BANKING SUPERVISION

AND RESOLUTION IN THE EU

EFFECTS ON SMALL HOST COUNTRIES
IN CENTRAL, EASTERN AND
SOUTH EASTERN EUROPE

Working Paper | April 2019

Ismael Ahmad Fontán · Senior Financial Sector Specialist · World Bank | FinSAC
Thorsten Beck · Professor · Cass Business School
Katia D’Hulster · Lead Financial Sector Specialist · World Bank
Pamela Lintner · Senior Financial Sector Specialist · World Bank | FinSAC
D. Filiz Unsal · Economist ·The International Monetary Fund

Disclaimer
The findings, interpretations and conclusions expressed in this document are entirely those of the au-
thors. They do not represent the views of the World Bank and its affiliated organizations or those of the
affiliated organizations of the authors. Although every effort was made to ensure that the information in
this document was correct at the time of publication, the authors do hereby disclaim any liability to any
party for any loss, damage, or disruption caused by errors or omissions, whether such errors or omissions
result from negligence, accident, or any other cause.

Acknowledgements
The document was prepared by Ismael Ahmad, Thorsten Beck, Katia D’Hulster, Pamela Lintner, and Filiz
D. Unsal. Special thanks go to the ECB, SRB, EC, EBA and IMF staff as well as to Andrew Gracie, Nicolas
Veron, Ezio Caruso and Maria Soledad Martinez Peria for valuable informal comments on an earlier ver-
sion of this paper. Finally, we would like to thank Milica Nikolic for excellent analytical work and overall
support. Any errors are exclusively the authors’ (see disclaimer above). Miquel Dijkman, FinSAC Coordi-
nator and Mario Guadamillas, Practice Manager provided oversight.
Background

A common feature of many FinSAC client countries’ banking system is the significant presence of sub-
sidiaries of multinational banks, generally with parent banks from the EU. Most of those subsidiaries
are of systemic importance in the host country, but in many cases these foreign operations are not
material for the parent bank and thus for the home country supervisory and resolution authority. The-
se “small hosts” face a unique set of policy and implementation challenges, which do not feature pro-
minently on the international and EU agenda. Moreover, EU directives and regulations are not designed
with these smaller host countries’ banking systems in mind, especially with non-EU countries where
the hosts rights don’t apply, and their implementation is often challenging. To address these challenges,
FinSAC provides bilateral independent and tailored expertise, technical assistance and advice.

For example, FinSAC has assisted bank supervisors and resolution authorities in the CESEE with the
design of new supervisory and resolution approaches based on the EU legal and regulatory frameworks.
In the bank supervision area, FinSAC’s technical assistance has included business model assessment,
internal governance and risk management, credit risk, operational risk, market risk, interest rate risk in
the banking book, capital, and liquidity, as well as the operating model of the new supervisory approach.
Other bank supervision projects included the assessment of the prudential implications of IFRS 9 and
the implementation of the Supervisory Review and Evaluation Process (SREP).

For bank resolution, FinSAC has worked closely with several authorities to help them create and ope-
rationalize a resolution framework and advance the assessment of banks’ recovery plans. FinSAC has
assisted with drafting by-laws on valuation, recovery, and resolution planning – including the adaptati-
on of EBA bank information templates to the local framework – and helped clients prepare for the as-
sessment of the first recovery plans. FinSAC has also offered technical advice on internal information
and cooperation rules and provided example resolution fund calculations. A resolution plan manual has
also been developed by FinSAC, including analysis of good practice examples for resolution authorities’
internal decision making, recovery and resolution planning, and the taking of resolution decisions.

Many of our client countries are EU candidates, and hence are in the process of introducing and im-
plementing complex EU regulations. Notwithstanding this, candidate countries are still considered and
treated as third countries which means they are not part of the EU rule making process. Hence, our
clients often lack insight in the practical application of the EU legal and regulatory framework and
face difficulty in finding counterparts to share their experience candidly. FinSAC is well-aware of the
necessity for our client countries to fully understand the EU legal and regulatory framework.

In this publication, we have carefully analyzed the EU legal and regulatory framework for bank super-
vision, recovery and resolution from the perspective our client-countries. Our analysis distinguishes
between small hosts (i) inside the Eurozone, (ii) inside the EU but outside the Eurozone, and (iii) outside
the EU. The latter category consists of non- EU -candidate countries and third countries. The implica-
tions for cross-border supervision and resolution cooperation are distinct in each of these cases. We
hope this will be the basis for further discussions between authorities within and outside the region.

2
1. Introduction 6

2. Trends in cross border banking in the region 7

3. Supervision and resolution in cross-border banking

problem analysis 12
3.1. Cross-border externalities associated with a bank failure 12
3.2. Diverging interests and powers across home and host countries 13

4. Regulatory reforms in the EU and the Eurozone 15

4.1. Supervisory processes and procedures in the EU 16
4.2. Bank Resolution in the EU 18
4.3. Cross border expansion: legal structures 23

5. Cross-border cooperation in supervision

and resolution - practice 27
5.1. Eurozone small hosts 28
5.1.1. Overview: Banking supervision in the Eurozone 28
5.1.2. Overview: Bank resolution in the Eurozone 29
5.1.3. Implications of the new framework for “eurozone small hosts” 30
5.1.4. Challenges and potential reactions by “eurozone small hosts” 31
5.2. Non-Eurozone EU small hosts 31
5.2.1. Overview: Banking supervision in the EU 32
5.2.2. Overview: Bank resolution in the EU 32
5.2.3. Implications of the new framework for non-Eurozone EU hosts 34
5.2.4. Potential reactions by non-Eurozone EU small hosts 38
5.3. Non-EU host countries (EU candidates and non EU candidates) 39
5.3.1. Banking regulation and supervision in candidate countries 42
5.3.2. Potential issues faced by host authorities in candidate countries 43
5.3.3. Bank regulation, supervision, and resolution in non-candidate countries 47

6. Conclusions 48

Annex 50
References 52
Authors 54
Ismael Ahmad Fontán, Senior Financial Sector Specialist, The World Bank/FinSAC 54
Thorsten Beck, Professor Cass Business School 54
Katia D’Hulster, Lead Financial Sector Specialist, The World Bank 55
Pamela Lintner, Senior Financial Sector Specialist, The World Bank/FinSAC 55
D. Filiz Unsal, Economist, The International Monetary Fund 55

3
Figures
Figure 1: The number and the average share of foreign banks in ECA over time 7
Figure 2: Share of domestic and foreign-owned banks in the region over time 8
Figure 3: Foreign bank penetration across different country groups 9
Figure 4: Bank stability in the region over time 9
Figure 5: Resolution and resolvability planning 19

Tables
Table 1: Country classification 8
Table 2: Market share - As a percentage of bank‘s assets in total banking sector assets 10
Table 3: Market share in host country and share of assets in group‘s total assets in 2017 11
Table 4: EU Prudential Rulebook 15
Table 5: Supervisory processes and procedures 16
Table 6: Joint Supervisory Decisions within EU Colleges 17
Table 7: Minimum requirement for own funds and eligible liabilities and resolution strategies 22
Table 8: Joint decisions in resolution planning 22
Table 9: Regulatory classification of EU branches 25
Table 10: EU corporate structures for cross-border banking 26
Table 11: EBA professional secrecy and confidentiality regime supervisory authority
applicable to each third-country 41

Boxes
Box 1: Greek banks: parent companies in crisis 11
Box 2: Total loss absorbing capacity and minimum requirement for own funds 19
Box 3: Nordea 24
Box 4: Survey of host country supervisors’ views on cooperation with home country supervisors 34
Box 5: Why would EU small hosts join the European banking union? 35
Box 6: How mergers and acquisitions in the Eurozone may affect small hosts 37
in the EU and other non-EU European countries 37
Box 7: Benefits for non-EU hosts from participation in resolution and supervisory colleges 40
Box 8: Regulatory convergence in EU-candidate countries in bank supervision and resolution 42
Box 9: Can parent banks let subsidiaries fail;
how the new resolution framework can shape the debate 45
Box 10: Sovereign bond holdings in non-EU countries 47
Box 11: Steps for effective cross-border supervision in Central America 49

4
Acronyms

BRRD BANK RECOVERY AND RESOLUTION DIRECTIVE NPL NON-PERFORMING LOANS

CAPTAC-DR IMF REGIONAL TECHNICAL ASSISTANCE CENTER NRA NATIONAL RESOLUTION AUTHORITY

FOR CENTRAL AMERICA, PANAMA AND THE OeNB OESTERREICHISCHE NATIONALBANK

DOMINICAN REPUBLIC PONV THE POINT OF NON-VIABILITY

CCSBSO CENTRAL AMERICAN COUNCIL OF RCA RECAPITALIZATION AMOUNT

SUPERINTENDENTS OF BANKS, INSURANCE AND RWA RISK-WEIGHTED ASSETS

OTHER FINANCIAL INSTITUTIONS SEP SUPERVISOR EXAMINATION PROCESS

CET COMMON EQUITY TIER SI SIGNIFICANT INSTITUTION

CESEE CENTRAL AND EASTERN EUROPE SPE SINGLE POINT OF ENTRY

CMGS CRISIS MANAGEMENT GROUPS SREP SUPERVISORY REVIEW AND EVALUATION PROCESS

CRD CAPITAL REQUIREMENTS DIRECTIVES SRB SINGLE RESOLUTION BOARD

DGS DEPOSIT GUARANTEE SCHEME SRF SINGLE RESOLUTION FUND

DR DELEGATED REGULATION SRM SINGLE RESOLUTION MECHANISM

EBA EUROPEAN BANKING AUTHORITY SRMR SINGLE RESOLUTION MECHANISM REGULATION

EBRD EUROPEAN BANK FOR RECONSTRUCTION SSM SINGLE SUPERVISORY MECHANISM

AND DEVELOPMENT TLAC TOTAL LOSS ABSORBING CAPACITY

EC EUROPEAN COMMISSION US UNITED STATES

ECA WORLD BANK EUROPE AND CENTRAL ASIA REGION WDCC WRITE-DOWN OR CONVERSION OF CAPITAL

ECB EUROPEAN CENTRAL BANK

EDIS EUROPEAN DEPOSIT INSURANCE SCHEME

ELA EMERGENCY LIQUIDITY ASSISTANCE

EU EUROPEAN UNION

FOLTF FAILING OR LIKELY TO FAIL

FINSAC FINANCIAL SECTOR ADVISORY CENTER

FSB FINANCIAL STABILITY BOARD

GFSA GROUP FINANCIAL SUPPORT AGREEMENT

GLRA GROUP LEVEL RESOLUTION AUTHORITY

G-SIBS GLOBAL SYSTEMICALLY IMPORTANT BANKS

HQLA HIGH QUALITY LIQUIDITY ASSETS

HVB HYPOVEREINSBANK

ICAAP INTERNAL CAPITAL ADEQUACY ASSESSMENT

ILAAP INTERNAL LIQUIDITY ADEQUACY ASSESSMENT

IMF INTERNATIONAL MONETARY FUND

IRT INTERNAL RESOLUTION TEAM

JST JOINT SUPERVISORY TEAM

LAA LOSS ABSORPTION AMOUNT

LCR LIQUIDITY COVERAGE RATIO

M&A MERGERS AND ACQUISITIONS

MLE MATERIAL LEGAL ENTITY

MoU MEMORANDUM OF UNDERSTANDING

MPE / MPOE MULTIPLE POINT OF ENTRY

MREL MINIMUM REQUIREMENT FOR OWN FUNDS

AND ELIGIBLE LIABILITIES

5
1. Introduction

The objective of this paper is to explain and analyze the implications of recent supervisory and resolu-
tion reforms in the European Union (EU) for small host countries in Central, Eastern, and South Eastern
Europe (CESEE) whose banking systems are dominated by Western European banks. By “small host
countries” we understand those in which subsidiaries of multinational parent banks are of systemic
importance, while the foreign operation is not material for the parent bank and thus for the home
country authority. These “small hosts” face a unique set of challenges, that do not feature much on the
international agenda.

Small hosts are very common in the CESEE region where the banking systems are dominated by
subsidiaries of multinational banks, many with parent banks from the EU. The dominance of foreign
banks in the region has often been the result of banking crises and/or privatization waves and has, in
most cases, helped form market-based local banking systems and in turn contributed to these econo-
mies’ overall development and rapid growth. However, foreign banks were also part of a credit boom
in the early to mid-2000s, often accompanied by housing price booms, increasing reliance on parent
and wholesale funding, and worsening net foreign asset positions. More recently, the post-crisis re-
trenchment of European parent banks has been blamed for falling lending and rising financing costs in
the region and led to the Vienna Initiative by the European Bank for Reconstruction and Development
(EBRD), World Bank, International Monetary Fund (IMF) and European Investment Bank to prevent a
large-scale and uncoordinated withdrawal of cross-border bank groups from the region.

The global financial crisis has led to significant reforms in cross-border supervisory cooperation. The
failure of big international financial institutions (e.g., Lehman Brothers) and cross-border banks (e.g.,
Fortis, Dexia, and the Icelandic banks), and their rather chaotic resolution played a prominent role in
the unfolding of the crisis. The resolution of these institutions had to be undertaken on a national level,
given the lack of tools to coordinate resolution across borders. Consequently, recognition grew that the
main tools of supervisory cooperation available during the financial crisis, such as supervisory colleges,
were not sufficient to deal with (distressed) large and systemically important cross-border financial
institutions. Since then, there have been multiple global initiatives by international standard setters
such as the Financial Stability Board and the Basel Committee to strengthen cross-border cooperation
during normal and crisis times. Still, these initiatives continue to rely on voluntary cooperation among
home and host supervisors.

This is different from the EU where binding cross border cooperation and consensus based joint de-
cision making among supervisors and resolution authorities exist. Within the Eurozone, the most im-
portant initiative has undoubtedly been the (incomplete)1 construction of the European banking union.
The Single Supervisory Mechanism (SSM) has been complemented by a largely centralized Single Re-
solution Mechanism (SRM), even though it is still reliant on national execution and provides European
Commission and European Council veto rights.

Our analysis distinguishes between small hosts (i) inside the Eurozone, (ii) inside the EU but outside
the Eurozone, and (iii) outside the EU. The latter category consists of EU-candidate countries and third
countries. The implications for cross-border supervision and resolution cooperation are distinct in each
of these cases, with increasing asymmetries between home and host countries.

The next section gives a short overview of trends in cross-border ownership across the World Bank
Europe and Central Asia (ECA) region with particular focus on CESEE countries. Section 3 discusses the
regulatory and supervisory challenges arising from cross-border banking. Section 4 describes recent
changes in bank regulation and supervision within the EU. Section 5 discusses cross-border cooperati-
on on supervision and resolution from the perspective of small hosts. The final section concludes with
recommendations.
1
The funding mechanisms, including a European Deposit Insurance Scheme (EDIS) as proposed by the European Commission
and a backstop mechanism, await political approval and implementation.

6
2. Trends in cross border banking in the region

The late 1990s and early 2000s saw a surge in foreign banks and their share in local banking systems
in the transition economies of selected ECA countries (Figure 1 and Figure 2).2 During this period, the
region transformed from having the lowest share of foreign banks to having the highest share. Some
smaller countries like Croatia have seen their foreign bank share reach 100%. In general, it is the smal-
ler countries of the region that have higher shares of foreign banks. While a few regional banks have
expanded across other countries of the region, most foreign banks are from Western Europe.

The transformation of the banking system, from a state-owned mono-bank system towards a priva-
tely-owned market-based financial system was key to achieving macroeconomic stability in the late
1990s. Successful transformation – mostly into a foreign-bank dominated system – served as disci-
plining tool to break the links between banks and incumbent, formerly state-owned enterprises and
thus the cycle of non-performing loans, bank recapitalization and inflation. At the same time the ban-
king system supported sectors of the economy that were hitherto de facto un(der)served (e.g. SMEs,
households), thereby promoting economic convergence. The entry of foreign banks also helped to raise
the bar with respect to risk management practices as well as bank and corporate governance and the
break with the past. Indeed, the countries that finalized the ownership transformation process the
fastest were also the first ones to emerge from the systemic banking crises of the 1990s.

Figure 1: The number and the average share of foreign banks in ECA over time3

While in EU member states foreign bank penetration has happened using both branches and subsidiari-
es, non-EU countries have mainly licensed foreign banks as subsidiaries. However, there have also been
significant differences across countries in the region.

2
We define “foreign bank” as a banking institution with at least 50% ownership stake held by a non-domestic parent bank. Most foreign banks in the
region are fully-owned by their parent banks. Foreign banks can enter a country through establishing subsidiaries or branches. Subsidiaries are incorpo-
rated in the host country and regulated and supervised by the host country supervisors. If the parent bank exercises control over the subsidiaries, their
balance sheet and income statements are consolidated into the parent bank’s financial statements. A branch, on the other hand, is part of the parent
bank, operationally, legally and financially. A branch is typically regulated and supervised by the home country supervisor and is subject to home country
deposit insurance (though the host country can top-up the deposit insurance).
3
Figure 1 includes all countries from Table 1 except Kosovo due to the lack of comparable data.

7
Figure 2: Share of domestic and foreign-owned banks in the region over time4

In the subsequent analysis, we differentiate between EU and non-EU countries. Within the group of EU
countries, only some have adopted the euro as currency. Within the non-Euro countries, some (most
prominently Bulgaria) have a currency board arrangement that effectively ties the local currency to
the euro.5 Further, among the non-EU member countries, we differentiate between EU candidate and
non-candidate countries. Table 1 provides a list of countries that we consider in this policy paper and
the respective classification. In addition to the above classification, we also denote FinSAC client
countries.

Table 1: Country classification

Non-EU members
Eurozone Non-euro EU members Non-candidate FinSAC clients
Candidate countries
countries
ALBANIA
ARMENIA
AZERBAIJAN
BELARUS
BOSNIA AND HERZEGOVINA
BULGARIA
CROATIA
CZECH REPUBLIC
ESTONIA
GEORGIA
HUNGARY
KAZAKHSTAN
KOSOVO
KYRGYZSTAN
LATVIA
LITHUANIA
NORTH MACEDONIA
MOLDOVA
MONTENEGRO
POLAND
ROMANIA
RUSSIA
SERBIA
SLOVAKIA
TURKEY
UKRAINE
UZBEKISTAN

4
Figure 2 includes all countries from Table 1 except Kosovo due to the lack of comparable data.
5
Two non-EU countries (Montenegro and Kosovo) use the Euro as currency despite not being Eurozone countries, which imposes specific challenges for
bank supervision. Both countries euroized unilaterally, i.e. bypassing the usual convergence criteria and against the recommendations of the European
institutions.

8
Figure 3 shows the highest foreign bank share for ECA countries are in EU candidate countries (all are
also FinSAC client countries), closely followed by Eurozone members of this region, while the third group
with very significant shares of foreign banks are EU member countries that are outside the Eurozone.

Figure 3: Foreign bank penetration across different country groups

In the run-up to the crisis, following their parent’s strategy, foreign subsidiaries relied increasingly on
funding from their parent bank and wholesale markets and reached loan-to-deposit ratios of almost
130%. In the fall of 2008, there was increasing concern, both among policy makers and banks themsel-
ves, that there would be a stampede towards the exit in the CESEE region, which might have led to a
meltdown of the financial systems of these countries and balance of payment crises. More specifically,
there were discussions that some of the large European institutions had to shed subsidiaries in CESEE
as a condition for receiving state aid. Starting in late 2008, the IMF, EBRD, EU Commission and others
therefore initiated a series of meetings in Vienna with banks, home and host supervisors to prevent this
from happening. Many banks made specific roll-over and recapitalization commitments.

Since the crisis loan-deposit ratios have decreased substantially, approaching 100% (Figure 4). At the
same time, the liquidity ratio declined until 2008 and has stayed constant at an average of around
25% since then. The non-performing loans (NPL) ratio increased rapidly until 2010 (reaching 10%).
It has been on a downward path since then, although it remains high in many countries (right-hand
side panel), while the NPL coverage ratio has been stable at an average of around 60%. However, the
crisis-induced jump in NPLs and the reduction in funding by parent banks have increased the risk of
many banking systems in the region getting stuck in a bad equilibrium of lending reduction and poor
real sector performance.

Figure 4: Bank stability in the region over time

Source: Global Financial Development Database

140,0% 129,88% 45,00% 12,0% 90,00%

41,33% 77,77%
10,09%
40,00% 80,00%
120,0% 10,0%
97,53% 8,28% 70,00%
35,00%
100,0%
30,00% 8,0% 60,00%
97,05% 62,92% 50,00%
80,0% 25,00%
28,60% 6,0%
20,00% 4,32% 40,00%
60,0%
15,00% 4,0% 30,00%
40,0% 3,09%
10,00% 20,00%
2,0%
20,0% 10,00%
5,00%
0,0% 0,00% 0,0% 0,00%
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Loan-Deposit Ratio Liquidity Ratio (left axis) NPL ratio Prov-NPL Ratio (right axis)

9
Rather than gauging the importance of foreign banks from a country perspective, another way to look
at the small host issues is to analyze the simultaneous presence of mainly Western banking groups
across many countries throughout the region. In Table 2, we show the share of several international
banks across countries in the region (as of 2017).

Table 2: Market share - As a percentage of bank‘s assets in total banking sector assets

Following the global financial crisis, the presence of foreign banks in the region has changed, in some
cases significantly. The retrenchment of Greek banks and some Austrian banking groups6 from several
Balkan countries mainly responded to liquidity and solvency issues in their home country. In addition,
many banks such as Commerzbank (2012) and Unicredit (2016) sold their operations in crisis-hit coun-
tries like Ukraine. In other cases, issues related to the banks’ business models and solvency prompted
strategic revisions of the banking group, resulting in the divestment of some of their foreign opera-
tions.7

The expansion of non-Western banking groups in CESEE has run in parallel to the Western European
banks’ retrenchment.8 Nevertheless, Western European banks continue to dominate in most countries
of the region. For example, Unicredit has very high market shares in Croatia, Bosnia, and Bulgaria, while
Raiffeisen Bank is also very relevant in Albania, Bulgaria, Kosovo, and Slovak Republic, and Intesa San
Paolo in Croatia, Albania, Serbia, and Slovakia.

Small hosts face the asymmetry between the systemic importance of the subsidiary in the host coun-
try and the irrelevance of the host country operation for the overall operation of the parent bank and
thus the home supervisor. Table 3 shows for five of the Western European banks present in the region
their market share in the different host countries and the share of the respective subsidiary in the
group’s total assets. The smaller the host country, the more pronounced the asymmetry becomes.

6
Most of the cross-border bank subsidiaries of the resolved Austrian bank Hypo Alpe Adria (Slovenia, Bosnia-Herzegovina, Serbia, Croatia, and Mon-
tenegro) were acquired by Advent (80%) and EBRD (20%) in December 2014 and renamed as Addiko. Another relevant case was the sale of Volksbank
International AG (VBI) cross-border subsidiaries to Sberbank in September 2011.
7
See the case of Raiffeisen Bank in 2015-2016 and its decision to divest its operations in Slovenia and Poland, or the case of Société Générale with the
selling of its Croatian, Albanian, Serbian, and Bulgarian operations.
8
See for example, the cases of OTP, Sberbank or, to a lesser extent, certain Turkish banks

10
Table 3: Market share in host country and share of assets in group‘s total assets in 2017

Box 1: Greek banks: parent companies in crisis

Greek banks: parent companies in crisis

Following the introduction of the Euro, many Greek banks expanded across countries using their high profitability
and access to wholesale funding. As Greek debt sustainability concerns started to rise, host supervisors in the
region grappled with two main problems.

First, a weak parent company could transmit its problems to its local subsidiaries across the region. This contagion
may be triggered by intra-group exposures, especially those that involve the upstream of host country deposits to
the parent entity. The reaction by many host authorities across the region involved ring-fencing these subsidiaries,
for example by imposing higher capital requirements, restricting upstream intra-group exposures, and requiring
specific profit retention policies. These measures ensured that the local subsidiaries remained relatively isolated
from the liquidity and solvency crisis of their parents.

Second, the Greek banks were forced to divest non-core assets to free up capital to absorb losses in their Greek
home market. In particular:

• National Bank of Greece sold its operations in Turkey (Finansbank) in 2016, Bulgaria
(UBB) and Serbia (Vojvodjanska Banka) in 2017, and in Albania (Banka NBG Albania) in
2018. NBG still retains its subsidiaries in North Macedonia and Romania.

• Piraeus Bank divested in 2017 its operations in Romania (Piraeus Bank Romania) and
Serbia (Piraeus Bank Beograd). It still owns subsidiaries in Albania (Tirana Bank),
Bulgaria (Piraeus Bank Bulgaria) and Ukraine (JSC Piraeus Bank ICB).

• Eurobank Ergasias sold its Ukrainian operation (Public JSC Universal Bank) in 2016 and
its Romanian subsidiary (BancPost Romania) in 2017, while keeping a relevant position in
Serbia and Bulgaria.

• Alpha Bank disposed of its Macedonian (Alpha Bank Skopje) and Bulgarian
(Alpha Bank Bulgaria) operations in 2016, and of its Serbian subsidiary (Alpha Bank
Srbija) in 2017. Alpha Bank still owns subsidiaries in Romania and Albania.

11
3. Supervision and resolution in cross-border banking
problem analysis

This section explains the need for cooperation and coordination between home and host supervisors,
especially in (the preparation of) bank resolution, but also asymmetries in information and interests,
which diverge even further during crisis situations. Discussing these issues will provide the analytical
fundament for the following discussion on regulatory reforms within the European Union and its impli-
cations for relationships between home and small host supervisors in CESEE.

3.1. Cross-border externalities associated with a bank failure

One of the main rationales for developing financial safety nets—consisting of regulation, supervision,
lender of last resort facilities, deposit insurance, and bank resolution frameworks—is that bank failures
and systemic crises can create sizeable negative externalities, which can make them very costly for the
real economy.9 Traditionally, financial safety nets are national. In a world with mostly domestic banking
systems and limited cross-border bank flows—so that the externalities of bank failure are limited to
domestic agents—there is little, if any, need for cross-border cooperation. However, in a world where
banks are involved in cross-border activities, distortions arise from a national supervisory process, as
we spell out in the following.

Ownership linkages: First, cross-border externalities arise from cross-border activities of specific fi-
nancial institutions. For example, a bank failure may impose costs on borrowers abroad by leading to
lower credit availability to foreign firms or may cause foreign depositors to lose access to their savings.
These externalities are not always considered by domestic supervisors who are focused on domestic
stakeholders and domestic financial stability as they are accountable to domestic governments and
taxpayers. To avoid these distortions, the geographic perimeter of the responsible supervisor should
match the geographic footprint of the bank. However, an alignment of supervisory mandates and
banks’ footprints is hard to implement in practical terms, given the size of banks’ geographic footprints
and their variability over time as well as legal, political, and economic limits set by national boundaries
(see also section 3.2.).10

Market linkages: Second, cross-border externalities can arise even if there is no direct cross-border
bank presence in a country through financial market integration. Specifically, direct interbank expo-
sures but also common-asset exposures can result in negative cross-border externalities from a bank
failure. Such cross-border spill-overs can be due to fire-sales of fragile banks and common asset ex-
posures, but also informational contagion among investors. For example, further sovereign problems in
the Eurozone may trigger a reassessment of all the sovereign risks across closely related economies,
decreasing market values and widening sovereign bonds spreads, which may in turn affect banks in the
region, even if they do not have direct exposure to the Eurozone. Even exposure by banks to the same
asset markets as the failing bank in another country may be sufficient for this type of externality to
occur. The more financially integrated financial systems are, the higher is this exposure.

Regulatory arbitrage: Third, cross-border externalities can arise from regulatory arbitrage. Banks have
incentives to move to jurisdictions with lighter regulation or weaker enforcement. This can result in

9
The concept of externalities refers to a situation where the costs or benefits of an activity are incurred by an unrelated third party. In unregulated mar-
kets, goods with positive externalities tend to be in short supply. Conversely, goods with negative externalities tend to be oversupplied since the full costs
of the activity (e.g. pollution, systemic risk) are not borne by the suppliers alone but by society at large.
10
Beck, Todorov and Wagner (2013) show that banks’ cross-border activities distort supervisory incentives as evidenced by actual intervention decisions
during the recent financial crisis. Specifically, cross-border banks with a high share of foreign deposits and assets were intervened at a later, more fragile
state by their home country supervisors, while cross-border banks with a high share of foreign equity were intervened earlier at a less fragile state. These
findings are consistent with the costs of bank failure being borne by foreign depositors and borrowers thereby providing the incentive for home country
supervisors to delay intervention by exercising forbearance.

12
negative externalities for the host countries, if lighter regulation or weaker enforcement leads to bank
fragility or [Link] circumvention of supervisory oversight due to regulatory arbitrage (e.g., on
licensing requirements, reporting standards and observance of prudential regulations) can have a per-
vasive impact on the solidity of the banking sector and is a major concern particularly in less developed
and smaller economies where supervisory capacity is limited.11

Currency unions: Finally, specific cross-border externalities arise within monetary unions since a coun-
try cannot simply devalue its currency to regain competitiveness following a shock and hence may
need to tap the resources of other countries in some form or other. The costs from asymmetric shocks
that affect members of a currency union to a different extent are thus much higher in monetary
unions.12 Further, relying on a common lender of last resort might result in a tragedy of the commons
problem, as it is in the interest of every member government with fragile banks to “share the burden”
with the other members through, e.g., drawing on liquidity support by the joint lender of last resort. It
is important to note that this externality applies at the systemic level, rather than just for individual
institutions. The costs arising from this burden sharing, or rather burden shifting, in monetary unions
increases in line with the size of the banking systems and the interlinkages across borders within the
union. This externality might be especially relevant for smaller countries in currency unions and for
countries with oversized (relative to their overall economy) banking systems. An example are countries
that are completely euroized, such as Montenegro or Kosovo, as they effectively cannot exercise their
lender of last resort responsibility.

In summary, cross-border externalities of bank failure distort the decision process of national regula-
tors and supervisors and not only in the context of bank resolution decisions, but possibly also during
regular supervision and regulation. These distortions are exacerbated by the asymmetry in interests
and powers between different supervisors as we will discuss next.

3.2. Diverging interests and powers across home and host countries

Home country supervisors of multinational banks supervise on a consolidated basis, which gives them
a more important role, more knowledge and thus also more power than host country supervisors. Con-
solidated information, however, relies to some degree on cooperation between home and host country
supervisors, especially for the exchange of soft information. Memoranda of Understanding (MoU) have
typically been used to facilitate the flow of information on a continuous rather than ad hoc basis and
authorize supervisors to exchange confidential information, even though they are not legally binding.
However, the protection of financial and national interest as well as asymmetric information availa-
bility across home and host country supervisors can skew decision-making processes in favor of the
former and at the expense of the latter.

Beyond the information asymmetries, there are other asymmetries in the interests and relative powers
of home and host country supervisors. The differences in interests are especially striking for small host
countries, for whom the subsidiaries of multinational parent banks are often of systemic importance
and thus the stability of the bank a high priority, while the host country operation of the bank is not
material for the parent bank and thus of lower priority for the home country supervisor.

The diverging interests become even clearer during times of distress (D’Hulster, 2012). If the problem
arises in the parent bank, the home country supervisor has strong incentives to delay and minimize

11
Since the global financial crisis, the Basel Committee has developed a Regulatory Consistency Assessment Program, focusing on timely and effective
implementation of its standards. The FSB is also performing regular peer and thematic reviews. These efforts have contributed to a clear improvement
in implementation monitoring in BCBS and FSB member countries.
12
A similar need to tap common resources might arise if the banking system is too large relative to fiscal revenue, and thus becomes too-big-to-save, as
the examples of several countries within the Eurozone have shown (Bertay, Demirguc-Kunt and Huizinga, 2011).

13
information sharing (especially if the host country subsidiary is of material importance for the parent
bank), while the host country supervisor has strong incentives to ring-fence and thus prevent local as-
sets from being up-streamed to offset losses in the parent bank’s financial position or in other parts
of the group. If the problems arise in the subsidiary, on the other hand, the home country supervisor
has incentives to share information with the host country supervisor (if the subsidiary is of material
importance for the parent banks), while the host country supervisor has incentives to overstate the
problem vis-à-vis the home country supervisor (possibly triggering capital and liquidity support from
the parent) but also to ring-fence. Ultimately, in times of distress the interests of home and host coun-
try supervisors are not aligned.

And while the host country supervisor can try to ring-fence the subsidiary in times of distress, this is
often difficult to do given the organizational interdependence (such as common IT platforms and cent-
ralized back offices) across the bank. This asymmetry in relative power is further intensified in Europe
by the fact that home country supervisors in Western Europe have longer and deeper experience in
supervisory cooperation and generally more financial strength than host country supervisors in CESEE.
There might also be a political dimension for EU candidate countries, it is delicate for a candidate coun-
try to challenge EU home supervisors and supervision practices, particularly while relying on EU-funded
technical assistance.

Before the global financial crisis, many countries did not have bank resolution regimes and most re-
lied on corporate insolvency regimes to deal with failing banks. Even when resolution regimes were in
place, for instance in the United States (US), they were nationally based, complicating the resolution
of failing financial institutions active in multiple jurisdictions. This resulted in many breakdowns in
supervisory cooperation during the crisis. Since then, international efforts have focused on developing
effective resolution regimes with the Financial Stability Board’s (FSB) Key Attributes laying out the
international standard. At the global level, crisis management groups (CMGs) and resolution colleges
have been established.

An effective resolution regime should be able to minimize the systemic damage caused by an orderly
collapse of an individual bank without exposing the tax payer to the risk of loss or creating perverse
incentives to take aggressive risks. One of the key features of post crisis resolution regimes is a move
away from bail-outs to ensure losses are born by shareholders and creditors (including via bail-in).

To do this, the regime must provide authorities with the tools to safely and quickly ensure the cont-
inued performance of the bank’s essential functions, including uninterrupted access of depositors to
their funds. Resolution plans and resolvability assessments are to be drawn up to facilitate the swift
resolution of a financial institution and to limit contagion effects and disruption in financial markets.
While this playbook sounds attractive and fair on paper, the practical application in a crisis remains
uncertain.

14
 4. Regulatory reforms in the EU and the Eurozone
4.
TheRegulatory reforms
post-crisis global reformin the EUhave
initiatives and thetranslated
been Eurozone into a new EU regulatory
framework. Prompted by the Eurozone crisis, the Eurozone countries moved beyond
these reforms and established a supranational supervision and resolution framework
to address
The the
post-crisis externalities
global arisinghave
reform initiatives frombeen
the translated
strong ownership
into a newlinkages across
EU regulatory banking
framework.
systems in the EU and the close integration of financial markets. While open to all EU
Prompted by the Eurozone crisis, the Eurozone countries moved beyond these reforms and established
amembers,
supranational supervision
currently andEurozone
only resolution framework
countriestoparticipate,
address the externalities
reflecting arising from the
the additional
strong ownership linkages across banking systems in the EU and the close integration 13 of financial
externality coming from cross-border banking within a currency union.
markets. While open to all EU members, currently only Eurozone countries participate, reflecting the
additional externality coming from cross-border banking within a currency union.13
This corpus of EU regulations, supervisory practices, and procedures, as well as cross-
border
This corpuscooperation requirements,
of EU regulations, y are and
supervisory practices, applicable
procedures,both
as wellatas an individual
cross-border and
coopera-
14
consolidated
tion requirements,level.
y are Hence,
applicablethey arean also
both at relevant
individual for non-EU
and consolidated [Link]. Thearemost
Hence, they also
relevant
relevant forregulations
non-EU [Link] EU
The banks
most andregulations
relevant the scopeforofEUtheir applications
banks and the scope areofsummarized
their applica-
in Table
tions 4.
are summarized in Table 4.

Table 4: EU Prudential Rulebook

Table 4: EU Prudential Rulebook
EU Rulebook

Scope of application

Area Main content Waiver

Consolidated
Solo basis Branches
basis

Parents and
• Definition of CET1, T1 and TCR
subsidiaries on a
• Minimum ratios
solo basis
• Definition of risk-weighted assets
Capital P P û
(RWAs)
• Capital buffers
• Pillar 2 capital requirements
Only domestic
• Leverage ratio (2019)

Parents and
subsidiaries on a
• Definition of High-Quality Liquidity solo basis
Assets (HQLA)
Liquidity P P û
• Minimum liquidity coverage ratio
(LCR) ratio
• Other requirements Domestic and
cross-border

• Fit & Proper for key function holders

Governance P P û No waiver
• Group remuneration
• Risk management standards
ICAAP & • Internal capital and liquidity planning
P P û No waiver
ILAAP • Economic needs
• Capital and liquidity targets
Supervisory • Sanctions
P P P N/A
powers • Early intervention measures
• Supervisory measures
Recovery • Definition of measures to overcome
P û û N/A
planning a crisis
• Recovery indicators

4.1. Supervisory processes and procedures in the EU

The European Banking Authority (EBA) has played a major role in the harmonization
of
13
Thissupervisory processes
externality became clear at the height ofand procedures
the Eurozone across
crisis in 2011/12, the EU.
when regulators across As parttried
the region oftothe banking
ring-fence local subsidiaries
union, the harmonization process has evolved to reach a uniform standard since the
and parent banks considering denomination risk..

establishment of the SSM. The supervisory work is based on regular processes and
procedures (

15
4.1. Supervisory processes and procedures in the EU

The European Banking Authority (EBA) has played a major role in the harmonization of supervisory
processes and procedures across the EU. As part of the banking union, the harmonization process
has evolved to reach a uniform standard since the establishment of the SSM. The supervisory work is
based on regular processes and procedures (Table 5 and Annex for details), which include an annual
Supervisory Examination Process (SEP), an annual Supervisory Review and Evaluation Process (SREP),
annual stress testing, an annual internal capital adequacy assessment program (ICAAP), an annual in-
ternal liquidity adequacy assessment program (ILAAP), an annual review of the recovery plan, and – ad
hoc – internal risk model validation and fit and proper assessments of the new members of the board,
senior managers or other key function holders.

Table 5: Supervisory processes and procedures

Table 5: Supervisory processes and procedures
Supervisor process and procedures

Area Frequency Content Outcome Scope

• Business Model Assessment

• Capital Adequacy • Supervisory score • Consolidated
SREP Annual
• Liquidity and Funding • Pillar 2 requirements • Not applicable to
• Internal Governance • Other requirements branches
• Risks to capital
• On-site inspections
SEP Annual • Planning for the • Consolidated
• Off-site inspections
supervisory year • Applicable to branches
• Others
• Consolidated
Stress Testing Annual • External evaluation
• Capital Adequacy in a stressed scenario • Not applicable to
of capital levels
branches
ICAAP / ILAAP • Consolidated
Annual • Opinion integrated
assessment • Review of the ICAAP / ILAAP process • Not applicable to
into SREP
branches
Recovery Plan • Consolidated
Annual • Supervisor
review • Revision of the group recovery plan • Not applicable to
assessment letter
branches15
Internal Model • Consolidated
N/A • Validation of capital, market, operational
validation • Validation report • Not applicable to
and others capital models
branches
• Assessment of the experience,
Fit & Proper N/A reputation and professional background • Consolidated
• Fit & Proper report
of members of the board and other key • Individual
function holders

Most of these supervisory activities are conducted on a consolidated basis for banking groups with
aMost
cross-border
of thesepresence. EU regulations
supervisory activities impose mandatory
are conducted oncooperation and coordination
a consolidated between
basis for banking
supervisors when performing these tasks,16 including obligations on information exchange and sharing,
groups with a cross-border presence. EU regulations impose mandatory cooperation
both in normal times and in emergency situations. EU consolidating supervisors and 16 the other EU
and coordination between supervisors when performing these tasks, including
competent authorities involved in the supervision of the banking group shall also enter into written
obligationsand
cooperation oncoordination
information exchange and sharing, both in normal times and in
agreements.
emergency situations. EU consolidating supervisors and the other EU competent
authorities
The involved
EU regulatory in theestablishes
framework supervision thatofseveral
the banking
decisionsgroup
shouldshall alsojointly
be taken enterbyinto
thewritten
consoli-
17
cooperation
dating and hostand coordination
supervisors agreements.
within colleges. The main features of these decisions are outlined in Table
6 and further described in the Annex.
The EU regulatory framework establishes that several decisions should be taken
jointly by the consolidating and host supervisors within colleges. The main features of
these decisions are outlined in Table 6 and further described in the Annex.
Table 6: Joint Supervisory Decisions within EU Colleges Kommentie
15
Recovery plan review could be asked on individual basis in the case of EU cross-border banking groups.
INSERT NEW
16
This externality became clear at the height of the Eurozone crisis in 2011/12, when regulators across the region tried to ring-fence local subsidiaries
and parent banks considering denomination risk.

16
Table 6: Joint Supervisory Decisions within EU Colleges
Joint decisions

EBA
Joint decision Content Deadline No joint decision is reached
mediation

1) Consolidated supervisor to
Consolidated and individual /
decide on consolidated basis
1. Joint decision on capital subconsolidated institution specific capital 4 months Binding
2) Host supervisors to decide on
requirements
individual / subconsolidated basis

1) Consolidated supervisor to
Consolidated and individual /
decide on consolidated basis
2. Joint decision on liquidity subconsolidated institution specific 1 month Binding
2) Host supervisors to decide on
liquidity requirements
individual / subconsolidated basis

Validation of first use and significant

Consolidated supervisor to decide on
changes on Credit Risk (IRB), Counterparty
3. Joint decision on internal models 6 months both consolidated and individual / Binding
Credit Risk (IMM), Market Risk (VAR) and
subconsolidated basis
Operational Risk (AMA)

Determination of a cross-border EU branch

4. Joint decision on significant branches 2 + 2 months Host supervisor decides No mediation
as “significant”

Assessment of the Group Recovery Plan, Home supervisor decides on

Binding
and measures regarding Recovery Plan consolidated basis.
5. Recovery Plan 4 months
Decision to require an individual Host decides on individual Recovery
Non Binding
Recovery Plan Plan

6. Group financial support Approval of the agreement 4 months Consolidated supervisor decides Binding

7. Authorisation of actual group financial Allowing or prohibiting/restricting the Decision is made by the providing
5 days Non-binding
support requested support entity

EU regulations require the following tasks to be conducted by the EU supervisory colleges: 1

• The mapping exercise performed by the home supervisor, which allows a better understanding of the
geographic footprint of the banking group and its corporate structure.

• Agreement on the SEP, which covers the activities to be carried out during the year by both the consoli-
dating supervisor and the host authorities.

• The potential entrustment of tasks and delegation of responsibilities across the different authorities of
the supervisory college, to avoid the unnecessary duplication of supervisory requirements.

• The group risk assessment, including the liquidity risk assessment as performed by the home supervisor.
The risk assessment is needed to reach a joint decision on the capital and liquidity requirements, since
additional Pillar 2 requirements and other supervisory measures are based on the outcome of this risk
assessment. The group risk assessment should follow a standardized reporting template, usually inclu-
ding the evolution of all risk categories and forward-looking views on all the risk elements. It (the risk
assessment) also covers conclusions on the risks and vulnerabilities faced by the group. In most of the
supervisory colleges, a significant amount of time is devoted to discussing the risk assessments perfor-
med by the different participants in the supervisory college.

• Making joint decisions (open to binding EBA mediation as outlined in Table 6) on institution-specific capi-
tal and liquidity requirements, on the assessment of the group recovery plan, on group financial support,
and on internal model validation.

• Contingency planning for emergency situations, including the exchange of information on early warning
signs, potential risks and vulnerabilities, and establishing a college framework for emergency situations.

• Management of emergency/crisis situations: including leading a coordinated response by the

consolidating supervisor to crisis situations.

17
Progress in the development of some of these tasks is more advanced than in others. In most of the
supervisory colleges, the group and subsidiaries risk assessments are performed, and a significant
amount of time is devoted to these discussions. The members of the college also participate in the
joint decisions on capital, liquidity, and recovery plan; with the capital decision more developed that the
others. The requirement to produce a SEP is only complied with through the compilation of the indivi-
dual programs of the consolidating and host supervisors. There has been little progress on the roles of
the supervisory colleges on crisis contingency planning and crisis management.

The EU home supervisor is responsible for establishing a supervisory college, with additional members
comprising: (i) the competent authorities responsible for the supervision of subsidiaries, (ii) the compe-
tent authorities of host Member States where significant branches are established, (ii) central banks of
Member States that are involved (in accordance with their national law) in the prudential supervision
of legal entities but are not competent authorities, and (iv) the EBA. In addition, the home supervisor
can, with the previous agreement of the other members, invite other authorities as observers. While
their status varies between colleges and authorities, observers typically receive only partial information.

4.2. Bank Resolution in the EU

The EU framework for bank resolution, recovery, and partly also early intervention, is described in the
Bank Recovery and Resolution Directive (BRRD).18 It lays out the establishment of an independent
resolution authority and the powers to ensure that failing banks are resolved in accordance with the
resolution objectives. The policy objective of resolution is the continuity of critical functions to avoid
significant effects on financial stability. This includes the protection of public funds by minimizing
reliance on extraordinary public financial support, the protection of insured depositors and investors,
and the protection of client funds and client assets. The so-called “failing or likely to fail” (FOLTF) test
creates a standardized but flexible set of resolution triggers to ensure the application of the special
resolution powers are taken early enough and justified by the “public interest test.”

The BRRD has introduced several resolution tools not previously available to most Member States
authorities. Through the sale of business, the resolution authority can transfer the shares of the bank
or (part of) its assets and liabilities to a third-party private sector buyer. If no private sector buyer is
available but the resolution authority still considers it is possible to sell the bank in the future, it may
decide to transfer the (good part of) assets and liabilities of the failing bank, or its shares, to a bridge
institution. The latter will hold those assets and liabilities until a sale and purchase transaction can
be arranged. Importantly, the bridge bank will need to be managed and it is often challenging to find
the relevant skills in the public sector. However, if there is no realistic prospect of finding a buyer, the
resolution authority may pursue an open-bank bail-in through the bail-in tool, whereby the claims of
shareholders and creditors of the bank are written down or converted into equity to absorb losses and
provide the fresh capital the bank needs for ensuring the continuity of critical functions. Finally, any
of these tools can be combined with the asset management company tool that, in some cases, allows
the separation of illiquid and impaired assets from the failing bank to be transferred to a separate
company or “bad bank”.19

When applying resolution tools, the resolution authority has extensive resolution powers to take control
of the bank and take over ownership rights. The resolution authority can write down, restructure, or
convert the debt instruments issued by the entity, apply a moratorium to debt instruments, and stay
termination rights. It can also require the continuity of operating services by third-party providers,
appoint special managers, etc.

18
For a detailed analysis see: Lintner, P (Hg), Understanding bank recovery and resolution in the EU: a guidebook to the BRRD, December 2017, World
Bank, FinSAC.
19
These resolution tools can be used in combination or individually, except for the asset management company which can be used only in combination
with other resolution tools. For example, the resolution authority may bail-in shareholders and subordinated creditors of a resolved bank to absorb
losses, and then use the sale of business tool to sell the bank to another private sector entity.

18
The BRRD requires several steps for resolution authorities to prepare for resolution including develo-
ping resolution plans that identify the preferred resolution strategy for the banking group. Resolution
planning also requires the determination of the “points of entry” (the legal entities where the resolution
powers are expected to be applied), resolvability assessments undertaking, and determination of mini-
mum requirement for own funds and eligible liabilities (MREL) requirements (Box 2).

Figure 5: Resolution and resolvability planning

STRATEGIC
BUSINESS OPERATIONAL INFORMATION
RESOLUTION RESOLUTION AND AND RESOLVABILITY
ANALYSIS AND STRATEGY TOOLS ASSESSMENT
FINANCIAL COMMUNICAT
INFORMATION
ON LIABILITIES CONTINUITY ION PLAN

NORMAL POINTS OF BARRIERS

STABILIZATI RESTRUCTURI FINANCIAL OPERATIONAL INFORMATI COMMUNICA
INSOLVENCY ENTRY: SPOE AND MREL
PROCEEDINGS vs MPOE ON TOOLS NG TOOLS CONTINUITY CONTINUITY ON TION PLAN MEASURES

Box 2: Total loss absorbing capacity and minimum requirement for own funds
and eligible liabilities, including setting criteria

TLAC and MREL and MREL-setting criteria

Total loss absorbing capacity (TLAC) for global systemically important banks (G-SIBs) vs the MREL under
the BRRD

During the global financial crisis and the Eurozone crisis, regulators were reluctant to bail-in bond holders fearing
contagion effects on financial markets. Consequently, post-crisis reforms included the introduction of TLAC and
MREL requirements to ensure enough high-quality loss absorbing instruments are held by banks to facilitate bail-in.

The TLAC is required only for G-SIBs according to criteria defined by the FSB. A G-SIB should hold a TLAC amount
no lower than 16% of its risk-weighted assets (RWAs) (or 6% of their leveraged exposure) from 2019 onwards and
at least 22% (or 6.75% of their leveraged exposure) from 2022 onwards. For single point of entry institutions, the
rules are complemented with the so-called “internal TLAC”, by which the TLAC resources raised by the parent
entity are also down-streamed, to a certain extent, to the “material subgroups” that make up the banking group
(between 75% - 90% of the external minimum TLAC requirement that would apply to the material sub-group if it
were itself a resolution group). In addition, the TLAC standards provide specific rules to calculate TLAC require-
ments for multiple point of entry institutions, considering that each point of entry within the group should raise
its own resources.

The criteria for the eligible instruments include that the instruments should be paid-in, have a residual maturity
of more than one year, should be directly issued by the entity (not through special purpose vehicles). A significant
share of the eligible TLAC instruments should have a certain degree of subordination to operational liabilities on
which the performance of critical functions depends (in general terms this are deposits and also derivatives), in
order the enhance resolvability and reduce NCWOL risks.

MREL was first introduced by the BRRD in 2014, with the calibration criteria developed in a European Commission
Delegated Regulation (DR 2016/1450 or DR). In principle, MREL is applicable to all EU banks and banking groups,
not only G-SIBs. As opposed to TLAC, the calibration of MREL is flexible, lacking a specific minimum requirement.
This means it has to be calibrated for each individual bank – at consolidated and individual entity level - by the
competent resolution authority.

19
 A bank’s MREL requirement, is guided by the above-mentioned DR and should be calibrated taking into
account two components:

• A default loss absorption amount (LAA), representing the quantum of losses an institution or
group should be capable of absorbing. It is equivalent to the bank´s capital requirements; and

• A recapitalization amount (RCA), defined as the amount necessary to implement a bank‘s resolution
strategy. In other words, if the bank is expected to be resolved under a bail-in resolution strategy, the
RCA represents the resources that the bank would need to generate fresh capital through
recapitalization to comply with the applicable capital requirements.

• There is room to tailor MREL to bank-specific features.

° The leeway for adjusting the default LAA upwards or downwards is closely related to
supervisory stress tests, the resolvability assessment and the SREP. For example, the
resolution authority may not include in the LAA of MREL the capital requirements
that are imposed as the outcome of a stress testing exercise.

° The recapitalization amount can be adjusted downwards based on the preferred resolution
strategy, the bank’s business model, funding and risk profiles, as well as its resolvability
assessment. For example, the default recapitalization amount may be adjusted downwards if
the resolution strategy of the bank is expected to be based on a sale of business rather than
on an open bank bail-in.

° Additional adjustments to account for the contributions from deposit-guarantee

schemes (DGS), a bank‘s size and systemic risk, and the exclusions of liabilities from
a bail-in are possible.

The BRRD also defines the eligible instruments that comply with MREL, with requirements more flexible than for
TLAC. An instrument would be eligible for MREL if it has been issued and paid-in, its residual maturity is longer
than one year, and it does not arise from a derivative exposure. The ranking of the instrument should not be lower
than senior – introducing the new subordinated class of “senior non-preferred”20 - and the resolution authorities
may introduce further criteria for ensuring that the instruments can absorb losses.

Finally, negotiations to modify MREL standards and to introduce TLAC for EU G-SIBs are ongoing. More detailed
rules are expected in several areas, for instance clearer rules on MPE and SPE banking groups (including internal
MREL), specific TLAC requirements (more consistency between capital requirements and MREL/TLAC require-
ments), and more criteria, especially regarding subordination, on eligible instruments for TLAC and MREL.

One critical issue in resolution planning, especially in the context of cross-border banks is the concep-
tual discussion of SPE vs. MPE resolution strategies.21

• An SPE strategy identifies a single resolution group, composed of one resolution entity, i.e. the entity to
which a single resolution authority would apply resolution tools, and the non-resolution entities controlled
or owned by the resolution entity.

• An MPE strategy foresees the application of resolution tools by two or more resolution authorities to two
or more resolution entities. In these cases, the banking group would be considered as divided in more reso-
lution groups (each one with one resolution entity and the non-resolution entities controlled or owned by it).

• A combination of the approaches is also possible and foreseen under FSB standards and the Single Reso-
lution Board (SRB) resolution planning manual.

20
Senior non-preferred debt ranks senior to regulatory capital instruments (CET1, AT1 and Tier 2) and other subordinated debt but junior to senior debt
(such as deposits and ordinary creditors). See Directive (EU) 2017/2399 as regards the ranking of unsecured debt instruments in insolvency hierarchy
21
FSB, Guiding Principles on the Internal Total Loss-Absorbing Capacity of G-SIBs (‘Internal TLAC’), July 2017.

20
The choice of a particular strategy takes account of the existing structure and business model of the
individual firm and relevant characteristics. In general, the SPE approach is more likely to be chosen by
resolution authorities for banks that are centrally structured and operated, that are mainly funded th-
rough the parent company, and that can transfer losses from other entities of the bank to the parent
company. On the other hand, the MPE approach is, in general, more likely to be suitable for banks that
have material subsidiaries that are operating with limited interconnections to other resolution groups.

An SPE strategy is based on a consolidated approach to resolution where resolution tools are expected
to be applied only at the level of the parent entity, which is called “resolution entity” for these purpo-
ses. The SPE approach is therefore, in principle, more adequate for closely integrated banking groups.
Debt is issued by the resolution entity and downstreamed to other entities of the group, so only the
shareholders and the creditors of the resolution entity are expected to absorb losses in resolution. To
ensure that losses in the subsidiaries can be up-streamed to the resolution entity, an amount of so
called “internal loss absorbing capacity” should be prepositioned on the subsidiaries’ balance sheet or
readily available to recapitalize any direct or indirect subsidiaries (for example, via collateralized gua-
rantees). Currently, SPE is the predominant resolution strategy agreed among resolution authorities for
G-SIBs and also for banking groups within the banking union. In case of capital needs of a subsidiary
under SPE, the expectation is that the parent would step up to its rescue and therefore wouldn’t be
expected to enter resolution. However, in the extreme case, resolution action (or WDCC) may be needed
to be taken by the host authority in case internal LAC needs to be formally converted and the internal
resources provided by the resolution entity are not enough to restore the subsidiary.

If the subsidiary is identified as a resolution entity as part of a MPE strategy, shareholders and credi-
tors of that subsidiary are expected to assume the losses following the application of the resolution
tools. MPE is generally suitable for banking groups consisting of self-sufficient, autonomous financial
subgroups operating in different countries with limited interconnections with other resolution groups.
Hence, MPE banking groups typically consist of subsidiaries instead of branches, with local deposit
bases or wholesale funding, raised using their own financial strength. There is typically a high degree
of operational and IT flexibility through independent IT or centralized entities. From a resolution plan-
ning standpoint, this implies that the MREL of the subsidiary is in principle expected to be issued to
third-party investors (external MREL) and both the group-level resolution authority and the local re-
solution authority would identify obstacles, related to the interconnections among the different legal
entities that may impede the separate resolution of the subsidiary. Though it should be stressed, that
internal LAC coming from the parent does not per se create an impediment to the subsidiaries’ resol-
vability under MPE; it may create an obstacle for the resolvability of the parent if it entails a risk of
contagion. So far, MPE has been selected as a resolution strategy in only a relatively small number of
G-SIBs, most prominently Santander and HSBC.

While the legal requirements are the same under the BRRD for both SPE and MPE strategies , the
MREL (and TLAC) structures will be different. As can be seen in Table 7, within a SPE banking group,
the parent company (the “entry point”) is expected to raise the MREL resources from third-party inves-
tors and then downstream them to other parts of the banking group. However, within an MPE banking
group, the different resolution entities are in principle expected to raise MREL from investors (though
there is no legal requirement). Other solutions may also be available to increase the LAC at local level
(e.g. by retention of profits). These differences influence the calibration of the MREL amounts at the
consolidated and bank individual level.

22
The proposed amendments under Art.72b(2b) EU Capital Requirements Regulation 2013 do not exclude the option that MREL would come from
within the banking group also under an MPE strategy.

21
is no legal requirement). Other solutions may also be available to increase the LAC at
local level (e.g. by retention of profits). These differences influence the calibration of
the MREL amounts at the consolidated and bank individual level.

Table 7: Minimum requirement for own funds and eligible liabilities and
resolution strategies
Table
MREL7: Minimum requirement for own funds and eligible liabilities and resolution strategies

Block SPE MPE

Sub-consolidated

Scope of application Consolidated RWAs RWAs (“resolution group”)

Entry-point based

Capital requirements (Sub-consolidated) +

Capital requirements for MREL injected in other
Calibration Capital requirements + Recapitalization needs
points of entry + Recapitalization amount for
the same point of entry

Yes, for subsidiaries within the same resolution

group to the resolution entity
Internal MREL Yes, for subsidiaries to the resolution entity
Not legally prohibited for resolution entry
subsidiaries to another resolution entry entity
i.e. the parent

As in supervision, there are many areas in resolution where cross-border coordination is needed
and
As inprescribed by thethere
supervision, BRRD (Table
are many 8).23
areas in resolution where cross-border coordination
is needed and prescribed by the BRRD (Table 8)23.
Table 8: Joint decisions in resolution planning
No Joint Decision is
Table 8: Joint decisions
Joint Decision Content in resolution planning
Deadline
reached
EBA mediation

Joint decisions in resolution planning Group level resolution

authority decides at group
Joint No Jointlevel
Decision is
Group Content
Decision on group Deadline EBA mediation
Decision reached Binding, but fiscal
Resolution resolution, individual parent 4 months
impingement clause
Plan company, and subsidiaries
Group
Local level resolution
resolution authority
authority
decides decides
on an at
individual
resolution plan
group level
Decision on group
Group
resolution, individual Binding, but fiscal
Resolution 4 months Group level resolution
parent company, and impingement clause
Plan authority decides at group
subsidiaries
Decision on the resolvability Local resolution
level authority
Group decides on an individual
of the banking group and Binding, but fiscal
Resolvability 4 months resolution plan
24 identifying obstacles to impingement clause
Assessment
resolution
Local resolution authority
Group level
decides resolution
an individual
authority decides
resolution plan at
Decision on the group level
Group resolvability of the Group level resolution Binding, but fiscal
Resolvability banking group and 4 months authority decides on
24 impingement clause
Assessment identifying obstacles to measures taken at parent
Remove resolution Local resolution
level authority
Adopt measures to remove decides an individual
impediments 4 months Binding
impediments resolution plan
to resolution
Local resolution authority
decides on measures taken
at local level

Group level resolution

authority decides on MREL
22 at parent level
The proposed amendments under Art.72b(2b) EU Capital Requirements Regulation 2013 do not
exclude
MRELthe option that MREL
Determine would come from
MREL levels within the banking group also under an MPE
4 months strategy.
Binding

Local resolution authority

decides on MREL at local
23For procedural details see for example: Lintner, P. (2018): De/centralized
level
Decision Making Under the
European Resolution Framework: Does Meroni Hamper the Creation of a European Resolution Authority?

24
23 Technically
For the resolvability
procedural details see for example:assessment
Lintner, P. (2018):isDe/centralized
part of the Decision
same joint
Makingdecision
Under theof the group
European resolution
Resolution plan
Framework: Does Meroni
(as perthedelegated
Hamper Creation of aregulation 1075/2016).
European Resolution Authority?
24
Technically the resolvability assessment is part of the same joint decision of the group resolution plan (as per delegated regulation 1075/2016).

SMALL HOST SUPERVISORS 21

22
Resolution colleges are the main instruments in the EU for cross-border cooperation, both for prepara-
tion and for the execution of resolution actions (Table 7). EU host authorities (i.e. resolution authorities
but also finance ministries, competent authorities, and DGS authorities) of subsidiaries and significant
branches have member status in the college. For EU G-SIBs, the BRRD allows that the functions of
the resolution college can be assumed by the CMGs25, without the need to set up two different and
parallel structures.

Within resolution colleges, the group level resolution authority (e.g. the resolution authority of the
ultimate parent entity) and the members (i) exchange information required for resolution planning
purposes, (ii) develop the group resolution plan (including the joint decision), (iii) discuss the group re-
solvability assessment (including the joint decision), (iv) discuss the measures to remove the obstacles
to resolution (including the joint decision), and (v) discuss MREL requirements (including the joint de-
cision) as outlined in Table 7.

Colleges are also the forum for coordination and cooperation when resolution decisions involving
cross-border banking groups are made. In resolution colleges, the group-level resolution authority and
the other members decide on the need for a group resolution scheme, coordinate the public commu-
nication process of group resolution strategies and schemes (if applicable), and coordinate the use of
financial arrangements (if needed). If there is no agreement on the group resolution scheme, the EBA
may assist in reaching a joint decision.26 In the end, each resolution authority may make its own deci-
sion for financial stability reasons, but it is obliged to provide a detailed reasoning to college members.
Resolution authorities which reach a joint decision may go ahead in unison for the parts of the group
under their respective jurisdictions.

Banking union. The Eurozone countries have gone further in centralizing supervision and resolution
functions than the EU. The SSM was established when the ECB took over responsibility for bank super-
vision in the euro area in late 2014, following a year-long ‘comprehensive assessment’ to identify legacy
issues. By that time, European authorities had agreed on the second pillar of the banking union – the
SRM – to come into effect in 2016. The SRM is based on centralized decision making for resolution by
the SRB. The latter has the national resolution authorities (NRA) as its constituent members, the Euro-
pean Commission and the ECB as observers, and the European Council in cooperation with the Europe-
an Commission with veto powers. A centralized solution is yet to be established for deposit insurance.27

4.3. Cross border expansion: legal structures

The cross-border expansion of European banks has been achieved mainly through (fully-owned) sub-
sidiaries. This has occurred despite the “passporting regime” allowing an institution that has been
licensed as a bank in one Member State to use the right of establishment and the freedom to provide
services within other Member States. European banks currently use branches mainly for very specific
purposes, for example to (i) conduct certain wholesale or investment banking activities, (ii) raise who-
lesale funding in relevant financial centers, (iii) enter the banking market of a country (before acquiring
a local bank or, less frequently, asking for a banking license to grow the business organically), and (iv)
exclusively undertake cross-border commercial and retail banking, mainly in Member States.29 30

25
CMGs have usually more competences than the resolution colleges. For example, one of their responsibilities is the assessment of the
group recovery plan, which is typically done, in the supervisory colleges according to the European regulations.
26
But no binding mediation is stipulated.
27
Proposal for a Regulation of the European Parliament and of the Council amending Regulation (EU) 806/2014 in order to establish
a European Deposit Insurance Scheme, COM/2015/0586 final.
29
Nordic banks in their cross-border expansion to certain Baltic countries (e.g SEB, Danske Bank) and some Benelux banks
in other EU countries (e.g ING).
30
The use of branches for cross-border banking activities by EU banks outside the EU to engage in deposit-taking activities is virtually non-existent.

23
Box 3: Nordea

Nordea: branching and regulatory arbitrage

Nordea is a multinational banking group with a business model focused on retail and commercial banking throug-
hout the Scandinavian region. It was the result of the merger of banking groups of Finland, Sweden, Norway and
Denmark between 1997 and 2000. It has a relevant presence in the Baltic States and, to a lesser extent, Russia.
It divested its Polish banking operations in 2013. It has been identified as a domestically important systemic ins-
titution in many of the countries where it is currently present.

Nordea´s model has been characterized by an integrated capital, funding, operating, and commercial model across
the Scandinavian countries. Intra-group liquidity flows represent a very important share of the assets and liabili-
ties of the main banking entities. In fact, the entities are so tightly integrated that they resemble branches instead
of subsidiaries.

Nordea group has traditionally been headed by the Swedish banking entity, which means that the Swedish FSA
(Finansinspektionen) has been its consolidating supervisor, while the European Central Bank (ECB) (for the Finnish
subsidiary), Norway´s Finanstylsinet, and the Danish FSA have been, among others, its host supervisors.

Recently, Nordea´s managers were seeking to restructure the bank, including via mergers, with the objective to
change consolidating supervisory responsibility of the Swedish FSA. In 2016, according to the press and public
statements by Nordea´s chairman, the bank made an offer to take over Dutch state-owned bank ABN Amro. The
potential transaction was justified by the possibility of some regulatory relief, since the combined bank would be
in the Netherlands and, consequently, the consolidating supervisor would be the ECB.

However, the transaction was not successful. In 2017 and 2018, Nordea took two major restructuring steps. First,
in January 2017, it merged its main banking entities into its Swedish bank. With the resulting structure, it became
the most significant case of cross-border branching in the EU as it started to operate in Finland, Norway, and
Denmark with branches instead of subsidiaries. Second, in September 2017 the board of directors of Nordea de-
cided to start a re-domiciliation process from Sweden to Finland, with the objective to domicile its parent entity
in the banking union.

It applied for a banking license in Finland, which was granted by the ECB in March 2018. And it merged the Swe-
dish bank (the former parent entity) into the Finnish new legal entity. After merging the two banks, the parent en-
tity would be the Finnish Bank (Nordea Bank AB) and, as a result, Nordea would perform also its banking activities
in Sweden through a branch. The consolidating supervisor shifted from the Swedish FSA to the SSM.

According to Nordea, the decision to re-domicile in Finland, part of the banking union, would secure a fair, stable,
and predictable regulatory environment for the bank on par with its peers. It was done to promote the interests
of customers, shareholders, and employees. In particular, Nordea justified that the movement would result in
material cost savings with a net present value between 900-1,200 million EUR, taking into account fees from the
resolution fund, contributions to deposit guarantees, and other transitional effects. Whether the stricter approach
to capital requirements by the Swedish FSA played a role in the relocation decision is of course difficult to judge.

In recent years the relative power of EU branch host supervisors, and particularly Eurozone hosts,
vis-à-vis home country supervisors has improved. Branches may now be considered, for EU purposes, as
(i) non-significant, (ii) significant, and (iii) significant-plus, according to criteria defined in article 51 of
the Capital Requirements Directive (CRD) IV (Table 9). The consideration of a branch as significant
does not alter the rights and responsibilities of the authorities under EU regulations and directives.

24
Table 9 illustrates the typical structures used in cross-border banking across the EU and the impact
on supervision and resolution

Table 9: Regulatory classification of EU branches31

Recent strengthening of the relative power of EU branch host supervisors can be explained by several
factors. First, the Solvency Directive32 introduced the concept of a significant branch and gave host
supervisors the right to participate in the supervision of the branch, and to receive timely information
from the home supervisor. This amendment, now included in article 51 of CRD IV (see Table 9), de-
fines the criteria to determine when a branch is significant. It then relies on a joint decision process
to identify the branch as significant. This regime has been further strengthened by the new concept
of the significant-plus branch, that requires a deeper involvement from both the home and the host
supervisor in the oversight of activities performed through the branch.33

Second, the EU has continued to harmonize regulations ranging from prudential requirements to consu-
mer protection, thus reducing compliance costs for banks and making the sharing of central services
across borders easier.

Third, the implementation of the European banking union—with the abolishment of the traditional
home-host division of tasks for significant institutions (SIs) and the harmonization of supervisory
practices across the Eurozone—makes the branch structure more feasible, because responsibilities for
supervision of both home and host functions are taken over by one supervisor, the ECB. Subsidiaries
can thus be more easily converted into branches. However, the difference in regulatory regimes, most
importantly the deposit insurance regime, remains the main obstacle to branching34. Moreover, the
impact of digitalization on banks may render a network of physical branches less necessary with the
massive use in retail banking of new digital distribution channels (internet, mobile banking, etc.).

31
Note that for banking union countries the ECB does not take the role as host supervisory authority for a branch from a non-Eurozone bank.
This role is assumed by the host Member State’s National Competent Authority.
32
Directive 2009/111/EC of the European Parliament and the Council of 16 September 2009 amending Directives 2006/48/EC, 2006/49/EC
and 2007/64/EC (CRD II).
33
The EBA has issued Guidelines on how the criteria and process to determine significant-plus branches and also on the content of the reinforced
supervisory activities on the branch (EBA Guidelines on supervision of significant branches- EBA/GL/2017/14 of 16/02/2018).
34
As it is explained below, the creation of the European Deposit Insurance Scheme (EDIS) is expected to be a decisive factor to take into account when
deciding whether to establish a branch or a subsidiary.

25
Cross-border branching will most likely become more prominent in the Eurozone. Fostering concentra-
tion among Eurozone banks and creating truly pan-European banks is a European policy goal. In other
words, the objective is to have fewer banking groups with a wider geographic reach and substantial
use of cross-border branching. However, significant obstacles to cross-border branching in the rest of
the EU remain. Despite recent strengthening, the rights and obligations of home and host supervisors
with regard to branches still appear unbalanced. Furthermore, memories of the failed Icelandic banks
raising deposits in EU countries through branches (where host deposit insurance schemes had to step
in to protect local depositors) are still alive, resulting in reluctance against expansion through branches
rather than subsidiaries.

For host countries outside the EU, where no passporting rights exist, supervisors will continue to op-
pose the provision of commercial banking services through branches, since it will limit their ability to
supervise the local operations of the bank. In addition, from a home perspective, some EU supervisors
have shown a clear preference for cross-border banking services to be conducted through separately
capitalized and funded subsidiaries.

When non-Member States participate in EU supervisory colleges as hosts, they always do so as ob-
servers. EU host supervisors of non-significant branches, and other non-prudential supervisory autho-
rities (like market/conduct supervisors or authorities in charge of anti-money laundering and terrorist
financing supervision) may also be designated as observers.36 The EU home supervisor also ensures
that the members and observers of the supervisory college sign written cooperation and coordination
agreements.

We discuss details of both SSM and SRM in Section 5.1 in the context of cross-border cooperation
from the viewpoint of small Eurozone hosts countries. Table 10 summarizes the supervisory and regu-
latory arrangements for different corporate structures of cross-border banks within the EU, including
responsible authorities, capital and liquidity requirements, MREL requirements, and which country is
responsible for deposit insurance

Table 10: EU
EU corporate corporate
structures structures
for cross-border for cross-border banking
banking

Authority Authority
Legal Capital / Liquidity MREL
responsible for responsible for Deposit insurance
Entity/Structure requirements requirements
supervision resolution

Subject to capital
Host supervisor Host resolution
MPOE subsidiary and liquidity Subject to MREL Host country
authority authority
requirements

Capital requirements
Host supervisor and potential waiver Host resolution
SPE subsidiary Internal MREL Host country
authority on liquidity authority
requirements

Capital requirements
(future potential
SPE subsidiary Internal MREL, if
SSM waiver) and potential SRB Host country
(Eurozone) applicable
waiver on liquidity
requirements

Home supervisor
authority,
Home resolution
Significant branch information sharing Not applicable Not applicable Home country
authority
with host supervisor
authority

Home supervisor Home resolution

Non-significant Not applicable Not applicable Home country
authority authority
branch

35
That has been the case, for example, of the Austrian authorities (the Financial Market Authority and the OeNB). They publicly stated in 2011 that the
foreign subsidiaries of the Austrian banks should restrict their loan-to-deposit ratio to levels lower than 110%, effectively limiting the amount of funding
that can be down-streamed from the parent companies to those subsidiaries. Other supervisors have even more experience limiting the use of branches
and intragroup funding for subsidiaries in banking groups, as the model for cross-border banking in Latin America developed by Spanish banks shows.
36
In the context of the European banking union, national competent authorities represented in the Joint Supervisory Team also participate in the super-
visory colleges as observers.

26
5. Cross-border cooperation in supervision
and resolution - practice

Geographically, a legal and regulatory distinction can be made between three types of home-host coun-
try relations: (i) parent bank and subsidiary are both located within the Eurozone, (ii) parent bank and
subsidiary are both located within the EU, but the subsidiary is not located in the Eurozone and (iii) the
parent bank is located in the EU, the subsidiary is outside the EU. In the latter category, we distinguish
between candidate and non-candidate host countries.

The distinction is critical in terms of institutional responsibility as well as application of the EU regu-
latory framework. For all EU authorities the same regulatory framework and automatic membership in
colleges including binding EBA mediation applies. For the euro area the traditional home-host divided
responsibilities are, however, largely centralized at the ECB/SRB. Non-EU subsidiaries or branches are
within the scope of the home supervisor’s consolidated EU requirements. Moreover, the consolidated
and requirements may have an impact on the activities as well as capitalization and liquidity positions
in the host country. Although, the binding legal and regulatory basis for coordination and cooperation
in the EU colleges does not apply to non-EU hosts they may be granted observer status.

Particularly worrisome for host supervisors are those situations where they do not have timely and
comprehensive access to relevant group information. This may occur when the consolidating supervi-
sor is paying less attention to the bank’s activities in the host country (due to the fact of its non-mate-
riality to the group), is not conducting preparations for group-wide emergency situations, and no strong
relationship between the consolidating supervisor and the host supervisor exists. In these cases, the
host supervisor may decide to take ring-fencing measures to separate the bank from the rest of the
banking group as a second line of defense.

That said, changes in business models following the crisis have made subsidiaries less reliant on fun-
ding from their parent companies, which can increase the effectiveness of potential ring-fencing mea-
sures taken by both home and host supervisors. It is important to differentiate between ring-fencing
that is part of a policy framework, where the requirements have been consulted upon and are publicly
disclosed, and sudden ad hoc ring-fencing decisions during a crisis. While legitimate and, in some cases,
needed to protect host country interests, sudden ring-fencing measures usually result in suboptimal
outcomes for both home and host countries in the longer term. In the case of host countries, these
actions can result in underinvestment by foreign groups in their host country subsidiaries or even force
them to divest their ring-fenced banking businesses. In the case of home countries, ringfencing may
produce a suboptimal capital allocation, where ringfenced subsidiaries may allocate too much capital
at the expense of the parent institution, and even force the subsidiary to take on too much risk to
reach an acceptable return on the invested capital.37

In the next section, effects of EU supervision and resolution regulation of cross-border banks will be
analyzed from the viewpoint of (i) small host countries inside the Eurozone, (ii) small host countries
that are non-Euro EU member states, and (iii) small host countries that are not members of the EU.
In each section, we discuss first the framework for supervisory cooperation, then for cooperation bet-
ween resolution authorities, and conclude with specific concerns of small host countries and possible
reactions by these countries.

37
A more detailed analysis of ring-fencing measures for the home supervisor, the host supervisor, and financial stability can be found
in D’Hulster and Otker-Robe (2015).

27
5.1. ”Eurozone small hosts”38

The Eurozone countries have gone further in centralizing supervision and resolution functions than
the EU. The implementation of the banking union has resulted in the elimination of the traditional
“home-host” distinction for significant institutions as all responsibilities are attributed to the SSM/
SRM.39 Thus, for a cross-border banking group headed by a Eurozone bank where one of its subsidiaries
is based in a host Eurozone Member State, the ECB (SSM) is the supervisor and the SRB (SRM) the
resolution authority for both the parent entity and the subsidiary, even if they are located in different
Eurozone Member States. The SRM is supported by a single resolution fund (SRF), ex-ante financed by
annual contributions paid at solo level by all credit institutions and certain investment firms establis-
hed in the 19 SRM Member States.40

5.1.1. Overview: Banking supervision in the Eurozone

The SSM is the authority responsible for the supervision of significant Eurozone banks. It relies heavily
on the resources of national supervisory authorities, which gives them an opportunity to be heard as
they can voice any concerns through their membership of the Joint Supervisory Teams (JSTs) tasked
with the day-to-day supervision of banks and banking groups. Hence, the national host supervisors
retain access to relevant information, even though in practice they have lost decision power. National
supervisory authorities still participate as observers in ECB supervisory colleges as they are members
of the JSTs. Observers cannot vote in the decisions of the college.

The SSM exercises supervision both at the group level as well as at individual entity level ensuring that
the subsidiary has sound levels of capital and liquidity and that its internal governance is effective.
The SSM conducts the SREP and decides on other supervisory measures with regard to the group at
consolidated level, but also to the subsidiary at solo or sub-consolidated level. Similarly, the ECB is the
authority in charge of deciding on intra-border capital and liquidity waivers on individual basis to parent
entities and banking subsidiaries across banking groups.41

The SSM is now also the responsible body for ensuring that distressed banks are addressed in a timely
manner. It reviews the annual group recovery plan, demanding an adequate coverage of all the material
legal entities (MLE) included in the group regardless of whether they are located within or outside the
EU.42

38
The term is used for mere simplification purposes. Conceptually Eurozone small host authorities are non-existent as both the responsibility for the
group as well as for the individual entities located in the Eurozone has been assumed by the SSM/SRM.
39
Note that for BU countries the ECB does not take the role as host supervisory authority for a branch from a non-BU bank. This role is assumed by the
host Member State’s National Competent Authority.
40
Banks that don’t come under the SRBs remit are to pay (lump sum) contributions to the SRF. If there is a need for resolution funding for these banks,
they would ad hoc come under the SRBs responsibility rather than NRAs.
41
Currently, the CRR only considers the possibility of granting a capital waiver to a subsidiary when both the subsidiary and the parent entity are located
in the same Member State (intra-border capital waiver). Nevertheless, under CRR2 EBA is mandated to review this issue (the original Commission pro-
posal included the possibility for waivers to a subsidiary when its head office is in a different Member State than the parent entity (cross-border capital
waiver). With regard to liquidity requirements, the CRR already considers the possibility of granting both intra-border and cross-border waivers, through
what is known in the regulation as “liquidity sub-groups”. If granting a cross-border capital waiver to a subsidiary were to become reality, the SSM could
issue waivers for subsidiaries in small host countries on their solvency requirements. As a result, a cross-border banking subsidiary with waived capital
and liquidity requirements (and, with a guarantee from the parent entity) could become the prime example of future European banking group structure
with more prominent cross-border branches.
42
The consideration of a subsidiary being an MLE is unrelated to the decision if a third country subsidiary is considered “significant” in the host country
to access to the Supervisory College.

28
5.1.2. Overview: Bank resolution in the Eurozone

With the establishment of the SRM,43 the SRB is the authority responsible for ensuring that the
concerns of both home and host countries are considered.44 However, a strong role for the NRA in the
execution of resolution decisions as well as in the planning phase remains. NRAs, as part of internal
resolution teams led by the SRB, are deeply involved in drafting resolution plans, resolvability assess-
ments, communicating with banks, engaging with them, addressing their doubts, but in accordance
with the general policies and criteria approved by the SRB (following discussions in its plenary session
and sub-committees where every NRA within the banking union is represented).45

The general principles that underpin the SRM establish that its decisions and resolution actions should
take into consideration and carefully balance the interests of all participating Member States.46 Decisi-
ons on individual bank matters are taken by the SRB’s extended executive session formed by the four
SRB Board Members, the SRB Chair and the NRAs concerned. If no consensus can be reached, the five
permanent members have the power to decide on bank individual matters.47 In addition, the European
Commission and European Council have veto rights on SRB decisions on resolution actions (not in the
planning phase), which raises some concerns on whether decisions can be taken quickly enough and
whether political and non-stability concerns will reduce the efficiency of the decision process.48

The Eurozone NRA concerned, together with the SRB, make up the internal resolution team. They
jointly perform all the activities required to conduct resolution planning, and hence host country re-
solution authorities, even from small hosts, are also fully involved in the planning process.49 Internal
resolution teams have become a channel through which small Eurozone hosts can voice their concerns
and protect their interests in case these are not aligned with the SRB´s. The participation in internal
resolution teams by small hosts has resulted in increased group information flows for them. However, in
the resolution colleges established by the SRB as the home authority, NRAs have only observer status
and cannot vote in the decision-making processes as this power is centralized at the SRB.

The SRB is also the authority in charge of conducting group resolvability assessments, setting MREL
requirements, and identifying and removing obstacles to resolution. The SRB is thus the authority that
decides on the distribution of MREL requirements into the different legal entities that make up the
Eurozone group both at the consolidated and individual level.

43
See 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 SRM and a SRF and amending Single Resolution Mechanism
Regulation (SRMR) No 1093/2010.
44
For the specific distribution of responsibilities between SRB and NRAs, see article 7 of SRMR.
45
The cooperation obligations within the SRM are outlined by article 31 of SRMR.
46
See article 6(3) of SRMR. According to this article “when making decisions or taking action which may have an impact in more than one Member State,
and in particular when taking decisions concerning groups established in two or more Member States, due consideration shall be given to the resolution
objectives referred to in article 14 and all of the following factors:
a)
The interests of the Member States where a group operates and in particular the impact of any decision or action or inaction on the financial
stability, fiscal resources, the economy, the financing arrangements, the deposit guarantee scheme or the investor compensation scheme of
any of those Member States and on the Fund;
b)
The objective of balancing the interests of the various Member States involved and of avoiding unfairly prejudicing or unfairly protecting the
interests of a Member State;
c)
The need to minimize a negative impact for any part of a group of which an entity referred to in article 2, which is subject to resolution,
is a member.”
47
Like, for example, the setting of MREL or the taking of actual resolution schemes by simple majority without the NRAs concerned.
48
In the first and so far, only resolution case decided by the SRB on the failure of Banco Popular the decision making process proofed efficient, though it
has to be noted that this particular case did not involve (public) burden sharing i.e. the use of the SRF nor cross border issues.
49
Notwithstanding that, as can be expected, the NRA of the home country will be the most involved in the process.

29
Once a bank is considered FOLTF by the supervisor, the SRB will decide if the public interest test is
fulfilled before placing it in resolution. Then, it decides on the resolution scheme to be applied to the
resolution group, also selecting the specific resolution tools to be applied to the resolution entity under
its authority (limited by Commission and Council veto rights)50.

These tools and powers stipulated in the scheme will be exercised by the NRA, according to a model
of centralized decision-making and decentralized execution. The SRB also decides on the use of the
SRF, provided that the resolution scheme complies with the minimum requirements for tapping it for
capital restoration purposes (i.e. losses have been absorbed by shareholders and creditors up to 8% of
total liabilities and own funds). Nonetheless, there are limits to what the SRB can agree; in particular, it
cannot make decisions or take actions that either require Member States to provide extraordinary pu-
blic financial support or impinge on the budgetary sovereignty and fiscal responsibilities or the Member
States.51 This is aligned with the overarching principle of the BRRD as well as international standards
to avoid public bail-outs. Still, the lack of burden sharing arrangements and the limitations to use the
SRF might become an obstacle to taking timely and effective resolution decisions in case of contagion.

One of the outstanding issues in the design of bank resolution in the Eurozone is how to ensure suf-
ficient resolution funding including a possible public backstop. The current legal and institutional fra-
mework has the proper mechanisms to ensure that losses can be absorbed, and fresh capital can be
provided by shareholders, creditors or the SRF. However, if a bank under resolution becomes illiquid, it is
not clear where the funds will come from, which presents a fundamental risk to bank resolution in the
Eurozone. Although funding could come from different sources (deposit guarantee scheme, the SRF, or
national central banks through emergency liquidity assistance) each of these comes with drawbacks,
in particular when funding for entities from different Member States is required.

5.1.3. Implications of the new framework for “eurozone small hosts”

The current model of bank supervision and resolution in the Eurozone has benefits for small hosts, as
the integration into the SSM/SRB has resulted in harmonized and more comprehensive supervisory
and resolution methodologies52.

One main goals of the Eurozone is to foster integration between Eurozone banks, thus creating a truly
single banking market, and to encourage the creation of truly Eurozone banks. Although not explicitly
expressed, this goal also influences supervision and resolution planning which has some consequences
for the small host authorities within the Eurozone. In particular:

• More centralized supervision is more biased in favor of the consolidated level and may favor branches
against subsidiaries. Many supervisory processes are performed at the consolidated level (SREP, stress
testing, etc.); and while they also cover Eurozone subsidiaries, the attention paid to the sub-consolida-
ted or solo level has decreased significantly. This may also influence the possibility of granting waivers to
cross-border Eurozone subsidiaries (currently available for liquidity but not for capital) and, more import-
antly, the conversion of subsidiaries into branches across Eurozone countries.

• Host authorities of eurozone subsidiaries of parent banks that fall under the SRBs remit have no longer
legal responsibility on resolution planning nor the taking of resolution decisions as this is centralized under
the SRM. Resolution planning also includes decisions on the quantity and quality of (internal) MREL. If no
agreement can be found between the authorities concerned, it is finally up to the 5 permanent Members of
the Executive Session of the Board to decide on a resolution plan under the SRMR and to strike a balance
between different national interests within the eurozone authorities including on entity individual MREL
setting and the preparation of a resolution scheme (for the Commission) at parent and subsidiary level.53
In most cases, resolution action is envisaged at parent level as most eurozone banks are following an SPE
strategy within the Eurozone.

50
For details, see: Lintner, P. (2017) De/centralized Decision Making Under the European Resolution Framework:
Does Meroni Hamper the Creation of a European Resolution Authority?
51
See article 6(6) of the SRMR.
52
Note that entity individual MREL is so far not set by the SRB and the framework in the Eurozone for internal MREL still needs to be developed,
especially in light of the upcoming new EU regulations introducing the concept of the concept of ‘resolution entity’ and ‘resolution group’.
53
See Article 54 SRMR.

30
Requiring banks to structure themselves to comply with and be resolvable under a SPE at Eurozone
level is based on banks’ business structures, but could also contribute to changes in those business
structures, if necessary. It may encourage further operational and financial integration in groups. This
may translate to fewer restrictions on intragroup financial flows (subject to prudential rules, e.g. large
exposures, and supervisory approval), in the choice of the group´s operating and IT model, etc. And, in
the medium to long term, it may contribute to the conversion of subsidiaries into branches.

5.1.4. Challenges and potential reactions by “eurozone small hosts”

Despite the benefits of the Banking Union, small Eurozone hosts may still perceive that their concerns
are not being dealt with properly. For example, they may consider that the liquidity or capital levels
of the subsidiary are unsatisfactory, that the liquidity and funding model of the group is not in the
interest of the subsidiary, that the dividend policy of the subsidiary is too aggressive, or that the crisis
management capacities of the SSM as supervisor are not sufficiently well-developed. In these situa-
tions, they still might want to rely on their domestic powers to force banks to increase the capital, and
liquidity, and to lower their risk profiles.

A potential future concern for Eurozone small hosts is that they may feel powerless to oppose the
granting of cross-border liquidity and, when legally possible, capital waivers for their subsidiaries, since
these authorization powers are, or may in future be with the SSM. They may feel equally powerless
when the banking subsidiaries that received a waiver are converted into branches. And in resolution,
when the SRB decides that the Eurozone banking subsidiary is part of the same resolution group/
point of entry, this may implicitly allow for free flows of capital and liquidity and thus the integration
of different legal entities within the euro area.

While one goal of banking union is to facilitate more integrated banking groups and markets, small host
country supervisors can be concerned that this goal is achieved prematurely as significant issues re-
main unresolved, such as the setting up of the European Deposit Insurance Scheme (EDIS), emergency
liquidity assistance (ELA) provision by national central banks or public backstops in resolution funding.
Small Eurozone hosts however still retain certain competences and powers. In particular:

• The macroprudential authorities of Eurozone Member States (“Designated Authority”) identify systemic
institutions and set other systemic important institutions buffers. Some small host authorities have occa-
sionally required higher capital buffers than other supervisors.

• Small Eurozone hosts may also adopt measures to address macroprudential risks at the level of that Mem-
ber State, which may include gold-plating the capital, liquidity, or large exposures requirements, among
others.54

• Host regulation may still require subsidiaries to draft individual recovery plans, even when a group recovery
plan including the subsidiary has been drafted.55 However, this is not applicable for branches or subsidiaries
with a waiver.

5.2. Non-Eurozone EU small hosts

Contrary to Eurozone hosts, other EU Member States retain their full competences as individual or
sub-consolidated host supervisors and resolution authorities. This means that all the supervisory pro-
cesses discussed in section 4.1. and applicable to the subsidiary are planned and executed by the host
authority. Similarly, in terms of resolution, non-Eurozone EU host resolution authorities are responsible
for the resolution of subsidiaries.

54
See article 458 of CRR.
55
The individual recovery plan may be only required in case the local regulations demand it. Otherwise, the requirement of an individual recovery plan
would be subject to the final decision of the ECB, since it constitutes a joint decision, where the ECB has taken the competences from the national su-
pervisory authority.

31
5.2.1. Overview: Banking supervision in the EU

The European regulatory framework provides for compulsory cooperation between home and host
supervisors within the EU.56 Small EU hosts are members of the supervisory college57 of the banking
group that the consolidating supervisor should establish.58 Participation in these structures is essential
as they are the legal forum where joint decisions are taken and coordination and cooperation between
supervisors takes place. Joint decisions for instance on capital, liquidity, internal model validation, both
on a consolidated and on an individual basis, are taken within the colleges.59

The involvement in these processes allows the small hosts to have a more comprehensive view of the
capital and liquidity available at group level and how they are distributed across the different units
of the group. The sharing of information by the home supervisor in the colleges is also important
for another reason: the access to recovery plans, to ICAAPs, to ILAAPs, to SREP reports, and to risk
assessments allows small EU hosts outside the Eurozone to gain insights into these SSM supervisory
processes.

5.2.2. Overview: Bank resolution in the EU

The resolution authorities of non-Eurozone EU subsidiaries or significant branches (depending on their

relevance in the host country60) are members of the resolution colleges, organized by the group-level
resolution authority (usually, the SRB). As college members, they are granted access to group-level
information, which is valuable for resolution planning purposes, and, at the same time, they participate
in the joint decisions within the college, including those with regard to resolution plans, resolvability
assessments, and MREL requirements.61 However, the development of resolution plans, resolvability as-
sessment, and MREL requirements is at an early stage for many EU countries, even though significant
progress is expected in the coming years.

The typical small host concern of insufficient coverage of local subsidiaries or significant branches in
the group resolution plan has been addressed by how the EU regulations define material legal enti-
ties.62 One of the criteria used for determining the materiality of legal entities and branches (not to
be confused by the “significant branch criteria in supervision) is “the importance of the entity to the
financial system of at least one of the Member States in which they have their registered offices or
operate”.63 This definition ensures that subsidiaries and branches are covered in the group resolution
plan and resolvability assessment, even when they do not represent a significant share of the assets
of the group.64

56
See articles 114 and 117 of CRD IV.
57
The consolidated supervisor should establish supervisory colleges for banking groups.
58
To facilitate and establish effective supervision, the consolidated supervisor and the other competent authorities shall have written coordination and
cooperation agreements (WCCA) in place, in according with article 115 of the CRD IV.
59
This may not be the case if the group has set up a branch instead of a subsidiary in the host country. In this particular case, the supervisor may only be
part of all these processes and receive information regularly if the branch is considered as “significant”, according to the definition and decision-making
process outlined by the CRD IV.
60
For the significance criteria see Table 7. Other authorities from the host country will also be members of the resolution college: (i) the competent
authority (host supervisor), (ii) the Minister of Finance, (iii) and the authority in charge of the deposit guarantee scheme.
61
Until now there hasn´t been required to adopt a joint decision on a group resolution strategy for EU cross-border banks.
62
See article 7(2) of the Commission Delegated Regulation (EU) 2016/1075 on the content of the recovery plan.
63
ibid
64
Like supervisory colleges, the inclusion of a subsidiary as a material legal entity in resolution or recovery plan has no effect on the consideration of a
subsidiary as significant (in terms of its importance in the host country) to provide the basis for access to the College under Art 88 BRRD/51 CRD.

32
Even though it is not explicitly stated in the in the EU legal framework, authorities must make a choice
between an SPE, an MPE, or a mixed approach for cross-border banking groups. They need to explain
in the resolution plan to which legal entity resolution tools would be applied. For subsidiaries in the EU,
but outside the Eurozone, the home (usually the SRB) and host authorities must find a common un-
derstanding for determining the resolution strategy. If disagreements arise, EBA mediation is available
(see Table 9). Joint decisions between Group Level Resolution Authority (i.e. the SRB in most cases) and
NRA are taken on (external) MREL at solo level for subsidiaries, and where there is disagreement it is
the “host” NRA to decide, with the possibility for the Group Level Resolution Authority (GLRA) to recur
to EBA binding mediation.65 Vice versa the GLRA decides on consolidated MREL /at parent level open
to EBA binding mediation by the host authority.

If the SRB / other EU home authorities and the local resolution authority jointly decide to include the
EU banking subsidiary into the same resolution group as the parent (i.e. the resolution entity in this
case), host supervisory authorities have incentives to give up ring fencing strategies. Still, also under an
SPE approach, in a worst-case scenario, (prepositioned) internal MREL could not be sufficient and the
parent not able or willing to provide further support, triggering the host authority to take independent
resolution action being left with limited options/additional difficulties in light of the interconnected-
ness underlying an SPE approach.

Under an MPE strategy, the host resolution authority might still prefer a situation where it demands
that the MREL is provided by the parent entity. However, for home authorities, this may be an obstacle
as it may entail a risk of contagion. This is especially the case in small host countries with shallow ca-
pital markets where issuing loss-absorbing debt at sustainable prices is not always possible. Regardless
of the underlying approach, host authorities have incentives to ensure high amounts and high quality of
ex-ante pre-positioned internal MREL instruments are available in case of need.66, 67 The SRB and other
group-level resolution authorities can aim to restrict the risk of contagion (i.e. financial interdepen-
dencies including MREL) between resolution groups. Though host authorities are competent to define
MREL for the subsidiary, the parent’s supervisory or resolution authority could potentially prevent the
parent from down-streaming capital or debt. In any case, decisions on MREL for both individual entity
and consolidated level as well as general decisions on banks resolvability are subject to EBA binding
mediation in case of disagreement between “home” and “host” authorities.68

If the subsidiary is assessed as a separate point of entry (MPE), the host resolution authority may
identify obstacles stemming from the interconnections of the subsidiary to the rest of the banking
group (IT connections, upstream financial exposures, centralized liquidity management by the parent

65
Unless the host NRA has calibrated the external MREL of the subsidiary at a level within 1% of the consolidated MREL (safe-harbour clause). Also the
BRRD2 Commission proposal foresees that a joint decision on the resolution strategy is taken by the home resolution authority and the NRAs of the
non-resolution entities which form part of the same resolution group. If no agreement can be found between the group-level resolution authority and
the NRAs concerned, the decision on external MREL is to be taken by the group-level resolution authority (taking into account the “host” opinion) and the
host NRAs can refer the decision to the EBA for its binding mediation. The decision on internal MREL would be taken by the host NRAs.
66
According to this model, the parent company may be effectively guaranteeing the balance sheet of the subsidiary, ensuring a recapitalization if losses
arise and isolating the subsidiary´s creditors from these losses. The group level resolution authority may be, up to a point, interested in this model since
these subsidiaries are usually small and if one of them fails the parent company would be, for different reasons (mainly reputational, but also financial),
reluctant to let it fail. However, it may oppose any measure taken by the host resolution authority demanding a higher degree of autonomy of the sub-
sidiary with regard to the parent company.
67
Legal MREL requirements in such cases could be complied with, in form of write-down or conversion of capital (WDCC) of eligible capital instruments
being triggered outside resolution at the point of non-viability (PONV)/FOLTF by the host authority (the expectation would be for the parent to recapitalize
the bank without a conversion action by the resolution authority). This case may imply fewer restrictions on the interconnections between the subsidiary
and the rest of the group, and the host may have additional flexibility to trigger WDCC/bail-in of the pre-positioned amount early enough. However, the
unlikely case of lack of support from its (solvent) parent company (above the prepositioned amount) may result in separate resolution of the subsidiary,
making it difficult to separate the subsidiary from the group. This situation is highly unlikely due to the high reputational risk incurred by the parent com-
pany but is not impossible, and can arise in situations where the failure of the subsidiary is mainly explained by political events in the host country or when
the parent company does not have the financial capacity to support the subsidiary or several subsidiaries at the same time. See Box 6.
68
The EBA published its first binding mediation decisions between two resolution authorities, the Single Resolution Board (SRB) and the National Bank
of Romania (NBR) in June 2017.

33
company, centralized treasury, etc.) as it may have to enforce the separation of the subsidiary in case
of resolution. In an MPE strategy, the services ensured by one resolution group to the other resolution
groups should rather be the exception than the rule. However, some limited services could continue to
be provided if there is no risk of disruption and contagion. The group-level resolution authority may
find opposing these (ring-fencing) measures difficult if the subsidiary constitutes a separate point of
entry. The provision of internal MREL (by the parent company) under an MPE approach to a certain
extent dilutes the theoretical concepts underlying the difference to SPE and undermines the subsidia-
ries financial self-sustainability. However, if internal MREL is prepositioned in the same way as external
MREL, the mere fact of parental support does not per se negatively affect a subsidiary’s resolvability
under MPE. However, this may be an obstacle for the parent in case it entails risk of contagion.

If resolution authorities do not agree with the group resolution plan, the resolvability assessment, the
decision to remove obstacles to resolution, or the type or amount of MREL decision, responsibility falls
back to the host authority to decide for the individual entity and to the home for the consolidated level.
If the home authority (SRB) does not agree with the host authorities’ decision or vice versa, it can refer
the decision to EBA´s binding mediation, which is open to the fiscal impingement clause in some cases
(see Table 8).

5.2.3. Implications of the new framework for non-Eurozone EU hosts

A common concern of non-Eurozone EU small hosts is that insufficient attention is paid by the SSM/
SRB to the systemically important subsidiaries in the host country if they do not represent a signi-
ficant share of the assets, liabilities, revenues, or capital of the banking group. As illustrated by the
survey responses reported in Box 4, typical examples are poor communication with host supervisory
authority and poor follow up on proposed joint activities by the home authorities

Box 4: Survey of host country supervisors’ views on cooperation with home country supervisors

Survey

This box presents the results of a survey among host country supervisors in Central, Eastern and Southeastern
Europe (both EU and non-EU members).69 Supervisory authorities were asked to rate the effectiveness of the co-
operation with home country supervisors for systemically important foreign financial institutions. In all cases, for-
eign-owned banks are present in the form of subsidiaries. Austria, France, Germany, Hungary, Italy, but also Russia,
are among the home countries of the most systematically important foreign-owned banks in these countries.
Only 50% of the supervisory authorities are part of the relevant supervisory colleges and 40% of the resolution
college. In the case of supervisory colleges, in about half the cases, the host supervisor is a member of the core
college and in about half the cases only of the general college.

Asked about the effectiveness of supervisory cooperation, six host supervisors answer “average”, one “effective”
and another “very effective”. Two thirds of the host supervisors consider the information they receive in the super-
visory colleges as sufficient and useful to undertake effective supervision in their country.

There are information asymmetries between home and host supervisors. Few respond positively when asked
whether home supervisors are willing to provide host supervisors with information they need or want. There are
only a few cases where the home country supervisor performed an inspection of the subsidiary in the host country
and informed the host supervisor. When it comes to joint supervision missions, in half of cases this is rare and in
the other half it has yet to happen. In half of all cases, the host country supervisor could observe an inspection
by the home country supervisor, but less than half stated that they receive the inspection report from the home
supervisor. In most cases, host country supervisors often or always meet with home country supervisor, yet this
is not necessarily the case for EU host country supervisors.

69
Respondents were from Albania, Belarus, Bulgaria, Kosovo, North Macedonia, Moldova, Montenegro, and Romania.

34
 When the host supervisor undertakes an inspection of a systemically important foreign subsidiary, half of the
host supervisors communicate with the home supervisors in written form, none of them meets with the home
supervisor. In only half of the cases, the host supervisors share the inspection report with the home supervisors.
Half of the host supervisors invite the home supervisor to either observe or participate in an inspection. When
addressing material supervisory matters regarding foreign-owned domestic systemically important banks, howe-
ver, more than half of the host supervisors never or rarely communicate or consult with the home supervisors and
none decides jointly on enforcement actions. This reflects the national mandate of supervisors.

When it comes to ring-fencing, most of the host supervisors state that the current set up does not allow for
restructuring without the help of the parent group. According to the host supervisors, the parent bank would
support its subsidiary with capital, liquidity and technical support and (to a lesser extent) statutory funds and
subordinated loans. Foreign-owned domestically important banks are (so far) in none of the countries required to
hold Loss Absorbing Capacity (LAC) which could be converted into equity in case of insolvency or bail-in. This is
surprising since the BRRD was issued in 2015. And there is no practical experience with a foreign bank failure in
the eight surveyed countries.

In summary, there is still only limited participation in supervisory and resolution colleges and limited information
exchange, especially down-stream. On the other hand, there are not many options for host supervisors to ringfen-
ce their systemically important foreign bank subsidiaries.

In the following paragraphs, we discuss some recurrent issues.

a) Potential inadequacy of the branch regime for cross-border banking. The lack of attention and com-
munication by consolidating authorities is even more important when the parent company performs its
activities in the host country through a branch rather than a subsidiary. While these concerns have been
partly addressed by amendments to the prudential regime in the EU70, local authorities may still perceive
the attention paid by the home supervisor to the host country as insufficient. In addition, the development
of cross-border commercial banking activities through branches may place significant stress on the finan-
cial safety net of the home country. Deposits raised in the host country will be covered by the home deposit
insurance scheme, and creditors and depositors of the parent entity in the home country will be exposed to
the losses of the branch.

b) Dealing with powerful central institutions: The EU small hosts must deal with the SSM or the SRB rather
than national authorities. On the plus side, in countries with a significant share of Eurozone parents like
Croatia, Romania, Czech Republic, and Hungary, the local authorities now have a single counterpart for all
their subsidiaries, which should ensure more consistent approaches.

For some authorities that have, implicitly or explicitly, stated their intention to join the Eurozone and the
SSM/SRB mechanisms (Bulgaria, Croatia), differing with the Eurozone institutions comes with political
trade-offs. For example, there may be reluctance to call for EBA mediation. Box 5 summarizes some of the
arguments in favor and against small EU non-Eurozone host countries joining the banking union.

Box 5: Why would EU small hosts join the European banking union?

WHY WOULD EU SMALL HOSTS JOIN THE EUROPEAN BANKING UNION?

Some non-Eurozone EU host countries have considered joining the European banking union prior to being part
of the Eurozone. Romania, Bulgaria, and Croatia are considering applying for banking union membership. Some
other countries (Poland, Czech Republic, and Hungary) remain opposed to joining the banking union, which (consi-
dering the high shares of foreign ownership in their banking sectors) may seem inconsistent at first.

70
These include introduction of the prudential regime for significant branches, enhancement of the joint decision regimes, clarification of the roles of
the supervisory colleges, and mandatory coverage of the material legal entities in the group recovery plan based also on the systemic relevance for the
host country.

35
 There are several arguments in favor of joining the banking union, including:

• Given the high share of ownership of Eurozone banking groups, contagion risks from the Eurozone may
be elevated. Becoming part of the SSM and SRB can help to influence policy-making and participation
in supervisory and resolution issues.

• The SRB may adopt a SPE approach for these subsidiaries, resulting in a „de facto“ guarantee of the
subsidiaries’ external creditors by the parent company.

Nevertheless, there are also arguments against joining, such as:

• Since all the supervisory and resolution powers are (semi)centralized at the Eurozone authorities, there is
a high risk of the “branchization” of their financial systems, especially when the banking groups are set
in the Eurozone. The decreasing number of barriers (and having the same supervisory authority removes
a significant one) to cross-border banking may contribute, as explained in previous sections, to the use
of branches to conduct banking activities across the euro area.

• Host supervisors would largely lose their ability to impose higher institution-specific capital and liquidity
requirements, among others, under the SREP and Pillar 2 rules.

• The host countries, considering the relatively small size of their financial systems when compared to
others in the Eurozone (Germany, Italy, France, or Spain) may fear that their banks would not receive the
required attention by the SSM. This may be the case particularly for domestic-owned banks and ban-
king groups (e.g. PKO in Poland or OTP Bank in Hungary) that are under intense supervision by national
supervisors but would be considered medium-size or small banking groups in the Eurozone.

• Poland and Hungary have targeted an increase in the domestic ownership levels of their banking sys-
tems in so called “banking nationalism” (Méro and Piroska, 2016). The transfer of banking supervision
powers to the SSM may interfere with these objectives and the remaining domestic banks may be taken
over by Eurozone banks since the authorization powers for mergers and acquisitions (M&A) transactions
are transferred to the ECB.

• Many EU non-Eurozone countries, such as Czech Republic and Poland, have strong supervision practices
and hence were not so affected by the negative fall out of the global financial crisis.

• Their small size relative to other participating countries may reduce their leverage and influencing pow-
ers in the ECB/SRB.

• The current weaknesses of the Eurozone, mainly the lack of a European Deposit Insurance Scheme and
the lack of a liquidity backstop for banks under resolution, may make it less attractive to join.

c) Lack of representation of host countries interests in EU supervisory colleges: As part of their participa-
tion as members in the SSM supervisory colleges, EU small hosts typically voice the following concerns:

• In the annual SEP the “combined component” is usually the simple consolidation of the individual SEPs of
the supervisors of the college rather than a truly joint work program. For example, it is unusual to include
joint inspection/examination programs that involve officers from both the home and the host supervisor.

• Crisis preparedness/contingency planning is still in a premature phase. Correspondingly, discussions bet-

ween home and host supervision over intra-group financial support agreements have yet to take place.

• In some cases, discussion on the capital and liquidity levels of the banking group as part of the joint deci-
sion has not been comprehensive or some small hosts have not been invited to the college at all.71 This is
especially important for group liquidity when the liquidity and funding model of the subsidiary is linked to
the group (i.e. a subsidiary dependent on the funding of the parent company, a subsidiary that up-streams
the excess liquidity to the parent company, a centralized model for managing collateral, etc.).

71
See EBA Report on the Functioning of the Supervisory Colleges in 2017, published on the 16th March 2018.

36
 d) EU resolution colleges: The small EU hosts of both subsidiaries and significant branches have the status
of members of the college, so they can attend the college meetings, receive the information shared within
the structures of the resolution college, and participate in the joint decisions. Resolution colleges, alongside
resolution authorities, are novel in the EU. So far, the participation of the non-Eurozone EU hosts has been
positive in terms of information sharing between authorities and exchanging views on resolution planning,
resolvability assessments and MREL.

The participation of the EBA in the colleges and the number of college meetings by the SRB have con-
tributed to harmonization. The role of the colleges in resolution cases is still to be tested as no relevant
cross-border bank failure has taken place since bail-in entered into force.72 In addition, there is room for
improvement regarding the content and the dynamics of the resolution colleges organized by the SRB
acknowledging that cooperation is depending also on host authorities willingness to cooperate and the
building up of capacities. Each part of the resolution planning process can be improved in terms of, for
example, information shared, technical developments, and policy issues.

e) Different views on resolution strategies: As discussed above, the home and host resolution authorities
may have a different understanding of what the main implications of being considered a separate point of
entry are in terms of resolvability and the interconnections between the subsidiary and the other parts of
the group, open to EBA mediation. For example, host resolution authorities may consider that a separate
point of entry implies that the subsidiary should have independent funding and treasury functions, while
the home resolution authority and other host authorities may consider it appropriate to adopt a propor-
tionate approach with measures strictly adapted to the mitigation of the risk of contagion and to ensure
operational continuity in case of resolution with MPE.

These differences in view are increasingly important, and tensions between some authorities are mounting,
as shown by public statements by certain resolution authorities. For example, the Austrian Central Bank
(OeNB) Vice Governor recently stated that resolution authorities must cope with regional failures them-
selves; and that if the local resolution authorities insist that the MREL should be provided by the parent
banks, then there would be a strong incentive for converting those subsidiaries into branches (reflecting
the idea that MPE and provision of internal MREL between two resolution entities of the MPE are in prin-
ciple conceptually incompatible). The host perspective was publicly raised by the Governor of the Central
Bank of Romania, declaring that at least until the conditions for MPE strategies are met, parent banks
should continue to provide support to their subsidiaries. Fully fledged MPE strategies might not be feasible
in many small host countries considering the shallow capital markets and related difficulties in raising
external MREL.

f) Excessive levels of NPLs and material sovereign concentrations through their subsidiaries are currently
two of the main common risks in the EU banking system. In this context, the SSM may decide that a bank
should reduce its risk profile through the reduction of NPLs and the sovereign exposures. While in principle
the SSM may enforce these actions at the consolidated level, these requirements shall be carefully consi-
dered to avoid negative spill-over effects in the host financial systems. Even though a joint decision does
not exist for these measures, the consolidating supervisor should discuss bilaterally with the affected host
supervisors and, when needed, within the supervisory college.

Box 6: How mergers and acquisitions in the Eurozone may affect small hosts
in the EU and other non-EU European countries

HOW M&A IN THE EUROZONE MAY AFFECT SMALL HOSTS IN THE EU AND OTHER NON-EU EUROPEAN COUNTRIES

As explained in Section 2, a handful of Eurozone banking groups (Unicredit, Intesa SanPaolo, Raiffeisen Bank
International, Erste Bank Group, Société Generale, etc.) have relevant market shares in many small host countries
across Europe. Some of these banks were the result of several cross-border banking mergers. For example, Unicre-
dit was created following the acquisition by the Italian Group Unicredit (itself the result of merging a large number
of Italian former savings banks) of several banks in other European countries.

72
Spanish-based Banco Popular, resolved in 2017 by the SRB, had one banking subsidiary in Portugal and another one in the US. However, no resolution
college had been set up at the time of the resolution.

37
 In 2005, Unicredit took over the German bank Hypovereinsbank (HVB), which alongside a relevant market share in
retail banking in Southern Germany, had many subsidiaries in the then fast-growing CESEE region. In 2000, HVB
took a controlling interest in Bank Austria, a systemic Austrian bank with many subsidiaries across the region.
These stakes, combined with the ones owned by Unicredit, created a banking group with very high market shares
across the CESEE region. For example, in Croatia Unicredit had 25.7% market share and HVB 8.3% (34% combined
market share) and in Bulgaria (market share of 24.3%, combining 14.3% of Unicredit and 10% of HVB).

One of the goals of the European banking union (and particularly of the SSM) is to foster cross-border banking
M&A within the Eurozone, creating truly pan-European players that provide banking services across the region.
However, in view of the many obstacles to these goals, not least the unfinished nature of the banking union itself,
few cross-border transactions have taken place between banks since the European banking union was created
(with some exceptions, like the takeover of Portuguese BPI by Spanish CaixaBank).

Some of the most relevant Western European banking groups for the region may further merge their operations,
indirectly combining their already high market shares in many small host countries. This outcome would be far
from ideal for the small host supervisors, reducing the number of players, creating even more systemic institu-
tions, and having a potential negative impact on local customers that might see their options for banking services
curtailed.

If this were to occur, consolidating supervisors (especially, the SSM) should pay special attention to the concerns
of the host countries. They should engage with host authorities (EU and non-EU) at an early stage of the dis-
cussions and try to devise solutions to address the issues faced by small hosts. A possibility may be forcing the
divestments of part of the business by the combined banking group. But this solution also has relevant challenges,
especially in countries where there are not many relevant players interested in entering the banking market.

5.2.4. Potential reactions by non-Eurozone EU small hosts

In cases where EU small hosts perceive that their interests are not sufficiently protected by partici-
pation in supervisory colleges and coordination and cooperation by home authorities, they may adopt
additional measures to ensure the subsidiary or sub-group under their supervision is well-protected
from a potential crisis in the parent company. In particular:

• Like Eurozone hosts, the designated authority of the Member State is responsible for identifying systemi-
cally important domestic institutions and for setting the capital buffers for them. Some EU small hosts
have included relatively high levels of capital buffers compared to other Member States, and some have
used other regulatory instruments to increase the capital levels of their financial system. For example, in
the case of Intesa San Paolo, the Italian authority has determined a 0.75% D-SIB buffer on a consolidated
basis, while the Croatian authorities have placed a 2% D-SIB buffer on the local subsidiary. That said, host
authorities are facing a balancing act with their banks, as they have no defense against the threat of con-
version of subsidiaries into branches.

• EU hosts retain the ability to impose supervisory measures and to require institution-specific Pillar 2 re-
quirements in terms of capital, liquidity, and other supervisory measures, even for cross-border banking
groups, in case they cannot reach joint decisions on these requirements within the supervisory college.73

• Some EU hosts may seek to ensure more independence from the parent companies. This may include
requiring a higher number of independent members of the board of directors/supervisory board of the
subsidiary, more independent IT and operations structure, or a self-sufficient funding structure. Some su-
pervisors have also required the listing of the subsidiary on the local stock exchange.74

73
Experience on the operation of supervisory colleges shows that the consolidated supervisory authority is unlikely to challenge the capital and/or liquid-
ity levels required of the bank on an individual and consolidated basis while host supervisors are also unlikely to challenge the consolidated capital and
liquidity requirements of the banking group, as proposed by the home supervisory authority.
74
See the case of Poland. Through the listing of the subsidiary, the autonomy and independence of the subsidiary with regard to the rest of the group
may increase naturally due to corporate governance and disclosure requirements that are applicable to listed companies. A situation where some for-
eign-owned banks also have minority shareholders and are listed in the stock market also happens, like Turkey.

38
 • When parent companies are suspected to be distressed, some EU host supervisory authorities have en-
forced a stricter ring-fencing regime.75, 76

• Many EU hosts still prefer that their subsidiaries draft their own individual or sub-consolidated recovery
plans.77 Some host authorities require the mandatory inclusion of potential support from the parent com-
pany in the individual recovery plan of the subsidiary.

• Host authorities may require mediation (binding or non-binding) by the EBA, especially with respect to
joint decisions. EBA´s non-binding mediation has included joint decisions, liquidity, recovery planning, and
supervisory measures.78 Home authorities have also demanded the non-binding mediation by the EBA for
ring-fencing measures adopted by EU hosts. EBA´s binding mediation is seen only as a last resort and has
been recently called on by some EU resolution authorities when agreeing on resolution planning, resolvabi-
lity assessments and MREL requirements.

However, these powers are generally not available to EU host authorities of branches. As a result, many
EU host supervisors remain opposed to branching, especially if the foreign bank wants to raise retail
deposits in the host country.79 This is despite CRD IV’s requirements for home supervisors to share
information80 and for a mechanism to determine a branch as significant (article 51) resulting in more
information being shared with the host supervisor and the host’s membership in resolution colleges81.
Moreover, a branch can be designated as a “significant-plus branch” and made subject to strengthened
supervisory procedures and even more intense home-host coordination.

5.3. Non-EU host countries (EU candidates and non EU candidates)

The following discussion focuses specifically on small non-EU host countries in CESEE. These countries
can be divided into two different groups EU candidate (including potential candidates) and non-candi-
date countries. Naturally, the former has significantly stronger incentives to align with EU regulations
and supervisory practices. However, it is important to stress that the candidate status only affects the
host, not the home supervisor, as from the perspective of the EU both candidate and non-candidate
countries are considered third country.

As a result, for participation in EU supervisory and resolution colleges, all non-EU countries are third
countries, regardless of their candidate status. At the request of a third country authority, the EU
home authority may invite it as an observer to the supervisory and resolution colleges.82 Such partici-
pation offers benefits for host country supervisory and resolution authorities (Box 7). In this respect,
both EU regulations and EBA and other EU bodies encourage EU home supervisors and resolution
authorities to closely cooperate with non-EU European host supervisors83, on a bilateral basis (informa-
tion sharing) but especially within supervisory colleges.

75
These regimes may entail different sets of measures such as restricted and/or subject to authorization upstream transactions between the subsidiary
and the parent company, involving the payments of dividends, intra-group lending agreements, derivative contracts.
76
More clarity on crisis preparedness and contingency planning within the supervisory colleges may help to avoid these measures if a crisis situation of
the parent company or the group as a whole arises.
77
Notwithstanding that most group recovery plans already include the information (on the recovery indicators and recovery measures, among others)
related to the subsidiary or subgroup under supervision of the host supervisory authority.
78
See 2016 EBA Annual Report
79
The migration to a single deposit guarantee scheme would not apply in this case, so deposits raised through branches in host countries would be
insured by the deposit guarantee scheme of the home country, creating risks and unbalances in the process.
80
See Article 50 of CRD IV.
81
Article 51(3) even includes the obligation of set up resolution colleges made up only by the consolidating supervisor and the host supervisors of signif-
icant branches, when there are no other members of the college according to article 116 of the CRD IV.
82
See Articles 88 (3) BRRD and 51 of CRD IV.
83
See, for example, the priorities for 2018 EU supervisory colleges as contained in the EBA Report of the Functioning of Supervisory Colleges in 2017. It
contains as main tasks for 2018 supervisory colleges “considering expanding the list of authorities with observer status in the light of the outcome of EBA´s
work on equivalence provisions of non-EU supervisory authorities, following the process envisaged in the Level 1 and Level 2 provisions”.

39
Box 7: Benefits for non-EU hosts from participation in resolution and supervisory colleges

BENEFITS FOR NON-EU HOSTS FROM PARTICIPATION IN RESOLUTION AND SUPERVISORY COLLEGES

For non-EU European supervisors, it is essential to participate in discussions within the supervisory and re-
solution colleges to mitigate the problems of asymmetric information between home and host authorities. Many
benefits can be identified for hosts from their participation in these colleges:

• First, within the colleges the information shared by the home supervisor or the group-level resolution
authority and the rest of the participants enables the non-EU hosts to improve their understanding
of the whole group and the position of the subsidiary in the group. For example, in the supervisory
colleges it is expected that the home supervisor shares (at least annually) the mapping of the group,
the supervisory examination program, the group risk assessment (including liquidity risk assessment)
according to SREP standards, and the group recovery plan. The consolidating supervisor may also share
other information, like ICAAP and ILAAP reports.

• Host supervisors are grappling with many new EU regulations and guidelines. The supervisory and
resolution colleges are a forum for sharing experience on implementation. For example, college discus-
sions on group risk assessments or group recovery plans expose practical expertise and knowledge on
how to apply these new and complex requirements. This is even more important when dealing with ins-
titution-specific capital and liquidity requirements. Even though Pillar 2 add-ons have been available for
supervisors for a long time, few non-EU supervisors have used them as they generally have higher Pillar
1 requirements. Discussions in the supervisory colleges are the ideal places for understanding different
supervisory practices, especially for Pillar 2 capital requirements and Pillar 2 guidance. The same applies
for supervisory stress tests, ICAAP and ILAAP assessments and the assessment of SREP elements like
the viability and sustainability of business model, and what to expect from risk appetite frameworks.

• Finally, supervisory and resolution colleges may contribute to the building of trust between autho-
rities, a precondition but no guarantee that cooperation will occur in a crisis. In this respect, while
the EU regulations require colleges to hold at least one physical meeting a year; experience84 shows that
most supervisory colleges meet physically at least twice a year.85

That said, before a third-country host supervisor can join the supervisory college with observer status
the following requirements apply:

• There is a regulated procedure to be followed before a non-EU authority can gain observer status.86 The
consolidating supervisor should first notify the other members of the college of its intention to invite the
non-EU host authority. To participate in the college, the other members must not oppose this proposal.

• Before becoming an observer, the consolidating supervisor needs to assess the equivalence of the con-
fidentiality and professional secrecy requirements applicable to the third country supervisory authority
(Table 11), a process usually facilitated by the EBA.87

• The non-EU host supervisory authority should sign the written cooperation and coordination agreement in
accordance with the applicable minimum requirements as outlined by the EU regulations.

Observer status does not always have the same scope and content since the home supervisor is
responsible for proposing the scope of observers’ rights to access information. Non-EU small host su-
pervisors can be classified into two categories when it comes to participation in supervisory colleges.

84
Experience on the operation of supervisory colleges shows that the consolidated supervisory authority is unlikely to challenge the capital and/or
liquidity levels required of the bank on an individual and consolidated basis while host supervisors are also unlikely to challenge the consolidated capital
and liquidity requirements of the banking group, as proposed by the home supervisory authority.
85
See the case of Poland. Through the listing of the subsidiary, the autonomy and independence of the subsidiary with regard to the rest of the group
may increase naturally due to corporate governance and disclosure requirements that are applicable to listed companies. A situation where some for-
eign-owned banks also have minority shareholders and are listed in the stock market also happens, like Turkey.
86
See Commission Implementing Regulation (EU) 2016/99 of 16 October 2015, laying down implementing technical standards with regard to determin-
ing the operational functioning of the colleges of supervisors according to Directive 2013/36/EU of the European Parliament and the Council.
87
The EBA has made its own assessments and has concluded that some non-EU European countries (Bosnia Herzegovina, North Macedonia, Montene-
gro, Kosovo, Serbia, and Turkey) have an equivalent regime in terms of confidentiality. In any case, the opinion on the equivalence of the confidentiality
provisions shall be made by the consolidating supervisor.

40
In the first category are the many supervisory colleges where third-country supervisors still do not
have observer status. In the other category, supervisors participate as observers in the general college
where they can access relevant group information and participate in the risk assessment of the group,
but do not participate in joint decisions. Non-EU hosts have been increasingly invited to supervisory
colleges organized by the SSM.88 For this purpose, the ECB has been signing cooperation agreements
with other authorities, and in addition it has assumed the bilateral agreements that national authori-
ties had signed in the past.

Table 11: EBA professional secrecy and confidentiality regime applicable to each third-country
supervisory authority
EBA professional secrecy and confidentiality regime applicable to each third-country supervisory authority

State / Agency Competent authority Overall

Date
assessment
Equivalent 11/09/2015
Albania Bank of Albania

Australian Prudential Regulation Authority Equivalent 11/07/2017

Australia
Reserve Bank of Australia
Equivalent
Banking Agency of Republika Srpska 02/04/2015
Bosnia-Herzegovina
Banking Agency of the Federation of BiH Equivalent
Brazil Central Bank of Brazil Equivalent 02/04/2015
Canada Office of the Super-intendent of Financial Institutions Equivalent 02/04/2015
China China Banking Regulatory Commission Equivalent 02/04/2015
Bailiwick of Guernsey Guernsey Financial Services Commission Equivalent 20/06/18
Hong Kong Hong Kong Monetary Authority Equivalent 11/07/2017
Bank of Japan
Japan Equivalent 11/07/2017
Japan Financial Services Agency
Korea The Bank of Korea Equivalent 20/06/18
Kosovo Central Bank of the Republic of Kosovo Equivalent 11/07/2017
Macedonia National Bank of the Republic of North Macedonia Equivalent 02/04/2015
Bank of Mexico Equivalent 02/04/2015
Mexico
National Banking and Securities Commission Equivalent 02/04/2015
Montenegro Central Bank of Montenegro Equivalent 02/04/2015
Serbia National Bank of Serbia Equivalent 02/04/2015
Singapore Monetary Authority of Singapore Equivalent 02/04/2015
Switzerland Swiss Financial Market Supervisory Authority FINMA Equivalent 02/04/2015
Turkey Banking Regulation and Supervisory Agency Equivalent 02/04/2015
The Superintendence of the Financial Services of the Central
Uruguay Equivalent 20/06/18
Bank of Uruguay
Federal Deposit Insurance Corporation Equivalent
USA Office of the Comptroller of the Currency Equivalent 02/04/2015
Board of Governors of the Federal Reserve System Equivalent

Equivalence assessments for supervision are no guarantee for being granted observer status in resolu-
tion colleges. While the BRRD strongly encourages cooperation and coordination with non-EU authori-
ties, the participation of non-EU countries in resolution colleges has been so far limited. The rather less
developed and efficient resolution frameworks in these countries must be considered though. Like the
process for supervisory colleges, third country Resolution Authorities may, at their request, be invited
to participate in the resolution colleges as “observers” upon the following conditions:

• Subject to equivalent confidentiality requirements (as assessed by the group level resolution
authority i.e. the SRB);

• Invitation after group level resolution authority agreement, following consultation with
the college members;

• Only when the third-country entity is a subsidiary, or a branch that is significant if it were located in the EU.

88
According to the 2017 SSM Annual Report, the ECB had, until December 2017, joined existent MoUs of Eurozone National Competent Authorities
with 9 EU non-Eurozone EU Member States and 45 with third-country supervisory authorities. Moreover, the ECB has signed MoUs with 25 supervisory
authorities since its inception

41
If the application is accepted by the College, the non-EU resolution authority may become an “observer”
and thus becomes eligible to receive and share information for resolution planning purposes. It will not
be able to participate in the joint decisions of the college and will usually not receive the group reso-
lution plan, but will receive presentations and background material like CMGs. In addition, and before
getting observer status, bilateral cooperation agreements are concluded.

5.3.1. Banking regulation and supervision in candidate countries

EU candidate host countries, and those that have publicly stated that they seek to join the EU, have
implemented or are in the process of implementing regulations and supervision practices based on the
EU regulatory and supervisory framework. For example, many have signed written cooperation and
coordination agreements and have also participated in supervisory colleges as observers. This process
is essential for these authorities for a variety of reasons, from political (convergence is required to ad-
vance in the EU accession talks) to prudential (see Box 8).89

For example, a major step forward has been the signing of the Memorandum of Cooperation between
the EBA and six Balkan countries.90 These commit signatory supervisory authorities to, among other
things, take steps to adjust their regulatory and supervisory standards, as well as institutional arran-
gements, to those existing in the EU in a manner and timeframe that are appropriate considering
the specifics of each market. The adoption of the EU regulations and supervisory frameworks facili-
tate cooperation between host non-EU authorities and home EU consolidating supervisors. This helps
ensure that both authorities “speak the same language”. For example, they have developed common
methodologies to perform risk assessments, making the host individual risk assessment consistent to
the ones presented in the supervisory college.

Box 8: Regulatory convergence in EU-candidate countries in bank supervision and resolution

REGULATORY CONVERGENCE IN EU-CANDIDATE COUNTRIES IN BANK SUPERVISION AND RESOLUTION

Many of the non-EU hosts have implemented or are in the process of implementing EU-based regulations.
These standards are based on the ones approved by the Basel Committee on Banking Supervision and cover
the following areas:

• CRR-like capital requirements for credit, market and operational risk, including requirements for capital
buffers and leverage ratio. But some differences with EU regulations usually persist. For example, some
non-EU host countries have implemented stricter minimum Pillar 1 capital ratios across the board, higher
standardized risk-weights for certain portfolios (i.e. residential mortgages), and have not granted the
possibility to calculate capital requirements based on internal models (as in Basel II and Basel III).91

• LCR liquidity requirements have replaced or complemented progressively the local liquidity and funding
standards which were in force in many of these countries.

• Minimum governance standards on EU rules, regarding remuneration, internal control, role of the board
of directors, etc.

89
As explained above, in some jurisdictions most of the systemically important banks are integrated in EU banking groups..
90
The Banking Agency of the Federation of Bosnia and Herzegovina, the Banking Agency of the Republika Srpska, the National Bank of the Republic of
North Macedonia, the Central Bank of Montenegro, the National Bank of Serbia, and the Bank of Albania
91
The reluctance to allow capital requirements based on internal models is often explained by a number of reasons including: to avoid complexity and
due to lack of resources in the supervisory authority to perform model validation, small credit portfolios in the country do not justify the development
of internal credit models to calculate capital requirements, to avoid competitive disadvantages for local banks that do not have the capacity to develop
these models against foreign-owned entities, and some distrust in a regulation based on internal models.

42
 • Supervisory information based on EU standards.92

• Risk management standards close to EU ones, including, among others, requirements to develop ICAAP
and, in some cases, ILAAP reports.

• Supervisory powers to require institution-specific capital, liquidity, and other Pillar 2 measures to be
applicable to banks.

• EU-based requirements for banks to draft group recovery plans.

• Crisis management measures, including the ability of the supervisor to take early intervention
measures.

• In addition, many of these authorities are also adapting their processes to make them converge to EU
standards, as defined by the EBA and (complementary) by other supervisors, including amending their
supervisory assessment systems93 to ones based on SREP, as defined by the EBA Guidelines and de-
voting more time and resources to review documents and processes that are essential in the new EU
supervisory framework, like ICAAP, ILAAP, recovery planning or risk appetite framework.

Regulatory convergence with the EU framework is also taking place in bank resolution

• Many candidate countries have set-up new functionally independent resolution departments regu-
larly within existing supervisory authorities along the lines of the BRRD. Some important divergen-
ces from simple BRRD take over relate to resolution funding, phasing in of Bail-in and MREL or the
special treatment of “affiliated parties” exempting them from the pari passu principle.

• Drafting of resolution plans, resolvability analysis to identify ex-ante potential obstacles to the
resolution of the banks as well as imposition of MREL requirements still most authorities have not
yet started to prepare.

• Also, the starting point for introducing a special bank resolution regime, is quite different from most
EU countries in many FinSAC client countries as most already had an administrative based liquidati-
on process allowing for P&A and Bridge Banks.

5.3.2. Potential issues faced by host authorities in candidate countries

Even though candidate country host authorities are making important progress towards the implemen-
tation of regulatory and supervisory frameworks aligned with the EU, it is likely they will require more
stringent capital and liquidity requirements if the coordination and cooperation with home authorities
is not satisfactory. Some common problems faced by host supervisors when dealing with EU home
supervisors include:

a) Non-EU host countries must implement the European rulebook before they can qualify as candidates.
EU regulations are complex and comprehensive, and the transition from old regimes (usually based on
modified versions of Basel I and with limited regulations on resolution) is difficult. In many cases, the local
authorities must digest and understand the European regulations first, and then assess if the EU rules are
relevant to their financial systems. In many instances, the EU rulebook is not necessarily well adapted to
the relatively small and non-complex financial systems heavily based on banking intermediation94. Techni-
cal assistance providers can help with the interpretation and tailoring of EU regulations to the country
specific environment.

92
Supervisory reporting information and on own funds and other prudential rules.
93
Some of these countries have had other supervisory assessment methodologies; some of them based on the US CAMELS rating system.
94
Such as MREL, remuneration, and internal approaches for the trading book capital requirements.

43
 b) Many non-EU hosts do not participate in the EU Supervisory Colleges, particularly those organized by
the SSM. This may create or increase the information asymmetries between home and host supervisory
authorities. As part of informal feedback sought from host supervisors in FinSAC client countries, some
non-EU supervisors replied that the attention paid by the home supervisor has decreased since the esta-
blishment of the SSM. They believe that the colleges currently organized by Eurozone national supervisory
authorities have a higher engagement level with non-EU hosts than the colleges organized directly by the
SSM for significant institutions. It may well be that this observation proves to be temporary, as the SSM is
a relatively new institution and will need time to develop its engagement with host supervisors.

c) Participation in supervisory colleges remains much more widespread than in resolution colleges. This may
be explained by several reasons. First, the resolution authorities are generally young institutions, mainly
created after the national implementation of BRRD. The SRB was only established in 2015. Hence, they
have been focusing on organizational, logistical, and technical issues since most activities they conduct
(resolution planning, MREL, etc.) were completely new. Even though CMGs for global systemically import-
ant institutions have been organized since 2011, resolution colleges have been less relevant for small coun-
try hosts so far. For many of the issues discussed in the European resolution colleges a policy has yet to
be developed and discussed. Moreover, the confidentiality and professional secrecy regime assessment
as “EU-equivalent” a requirement to become observer is a very demanding and work intensive process for
EU authorities, especially for the SRB, given its workload and recent establishment. In addition, the EBA
has not (yet) issued any non-binding equivalence assessment for resolution, unlike for supervision (see
above). These EBA assessments shall be employed as a basis for the assessments by the different resolu-
tion authorities across the EU. Most host countries have only recently passed laws detailing their special
resolution regimes and their operationalization is at a relatively early stage of development. In many cases
the resolution authorities are still in the process of staffing and deciding on their organizational set-up.
They have yet to prepare resolution plans or resolvability assessments, and hence cannot contribute to the
development of the European colleges. The group level resolution authority should nevertheless ask these
authorities to join at this early stage – acknowledging that formal request has to come from the host au-
thorities- to start establishing relationships, building trust, and expanding good practices that can further
harmonize resolution planning.

The next few years will be very challenging for both EU and non-EU resolution authorities. They must
finish the institutional set-up, decide on policy issues such as MREL, draft first or new versions of re-
solution plans and resolvability assessments, and enforce requirements for banks (including measures
to remove obstacles to resolution). Though not legally obliged like EU hosts, small non-EU hosts might
want to determine, jointly with the EU home resolution authority, if the subsidiary constitutes a se-
parate point of entry in a MPE context or is included in a SPE banking group. If non-EU subsidiaries
were to be considered as separate points of entry, the subsidiaries need to be separable from the rest
of the group to ensure (i) the parent entity can enter resolution without the need for the subsidiary to
be resolved, and (ii) if the subsidiary is resolved, the parent entity can continue to perform as a going
concern because it has decided not to save the subsidiary (see Box 9). Small host authorities will pre-
fer ensuring enough loss absorbing capacity (internal LAC) is available in case of need, regardless of
whether the overall group resolution approach is SPE or MPE.

If significant disagreements between home and host authorities arise, the EU conflict-solving mecha-
nisms are not available to non-EU hosts and there is no joint decision for these cases. The bargaining
power of non-EU small host authorities may be rather limited in practice, especially as they may not
want to challenge or formally oppose approaches decided by the centralized Eurozone supervisors if
their countries are in the process of becoming members (or candidates) to the EU.

44
Box 9: Can parent banks let subsidiaries fail; how the new resolution framework can shape the debate

CAN PARENT COMPANIES LET SUBSIDIARIES FAIL?

HOW THE NEW RESOLUTION FRAMEWORK CAN SHAPE THE DEBATE

As the parent company and the subsidiary are different legal entities their interests are not always aligned, and
the subsidiary of the banking group may sometimes not be rescued by the parent company.

However, in practice parent companies normally have a clear interest in bailing out their subsidiaries:

• In many cases, not bailing them out is simply impossible. In many banking groups, the parent companies
guarantee the liabilities of their subsidiaries through financial guarantees, or comfort letters. This is
usually the case for centralized banking groups where decisions are made at parent level and the subsi-
diary is just a conduit for business in the host country. These subsidiaries may be, in some cases, very
close in nature to branches. Consequently, the performance of these groups is so correlated that the
subsidiaries cannot fail without the parent company failing, and vice versa.

• However, there may be cases where the links between the subsidiary and the parent company are not
so close and the parent company must decide if it supports the subsidiary. In most cases the parent
company will decide to grant support (in the form of capital or liquidity or both) to the subsidiary
for the following reasons:

° Financial interest: Often not letting the subsidiary fail is in the financial interest of the parent,
since the new investment made by the parent would serve to clean up the balance sheet and give
a new start. There are many examples of banks recapitalizing their subsidiaries following losses.

° Potential liability: If the decisions made by the management of the subsidiary (which in turn
reports to the management of the banking group) is responsible for the bank failing, the parent
company may be deemed responsible for the losses. If the liability risk is relevant, the parent
bank would decide to support the subsidiary. HSBC, for example, was effectively banned from
walking away from its US subsidiary Household Finance.

° Reputational risk: Even when the bank failure is not linked to decisions of the parent entity (for
example, in the case of a macroeconomic or systemic crisis in the host country), the expectation
of the host authority will be that the parent entity should cover losses if they arise. Particularly
in the case of cross-border banking groups with a presence in different host countries, any signs
of not supporting the subsidiary in one country may prompt other host supervisors to ring-fence
the subsidiaries in their country.

° Host country: Some countries explicitly state their expectation that the subsidiaries should be
supported by parent entities. In the US, for example, the “source of strength” doctrine has been
the main rule, and it has been recently complemented with the requirements for foreign banks to
set up separate capitalized and governed intermediate holding companies.

Nevertheless, there are cases where parent entities decided not to support their subsidiaries.
They can be divided into two different groups:

• The parent entity or group is in distress. In some situations, the whole group may be insolvent and the
parent company may not have placed enough loss-absorbing resources in the subsidiary. In these situa-
tions, the losses cannot be up-streamed to the parent entity. See, for example, the European subsidiaries
of Lehman Brothers.

• There is no financial interest in supporting it. The parent entity decides to not support the subsidiary,
because it considers that the failure cannot be attributed to its management of the subsidiary. In these
very rare cases, the subsidiary may fail or be bailed out by the host country´s public sector. See the cases
of Citibank or Credit Agricole during the Argentinian crisis or, more recently (to a more limited extent)
the case of the Greek subsidiary of Credit Agricole (Emporiki).

For group loss absorbing approaches under SPE/MPE resolution strategies see above 4.2.

45
Many non-EU resolution authorities have already implemented MREL legislation and are in the process
of implementing MREL calibration for individual banks. MREL requirements are mostly based on the EU
approach, under which the group’s consolidated MREL also includes the exposures that the group has
through its third-country subsidiaries and branches. When setting MREL, several scenarios have to be
considered from a third country small host perspective.

If, in case of non-viability, the resolution plan of a bank foresees that the bank enters resolution
proceedings, then the MREL should be calibrated considering the capital requirements the resolved
bank will face after resolution is enforced. For banks expected to be resolved under bail-in preferred
resolution strategies, it may well mean that the bank should hold roughly twice its minimum capital
requirements to comply with MREL requirements. Requiring MREL as high as twice the capital require-
ments can be challenging, however, and might have unintended consequences, particularly considering
shallow private debt markets and common higher local capital requirements across the region.95 Parent
companies may have to increase their investments in the subsidiaries significantly without increasing
the local assets, effectively reducing their returns.

However, considering the flexibility that EU-based MREL requirements provide and the specific featu-
res of non-EU banking systems, resolution authorities consider different aspects of the resolution stra-
tegy when defining institution-specific MREL requirements. They may consider sale of business96 or
other partial transfer resolution strategies which would trigger lesser MREL instead of bail-in, although
this resolution strategy has also its own limitations in small countries.

In most accession countries that have aligned their framework with the BRRD, it is not defined in
which form MREL instruments have to be provided for i.e. capital or debt instruments. It has also not
been specified if MREL would always need to be ‘pre-positioned’ on the balance sheet of the entity in
the form of a fully paid up instrument or not. Prepositioning will give highest legal certainty to host
countries to bail-in the instrument when needed. That said, fully, paid-up instruments might expand
banks’ balance sheets with potentially negative consequences on performance and increased invest-
ment risks. In case collateralized guarantees are allowed under certain specified circumstances (under
internal MREL approaches) considering SPE and MPE structures, authorities should pay attention to
ensure collateral is executable in a timely manner i.e. not governed by foreign country law.

The strengthening of prudential standards across the banking system reduces the probability of bank
failures but there is a cost, especially for small jurisdictions with limited business opportunities. Banks
will argue that raising the levels of capital and liquidity or restricting the exposures of the subsidiary
reduces the attractiveness of the local financial sector to foreign direct investment. Some EU pa-
rent entities have been selling their subsidiaries in many non-EU countries to focus on their domestic
banking markets or on a more limited number of markets. The number of western European banking
groups interested in engaging in some banking markets has decreased, significantly in some cases.
Raising the capital requirements of the local subsidiaries (for example, through a strict implementation
of the MREL requirements) may prompt some parent entities to reassess their presence in the local
market.97 A balance should be struck between host authorities’ objective of financial stability by ensu-
ring legal certainty, and comfort that the subsidiary can be resolved even without parental support and
the effects of (internal) MREL on the returns from a group perspective.

95
Capital requirements are often higher in non-EU countries than in the EU. This is for several reasons. First, it is not unusual for minimum Pillar 1 capital
requirements to be as high as 12%; and for newly implemented capital buffers and Pillar 2 requirements to be required on top of this 12%, without any
transitional period. Second, risk-weights are usually higher in non-EU countries since there is rarely the option to use internal models for calculating risk-
weights and standardized risk-weights are normally higher for certain portfolios (e.g. residential mortgage portfolios). Finally, capital requirements act as
a buffer for weaker data quality and granularity and capacity in supervision in some countries.
96
If the sale of business is the main resolution strategy, the resolved bank will be expected to be sold to a third-party buyer when resolved. In this sense,
it can be argued that the provision of the fresh capital required to recapitalize the bank would be provided by the buyer, so there may be no need for
MREL to consider the recapitalization amount. And the credibility of the application of this strategy in many of these countries is frequently high because
of the abundance of potential buyers with capacity (capital and liquidity) and interest in a potential takeover of the resolved bank. This is explained by
the fact that many of the potential buyers may be large foreign groups already operating in the country. To these buyers, the absorption of the bank will
not be challenging due to the small size of these subsidiaries in relation to the whole-group balance sheet.
97
These reassessments may consist of portfolio restructurings (e.g. reducing high risk-weight exposures for lower risk-weight ones), significant downsiz-
ing of their activities in the country, or even the outright selling of the full operation.

46
Like the small EU hosts, the home supervisor may decide to enforce stricter limits on sovereign holdings
of debt issued by non-EU hosts or to divest in certain non-EU NPLs to which the banks are exposed th-
rough their subsidiaries. These decisions (taken on a consolidated level by EU supervisors) may impact
local financial systems, triggering the sale of NPLs or locally-issued sovereign bonds and impacting the
credit conditions in the host country (Box 10).

Box 10: Sovereign bond holdings in non-EU countries

SOVEREIGN BOND HOLDINGS IN NON-EU COUNTRIES

CRD Article 114 states that government bonds of non-EU states attract a non-zero risk weight unless (i) their
risk is rated not lower than AA-/Aa3 or (ii) supervisory and regulatory arrangements are considered equivalent
to those applied in the EU. If the second condition is satisfied, the risk weights as applied by the host country for
their bonds in local currency are accepted in the EU.

This requirement creates a divergence between the consolidated and the solo capital calculations. In accordance
with the Basel regime, the non-EU host subsidiary applies a zero risk weight to sovereign bonds. The risk weight ap-
plied by the EU-based parent bank for the consolidated capital calculation will generally be higher, unless the host
country’s supervisory and regulatory regime has been deemed equivalent to the EU or the host country sovereign
has a high credit rating. Non-EU host countries argue that this divergence can have negative repercussions for the
development of government bond markets in their country, especially where foreign banks dominate the banking
system. Further, it can complicate the definition of high-quality assets for the LCR and ultimately undermine
access to liquid assets.

5.3.3. Bank regulation, supervision, and resolution in non-candidate countries

Unlike EU candidate or accession countries, non-candidate countries have no incentive to adopt the EU
regulatory and supervisory framework, but rather have the freedom to focus on the most immediate
needs in upgrading their financial safety net. While most of these countries aim to adopt Basel II and
III, they can, and should, be more selective in implementing reforms.

Unlike EU-hosts, non-EU hosts can take unrestricted actions regarding a foreign-owned subsidiary, to
effectively ring-fence it from the rest of the group, as they are not bound by the EU rulebook. They may
take several measures, including:

• Increase the capital requirements that are applicable to that subsidiary, making use of Pillar 1 and Pillar 2
requirements. Some host supervisors have already taken these measures to strengthen capital buffers and
thus strengthen the subsidiaries under their supervision from the problems in their parent entities.

• Restrict the dividend policy of the subsidiary, forcing the subsidiary to apply a specific profit retention po-
licy, especially if it is expected that the parent company would not be able to support the subsidiary when
needed.

• Restrict the liquidity and funding exposures from the subsidiary to other members of the group—the most
frequently used restriction is to impose large exposure limits to the group (usually more stringent that the
25% limit over own funds that the BCBS recommends).

• Reinforce the governance requirements of the subsidiary, asking for a specific policy on intra-group
transactions or increasing the number of independent directors in the subsidiary´s board.

Most non-candidate countries in the region face significant challenges with effective bank supervision
because of limited resources. A bank resolution framework might or might not exist, and the instituti-
onal foundations to implement different resolution approaches are often missing. The BRRD resolution
tools may not be feasible in non-candidate countries, with the practical approach mostly likely being a

47
purchase and assumption operation and possibly a bridge bank operation. The main challenge for the
first is the choice of acquiring bank and the implications for the market. The biggest challenge for a
bridge bank is the limited capacity in most countries to run such an operation and the risk that it will
last longer than expected. Discussions on MREL are premature in this context as bail-in is very difficult,
if not impossible, because there are few non-deposit liabilities and debt markets yet to be developed.
This situation is the same in many other developing countries across the globe.

Experiences from other regions, such as Central America and Africa (Beck et al., 2014) have pointed
to the importance of focusing on effective bank supervision (as gauged by, e.g., the Basel Core Princi-
ples of Effective Supervision), especially the principles related to cross-border supervisory information
exchange, coordination, and cooperation. It also points to the important effect that cooperation among
host countries with similar levels of supervisory capacity can have in terms of learning effects, and in
terms of a stronger voice of host countries vis-à-vis home supervisors. It also remains important to
continue to develop strong supervisory frameworks and skills to ensure the “ability” and the “will” to act
early and effectively. This requires appropriate resources, authority, a clear and unambiguous mandate,
operational independence and accountability, and constructive working relationships with the home
supervisor. Nevertheless, without a credible threat of resolution, prudential supervision remains hard
to enforce.

6. Conclusions

This section provides some conclusions of our analysis, which can also serve as recommendations for
different authorities.

1) Non-EU hosts should participate in EU resolution colleges and CMGs.

The participation of non-EU hosts in EU resolution colleges and CMGs as observers should be encouraged.
The EBA can facilitate this process by issuing non-binding opinions concerning the assessment of confi-
dentiality and professional secrecy regimes of non-EU hosts. Like supervision, EU group-level resolution
authorities could use EBA assessments as a starting point to conduct their own. The EBA can also issue
a non-binding recommendation to EU authorities by which the non-EU resolution authorities of systemic
subsidiaries of candidate countries should be conferred preferential status for becoming observers within
the EU resolution colleges98. European consolidating authorities, especially the ECB and SRB, can also
take direct initiatives and invite non-EU supervisors / resolution authorities of systemic subsidiaries and
branches to the EU colleges.

2) Non-EU hosts should be given observer status in EU supervisory and resolution colleges

It is important to expand the active participation of non-EU hosts in EU supervisory and resolution colleges
as observers (preferential status) and ensure the same treatment among all host (candidate) countries.
This can be facilitated by the EBA issuing a non-binding recommendation to EU authorities on the treat-
ment of observers ie as regards access to documents, provide a platform for sharing of common concerns.

3) EU authorities should improve cooperation with non-EU hosts and consider how decisions will impact them

It would be helpful for EU authorities to be more transparent and reflect more on the impact of their de-
cisions on hosts’ financial systems. This is especially relevant for the impact of SSM and SRM on non-EU
hosts, since the latter do not benefit either from participation as members of the EU colleges or from the
joint decision processes under EU regulations. Communication and cooperation with small hosts before

98
Currently candidate countries that have a significant percentage of their banking system owned by EU banks, are grouped together with other third
countries.

48
 making decisions, including on reducing NPL levels and exposure of sovereign bonds, would be desirable. It
is also critical that EU consolidating supervisors, and particularly the SSM, consider the effects on small
hosts when authorizing cross-border M&A transactions between EU banking groups to thus avoid further
concentration in these markets. Additionally, more transparency in defining the main criteria for selecting
the resolution approach for EU banking groups – SPE or MPE – would also be helpful for host resolution
authorities, including being open about the consequences of identifying a subsidiary as a separate point of
entry and their approach to internal MREL (for SPE groups).

4) Small EU host countries should make use of the EBA mediation service when relevant

Small EU host countries outside the banking union should consider using the mediation services of the
EBA strategically to ensure their interests are properly protected when it comes to joint decision making
in colleges.

5) Non-EU host candidate countries should adapt EU regulations to fit their national circumstances

Non-EU host candidate countries shall, to the extent possible, adapt EU regulations to the features of their
local financial systems rather than adopting them blindly. A regional forum for cooperation in banking
supervision and resolution, or further expanding and focusing existing structures99, would be an important
step towards sharing common experience with EU authorities in supervision and resolution and represen-
ting joint interests of small host countries (see Box 11).

6) Non-candidate countries can be flexible in adopting the most appropriate elements from international
frameworks

Non-candidate countries in the region would benefit most from a more flexible approach to adopting Basel
III and EU rules, focusing on fundamental upgrades to their financial safety net, including bank resolution.
Participating in regional structures as discussed for candidate countries can be as, if not even more, helpful
for these countries.

Box 11: Steps for effective cross-border supervision in Central America

STEPS FOR EFFECTIVE CROSS-BORDER SUPERVISION IN CENTRAL AMERICA

The rise in international and regional banking groups and related cross-border supervision and resolution chal-
lenges for host countries are not confined to ECA. Central America has also experienced a significant (and in
some cases a dominant) foreign/regional bank presence even prior to the global financial crisis. Expansion of
cross-border banking resumed soon after the crisis eased. Among 12 principal banking groups operating in the
region in 2010, half were supervised by authorities in Panama and Colombia, and about a third were supervised
by authorities in Guatemala and Dominican Republic. There have also been instances of withdrawal of banks from
advanced countries—for example, Davivienda (Colombia) took over HSBC’s investments in Costa Rica, Honduras,
and El Salvador in 2012. These developments put cross-border supervision challenges at the forefront of policy di-
scussions in the region. The Central American Council of Superintendents of Banks, Insurance, and Other Financial
Institutions (CCSBSO)—a forum for coordination among supervisors in the region—has since carried out several
forceful initiatives to improve cross-border supervision and exchange of quality information among supervisors.

CCSBSO first assessed priorities for effective cross-border consolidated supervision and, with support from IMF’s
regional technical assistance center (CAPTAC-DR), prepared a “Strategic Plan” incorporating a concrete timeline
of necessary steps over the 2010-2014 period. In 2011, CCSBSO established a Liaison Committee (with repre-
sentatives from all member countries) to meet bi-annually to identify material risks and coordinate consolidated
cross-border supervision, share relevant information and concerns as much as (and as soon as) possible, request
information from member countries’ (home/host) supervisors. CCSBSO also put in place yearly implementation
targets including for developing new methodology/procedures, pilot and final implementations, automated exchan-
ge of information (monthly/quarterly), and the establishment of a (pilot) college of supervisors.

99
For example, the group of Banking Supervisors from Central and Eastern Europe.

49
 In 2012, a new methodology for improving risk analysis and consolidated supervision of banking groups was de-
veloped, and a comprehensive set of reporting formats and accompanying guidelines was delivered to the supervi-
sors in each country through workshops. By the end of that year, consistent data collection, analysis, and sharing
across countries was broadly achieved. Nevertheless, some hurdles remained—for example, until recently, Panama
did not supervise its bank holding companies so it could not provide information on them to other countries.

A welcome development was Colombia joining the CCSBSO in 2012, but transforming this new membership
status into active collaboration has been difficult, partly as legal changes were required to allow exchange of
information with the other members. In 2013, the Liaison Committee started performing tests on the implemen-
tation of the new methodology with actual data from financial groups through the newly established „College of
Supervisors“, and later in 2014 conducted global off-site assessments. Overall, the 2010-2014 strategic plan of
the CCSBSO achieved significant advances in the implementation of consolidated cross-border supervision in the
region. CAPTAC-DR provided substantial support to CCSBSO during this process through comprehensive policy
advice, technical assistance, and training.

Further progress will continue to be needed, however. The CCSBSO’s Strategic Plan for 2015-2019 focuses on
defining and implementing action plans (again with support from CAPTAC-DR) to (i) close the regulatory and
supervisory gaps in cross-border supervision and (ii) require financial groups to manage liquidity and solvency at a
consolidated level and to manage risks to regional groups. Moving towards fully effective cross-border supervision
remains challenging for the region, but not impossible given the strength of the reforms proposed and the CCSB-
SO’s commitment (and track record) to implement them.

Annex
Joint Decisions

• Institution-specific capital requirements; the consolidating supervisor and the rest of EU competent au-
thorities shall agree on the level of Pillar 2 capital to be required on a consolidated basis and on an individual
or sub-consolidated basis for each institution within the group.100 The authorities have a 4-month period
to reach a common decision after the consolidating supervisor submits a group risk assessment report to
the other authorities. In the absence of a joint decision, the consolidating supervisor will decide on the con-
solidated level, where the host authorities will decide on the individual or sub-consolidated levels of capital.
Nevertheless, within the 4-month period, either the home supervisor or the host authorities may refer the
decision to the EBA, which will exert its binding mediation.101 These joint decisions are becoming the most
important piece of home-host coordination and cooperation in the EU.

• Institution-specific liquidity requirements. The home supervisor and the rest of the EU competent au-
thorities shall agree on the liquidity levels to be held by both the group on a consolidated level and by
each of the institutions on an individual or sub-consolidated level.102 The authorities have a one-month
period to reach a common decision after the consolidating supervisor submits a report on the liquidity risk
assessment of the group to the other participant authorities. In the absence of a joint decision, the home
supervisor will decide on the consolidated level, where the host authorities will decide on the individual or
sub-consolidated levels of capital. Within the one-month period, the consolidating supervisor or the host
authorities may refer the decision to the EBA, which will exert its binding mediation. Although much less
material (as not directly affecting leverage and shareholders’ returns) than the joint decision on capital, this
one-month deadline to come to an agreement is expected to gain relevance in the future.

100
See Article 113 of CRD IV and Commission Implementing Regulation No 710/2014 of 23 June 2014 laying down implementing technical standards with
regard to conditions of application of the joint decision process for institution-specific prudential requirements according to Directive 2013/36/EU of the
European Parliament and the Council.
See Article 19(3) of Regulation (EU) No 1093/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Su-
101

pervisory Authority (European Banking Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/78/EC
102
See Article 113 of CRD IV and Commission Implementing Regulation No 710/2014.

50
 • Validation of internal models and approving significant changes in them,103 through a Joint Decision on
internal models on credit risk, counterparty credit risk, market risk, and operational risk. The authorities
have a 6-month period to reach a common decision after the home supervisor receives the specific appli-
cation. These 6-month processes are very relevant for supervisory colleges and especially for home super-
visors because if the authorities fail to reach a joint decision the home supervisor may decide unilaterally to
allow the use of internal models for consolidated supervision purposes, not on a solo basis. However, either
the consolidating supervisor or the host authorities may refer the decision to the EBA for a decision under
binding mediation.

• Assessment of the significance of a cross-border branch in the EU. The competent authorities of a host
Member State may request the home supervisor to consider a branch under its supervision as “signifi-
cant”.104 The host supervisor shall make a request to the home supervisor and after this both will have two
months for reaching a joint decision. If no joint decision is reached, the host supervisor shall take its own
decision. This two-month process has not been very relevant until now but is likely to increase in relevance
if more cross-border banking groups start using branches instead of subsidiaries.

• Assessment of the group recovery plan, involving an individual recovery plan and specific requirements on
the group recovery plan.105 The authorities shall make a common decision within a 4-month period, starting
from the day the home supervisor submits the group recovery plan to the host authorities. However, the
home or the host authorities may refer the decision to the EBA for a decision under binding mediation. This
4-month assessment period is gaining relevance, becoming one of the main pillars of the supervision of
banking groups. Its nature as a bridge between going concern supervision and crisis management increa-
ses its usefulness as a supervisory tool.

• Authorization of a group financial support agreement (GFSA):106 An ex-ante agreement between mem-
bers of the group regarding the provision of liquidity or funding facilities in case of stress, subject to the
compliance with some requirements. The authorities have a period of 4 months after the home supervisor
submits the application to the rest of the authorities to agree on a common decision; if no decision is
reached after that period, the home supervisor will decide on the authorization of the GFSA. However, the
authorities may request binding mediation by the EBA advisors.

• Authorization of actual group financial support (based on GFSA):107 in case of triggering support under
an IGFSA, supervisors should reach a joint decision on allowing or prohibiting/restricting the requested
support within five days. The matter can be referred to the EBA for non-binding assistance, but the final
decision lies with the supervisory authority of the liquidity-providing entity.

103
See article 20(1) of CRR and Commission Implementing Regulation 2016/100, of 16 October 2015 laying down implementing technical standards spec-
ifying the joint decision process with regard to the application for certain prudential permissions pursuant to Regulation No 575/2013 of the European
Parliament and of the Council.
104
See article 51 of CRD IV. Further to the joint decision process, this article also include some non-exhaustive criteria for considering a branch as “signif-
icant”. In particular; (I) a deposit market share in the host Member State higher than 2%; (ii) the likely impact of the suspension of the operations of the
institution on the systemic liquidity and the payment, clearing, and settlement systems in the host Member State, and (iii) the size and importance of the
branch in terms of number of clients within the Member State.
105
See article 8(2) of BRRD.
106
See article 20 of BRRD.
107
See Article 25 of BRRD.

51
References

Beck, T., Todorov, R. and Wagner, W. (2013): Supervising cross-border banks: theory, evidence and
policy, Economic Policy, Vol. 28, Issue 73, p. 5-44

Beck, T., Senbet, L. and Simbanegavi, W. (2014): Financial Inclusion and Innovation in Africa: An
Overview, Journal of African Economies, Vol. 24, Issue suppl_1, p. i3-i11

Bertay, A. C., Demirguc-Kunt, A., and Huizinga, H. (2011): Is the Financial safety net a barrier to
cross-border banking?, CEPR Discussion Paper No. 8712

Commission Implementing Regulation (EU) No 710/2014 of 23 June 2014 laying down implemen-
ting technical standards with regard to conditions of application of the joint decision process for
institution-specific prudential requirements according to Directive 2013/36/EU of the European
Parliament and of the Council

Commission Implementing Regulation (EU) 2016/99 of 16 October 2015 laying down implemen-
ting technical standards with regard to determining the operational functioning of the colleges
of supervisors according to Directive 2013/36/EU of the European Parliament and of the Council

Commission Implementing Regulation (EU) 2016/100 of 16 October 2015 laying down implemen-
ting technical standards specifying the joint decision process with regard to the application for
certain prudential permissions pursuant to Regulation (EU) No 575/2013 of the European Parlia-
ment and of the Council

Commission Delegated Regulation (EU) 2016/1075 of 23 March 2016 supplementing Directive

2014/59/EU of the European Parliament and of the Council with regard to regulatory technical
standards specifying the content of recovery plans, resolution plans and group resolution plans,
the minimum criteria that the competent authority is to assess as regards recovery plans and
group recovery plans, the conditions for group financial support, the requirements for independent
valuers, the contractual recognition of write-down and conversion powers, the procedures and
contents of notification requirements and of notice of suspension and the operational functioning
of the resolution colleges

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 require-
ment for own funds and eligible liabilities

D’Hulster, K. (2012): Cross-border banking supervision: Incentive conflicts in supervisory informa-

tion sharing between home and host supervisors, Journal of Banking Regulation, Vol. 13, Issue 4,
p. 300-319

D’Hulster, K. and I. Otker-Robe (2015): Ring-fencing Cross-Border Banks: An Effective Supervisory

Response?, Journal of Banking Regulation, Vol. 16, Issue 3, p. 169-187

Directive 2009/111/EC of the European Parliament and of the Council of 16 September 2009
amending Directives 2006/48/EC, 2006/49/EC and 2007/64/EC as regards banks affiliated to
central institutions, certain own funds items, large exposures, supervisory arrangements, and cri-
sis management

52
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

Directive 2014/59/EU of the European Parliament and of the Council of 15 May 2014 establis-
hing 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

European Banking Authority (2016): EBA Annual Report

European Banking Authority (2018): Guidelines on Supervision of Significant Branches – EBA/

GL/2017/14

European Banking Authority (2018): EBA Report on the Functioning of Supervisory Colleges
in 2017

European Central bank (2018): 2017 SSM Annual Report

FSB, Guiding Principles on the Internal Total Loss-Absorbing Capacity of G-SIBs (‘Internal TLAC’),
July 2017.

Lintner, P (Hg.), Understanding bank recovery and resolution in the EU: a guidebook to the BRRD,
December 2017, World Bank, FinSAC

Lintner, P. (2017): De/centralized Decision Making Under the European Resolution Framework:
Does Meroni Hamper the Creation of a European Resolution Authority?, European Business Orga-
nization Law Review, Springer, Volume 18, Issue 3, pp 591–616.

Regulation No 1093/2010 amending Decision No 716/2009/EC and repealing Commission Deci-

sion 2009/78/EC – SRMR

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

Regulation (EU) No 806/2014 of the European Parliament and of the Council of 15 July 2014 esta-
blishing 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

Mero, K and D. Piroska (2015): Banking Union and Banking Nationalism – Explaining opt-out
choices of Hungry, Poland and the Czech Republic, Policy and Society, Vol. 35, Issue 3, p. 215-226

53
Authors

Ismael Ahmad Fontán, Senior Financial Sector Specialist, The World Bank/FinSAC

Ismael Ahmad Fontan is a Senior Financial Sector Specialist at the World Bank based in Vien-
na, Austria. Ismael Ahmad Fontán specializes in banking supervision, regulation, and recovery
and resolution planning. He worked in the G-SIB Supervision Department of the Bank of Spain,
where he engaged in on-site and off-site supervisory activities, including on-site inspections,
and SREP assessments. He also participated in the negotiations of key European regulations
(BRRD, CRD, etc.) and joined several international working groups. As a consultant in financial
services for PwC and then Bluecap Management Consulting he has experience of working in
various regions of the world. He has advised a number of banks and financial institutions, both
cross-border and domestic, mainly in developing SREP, ICAAP, ILAAP, Risk Appetite Frame-
work, Stress Testing, Risk Policies, Governance, and NPL management and resolution. He has
also assisted banks in developing recovery plans and complying with resolution planning requi-
rements set forth by resolution authorities. He has worked in Bosnia and Herzegovina, Albania,
Turkey, Indonesia, Ukraine and North Macedonia.

He holds a double bachelor’s degree in law and business administration from the Comillas Pon-
tifical University in Madrid and regularly provides training on financial regulation and banking
risks in seminars for professionals in the financial sector (i.e.: banks, law firms, public institu-
tions, etc.) and to post-graduate students.

Thorsten Beck, Professor Cass Business School

Thorsten Beck is professor of banking and finance at Cass Business School in London. He is
also a research fellow of the Centre for Economic Policy Research (CEPR) and the CESifo. He
was professor of economics from 2008 to 2014 at Tilburg University and the founding chair of
the European Banking Center from 2008 to 2013.

Previously he worked in the research department of the World Bank and has also worked as
consultant for – among others - the European Central Bank, the Bank of England, the BIS,
the IMF, the European Commission, and the German Development Corporation. His research,
academic publications, and operational work have focused on two major questions: What is the
relationship between finance and economic development? What policies are needed to build
a sound and effective financial system? Recently, he has concentrated on access to financial
services, including SME finance, as well as on the design of regulatory and bank resolution
frameworks.

In addition to numerous academic publications in leading economics and finance journals, he

has co-authored several policy reports on access to finance, financial systems in Africa, and
cross-border banking. His country experience, both in operational and research work, includes
Bangladesh, Bolivia, Brazil, China, Colombia, Egypt, Mexico, Russia, and several countries in
Sub-Saharan Africa. In addition to presenting at numerous academic conferences, including
several keynote addresses, he is invited regularly to policy panels across Europe. He holds a PhD
from the University of Virginia and an MA from the University of Tübingen in Germany

54
Katia D’Hulster, Lead Financial Sector Specialist, The World Bank

Katia D‘Hulster is a Lead Financial Sector Specialist at the World Bank based in Sydney, Aus-
tralia. Her role consists of providing technical assistance in prudential bank and conglomerates
supervision, crisis management, and bank recovery and resolution frameworks to financial
sector authorities. She has held senior positions with the World Bank in Washington DC, and
Vienna, Austria. Before joining the World Bank, she was a Senior Manager with the Australian
Prudential Regulation Authority responsible for the prudential supervision of one of the four
major banking groups in Australia. In that capacity, she was seconded to the Bank for Inter-
national Settlements in Basel. She was earlier a Policy Director at the International Swaps
and Derivatives Association (ISDA) in London, where she was responsible for developing policy
positions on risk management and IFRS accounting.

She holds a Doctorate in Applied Economics (Commercial Engineer) on the topic of cross border
banking supervision from the Brussels Free University (Solvay Business School). She passed
the professional exams and took the oath of Chartered External Auditor in Belgium. She also
holds the Global Association of Risk Professionals FRM and Professional Risk Management
(PRMIA) risk management certifications.

Pamela Lintner, Senior Financial Sector Specialist, The World Bank/FinSAC

Pamela Lintner provides consulting on the establishment and enhancement of crisis manage-
ment frameworks in FinSAC’s client countries with a focus on bank recovery and resolution.
Her role includes the development of financial sector reform advice and leading technical as-
sistance missions. She has participated in several FSAPs and World Bank crisis simulation
exercises. In her previous works as a lawyer for the Austrian Central Bank (Unit for Financial
Markets Regulation and Payment Systems Oversight) and before, from a different angle, at
the Austrian Bankers Association she was closely involved in European regulatory policy in-
cluding CRD IV, crisis management and deposit insurance. In the course of EMIR implemen-
tation, she headed the on-site inspections for the licensing process for the Austrian Central
Counterparty. Pamela studied law the Universities of Vienna and Madrid, received a research
scholarship at the European Institute of Public Administration (EIPA)/Maastricht and holds a
Doctorate from the University of Linz.

D. Filiz Unsal, Economist, The International Monetary Fund

D. Filiz Unsal is an economist at the Research Department of the International Monetary Fund.
When this paper was written, she was visiting FinSAC and provided consultancy on macropru-
dential policy, microprudential supervision, and bank resolution issues. Her research interests
and published articles are in the areas of monetary policy and financial sector governance,
frameworks and policies; as well as on development and open economy macroeconomics. She
has a PhD in Economics and a MSc in Economic and Finance from the University of York.

55