemically Important Institution
Required Stable Funding
81
 kom (2016) 0853 - Ingen titel
SA-CCR
SM
SME
SME SF
SSM
SREP
SRMR
TLAC
Standardised Approach for Counterparty Credit Risk
Standardised Method
Micro, small and medium-sized enterprise
SME Supporting Factor
Single Supervisory Mechanism
Supervisory Review and Evaluation Process
Regulation on Single Resolution Mechanism
Total Loss-Absorbing Capacity
82
 kom (2016) 0853 - Ingen titel
1715591_0083.png
A
NNEX
1. PROCEDURAL ISSUES AND CONSULTATION OF INTERESTED
PARTIES
The meeting of the Regulatory Scrutiny Board took place on 7 September 2016 to
discuss the draft impact assessment. The Regulatory Scrutiny Board issued a positive
opinion on this impact assessment on 27 September 2016.
Possible impact of the CRR/CRD IV on financing of the economy ("CRR
consultation")
Consultation activity
The CRR required the Commission to review the impact of own funds requirements on lending to
SMEs and long-term financing, including infrastructures.
93
As a result, the Commission services
consulted on the potential impact of the CRR and CRD IV on the financing of the economy,
including SME lending and long-term financing, in 2015.
94
This consultation has fed into the
preparation of the legislative initiative accompanying this impact assessment.
Stakeholder groups
There were 84 responses to the consultation. The majority of responses came from the financial
industry. Half of the responses came from three Member States: Belgium (a vast majority of
industry associations), the United Kingdom and Germany.
Chart 1: Type of respondent
Chart 2: Location of respondent
93
94
Article 501: the impact of own funds requirements on lending to SMEs and natural persons; article 505:
the appropriateness of the CRR requirements in light of the need to ensure adequate levels of funding
for all forms of long-term financing for the economy, including critical infrastructure projects; and,
Article 516: the impact of CRR on the encouragement of long-term investments in growth-promoting
infrastructure.
The public
consultation
was launched in July 2015 and closed in October 2015. A summary of the
responses is published on the Commission's
website.
83
 kom (2016) 0853 - Ingen titel
1715591_0084.png
Results
In terms of substance, the consultation asked stakeholders for their views on the impact and role
of the CRR/CRD IV on the recapitalisation process; lending to corporates in general and SMEs in
particular; and, lending to infrastructures. It also asked questions related to proportionality,
simplification and the single rulebook. While views differed substantially between type of
respondents, a number of high-level messages can be extracted:
Role of CRR/D in recapitalisation process:
All stakeholder groups shared the
view that the CRR/D have increased the resilience of the European banking
sector. As to what drove the increase in the capital levels, most banks ranked
regulatory demands as the most important ones, at the same time highlighting the
importance of supervisory and market demands, which in general frontloaded full
CRR requirements or even went beyond those. The views of other stakeholders
were mixed, emphasizing more the role of supervisory and market demands. The
financial industry did not portray capital requirements as excessive in general
(some contrasting views on the securitisation and credit valuation adjustment
(CVA)
95
frameworks). By contrast to the general acceptance for the minimum
regulatory requirements, the banking industry criticised additional supervisory
capital requirements, largely stemming from supervisory stress tests (Pillar II add-
ons), and macro prudential buffers, primarily because of their unpredictability,
complexity, lack of transparency and uneven implementation across the EU;
Lending to corporates:
all stakeholder groups argued that regulation was not key
in driving lending. Other factors, such as demand-side factors (slowing economic
activity) and monetary policy, affect the actual level of lending more than
regulatory requirements. Nevertheless, stakeholders generally agreed that
increased capital requirements have had a negative impact on the overall capacity
to lend, at least during the transitory period to adjust to the new capital
requirements. A few banks however stated that those banks which raised their
capital and retained earnings and those banks that already had high capital levels
were able to maintain their lending supply unaffected. A vast majority of
respondents agreed that the impact on corporate lending is in part structural
(permanent), and in part transitional (temporary). Banking industry most often
referred to the structural increase in refinancing costs. Banking industry
highlighted the important role of other financial sector regulations besides CRR,
notably, the BRRD, in affecting the cost/availability of lending. There were also
references to not yet implemented standards, such as the NSFR or the leverage
ratio, as having the potential to affect the availability and cost of lending. Banks
95
CVA risk is the risk of mark-to-market losses on OTC derivatives that are due to a deterioration in the
credit quality of the counterparty.
84
 kom (2016) 0853 - Ingen titel
1715591_0085.png
referred in general to instruments bearing more risk and with longer maturities
most affected by regulatory requirements. More specifically, they cited among
others: securitisations, trade finance, repos and derivatives and real estate
exposures;
Lending to SMEs:
views differed on the effectiveness of the SME supporting
factor (SF) in providing more lending to SMEs
96
. The financial industry was
largely of the view that the SF has been effective to incentivise SME lending.
SMEs and other corporates also supported capital relief for these SME exposures
and asked for an extension of the scope of application of the SF.
97
SMEs and
corporates also highlighted the importance of bank lending, as market financing is
associated with high fixed costs. However, a majority of supervisors and
regulators did not notice any clear impact of the SF on lending, with some noting
that the SF distorts the perception of actual risk arising from SME exposures.
Many respondents thought that concerns about SME funding should be solved by
other means (e.g. creating a credit register for SMEs and developing specific
public subsidies or guarantees for SME loans). Some respondents also provided
alternative proposals to change CRR in favour of SMEs, such as improving the
risk-sensitivity of the standardised approach, making the standardized approach
dependent on SME specific factors (profitability/turnover) or reviewing the
prudential calibration of some market segments, especially securitisations;
Lending to infrastructures:
According to the banking sector, CRR requirements
for infrastructure projects, especially capital and liquidity ones, do have an impact
on the capacity of banks to provide loans to this sector. Moreover, some indicated
that supervisory practices in approving banks’ risk measurement for infrastructure
puts some banks at a disadvantage and creates an uneven playing field. Public
authorities and supervisors were split on whether CRR requirements actually have
an impact on infrastructure lending. Those who answered positively regarded the
NSFR and the leverage ratio as having the greatest potential impact, arguing that
these would affect longer-maturity and lower-risk instruments, such as
infrastructure, relatively more. On the issue of whether infrastructure projects
should continue to be treated as loans to corporate borrowers, a majority of
respondents from all stakeholder groups answered negatively, except for public
authorities and supervisors, where the answers were more split. Justifications
given for a specific treatment were based on the alleged different features
98
and
presumable different risk of these two types of exposures. In particular, the
banking industry and corporate sector argued in favour of a specific supporting
factor for infrastructure, similar to the one for SMEs. There were also, from these
two groups, calls for greater risk-sensitivity of the standardised approach and
more harmonization amongst supervisory practices as regards these loans;
Proportionality:
within the banking sector, there was a divide between big and
small banks. The former generally argued against more proportionality, claiming
that there are already additional requirements for big banks and systemic
institutions. The latter favoured increased proportionality, arguing that
96
97
98
24% lower capital requirements for banks' SME exposures subject to certain conditions (see Article 501
of the CRR).
At the moment, only small SMEs are likely to benefit in practice given a limit of €1.5 million on the
total amount led to each SME.
Among those cited: longer maturity, higher collateralisation, higher recovery rate and lower volatility
of infrastructure exposures compared to corporate exposures.
85
 kom (2016) 0853 - Ingen titel
1715591_0086.png
compliance costs for small- and medium-sized banks can be disproportionate. For
those banks which wanted enhanced proportionality, there were different views
on how to select the target institutions and on which areas of the CRR should
allow a more proportional treatment. On the former, size and risk profile were the
most prominent responses. On the latter, there were a few singling out reporting
requirements; otherwise, responses were much dispersed, including: leverage,
market risk, operational risk or remuneration, among others. Supervisors and
public authorities were of the opinion that proportionality is already embedded in
the CRR through: risk-based rules, the possibility to choose between standardised
approaches and internal models and the additional requirements for systemic
institutions. However, supervisors and public authorities suggested simplifying
reporting requirements for small banks or simpler institutions. They also saw the
need to alleviate disproportionate compliance costs for these institutions by
simplifying complex rules, provided simpler rules are not less conservative;
Simplification:
the banking industry generally supported greater simplicity of the
rules. However, bigger banks were of the view that greater simplicity should be
promoted for all banks, while smaller ones thought they should be the main
target. Specific areas mentioned for simplification were: supervisory reporting,
using accounting values (dismissing prudent valuation or credit risk adjustments),
and governance and risk management requirements. Several private persons and
think tanks argued, as a general policy option, that internal ratings based
approaches should be replaced by a simple leverage ratio in combination with the
standardised approach; and
Single rulebook:
a clear majority of respondents from the banking industry were
supportive of greater harmonization and against national discretions. This
harmonization was understood by many respondents not only as harmonizing
Pillar I requirements, but also supervisory practices (Pillar II). There was some
recognition of the need to maintain certain national flexibility regarding the
macroprudential toolkit, given that different EU jurisdictions may not follow the
same financial cycles. However, these tools should be solely used to address
systemic risks, and not to address risks covered by the other CRR requirements.
Call for Evidence
Consultation activity
On 30 September 2015, the European Commission launched a public consultation
entitled the Call for Evidence: EU regulatory framework for financial services. The
consultation closed on 31 January 2016. The purpose of the Call for Evidence, which is
part of the Commission's 2016 work programme as a REFIT item, was to consult all
interested stakeholders on the benefits, unintended effects, consistency, gaps in and
coherence of the EU regulatory framework for financial services. It also aimed to gauge
the impact of the regulatory framework on the ability of the economy to finance itself and
grow.
Stakeholder groups
86
 kom (2016) 0853 - Ingen titel
1715591_0087.png
The Commission received 288 responses.
99
Most responses came from the UK, Belgium,
France, Germany and the Netherlands (figure A1, table A1). Responses came from
various sectors (tables A2-A3). The majority of respondents came from the financial
sector, including banking, investment management, insurance and market infrastructure
operators. The majority of respondents were providers of financial services, or
associations representing them. In contrast, responses from consumers of financial
services were more limited.
Figure A1.
Respondents by country
Table A1.
Respondents by country
Country of respondent
United Kingdom
Belgium
France
Germany
The Netherlands
Sweden
Italy
Spain
Finland
Greece
Denmark
United States
Ireland
Croatia
Austria
Czech Republic
Norway
Switzerland
Malta
Luxembourg
Hungary
Slovakia
Poland
Guernsey and Jersey
South Africa
TOTAL
No.
75
52
42
27
13
9
8
7
5
5
5
4
4
4
4
4
4
4
3
3
2
1
1
1
1
288
DK
EL
FI
ES
IT
SE
Others
UK
NL
DE
BE
FR
Source: Call for Evidence database
Source: Call for Evidence database
99
Responses to the Call for Evidence as well as the summary feedback statement can be found on the
Commission's
website.
87
 kom (2016) 0853 - Ingen titel
1715591_0088.png
Table A2.
Respondents by type
Type of respondent
Public Authority
Regulatory authority, Supervisory
Authority or Central bank
Government or Ministry
Regional or local authority
Organisation
Industry association
Company, SME,
micro-enterprise, sole trader
Consultancy, law firm
Consumer organisation
Non-governmental organisation
Think tank
Trade union
Academic institution
Private Individual
TOTAL
No.
29
15
13
1
246
218
89
7
7
6
4
3
2
13
288
Table A3.
Respondents by sector
Sector of respondent
Banking
Investment management
Insurance
Market infrastructure
operator
Pension provision
Auditing
Consumer protection
Accounting
Civil society
(advocacy, unions, NGOs)
Other Financial services
Credit rating agencies
Corporate
(governance, issuers, treasuries)
Consultancy, law firm
Telecommunication
Social entrepreneurship
Academia
Energy
Auto
Real estate
News
Transport
TOTAL
Source: Call for Evidence database
No.
100
79
50
39
30
21
20
19
19
19
11
11
8
8
7
7
6
2
2
1
1
288
Results
Respondents referred to all the main legislative acts in financial services, but most replies
concerned the Capital Requirements Regulation and Directive (CRR/CRD IV). While the
Call for Evidence aimed to assess the effect of existing legislation, respondents also
expressed views on possible forthcoming regulation. This was particularly pronounced in
the area of banking, where a large number of claims were focused on the leverage ratio,
NSFR, FRTB, TLAC and BSR. Responses covering the CRR/D raised the following
issues that are of relevance for this impact assessment:
Unnecessary constraints on financing:
some public authorities and other non-
industry respondents argued that higher regulatory capital requirements for
institutions may have a net positive effect on the financing of the economy in the
longer term, while adverse effects on loan supply may occur in the short term.
They further argued that the slowdown in lending observed in some Member
States is more likely due to factors other than regulation (e.g. lower demand for
loans). Many respondents sought improvements in financing conditions for
SMEs. They suggested providing further support to SME financing, for instance
by continuing with the current ‘supporting factor’ for loans to SMEs. They also
expressed concerns about the potential impact of capital requirements for interest
rate risk in the banking book, especially if those would be introduced in the form
of a Pillar 1 requirement..
Negative impact on market liquidity:
many respondents stressed the combined
impact of the leverage ratio, the NFSR and the revised capital requirements for
market risk on market liquidity in general and, in relation to the revised market
risk rules, the particularly negative impact on e.g. market-making. Hence, as
regards the latter respondents proposed to reconsider specific aspects of
88
 kom (2016) 0853 - Ingen titel
calibration and making it less operationally difficult to apply the standardised
approach.
Proportionality:
A large number of respondents called for a more proportionate
application of the rules, in particular on: (i) reporting and disclosure
requirements; and, (ii) prudential requirements. As regards the former,
respondents highlighted the difficulty for smaller and less complex credit
institutions to comply with these requirements, including those that will
eventually apply in relation to the NSFR. There were additional concerns that
requirements would be "gold-plated" by some Member States (e.g. require
subsidiaries of international groups to report additional financial information at
individual level). As regards the latter, some respondents argued that capital
requirements should take better into account firms' size and business model, in
particular with regard to smaller and less complex credit institutions. Finally,
some respondents also argued that the leverage ratio could reduce diversity, as it
would have a disproportionate negative impact on low risk-weighted business
models (e.g. specialised community banks, building societies, mortgage banks).
Reporting and disclosure obligations:
Banking associations and individual
institutions frequently pointed to reporting burdens imposed by various regulatory
and supervisory bodies (national competent authorities, the SSM, EBA, etc., to
perceived inconsistencies between various reporting requirements and respective
templates, as well as to wide-spread ‘gold-plating’ by competent authorities in a
context of maximum harmonisation.
Interactions:
many claims stressed possible inconsistencies arising from the
interaction between EMIR and the CRR. Specifically, respondents argued that the
introduction of the leverage ratio would be penalising for institutions offering
clearing services, as it does not take into consideration the risk-reducing effect of
(segregated) initial margin provided by the institutions' clients in relation to the
CCP-cleared transactions.
Targeted consultations
On the NSFR,
to complement the EBA report and the responses to the call for evidence,
the Commission services conducted an additional targeted consultation to gather
stakeholder's views on some specific aspects of this requirement:
- the potential adjustments resulting from complying with the NSFR;
- the treatment of derivative transactions;
- the treatment of short term transactions with financial institutions;
- the effective application of the principle of proportionality.
Respondents expressed concerns that the cost of compliance with the NSFR requirement might
be excessive for certain specific business models or activities, in particular for short term and
market activities.
The treatment of derivative transactions is one of the main sources of concern in a vast majority
of answers to the consultation. The additional requirement to hold 20% of stable funding against
gross derivatives liabilities is very widely seen as a rough measure that overestimates additional
funding risks related to the potential increase of derivative liabilities over a one year horizon. The
rules underpinning the calculation of NSFR derivative assets and liabilities and in particular the
89
 kom (2016) 0853 - Ingen titel
asymmetric treatment between variation margins received and posted and of initial margins
received and posted is also cited as detrimental to derivatives markets.
Regarding short term transactions with financial institutions, the vast majority of respondents are
concerned that the asymmetric treatment of short term (less than 6 months) secured funding (0%
ASF) and lending (10% or 15% depending on the quality of the underlying collateral)
transactions with financial counterparties could be very detrimental to market making activities
and, as a consequence, to the liquidity of repo market and of the underlying collateral. Repo
markets are presented as essential for the smooth functioning of both bank liquidity management
and market makers' inventory management. This treatment also raises some concerns regarding
the impact on the interbank market, in particular for liquidity management purposes.
The 5% RSF factor that applies to Level 1 HQLA and the high RSF factor that applies to non-
HQLA equities are criticised as being too high. For the Level 1 HQLA, this is deemed as not
being consistent with the LCR that recognize the full liquidity of these assets even in time of
severe stress. For non-HQLA securities, they think that funding requirements for a particular
asset should depend on the purpose for which the bank holds the asset (eg securities held as a
market hedge for a derivative transaction).
Secured issuances, and covered bonds issuances in particular, are single out as being
unintendedly penalised by the NSFR. Concerns are also raised about the continued
ability to operate pass-through structures, amongst which the distribution of promotional
loans feature prominently. Doubts are also voiced about the relevance of a stable funding
requirement for business models that, even though they require a banking license, engage
into maturity transformation to a very limited extent. The respondents are in favour of
taking into account European specificities and raise some more technical issues on the
design of the NSFR.
Finally, the majority of respondents do not favour a reduced scope of application or
differentiated treatment for small banks to make NSFR requirements more proportionate.
An exemption from NSFR requirements or the introduction of simplified metrics for
either smaller or 'low funding risk' institutions do not have a wide support. They are
furthermore in favour of applying the NSFR on a consolidated basis only or, at least, of
defining a preferential symmetric treatment of intragroup transactions if the NSFR is also
applied on an individual basis.
On the FRTB,
to complement the EBA report and the responses to the call for evidence,
the Commission services conducted an additional targeted consultation to gather
stakeholder's views on the application of the principle of proportionality under the
revised market risk framework, including:
- potential changes to the current derogation for small trading book businesses; and
- potential options for a simplified calculation of the market risk capital requirements for
small banks.
A majority of respondents choose as preferred policy option a combination of the derogation for
small trading book businesses and a simplified standardised approach.
First, respondents agree that the new standardised approach of the FRTB framework, the
sensitivities based approach (SBA), is far more complex than the existing approach under CRR
and this additional complexity would be inappropriate for banks with small or medium trading
books. In particular, respondents consider that the granularity of data requirements under the
90
 kom (2016) 0853 - Ingen titel
1715591_0091.png
SBA would be too extensive which would complicate its use on an on-going basis. In addition,
the one-off costs of implementing the SBA could be substantial
100
.
A majority of respondents agree that treatment of capital requirements of trading positions of
institutions granted with the derogation for small trading book businesses would be inadequate
for institutions with medium-sized trading books, due to its crudeness, implying that it would not
be a solution to raise the thresholds of this derogation to capture more institutions that could
suffer from the introduction of the SBA.
An alternative, simplified standardised approach is envisaged as the solution for institutions with
medium-sized trading books for most of respondents. To the question on what this simplified
standardised approach should consist of, there is also a clear consensus among respondents to use
the current standardised approach as the basis for the new simplified standardised approach. In
this case, institutions would avoid any implementation costs. However, some respondents
highlight that some recalibration of the current standardised would be necessary, in order to keep
incentives to move to the SBA but also to avoid cliff-effects. Finally, no respondents provided
clear proposals for the eligibility criteria that would grant institutions with medium-sized trading
books the permission to use a simplified standardised approach instead of SBA.
Regarding the derogation for small trading book businesses, most respondents support keeping it.
There is not strong support for raising the threshold of the derogation, at least not significantly,
but some say it should be explored, based on data. On the other issues concerning the definition
of the derogation, respondents are not very specific. Some support clarifying the definition of the
size of trading assets and the application of the treatment provided by the derogation. Others
highlight the fact that the scope of the current derogation does not include positions in FX and
commodities and need to apply a simpler regime for these positions for banks under the
derogation, especially for FX.
On the introduction of SACCR,
to complement the EBA report and the responses to
the call for evidence, the Commission services conducted an additional targeted
consultation to gather stakeholder's views on the overall complexity and operational
burden to implement SACCR and whether it would be preferable to maintain some of the
current standardised approaches for counterparty credit risk exposures, simpler than
SACCR, for small banks.
The majority of respondents see some merits in introducing SACCR in the EU, mostly
because it would increase the risk-sensitivity of the capital requirements for counterparty
credit risk and align capital requirements with the true risks faced by institutions.
However, all respondents recognised that the SACRR approach would impose undue
complexity to institutions with small trading portfolios and therefore a simpler alternative
approach should be maintained for them. Respondents have diverging views whether this
simpler alternative should be the current Original Exposure Method, the current Mark-to-
Market method or a revised version of the Original Exposure Method to align certain of
its assumptions with SA-CCR.
Finally, most of the respondents considered that more institutions should be able to use a
simpler alternative to SACCR than institutions that are currently permitted to use the
OEM (ie the institutions that are eligible for the derogation for small trading book
business under CRR article 94).
100
One member reported that the cost of implementation for a bank would be at least 1 Million, which is a
considerable investment for an institution with small trading book activities.
91
 kom (2016) 0853 - Ingen titel
1715591_0092.png
Remuneration
The Commission has engaged in several work streams in order to carry out its assessment of the
CRD IV remuneration rules and to collect the information underpinning this impact assessment.
The strategy consisted of a mix of the following: stakeholders’ consultation (through a
stakeholders’ event, bilateral meetings and a public consultation), own research, input from the
European Banking Authority (EBA) and a study on a number of aspects relevant for evaluating
the CRD IV remuneration rules commissioned to an external contractor.
On 16 December 2015, the European Commission hosted a fact-finding stakeholder
event in the context of its ongoing review of the remuneration rules of the Capital
Requirements Directive (2013/36/EU) and Regulation (No 575/2013). The objective of
the event was to gather evidence on the effects of the CRD IV remuneration rules in
terms of contributing to curbing excessive risk taking and by impacting on the incentives
for the so-called “material risk takers”.
The effectiveness of deferral requirements in term of risk-adjustment was generally
acknowledged, with a number of participants arguing in favour of a possible
differentiated regime for certain types of investment firms, or in favour of a proportional
application depending on the size of the firm and on the level of individual remuneration.
On the issue of pay-out in instruments, it was expressed that the use of share-linked
instruments should be allowed for listed companies (as it is for non-listed ones), as the
process to pay-out in shares is considered burdensome, unsuited for certain types of
entities (e.g. cooperative banks) and is subject to certain restrictions in some foreign
jurisdictions. Non-listed companies on the other side can be faced with high costs for
creating suitable instruments and this cost, it was argued, can be disproportionately high
for small firms.
A recurrent theme throughout the discussion was an argument about the allegedly excessively
wide application of the CRD IV remuneration rules, and a plea from some of the participants to
allow disapplication of those rules on the basis of the principle of proportionality with respect to
certain entities, staff or awards.
Public consultation
The public consultation on the impacts of the maximum remuneration ratio under the Capital
Requirements Directive 2013/36/EU (CRD IV), and on the overall efficiency of the CRD IV
remuneration rules, ran from 22 October 2015 to 14 January 2016. By the set deadline 35 online
contributions were received from a variety of stakeholders such as credit institutions, investment
firms, industry or employee representation organizations and public authorities. A summary of
the contributions is provided below for each of the topics covered by the public consultation.
Figure A2.
Country overview of the respondents
92
 kom (2016) 0853 - Ingen titel
1715591_0093.png
Figure A3.
Profile overview of the respondents
The requirement to defer part of the variable remuneration
Most respondents agreed with the deferral requirement and positively appreciated its
effectiveness in ensuring alignment with long-term performance and deterring excessive risk-
taking behaviour. A few respondents highlighted its usefulness in conjunction with the
application of
malus
and one respondent considered that deferral is useful in retaining employees.
Regarding the percentage of variable remuneration to be deferred, a few respondents supported a
higher deferred portion (i.e. 60%) for senior managers and the highest paid material risk takers.
Those who assessed the deferral period generally supported the appropriateness of 3 to 5 years,
but certain investment firms (e.g. proprietary trading firms) argue in favour of shorter deferral
periods, better aligned with the time horizon of their investments and associated risks. Asset
managers consider that the UCITS V and AIFMD rules contain provisions regarding deferral
periods that appropriately account for the fund strategy, risk and lifecycle.
93
 kom (2016) 0853 - Ingen titel
Many respondents argued that it is important to preserve flexibility in the application of the rules
and to maintain the possibility to exempt some entities and staff from the application of some of
the remuneration rules. Some ask for the definition at EU level of uniform thresholds to apply the
requirements on variable remuneration. Another respondent considered that national supervisory
authorities are the best placed to assess to which extent a particular bank should comply with
each of the rules.
It is argued that deferral is not appropriate for staff members receiving only low amounts of
variable remuneration (or for small, non-complex institutions that generally pay out only limited
amounts of variable remuneration) for a number of reasons: (i) the deferral of small amounts of
variable remuneration would have detrimental motivational effects on staff and erode the
perceived value of the award, leading in some cases to increases in either fixed or variable
remuneration; (ii) it would make it less attractive for lower-paid identified staff with transferrable
skills (e.g. in control functions or middle management) to retain or take up jobs in CRD IV-
regulated sectors; (iii) deferral would be particularly difficult to apply with respect to staff in
non-EEA subsidiaries.
Some respondents consider the deferral requirement costly and administratively difficult, with
costs affecting disproportionately smaller firms and the multiple, small instalments of deferred
variable remuneration in the case of staff with low bonuses. Deferral is also said to have only
negligible (if any) influence on the risk-taking behavior of staff receiving only low variable
remuneration.
A vast majority of respondents therefore argued in favour of a proportionate application of the
deferral requirement.
The requirement to pay out part of the variable remuneration in instruments
A number of respondents considered that pay-out in instruments is an efficient tool in terms of
aligning the remuneration of material risk takers with the performance and risks of the institution.
Nevertheless, most respondents pleaded for a more proportional application of the pay-out in
instruments requirement and considered that its administrative burden outweighs its benefits in
the case of staff earning only low levels of variable remuneration and in the case of institutions
that are small, non-complex or of a certain legal form (e.g. public bank, building society, savings
or cooperative bank, principal trading firm).
Pay-out in instruments is considered particularly problematic for institutions with a specific legal
form or ownership structure, for which there would be legal and factual barriers preventing them
from issuing such instruments. Where institutions cannot issue shares, requiring to issue
“equivalent non-cash instruments” is said to create additional risk for these firms (they cannot
readily hedge against the additional cost that may be associated with an increase in the value of
the underlying instrument) and additional cost (they would need to ”value” the instruments).
Respondents stated that it is important to maintain flexibility in the type of instruments used, as
long as they have the same efficiency. It is argued in this respect that listed institutions should
have the choice between shares and share-linked instruments. Share-linked instruments are said
to be as efficient as shares in terms of alignment with the institutions’ performance and risks,
while it can be applied more easily, in a less costly way and in a uniform manner worldwide.
94
 kom (2016) 0853 - Ingen titel
1715591_0095.png
It is said that the payment in shares cannot be put in place in a uniform way in all countries, given
the different legal, regulatory, accounting and tax constrains and formalities (in some countries
payment in shares would even be forbidden). From an operational standpoint (IT, HR,
governance accounting and tax, external and internal communication) pay-out in shares is also
considered complicated and costly. Institutions would need to either create new shares or buy
them in the market. Creating new shares would mean that existing shareholdings get diluted,
whereas buying shares is said to have the possible disadvantage of triggering speculation, thus
resulting in the institution needing to pay a hefty premium.
On the other hand, one respondent considers that shares are commonly known financial
instruments for which staff members may appreciate their link to the institution’s performance,
while share-linked or debt instruments may be too opaque and difficult to understand, hindering
the staff member’s ability to assess their value against the institution’s performance.
Regarding the use of bail-in-able debt instruments, it was argued that they should be limited to
top staff and that they could be costly if the existing instruments are not adapted to paying
remuneration (e.g. because they are meant for large institution investors, with no secondary
markets available and not aligned with remuneration schemes in terms of maturity).
Some investment firms (in particular employee-owned or controlled by a small group of
employees or founders, and where risks are said to be effectively aligned with those of the long-
term interest of the firm) consider that the rules on payment in instruments are too complex and
expensive for their kind of firm. They consider that deferred cash bonuses that remain subject to
full forfeiture serve as a far more effective disincentive to imprudent risk-taking.
Cooperation with the European Banking Authority
EBA was closely associated with the process of evaluating the CRD remuneration rules, by
gathering and providing information and data through annual reports on Benchmarking
remuneration practices at EU level
101
, a Public Consultation on its draft
Guidelines on sound
remuneration policies,
which contained a number of questions directly relevant for the issue of
the proportionate application of the rules
102
, as well as a
Report on the Member States’
implementation of the rules under the principle of proportionality,
accompanied by the
Opinion
on proportionality
to the Commission advising on a CRD IV legislative change.
A
NNEX
2. P
ARTIAL EVALUATION OF THE EXISTING POLICY FRAMEWORK
The CRR and CRD IV entered into force on 1 January 2014. Therefore, at this stage there is
insufficient available data and experience for conducting a full evaluation. Nevertheless, the need
of amending these instruments in order either to introduce new provisions or to review the
existing ones has emerged as a result of the work carried out by the BCBS, obtaining evidence on
the national implementation of the Directives or as an outcome of specific consultations and
studies, solicited by the Commission.
The focus of the analysis below is limited to providing early and targeted assessment of two
specific areas: the rules on remuneration and the impact of CRR on bank financing of the
economy, including SMEs.
103
101
102
Available at https://www.eba.europa.eu/regulation-and-policy/remuneration
Consultation on
Guidelines on sound remuneration policies (EBA/CP/2015/03)
103
The other areas covered in the problem definition are not in the scope of the existing policy framework
95
 kom (2016) 0853 - Ingen titel
1715591_0096.png
A full evaluation will be conducted after sufficient experience with the functioning of the new
rules has been gathered.
Annex 2.1. Evaluation of rules on remuneration
As required under Article 161(2) of the CRD IV, the Commission has reviewed the
efficiency, implementation and enforcement of the remuneration rules. In carrying out
this review, the Commission engaged in several work streams. It studied available
academic literature and commissioned a study from an external contractor to assist with
its assessment
104
. It sought stakeholders’ input through a public consultation
105
, a fact-
finding stakeholder event and bilateral meetings with industry representatives. Moreover,
the Commission engaged with Member State representatives and supervisory authorities.
In accordance with the CRD IV mandate, the European Banking Authority was closely
associated with the review process and delivered valuable information. In particular, the
European Banking Authority reports on high earners and on benchmarking of
remuneration practices at EU level
106
were a valuable source of data covering the years
2010-2014. The findings of the Commission's evaluation are reflected in the Commission
Report COM(2016) 510
107
.
Other than for the maximum ratio between variable and fixed remuneration, for which
the review found that for the time being there is insufficient evidence to draw final
conclusions on its impact, the review allowed for a largely positive assessment of the
remuneration rules.
The rules on the governance of remuneration processes, performance assessment,
disclosure and pay-out of the variable remuneration of identified staff, introduced by
CRD III are overall well received by stakeholders and thus can be positively assessed in
terms of acceptability.
These rules are found to contribute to the overall objectives of curbing excessive risk-
taking and better aligning remuneration with performance, thereby contributing to
enhanced financial stability. These objectives are still fully relevant today. The rules can
thus be positively assessed in terms of effectiveness and relevance.
The CRD IV remuneration rules and associated delegated acts
108
brought about a set of common
requirements on remuneration. The rules continue to require action at EU level in order to ensure
the level-playing field, avoid fragmentation of the internal market and eliminate the risk of
similar institutions being treated differently depending on the jurisdiction in which they are
104
institut für finanzdienstleistungen e.V.
,
study on the remuneration provisions applicable to credit
institutions and investment firms (2016).
Public consultation on impacts of maximum remuneration ratio under Capital Requirements Directive
2013/36/EU (CRD IV), and overall efficiency of CRD IV remuneration rules (22.10.2015 –
14.01.2016).
All publications are available at https://www.eba.europa.eu/regulation-and-policy/remuneration/-/topic-
documents/ ckV8kFRsjau9/more.
Report from the Commission to the European Parliament and the Council of [28 July 2016] –
Assessment of the remuneration rules under Directive 2013/36/EU and Regulation (EU) No 575/2013.
Commission Delegated Regulation (EU) No 527/2014
introduced a harmonised definition of classes of
instruments that adequately reflect the credit quality of an institution as a going concern and are
appropriate to be used for the purposes of variable remuneration, whereas
Commission Delegated
Regulation (EU) No 604/2014
) laid down qualitative and quantitative criteria to identify categories of
staff whose professional activities have a material impact on an institution’s risk profile
96
105
106
107
108
 kom (2016) 0853 - Ingen titel
1715591_0097.png
located. A common binding framework is all the more relevant given that some institutions are
active in more than one EU Member State. Thus, the rules on remuneration can overall be
positively assessed in terms of their EU added value.
The review nevertheless also revealed shortcomings with respect to the rules on deferral
and pay-out in instruments in certain specific circumstances. The Commission therefore
carried out a detailed evaluation of these two rules in function of their relevance,
effectiveness, efficiency, coherence, acceptability and EU added value, the findings of
which are set out in a Staff Working Document
109
. The evaluation yielded positive results
with regard to the relevance, effectiveness, efficiency and acceptability of the two rules
overall, but revealed significant reservations in the particular cases of small and non-
complex institutions and of staff with low levels of variable remuneration. A negative
assessment was also made with respect to the efficiency and acceptability of the
provision requiring listed institutions to use shares (and not share-linked instruments) for
meeting the requirement under Article 94(1)(l)(i) CRD IV. The coherence of the
analysed provisions in the absence of implementation flexibilities for the above-
mentioned types of institutions and staff was assessed as rather low, whereas the overall
EU added value was assessed positively.
109
Staff Working Document SWD(2016)266
- Evaluation of the deferral and pay-out in
instruments rules under Directive 2013/36/EU
97
 kom (2016) 0853 - Ingen titel
1715591_0098.png
Annex 2.2. Impact on the bank financing of the economy, including SMEs
Main conclusions on the impact of CRR on bank financing of the economy and
infrastructure
As far as the impact of CRR on the long-term financing and investment is concerned, the
Commission commissioned a study to London Economics to assess the impact of CRR
on bank financing of the economy. [Impact of the Capital Requirements Regulation
(CRR) on the access to finance for business and long-term investments insert a link to a
website where the study will be published].
The following main conclusions can be drawn from the report:
Main estimate of the transitional effect, derived in this study using data for the
period 1985-2014, shows that for a one percentage point increase in the Total
Capital Ratio the impact on lending flows of banks in the EU is -0.8% over one
year with the implied impact over a three-year period being -1.5%.
Macroeconomic environment matters a lot for the credit flows to the economy. A
one percentage point increase in the output gap results in a 0.95% reduction in
bank lending flows.
An analysis carried out for subsamples of banks based on pre-crisis business
models proxied by size, capitalisation, and funding, showed that the impact of the
Total Capital Ratio on bank lending flows was greater for banks that have
historically been less capitalised and are funded to a greater extent through non-
deposit liabilities.
Estimated impact of the Total Capital Ratio on bank lending stocks in long-run is
negative (of -2.2%), but the effect is not statistically different from zero.
There is not clear evidence of a major impact of increased capital requirements
under the CRR on bank financing of infrastructure, a result which is consistent
with findings from the consultations and survey. The results highlight further that
the impact of changes in the Total Capital Ratio on bank lending flows in general
(as per the transitional effects analysis) are economically more significant than on
bank financing of infrastructure in particular.
These conclusions have been taken into account once estimating different options,
particularly on their impact to maintain sustainable bank financing of the economy.
Main conclusions from the EBA report on SMEs
Art 501 of CRR introduced a capital reduction factor for exposures to SMEs under both the SA
and IRB approach. The introduction of this factor was accompanied by a review clause according
to which the Commission, on the basis of an advice from the EBA, should have assessed by the
28 June 2016 the impact of the measure on SME lending.
The EBA in its 2016 report on SMEs
110
, using the data made available by national supervisors,
highlighted that the capital reduction stemming from the CRR did not make SMEs to benefit
110
EBA report on SMEs
98
 kom (2016) 0853 - Ingen titel
1715591_0099.png
more than large corporates in the provision of new loans:
there is no evidence yet that the SME
SF has provided additional stimulus for lending to SMEs compared to large corporates. In
particular, according to the results presented, SMEs have faced the same probability of being
credit constrained as large firms.
The EBA, however, also recognised
that it might be too early
to draw conclusions, given the limitations of the data available and the relatively recent
introduction of the SME SF. In order to be effective the SME SF has to be fully integrated into the
decision process of institutions which is not yet the case
111
.
EBA in the SME report also highlighted that the use of SF is consistent with the empirical
riskiness of SME exposures, except for the retail asset class in IRB banks:
"The results for
France and Germany suggest that, under CRR/CRD IV, the SME SF is consistent with the lower
systematic risk of SMEs for all exposure classes in the SA, and for corporate SMEs in the IRBA.
However, for IRBA retail loans, the capital reductions associated with the SME SF lead to
relative capital requirements that are lower than those suggested by the systematic risk. As a
result, after the application of the SME SF, the relative regulatory RWs are in line with the
empirical ones in the IRBA corporate exposure class and the SA, but are lower than the
empirical ones in the IRBA retail class, suggesting that these exposures may not be sufficiently
capitalised relative to large corporates"
112
.
Regional differences in the EU
For the EBA reporting banks, the highest impact in CET 1 ratios is observed in
most Easter EU Member States>0.45% point in CET1 ratios: Estonia, Latvia,
Lithuania, Slovenia;
For smaller banks, which do not report regularly to EBA, the highest impact
(>0.4% point change in CET 1 capital ratio) is observed on Italy, Germany,
Poland, Sweden, Belgium.
Table A4.
Increase in CET1 capital ratio due to SME SF
111
EBA report on SMEs,
March 2016, p. 11
112
EBA report on SMEs,
March 2016, p. 95
99
 kom (2016) 0853 - Ingen titel
1715591_0100.png
100
 kom (2016) 0853 - Ingen titel
1715591_0101.png
EBA conclusions on the consistency of risk weight with the actual riskiness of SMEs
Table A5.
Risk weights under IRB and SA and their comparison with the empirical riskiness in
France and Germany over the full economic cycle
France
Retail
0.75 -
1.5
IRBA
SA
IRBA
SA
Corporate
Turnover (€mio)
Basel III
CRR/CRD IV
1.5 - 5
5 - 15
15 - 50
BM
Applicable risk weights
46%
78%
80%
91%
100%
75%
100%
100%
100%
100%
35%
59%
61%
70%
100%
57%
76%
76%
76%
100%
% points difference from the applicable risk
weights
57%
11%
-19%
22%
-1%
58%
-20%
-42%
-2%
-19%
59%
-21%
-41%
-2%
-17%
63%
-28%
-37%
-6%
-13%
100%
0%
0%
0%
0%
Estimated RWs on the actual
riskiness of loans
Difference to Basel III
Difference to CRR/CRD IV
IRBA
& SA
IRBA
SA
IRBA
SA
Germany
Turnover (€mio)
Basel III
CRR/CRD IV
Corporate
5 - 15 15 - 50
BM
0.75 - 1.5 0.75 - 1.5
1.5 - 5
Applicable risk weights
IRBA
46%
47%
78%
82%
93%
100%
SA
75%
75%
100%
100%
100%
100%
IRBA
35%
36%
59%
62%
71%
100%
SA
57%
57%
76%
76%
76%
100%
% points difference from the applicable risk weights
IRBA
IRBA
SA
IRBA
SA
48%
2%
-27%
13%
-9%
47%
1%
-28%
12%
-10%
44%
-34%
-56%
-15%
-32%
58%
-24%
-42%
-4%
-18%
63%
-30%
-37%
-7%
-13%
100%
0%
0%
0%
0%
Retail
Estimated RWs on the
actual riskiness of loans
Difference to Basel III
Difference to CRR/CRD IV
In comparison to empirical riskiness, in relative terms:
RWs too low
RWs are about right
RWs are too high
Source:
EBA report on SMEs,
figures 47 and 48
101
 kom (2016) 0853 - Ingen titel
EBA conclusion on the relevance of the €1.5 mio threshold,
EBA report p. 92
"No empirical evidence supporting the limit of €1.5 million currently implemented in Article 501
of the CRR is found for either Germany or France. This means that the limit of €1.5 million for
the amount owed set in the Article 501 of the CRR does not seem to be indicative of any change
in riskiness for firms. Hence, further work would be required to understand whether the limit is
justified, compared to the €1 million threshold already existing in the CRR for the allocation of
retail/corporate exposures or a different threshold".
EBA conclusion on the prudential soundness of the SF,
EBA report, p. 88-89, 94
"The results for France and Germany suggest that, under CRR/CRD IV, the SME SF is consistent
with the lower systematic risk of SMEs for all exposure classes in the SA, and for corporate
SMEs in the IRBA. However, for IRBA retail loans, the capital reductions associated with the
SME SF lead to relative capital requirements that are lower than those suggested by the
systematic risk. As a result, after the application of the SME SF, the relative regulatory RWs are
in line with the empirical ones in the IRBA corporate exposure class and the SA, but are lower
than the empirical ones in the IRBA retail class, suggesting that these exposures may not be
sufficiently capitalised relative to large corporates"
Bank lending to natural persons
Article 501 (4) requires the Commission to report on the impact of the own funds
requirements also on lending to natural persons. The figure below does not indicate any
material drop in lending to natural persons in the aftermath of the crisis, except mild
reduction in the stock of consumer loans during 2014. However, lending to natural
persons increased in all the relevant categories of lending over 2015, namely in the
category of
households, mortgage lending
and
credit for consumption,
leading to a
continuing increase in the stock of lending in these categories. While it is cannot be
excluded that an increase in the overall capital requirements could have had a negative
net impact on lending to households, particularly over 2014, overall macroeconomic
environment, such as interest rate policy by monetary institutions, can adjust or be
adjusted effectively so as to maintain a sustainable level of credit flow to households.
Moreover, the success of the peer-to-peer lending platforms over the recent years
suggests that households can also obtain credit, particularly consumer credit, outside the
banking system. Finally, no respondent to the CRR consultation or to the Call for
Evidence raised an issue of the lack of credit to natural persons.
Figure A4.
Evolution of bank lending in the EU
102
 kom (2016) 0853 - Ingen titel
1715591_0103.png
Source: ECB data warehouse.
103
 kom (2016) 0853 - Ingen titel
1715591_0104.png
A
NNEX
3. ASSESSMENT OF OTHER PROPOSED AMENDMENTS TO CRR/CRD
IV/BRRD
As mentioned in annex 2, the CRR and CRD IV entered into force on 1 January 2014. Therefore,
at this stage there is insufficient data for a full evaluation of most topics included in this impact
assessment, even though the CRR included some early "review" clauses.
However, the call for evidence launched by FISMA in 2015
113
(please add reference) has allowed
to identify shortcomings on some areas/provisions warranting some fine-tuning of existing rules.
Moreover, for several of these issues, a review has been recently finalised by the BCBS and
needs to be reflected in EU legislation.
Therefore, an early review of a number of provisions becomes necessary even in the absence of
data allowing completing a full evaluation. A preliminary analysis of the functioning of the
provisions/areas referred above is summarised below. The amendments suggested in relation to
these provisions/area are of limited scope/impact or, in other cases, implement a solution which is
straightforward and uncontroversial. In a few cases the amendment is basically required to
ensure a better alignment with the applicable international standards.
113
See http://ec.europa.eu/finance/consultations/2015/financial-regulatory-framework-review/index_en.htm
104
 kom (2016) 0853 - Ingen titel
1715591_0105.png
Annex 3.1. Calculation of derivative exposures in the counterparty credit risk
framework
Problem definition
Under the current CRR institutions have the choice to use among three different standardised
approaches for calculating the exposure value of derivative transactions under the counterparty
credit risk framework: the Standardised Method ('SM'), the Mark-to-Market Method ('MtM
method') and the Original Exposure Method ('OEM') (see Articles 276 to 282, Article 274 and
Article 275 of the CRR, respectively). These approaches are also used in other areas of the CRR
that need to measure the exposure value of derivative transactions. Among the own fund
requirements, the MtM method, the SM or OEM can also be used in the own fund requirements
for CVA risks and the own fund requirements for trade exposures to CCP. Otherwise the use of
these standardised approaches is allowed in the large exposure framework while the leverage
ratio and the own fund requirements for default funds exposures to CCP impose the MtM
method.
The proportion of own fund requirements under CRR calculated with one of the standardised
approaches for derivative exposures is generally small: less than 5.28% of the total own fund
requirements for 75% of large EU institutions and less than 0.06% of the total own fund
requirements for 75% of small EU institutions. For large EU institutions, this proportion varies
materially depending on the business models of the institution while this proportion remains
relatively low for all EU institutions with small trading activities irrespective of their business
models.
Table A6.
Materiality of the own fund requirements for Counterparty Credit risk and CVA risk
for a representative sample of European institutions
Sample of EU institutions with small trading
activities
115
Min
25% percentile
Median
75% percentile
Max
0.00%
0.00%
0.01%
0.06%
18.21%
Sample of large EU institutions
114
Min
25% percentile
Median
75% percentile
Max
0.00%
0.85%
2.17%
5.28%
55.75%
Source: EBA report on SACCR and FRTB implementation, November 2016
Table A7.
Materiality of the own fund requirements for Counterparty Credit risk and CVA risk
for a representative sample of European institutions depending on their business models
114
115
This sample consists of 193 large European institutions that the EBA receives COREP and FINREP
reporting from.
This sample consists of 1094 European institutions with a presumption of small trading activities as
identified by the EBA as the institutions with less than euros 500 million of fair valued assets and
liabilities.
105
 kom (2016) 0853 - Ingen titel
1715591_0106.png
Large EU institutions
Number
Auto & cons.
CCP
Co-operatives
Custodien inst.
Div. no retail dep.
Local Universal
Mrtg. & Build.Soc.
Other
Other no retail dep.
Pass-through
Savings
Sec. trading house
Univ. Cross-Border
Unclassified
TOTAL
5
2
15
3
14
52
13
2
8
1
10
4
33
31
193
Average
3.11%
8.08%
6.46%
1.41%
0.05%
4.37%
12.19%
0.19%
1.95%
42.52%
1.30%
5.63%
4.49%
8.12%
5.55%
EU institutions with small trading activities
Number
16
0
505
3
3
210
26
28
12
0
144
6
13
47
1013
Average
0.06%
-
0.28%
0.42%
0.33%
0.29%
0.56%
1.38%
0.64%
-
0.16%
0.35%
0.31%
0.34%
0.31%
Source: EBA report on SACCR and FRTB implementation, November 2016
The EBA identified only 6 EU institutions currently using SM, 372 EU institutions
116
using the
OEM. Knowing that the EBA also identified 20 EU institutions currently permitted to use the
Internal model method ('IMM') - the alternative model approach to the standardised approaches -
all the other EU institutions subject to own funds requirement for counterparty credit risk –
potentially few thousands - are supposed to currently use the MtM method.
116
The EBA launched an ad-hoc survey to identify all the EU institutions that currently use OEM.
Therefore, the 372 EU institutions identified as currently using OEM come from a broader population
than the 1094 European institutions with a presumption of small trading activities (for this population
106
 kom (2016) 0853 - Ingen titel
1715591_0107.png
Table A8.
Number EU institutions currently using OEM and SM per jurisdictions
Member State Institutions using OEM
AT
BE
BG
CZ
CY
DE
DK
EE
EL
ES
FI
FR
HR
HU
IE
IT
LT
LU
LV
MT
NL
PL
PT
RO
SE
SI
SK
UK
TOTAL
176
1
0
0
NA
117
0
0
0
4
NA
5
26
3
0
1
0
27
1
NA
NA
NA
2
NA
NA
1
NA
8
372
Institutions using
SM
NA
0
5
0
NA
0
0
NA
NA
0
NA
0
0
NA
NA
NA
0
NA
NA
NA
NA
NA
0
NA
1
0
NA
0
6
Source: EBA report on SACCR and FRTB implementation, November 2016. 'NA' means that the jurisdictions did not provide the
relevant answer to the EBA during the survey
.
The need for an early review without evaluation
The MtM method and the SM have been criticised
117
for several limitations, mainly: they do not
recognise appropriately the risk-reducing nature of collateral in the exposures (an issue in light of
the forthcoming international clearing/margin obligations); their calibrations are outdated and do
not reflect the high level of volatility observed during the financial crisis; they do not recognise
appropriately netting benefits. While at this stage we do not have the relevant data to quantify the
inefficiencies of that result from the current approach, it has to be noted that, in March 2014, the
Basel Committee for Banking Supervision (BCBS) adopted a new standardised methodology to
compute banks' derivatives exposures in the Basel framework – the Standardised Approach for
Counterparty Credit Risk ('SA-CCR'). It agreed that SA-CCR would replace the two existing
methodologies allowed in the Basel framework (the Current Exposure Method (CEM) and the
117
See Section B in http://www.bis.org/publ/bcbs254.pdf
107
 kom (2016) 0853 - Ingen titel
Standardised Method (SM)) for computing banks' derivatives exposures in Basel counterparty
credit risk framework from 01 January 2017.
To better capture the exposure value of derivative transactions under the counterparty credit risk
framework, and to comply with international agreed standards, SA-CCR should be introduced in
the EU.
Proposed solution
Under the proposed amendment, institutions would use the SA-CCR in the counterparty credit
risk framework while, under revised conditions, institutions with small trading activities would
have the possibility to use a revised version of OEM. A simplified version of SA-CCR will also
be available for banks that would face some operational difficulty to implement SA-CCR but
have sizeable derivative activities that would not warrant them the use of the revised OEM.
SA-CCR would provide institutions not permitted to use an internal model approach for
calculating the exposures of derivative transactions (the vast majority of EU institutions since
only 20 institutions have been permitting to use such a model according to a survey performed by
the EBA) with a more risk-sensitive approach than the current ones, calibrated to stress
conditions and differentiating between collateralised and uncollateralised derivative transactions.
Under this option, SA-CCR would be implemented under CRR without any material deviations
from the Basel rules.
However, it is clear from the responses to the consultation paper that the implementation of SA-
CCR would be too challenging for banks with small trading activities that currently use OEM or
the MtM method. The responses are split whether OEM, the MtM method or a simplified version
of SA-CCR should be used for those banks. However, it is also clear that some of the features of
OEM and the MtM method are too different from the features SA-CCR, leading to different
exposures amount for the same transactions, which would create an unlevel playing field between
the institutions applying SA-CCR and institutions that would be allowed to use these approaches
if they were kept alongside SA-CCR.
Based on the above evidence and to maintain the level playing field across all institutions not
permitted to use an internal model for calculating derivative exposures, only one simple
alternative to SA-CCR will be maintained under CRR – OEM – with revised assumptions to
ensure its consistency with SA-CCR (the revised assumptions have been designed to limit
additional undue complexity for the targeted institutions).
In addition, new eligibility criteria would be set to permit institutions to use the revised OEM,
based on the size of the market values of their gross derivative activities for trading purposes. A
combination of absolute and relative threshold may be maintained to ensure that only institutions
with small derivative portfolios, as compared to their entire balance sheet, would be eligible for
the application of the revised OEM.
The EBA has assessed different level of thresholds based on the new eligibility criteria for the
application of the revised OEM based on a small sub-sample of the EU institutions with small
trading activities (the sample has been reduced to 134 EU institutions due to lack of data
available). The majority of EU institutions in this sub-sample have a size of the market values of
their gross derivative activities for trading purposes below euros 20millions and a relative size of
the market values of their gross derivative activities for trading purposes to total assets below 5%.
The remaining few EU institutions in this sub-sample have a size of the market values of their
gross derivative activities between euros 20 and 300 million and a relative size of the market
108
 kom (2016) 0853 - Ingen titel
values of their gross derivative activities for trading purposes to total assets between 5% and
15%.
In order to allow enough leeway to capture those institutions that would face some operational
difficulty to implement SA-CCR, it would be preferable not to set the level of the absolute and
relative thresholds too low. Based on the EBA data, we propose that the new eligibility criteria
for the application of the simplified SA-CCR would be based on an absolute threshold of EUR
150 million and a relative threshold of 10%.
109
 kom (2016) 0853 - Ingen titel
1715591_0110.png
Annex 3.2. Disclosure
Problem definition
The specific policy objective is for institutions to provide meaningful, consistent disclosures of
prudential information at a reasonable cost. These disclosures complement the mandatory
solvency and liquidity requirements and the supervisory review process and are in combination
the bedrock for safe-guarding the financial stability of institutions established in the Union.
Under the current CRR institutions have to disclose information to allow users (investors and
other stakeholders) to form a view on the risk profile of the institution and exercise market
discipline. Whilst overall the substantial disclosure requirements of the CRR can be considered
sufficient, the lack of harmonised disclosure formats hampers the comparability of disclosures
between institutions and over time thereby reducing market discipline. Moreover, the existing
disclosure requirements are mainly a "one size fits" allowing for hardly and differentiation based
on the size of the institution and are therefore not optimally proportionate.
Maintaining the status quo of Part Eight "Disclosure" of the CRR would imply that the disclosure
requirements 1) would have very little proportionality thereby neglecting the claims for more
proportionate disclosure requirements made during the call for evidence, 2) would not use the full
potential of disclosures by facilitating efficient comparability and 3) create divergence from the
revised Basel disclosure requirements in an area where the EU is currently fully compliant
The Basel Committee adopted in January 2015 revised Pillar 3 disclosure requirements for
financial years starting on or after 2016
118
requiring common formats for any disclosure in
relation to the Basel mandatory ("Pillar 1") requirements.
The need for an early review without evaluation
The current CRR disclosure requirements were applicable from 1 January 2014 onwards. The
EBA issued a report on Pillar 3 disclosures by banks for the year 2014 in 2015 which included a
synthetic overview of the missing CRR disclosures compared to the revised Basel Pillar 3
disclosure framework. In order to be aligned with the revised Basel international requirements,
the CRR should be amended.
In addition, disproportionality of disclosure requirements was mentioned by many respondents to
the call for evidence indicating the unnecessary administrative burden for smaller banks. In light
of this Commission's better regulation agenda, these call for proportionality should be addressed
before the evaluation of the CRD IV/CRR.
Proposed solution
In order to alleviate the current disproportionate operational burden and to be aligned with the
revised Basel Pillar 3 disclosure framework institutions should be categorised on the basis of
their significance. "Significant institutions" would be defined along the lines of the SSM
Regulation criteria for identifying significant banks and "small institutions" would be defined on
118
http://www.bis.org/bcbs/publ/d309.htm
110
 kom (2016) 0853 - Ingen titel
the basis of total asset size. A further proportionality criterion would be whether the institution
has issued securities listed on an EU regulated market or not.
Institutions would either be significant, small or "other" with or without being "listed". The
disclosure requirements would be a sliding scale with differentiations in the substance and
frequency of disclosures whereby for all types of institutions disclosure templates developed by
the EBA would be mandatory.
At the upper end of the sliding scale would be significant institutions that would be required to
quarterly disclosure of approximately 1 to 2 pages key metric tables of prudential information,
semi-annually disclosure of key metrics plus some selected more substantial disclosures and a
fully-fledged annual disclosure small banks with no securities listed would be required to
disclose only annual key-metrics.
111
 kom (2016) 0853 - Ingen titel
Annex 3.3. Supervisory reporting
Problem definition
Reporting prudential information by institutions to supervisors is an essential prerequisite for
effective ongoing supervision and monitoring of risks. The CRR constitutes a single rulebook for
supervisory reporting whereby EBA develops Implementing Technical Standards (ITSs) for
supervisory reporting with common data definitions, common templates, common reporting
frequencies and common remittance dates as well as a common IT solution.
The Commission adopts the ITSs prepared by EBA as implementing regulations. So currently the
Union has a legally enforceable common supervisory reporting system applicable to any
institution established within the EU. The single rulebook of supervisory reporting in
combination with the underlying notion of maximum harmonisation implies that supervisors
cannot impose additional systematic reporting requirements on institutions. However, supervisors
have the power to request ad hoc information from individual institutions which is one the
minimum supervisory powers laid down in the CRD IV.
Although the CRR mandate on supervisory reporting specifically mentions that supervisory
reporting shall be proportionate to the scale, nature and complexity of the activities of the
institutions and where there is "implicit proportionality" in the sense that if an institution has a
simple business model it only has to report a fraction of the data points from the supervisory
reporting package, several claims have been made during the call for evidence that supervisory
reporting in the EU has become disproportionate.
Respondents to the call for evidence highlighted in particular the high administrative
burden caused by 1) disproportionate reporting requirements generally and for smaller
banks in particular in terms of content and reporting frequency and 2) supervisors
requiring additional reporting on top of the regular EU reporting requirements. Some
respondent also expressed strong concerns about further disproportionate reporting
requirements based on forthcoming initiatives such as the ECB's AnaCredit and the ECB
European Reporting Framework.
So based on the call for evidence there seems to be two main sources of potential
disproportionality in the area of supervisory reporting:
1. EBA is not considering proportionality optimally when developing ITSs on
supervisory reporting; (Note: this includes some detailed "level 1" reporting
requirements in the CRR that do not fulfil a clear supervisory purpose);
2. Supervisors requesting systematic reporting of prudential information on top of
the EU agreed supervisory reporting package; (Note: this is partly driven by the
Commission's non-timely adoption of ITSs creating a misalignment between the
applicable level 1 prudential requirements and the reporting requirements or
Commission decisions to reject certain reporting requirements that EBA decided
necessary for effective supervision).
The need for an early review without evaluation
Disproportionality of reporting requirements was invoked by many respondents to the call for
evidence as a cause of unnecessary administrative burden for smaller banks. In light of this and
taking into account the Commission's better regulation agenda, these calls for proportionality
should be addressed before an evaluation of the CRD IV/CRR can be undertaken.
112
 kom (2016) 0853 - Ingen titel
1715591_0113.png
Proposed solution
Non optimal proportionality in level 1 and level 2 legislation
As to the first source of disproportionality it is not straightforward for EBA to determine
the appropriate trade-off between the cost of reporting relevant prudential data by
institutions and the benefits for effective supervision. EBA makes this trade-off inter alia
via public consultations of draft ITSs. However, since proportionality is a qualitative
concept open to different views on the cost-benefit, the trade-off may not always be right
depending on the stakeholder perspective.
In particular the frequency of reporting for smaller institutions could be reduced leading
to less reporting burden without undermining overall the supervisory effectiveness or
financial stability risk. In order to achieve reduced frequency for smaller institutions the
proportionality concept in the CRR it is proposed to better frame in the CRD IV and
CRR the proportionality mandate of EBA when developing ITSs.
Additionally the extant body of reporting requirements should be reduced for some or all
institutions depending on their size or other quantitative criteria. This can be achieved by
including in the CRR a specific requirement for EBA to report to the Commission on
concrete proposals for reducing the current supervisory reporting package without
sacrificing supervisory effectiveness or directly in the level 1 text.
As regards too detailed level 1 reporting requirements, the CRR review proposal should
include amendments to delete or reduce some specific reporting requirements in the level
1 text. In particular those for which supervisory experience has shown they are dis-
proportionate but which EBA has to include in the ITSs since they are specified in the
level 1 text.
Supervisors requesting additional systematic reporting of prudential information
In order to address this supervisory behaviour, the CRR review proposal should include a
mandate for an independent study of any such systematic additional reporting
requirements that would infringe on the single rulebook. On the basis of the conclusion
drawn by the study the Commission would consider whether the additional reporting
imposed by supervisors is infringing on the single rulebook and take corrective actions if
deemed necessary.
In addition, CRD IV rules entrusting supervisors with supervisory reporting powers
should limit those powers to
ad hoc
reporting by individual institutions (thus eliminating
the possibility for supervisors to impose systematic reporting by all or a subset of
institutions).
113
 kom (2016) 0853 - Ingen titel
1715591_0114.png
Annex 3.4. Pillar 2 additional capital
Problem definition
Pillar 2 capital requirements are additional capital requirements that supervisors may impose on
individual banks in excess of Pillar 1 capital requirements (i.e. "minimum" requirements
applicable to all banks) and the combined buffers requirement (i.e. the combination of various
buffer requirements related to certain macro-prudential risks applicable to all banks or a subset of
banks). According to CRD IV
119
, a bank that doesn't meet Pillar 1 or Pillar 2 capital requirements
may lose its license, whilst the consequence of breaching the combined buffer requirement is the
automatic restriction of dividend payments, bonus pay-outs and the remuneration of Additional
Tier 1 (AT1) instruments to a certain share of the bank’s profits (i.e. Maximum Distributable
Amount – MDA)
120
.
The current text of the CRD IV sets the broad parameters of the exercise of Pillar 2 powers,
whilst leaving to supervisory authorities a wide margin of discretion when exercising their
powers.
The need for an early review without evaluation
Input received from industry and supervisory authorities during the 2015 public consultation on
the CRR/CRD IV review, the Call for evidence and from bilateral contacts between the
Commission and various parties concerned hinted to some discrepancies and weaknesses in the
way Pillar 2 capital requirements are applied across jurisdictions and to the sometimes not
transparent way supervisors' decisions on the additional capital imposed on individual banks are
made. This is due to the ambiguities generated by the legal text as currently drafted. The way
Pillar 2 capital add-ons are defined and calculated are an important driver of an institution’s
overall level of capitalisation and are relevant for market participants since the level of additional
capital imposed by supervisors as a Pillar 2 measure may impact on the triggering of restrictions
of dividend payments, bonus pay-outs and the remuneration of AT1 instruments (MDA).
Despite the lack of sufficient data for a full evaluation, a clarification of the current rules is thus
needed to ensure the proper functioning of the market, especially for those financial instruments
directly linked to the automatic restrictions of distributions (e.g. AT1). Moreover, with regards to
the interest rate risks for banking book positions, the modification of the current text is justified
by the shortcomings identified at international level and the solutions developed in the standard
adopted by the BCBS in April 2016, to which the proposal seek to align.
Solution proposed
The relevant articles of the CRD IV and CRR will be modified to clarify: the relation between
Pillar 1, Pillar 2 and buffer capital requirements (so called "stacking order" of capital
requirements); the distinction between Pillar 1 (applicable to all banks) and Pillar 2 (bank
specific) capital requirements; the difference between Pillar 2 capital requirements (to be met by
the bank at all time and subject to public disclosure) and Pillar 2 capital guidance (which implies
an expectation that the institution have additional capital beyond mandatory capital
requirements); the fact that the MDA shall be calculated by taking into account Pillar 1, Pillar 2
119
Article 18(1)(d) of CRD.
120
Article 141 of CRD.
114
 kom (2016) 0853 - Ingen titel
and buffers capital requirements (but not Pillar 2 capital guidance) and that the AT1 instruments
should be given priority if as a result of the MDA calculation distributions have to be limited. In
addition, the framework for capturing interest rate risks for banking book positions under Pillar 2
measures will be included.
These proposed amendments are expected to promote consistency in the application of rules,
improve transparency and legal certainty on the use of Pillar 2 capital instruments.
Impact of the proposed solution
The modifications of the CRD IV and CRR proposed are not expected to impact on the total
amount of capital hold by credit institutions or on their ability to lend. As a consequence of the
clarifications proposed credit institutions are likely to meet the different capital requirements or
capital expectations by reallocating the capital they have.
115
 kom (2016) 0853 - Ingen titel
1715591_0116.png
Annex 3.5. Equity investments into funds
Problem definition
The CRR contains specific rules governing capital requirements for banks' banking book
exposures in the form of units or shares in collective investment undertakings (CIUs) – basically
undertakings for collective investment in transferable securities (UCITS) and alternative
investment funds (AIFs).
There is a separate set of rules for banks applying the standardised approach (SA) for credit risk
on the one hand and those applying the internal ratings-based (IRB) approach on the other hand,
both with different methods for calculating risk weights.
Both SA and IRB banks may apply a "look-through approach", whereby they look through to a
CIU's underlying exposure in order to calculate an average risk weight for their exposures in the
CIU; this is the most risk-sensitive and transparent approach. The other, less risk-sensitive
methods differ for SA and IRB banks and tend to assign risk weights based on crude criteria. For
SA banks, the look-through approach is optional, whereas IRB banks must use it if certain
criteria are met.
The abovementioned rules are the EU implementation of the internationally agreed standards
published by the Basel Committee as currently applicable.
[1]
During the crisis, concerns were raised regarding the oversight and regulation of "shadow
banking" entities and activities, as well as their indirect regulation through banking regulation. In
this context, the Financial Stability Board recommended
[2]
in 2011 that "the risk-based capital
requirements for banks’ exposures to shadow banking entities should be reviewed to ensure that
such risks are adequately captured", specifically referring to the treatment of investments in
funds. Such review was conducted by the Basel Committee, resulting in a new standard published
in December 2013, but not yet implemented in EU legislation.
The concerns raised with respect to the current framework notably relate to risk sensitivity and
transparency. The framework lacks risk sensitivity notably in the sense that it does not require
banks to reflect a fund's leverage when determining capital requirements associated with their
investment, even though leverage is a very important risk driver. This creates undesirable
incentives by encouraging investments in higher-risk funds and may result in an insufficient
capitalisation of such higher-risk exposures.
Also, the framework does not promote transparency and appropriate risk management of the
relevant exposures, as there is no clear rank ordering between the different approaches, with
different degrees of prescriptiveness for SA banks compared to IRB banks and insufficient
incentives to apply the look-through approach.
The need for an early review without evaluation
The Commission proposes to adopt the main aspects of the new Basel standards, which would
help address a number of the aforementioned weaknesses of the current rules while allowing us
to comply with our international obligations. However, at this stage, we do not have the relevant
data to quantify the shortcomings of the current EU approach.
[1]
Basel Committee on Banking Supoervision, "International Convergence of Capital Measurement and
Capital Standards", 2006
[2]
Financial Stability Board, "Shadow Banking: Strengthening Oversight and Regulation", 2011
116
 kom (2016) 0853 - Ingen titel
Proposed solution
The proposed framework consists of three approaches, which would apply to both SA and IRB
banks' exposures. The look-through approach (LTA) requires banks to risk weight the fund's
underlying exposures as if they were held directly; the mandate-based approach (MBA) assumes
that the underlying portfolios are invested to the maximum extent allowed (as per the mandate,
regulations, or other disclosures) in the assets attracting the highest risk weights; and the fall-
back approach (FBA) – used for funds with insufficient transparency – requires the application of
a 1,250% risk weight. It provides a hierarchy of approaches as a function of the degree of due
diligence performed by banks, with an appropriate incentive structure, whereby the degree of
conservatism increases with each successive approach as risk sensitivity and transparency
decrease. This promotes appropriate risk management of bank exposures to funds by providing
incentives to use the more risk sensitive and transparent approaches.
A leverage adjustment is added, whereby banks using the LTA or the MBA must adjust the
average risk weight for an equity investment upwards by the fund's leverage (LTA) or permitted
leverage (MBA). This will allow an improved reflection of the actual risks faced by the banks as
concerns of double gearing are addressed.
According to Basel Committee data, banks' equity investment in funds do not appear to be
material exposures in most jurisdictions, as risk weighted assets arising from these investments
represent less than 2 per cent of total RWAs in most jurisdictions, even though a considerable
degree of heterogeneity across jurisdictions is observed.
117
 kom (2016) 0853 - Ingen titel
1715591_0118.png
Annex 3.6. Bank financing of infrastructure projects
Problem Definition
One of the goals of the Capital Markets Union is to help mobilise capital in Europe and channel it
to the infrastructure and long term sustainable projects that Europe needs to create jobs. The
European Investment Bank estimates that the EU may need up to €2 trillion in investment in the
period up to 2020. Public support through measures such as the €315 billion Investment Plan for
Europe (IP/15/5420) will help, but there is a need for more private investment in such projects in
the longer term.
Despite the growing role of large institutional investors in providing long-term funding for
infrastructure investments, banks continue to play an important role and being the most relevant
source of funding of infrastructure projects in the EU. In 2014 the
proportion of the value of
infrastructure financed through bank debt in the EU of the total value of infrastructure deals was
equal to 65.9% (it was 82,7% in 2014).
In absolute terms, bank lending for infrastructure has grown markedly from 2009 to 2014, almost
reaching the pre-crisis peak of 2006.
Figure A6.
Proportion of infrastructure finance lent by banks in total volume of infrastructure
funding
100
90
80
70
60
50
CAPEX in € bn
40
30
20
10
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Total
Financed by Bank Debt
Source: InfraDeals and Infrata calculations
118
 kom (2016) 0853 - Ingen titel
1715591_0119.png
Figure A7.
Total value of EU Infrastructure projects for which banks provided financing – 2000-
2014
Source: InfraDeals and Infrata calculations
The need for an early review without evaluation
Although the limited time elapsed since the entry into force of CRD IV/CRR didn't allow an in-
depth evaluation of rules applicable to specialised lending, several replies in the call for evidence
suggest the need for a more risk-sensitive approach to the credit risk attached to infrastructure
projects. These calls highlight that in order to boost long-term funding for infrastructure
investment to respond to the needs of the EU economy is therefore important, besides promoting
the role of non-bank investors, to make infrastructure investments, in particular high-quality
ones, more attractive to banks and allow banks to better understand and manage risks attached
infrastructure projects.
Proposed solution
A specific 'population' of specialised lending exposures will be identified which aim at funding
infrastructure projects and fulfil certain criteria able at reducing the different risks a bank would
incur in providing such funding (financial, political, legal, operating, etc.). This new asset class of
qualifying specialised lending exposures would benefit from a discount factor of 25% The
criteria will denote safer infrastructure projects and ensure that lending banks understand the
associated risks.
These criteria, largely derived from those used for the Category I exposures in the so-called
'slotting approach' in the IRBA (see draft EBA RTS on specialised lending exposures
121
), would
119
 kom (2016) 0853 - Ingen titel
1715591_0120.png
be consistent with the criteria developed in the insurance framework for the prudential treatment
of qualifying infrastructure investments (Commission Delegated Regulation 2016/467).
The proposed amendment, while promoting high-quality, sustainable infrastructure investments,
would enhance cross-sectoral harmonization and comparability between SA and IRBA banks
122
.
121
Draft RTS on Assigning Risk Weights to Specialised Lending Exposures under Article 153(9) of
Regulation (EU) No 575/2013 (Capital Requirements Regulation – CRR)
Currently the slotting approach is used by only 23% of IRBA Banks but the BCBS is currently
considering requiring in future all IRBA banks to apply the slotting approach for specialised lending
exposures.
120
122
 kom (2016) 0853 - Ingen titel
1715591_0121.png
Annex 3.7. Large exposure framework (alignment with Basel rules)
Problem definition
The purpose of the large exposure limits is to protect banks from significant losses caused by the
sudden default of an individual counterparty or a group of connected counterparties. It thus
targets exposures that are large compared to a bank’s capital resources. The current text set a
general limit to large exposures of 25% of institutions' eligible capital
123
(which is the sum of
Tier 1 capital and an amount of Tier 2 capital equal to one third of Tier 1 capital
124
).
The general limit of 25% is not sufficiently prudent, especially for larger banks, since it only
capture a small part of the overall large exposures that European institutions have. In fact, the
25% limit addresses only a limited number of the exposures. Moreover, it results in a higher limit
for smaller banks since larger banks have usually more Tier 2 capital than smaller ones. This
doesn’t ensure that the maximum possible loss a bank could incur if a single counterparty or a
group of counterparties were to suddenly fail would not endanger the bank’s survival as a going
concern.
Moreover, the current limit doesn’t take into account the higher risks carried by the exposures
that globally systemically important Banks (G-SIBs) have to single counterparty or groups of
connected clients and, in particular, as regards exposures to other G-SIBs. The financial crisis
has, in fact, demonstrated that material losses in one systemically important institution (SIFIs)
can trigger concerns about the solvency of other SIFIs with potentially serious consequences on
financial stability.
Finally, the BCBS has developed in 2014 a new methodology (i.e. Standardised Approach for
Counterparty Credit Risk, SA-CCR) for computing banks’ derivatives exposure (i.e. Over The
Counter, OTCs) that better capture the risks carried by this type of exposures. The current large
exposures framework relies instead on less accurate methods, which could lead to underestimate
the risks linked to derivatives exposures.
The need for an early review without evaluation
At international level the BCBS identified some shortcomings in the large exposure regime that
the BCBS standard, published in 2014 and expected to be implemented by jurisdictions by 2019,
aims to address. The standard has been published for 2 years and market participants, which have
participated in the public consultation
125
and the quantitative impact assessment (QIS)
126
launched by the Basel committee, expect the EU system to be aligned with the standard. It is thus
proposed to modify the CRR to reflect the Basel framework.
Solution proposed
The measures proposed to address the loopholes identified in the current large exposures
framework are essentially three. First, increasing the quality of capital that can be taken into
account for limiting large exposures, by limiting the eligible capital only to Tier 1 capital (no
123
Art. 395 of CRR.
124
This is the definition of eligible capital that applies as from 2016, after the transitional period set out in
Article 494 of the CRR has expired.
125
http://www.bis.org/publ/bcbs246.htm.
126
http://www.bis.org/publ/bcbs246/instructions.pdf
121
 kom (2016) 0853 - Ingen titel
1715591_0122.png
more Tier 2 capital). The higher quality of capital used as capital base will improve the ability of
institutions to absorb losses. At the same time, the change in the capital base will reduce the
quantity of exposures that a bank can have - and thus the risk of losses in case of default of the
counterparty – and introduce a more proportionate system for smaller banks compared to larger
ones. In the same direction, a second proposal is to introduce a lower limit for G-SIBs exposures
to other G-SIBs (15% of banks’ Tier 1 capital instead of the 25% of banks’ Tier 1 capital
required for other banks) in order to reduce systemic risks related to the interlink among large
institutions and the probability that the default of G-SIBs counterparty may have on financial
stability. Finally, it is proposed to impose to use the SA-CCR methods for determining exposures
to OTC derivative transaction, even for banks that have been authorised to use internal models.
These interventions will overall increase the risk-sensitivity of the large exposures regime and
better tailor the requirements to specific types of exposures and to the size of banks. At the same
time, the modifications introduced in the current framework will align the European system to the
BCBS standard on large exposures issued in 2014
127
, thus increasing international comparability
and consistency across jurisdictions.
Impact of the proposed solutions
The enhanced quality of capital (only to Tier 1 capital taken into account as capital base) is not
expected to have a significant impact on the number of the exposures exceeding the large
exposure limit. According to the available data the total number of exposures in breach of the
25% limit of the capital base increases from 25 (considering eligible capital) to 63 (considering
Tier 1 capital).
Table A9.
Changes in the number of large exposures due to the enhanced quality of capital
Eligible capital
Exposure bucket
≤ 10%
> 10% ≤ 15%
> 15% ≤ 20%
> 20% ≤ 25%
> 25%
Total
Total
14 096
384
167
62
25
14 734
Group 1
10 885
130
56
13
13
11 097
Group 2
3 211
254
111
49
12
3 637
Total
13 953
417
210
91
63
14 734
Tier 1 capital
Group 1
10 800
183
63
32
19
11 097
Group 2
3 153
234
147
59
44
3 637
Source: EBA (Group 1 and Group 2 institutions differentiate between respectively big and less big banks. The smallest
bank in Group 2 has total assets of EUR 5 billion
There would be 11 additional credit institutions out of 198 credit institutions analysed which
would exceed the large exposures limit when the capital base changes from eligible capital to
Tier 1 (of which 4 are Group 1 and 7 Group 2 institutions). This would imply that 6 Member
States that would not have reported exposures above the large exposure limit (25% or 15%)
would be affected by the change in capital base (namely, ES, AT, LUX, IE, DE and SE).
127
BCBS,
Supervisory framework for measuring and controlling large exposures,
April 2014, available at
http://www.bis.org/publ/bcbs283.pdf.
122
 kom (2016) 0853 - Ingen titel
1715591_0123.png
Figure A8.
Changes in the number of large exposures per country due to the enhanced quality of
capital
Source: EBA (Group 1 and Group 2 institutions differentiate between respectively big and less big banks. The smallest
bank in Group 2 has total assets of EUR 5 billion)
The modification of the large exposure limit for exposures between G-SIBs (from 25% to 15%)
would in practice have no impact, since there are no G-SIBs in the EU that have reported
exposures greater than 10% both in terms of eligible capital and Tier 1 capital. The imposition of
a 15% limit will however prevent in the future G-SIBs to increase the number of exposures
towards other G-SIBs.
Table A10.
Changes in the number of large exposures due to the change of limit for G-SIBs v G-
SIBs exposures
Eligible capital
Exposure bucket
> 0% ≤ 5%
> 5% ≤ 10%
Source: EBA
Tier 1 capital
Number of
Institutions
13
8
Number of
Exposure
186
12
Number of
Institutions
13
4
Number of
Exposure
193
5
As concerns the use of the
standardised approach for measuring exposure at default for
counterparty credit risk
(SA-CRR), it was not possible to gather information concerning the
impact of this change on the large exposure regime. A data collection on simulated data would
need to be conducted given that the SA-CRR is not yet implemented in the CRR.
Overall, the modifications to the large exposure framework are however not expected to impact
on the ability of credit institutions to lend since the large exposure regime only impose a
diversification of clients to which credit institutions' have exposures. The number of clients or of
exposures are not limited.
123
 kom (2016) 0853 - Ingen titel
Annex 3.8. Exemptions on large exposures
Problem definition
Article 400 (2) of the CRR lists a number of exposures that competent authorities may fully or
partially exempt from the scope of application of the large exposures limit. These exposures can
only be exempted if the conditions laid down in paragraph 3 of the same article are met.
By way of derogation from Article 400 (2) and (3) of CRR, Article 493(3) of CRR provides for a
temporary possibility for Member States to grant an exemption from the large exposures limit for
the same exposures listed in Article 400 (2) of CRR, however without having to meet the
conditions set out in paragraph 3 of Article 400 of CRR.
The concurrent possibility of Members States and competent authorities of granting exemptions
to the same exposures has proved to be problematic after the introduction of the Single
Supervisory Mechanism (SSM).
Since November 2014 the SSM (and not anymore national supervisors) has become the
competent authority for significant institutions established in the banking union. It is therefore the
SSM which has to decide whether one of the exposures listed in Article 400 (2) of the CRR
should be partially or fully exempted from the large exposures limit for significant institutions.
The fact that Article 493(3) of CRR entrusts Member States with the same power interferes with
the ability of the SSM to perform its tasks in a consistent and coherent manner. In fact, the same
exposure may or may not be exempted from the large exposure limit depending on whether the
Member State where the significant institution is established has exercised the option set out in
Article 493(3) of CRR or not.
The need for an early review without evaluation
Article 493(3) of CRR entered into force in 2013 and was elaborated when supervisory
authorities were only national supervisors. The adoption of the SSM regulation (Regulation EU
No 468/2014) and the start of the functioning of the SSM revealed the shortcomings created by
the use of Article 493(3) of CRR by Member States. Moreover, Article 493(3) of CRR is a
transitional provision and its modification is expected.
Solution proposed
We would propose to end the transitional period allowing Member States to grant exemptions for
certain exposures to the large exposure limit set out in Article 493(3) of CRR.
The modification will allow for a more coherent application of large exposures rules, foster
harmonisation across Member States and promote a level playing field among significant banks
established in the banking union.
At the same time, ending the transitional period is not expected to have negative impacts on the
EU system since competent authorities – including national supervisors for banks not falling
under the supervisory competence of the SSM – will still be able to grant exemptions to the same
types of exposures according to Article 400 (2) and (3) of CRR.
Ending the transitional period for exemptions to the large exposure regime would also be
more prudent since exemptions to the large exposures limit could only be allowed when
the conditions of paragraph 3 of Article 400 of the CRR are met. Finally, ending the
124
 kom (2016) 0853 - Ingen titel
Member States transitional period could also enhance the further integration of the single
market in banking services.
Impact of the proposed solutions
The proposed measure is not expected to have any impact on credit institutions (including their
lending capacity) since they will not be deprived of the possibility of being exempted from the
large exposure limit. Competent authorities will still be able to exempt form the application of the
large exposures limit the same exposures set out in Article 493(3) of CRR according to Article
400 (2) and (3) of CRR.
125
 kom (2016) 0853 - Ingen titel
1715591_0126.png
Annex 3.9. Rules on exposures to CCPs
Problem definition
The CRR introduced specific rules on capital requirements for institutions' exposures to central
counterparties (CCPs). The introduction of these rules represented an important change in terms
of the measurement, monitoring and management of such exposures as they had previously
attracted no capital requirements.
In a nutshell, the CRR contains capital requirements for two types of exposures to CCPs: trade
exposures and exposures due to default fund contributions. The size of the requirement depends
on whether a CCP is labelled as a 'qualifying' (QCCP) or not (non-QCCP). Requirements for the
former are lower than requirements for the latter (in fact, the requirements for exposures to non-
QCCPs were deliberately designed to be penal to disincentivise institutions from using them). In
order for a CCP to be considered a QCCP, it has to be either authorised (for CCPs established in
the EU) or recognised (for CCPs established outside the EU) in accordance with EU rules.
In order to achieve a certain degree of risk sensitivity in the level of capital requirements for
exposures due to default fund contributions, the CRR sets out a method ('Method 1') that
compares a “hypothetical” level of resources that a QCCP should have in order to cover potential
losses resulting from the default of one or more of its members with the actual resources the
QCCP has at its disposal. The capital requirement depends on the difference between those two
amounts (the requirement is higher if the "hypothetical" resources exceed the actual resources
than in the opposite case). The CRR also contains an alternative method for calculating the
capital requirements for such exposures, which depends solely on the size of the exposures
('Method 2'). Institutions are free to choose which of the two methods to apply.
The abovementioned rules are the EU implementation of the internationally agreed interim
standards published by the Basel Committee.
128
Under the CRR, Method 1 relies on the application of the Mark-to-Market Method (MtMM)
when calculating the "hypothetical" resources in relation to derivatives exposures.
129
One concern
that was expressed in relation to the use of MtMM in that context was that, given that the MtMM
was designed for simpler and more directional derivatives positions, it was not suitable for the
centrally cleared space. This is because CCPs have, by definition, balanced positions (i.e. the
amounts owed by the CCP to its members and the amounts the CCP's members owe to the CCP
perfectly offset each other)
130
and clear also more complex derivatives. Impact studies carried out
at international level found that calculating the hypothetical level of resources using the MtMM –
combined with the nature of the formula for determining the capital charge – meant that capital
requirements on member contributions to default funds varied significantly between QCCPs: in
many cases the charges were very small, and in some cases they were very large. That degree
variation could not be explained solely by differences in the risk profiles of the different QCCPs.
In other words, the results showed that the method did not capture risks sufficiently well, i.e.
either leading to too low or too high requirements.
128
Capital requirements for bank exposures to central counterparties, July 2012. Available at
http://www.bis.org/publ/bcbs227.pdf.
129
Within the Basel framework it is known as the Current Exposure Method (CEM).
130
This balance can be disrupted only in case one or more of the CCP's members default.
126
 kom (2016) 0853 - Ingen titel
1715591_0127.png
There were also concerns that the rules did not take a sufficiently holistic view of how the
different types of exposures to a QCCP interrelate and were therefore not sufficiently sensitive to
the aggregate risk of those exposures and how that risk is distributed. More specifically, the
concerns were that the rules did not recognise sufficiently the fact that increasing members'
contributions to a QCCP's default fund would, all else equal, make the QCCP safer. These
concerns were due to the fact that the capital requirements proportionately increase with the size
of the contribution to the default fund of the QCCP.
131
The abovementioned Method 2 was introduced in the interim international standard (an
implemented in the CRR) as a temporary solution intended to address situations where Method 1
was deemed to lead to excessively high capital requirements. It was meant to buy time to allow
for the development of a permanent solution that would address the abovementioned problems.
132
The permanent solution was published by the Basel Committee in April 2014.
133
The need for an early review without evaluation
Despite the lack of sufficient implementation experience for an in-depth evaluation of existing
rules, impact studies at international level and work conducted by the BCBS have showed the
shortcomings of current rules which will be fixed with the amendment described below. The
review is also needed to comply with BCBS standards.
Proposed solution
The revised standards adopted by the Basel Committee will be implemented. Notable revisions to
the Basel standards include the use of a single method for determining the capital requirements
for exposures to QCCPs stemming from default fund contributions, an explicit floor for those
requirements, as well as an explicit cap on the overall capital requirements applied to exposures
to QCCPs (i.e. those charges will not exceed the charges that would otherwise be applicable if the
CCP were a non-qualifying CCP). They did not change the treatment of exposures to non-
QCCPs.
Under the new method for capital requirements for default fund contributions a more holistic
approach is taken that ensures that the capital requirement no longer increases in proportion with
the size of the contribution. The method also applies a more risk-sensitive approach for
calculating the "hypothetical" resources (called the standardised approach for counterparty credit
risk or SA-CCR). While this new approach is more complicated, the fact that the calculation of
the hypothetical resources is actually required from the QCCP and not from the institutions,
means that there is no increase in compliance costs for institutions and the increase in costs for
the QCCP should be fairly limited (mostly due to the one-off cost of changing their systems to
accommodate this new approach) but this is more than outweighed by the benefits brought from
the higher risk-sensitivity of the approach. The method also introduces a floor to the capital
requirements to ensure that there is at least a small capital requirement for those exposures and
hence that institutions still monitor them and manage them.
131
This is true under both Method 1 and Method 2.
132
Its introduction created a new problem: since institutions were left full freedom of choice between the
two methods, this meant that they were allowed to choose the method that delivered the lower capital
requirement and not necessarily the one that reflected the inherent risks better.
133
Capital requirements for bank exposures to central counterparties - final standard. Available at
http://www.bis.org/publ/bcbs282.pdf.
127
 kom (2016) 0853 - Ingen titel
The revised standard also provides for an explicit cap on the capital requirements for exposures
to QCCPs: the latter cannot be higher than in case if the same CCP would be deemed a non-
QCCP.
The fact that only one method is used instead of two has the additional benefit of decreasing the
complexity of the rules and removes the arbitrage possibility present in the current rules (because
of the two methods).
The revised rules further reduce the administrative burden for institutions by dropping the
requirement for legal opinions from the conditions that need to be met by institutions in order to
be able to use the more favourable treatment for trade exposures (it was replaced by the condition
for the institution to conduct sufficient legal review).
128
 kom (2016) 0853 - Ingen titel
Annex 3.10. Contractual recognition of bail-in (article 55 BRRD)
Background/introduction:
Stakeholders raised practical concerns with respect to Article 55 of Directive 2014/59/EU
establishing a framework for the recovery and resolution (BRRD). The provision requires credit
institutions and other entities falling under the scope of the BRRD to include in contracts to
which they are party and which are governed by the law of a third country a clause by which the
creditor recognises the bail-in power of Union resolution authorities. This obligation is
particularly relevant for branches of Union banks in third countries, as their business, and in
particular concluded contracts, are usually governed law of those third countries.
Problem definition
Stakeholders reported that compliance with Article 55 BRRD raises two types of difficulties.
First, certain third country counterparties refuse to include a contractual clause recognising a
Union bail-in power in financial contracts concluded with Union banks. These third country
entities often have a high degree of negotiating power against Union banks, or apply
internationally agreed standard contractual terms in their banking contracts, e.g. with respect to
liabilities to non-Union financial market infrastructures or trade finance liabilities (letters of
credit, bank guarantees and performance bonds). As a result, the only way for Union banks to
comply with Article 55 BRRD in these cases would be not to enter into the contract at all. In
extreme cases this could entail that a certain portion of their business would need to be ceased.
Secondly, even when third country counterparties are prepared to accept bail-in related clauses in
their contracts with Union banks, in some cases the local supervisor may forbid this. In this case
the only way for banks to comply with Article 55 BRRD would be to either contravene to the
rules imposed by the local supervisor or exit the relevant part of their business.
The need for an early review without evaluation
The BRRD provisions entered into force in 2016. Despite the lack of sufficient implementation
data to conduct an evaluation, in particular the provisions of Article 55 generated an extensive
feedback from the industry and resolution authorities. Data on the magnitude of the problem has
been made available by banks under the coordination of the European Banking Federation. It was
not verified to what extent this data was representative, but it was nevertheless suitable to
demonstrate the different degrees of impact for banks, since their share of liabilities governed by
third-country law widely differs. Hence, an early adjustment of rules seemed necessary and the
proposed solution to grant discretion to the resolution authority in applying the requirement
seemed the most suitable approach.
Objective
A better environment for jobs and growth across Europe is the ultimate goal. In this respect, it is
worth mentioning that a credible and stable financial system is key. To achieve this goal, there
must be reassurance that global institutions can be resolved in an orderly manner without causing
disruptions to the financial system and to the economy in general, therefore avoiding the use of
taxpayers’ money. This is only possible if institutions hold sufficient liabilities that can actually
be bailed-in in resolution. It is within this spirit that the EU agreed initially on a broadly worded
provision (Article 55 of BRRD) whereby any liability which is subject to the law of a third
country would not escape the normal loss absorption cascade in resolution, and therefore, would
not be treated more favourably than other liabilities of the same type only for the reason that they
129
 kom (2016) 0853 - Ingen titel
are not subject to EU law. Still, this should not be seen as a one-size- fits-all-approach. A series
of instruments, including trade finance instruments are of the utmost importance for international
trade, in particular, for small and medium sized EU companies. In this regard, article 55, should
not affect access of European manufacturers and service providers to trade finance instruments,
in particular, and should not weaken their competitiveness in international markets with potential
adverse economic effects in the EU.
Furthermore, the effectiveness and practicability of the Article 55 provisions need to be judged in
the context of their ultimate purpose, i.e. the facilitation of bail-in. To that end, Commission staff
has gathered evidence from the European Banking Federation, jointly with the Bankers
Association for Finance and Trade (BAFT) and the International Chamber of Commerce (ICC).
The evidence includes a survey to banks which aims at quantifying the potential effects of
application (or waving of) Article 55 for two types of liabilities: (i) information on subordinated
and senior unsecured debt, i.e. debt that would likely be available for bail-in, (ii) other liabilities,
which may impede the effectiveness of the bail-in tool due to e.g. operational challenges.
For the first type of liabilities, the submitted data shows a large margin of deviation regarding the
share of subordinated and senior debt governed by third country law in comparison to debt of
that category governed by Union law. For some banks this category is entirely irrelevant, whereas
some banks claim that up to 35% of these securities are governed by 3rd country law.
The conclusions are three-fold: Firstly, the issue at stake is generally sizable for parts of the
industry. If third-country counterparties would not be willing to enter into contractual recognition
clauses, banks may struggle to roll-over significant parts of their liabilities at maturity. Secondly,
for some banks the resolution authority will need to assess the effectiveness of these clauses for a
significant part of the liabilities, as well as managing the related risks. Thirdly, the data exhibits a
very heterogenous picture that suggests enabling the resolution authority to conduct a case-by-
case approach.
For the second type of liabilities, stakeholders have singled out three particular classes for which
the application of Article 55 would likely lead to costs without equivalent benefits:
a) Contingent liabilities arising from e.g. trade finance products (e.g. letters of credit)
Trade finance generates mostly contingent exposures; hence its potential value in resolution is
difficult to evaluate ex-ante. Moreover, a reduction of the liability under a letter of credit vis-à-
vis the bank under resolution would automatically result in a corresponding reduction of the
counterclaim of this bank against the third party in whose interest the trade finance product in
question has been issued, therefore it is unlikely to create any loss absorption or recapitalisation
capacity. No industry-wide data has been available, but banks who have responded to the survey
indicated that the value of the (contingent) liabilities stemming from trade finance does not
exceed 1-2% of MREL.
b) Liabilities vis-a-vis Financial Market Infrastructures/Central Counterparties
(FMIs/CCPs)
FMI/CCP participation is generally governed by standard contracts that individual banks are
incapable of changing on their own. Hence, Article 55 appears to be a particularly inappropriate
mechanism for introducing bail-in rights vis-à-vis FMIs. In addition, according to Article 44(2)(f)
130
 kom (2016) 0853 - Ingen titel
BRRD, liabilities with a remaining maturity of less than seven days owed to operators of such
infrastructures are excluded from bail-in and already not subject to Article 55.
c) Derivatives
Respondents to the survey indicated that the immense majority of derivative contracts are written
under standard (ISDA) terms into which individual banks cannot insert Article 55 clauses. Those
few respondents who provided data indicated that contracts which are not governed by such
standard contracts would not exceed 3% of the nominal value of all derivatives, including
collateralised positions. The costs of inserting Article 55 hence seem to outweigh the benefits
significantly.
Proposed solution
Article 55 can only be amended by a new legislative proposal. It cannot be revised by way of a
technical 'Level 2' measure or through interpretative guidance. Even before the proposal would
be finally adopted by the co-legislators, it would produce some benefits for the authorities and
industry concerned as it is common practice for the Commission not to pursue violations of
Union law provisions after it has adopted a proposal which aims at amending the provision in
question in a way that would eliminate the violations.
The amendment to article 55 BRRD would entail an application of the requirement by the
resolution authority in a proportionate manner. The resolution authority can exclude the
obligation by means of a waiver if it determines that this would not impede the resolvability of
the bank, or that it is legally, contractually or economically impracticable for banks to include the
bail-in recognition clause for certain liabilities. In these cases, those liabilities should not count as
MREL and should rank senior to MREL to minimize the risk of breaking the No-Creditor-Worse-
Off (NCWO) principle. In this regard, the proposal will not to weaken the bail-in.
131
 kom (2016) 0853 - Ingen titel
Annex 3.11. Changes to MREL
Problem definition
The incorporation of TLAC in the EU legislative framework should not materially affect the
burden of non G-SIBs to comply with the current MREL framework. Fundamentally, TLAC and
MREL aim to achieve the same policy objective of ensuring that banks hold a sufficient amount
of bail in-able liabilities that allow for smooth and quick absorption of losses and bank
recapitalisation. Some technical differences exist however between the 2 frameworks regimes in
terms of eligibility criteria (excl. subordination) and in terms of basis of calculation of the
requirement. Additionally, MREL is not specific as to how bial-in capacity should be allocated
within groups depending on the choses resolution strategy. MREL is set on an individual basis to
each institution. As resolution policies are developing, thereby distinguishing between a Single
Point of Entry (SPE) and a Multiple Point of Entry (MPE), it becomes clear that resolution tools
will be applied at the level of the resolution entity, covering all material subgroups (subsidiaries)
that compose the resolution group. Firstly, the concepts of resolution entities, resolution groups
and material subgroups are not defined in the current BRRD level 1 legislation. Secondly, the
prepositioning of internal/external loss absorbing capacity at subsidiary level is not defined in the
current MREL framework.
Maintaining the status quo would imply that the existing differences between TLAC and MREL
would be maintained which would result in 2 technically inconsistent frameworks which pursue
similar objectives. Moreover, the lack of an adequate approach to resolution policies within
groups could lead to divergence of practices across member states and reduced confidence by
resolution entities in charge of subsidiaries, leading to ring fencing measures. In the call for
evidence, several claims were received on the lack of consistency between the TLAC and the
MREL framework and the complexity that this would entail.
The need for an early review without evaluation
BRRD provisions requiring banks to comply with MREL entered into force in 2016. Reviewing
them is consistent with the directive, Article 45(18) of which includes a mandate to the European
Commission by December 2016 to submit, if appropriate, a legislative proposal to the European
Parliament and the Council on the harmonised application MREL, including proposals for the
introduction of an appropriate number of minimum levels of MREL and any appropriate
adjustments to the parameters of the requirement. The proposal takes into account the analytical
report of the EBA on a wide range of MREL-related aspects listed in Articles 45(19) and (20) of
the BRRD, including consistency of MREL with the minimum requirements relating to any
international standards (such as TLAC) developed in the international fora.
Proposed solution
MREL will be amended to address some shortcomings, notably to (1) create 1 set of eligibility
criteria for MREL/TLAC eligible instruments (except for subordination), (2) clarify the internal
loss absorbing capacities within banking groups, independent of the chosen resolution strategy
132
 kom (2016) 0853 - Ingen titel
through introduction of the concepts of resolution groups, resolution entities and material
subgroups, and (3) the alignment of the basis for calculation on the RWA and the leverage ratio
exposure measure.
133
 kom (2016) 0853 - Ingen titel
Annex 3.12. Application of IFRS 9 by the EU banks
Problem definition
After the financial crisis, the G20 and the BCBS pushed international accounting standard setters
to enhance the "too little too late" credit loss provisioning rules under the IAS 39 and the US
GAAP models. IFRS 9 will bring improvements in that respect by introducing a forward looking
model for the provisioning of loan losses that should lead banks to book higher and earlier
provisions than in the past. IFRS 9 which was endorsed by Member States on 27 June 2016 will
be applied from the beginning of 2018.
The move from IAS 39 to IFRS 9 will affect CRR capital requirements. The impact will depend
on:
1. the amount of the increase in provisions due to the change in accounting;
2. the type of regulatory approach the bank follows to calculate its capital
requirements;
3. for IRB banks, their present level of provisions compared to the regulatory
expected loss.
In relation to the first issue, there is still
uncertainty about the difference in levels of
provisioning between current IAS 39 and IFRS 9.
Banks are still working on the
implementation of the new IFRS. They have therefore not been able to produce precise figures so
far. Some analysts have however pointed to a 20% increase in provisions. A 20% - 30% increase
of loan loss provisions seems to be confirmed by an analysis the EBA is currently carrying out
for a sample of banks. The EBA provisionally estimates a reduction of 50 basis points on average
for the CET1 ratios. EBA has however already pointed out that data are not fully reliable.
In relation to the second issue, banks on the
Standardised Approach (SA)
will probably be the
most affected. They will see a reduction in their CET1 capital equal to the increase in provisions
following the introduction of IFRS 9 which is only very partially compensated by reduced capital
requirements following a reduction of exposure values from increased deductions of specific
credit risk adjustments.
In order to partially limit the impact on accounting provisioning, Basel and CRR (Art. 62) allow
banks that use the standardised approach to add back provisions which are "general" in nature
(i.e. not linked to any particular position) and deducted from CET1, as (lower quality) Tier 2
capital to meet some of the bank's capital requirements. The adding back of general provisions as
Tier 2 capital is subject to a cap of 1.25% of the bank's risk weighted assets RWA.
In contrast to the US, where all credit provisions are considered as "general", the EU has adopted
a RTS that labels any credit risk provision under the current IAS 39 incurred loss model as a
"specific" provision which cannot be added back to Tier 2 capital.
In relation to
Internal Ratings Based (IRB) banks,
it is necessary to differentiate the analysis
on whether present accounting rules result for certain assets (e.g. a loan portfolio) in a provision
that is higher or lower than the Expected Loss calculated according to CRR / Basel prudential
rules. There are two possible cases.
134
 kom (2016) 0853 - Ingen titel
1715591_0135.png
The "shortfall"
case:
if the accounting loan loss provision is lower than the prudential
expected losses. In this case banks must, when calculating their capital ratio, deduct
from CET1 capital the difference between the prudential expected losses (higher) and
the accounting provisions (lower). Obviously, this reduces for a bank the available
amount of CET1, the highest quality of capital, and makes it for the bank more
difficult to maintain a certain surplus on top of capital requirements and therefore a
certain rating.
This deduction from CET1 is imposed by bank regulators to prevent insufficient levels of
provisions from accounting standards.
The "excess"
case:
if, in the alternative case, the accounting provision for credit
losses is higher than the prudential expected loss, when calculating their capital ratio
IRB banks can add the "excess" in provisioning back as Tier 2 capital up to a limit of
0.6% of the Risk Weighted Assets.
134
The impact on an IRB bank of an increase in accounting provisions due to the introduction of
IFRS 9 will therefore depend on whether the bank is in a "shortfall" or in an "excess" case. If this
increase happens for a bank with a "shortfall" (i.e. accounting provisions are less than the
prudential expected loss), IFRS 9 will increase accounting provisions, but the effect on CET1
capital of these higher provisions will be normally compensated by fewer deductions according
to Basel / CRR rules described above. The impact can therefore be expected to be - by and large -
limited.
If the increase in accounting provisions occurs instead for a banks in an "excess" case, (i.e.
accounting provisions are already higher than the prudential expected loss), any increase in
accounting provisions will directly determine a reduction in CET1 capital, and the bank CET1
capital ratio will be reduced accordingly, with all expected consequences (higher funding costs,
lower rating, etc.). The possible increase in lower quality Tier 2 capital would not provide a
sufficient compensation.
EBA has informally estimated that some 2/3 of the EBA sample (large) banks using the IRB
approach are in a situation of "shortfall" of accounting provisions, while one third would be in a
situation of "excess" of accounting provisions. These some 38 banks in the "excess" case, plus all
those on the SA approach not comprised in the EBA sample would be the one most probably
affected by any increase in provisioning due to the introduction of IFRS 9 in 2018.
We suspect that banks more active in commercial banking activity and focused on Member States
with high levels of NPLs might be the most affected. Although there are not entirely reliable
numbers on the impact, we have understood that for some banks the capital ratios might be
reduced by 0.5 – 1.5 percentage points (i.e. up to minus 15% for a bank with a 10% capital ratio).
This would most probably have a direct impact on those banks' lending practices.
The Basel Committee will not finalise a revised specification of how IFRS 9 accounting interacts
with the calculation of bank capital requirements until after IFRS 9 becomes effective on 1
January 2018.
The need for an early review without evaluation
134
IRB banks in the "excess" case and SA banks are therefore in a "similar" situation in relation to how the
introduction of IFRS 9 can impact their capital (higher deductions from CET1, possibly reconsidered in
Tier 2).
135
 kom (2016) 0853 - Ingen titel
The mandatory application of IFRS 9 starts from 1 January 2018 onwards, before the CRR will
be evaluated. In light of the potential significant and sudden impact on banks' capital ratios it is
opportune to include the transitional phasing in of expected credit loss provisions in the CRR
review.
Proposed solution
The potential significant impact of IFRS 9 expected loss provisioning on CET1 capital creates a
need for action now so that possible measures can avoid any sudden unwarranted impact on
banks' capital ratios and lending in 2018.
The best possible solution seems to introduce in CRR a transitional regime so that IFRS 9
changes will be phased-in progressively over a few years. Treatment would need to be adapted
according to the approach for calculating capital requirements used by banks.
The CRR review would need to include a separate article (e.g. Art. 473a) for a transitional regime
on the phasing in of the higher loan loss provisioning of IFRS 9 compared to IAS 39 from 2018
onwards.
136
 kom (2016) 0853 - Ingen titel
1715591_0137.png
Annex 3.13. Comparative analysis of characteristics of EU G-SIIs and O-SIIs
The calibration of MREL should be closely linked to and justified by the institution’s resolution
strategy and should take into account such criteria as the business model, size, risk profile,
funding model or the extent to which the Deposit Guarantee Scheme could contribute to the
financing of resolution
135
. It can be observed from the figure below that in terms of size
compared to GDP (of the member state in which a O-SII is recognised) there is a large
heterogeneity between the different O-SIIs both within and across member states.
Figure A9.
Size of EU G-SIIs and O-SIIs compared to GDP
Source: The European Commission calculations based on SNL and Eurostat
Also in terms of business models and activities there is a large disparity between the different O-
SIIs. In the figure bellow we split up both assets and liabilities of EU G-SIIs in four main activity
categories (interbank, customer loans and deposits, derivatives and securities/senior debt). It can
be observed that O-SIIs on average have on the one hand more interbank activities, more
customer loans and deposits, on the other hand on average they have less derivative and securities
activities. Senior debt levels are similar in both groups. Looking beyond the averages gives a
good insight in the variety in business models. For each activity, there are O-SIIs which have
more than 60% of their total balance sheet volume in this activity, but there are also O-SIIs which
have no volume in it. This a consequence of the different types of banks which have been
identified as O-SIIs ranging from retail savings banks, depositary banks, investment banks,
captive banks, building societies, etc.
Figure A10.
Importance of selected liabilities for EU G-SIIs and EU O-SIIs
135
Article
45(6) of the BRRD
137
 kom (2016) 0853 - Ingen titel
1715591_0138.png
The figure shows, based on data for almost all EU-SIIs, for selected balance sheet components the minimum, maximum, interquartile
range and average compared to total assets for respectively EU G-SIIs and EU O-SIIs
Source: The European Commission calculations based on SNL
138
 kom (2016) 0853 - Ingen titel
1715591_0139.png
Annex 3.14.
Analysis of a leverage ratio requirement for different business models and
exposure types
The EBA has assessed the impact of the leverage ratio for different levels of calibration for
twelve different business models and found some differences in impact of the leverage ratio on
certain business models, including public development banks. However, EBA's advice is that
these differences are not so material that they would justify a differentiation in the calibration of
the leverage ratio requirements. Hence the EBA advises on a single Tier 1 capital calibration for
the leverage ratio for any institution irrespective of its business model.
A possible lower calibration for public sector lending by public development banks
As evidenced by the Call for Evidence some public development banks have to date a 1%
leverage ratio. Normally public development banks are majority owned by the State or the public
sector, are not for profit, do not take retail deposits and are subject to legal constraints on their
lines of business which limits their possibilities to meet a leverage ratio requirement compared to
other types of banks. The latter implies that public development banks have much less discretion
to manage their balance sheet or income compared to universal banks that can freely choose their
business model. The EBA recognises this special situation for public development banks in its
report.
The EBA points out that it is difficult to provide a common European definition of public
development banks given the broad diversity of types of public development banks. It may be
difficult indeed to define public development banks and one should be wary of distorting
competition through favourable prudential treatments and favourable treatments being considered
State aid. However, there are common criteria that could define a public development bank. A
public development banks is a credit institution that:
is organised as a credit institution under public law;
has a legally defined mandate defining the public policy objective of its
business and defining the business areas in which it is allowed operate;
operates on a not for profit basis;
does not take deposits from retail clients.
Moreover, there is already a definition of promotional banks in the Commission delegated
regulation (EU) 2015/63 on the ex-ante contribution to resolution financing arrangements.
In a context of enhancing economic growth and jobs creation, it seems contradictory to impose
prudential requirements on public development banks through a leverage ratio that would
increase the cost of public sector lending.
The leverage ratio requirement should therefore be adjusted by excluding from the leverage ratio
exposure measure public development loans and pass-through promotional loans provided by
public development banks set up by a Member State, central or regional government or
municipality.
A possible higher calibration for G-SIIs (Globally Systemically Important Institutions)
If a bank is categorised as a G-SIB its CET1 risk sensitive capital requirement will increase
depending on its degree of SIFI-ness. In order to maintain the same level of backstop of the
139
 kom (2016) 0853 - Ingen titel
leverage ratio for G-SIIs subject to higher risk sensitive capital requirements, the leverage ratio
requirement would need to be increased in proportion to the additional G-SIB capital surcharges.
The Basel Committee has not yet decided on possible leverage ratio surcharges for G-SIBs. The
EBA has indicated in its report on the leverage ratio that it is carefully monitoring the possible G-
SII surcharge in terms of design and calibration. At this point the outcome of the Basel
Committee work on G-SII surcharges should be awaited and upon the Basel Committee adopting
G-SIB surcharges should be decided whether the Basel G-SIB surcharges would be appropriate
for European G-SIIs and whether these surcharges should be considered also for large O-SIIs
(Other Systemically Important Institutions).
Several claims have been made in the Call for Evidence about undesired effects of a binding
leverage ratio on other prudential or economic objectives. In particular claimants asserted
(without providing further evidence on the impact) that the leverage ratio:
would run counter the Liquidity Coverage Requirement objective of holding
highly liquid assets as the leverage ratio would impose constraining capital
requirements on these low risk weighted highly liquid assets;
would undermine central clearing of derivatives by banks for clients due to not
recognising segregated collateral for reducing the leverage ratio exposure
measure including initial margins;
would require holding more capital for trade finance exposures secured by Export
Credit Agencies (ECAs);
Although the comments on the LCR interaction are understandable, the LCR and leverage ratio
pursue different prudential objectives (liquidity and capital) which credit institutions have to meet
in parallel. EBA concluded in its report that correlations between the LCR and the leverage ratio
are very weak. Holding buffers on top of the prudential minimum requirements for a particular
ratio, such as the LCR, is not necessarily accompanied by a low Leverage ratio. On the contrary,
the EBA results give evidence that many institutions manage to hold significant buffers on top of
all prudential requirements at the same time. The leverage ratio requirement should therefore not
be adjusted for the interaction between the LCR and the leverage ratio.
Banks acting as clearing member have argued that if the leverage ratio requirement does not
allow the initial cash margins received from clients and properly segregated from their own cash,
to reduce the potential future exposure on the client leg of the centrally cleared client derivative
transaction this would result in a disproportionate increase in capital requirements for this low
margin business. This would adversely affect the provision of central clearing services to clients
which is contrary to the G20 objective of promoting central clearing. The Basel Committee is
currently considering this issue carefully and seeking further evidence on the potential impact of
the Basel III leverage ratio on clearing members’ business models during the consultation period.
The leverage ratio should not adversely impact the provision or pricing of centrally cleared
derivative transactions that credit institutions as a clearing member to a CCP (Central Clearing
Party) offer to clients. From this perspective the Commission looks forward to the forthcoming
Basel decision on the treatment of initial margins for inclusion in the CRR review.
Short term trade finance exposures such as letters of credit are often subject to higher capital
charges than the implicit leverage ratio capital requirement of 35% or in case of off balance sheet
trade finance positons subject to the same credit conversion factor and hence the leverage ratio
would not be constraining compared to the risk weighted capital requirements. This is however
different for export credits guaranteed by sovereigns or export credit agencies which receive a
considerably lower risk weight. In these instances the leverage ratio would be constraining capital
requirement leading to higher capital charges. The leverage ratio would in any case affect the
140
 kom (2016) 0853 - Ingen titel
internal allocation of capital costs within banks to the specific business line of export credit and
depending on the relative importance of export credit within the bank overall business impact the
overall leverage ratio. (Albeit that since virtually all banks operate well above a 3% Tier 1
leverage ratio the inclusion of guaranteed export credits in the leverage ratio exposure measure
will not necessarily create a sudden need for banks to raise capital or to deleverage.) However,
export credits are important for jobs and growth and therefore guaranteed export credits need to
be excluded from the leverage ratio exposure measure.
141
 kom (2016) 0853 - Ingen titel
1715591_0142.png
A
NNEX
4. E
STIMATED IMPACT OF POLICY OPTIONS
On funding Risk
Comparison of the different policy options to provide a complementary stable funding requirement to
capital and liquidity requirements
Effectiveness
Efficiency
n.a.
Option 1: No policy change
n.a.
Option 2: A single NSFR requirement
as per Basel for all banks
(+) would be an internationally recognised
credible stable funding measure for all EU
banks
(-) may disproportionally "hit" business
models of banks that are outside the scope
of the international Basel framework
(-) may have a disproportionate impact on
some specific activities
(+) since institutions have already to
ensure that their long term
obligations are adequately met with a
diversity of stable funding
instruments there is hardly
incremental operational cost to
determine the requirement
Option 3: A single NSFR requirement
as per Basel with some adjustments for
all credit banks
(+) could allow taking into account some
European specificities
(+) could help to alleviate the unintended
consequences of the NSFR on some specific
activities
(-) difficult to define these specific activities
and an appropriate calibration
(+) less operational cost if
adjustments to the NSFR are closer to
the economic reality of the operations
(-) uncertainties linked to the
potential different calibration of the
European NSFR
Impact on stakeholders
Option 1 (no policy
changes)
Option 2
(A single NSFR
requirement as per
Basel for all banks)
n.a.
(+/-) Improves the stability of funding of
banks'
activities on an ongoing structural
basis and reduces the maturity mismatches but can disproportionately impact some
specific activities that are not adequately recognised by the Basel standard.
(-)
Banks having a low funding risk profile
on the one hand benefit from a further
improvement in the stability of their funding but, on the other hand, the adjustment
costs that they face are disproportionate compared to the marginal benefits.
(+)
Supervisors
gain an additional instrument to monitor and limit the ongoing,
structural dimension of funding risk that was not properly captured so far.
(+)
Companies and households
benefit from this requirement as more stable
funding sources increase banks' resilience at times of funding stress, reducing the
likelihood of systemic stress with adverse macroeconomic consequences, and
enhancing the ability of banks to continue lending in a challenging liquidity
environment.
Option 3
(++) Improves the stability of funding of
banks'
activities on an ongoing structural
basis and reduces the maturity mismatches, while preserving the ability to run
142
 kom (2016) 0853 - Ingen titel
1715591_0143.png
Impact on stakeholders
(A single NSFR
requirement as per
Basel with some
adjustments for all
banks)
activities that would be disproportionately impacted, as not adequately recognised,
by the introduction of the Basel standard.
(+/-)
Banks having a low funding risk profile
on the one hand benefit from a
further improvement in the stability of their funding but, on the other hand, the
adjustment costs that they face are disproportionate compared to the marginal
benefits if the adjustments to the Basel standard do not take enough account of their
specific business models.
(+)
Supervisors
gain an additional instrument to monitor and limit the ongoing,
structural dimension of funding risk that was not properly captured so far.
Adjustments applicable to all institutions are unlikely to have a significant impact on
supervisory work.
(++)
Companies and households
benefit from this requirement as more stable
funding sources increase banks' resilience at times of funding stress as described in
option 2. Moreover, the adjustments to the Basel NSFR standard prevent a negative
impact on the financing of the economy in the activities that would otherwise be
disproportionately affected.
On leverage ratio
Effectiveness
Efficiency
Option 1 (baseline)
Option 2
A single leverage
ratio requirement as
per Basel for all
institutions
n.a.
(+) would act as a backstop to model risks and the build-up of excessive leverage
(-) one size fits all implies that the leverage ratio is more constraining on
institutions with low risk business models
n.a.
(+) same measure
and method of
calculation applies
to all banks
(+) common
backstop
irrespective of the
type of business
model
Option 3
A leverage ratio
requirement
differentiated for
business models or
adjusted for
exposure types
(+)Would not have
disproportionate effects
on low risk business
models
(+) would prevent
undesirable impact on
other prudential policy
objectives such as the
LCR, or central clearing
or specific types of
lending (public sector , .
(-)potentially reduces
the backstop function of
the leverage ratio
(+)Would not have
disproportionate effects
on low risk business
models
(+) would prevent
undesirable impact on
other prudential policy
objectives such as the
LCR, or central clearing
or specific types of
lending (public sector , .
(-)potentially reduces
the backstop function of
the leverage ratio
(+)Would not have
disproportionate effects
on low risk business
models
(+) would prevent
undesirable impact on
other prudential policy
objectives such as the
LCR, or central clearing
or specific types of
lending (public sector , .
(-)potentially reduces
the backstop function of
the leverage ratio
(- )users of leverage
ratio information
would need to take
into account varying
calibrations of the
leverage ratio
(-) definition of
business models
would need to be
developed
(-) makes the
calculation of the
leverage ratio
slightly more
complex
143
 kom (2016) 0853 - Ingen titel
1715591_0144.png
Effectiveness
Efficiency
(-) would create a
divergence from the
international agreed
leverage ratio
Impact on stakeholders
Option 1 (no policy
changes)
Option 2
(A single leverage ratio
requirement as per for
all institutions)
n.a.
(+) Improves the financial stability of
banks'
due to underestimating risks in
particular during economic upswings
(-)
Banks
with low risk weighted business models may be disproportionally affected
by a one size fits all leverage ratio requirement
(+)
Supervisors
will have an additional minimum benchmark for assessing risk of
excessive leverage of institutions during their supervisory review process.
(+)
Companies and households
benefit through enhanced financial stability of the
banking system
(-) for some
companies
loans and other services provided by institutions may
become more expensive due to capiadditonal capital charges stemming from the
leverage ratio.
Option 3
A leverage ratio
requirement
differentiated per
business model or
exposure type
(+) Would prevent creating disproportionate effects from the leverage ratio
requirement for
banks
with low risk weighted business models due to enhanced risk
sensitivity of the leverage ratio.
(-) would water down the main feature of the leverage ratio for
banks
as a non-risk
based back-stop to risk sensitive capital requirements
(-)
Supervisors
have to deal with differently calibrated leverage ratios which lessen
supervisory effectiveness due to reduced comparability of the leverage ratio of
supervised entities
(+)
Companies and households
would benefit from enhanced financial stability
without unnecessarily increasing costs for banking business that have a truly low
risk character.
(-) comparing leverage ratios across banks would become more complex for
investors
Option 4
(A leverage ratio
adjusted to prevent
undermining other
policy objectives
(+/-) Improves the financial stability of
banks'
due to risk of excessive leverage
albeit to a lesser extent than under option 2 and 3 because more adjustments to the
leverage ratio would be made.
(+) The leverage ratio requirement as a prudential measure would not have
unintended or unwanted adverse impacts on other prudential objectives or jobs and
growth policy objectives to the benefit of
companies and households.
144
 kom (2016) 0853 - Ingen titel
1715591_0145.png
Impact on stakeholders
(--)
Supervisors
have to deal with differently calibrated leverage ratios and
adjusted leverage ratio exposure measures which lessen supervisory effectiveness
due to reduced comparability of the leverage ratio of supervised entities
(+)
Companies and households
would benefit from enhanced financial stability
without unnecessarily increasing costs of for banking business that have a truly low
risk character.
(-) comparing leverage ratios across banks would become more complex for
investors
On SME exposures
Comparison of the impact of policy options on stakeholders
Impact on stakeholders
Option 1 (no
policy changes)
Option 2
Alignment with
the Basel rules
n.a.
(-)
Banks using standardised approach for SME exposures
would be worse off. While on
average the change is only 0.16% difference in the capital ratios, there is a high variation
among individual banks. The effect could be partially offset by the new Basel
Standardised Approach, which is currently being revised, as BCBS intends to reduce
exposures to SME loans from the 100% to 85%.
(– –)
Banks using internal ratings-based approach for SME exposures
would be worse
off. While on average the change is only 0.16% difference in the capital ratios, there is a
high variation among individual banks. Moreover, currently BCBS does not foresee any
change to the internal ratings-based approach.
(–/≈)
Companies and households
might be affected by the additional funding constraints
by banks, particularly from the most capital constraint banks.
(≈) Some
regulators/supervisors
might see the benefit of aligning risk weight calibration
with the Basel rules, while others might be concerned by moving back to Basel rules
which could go against the evidence seen on the actual riskiness of SME loans in the EU.
Option 3
Introducing
additional capital
reduction for SME
exposures above
€1.5 million
(+)
Banks using SA and IRBA.
Both banks using SA and IRBA would be better off as
they would obtain additional capital relief and thus would have incentives to provide
more financing to the economy and, in particular, SMEs.
(+/≈)Companies
and households.
They might benefit, in terms of both volume and price,
from increased incentives for banks, particularly the most capital-constraint banks, to
provide additional financing of the economy,
(≈)
Some
regulators/supervisors
might be concerned by moving back to Basel rules which
could go against the evidence seen on the actual riskiness of SME loans in the EU, while
others might see the benefit of aligning risk weight calibration with the Basel rules.
145
 kom (2016) 0853 - Ingen titel
1715591_0146.png
On loss absorption and recapitalisation capacity
Effectiveness
Efficiency
n.a.
Option 1: No policy change
n.a.
Option 2: Integrate TLAC standard in MREL for EU G-SIIs
(+) Enhances global financial
stability by promoting
implementation of framework
for bail-inable liabilities
across jurisdictions
(+) Enhances level playing
field and enhances clarity
(-) Depending on calibration
of MREL, this could represent
an additional funding cost for
banks
(+) Common backstop for
all G-SIIs
(+) Clarity on applicable
regulatory framework
Option 3: Integrate TLAC standard in MREL for EU G-SIIs
and O-SIIs
(+) Enhances global financial
stability by promoting
implementation of framework
for bail-inable liabilities
across jurisdictions
(+)Enhances level playing
field and enhances clarity
(+) Common backstop for
all SIIs
(+) Clarity on applicable
regulatory framework
(-)Would be
disproportionate for a
large number of banks as
TLAC standard developed
for G-SIIs and minimum
calibration might
overshoot actual needs
based on resolution
strategy
(- )Disproportionate to impose
specific G-SII standards to O-
SIIs and additional funding
cost impact compared to
option 2
Stakeholder
Banks
Policy option
Bank debt- and
shareholders
Supervisors
Companies and
households
Option 1: No policy
change
Option 2: Integrate
TLAC standard in
MREL for EU G-SIIs
0
0
0
0
+
+
+
+/-
Option 3: Integrate
TLAC standard in
MREL for EU G-SIIs
and O-SIIs
+/-
+
+
+/-
Impact on stakeholders
146
 kom (2016) 0853 - Ingen titel
1715591_0147.png
Impact on stakeholders
Option 1 (no policy
changes)
Option 2
(Integrate TLAC
standard in MREL
for EU G-SIIs)
n.a.
(+) Improves the financial stability of
banks'
due to increased loss
absorption and recapitalisation capacity in the global banking system and
avoids overlaps in regulation between EU MREL and international TLAC
(+)Provides increased clarity for bank
debt- and shareholders
on the order
in which instruments could be bailed in
(+)
Supervisors
will have a minimum benchmark to set loss absorption and
recapitalisation capacity
(+/-)
Companies and households
benefit through enhanced financial
stability of the global banking system but loans and other services provided
by institutions may become more expensive due to increased funding costs.
Option 3
(Integrate TLAC
standard in MREL
for EU G-SIIs and O-
SIIs)
(+/-)As under option 2 banks' benefit from increased financial stability in the
global banking system but for O-SIIs this could have disproportionate effects
(+)Provides increased clarity for bank
debt- and shareholders
on the order
in which instruments could be bailed in
(+)
Supervisors
will have a minimum benchmark to set loss absorption and
recapitalisation capacity.
(+/-)
Companies and households
benefit through enhanced financial
stability of the banking system but loans and other services provided by
institutions may become even more expensive compared to option 2 due to a
further increase in funding costs stemming from the disproportionate impact
on O-SIIs.
On remuneration
Problem 1
Impact on stakeholders
Option 1: No policy
change
Option 2: Allow
Member States or
supervisory authorities
to exempt some
institutions and staff
from the rules on
deferral and pay-out in
instruments
n.a.
(+)
Positive effect on
Regulators / supervisory authorities,
as the level
of supervision would be tailored to the riskiness of institutions and staff;
some Regulators / supervisory authorities may find it an advantage that
they could fix their own exemption criteria and thresholds
(+)
Positive effect on
Institutions and Employees,
as the rules would be
more proportionate
(?)Uncertain
effect on
Taxpayers / Consumers,
as the effectiveness of
147
 kom (2016) 0853 - Ingen titel
1715591_0148.png
Impact on stakeholders
the future different national criteria for exemptions in terms of coverage
of the prudentially-relevant entities and staff cannot be assessed
Option 3: Exempt small
and non-complex
institutions and staff
with low variable
remuneration from the
rules on deferral and
pay-out in instruments,
based on harmonised
exemption criteria
defined at EU level
(+)
Positive effect on
Regulators / supervisory authorities,
as the level
of supervision would be tailored to the riskiness of institutions and staff;
some Regulators / supervisory authorities may prefer that the exemption
criteria and thresholds be established at EU level
(++)
Very positive impact on
Institutions,
as the rules would be more
proportionate and also uniform across the EU
(+)
Positive effect on
Employees,
as the rules would be more
proportionate
(≈)
Neutral effect on
Taxpayers / Consumers,
as the prudentially-
relevant entities and staff will continue to be covered
Regulators / supervisory authorities
Under
Option 1
(baseline), regulators / supervisory authorities are not in a position to allow for a
proportionate application of the rules on deferral and pay-out in instruments, in the sense of going
below or dis-applying the
de minimis
thresholds of the Directive. This lack of flexibility would
put them in a position whereby they would need to enforce requirements vis-à-vis institutions and
staff where this might not be warranted from a prudential supervision perspective.
Options 2 and 3
are positively assessed, as they would bring about a level of supervision better
tailored to the prudential riskiness of those supervised (institutions and staff). Regulators /
supervisory authorities may find it an advantage that under
Option 2
they could fix their own
exemption criteria and thresholds. Others could, however, see an added value in having the
exemption criteria defined at EU level as proposed under
Option 3,
in particular when these EU
harmonised exemption criteria would be combined with a possibility for supervisory authorities
to adopt a stricter approach.
Institutions / Shareholders
Under
Option 1
(the current CRD IV provisions), institutions are required to comply with the
deferral and pay-out in instruments requirements in a manner which for some of them triggers
costs/burden disproportionate when compared to the prudential benefits.
Options 2 and 3
are positively assessed from an institutions' perspective, as they would bring
about a higher level of proportionality and a reduction in the institutions’ compliance burden.
Under
Options 2 and 3,
small and non-complex institutions would still need to ensure that their
remuneration practices do not have a negative impact on their long term interest and sound risk
management. However, they would have more flexibility when setting their remuneration
schemes and practices, thereby potentially benefiting from one-off and on-going savings on the
costs, currently estimated to range from € 50 000 to € 500 000. Also large institutions would
benefit from savings on costs currently estimated between € 400 000 and € 5 million with regard
to staff with non-material levels of variable remuneration.
148
 kom (2016) 0853 - Ingen titel
1715591_0149.png
Under
Option 2,
different national exemption regimes would exist, thereby potentially creating
regulatory complexity and unwarranted compliance costs for cross-border activities.
Option 3,
is
in principle not associated with such risks and therefore is assessed as strongly positive.
Employees
Under
Option 1
(baseline), all Identified Staff need to comply with the deferral and pay-out in
instruments requirements, regardless of their level of variable remuneration and the incentives for
excessive risk-taking this may or may not entail. As a result, in cases of staff with low levels of
variable remuneration, there can be instances of perceived decrease of the overall value of
remuneration (because of its deferral in time and its pay-out in instruments, as opposed to cash),
and resulting from this detrimental motivational effects, without this being associated with clear
prudential benefits.
Options 2 and 3
would ensure an application of the remuneration rules that is proportionate
given the rather limited prudential usefulness of deferral and pay-out in instruments in the case of
these staff with low levels of individual variable remuneration.
Option 3
has the advantage of
ensuring that staff with low levels of remuneration is in principle subject to equal treatment
across the EU.
Tax payers / Consumers
Option 1
(current CRD IV provisions), by requiring institutions and their Identified Staff to
comply with certain remuneration rules, contributes to enhancing risk management through
remuneration policies and thus contributes to fostering financial stability to the benefit of tax-
payers / consumers.
The impact of
Option 2
is uncertain, as the effectiveness of the future different national criteria
for exemptions in terms of coverage of the prudentially-relevant entities and staff cannot be
assessed.
Option 3
is expected to also preserve the interests of tax-payers and consumers, by ensuring that
all the prudentially-relevant (potentially risky) institutions and staff will continue to be subject to
the rules. They are therefore assessed as having a neutral effect on tax-payers / consumers
compared to the baseline scenario.
Problem 2
Impact on stakeholders
Option 1: No policy
change
Option 2: Allow listed
institutions to use
share-linked
instruments in addition
or instead of shares in
fulfilment of the
requirement under
Article 94(1)(l)(i) CRD
IV
n.a.
(≈)
Neutral impact on Regulators / supervisory authorities, as the extent of
supervision stays the same
(++)
Very positive effect on Institutions, as listed firms will be allowed to reach
the same prudential results through the less costly means of using share-linked
instruments instead of or in addition to shares; moreover, shareholders will no
longer be faced with shareholdings dilutions each time shares are issued for
remuneration purposes
149
 kom (2016) 0853 - Ingen titel
1715591_0150.png
Impact on stakeholders
(+)
Positive impact on Employees through greater flexibility induced by share-
linked instruments (e.g. Employees no longer faced with potential insider trading
problems when selling their shares)
(≈)
Neutral impact on Taxpayers / Consumers, as the prudential objectives of the
rule will be met to the same extent
Regulators / supervisory authorities
Under
Option 1,
regulators / supervisory authorities would not be in a position to allow listed
institutions a proportionate application of the rule on pay-out in shares.
Option 2
is assessed as having a neutral effect on regulators / supervisory authorities, as the
extent of supervision and the risk profile of supervised entities would not change compared to the
current situation.
Institutions / Shareholders
Under
Option 1
(current CRD IV provisions), listed institutions are required to comply with the
pay-out in instruments requirement by means of shares only. This triggers difficulties and
administrative burden for the institution and its shareholders. Institutions would need to either
create new shares or purchase them in the market. Both are complex processes.
Existing shareholders can be confronted with a dilution of their rights. The staff members that
receive shares can be confronted with problems of insider trading (e.g. problems with selling
shares received as remuneration).
Under
Option 2,
allowing listed institutions to use share-linked instruments in addition to or
instead of shares in fulfilment of the requirement under Article 94(1)(l)(i) CRD IV would reduce
compliance cost and administrative burden for these institutions. Moreover,
Option 2
is
positively assessed from the shareholders’ perspective, as it prevents the dilution of
shareholdings and the disruption of shareholding structures through repeated awards of shares for
remuneration purposes.
Employees
Under
Option 1,
all Identified Staff of listed institutions would receive part of their variable
remuneration in shares. Depending on their role in the institution, they might be confronted with
insider dealing problems if trying to sell these shares. This method of paying out variable
remuneration might not be the most flexible/convenient from the employees’ perspective, and
might lead to a perceived deterioration in the overall value of remuneration and/or detrimental
motivational effects.
Option 2
would allow Identified Staff to be remunerated in share-linked instruments in addition
to or instead of shares. This would provide staff with more flexibility in benefitting from the
awarded instruments (for instance by avoiding a potential situation in which staff may not be able
to sell the shares after the retention period because of insider dealing concerns).
Tax payers / Consumers
150
 kom (2016) 0853 - Ingen titel
1715591_0151.png
Option 1,
by requiring listed institutions to pay part of their variable remuneration in shares,
contributes to enhancing risk management through remuneration policies and thus contributes to
fostering financial stability to the benefit of tax-payers / consumers.
Option 2
is expected to preserve the interests of tax-payers and consumers, as the prudential
objectives of the pay-out in instruments requirement are reached to the same extent through pay-
out in shares as they are through pay-out in share-linked instruments.
A
NNEX
5. B
ACKGROUND TO CUMULATIVE IMPACT ASSESSMENT
Annex 5.1. Estimation of costs of FRTB and LR using the QUEST model
Introduction
Table A11.
Regulatory measures and their impact on bank capital requirements.
Policies
FRTB
FRTB only (change
+ Leverage Ratio (
in RWA calculation
at 3% Total
method)
Assets)
10.77%
11.17%
Baseline
Average capitalization as a share of
RWA
average absolute variation in PP wrt
baseline (share of RWA)
Average capitalization as a share of
TA
average absolute variation in PP wrt
baseline (share of TA)
Source: Commission calculations
10.50%
0.27
0.67
3.81%
4.05%
0.24
The QUEST model is well suited to assess the costs of regulatory constraints but is less
developed to also assess the benefits. For this purpose the model would need to be extended to
allow for a better modelling of how regulation affects risk taking by banks. By taking into
account the risk taking channel the model could also be used to assess how regulations affect the
probability of the economy being hit by large negative shocks (financial crises). The model can
still be used to look at the cost of regulation in normal times. The major effect of regulation
which is captured by the model is the impact of bank funding costs which are then transmitted
onto lending rates and increase capital costs for non-financial firms with negative effects on their
investment. There is a cost effect because an increase in capital requirements shifts funding from
deposits to bank capital and the cost of capital for banks is larger than the cost on deposits.
The size of this cost effect from changing the financing structure of banks is, however, not
undisputed among economists. For example Admati and Hellwig (2012)
136
argue that because of
136
Admati, A., DeMarzo, P., Hellwig, M. and Pfleiderer, P. (2010). “Fallacies, irrelevant facts, and myths
in capital regulation: why bank equity is not expensive”. Stanford University Working Paper no. 86.
151
 kom (2016) 0853 - Ingen titel
1715591_0152.png
the change in the composition of liabilities of the bank does not fundamentally change the
riskiness of lending a larger share of bank capital should reduce the risk premium since the total
risk of the bank is now borne by a larger equity base. This argument is based on the Modigliani
Miller (MM)
137
theorem. However, others argue that MM does not apply for banks because of an
implicit bail out subsidy. Therefore increasing the capital base is shifting the risk from the public
to shareholders. Assessments of bank regulations carried out by the BIS (BIS (2010a
138
, 2010b
139
,
2010c
140
) follow this argument and they assume that there is no offsetting effect on risk premium.
There are also many micro banking studies who look at this effect. They usually come to the
result that there is at least a partial reduction of the risk premium on capital if capital
requirements are increased (see, for example, Miles et al. (2013) and Kashyap et al. (2010) ). The
relatively detailed study by Miles et al. suggests that the risk premium effect is such that it offsets
about 50% of the increase in funding costs compared to a situation where the equity premium is
kept unchanged.
To calibrate the key features of the model the following assumption have been made: the ratio of
loans to GDP is set at 108%; the ratio of bonds to loans is set at 28%; the ratio of bank capital to
total assets (leverage ratio) is set at 3.81% and the ratio of bank capital to risk-weighted assets at
10.5%. In the simulations shown below, the combined effect of the change in FRTB and leverage
is presented. We consider two scenarios. In the first scenario it is assumed that the equity
premium of bank capital remains unchanged and in the second scenario we present results where
the equity premium is reduced in such a way that the funding cost increase under the first
scenario is halved. Notice, under the first hypothesis discussed above, namely that MM holds
fully, there would be no macroeconomic cost associated with an increase in capital requirements.
In principle the macroeconomic effects from changes in RWA and changes in TA should be very
similar, since the change in the two ratios represents identical policy measures which are only
expressed in a different metric. We have conducted the policy experiment both w. r. t a change in
RWA and a change in TA but in the note we only report results related to the RWA experiment,
which gives a slightly larger cost estimate in terms of GDP.
Scenario 1:
Increase in capital requirements with constant equity premium on bank capital
The regulation induces banks to increase capital relative to deposits. This has two opposing
effects on funding costs. Shifting to bank capital and paying an equity premium, increases
funding costs, while lowering the demand for deposits reduces the deposit rate, which lowers
funding cost. The latter effect is, however, extremely small, this applies especially in the current
juncture with effectively zero deposit rates, thus the first effect dominates. Optimising banks
shift the higher funding costs onto the non-financial private sector in the form of higher loan
rates. This increases capital costs for firms which partly finance their investment with loans.
Consequently the cost of the regulatory measures affect the real economy via reduced investment.
Since capital costs are permanently increased the economy moves to a lower capital output ratio
and a permanently lower (relative to the baseline) level of GDP. As shown in Table A12, higher
137
138
139
140
Modigliani, F. and Miller, M. (1958). “The cost of capital, corporation finance and the theory of
investment”, American Economic Review, vol. 48(3), pp. 261-97.
BIS (2010a). “An assessment of the long-term economic impact of stronger capital and liquidity
requirements”, Basel Committee on Banking Supervision, Bank for International Settlements.
BIS (2010b). “Assessing the macroeconomic impact of the transition to stronger capital and liquidity
requirements”, Basel Committee on Banking Supervision, Bank for International Settlements.
BIS (2010c). “Results of the comprehensive quantitative impact study”, Basel Committee on Banking
Supervision, Bank for International Settlements.
152
 kom (2016) 0853 - Ingen titel
capital requirements in terms of risk weighted assets of 0.67pp reduces the level of GDP in the
long run by 0.06%. This effect is mostly generated by a decline of investment, which is reduced
by 0.15%. GDP falls less than investment (capital) in the long run since long run employment
levels are hardly affected. This is due to the fact that real wages are adjusted downward (relative
to the baseline) because of the decline in productivity associated with a fall in capital, this wage
behaviour stabilises employment.
Table A12.
Increase in ratio of bank capital-to-risk-weighted assets (FRTB+leverage, 0.67pp)
2014
Y
I
C
LO
L
RLO
-0.02
-0.14
0.01
-0.04
-0.02
2.71
2015
-0.01
-0.19
0.03
-0.07
-0.01
0.59
2016
-0.01
-0.2
0.03
-0.07
-0.01
3.72
2017
-0.02
-0.2
0.02
-0.07
-0.01
2.91
2020
-0.02
-0.2
0.01
-0.08
-0.01
2.61
2030
-0.04
-0.17
-0.02
-0.11
-0.01
2.64
2050
-0.06
-0.16
-0.05
-0.14
-0.01
2.69
2150
-0.06
-0.15
-0.06
-0.15
-0.01
2.71
Note: Y: GDP, I: Investment, C: Consumption, LO: Stock of loans, RLO: Loan rate, L: employment.
Y, C, I, LO, L are % deviations from baseline levels. RLO is the deviation from the baseline level in BP.
In the simulation experiment where the capital requirement in terms of total assets is increased by
0.24pp yields a long term GDP effect of -0.05%.
Table A13.
Increase in ratio of bank capital-to-risk-weighted assets (FRTB+leverage, 0.67pp)
(with 50% MM offset)
2014
Y
I
C
LO
L
RLO
-0.01
-0.07
0.01
-0.02
-0.01
1.37
2015
-0.01
-0.1
0.02
-0.04
-0.01
0.3
2016
-0.01
-0.1
0.02
-0.04
0
1.89
2017
-0.01
-0.1
0.01
-0.03
-0.01
1.48
2020
-0.01
-0.1
0.01
-0.04
0
1.33
2030
-0.02
-0.09
-0.01
-0.06
0
1.34
2050
-0.03
-0.08
-0.02
-0.07
0
1.36
2150
-0.03
-0.08
-0.03
-0.08
0
1.38
Note: Y: GDP, I: Investment, C: Consumption, LO: Stock of loans, RLO: Loan rate, L: employment.
Y, C, I, LO, L are % deviations from baseline levels. RLO is the deviation from the baseline level in BP.
153
 kom (2016) 0853 - Ingen titel
1715591_0154.png
Annex 5.2. Estimation of benefits of FRTB and LR using the SYMBOL model
This report is an assessment of the effects of the implementation of the fundamental review of the
trading book (FRTB, henceforth) as envisaged by CRR 2 proposals and of requirements on
leverage ratio (LR, henceforth). This analysis includes the following steps:
1.
Estimation of average risk-weights for trading activities and non-trading activities.
This analysis grounds on a panel regression methodology already developed for the
impact assessment of bank structural separation.
Estimation of impacts of the FRTB on RWAs for banks based on expected changes
to risk-weights (the median impact estimated by the EBA is used as input).
RWAs estimated as per points 1 and 2 are used as inputs to run simulations of bank
losses through the SYMBOL model.
2.
3.
We aim to estimate the potential benefits for public finances of implementing:
1.
2.
the new rules established by the FRTB
the new binding requirements concerning LR.
Benefits for public finances are measured as a decrease in the potential costs due to bank defaults
and recapitalization needs that would remain uncovered by the available tools setup in the EU
legislation, thus potentially hitting Public Finances.
Section 1 - Panel analysis to estimate risk weighted assets for trading activities and non-
trading activities
In this section, we provide the description of the dataset and the empirical application in order to
estimate risk weighted assets. This panel regression analysis builds on a work developed for the
impact assessment of bank structural separation.
Dataset
In order to predict individual banks’ RWAs, we identify 9 categories of assets and
liabilities, summarised in Table A14.
Table A14.
List of assets and liabilities included in the preferred model to estimate
RWAs
Short
name
LB
NCL
AMZ
HTM
AFS
FV
TSA +
TSL
DA+DL
Description
Net loans to banks
Net loans to customers
Total assets held at amortised cost excluding loans to banks and customers held at amortised
cost
Securities held to maturity
Available for sale assets excluding loans
Assets held at fair value excluding loans
Securities held for trading excl. derivatives (volume in assets and liabilities side)
Derivatives held for trading (volume in assets and liabilities side)
154
 kom (2016) 0853 - Ingen titel
1715591_0155.png
DHV
Derivatives held for hedging purposes (volume in assets and liabilities side)
Empirical results
(1
)
where
of Basel III, and
is the dummy variable that represents the entry into force
are dummy variables for time-fixed effects. We estimate the model by
running a fixed-effect regression and we build standard errors through a robust clustered variance
estimator on the 194 banks. Table
A15A15
reports estimated coefficients of Equation (1) from
panel regression of RWAs on the categories of assets listed in Table A1A14. The coefficients for
net loans to banks and customers, and the total assets held at amortised cost are positive and
statistically significant (i.e., p-value are less than 1%). The estimated coefficients
and
for
the securities held to maturity, and for assets held at a fair value, have a positive effect on RWA
but they are not significant at a statistical significant level of 10%. As expected by the economic
theory, the available for sale assets are positive. We observe that the estimated coefficient for the
volume of trading assets gets a positive small value: it decreases from 14% to 3% when we
introduce the interaction with the dummy variable for Basel III. However, this estimate is not
significant. The volume of derivatives for trading is always positive and significant. Finally, the
coefficient for the derivatives held for hedging indicates a negative relation w.r.t. the RWA.
In order to save space, the estimates for the dummies
, with
, are not reported.
The estimated coefficients are all statistically significant at a level of 10%. (Results are
available on request). In table A15, the coefficient of determination R-squared is also reported.
Since the dependent variable RWA is built on the balance sheet values, R-squared of regression
(1) is very high.
Table A15.
Coefficients from the panel regression, P-values are reported in parentheses and *,
**, *** denote significance at 10, 5 and 1 percent significance level, respectively. Moreover,
+
denote significance at 20 percent
.
Eq. (1)
LB
NCL
AMZ
HTM
AFS
FV
0.5 * (TSA + TSL)
0.5 * (TSA + TSL) *
d
B3
155
0.3378***
(0.001)
0.4409***
(0.000)
0.5022***
(0.000)
0.4640
(0.355)
0.1649*
(0.084)
0.1426
+
(0.113)
0.1369
+
(0.161)
0.0348
(0.684)
 kom (2016) 0853 - Ingen titel
1715591_0156.png
0.5 * (DA + DL)
0.5 * (DA + DL) * d
B3
DHV
Number of obs
Number of groups
R-squared:
within
between
overall
0.0670***
(0.001)
0.1184***
(0.005)
-1.292***
(0.000)
1.462
194
0.6760
0.9590
0.9491
C
onsidering a significance level of at least 20 percent, the coefficients of
and
can be dropped. Thus, let us consider the following hypothesis:
, the constrained model used to predict RWAs is:
(2
)
Model (2) is nested within model (1). That is, model (2) has a smaller number of
parameters than (1). We compute the F-test in order to ensure that the model (2) fit to the
data. We test the null hypothesis
, and we do not reject it. The
estimation results from regressions (2) and (1) are very similar. Models (1) and (2) fit in
a similar way the data.
Estimation of trading activities RWA and non-trading activities RWA
This step of the analysis builds on the results of the panel regression to estimate the
portion of RWA that can be affected by the FRTB (proxied by “market risk RWA”). To
this end, each category of assets is attributed to one of the two lines of activities (i.e.
trading and all the rest) and RWAs of each activity are predicted according to the
relevant coefficients obtained in the econometric model.
Table A16.
Allocation of assets and liabilities categories to trading and non-trading
activities
Short
name
LB
NCL
AMZ
HTM
AFS
FV
Category
Net loans to banks
Net loans to customers
Total assets held at amortised cost excl. loans to banks and customers
held at amortised cost
Securities held to maturity
Available for sale assets excluding loans
Assets held at fair value excl. loans
156
Approach for RWA
allocation
non-trading
non-trading
non-trading
non-trading
non-trading
non-trading
 kom (2016) 0853 - Ingen titel
1715591_0157.png
TSA +
TSL
DA+DL
DHV
Securities held for trading excl. derivatives (assets & liabilities)
Derivatives held for trading (assets & liabilities)
Derivatives held for hedging purposes (assets & liabilities)
trading
trading
Proportional allocation
Predicted trading and non-trading RWAs are calculated for each bank and are then re-
normalised to sum up to the total RWAs as reported in balance sheet.
Results
As shown in Figure 1, the scatterplot shows an increasing relationship between the
dimension of trading assets (X-axis) and the related RWAs (Y-axis). Moreover, the share
of trading RWAs ranges from 0 to 10%/15% of total RWSs for most of the institutions
analysed, and only few of them present higher shares, from 15% to around 40%. Figure
A12 complements this information: the vast majority of small and medium size banks
have a share of trading RWAs below 5%, while large banks are the ones most involved in
trading activities. It can also be observed that the share of trading RWAs is in general
quite steady over time, with the exception of 2014 where it seems to be increased,
especially for the banks in the sample (this is probably due to the increased coefficient
for 2014 wrt the previous years, which is confirmed by alternative model specifications).
Figure A13 focuses on the 2014 by showing the frequency distribution of the share of
trading RWAs. The histograms confirms that the distribution is very skewed: most of the
considered sample has a share below 5% and only few outliers present a share of trading
RWAs that spans from 10% to 35%.
These results are in line with the composition of RWAs reported by the EBA in the
“CRD IV–CRR/Basel III monitoring exercise report”.
141
The EBA splits total RWA into
5 components: credit risk (attributable to non-trading activities), CVA (trading activities),
market risk (trading activities), operational risk (that can be proportionally attributed to
trading and non-trading activities) and other RWA. In 2014 the sum of market risk, CVA
and the share of operational risk is around 10% for group 1 banks (roughly 4% for group
1 banks). This number is consistent with the average share of trading RWA as estimated
by our model for the large banks (13%) and for medium/small banks (around 3%).
141
https://www.eba.europa.eu/documents/10180/950548/CRD+IV++CRR+-
+Basel+III+monitoring+exercise+report.pdf/f414a01e-4f17-4061-9b88-4e7fb89cc355
In particular, see figure 7
157
 kom (2016) 0853 - Ingen titel
1715591_0158.png
Figure A11.
Scatter plot of estimated share of trading assets and estimated share of
trading risk weights (2006-2014).
.5
0
.1
.2
.3
.4
0
.2
.4
Share of Trading Assets
.6
.8
Figure A12.
Box plots of share of trading RWAs by groups of institutions: small (total
assets below 30 bn €), medium (total assets from 30 to 500 bn €) and large (total assets
above 30 bn €). One box-plots for each year from 2006 to 2014.
.45
.05
0
06
07
Share of trading RWAs
.1
.15
.2
.25
.3
.35
.4
08
09
10
11
12
13
14
06
07
08
09
10
11
12
13
14
06
07
08
09
10
11
12
13
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Small (<30bn)
Medium (30-500bn)
Large (>500bn)
158
20
14
 kom (2016) 0853 - Ingen titel
1715591_0159.png
Figure A13.
Histogram of the share of trading RWAs in 2014
150
2014
Frequency
0
0
50
100
.1
.2
Share of trading RWAs
.3
.4
Share of trading RWAs (predicted, 2014)
liiksizeclass
variable
min
0
0
.0099959
0
p25
.0002282
.0043999
.0500465
.0012238
mean
.0111138
.0296267
.1251117
.0297103
p50
.002609
.0161946
.0850369
.0099578
p75
.0118071
.0411101
.2369824
.0354806
max
.1332347
.1706299
.3500271
.3500271
iqr
.0115789
.0367101
.1869359
.0342568
N
92
76
18
186
Small (<30bn)
rwatrb..
Medium (30-500bn) rwatrb..
Large (>500bn)
rwatrb..
Total
rwatrb..
Total trading RWAs (predicted, 2014)
liiksizeclass
variable
min
0
0
2326904
0
p25
392.4134
99195.16
1.48e+07
7776.5
mean
66632.49
1729398
5.55e+07
6107522
p50
10242.5
508756
2.30e+07
86899.3
p75
64110.25
2214900
1.22e+08
893023
max
874550.4
1.86e+07
1.52e+08
1.52e+08
iqr
63717.84
2115704
1.07e+08
885246.5
N
92
76
18
186
Small (<30bn)
r~rb_fvc
Medium (30-500bn) r~rb_fvc
Large (>500bn)
r~rb_fvc
Total
r~rb_fvc
Total Trading RWAs (predicted, 2014)
liiksizeclass
variable
N
92
76
18
186
sum
6130189
1.31e+08
9.98e+08
1.14e+09
Small (<30bn)
r~rb_fvc
Medium (30-500bn) r~rb_fvc
Large (>500bn)
r~rb_fvc
Total
r~rb_fvc
Section 2 - Estimation of impacts on RWAs based on expected changes to risk-
weights due to the introduction of the FRTB and LR requirements
RWAs can also be modified to obtain a counterfactual scenario representing the full
implementation of CRD IV–CRR requirements.
159
 kom (2016) 0853 - Ingen titel
1715591_0160.png
In order to develop a scenario representing the full implementation of Basel III rules, we
apply correction factors to RWA and Total capital of banks in the sample. Such
correction factors come from the EBA’s report “CRD IV–CRR/Basel III monitoring
exercise report”. The study conducted by the EBA analyses that banks are still subject to
transitional arrangements at the current
142
implementation stage of CRD IV–CRR. This
results in a reduction in the level of capital for both Group 1 and Group 2 banks and a
slight increase in RWAs under full implementation.
Table A17.
Changes in total capital and RWA relative to the current amounts
Total
Capital
Group 1
G-SIBs
Group 2
Large Group 2
Medium Group 2
Small Group 2
Source: EBA
Potential impact of FRTB on market risk RWAs
The EBA estimated the potential impact on RWAs coming from the new requirements of the
FRTB. According to the EBA, the median impact is a 27% increase of market risk RWAs.
RWA
0.1%
0.0%
0.9%
1.3%
0.6%
0.0%
-12.8%
-13.3%
-7.0%
-7.1%
-7.2%
-6.1%
Simulations are conducted using input data as of 2014. This analysis is based on the
expected changes to risk-weights due to the introduction of the FRTB and LR.
142
as of December 2014
160
 kom (2016) 0853 - Ingen titel
1715591_0161.png
Section 3 - Potential impact on public finances
The analysis presented in this section estimates the potential benefits for public finances of
implementing the new rules established by the FRTB and of the new requirements on LR.
Benefits for public finances are measured as a decrease in the potential costs due to bank defaults
and recapitalization needs
143
that would remain uncovered by the available tools (i.e. the safety-
net) setup in the EU legislation, thus potentially hitting Public Finances.
We assume that the safety-net that can intervene to cover losses and recapitalization needs
includes the bail-in tool, Resolution Funds (RF), as well as the improved standards on minimum
capital requirements and capital conservation buffer set up in the CRR/CRD IV package.
Banking losses are simulated using the SYMBOL model (Systemic Model of Banking Originated
Losses). SYMBOL simulates losses for individual banks using information from their balance
sheet data. Capital is the first source to absorb losses. We assume that a bank goes into
insolvency when simulated losses are larger than the available level of capital (the difference
between the loss and capital is the excess loss). Moreover, we also consider recapitalization needs
to reflect the minimum capitalization under which a bank can be considered viable. We refer to
excess losses plus recapitalization needs as
financing needs
hereafter. In case capital is not
sufficient, the bank makes use of its bail-in-able liabilities. Since data on the actual amount of
bail-in-able liabilities held by banks are not available, we assume that each bank has a total loss
absorbing capacity that is twice the minimum capital requirement. In other terms, banks are
assumed to hold an amount of bail-in-able liabilities that is equal to the minimum amount of
capital. In a next step, in case there would be financing needs after the bail-in intervention, the
RF can intervene. We assume that a single RF has at its disposal a target fund equal to 1% of the
amount of covered deposits of banks in the sample. Moreover, the RF can cover financing needs
up to a ceiling equal to 5% of each bank’s total assets. The remaining financing needs will remain
uncovered.
Dataset
Data used for the present exercise are as of 2014. The sample has 183 banks in EU and covers
83% of the EU TA.
144
Scenarios
The scenarios implemented in this analysis aim to represent the case where the FRTB and LR
requirement are not in place (baseline), the implementation of the FRTB (scenario 1), and a final
situation where both FRTB and LR requirements are in force (scenario 2 and scenario 3).
In all scenarios, the real riskiness of bank assets is assumed to be in line with an RWA amount
fully compliant with Basel III rules and the FRTB. This means that the balance sheet value of
RWA is adjusted by applying the following correction:
The baseline and the alternative policy scenarios differ from each other by the assumed level of
capital held by banks and the amount of recapitalisation needs.
143
144
The recapitalization need is the amount necessary to allow banks suffering from losses to continue
operating on an on-going basis.
We use the amount of total assets in the banking sector excluding branches as provided by ECB as
reference for the population.
161
 kom (2016) 0853 - Ingen titel
1715591_0162.png
Baseline:
all banks are assumed to be fully compliant with Basel III rules only. This implies the
Basel III correction to the definitions of risk-weighted assets and capital, as per Table A17A17.
As for the initial capital, we consider both the case where all banks hold the minimum capital
requirement (MCR) plus the capital conservation buffer (CCB), as per CRR/CRD IV (i.e. 10.5%
of RWA under a full implemented Basel III environment) and an alternative case where each
bank is assumed to hold at least 10.5% of RWA, while keeping any excess buffer (topping the
capital up to 10.5% of RWA). As for the recapitalisation needs, the viability requirement is set to
8% of RWA.
Scenario 1:
The FRTB is in place. The amount of capital has been set both to the minimum (i.e.
10.5% of RWA fully compliant with the FRTB) and to the maximum between the actual level
and the minimum. The viability requirement for recapitalisation is 8% of RWA under FRTB.
Scenario 2:
The leverage ratio requirement is in place in addition to the FRTB. All banks hold
the minimum regulatory requirement for total capital (10.5% of RWA compliant with the FRTB)
or an amount in line with the minimum requirement for the LR (i.e. 3% of total assets
145
),
whichever is higher. Also in this scenario two alternative options have been considered: staying
at the minimum and topping up actual capital. The level of recapitalisation takes into account also
the LR requirement (i.e. recapitalisation is at the highest of 8% RWA FRTB and 3% TA).
Scenario 3:
This scenario differs only in the chosen level of the LR requirement from scenario 2.
We assume that banks may hold an extra buffer on top of the 3% minimum, that we set to 4%
146
of total assets.
Table A18.
Scenarios implemented
Scenario
Total regulatory capital
No buffers
Baseline
Top up
Scenario 1
No buffers
Top up
Scenario 2
Top up
Max{K, 10.5% RWA
10.5% RWA
no FRTB
no FRTB
Recapitalization levels
8% RWA
no FRTB
}
8% RWA
FRTB
}
Max(8% RWA
FRTB
, 3%TA)
10.5%∙RWA
FRTB
Max{K, 10.5% RWA
FRTB
No buffers Max{10.5%∙RWA
FRTB
, 3%TA}
Max{K, 10.5%∙RWA
FRTB
,
3%TA}
Max{K, 10.5%∙RWA
FRTB
,
4%TA}
No buffers Max{10.5%∙RWA
FRTB
, 4%TA}
Scenario 3
Top up
Max(8% RWA
FRTB
, 3%TA)
Results
145
146
The policy refers to Tier 1 capital. However, due to technical reasons, the current version of the
SYMBOL model is not able to keep track separately of T1 and T2 capital. The requirement is thus
considered with respect to Total Regulatory Capital, leading to a slight under-estimate of the increase in
capital needs.
4% is close to the same amount of “relative” buffer afforded by the CCB on the 8% MCR (i.e.
3*10.5/8=3.9)
162
 kom (2016) 0853 - Ingen titel
1715591_0163.png
A set of simulations has been run for each scenario. The resulting distributions are
presented showing percentiles of the simulated distribution in the tables below, both in
terms of share of EU GDP and in billion Euro.
The simulation model runs on a representative sample of EU banks. In order to show
results related to the entire population of banks, results based on the sample are upscaled
by using a sample coverage ratio based on total assets.
Table A19.
Distributions of financing needs
FN
(i.e. potential costs for public finances
due to bank defaults and recapitalization needs) for all scenarios,
no buffers. FN
are
reported as a share of EU GDP
Baseline:
Percent
iles
Scenario 1
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.02%
0.02%
0.02%
0.03%
0.04%
0.05%
0.09%
0.16%
0.24%
0.34%
0.63%
2.16%
2.75%
2.81%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.33%
0.79%
1.87%
1.95%
2.08%
2.21%
2.32%
2.49%
2.71%
3.01%
3.43%
3.75%
4.23%
5.18%
8.31%
9.43%
9.55%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.02%
0.02%
0.03%
0.03%
0.05%
0.08%
0.13%
0.21%
0.29%
0.57%
2.05%
2.61%
2.67%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.60%
1.17%
2.38%
2.47%
2.63%
2.76%
2.89%
3.08%
3.32%
3.64%
4.10%
4.44%
4.93%
5.93%
9.13%
10.28%
10.40%
Scenario 2
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.04%
0.25%
0.26%
0.28%
0.32%
0.34%
0.39%
0.44%
0.53%
0.65%
0.78%
0.92%
1.25%
2.85%
3.45%
3.52%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.02%
0.02%
0.02%
0.03%
0.03%
0.05%
0.06%
0.10%
0.17%
0.26%
0.37%
0.68%
2.27%
2.88%
2.94%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.28%
0.70%
1.72%
1.79%
1.91%
2.02%
2.14%
2.29%
2.50%
2.78%
3.19%
3.50%
3.95%
4.87%
7.92%
9.03%
9.14%
Scenario 3
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.02%
0.15%
0.16%
0.17%
0.19%
0.22%
0.25%
0.28%
0.35%
0.44%
0.54%
0.64%
0.89%
2.11%
2.58%
2.63%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.01%
0.01%
0.02%
0.02%
0.03%
0.05%
0.08%
0.13%
0.18%
0.35%
1.53%
2.00%
2.05%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.34%
0.81%
1.90%
1.98%
2.11%
2.24%
2.36%
2.52%
2.75%
3.05%
3.47%
3.80%
4.27%
5.23%
8.36%
9.49%
9.61%
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.04%
0.23%
0.24%
0.26%
0.30%
0.32%
0.37%
0.41%
0.50%
0.62%
0.74%
0.88%
1.20%
2.74%
3.32%
3.38%
FN after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.03%
0.21%
0.22%
0.24%
0.28%
0.30%
0.34%
0.38%
0.47%
0.58%
0.70%
0.82%
1.13%
2.62%
3.19%
3.25%
80
82
84
86
88
90
92
95
97.5
99
99.5
99.9
99.91
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.985
99.99
99.995
99.999
99.9999
100
Table A20. Distributions of financing needs
FN
(i.e. potential costs for public finances
due to bank defaults and recapitalization needs) for all scenarios, no buffers.
FN
are
reported in billion Euro
Baseline:
Percent
iles
FN after
capital
80
82
84
0.00
0.00
0.00
Scenario 1
FN
after
RF
0.00
0.00
0.00
FN
after
RF
0.00
0.00
0.00
163
FN
after
capital
0.00
0.00
0.00
Scenario 2
FN
after
bail-in
0.00
0.00
0.00
FN
after
RF
0.00
0.00
0.00
Scenario 3
FN
after
bail-in
0.00
0.00
0.00
FN
after
RF
0.00
0.00
0.00
FN
after
bail-in
0.00
0.00
0.00
FN after
capital
0.00
0.00
0.00
FN after
bail-in
0.00
0.00
0.00
FN after
capital
0.00
0.00
0.00
 kom (2016) 0853 - Ingen titel
1715591_0164.png
86
88
90
92
95
97.5
99
99.5
99.9
99.91
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.985
99.99
99.995
99.999
99.9999
100
0.00
0.00
0.00
0.00
0.00
0.00
46.65
111.40
260.61
271.66
289.83
307.21
323.14
346.32
377.11
418.29
476.28
520.92
586.12
717.66
1,147.55
1,301.90
1,317.90
0.00
0.00
0.00
0.00
0.00
0.00
1.52
5.15
31.63
33.60
35.27
41.27
44.12
50.68
56.21
68.89
85.19
102.11
120.76
164.13
375.45
455.82
464.18
0.00
0.00
0.00
0.00
0.00
0.00
0.03
0.11
1.98
2.16
2.42
3.15
4.23
5.49
7.47
12.85
21.27
32.52
46.29
86.31
296.93
377.25
385.60
0.00
0.00
0.00
0.00
0.00
0.00
45.21
108.87
256.61
267.53
285.43
302.70
318.45
341.48
371.99
412.92
470.45
514.93
579.72
710.64
1,139.57
1,293.64
1,309.61
0.00
0.00
0.00
0.00
0.00
0.00
1.47
4.57
28.91
30.72
32.31
37.83
40.48
46.64
51.96
64.19
79.36
95.53
113.13
155.14
359.28
437.12
445.21
0.00
0.00
0.00
0.00
0.00
0.00
0.03
0.11
1.73
1.94
2.18
2.66
3.66
4.67
6.48
11.24
18.39
28.65
40.38
77.69
280.68
358.43
366.51
0.00
0.00
0.00
0.00
0.00
0.00
82.76
159.86
326.45
338.64
360.13
379.24
396.48
421.91
455.75
499.87
562.68
608.49
676.47
814.00
1,253.08
1,409.92
1,426.18
0.00
0.00
0.00
0.00
0.00
0.00
1.62
5.98
34.12
35.84
37.90
43.83
46.63
53.55
59.82
72.79
89.61
106.66
126.66
171.12
390.35
473.60
482.26
0.00
0.00
0.00
0.00
0.00
0.00
0.03
0.12
2.35
2.56
2.97
3.78
4.55
6.23
8.46
14.31
23.67
35.86
51.10
93.20
311.87
395.07
403.72
0.00
0.00
0.00
0.00
0.00
0.00
38.32
95.76
235.49
245.45
261.40
277.54
293.00
314.49
343.47
381.74
437.26
479.85
541.81
667.65
1,087.12
1,238.38
1,254.06
0.00
0.00
0.00
0.00
0.00
0.00
0.90
2.82
20.02
21.73
23.04
26.41
29.65
34.36
37.92
47.93
60.37
74.08
87.14
121.67
289.24
353.86
360.59
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.09
1.04
1.15
1.36
1.55
2.47
2.87
4.06
7.24
11.24
18.05
24.66
48.43
210.51
274.30
280.94
Table A21.
Distributions of financing needs
FN
(i.e. potential costs for public finances
due to bank defaults and recapitalization needs) for all scenarios,
top up of capital. FN
are reported as a share of EU GDP
Baseline:
Percent
iles
Scenario 1
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.01%
0.03%
0.04%
0.06%
0.09%
0.15%
0.84%
1.12%
1.15%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.11%
0.37%
1.09%
1.15%
1.22%
1.30%
1.40%
1.51%
1.66%
1.86%
2.16%
2.39%
2.74%
3.46%
6.06%
7.01%
7.11%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.02%
0.03%
0.04%
0.07%
0.09%
0.16%
0.87%
1.17%
1.20%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.18%
0.51%
1.33%
1.40%
1.49%
1.58%
1.68%
1.81%
1.99%
2.22%
2.55%
2.80%
3.19%
3.99%
6.77%
7.78%
7.88%
Scenario 2
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.10%
0.11%
0.12%
0.14%
0.15%
0.18%
0.20%
0.25%
0.31%
0.38%
0.47%
0.66%
1.60%
1.97%
2.01%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.01%
0.02%
0.02%
0.04%
0.06%
0.09%
0.12%
0.21%
1.04%
1.38%
1.42%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.13%
0.42%
1.19%
1.26%
1.34%
1.42%
1.52%
1.64%
1.81%
2.03%
2.35%
2.60%
2.97%
3.75%
6.49%
7.49%
7.59%
Scenario 3
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.08%
0.08%
0.09%
0.10%
0.11%
0.13%
0.16%
0.20%
0.26%
0.32%
0.39%
0.56%
1.43%
1.77%
1.80%
FN
after
RF
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.01%
0.03%
0.04%
0.07%
0.09%
0.15%
0.88%
1.19%
1.22%
FN
after
capital
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.10%
0.35%
1.04%
1.10%
1.17%
1.24%
1.34%
1.44%
1.59%
1.79%
2.08%
2.31%
2.64%
3.35%
5.92%
6.86%
6.96%
FN
after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.07%
0.08%
0.09%
0.10%
0.11%
0.13%
0.16%
0.20%
0.25%
0.31%
0.39%
0.55%
1.38%
1.70%
1.73%
FN after
bail-in
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.08%
0.09%
0.09%
0.11%
0.12%
0.14%
0.17%
0.21%
0.27%
0.32%
0.40%
0.58%
1.42%
1.75%
1.78%
80
82
84
86
88
90
92
95
97.5
99
99.5
99.9
99.91
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.985
99.99
99.995
99.999
99.9999
100
164
 kom (2016) 0853 - Ingen titel
1715591_0165.png
Table A22.
Distributions of financing needs
FN
(i.e. potential costs for public finances due to
bank defaults and recapitalization needs) for all scenarios,
top up of capital. FN
are reported in
billion Euro
Baseline:
Percent
iles
Scenario 1
FN
after
RF
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.28
0.44
0.50
0.58
1.25
1.50
2.02
4.01
5.51
8.86
12.12
20.70
114.68
153.98
158.09
FN
after
capital
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
14.69
50.73
149.73
158.02
167.69
178.64
191.68
207.04
227.30
255.39
296.72
328.28
375.58
475.27
831.77
962.16
975.69
FN
after
RF
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.28
0.47
0.54
0.60
1.36
1.65
2.24
4.23
5.95
9.47
12.89
22.11
119.64
160.23
164.47
FN
after
capital
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
24.41
69.30
182.12
192.15
204.60
217.36
230.37
248.15
272.38
304.79
350.05
384.65
437.64
547.75
928.45
1,066.63
1,080.96
Scenario 2
FN
after
bail-in
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.39
0.88
13.64
15.31
16.02
18.95
20.20
24.03
27.96
34.43
43.17
51.58
64.34
90.67
220.01
270.06
275.28
FN
after
RF
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.34
0.68
0.84
1.09
1.83
2.35
3.05
5.25
7.70
11.89
15.87
28.29
142.77
189.64
194.53
Scenario 3
FN
after
bail-in
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.34
0.75
10.30
11.25
11.74
13.44
15.40
18.43
21.60
27.60
35.25
43.98
54.10
77.51
196.09
242.20
247.01
FN
after
RF
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.32
0.48
0.55
0.63
1.10
1.33
1.67
3.71
5.07
8.93
11.75
21.06
121.18
162.84
167.19
FN
after
capital
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
13.51
47.35
142.80
150.68
159.96
170.70
183.30
198.22
217.76
245.13
285.29
316.24
362.40
460.12
812.21
941.14
954.52
FN
after
bail-in
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.31
0.66
10.00
11.35
11.90
13.64
15.49
18.46
22.04
27.19
34.83
42.57
53.30
76.09
188.98
232.95
237.53
FN after
bail-in
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.31
0.67
10.65
12.10
12.70
14.69
16.43
19.58
23.22
28.62
36.57
44.40
55.52
78.92
194.61
239.58
244.27
FN after
capital
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
17.36
57.82
163.50
172.41
183.66
195.48
208.93
225.60
248.14
279.12
322.74
356.99
407.93
514.54
890.80
1,027.66
1,041.86
80
82
84
86
88
90
92
95
97.5
99
99.5
99.9
99.91
99.92
99.93
99.94
99.95
99.96
99.97
99.98
99.985
99.99
99.995
99.999
99.9999
100
As a reference point, a crisis comparable to the last global one is approximately placed
on percentile 99.95 when considering excess losses and recapitalization needs based on
pre-crisis data.
Table A23.
Variation in financial needs when moving between scenarios, percentile
99.95, no buffers
Baseline
to
Scenario 1
-1.40%
-7.97%
-14.95%
Baseline
to
Scenario 2
+21.83%
+5.67%
+13.30%
Baseline
to
Scenario 3
-9.19%
-32.20%
-47.85%
Scenario 1
to
Scenario 2
+23.55%
+14.81%
+33.21%
Scenario 1
to
Scenario 3
-7.90%
-26.33%
-38.68%
Scenario 2
to
Scenario 3
-25.46%
-35.83%
-53.97%
Financial needs after capital
Financial needs after bail-in
Financial needs after RF
Table A24.
Variation in financial needs when moving between scenarios, percentile
99.95, top up capital
Baseline
to
Scenario 1
+4.45%
Baseline
to
Scenario 2
+25.19%
Baseline
to
Scenario 3
+13.81%
165
Scenario 1
to
Scenario 2
+19.86%
Scenario 1
to
Scenario 3
+8.96%
Scenario 2
to
Scenario 3
-9.09%
Financial needs after capital
 kom (2016) 0853 - Ingen titel
1715591_0166.png
Financial needs after bail-in
Financial needs after RF
+6.05%
+10.19%
+30.18%
+56.33%
-0.17%
-11.24%
+22.75%
+41.87%
-5.86%
-19.45%
-23.31%
-43.22%
By looking at the results for the “no buffers” scenarios at percentile 99.95, Table reports
the split of financing needs into losses and recapitalisation needs, the amount of total
financing needs absorbed by capital, bail-in-able liabilities and the RF, and the
Table A25.
Initial Financing needs and absorbed by the safety-net tools, “no buffers”
scenarios, percentile 99.95, billion Euro
Baseline
Financing needs = Losses + Recap needs (FN =
L+R)
of which: Losses (L)
of which: Recap (R)
FN absorbed by Capital
FN absorbed by bail-in-able liabilities
FN absorbed by RF
Leftover FN after RF intervention
862.40
595.20
267.19
516.07
295.64
45.18
5.49
Scenario 1
862.37
595.20
267.17
520.89
294.84
41.97
4.67
Scenario 2
945.32
595.20
350.12
523.41
368.36
47.32
6.23
Scenario 3
850.84
595.20
255.63
536.35
280.13
31.49
2.87
References
Acharya, V and S Viswanathan (2011): "Leverage, moral hazard, and liquidity", Journal of
Finance, vol 66, no 1, pp 99–138.
Allen, F, A Babus and E Carletti (2012): "Asset commonality, debt maturity and systemic risk",
Journal of Financial Economics, vol 104, no 3, pp 519–534.
Angelini, P and A Gerali (2012): “Banks’ reactions to Basel-III”, Bank of Italy Working Paper,
no 876, papers.ssrn.com/sol3/papers.cfm?abstract_id=2118496.
Basel Committee on Banking Supervision (2010): "An assessment of the long-term economic
impact of the stronger capital and liquidity requirements."
Bech, M and T Keister (2013): "Liquidity regulation and the implementation of monetary
policy", BIS Working Papers No 432
Bonfim, D and M Kim (2012): “Liquidity risk in banking: is there herding?”, European Banking
Center Discussion Paper vol 2012-024, Tilburg University European Banking Center.
Brooke, M, O Bush, R Edwards, J Ellis, B Francis, R Harimohan, K Neiss and C Siegert (2015):
"Measuring the macroeconomic costs and benefits of higher UK bank capital requirements", FSA
Paper, no 35
Calomiris, C W, F Heider and M Hoerova (2015): “A theory of bank liquidity requirements”,
CBS Research Paper, vol 14-39, Columbia Business School.
Calomiris, C W and C M Kahn (1991): “The role of demandable debt in structuring optimal
banking arrangements”, American Economic Review, vol 81, no 3, pp 497–513.
Chiaromonte, L. and Casu, B. (2016): "Capital and liquidity ratios and financial distress.
Evidence from the European banking industry". The British Accounting Review, online.
Cornett, M M, J J McNutt, P E Strahan and H Tehranian (2011): “Liquidity risk management and
credit supply in the financial crisis”, Journal of Financial Economics, 101, no 2, pp 297–312,
www.sciencedirect.com/science/article/pii/S0304405X11000663.
166
 kom (2016) 0853 - Ingen titel
de Bandt, O and M Chahad (2015): "A DSGE model to assess the post crisis regulation of
universal banks", Mimeo, University of Paris Ouest and Banque de France-Autorité de Contrôle
Prudentiel et de Résolution.
De-Ramon, S, Z Iscenko, M Osborne, M Straughan and P Andrews (2012): “Measuring the
impact of prudential policy on the macroeconomy: a practical application to Basel III and other
responses to the financial crisis, FSA Occasional Paper, no 42, ..
Dewatripont, M. (Chair), Hancock, D. (Chair) et al. (2016): "Literature review on integration of
regulatory capital and liquidity instruments. Research Task Force of the Basel Committee on
Banking Supervision." Working Paper No. 30. Bank of International Settlements
Diamond, D W and R G Rajan (2001): “Banks and liquidity”, American Economic Review, vol
91, no 2, pp 422–5.
EBA (2015), "EBA report on Net Stable Funding Requirements under article 510 of the CRR"
Elliott, D, J Balta, E Abbhinand, V Korostelina et al. (2016): "Interactions, coherence, and
overall calibration of post crisis Basel reforms.", Oliver Wyman.
Farhi, E and J Tirole (2012): “Collective moral hazard, maturity mismatch, and systemic
bailouts”, American Economic Review, vol 102, no 1, pp 60–93.
Goodhart, C (2011): “Global macroeconomic and financial supervision: where next?”, NBER
Working Paper, no 17682, National Bureau of Economic Research, Cambridge, MA.
Iyer, R., J-L Peydrό, S da-Rocha-Lopes and A Schoar (2013): "Interbank Liquidity Crunch and
the Firm Credit Crunch: Evidence from the 2007-2009 Crisis", Review of Financial Studies,
27(1), 347-372
Kapan, T. and C Monoiu (2014): "Liquidity Shocks and the Supply of Credit after the 2007-2008
Crisis", International Journal of Finance and Economics, 19, 12-23
King, M R (2013): "The Basel III Net Stable Funding Ratio and bank net interest margins",
Journal of Banking & Finance, vol. 37, 4144–4156.
Miles, D, J Yang and G Marcheggiano (2013): “Optimal bank capital”, The Economic Journal,
vol 123, no 567, pp 1–37, onlinelibrary.wiley.com/doi/10.1111/j.1468-0297.2012.02521.x/pdf.
Perotti, E C and J Suarez (2011): “A Pigovian approach to liquidity regulation”, CEPR
Discussion Paper, no 8271, Centre for Economic Policy Research (CEPR), London.
Pessarossi, P. and F Vinas (2015): "Banks’ supply of long-term credit after a liquidity shock:
Evidence from 2007-2009", Débats économiques et financiers n°16, February
167
 kom (2016) 0853 - Ingen titel
1715591_0168.png
A
NNEX
6. I
MPLEMENTATION OF PROPOSED MEASURES
Indicative list of amendments to the CRR, CRD IV and BRRD per topic addressed in the impact assessment
Problem definition / Origin
of the problem
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Area
Solution
Core issues analysed in the main body of the impact assessment
Funding Risk
Excessive
reliance
by
institutions on short-term
wholesale
funding
to
finance their long term
activities.
Existing
requirements in the CRR do
not provide an adequate
framework to ensure that
institutions’
assets
are
sufficiently stably funded by
their liabilities
Origin of the problem:
BCBS, review clause in
article 501 CRR, responses
to the Consultation on the
impact of CRR and CRD
IV on bank financing of the
economy, responses to the
A
single
NSFR
requirement as per Basel
with some adjustments for
all banks.
Some adjustments are
recommended by the
EBA NSFR report to take
into account European
specificities and relate
mainly
to
specific
treatments for:
-Pass-through models in
general
and
covered
bonds
issuance
in
particular;
-Trade
finance
factoring activities;
and
Amending
CRR
Articles 6, 412,
413, 414, 415
New Title IV in
CRR Part 6
New RTS on pass-through
models and extendable
maturities
New Delegated Act on
Derivatives future funding
risk
 kom (2016) 0853 - Ingen titel
1715591_0169.png
Area
Problem definition / Origin
of the problem
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
regulated
guaranteed
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
-Centralised
savings;
-Residential
loans;
-Credit unions.
Other adjustments needed
not to hinder the good
functioning
of
EU
financial markets and the
liquidity of sovereign
bonds markets relate to
the treatment of:
- derivatives transactions;
- short term transactions
with
financial
counterparties;
- Level 1 High Quality
Liquid Assets as defined
in the LCR.
Excessive
Institutions'
leverage
can
A
leverage
169
ratio
CRR Articles 92,
Maintaining empowerment
 kom (2016) 0853 - Ingen titel
1715591_0170.png
Area
Problem definition / Origin
of the problem
increase to unsustainable
levels and have a pro-
cyclical effect on the
financial system
Existing risk-based capital
measures
are
not
sufficiently
reliable
to
address systemic risk and
calls for the introduction of
a simpler and non-risk-
sensitive back-stop measure
In the EU, the leverage ratio
was introduced in the
prudential framework in
2013 but not as specific
capital requirement that
banks must meet.
Origin of the problem: G-20
declarations
147
,
BCBS,
Solution
Level 1 text to be
amended
and
relevant articles
429 to 430 and 511
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
for delegated act under
Article 456(1)(j)
leverage
requirement differentiated
for business models (e.g.
public
development
banks' lending to the
public sector)
or
adjusted for exposure
types (export credits)
147
"Risk-based capital requirements should be supplemented with a simple, transparent, non-risk based measure which is internationally comparable, properly takes into account off-
balance sheet exposures, and can help contain the build-up of leverage in the banking system", Declaration on strengthening the financial system, London summit, 2 April 2009.
London; "developing the leverage ratio as element of the Basel framework", Declaration on Further Steps to Strengthen the Financial System, September 5, 2009, London
170
 kom (2016) 0853 - Ingen titel
1715591_0171.png
Area
Problem definition / Origin
of the problem
review clause in CRR,
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
SME
exposures
The current calibration of
the requirement to address
the credit risk of exposures
to SMEs is not sufficiently
risk-sensitive and reduce the
ability of bank to lend to
SMEs
Origin of the problem:
review clause under CRR,
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Maintaining the SF for
exposures in its current
form (i.e. up to €1.5
million for SA and IRB
banks)
and
complementing it with a
discount of 15% in capital
charges for loans to SMEs
above €1.5 million euros.
CRR (Articles 123,
147, 505)
171
 kom (2016) 0853 - Ingen titel
1715591_0172.png
Area
Problem definition / Origin
of the problem
In the EU there's no
harmonised
minimum
requirement
on
loss
absorption
and
recapitalisation capacity, to
ensure that G-SIBs hold a
sufficient amount of bail in-
able liabilities and make
sure that they can absorb
losses internally without
worldwide
societal
implications or a fiscal
intervention in their favour.
Origin of the problem:
Financial Stability Board,
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
CRR (new Articles
on
eligibility
criteria, deduction,
holdings
and
TLAC
requirement)
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Loss
absorption and
recapitalisation
capacity
Integrate TLAC standard
in MREL rules for EU G-
SIIs
Market risk
The scope of application of
the market risk capital
requirements which is not
Adopt FRTB standards
with a) adjustments to the
calibration and to reflect
172
CRR (Articles 102-
106,
325-377),
CRD IV (Articles
New technical standards on
technical issues
 kom (2016) 0853 - Ingen titel
1715591_0173.png
Area
Problem definition / Origin
of the problem
defined sufficiently clearly.
This allows institutions to
engage
in
regulatory
arbitrage, i.e. they can
allocate some of their
instruments
to
the
regulatory
book
that
generates the lower capital
requirements.
Many features of market
risk are not reflected in the
capital requirement. As a
consequence, the amount of
capital required for certain
instruments is not aligned
with the real risks that
institutions face for these
instruments.
Internal models used by
institutions to calculate
capital requirements for
market risk may generate
very different estimates of
the amount
of capital
Solution
Level 1 text to be
amended
and
relevant articles
83-101)
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
European specificities and
ensure consistency with
other parts of the CRR
(e.g. STS securitisations
and sovereign exposures
)and b) a revised regime
for small trading book
businesses
173
 kom (2016) 0853 - Ingen titel
1715591_0174.png
Area
Problem definition / Origin
of the problem
required
portfolios.
Origin
BCBS
of
for
the
similar
problem:
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Remuneration
Excessive compliance costs
for institutions arising from
the rules on deferral and
pay-out in instruments.
Excessive compliance costs
arising from the requirement
for listed institutions to pay
out part of the variable
remuneration in shares.
Origin of the problem:
review clause in CRD IV,
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
1) Exempt small and non-
complex institutions and
staff with low variable
remuneration from the
rules on deferral and pay-
out in instruments, based
on harmonised exemption
criteria defined at EU
level, combined with a
possibility for competent
authorities to adopt a
stricter approach;
2)
Allow
listed
institutions to use share-
linked instruments in
addition or instead of
shares in fulfilment of the
requirement under Article
CRD IV (Articles
92, 94 )
174
 kom (2016) 0853 - Ingen titel
1715591_0175.png
Area
Problem definition / Origin
of the problem
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
94(1)(l)(i) CRD IV
Insolvency
ranking
MS implement divergent
approaches to the statutory
insolvency ranking of bank
creditors
which
create
uncertainty for issuers and
investors alike and makes
more
difficult
the
application of the bail-in
tool
for
cross-border
institutions.
This
uncertainty can also result
in competitive distortions in
the sense that unsecured
debt holders could be
treated
differently
in
different jurisdictions and
Creation
preferred
category.
of a
senior
non-
debt
BRRD
108)
(Article
This approach would
result in two categories of
unsecured debt, both
ranking
above
subordinated debt: a
newly created category of
non-preferred unsecured
senior and a preferred
unsecured
senior
category. As opposed to
subordinated debt which
can be written-down or
175
 kom (2016) 0853 - Ingen titel
1715591_0176.png
Area
Problem definition / Origin
of the problem
the costs to comply with the
TLAC
and
MREL
requirement for banks may
be
different
from
jurisdiction to jurisdiction.
Origin of the problem:
issues
in
consistent
implementation of BRRD,
responses to the Call for
Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
converted into equity
outside resolution as well
as during resolution, the
new non-preferred senior
category would be bailed-
in only in resolution
Moratorium
The diversity of national
approaches
to
the
implementation of the tool
as well as the lack of clarity
of certain elements reduces
the effectiveness of this tool
in resolution and of
resolution tools in a cross-
border scenario.
Origin of the problem:
issues
in
consistent
implementation of BRRD
Further harmonisation of
moratorium tools.
BRRD (Articles 27,
29a-new
article,
63)
176
 kom (2016) 0853 - Ingen titel
1715591_0177.png
Area
Problem definition / Origin
of the problem
The current EU regulatory
framework
does
not
sufficiently
differentiate
between the very large
institutions and very small
institutions, particularly as
regards
reporting
and
disclosure obligations. In
addition, compliance costs
due to the complexity and
large volume of rules are
more
burdensome
for
smaller banks. Some of the
prudential requirements in
the CRR and CRD IV
impose a disproportionate
burden on smaller and less
complex institutions.
Origin of the problem:
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Solution
Level 1 text to be
amended
and
relevant articles
Targeted measures
on market risk:
CRR (Article 94)
plus new CRR
article
for
the
application of the
simplified
standardised
approach;
New article on
mandate for the
EBA to set up an
IT tool;
See
below for
counterparty credit
risk
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Proportionality
(see
also
below
counterparty
credit
risk,
supervisory
reporting and
disclosure)
Specific reporting and
disclosure framework for
smaller institutions with
reduced frequency and
content. In addition, the
EBA would be mandated
to develop an IT tool to
guide credit institutions
through the rules which
are relevant to their size
and
business
model.
Finally, it is proposed to
introduce
tailored
measures for different
metrics
(e.g.
TLAC,
lending to SMEs, trading
book, leverage ratio,
NSFR and remuneration)
that take into account the
size and business model
of credit institutions.
177
 kom (2016) 0853 - Ingen titel
1715591_0178.png
Area
Problem definition / Origin
of the problem
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Issues included in the annexes
Counterparty
credit
risk
framework
The
standardised
approaches to calculate the
exposure value of derivative
transactions
under
the
counterparty credit risk
framework suffer from
several limitations: they do
not recognise appropriately
the risk-reducing nature of
collateral in the exposures
(an issue in light of the
forthcoming international
clearing/margin
obligations);
their
calibrations are outdated
and do not reflect the high
level of volatility observed
during the recent financial
crisis; they do not recognise
appropriately
netting
benefits.
Under
the
proposed
amendment, institutions
would use the SA-CCR
recently developed by the
BCBS in the counterparty
credit risk framework
while, under revised
conditions,
institutions
with
small
trading
activities would have the
possibility to use a
revised version of OEM.
A simplified version of
SA-CCR will also be
available for banks that
would
face
some
operational difficulty to
implement SA-CCR but
have sizeable derivative
activities that would not
warrant them the use of
CRR (Articles 272-
282,
298-299,
429a)
178
 kom (2016) 0853 - Ingen titel
1715591_0179.png
Area
Problem definition / Origin
of the problem
Origin of the problem:
BCBS, responses to the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
The lack of harmonised
disclosure formats hampers
the
comparability
of
disclosures
between
institutions and over time
thereby reducing market
discipline. The existing
disclosure requirements are
mainly a "one size fits"
allowing for hardly and
differentiation based on the
size of the institution and
are therefore not optimally
proportionate.
Origin of the problem:
BCBS, responses to the
consultation on the impact
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
the revised OEM.
Disclosure
In order to alleviate the
current disproportionate
operational burden and to
be aligned with the
revised Basel Pillar 3
disclosure
framework
institutions
will
be
categorised on the basis
of their significance.
Institutions would either
be significant, small or
"other" with or without
being
"listed".
The
disclosure requirements
will be a sliding scale
with differentiations in
the
substance
and
frequency of disclosures
179
CRR (Articles 13,
431 – 455) Plus
additional
delegated
act
power
on
disclosure
requirements CRR
Article 456
Broaden
EBA
ITS
mandate to all disclosure
articles in Part Eight
 kom (2016) 0853 - Ingen titel
1715591_0180.png
Area
Problem definition / Origin
of the problem
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
whereby for all types of
institutions
disclosure
templates developed by
the
EBA
will
be
mandatory.
The
frequency
of
reporting for smaller
institutions
will
be
reduced leading to less
reporting burden without
undermining overall the
supervisory effectiveness
or financial stability risk.
Additionally the extant
body
of
reporting
requirements will be
reduced for some or all
institutions depending on
their size or other
quantitative criteria.
CRR: (Articles 99
– 101), Study on ad
hoc
reporting
requirements
(Article
519a),
EBA report on
enhanced
proportionality
(Article 519b)
CRD IV Article
104 clarification of
ad hoc reporting
powers
Supervisory
reporting
High administrative burden
caused
by
1)
disproportionate reporting
requirements generally and
for smaller banks in
particular in terms of
content
and
reporting
frequency
and
2)
supervisors
requiring
additional reporting on top
of the regular EU reporting
requirements.
Origin of the problem:
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
180
 kom (2016) 0853 - Ingen titel
1715591_0181.png
Area
Problem definition / Origin
of the problem
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Pillar
additional
capital
2
The current text of the CRD
IV sets the broad parameters
of the exercise of Pillar 2
powers, whilst leaving to
supervisory authorities a
wide margin of discretion
when
exercising
their
powers. This leads to
discrepancies
and
weaknesses in the way
Pillar 2 capital requirements
are
applied
across
jurisdictions and to the
sometimes not transparent
way supervisors' decisions
on the additional capital
imposed
on
individual
banks are made.
Origin of the problem:
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
The relevant articles of
the CRD IV and CRR will
be modified to clarify the
nature of Pillar 2 capital
add-ons, the cases in
which these should be
imposed as requirements
or
as
non-binding
expectations and their
relationship with other
capital
requirements
(buffers and Pillar 1).
CRR (Articles 28,
428) – CRD IV
(Articles 104, 104a
new, 104b new,
113, 140a new,
141)
IRRBB: CRR (art.
448), CRD IV (art.
84, 98)
New RTS on additional
own fund requirements (art.
104a)
New
RTS
on
the
standardised approach for
IRRBB (CRD IV art. 84);
New
RTS
on
the
calculation of NII for
reporting purposes (CRR
art. 448);
Update guidelines (CRD
IV art. 84) for the capture
of IRRBB
181
 kom (2016) 0853 - Ingen titel
1715591_0182.png
Area
Problem definition / Origin
of the problem
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Equity
investments
into funds
The current framework for
the credit risk attached to
exposures in the form of
units or shares in collective
investment
undertakings
(CIUs)
[basically
undertakings for collective
investment in transferable
securities (UCITS) and
alternative investment funds
(AIFs)] lacks risk sensitivity
and
transparency.
The
framework
lacks
risk
sensitivity notably in the
sense that it does not require
banks to reflect a fund's
leverage when determining
capital
requirements
associated
with
their
investment, even though
leverage is a very important
risk driver. This creates
undesirable incentives by
A new Basel standard will
be implemented. The
proposed
framework
consists
of
three
approaches, which would
apply to both SA and IRB
banks' exposures. The
look-through
approach
(LTA) requires banks to
risk weight the fund's
underlying exposures as if
they were held directly;
the
mandate-based
approach (MBA) assumes
that
the
underlying
portfolios are invested to
the maximum extent
allowed (as per the
mandate, regulations, or
other disclosures) in the
assets
attracting
the
highest risk weights; and
the fall-back approach
182
CRR Articles 128,
132, 152
 kom (2016) 0853 - Ingen titel
1715591_0183.png
Area
Problem definition / Origin
of the problem
encouraging investments in
higher-risk funds and may
result in an insufficient
capitalisation
of
such
higher-risk exposures.
Also, the framework does
not promote transparency
and
appropriate
risk
management of the relevant
exposures, as there is no
clear rank ordering between
the different approaches,
with different degrees of
prescriptiveness for SA
banks compared to IRB
banks
and
insufficient
incentives to apply the look-
through approach.
Origin of the problem:
BCBS, responses to the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
(FBA) – used for funds
with
insufficient
transparency – requires
the application of a
1,250% risk weight. It
provides a hierarchy of
approaches as a function
of the degree of due
diligence performed by
banks,
with
an
appropriate
incentive
structure, whereby the
degree of conservatism
increases
with
each
successive approach as
risk
sensitivity
and
transparency
decrease.
This promotes appropriate
risk management of bank
exposures to funds by
providing incentives to
use the more risk
sensitive and transparent
approaches.
183
 kom (2016) 0853 - Ingen titel
1715591_0184.png
Area
Problem definition / Origin
of the problem
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Bank
financing of
infrastructure
projects
The
existing
capital
requirements on exposures
for infrastructure projects
lacks risk-sensitivity and
hamper the capacity of
banks to finance high-
quality, sound infrastructure
projects
Origin of the problem:
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
A specific 'population' of
specialised
lending
exposures
will
be
identified which aim at
funding
infrastructure
projects and fulfil certain
criteria able at reducing
the different risks a bank
would incur in providing
such funding (financial,
political, legal, operating,
etc.). This new asset class
of qualifying specialised
lending exposures would
benefit from a discount
factor of 25% The criteria
will
denote
safer
infrastructure projects and
ensure that lending banks
understand the associated
risks.
Three main measures are
CRR (new Article)
The current general limit to
large exposures of 25% of
CRR
(Articles
4(1)(71) and (91),
184
 kom (2016) 0853 - Ingen titel
1715591_0185.png
Area
Problem definition / Origin
of the problem
institutions' eligible capital
(which is the sum of Tier 1
capital and an amount of
Tier 2 capital equal to one
third of Tier 1 capital ) is
not sufficiently prudent,
especially for larger banks,
since it only capture a small
part of the overall large
exposures that European
institutions have. Moreover,
it results in a higher limit
for smaller banks since
larger banks have usually
more Tier 2 capital than
smaller ones. Moreover, the
current limit doesn’t take
into account the higher risks
carried by the exposures
that globally systemically
important Banks (G-SIBs)
have to single counterparty
or groups of connected
clients and, in particular, as
regards exposures to other
Solution
Level 1 text to be
amended
and
relevant articles
390, 391, 394, 395,
399, 400, 401, 403)
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Large
exposure
framework
proposed:
- reduced capital base
(only
Tier
1)
for
calculating the large
exposures limit;
- lower large exposures
limit for exposures of G-
SIIs v. G-SIIs (15% of
Tier 1 capital);
- use of the new
developed
SA-CRR
method for the calculation
of the exposure value of
exposures
towards
derivatives.
185
 kom (2016) 0853 - Ingen titel
1715591_0186.png
Area
Problem definition / Origin
of the problem
G-SIBs.
Origin of the problem:
BCBS, responses to the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Article 400 (2) of the CRR
lists a number of exposures
that competent authorities
may fully or partially
exempt from the scope of
application of the large
exposures limit. These
exposures can only be
exempted if the conditions
laid down in paragraph 3 of
the same article are met. By
way of derogation the CRR
provides for a temporary
possibility for Member
States to grant an exemption
from the large exposures
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Exemptions on
large
exposures
To end the transitional
period allowing Member
States to grant exemptions
for certain exposures to
the large exposure limit
set out in Article 493(3)
of CRR.
CRR (Articles 493,
507)
186
 kom (2016) 0853 - Ingen titel
1715591_0187.png
Area
Problem definition / Origin
of the problem
limit for the same exposures
listed in Article 400 (2) of
CRR, however without
having
to
meet
the
conditions set out in
paragraph 3 of Article 400
of CRR.The concurrent
possibility of Members
States
and
competent
authorities
of
granting
exemptions to the same
exposures has proved to be
problematic
after
the
introduction of the Single
Supervisory
Mechanism
(SSM) and can interfere
with the ability of the SSM
to perform its tasks in a
consistent and coherent
manner.
Origin of the problem:
BCBS, responses to the
consultation on the impact
of CRR and CRD IV on
bank financing of the
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
187
 kom (2016) 0853 - Ingen titel
1715591_0188.png
Area
Problem definition / Origin
of the problem
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Rules
exposures
CCPs
on
to
The
Mark-to-Market
Method does capture risks
sufficiently well, i.e. either
leading to too low or too
high requirements. The
current framework does not
take a sufficiently holistic
view of how the different
types of exposures to a
Qualifying CCP interrelate
and are therefore not
sufficiently sensitive to the
aggregate risk of those
exposures and how that risk
is distributed.
Origin of the problem:
BCBS, responses to the Call
for Evidence
The revised standards
adopted by the Basel
Committee
will
be
implemented.
Notable
revisions to the Basel
standards include the use
of a single method for
determining the capital
requirements
for
exposures to QCCPs
stemming from default
fund contributions, an
explicit floor for those
requirements, as well as
an explicit cap on the
overall
capital
requirements applied to
exposures to QCCPs (i.e.
those charges will not
exceed the charges that
would
otherwise
be
applicable if the CCP
were a non-qualifying
188
CRR Articles 300
to 311 and 497,
EMIR Articles 50a
to 50d
Maintaining empowerment
for delegated act under
Article 456(1)(h), update of
RTS mandated by Article
304(5) of CRR and update
of ITS mandated by Article
50c(3) of EMIR
 kom (2016) 0853 - Ingen titel
1715591_0189.png
Area
Problem definition / Origin
of the problem
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
CCP).
Contractual
recognition of
bail-in (article
55 BRRD)
Compliance with Article 55
BRRD raises two types of
difficulties. First, certain
third country counterparties
refuse
to
include
a
contractual
clause
recognising a Union bail-in
power in financial contracts
concluded
with
Union
banks. These third country
entities often have a high
degree of negotiating power
against Union banks, or
apply internationally agreed
standard contractual terms
in their banking contracts,
e.g.
with
respect
to
liabilities to non-Union
financial
market
infrastructures or trade
finance liabilities (letters of
credit, bank guarantees and
performance
bonds).
Secondly, even when third
Article 55 BRRD will be
amended in order to
enable the resolution
authority to exclude the
obligation by means of a
waiver if it determines
that this would not
impede the resolvability
of the bank, or that it is
legally, contractually or
economically
impracticable for banks to
include
the
bail-in
recognition clause for
certain liabilities. In these
cases, those liabilities
should not count as
MREL and should rank
senior to MREL to
minimize the risk of
breaking the No-Creditor-
Worse-Off
(NCWO)
principle. In this regard,
the proposal will not to
189
BRRD (Article 55)
 kom (2016) 0853 - Ingen titel
1715591_0190.png
Area
Problem definition / Origin
of the problem
country counterparties are
prepared to accept bail-in
related clauses in their
contracts with Union banks,
in some cases the local
supervisor may forbid this.
The incorporation of TLAC
in the EU legislative
framework
should
not
materially affect the burden
of non G-SIBs to comply
with the current MREL
framework. Fundamentally,
TLAC and MREL aim to
achieve the same policy
objective of ensuring that
banks hold a sufficient
amount of bail in-able
liabilities that allow for
smooth
and
quick
absorption of losses and
bank recapitalisation. Some
technical differences exist
however between the 2
frameworks regimes in
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
weaken the bail-in.
Changes
MREL
to
MREL will be amended
to
address
some
shortcomings, notably to
(1) create 1 set of
eligibility criteria for
MREL/TLAC
eligible
instruments (except for
subordination), (2) clarify
the internal loss absorbing
capacities within banking
groups, independent of
the chosen resolution
strategy
through
introduction
of
the
concepts of resolution
groups, resolution entities
and material subgroups,
and (3) the alignment of
the basis for calculation
190
BRRD Articles 2,
12, 13, 16, 18, 45,
59, 60, 89); SRMR
Article 12
Existing RTS on MREL to
be aligned with the new
level 1 provisions
 kom (2016) 0853 - Ingen titel
1715591_0191.png
Area
Problem definition / Origin
of the problem
terms of eligibility criteria
(excl. subordination) and in
terms of basis of calculation
of
the
requirement.
Additionally, MREL is not
specific as to how bail-in
capacity should be allocated
within groups depending on
the
choses
resolution
strategy.
Origin of the problem:
Financial Stability Board,
responses
to
the
consultation on the impact
of CRR and CRD IV on
bank financing of the
economy, responses to the
Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
on the RWA and the
leverage ratio exposure
measure.
191
 kom (2016) 0853 - Ingen titel
1715591_0192.png
Area
Problem definition / Origin
of the problem
The move from IAS 39 to
IFRS 9 will affect CRR
capital requirements. The
impact will depend on:
1. the amount of the
increase in provisions due to
the change in accounting;
2. the type of regulatory
approach the bank follows
to calculate its capital
requirements;
3. for IRB banks, their
present level of provisions
compared to the regulatory
expected loss.
There is uncertainty about
the impact of the difference
in levels of provisioning
between current IAS 39 and
IFRS 9 on CET1 capital of
EU Banks
Solution
Level 1 text to be
amended
and
relevant articles
CRR Article 473a new
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
Application of
IFRS 9 by the
EU banks
To introduce in CRR a
transitional regime so that IFRS 9
changes will be phased-in
progressively over a few years
192
 kom (2016) 0853 - Ingen titel
1715591_0193.png
Area
Problem definition / Origin
of the problem
Origin of the problem: responses to
the consultation on the impact of
CRR and CRD IV on bank
financing of the economy, responses
to the Call for Evidence
Solution
Level 1 text to be
amended
and
relevant articles
Level
2
measures
(Delegated, implementing
acts, RTS, ITS) envisaged
to be created or amended
193