Recovery and resolution
for the EU insurance
sector: a macroprudential
perspective
August 2017
Report by the ATC Expert Group on Insurance
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Contents 1
Executive Summary 3
Section 1 Introduction 9
Section 2 Systemic risks and the case for an effective recovery and resolution
framework for insurers 10
2.1 Introduction 10
2.2 Systemic risks in the insurance sector 10
2.2.1 Spillovers to other sectors 12
2.2.2 Cross-border spillovers 15
2.3 Systemic risks associated with a low interest rate environment 17
2.4 A recovery and resolution framework for insurers from a macroprudential
perspective 18
Section 3 International and European initiatives on recovery and resolution
for insurers 24
3.1 Introduction 24
3.2 Global initiatives 24
3.3 EU initiatives 28
3.4 Current status and recent initiatives at national level 30
Section 4 Recovery and resolution powers and tools 35
4.1 Introduction 35
4.2 Recovery and resolution planning 35
4.3 Recovery measures and early intervention tools 37
4.4 Resolution tools 38
4.4.1 Commonly used resolution tools 39
4.4.2 Resolution powers and tools that allow the business to be transferred
or separated 41
4.4.3 Resolution tools affecting contractual rights 45
4.5 Cross-sectoral implications of resolution measures 47
4.5.1 Stay on termination rights tool 47
4.5.2 Bail-in tool 48
Section 5 Resolution funding 50
Contents
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Contents 2
5.1 Introduction 50
5.2 Funding sources other than public funds 50
5.2.1 Resolution funded by a resolution fund 51
5.2.2 Resolution funded by an insurance guarantee scheme 52
5.3 Ex post and ex ante industry financing 55
Section 6 Role of the macroprudential authority 60
Section 7 Conclusions and possible ways forward 62
References 65
Abbreviations 69
Members of the drafting team 70
Imprint 71
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 3
1. This report focuses on a recovery and resolution (RR) framework for insurers
1
from a
macroprudential perspective. Covering primary insurers and reinsurers, the report
(i) discusses the need for comprehensive RR policies to complement supervisory and
macroprudential policies; (ii) identifies and describes a number of potential RR tools;
(iii) highlights funding aspects of the resolution process; and (iv) considers cross-sectoral and
cross-border implications and contagion channels that arise when resolution tools are applied.
2. The disorderly failure of an insurer or a group of insurers may pose financial stability
risks. Recent studies have shown that the contribution of the insurance sector, especially that
of the life insurance segment, to systemic risk has increased. This is, in particular, due to a
substantial common exposure to aggregate risk, caused partly by the ever more explicit
sensitivity of insurers’ balance sheets to interest rate volatility, and to its growing
interconnectedness with the rest of the system through financial markets, e.g. due to their
active role in the capital and, to some extent, derivatives markets. As a result, rather than
absorbing adverse shocks, the EU insurance sector may transmit and/or amplify shocks to
other parts of the financial system once negatively affected. Moreover, there are also strong
linkages between insurers. For example, the failure of a reinsurer will directly affect other
insurers. More generally, certain types of insurers, in particular non-life insurers offering
compulsory and sometimes niche insurance, provide essential services that are necessary for
the functioning of the real economy. The disorderly failure of an important insurer of this type
at short notice might lead to a temporary shortfall in supply and an inability to address the
needs of the real economy, leading to a stalling in goods supply.
3. The regular insolvency procedure might be unable to manage a failure in the EU
insurance sector in an orderly fashion. A broad set of tools, in addition to those related to
insolvency proceedings, may enable authorities to be better prepared to deal with situations
involving the distress and default of insurers. The regular insolvency procedure may not
always be consistent with policyholder protection and financial stability objectives. In contrast
to other resolution tools, it may not consider the continuity of critical functions or the
preservation of any other important functions.
2
As such, in the event of the failure of a large
insurer or the simultaneous failure of multiple insurers, it may not be possible to prevent
contagion spreading to other parts of the financial system. Moreover, following the regular
insolvency procedure, the settlement of policyholders’ claims could be delayed by several
years, possibly undermining the wider public’s trust in the EU insurance sector as whole.
4. Financial stability and policyholder protection objectives are equally relevant to an RR
framework in the insurance sector. The possibility of the simultaneous disorderly failure of
several life insurers cannot be disregarded at this juncture. A prolonged low interest rate (LIR)
1
Unless otherwise specified, this report uses the term “insurer” to collectively refer to insurers that directly insure businesses
and households (primary insurers) and insurers that provide insurance to other insurers (reinsurers). The terms “primary
insurer” and “reinsurer” are used when the text only applies to one type of insurer. Specific references are made to
differences in the business models of reinsurers and primary insurers, since reinsurers insure primary insurers (and other
reinsurers), but not private households.
2
The report advocates further discussion of critical functions in the insurance sector. For details, see Box 2.
Executive Summary
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 4
environment could lead to solvency problems throughout the life insurance sector, in particular
for life insurers that offered products with guaranteed returns at a time when interest rates
were higher.
3
In the very unlikely scenario that the LIR environment were combined with an
event of suddenly falling asset prices (a so-called double-hit scenario), there could be a risk of
life insurers in several countries coming under stress simultaneously. Even if no failing insurer
were systemically important on its own, the disorderly simultaneous failure of several insurers
could, collectively, pose a risk to financial stability. Against this background, this report argues
that financial stability aspects should be considered as part of insurance RR frameworks,
together with policyholder protection objectives.
5. An effective RR framework needs to take a sectoral view, while allowing for the
principle of proportionality. The international effort has so far focused on global
systemically important insurers – nine global systemically important insurers (G-SIIs) have
been designated by the FSB, with five of these domiciled in the European Union (EU). G-SIIs
are subject to an enhanced RR framework, used primarily as a tool to address the “too-big-to-
fail” issue. There are, however, multiple challenges, such as the LIR affecting the sector as a
whole. Moreover, in addition to considering the financial stability implications related to a
disorderly failure of a single insurer, systemic risks arising from common exposures should
also be taken into account. In the LIR environment some institutions may prove unable to
successfully adjust their business models to new challenges and, if all other regulatory
measures fail, their orderly exit should be assured. This suggests that regulatory attention
should focus on the overall sector, including smaller and less diversified insurers which might
threaten financial stability if they failed simultaneously in a disorderly manner. Nevertheless,
the benefits of a RR framework with a broad scope need to be weighed against the additional
costs, allowing national authorities to follow the principle of proportionality.
6. An effective RR framework also requires arrangements to fund resolution without
having to resort to public funds. In some EU Member States the existing national
frameworks might have difficulty to avoid the event of the disorderly failure of a large insurer
or the simultaneous disorderly failure of several insurers without having to resort to public
funds. An undesirable alternative to the use of public funds would be a policyholder-funded
bail-in, with policyholders losing a significant part of their investments, which could have
widespread negative implications for sovereigns and financial markets. Moreover, if the
insurance guarantee schemes (IGSs) are not properly equipped to compensate policyholders
for their losses, public trust might also suffer.
7. Differences between national legislations increase the complexity of ensuring that the
failure of an insurer active in several EU Member States can be managed in an orderly
manner. The recent crisis has illustrated the consequences of a lack of effective crisis
management for financial institutions that are active across borders. Following the saying
“international in life and national in death”, the authorities in individual jurisdictions have the
power which could be applied only at the level of local entity rather than at the level of cross-
border group. As in the banking sector, large and internationally active groups play an
important role in the insurance sector in the EU, while reinsurers from the EU play a leading
3
As shown by the EIOPA 2016 insurance stress tests, both explicit and implicit guarantees are of relevance. For the
distinction of contractual guarantees, please consult the IAIS report on Systemic Risk from Insurance Product Features
(IAIS 2016b).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 5
role in the global market. At the same time, however, national insolvency procedures,
insurance RR frameworks and national IGSs continue to differ widely across the EU. Although
this allows national authorities to account for national specificities, a patchwork of national
rules cannot fully take account of the cross-border implications of a complex failure. This could
result in legal uncertainty, unequal treatment of domestic and foreign policyholders, potential
spillover effects into host countries and competitive distortions of national actions. The cross-
sectoral implications of the failure scenario, particularly in the case of a financial conglomerate
active across borders, would increase the complexity even further. This indicates that the
harmonisation of the EU RR framework for insurers across the EU would be a step in the right
direction.
8. While a comprehensive RR framework for the banking sector is operational at EU level,
an EU-wide policy strategy addressing risks related to the insurance sector is lacking.
Whereas the Solvency II Directive is a major step forward in the enhancement of the Single
Market, no EU legislation has been proposed, in respect of an RR framework for insurers or
IGSs. Filling this regulatory gap will require a broad range of stakeholders in Europe to work
together, including EU and national legislators, EIOPA, macroprudential authorities and
microprudential regulators. The possibility of an LIR environment continuing for a protracted
period of time underlines the fact that the work to develop, strengthen and harmonise an
effective RR framework for insurers at EU level should be reinvigorated and include
consideration of the related funding aspects.
9. The report considers a number of RR tools for the EU insurance sector. The tools
considered in this report vary in terms of their costs and benefits, implications for the different
stakeholders and applicability in the EU insurance sector. They are grouped as shown below,
reflecting how, and at what stage of an insurer’s distress or failure, they may be used.
• RR planning. The Solvency II framework includes provisions that require insurers that
breach their Solvency Capital Ratio (SCR) to prepare recovery plans (“ex post recovery
plans”). However, this report stresses that pre-emptive RR plans can increase
awareness (e.g. in terms of recovery capacity, resolvability and obstacles to the
resolution of an insurer) and might thus support more decisive action by both insurers
and supervisors if the solvency position of an insurer deteriorates. This indicates that
more attention should be devoted to the RR planning phase during “good” or normal
times and that supervisors’ powers should be aligned with this principle.
• Recovery measures and early intervention tools. Financial stability and consumer
protection are best served if the failure of an insurer can be avoided by applying
recovery measures (designed and implemented by insurers) and early intervention tools
(available to home authorities). Some early intervention tools are already available in
individual EU Member States. However, they typically only allow supervisory authorities
to intervene once solvency requirements have been breached. This might prevent
supervisory authorities from intervening in a timely manner. The expansion and
harmonisation of early intervention tools would help supervisors in individual EU Member
States to limit disruption to the wider financial market and would limit the unnecessary
destruction of value.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 6
• Commonly used resolution tools. Liquidation (as part of regular insolvency proceedings)
and run-off,
4
which is an insurance-specific resolution tool, are the most frequently used
and broadly available tools across the EU. They come, however, with a number of
deficiencies. Liquidation does not ensure the continuity of critical functions or the
preservation of other functions, it sometimes has priorities that differ from policyholder
protection and financial stability, and it is a time-consuming process. Despite these
shortcomings, liquidation remains a valid resolution tool – e.g. in conjunction with other
resolution tools discussed in this report. A run-off can ensure continuity of cover for
existing policyholders. It also does not require the fire sale of assets, thereby reducing
the destruction of value for policyholders. However, it may result in a partial settlement of
claims if not used in conjunction with other resolution tools discussed in this report.
Given the benefits to financial stability of the continuity of critical functions, the use of
run-off should be available across the EU and should include solutions to address any
funding shortfalls that might occur.
• Resolution tools that allow the transfer or separation of all or part of the portfolio.
Portfolio transfer, separation of assets and liabilities, and the use of bridge institutions
are considered in this report. With the exception of portfolio transfers, these tools are not
currently widely available in the EU for the resolution of an insurer. They increase the
resolution authority’s flexibility in identifying and separating viable business and/or vital
economic functions and liquidating the remainder under ordinary liquidation proceedings.
• Tools affecting contractual rights. The bail-in tool and the power to impose restrictions on
the termination of contracts are also considered in this report. These tools interfere with
contractual rights and have therefore been used rarely in the EU. The rationale for
imposing restrictions on the termination of contracts is to give the resolution authorities
the time to deal with a distressed insurer. The bail-in tool involves the restructuring,
limiting or writing down of liabilities and, at the same time, allocates losses to
shareholders and creditors, including policyholders, in a transparent manner. In contrast
to the ordinary insolvency procedure, it ensures the continuity of the critical functions and
viable parts of the insurer. The introduction of the bail-in tool is under consideration in
the Netherlands, and a few EU Member States allow policyholders to restructure
liabilities for the purpose of portfolio transfer. Further consideration could be given to the
option of granting the authorities in charge of resolution the powers to restructure, limit or
write down liabilities, including both insurance and non-insurance liabilities. The report
recognises that, due to the particular structure of insurers’ balance sheets, the bail-in tool
is probably less effective in the insurance sector than in the banking sector, when
applied to capital or debt. Still, the change to insurance liabilities should be a measure of
last resort and with adequate safeguards and a reliable source of funding in place to
ensure protection of policyholders. This could include the power to modify the terms of
existing contracts, for example, for life and saving contracts by reducing guaranteed
rates of return or by reducing benefits by a specified percentage.
10. The report highlights the relevance of insurance industry-funded arrangements for an
effective resolution process. Any failure and the related resolution process, including the
normal insolvency procedure, is associated with significant costs. Against this background, the
4
In the context of resolution, a run-off as a voluntary decision of the company is not viewed as a resolution tool.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 7
expansion and/or creation of funding arrangements should be assessed as part of any
discussion on RR tools. Dedicated resolution funds (RFs) or insurance guarantee schemes
(IGSs) are two possible sources of funding for resolution, and the costs of these are directly
borne by the industry. At this juncture, Romania is the only country in the EU with an
operational RF, whereas IGSs operate in the majority of EU Member States, albeit often with
limited scope for specific insurance policies. Moreover, the possible use of IGSs beyond
compensating policyholders is mostly restricted to the use of a limited number of resolution
tools (mostly portfolio transfer) and focused on a single objective of policyholder protection.
Taken together, the current funding arrangements appear incomplete and, in the event that a
default of an insurer or several insurers were to pose risks to financial stability, the need to
resort to public funds would be very likely. The report argues that both IGSs and RFs can play
an important and complementary role in the resolution process
5
and that their use could be
further explored at EU level. Notwithstanding the role IGSs could play in resolution funding,
the question of adequate policyholder protection across the EU also warrants further attention
(ESRB 2017).
11. Any resolution action should pay attention to the significant cross-sectoral spillover
effects of the resolution tools applied. For example, the application of a stay on termination
rights could impact counterparties through open derivative positions. The report recognises
that the application of the bail-in tool increases interconnectedness across sectors and,
therefore, the consistency of resolution regimes across sectors should be further evaluated
and improved. This would guarantee the effectiveness of the tools and minimise the costs and
cross-sectoral spillovers of resolution actions, particularly in the case of financial
conglomerates.
12. The interaction between resolution and macroprudential authorities poses some
practical challenges. This report argues that ongoing interaction between the resolution
authority and other stakeholders, in particular the supervisory and macroprudential authorities,
is desirable, and this interaction should be decided prior to any crisis. However, the process of
assigning responsibilities during a failure could be hampered by the fact that a resolution
authority for insurers has not been designated in most EU Member States.
13. Against this background, this report advocates the development of a harmonised
effective RR framework for insurers across the EU. This includes the following:
• Existing RR frameworks should be evaluated and, if appropriate, enhanced and
harmonised at EU level. Furthermore, efforts should be made to ensure their consistent
implementation.
• The existing RR toolkit should be expanded and the multiple use of RR tools should be
allowed. A majority of ESRB member institutions take the view that this should include
giving resolution authorities the power to modify the terms of existing contracts as a
measure of last resort and subject to adequate safeguards.
• The RR framework should cover the whole insurance sector, while allowing for
proportionality.
5
The existence of IGSs can provide compensation for policyholders where losses are too burdensome and the existence of
a resolutions funds (RFs) for cases when compensation is needed in line with the no creditor worse off (NCWO) principle.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Executive Summary 8
• The financial stability objectives of the RR framework should be recognised, with a
majority of ESRB member institutions taking the view that it should be put on an equal
footing with the objective of policyholder protection. In addition, the interactions of the
resolution authority with the macroprudential authorities should also be clarified.
• Lastly, work on RR frameworks should go hand-in-hand with a discussion of how
resolution should be funded.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Introduction 9
14. The global financial crisis revealed, amongst other fault lines, the shortcomings from a
financial stability perspective of a government bailout and normal insolvency
procedure. Although a government bailout mitigates contagion both within a sector and in
terms of its spread to other financial sectors, the associated increase in sovereign debt
creates substantial cost to current and/or future taxpayers. It also increases the risk of moral
hazard. The crisis also showed that a normal insolvency procedure can lead to spillovers such
as concerns about the liquidity and solvency positions of other market participants with similar
business models or asset holdings. These self-perpetuating dynamics then become an
important driver of market developments, leading to a systemic crisis and a generalised loss
of confidence in the financial system.
15. Recovery and resolution frameworks have been developed to address these
shortcomings. Since the crisis, new legislation has been put in place – at both global and EU
level – with the aim of making financial institutions more robust and reducing their risk of
failure. Moreover, new recovery and resolution (RR) tools are being developed to enable
authorities to intervene more effectively prior to a failure and, when institutions do fail, to
resolve them in an orderly manner that does not require taxpayer support and that minimises
the impact on financial stability. While the EU-wide RR framework for the banking sector is
operational and some EU Member States are in the process of strengthening their national
RR frameworks for insurers, no EU legislative proposal has been put forward for an RR
framework in the insurance sector.
16. This report aims to contribute, from a macroprudential perspective, to the ongoing
debate in the EU on the RR framework for insurers. Building on previous ESRB positions,
6
and the recent ESRB Secretariat staff response to the EIOPA Discussion Paper (ESRB 2017),
the report (i) discusses the need for comprehensive RR framework to complement supervisory
and macroprudential policies; (ii) identifies and describes a number of potential RR tools;
(iii) highlights funding aspects of the resolution process; and (iv) considers cross-sectoral and
cross-border implications and contagion channels.
6
See ESRB (2015), ESRB (2016a) and ESRB (2016b).
Section 1
Introduction
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 10
2.1 Introduction
17. This section describes systemic risks in the insurance sector and makes the case, from
a macroprudential perspective, for an EU recovery and resolution framework for
insurers. The first subsection reviews the relevance of insurance from a macroprudential
perspective, thereby laying the foundations for the discussion in the remainder of the report. It
highlights the possible occurrence of spillover effects in other (financial) sectors and other
countries when an insurer is in distress or fails. The second subsection analyses the impact of
a low interest rate (LIR) environment, while the final subsection sets out arguments for and
against strengthening existing frameworks in EU Member States and supports the
development of an EU-wide, harmonised RR framework.
2.2 Systemic risks in the insurance sector
18. A well-functioning insurance sector contributes to economic growth and financial
stability. Insurers play an important role in the economy as providers of protection against
idiosyncratic, financial and economic risks. With liabilities standing at one-third of EU
households’ wealth, consumers depend on the insurance sector for their future income.
Moreover, with assets worth two-thirds of EU GDP, the EU insurance sector is a significant
part of the financial sector and one of the largest institutional investors (ESRB 2015). In
particular, insurers are an important source of long-term funding, and their long-term
investment strategy can – in principle – enable them to act as shock absorbers in financial
markets (IMF 2016).
19. However, the sector may also pose systemic risks. The ESRB previously identified four
main transmission channels of systemic risk: (i) insurers amplifying shocks due to their
involvement in so-called non-traditional and non-insurance activities (NTNI); (ii) insurers acting
procyclically in terms of investment and pricing; (iii) the collective failure of life insurers under
a scenario of prolonged low risk-free rates and suddenly falling asset prices (i.e. “the double
hit”); and (iv) a lack of substitutes in certain classes of insurance vital to economic activity
(ESRB 2015).
20. The contribution of the insurance sector to systemic risk has increased since the
financial crisis. Although the systemic risk associated with a default by individual insurers
has changed little, the contribution of the insurance sector as a whole to systemic risk has
increased in recent years (IMF 2016a). This increase is associated with higher commonalities
in exposures to aggregate risk within the insurance sector and with the rest of the financial
sector, as well as greater exposure to market risks through asset and liability (duration)
mismatches, induced by increased sensitivity of certain types of insurers to interest rates. The
changing nature of insurance activities, both in terms of investments and product offerings
(e.g. a switch to unit-linked/defined contribution models for life insurance or the increased use
of early cancellation clauses), has resulted in greater commonality across the financial
Section 2
Systemic risks and the case for an effective recovery
and resolution framework for insurers
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 11
system. It thus follows that insurers may be more likely to perform poorly when other parts of
the financial system are also facing difficulties.
Box 1
Systemic risks of reinsurers
Reinsurers traditionally transfer risk within the insurance sector. They are important to the
global insurance industry in that they provide a mechanism by which the risks of a cedent (local
primary insurers or other reinsurers that cede certain risks to a reinsurer) can be pooled. As a
result, the cedent is protected from extreme events and “tail losses” on its own exposures as
specific underwriting (or targeted market) risks are transferred to a reinsurer. This system not only
allows insurers to limit potential losses from an individual policy contract (or from a portfolio of
policies), but also to increase underwriting capacity and achieve a targeted risk profile (e.g. by
reducing risk concentration). By spreading insurance risks globally, reinsurance diversifies losses
stemming from local insurance markets, while providing capital relief and balance sheet protection.
While the reinsurance industry is small in comparison with the primary insurance industry, it still
plays an important role in the non-life sector.
Reinsurance may also be a source of concern for financial stability. Although the ways in
which reinsurers and primary insurers may pose systemic risks are similar, there may be features
specific to reinsurance which need to be monitored from a financial stability perspective (ESRB
2015). The ESRB previously identified the following systemic risks posed by reinsurance: (i) intra-
industry interconnectedness with both primary insurers and other reinsurers (known as ‘reinsurance
and retrocession spirals’),
7
increasing the risk of contagion within the insurance sector; (ii) the risk
arising from high market concentration of reinsurers, both globally and in the EU, and the related
substitutability issue, which may lead to a risk of market friction in the event of a reinsurer failing;
(iii) interconnectedness with the rest of the financial system and possible procyclical investment
behaviour and (iv) captive reinsurance (ESRB 2015).
Not much is known about the systemic risk posed by the failure of a reinsurer, although
there are indications that it might have increased. Little is known about the pattern and degree
of damage caused by the failure of a reinsurer as there have been few such events in the past.
Indeed, major risk events in recent years have demonstrated the well-designed risk-absorbing
capacity of the global reinsurance market, with little spillover effects. However, several metrics
indicate that the likelihood of a failure might have increased, and the spillover effects for the whole
insurance sector might be more severe than those observed in the past. First, the reinsurance
business has become more concentrated, with the top five reinsurers accounting for more than half
of the reinsurance capacity worldwide.
8
This process has led, to better diversified portfolios across
risks and regions, including additional benefits of economics of scale, although it might also
increase the risk that reinsurers have become too-big-to-fail. Second, the ratings of reinsurers’
7
Insurance market spirals, both reinsurance spirals and retrocession spirals, arise from the interplay of practices employed
by the insurance industry to disperse risk and spread it across other insurers. Reinsurance spirals refer to links between
primary insurers and reinsurers whereas, whereas retrocession spirals refer to links between reinsurers themselves.
8
The top five reinsurers accounted for 53% of the market in 2015, up from 51% in 2010 (A.M.Best 2010 and A.M.Best 2015).
From a longer-term perspective, the top five reinsurers in 2009 accounted for roughly the same amount of the market as
the top ten reinsurers in 1998 (roughly 52%). In 1991, the top ten reinsurers accounted for 35% of the market (Cummins
and Weiss 2014).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 12
financial strength have deteriorated in recent years and, although they are still characterised by a
high degree of robustness, their outlook is negative (A.M.Best 2016). Furthermore, a change in
business practices, such as the increased use of a special termination clause (e.g. with respect to
rating triggers in reinsurance contracts) might exacerbate the problem of reinsurance and
retrocession spirals.
9
2.2.1 Spillovers to other sectors
21. Insurers can transmit shock across financial markets. Within Europe and North America
respectively, there could be large spillovers a between different sectors, including insurers. In
Asia, non-life insurers and reinsurers seem to be highly interconnected with other sectors in
the region. In terms of spillovers across the regions, Europe and North America appear to be
the most interconnected, with insurers (in particular life insurers) from Europe having a high
potential to transmit spillovers to the American financial market (IMF 2016a).
10
A separate
analysis for Europe indicated that, before the global financial crisis, insurers were recipients of
spillovers from other sectors although, more recently, they seem to have become a source of
spillovers (IMF 2016a).
22. Insurers may pose systemic risks arising from their funding and investment activities.
Collectively, insurers are among the largest investors in financial assets in the EU. They can
contribute to systemic risks through various channels, which include taking up more risks,
increasing commonality in asset composition within the financial sector, leading to increased
exposure to common shocks (“tsunami risk”), or increasing procyclicality in their investment
behaviour. For instance, analysis by the Bank of England concludes that the systemic risk
associated with activities of the UK insurance sector that propagate or amplify shocks to
financial counterparties or markets may be the greatest source of systemic risk from insurers
for the UK (French et al 2015).
23. Disruption to systemically important financial counterparties can occur if these
institutions no longer have access to funding from EU insurers. Insurers hold large
amounts of debt securities and shares issued by banks and other financial institutions in the
EU. From the perspective of banks’ balance sheets, these accounted for 4% of total bank
funding in the euro area in 2014 (ESRB 2015), while, on average, around 13% of debt issues
by euro area banks is held by insurers domiciled in the euro area and Sweden.
11
This figure is
even higher in some EU Member States, e.g. 28% in Belgium, Greece and Slovakia and 37%
in France. The ECB has emphasised that contagion risks from ownership links to banks and
other financial institutions are among the most significant risks (ECB 2008). Insurers also
9
Insurance market spirals, whether reinsurance spirals or retrocession spirals, arise from the interplay of practices employed
by the insurance industry to disperse risk and spread it across other insurers. While reinsurance spirals may impact primary
insurers, retrocession spirals may trigger failures of multiple reinsurers at the same time, which would also have a further
impact on a wide range of primary insurers.
10
According to the IMF, spillovers are defined as the impact of changes in asset price movements (or their volatility) in one
region/sector on asset prices in other regions/sectors, and are measured as each region’s/sector’s contribution to the total
residual variance of the equity returns of all other regions/sectors.
11
From the perspective of the insurers, 18% of the total financial assets of EU insurers in 2014 were investments in bank
bonds (ESRB 2015).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 13
place deposits with banks which can easily be withdrawn
12
and may lend cash to banks and
other financial institutions through repo transactions. Taken together, if the funding activities of
EU insurers were to cease abruptly on large scale, this could pose a systemic risk to other
parts of the financial system.
24. Insurers’ growing participation in capital markets and their increased “non-core”
activities seem to be the main reasons for their growing links with the rest of the
financial sector. According to the IMF, as a result of increased exposure to man-made
catastrophes (such as terrorist attacks) and through the concentration of insurance risks in
danger zones, insurers have been turning more frequently to capital markets and alternative
risk transfer (ART) mechanisms to mitigate the impact of catastrophes on their balance
sheets.
13
This has increased their links with the rest of the financial sector, including the
banks, which act as counterparties to ARTs, albeit that the amount of these transactions is
low. Insurers are interconnected with other financial market participants through credit default
swaps and securities lending, which caused major spillovers during the financial crisis.
14
They
are also interconnected through derivatives trades, in particular interest rate swaps (Abad et al
2016). Moreover, insurers participate in capital markets to diversify risk via credit-linked
securities (Baluch et al 2011). It also seems likely that these links may be attributed to the
“non-core” or “banking-like” activities of insurers, such as the provision of financial guarantees,
asset lending, issuing credit default swaps, investing in complex structured securities, and an
excessive reliance on short-term sources of financing (Cummins and Weiss 2014).
25. Procyclical investment behaviour by insurers could make them more likely to transmit
shocks, albeit that evidence to date has been mixed. Procyclical investment behaviour by
insurers as a group may exacerbate the tendency for financial markets to experience “booms”
and “busts”. The capital position of other financial and non-financial institutions could then
deteriorate by a fall in the market prices of financial assets held, which could lead to second-
round effects resulting from fire sales and liquidity spirals (Brunnermeier and Pedersen 2009).
Evidence of insurers in the EU demonstrating procyclical investment behaviour has so far
been mixed (Bank of England and Procyclicality Working Group 2014; Bijlsma and Vermeulen
2015, Timmer 2016).
26. Some insurers are directly interconnected with other parts of the financial sector since
they are part of financial conglomerates. There were 83 financial conglomerates in Europe
in 2016, up from 75 in 2009 (Joint Committee of European Supervisory Authorities 2016).
Insurer-led conglomerates are the second most common type of conglomerate after bank-led
conglomerates (Chart 1) – an insurance entity often forms part of a bank-led conglomerate,
since those conglomerates traditionally follow a bancassurance model. They are also present
in conglomerates led by asset managers or pension funds, which have gained prominence
over the last decade. In general, insurer-led conglomerates tend to be smaller than bank-led
conglomerates (European Commission 2010c) and some of these only operate locally. At the
same time, the 30 largest financial conglomerates, most of which have an insurance entity,
12
This was, for example, the case in Greece in 2014 (ESRB 2015).
13
The data on ART is for the US insurance sector only. There is a lack of EU data on this.
14
See, for example, Cummins and Weis (2014); Dungey et al (2014) and Pierce (2014).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 14
belong to the biggest financial groups in Europe: ten of the 13 EU G-SIBs and three out of five
EU G-SIIs are also financial conglomerates.
15
Chart 1
Number of financial conglomerates by type and domicile (2016)
Source: JC of ESAs (2016) and ESRB based on national data.
Note: Data for the UK also include four conglomerates with the head of group outside the EU/EEA.
27. A financial conglomerate may pose high risks of contagion spreading from one part of
the conglomerate to the others. Conglomerates have certain benefits, e.g. in terms of risk
diversification and the reinforcement of commercial capacity. This comes, however, with
additional costs in terms of interdependencies and, therefore, higher risks of contagion
between group entities. The intensity of the interconnections within a conglomerate depends
on what strategy the conglomerate employs to combine activities in different sectors. There
could be financial (e.g. intra-group transactions), operational (e.g. shared services) and
commercial links
16
and, moreover, there could also be reputational risks. For instance, the
failure of the insurance part of a financial conglomerate could lead to the significant
degradation of the financial position of other parts through the reputational channel. Contagion
could also spread in the other direction, e.g. the financial position of the insurance part could
deteriorate if another part were no longer able to reimburse a loan granted by the insurer or if
a breakdown of the banking agent network had materially affected the distribution model of
the insurance part. As parts of financial conglomerates, the distress or failure of the insurance
part could have a direct impact on other parts, and vice versa.
28. The degree of interconnectedness is one of the key elements in the global assessment
of systemic importance of insurers. The global financial crisis highlighted the
consequences of the close interconnectedness of the financial sector. This resulted in the
FSB placing high importance on the degree of interconnectedness in its G-SIFI framework.
15
G-SIBs refer to global systemically important banks and G-SIIs to global systemically important insurers. Both G-SIBs and
G-SIIs are designated annually by the Financial Stability Board.
16
Financial interconnectedness is related to intragroup transactions, operational interconnectedness to shared supporting
systems and services (e.g. IT), and commercial interconnectedness arises when a common distribution channel is used by
different parts of a conglomerate.
0
2
4
6
8
10
12
14
16
UK FR DE NL SE FI IT ES AT BE PT DK MT BG IE NO
insurer-led
bank-led
asset manager-led
pension fund-led
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 15
The IAIS approach weighted the interconnectedness category for insurers with almost 50%
(IAIS 2016c). The indicator covers both direct exposures via the counterparty exposure
channel and indirect exposures of insurers to the macroeconomy via the macroeconomic
exposure channel.
2.2.2 Cross-border spillovers
29. Cross-border activity of primary insurers in the EU is high. Considering both subsidiaries
and branches, the cross-border activity is more prominent in the EU primary insurance sector
than in the EU banking sector (Chart 2). For their EU cross-border business, EU insurers
operate mostly via separate legal entities in the form of subsidiaries. On average, only 3.5% of
gross premiums is written by foreign EU branches. Branches are, however, significant in some
EU Member States and their connections still form a dense and diversified network within the
EU insurance sector. As Chart 3 shows, in some cases the direction of exposures of branches
goes from small EU Member States (in terms of the size of the economy and the financial
sector) to large EU Member States. This means that even if the exposure is small for the
country in which the branch is located it could be large for the country in which the insurer is
domiciled. This argument also applies to exposures via subsidiaries.
Chart 2
Cross-border activity in the banking and insurance sectors within the EU (2012, %)
Source: EIOPA (2016), based on Schoenmaker and Sass (2014).
Note: Data refer to 2012. Cross-border activity in the insurance sector is measured as the percentage of gross written premiums written by
subsidiaries and branches controlled by foreign enterprises located in the EU. For the banking sector, cross-border activity is measured as the total
amount of foreign lending (assets) as a percentage of total lending (assets). The chart illustrates the cross-border activity in the EU coming from other
EU Member States only. The EU uses a simple average across all EU Member States.
0
10
20
30
40
50
60
70
80
90
100
AT BE BG CY CZ DE DK EE ES FI FR GR HU IE IT LT LU LV MT NL PL PT RO SE SI SK UK EU
insurers - branches
insurers - subsidiaries
banks
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 16
Chart 3
Network of exposures in terms of insurance branches (2014)
Source: OECD Insurance Statistics 2015. ECB Exchange Rate Statistics used for currency conversion.
Note: Data refer to 2014, and to business written abroad by EU/EEA insurers through branches and agencies. The thickness of the lines is relative to
the size of the business. Non-EU EEA countries (CH, IS, NO) are shown in light green.
30. The reinsurance sector is traditionally an international business, with EU-based
reinsurers playing an important role. The international nature of the reinsurance business
serves to diversify exposures to risk in the insurance sector. In particular, the EU-domiciled
reinsurers account for more than 45% of gross premiums written worldwide (A.M. Best
2014).
17
Moreover, in terms of the relative transfer of risks between geographic regions,
aggregated gross reinsurance premiums assumed and ceded by regions show that European
reinsurers are – in contrast to any other region – “net insurance risk takers” (IAIS 2017).
18
31. The failure of an insurer could have cross-border spillover effects. Given the high degree
of cross-border activity in the EU insurance sector, the failure of a cross-border insurance
entity in one country could impact financial stability in other countries. Similarly to financial
conglomerates, the effects relate primarily to financial (e.g. intra-group transactions),
operational (e.g. shared services), commercial and reputational links. They will differ
depending on the nature and intensity of interconnectedness within an insurance group, and
on whether cross-border insurers operate in host countries through subsidiaries or through
branches. Analogously, the failure of a reinsurer with a large international portfolio could have
repercussions in several countries due to the nature of their business associated with
retrocession and reinsurance spirals, as well as a high level of concentration in the
reinsurance market.
17
This figure increases to more than 60% if Swiss reinsurers are also included.
18
This is a persistent finding over time. See also previous reports (e.g. IAIS 2012).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 17
2.3 Systemic risks associated with a low interest rate environment
32. The LIR environment increases the likelihood of insurers’ failures, particularly in the
life sector. The current LIR environment highlights how the systemic importance of insurers
may change over time. Globally, the outlook for many insurance companies continued to
deteriorate in 2016 as expectations of an extended period of low interest rates rose (IMF
2016) and, even though Europe has shown some economic recovery, interest rates are likely
to remain low for some time. The LIR environment is not a risk in itself, although it is a trigger
for vulnerabilities, in particular in respect of life insurance.
19
Therefore, if the LIR environment
extends into the future, it is likely to weaken the resilience of the insurance sector across the
EU. Despite the varying exposure of national insurance sectors to this risk, the impact on the
insurance sector would be noticeable in all EU Member States.
33. The 2016 EIOPA stress test showed that in a prolonged LIR environment a significant
number of EU insurers would lose a substantial part of their assets over liabilities.
20
In
a situation of secular stagnation with yields remaining low for a long period of time, insurers
would experience a highly negative impact on their excess of assets over liabilities and own
funds.
21
In particular, insurers with long-term life policies involving interest rate guarantees
could face difficulties keeping their financial promises. If rates stayed low for long, a situation
could arise where policies would continue to pay out a return that is higher than incoming
returns on assets. In such an environment, relatively high interest rate guarantees on liabilities
with a longer payout period would weigh on the profitability and solvency of these companies
over time, which could eventually lead to failures. The evidence shows that the insurance
sector is reacting to the LIR environment by shifting towards policies with lower or no
guarantees for new contracts (ESRB 2016b).
34. The likelihood of a “double-hit” scenario has also increased. If the LIR environment were
combined with a sudden increase in risk premia there would be a risk that life insurers in
several countries could simultaneously come under stress. The 2016 EIOPA stress test shows
that European insurers are highly sensitive to this scenario, which would negatively impact on
their excess of assets over liabilities and own funds.
22
However, the impact would not be
equally spread among the different insurers or national markets. Even though the failure of an
individual insurance company may not be systemically important, if it failed at the same time
as other insurers it might contribute significantly to systemic risk. Against this background, the
common vulnerability to a “double hit” from low interest rates and declining asset prices is
19
Non-life insurers are experiencing less pressure due to their shorter investment horizons and the possibility of repricing
existing contracts.
20
The exercise involved 236 insurance undertakings at solo level, from 30 European countries (EIOPA 2016c).
21
According to the 2016 EIOPA stress test results, at an aggregated level, the “low-for-long” scenario resulted in a fall in the
excess of assets over liabilities of about EUR 100 billion, with insurers representing 16% of the sample losing more than a
third of their excess of assets over liabilities. If long-term guarantees (LTG) and transitional measures were not included,
25% of the sample lost more than a third of their excess of assets over liabilities in the low-for-long scenario. No impact on
groups was considered.
22
According to the 2016 EIOPA stress test results, the so-called “double-hit” scenario (reflecting a sudden increase in risk
premia in conjunction with the low yield environment) had a negative aggregated impact on the undertakings’ balance
sheets of close to EUR 160 billion (-28.9% of the total excess of assets over liabilities) with more than 40% of the sample
losing more than a third of their excess of assets over liabilities. If LTG and transitional measures were not included, almost
75% of the sample would lose more than a third of their excess of assets over liabilities. It should be noted that the “double
hit” scenario reflects a very extreme and rare combination of events.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 18
viewed as one of the most prominent systemic risks, possibly leading to cascading failures in
the sector (ESRB 2016a).
35. The changing nature of the insurance business exposes insurers to new types of risks
and increases their interconnection with the rest of the financial sector. For example,
the ESRB has identified that, as a result of structural changes in investments and product
offerings by life insurers, liquidity risks in the life insurance sector could become more
prominent than in the past. This is due to (i) the risk of selective redemptions by policyholders
when insurers invest in less liquid and long-term assets (e.g. infrastructure and real estate),
(ii) the transfer of investment risk to policyholders, including the broader provision of unit-
linked/defined contribution models, which are more easily surrendered at short notice, and
(iii) a move into bank-like savings products without adequate expertise and risk management
(ESRB 2016b). Moreover, with greater integration into financial markets, life insurers are more
exposed to risks stemming from other sectors, in particular through (i) higher lending to banks
and (ii) an increase in the impact of risk factors shared with the investment fund sector due to
greater product similarity following the shift to unit-linked products (ESRB 2016b).
36. A protracted LIR environment could also induce some insurers to increase investments
in risky and/or less liquid assets with a higher return, thus exposing them to a higher
probability of distress (search-for-yield behaviour). The move towards less liquid and
higher-risk assets such as stocks, infrastructure and real estate could occur, in particular, in
life insurance and long-tail non-life (casualty) insurance, although risk-oriented capital
requirements in Solvency II could mitigate this development. Although no shift of portfolios
towards riskier categories of assets has been generally observed throughout the insurance
sector, firm-level case studies suggest that smaller life insurers, in particular those with
weaker capital positions and those with higher shares of guaranteed liabilities, tend to take on
more risk (IMF 2016). This riskier behaviour could be also relevant for reinsurers, given the
high level of competition they face as other investors offer an alternative to traditional
reinsurance and drive down prices.
2.4 A recovery and resolution framework for insurers from a macroprudential
perspective
37. In this context it is worth revisiting the need to strengthen the RR toolkit at national
level and to examine the case for a harmonised RR framework in the EU. New
challenges and developments in the insurance sector mean that the existing legal and
institutional frameworks should be revisited across the EU in order to assess their sufficiency
to address systemic risks in the insurance sector in an effective manner, without creating
unnecessary distortions to other financial sectors or countries. An effective RR framework
should provide authorities with a set of early intervention tools in order to deal with, in a timely
manner, distressed insurers under a “going-concern” approach, and a set of resolution tools to
deal effectively with a failing insurer under a “gone concern” approach.
38. An effective resolution toolkit could mitigate the financial stability implications of a
failure in the insurance sector, in comparison with normal insolvency procedure. As the
experience of the recent crisis has shown, the insolvency procedure has several
shortcomings. First, it may not provide continuity of critical functions (see Box 2 on critical
functions). This could mean that the key economic transactions facilitated by insurance might
not be possible without policyholders incurring significant additional costs. Second, it might not
be able to prevent possible contagion to other parts of the financial system. This is especially
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 19
so for large insurers (e.g. the failure of AIG), but is also the case for the simultaneous failures
of small insurers. Another disadvantage of liquidation is the time required, which could result
in delays of several years in the settlement of outstanding claims, possibly undermining
society’s trust in the insurance sector as a whole and the financial system in general.
39. If they are equipped with a broad set of tools in the recovery phase, authorities might
be better able to avoid a failure. Insurance risks are typically “slow-burning”, including
longevity risks as well as risks associated with the LIR environment. This means that they lead
to failure only after a number of years if not properly addressed under the “going-concern”
approach. This period of time allows supervisory action to be taken, within the limits of legal
and institutional frameworks. It follows that expanding the toolkit for early intervention periods
provides authorities with more options to avoid an insurer needing to be wound down, which
would benefit policyholders as well as financial stability.
40. It follows that the objectives of the resolution of insurers are both financial stability and
policyholder protection, and that they are interlinked. According to the FSB, the objectives
of resolution are (without ranking): (i) to avoid severe systemic disruption; (ii) to avoid
exposing taxpayers to losses; (iii) to protect vital economic functions;
23
and (iv) to protect
policyholders (FSB 2014a). These objectives are interlinked and overlap. If vital economic
functions are impaired because of a failing insurer, systemic disruptions may arise and the
government might step in to bail out the failing insurer, exposing taxpayers to losses.
24
In the
event of a capital shortfall at a life insurer, there could be an impact on consumer confidence
in the financial sector as well as political pressure to bail out this insurer rather than let it enter
liquidation, given the nature of its liabilities (ESRB 2015).
41. Moreover, an effective RR framework requires taking a sectoral view, albeit subject to
the principle of proportionality. The challenges the insurance industry is facing indicate that
the RR framework needs to look beyond the individual firms that could pose a systemic risk on
their own. Instead, a broad perspective, where all insurers (including smaller insurers) are
subject to the RR framework, is warranted.
25
The potential solvency problems of individual
firms may lead to cascading effects that could become systemic. Although the RR framework
should, in principle, cover the whole sector, its implementation should follow the principle of
proportionality. For example, the benefits related to the application of certain pre-emptive
measures, such as RR plans, should be considered against the additional costs of their
implementation, in particular with respect to small insurers and/or insurers with less diversified
product offerings and with strong capital positions. Furthermore, national authorities should
have the flexibility to apply the RR tools that best suit local conditions and should also have
the power to exempt some insurers from certain aspects of the RR framework, e.g. RR plans,
without preventing authorities from applying all the powers at their disposal should the need
arise. The principle of proportionality as explained above applies across the whole report.
42. Strengthening existing RR frameworks for insurers across the EU could limit systemic
risks following distress or failure. Strengthening the current RR framework at national level
could help mitigate financial stability concerns in the ways below.
23
The terms “vital economic functions” and “critical functions” are used interchangeably by the FSB (e.g. FSB 2014a).
24
The EU state aid rules must be complied with before a state bailout is approved by the European Commission.
25
As indicated by the IMF, firm-level case studies suggest that the behaviour of smaller and weaker insurers warrants the
attention of supervisors (IMF 2016a), since they tend to take on relatively more risk.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 20
• Critical functions could be identified in RR planning and could be better protected in
resolution.
• Strengthening frameworks could encourage better preparation for crises and the early
implementation of preventive actions. For example, measures such as RR planning and
resolvability assessments could reveal potential contagion channels.
• A strengthened set of early intervention tools may help to avoid an insurer’s failure,
allowing national competent authorities to address a potential issue with the right tool,
which would benefit policyholders as well as financial stability. Moreover, an enhanced
set of resolution tools could help to avoid a liquidation in which it might not be possible to
preserve critical functions, and which could also lead to a potential destruction of value.
• The exchange of confidential information in times of crisis and the requirements of
coordination between the relevant authorities within the country during the resolution
process would be defined and governed by a strengthened legal framework.
43. Increased harmonisation in Europe could limit systemic risks of cross-border
contagion. Strengthening national frameworks might not be sufficient, given that insurance is
an international business, with large cross-border insurers operating in several EU Member
States. At the same time, the fact that a failing insurer would typically not need to be resolved
as quickly as a failing bank, offers some room for national discretion. While it is recognised
that the harmonised framework could provide a certain flexibility to address the national
specificities of local insurance sectors, some degree of harmonisation across the EU may help
to mitigate systemic risks and create a level playing field in the ways described below.
• A harmonised EU-wide RR framework would enable better evaluation of the implications
for other jurisdictions of any national measure taken. This would mean that the protection
of critical functions in one country would not adversely affect the critical functions in
another country.
• Harmonised RR plans would make it easier for authorities to compare measures planned
for distressed insurers across the EU and enable them to better analyse and mitigate
any spillovers to other sectors or across borders.
• A common set of rules is particularly relevant in the event of the distress of a large
international insurer conducting significant business in several EU Member States or in
the event of the simultaneous distress of several unrelated insurers in several EU
Member States.
• A strengthened legal framework applicable EU-wide would increase coordination and
cooperation mechanisms, including information sharing, across borders. Enhanced
cross-border coordination could prevent cross-border spillovers. Although coordination
arrangements for supervisory colleges, including emergency planning, are currently in
place,
26
appointed administrators or liquidators at national level or national resolution
26
See Annex 1.E of EIOPA Guidelines on the operational functioning of colleges (EIOPA 2014b).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 21
authorities for insurers are not required to coordinate their resolution planning and
actions across jurisdictions.
27
• Investors and customers would be treated more similarly in all countries, which could
contribute to a higher level of financial integration. This could increase trust in the
insurance sector and in the EU internal market.
44. This report recognises that a few ESRB members might hold a different view on some
aspects of RR frameworks for insurers. It is broadly recognised that RR frameworks for
insurers need to be strengthened across the ESRB membership. Moreover, several
authorities initiated or are considering measures to strengthen their national frameworks,
either due to recent experience or to the fact that low systemic risk of insurers in the past does
not guarantee the absence of systemic risk in the future (IAIS 2015). However, a few ESRB
member institutions take the view that it might not be cost effective to introduce an ambitious
RR framework for insurers. This reflects sparse evidence to date that traditional lines of
business of insurance have generated or amplified systemic risk within the financial system or
in the real economy and, that existing national frameworks have functioned properly so far.
Moreover, they are of the view that national frameworks could be reinforced without the
introduction of a comprehensive single EU-wide RR framework. Furthermore, since most
failures of EU insurers have not impaired the EU financial system to any great extent, and
given concerns over potential conflict between policyholder protection and financial stability,
some authorities prefer to place higher importance on the policyholder protection objective
than the financial stability objective in the RR framework for insurers, while recognising the
importance of both (ESRB 2017).
Box 2
Critical functions in the insurance sector
The concept of critical functions has been developed by the FSB as part of its RR
framework. Critical functions are “essential and systemically important functions” (FSB 2013) or
“vital economic functions” (FSB 2014a), for which continuity should be maintained in the resolution
process while avoiding unnecessary destruction of value.
28
Where such functions have been
identified resolution tools other than regular liquidation proceedings should be considered. The
reason for this is that safeguarding critical functions is not an objective of the regular liquidation
proceedings; the liquidation process could, instead, lead to the destruction of their value. Moreover,
a regular liquidation process could also include the possible disruption of payments to policyholders
or other financial institutions (FSB 2016).
The FSB considers criticality in the context of the impact of a failure on the financial system
and the possible substitutability of the failed insurer. According to the FSB, the provision of
27
Title IV of the Solvency II Directive (“Reorganisation and winding-up of insurance undertakings”) provides regulation in this
area in the EU. The regulation largely refers to the right of the competent authorities of the home EU Member State to take
action in line with national legislation, while keeping the supervisory authorities of other EU Member States informed of the
decisions. It should be noted, however, that cross-border cooperation in the form of crisis management groups and
cooperation agreements are foreseen, and partially implemented, in the EU as a result of global standards for G-SIIs;
although these would not necessarily apply to a population broader than the five EU GSIIs.
28
Critical functions, systemically important functions, essential functions and vital economic functions are used
interchangeably.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 22
insurance includes three key functions carried out by insurers which may be viewed as critical:
(i) writing new business; (ii) providing insurance cover for existing business; and (iii) making
payments to policyholders (FSB 2016b). Insurers also provide important non-insurance functions to
third parties, such as asset management. The criticality of all functions provided by insurers varies
across insurance products, insurers and jurisdictions. The FSB defines a function of an insurer as
critical if it satisfies all of the following conditions: (i) it is provided by an insurer to third parties not
affiliated to the firm; (ii) the sudden failure to provide that function would probably have a material
impact on the financial system and/or the real economy as it would give rise to systemic disruption
or undermine general confidence in the provision of insurance; and (iii) it cannot be substituted
within a reasonable period of time and at a reasonable cost (FSB 2016b).
The identification of critical functions is dependent on time and context. The impact of an
insurance failure could vary, depending on the conditions in the financial system and the economy
at the time of the failure. It is, therefore, difficult to identify all critical functions prior to a resolution.
For example, during the distress of an insurer it might not be possible to find a substitute insurer
within a reasonable period of time and at a reasonable cost. As a result of this, certain types of
insurance might become critical if a lack of substitutability of an insurer within a reasonable period
of time is likely to have a significant impact on the financial system and/or the real economy.
Similarly, if a reinsurer fails, primary insurers could face a situation where – for a considerable
period of time – they are not able to buy reinsurance for certain risks at a reasonable cost and, as a
result, they might stop underwriting these risks. The lack of substitutability might therefore justify
resolution measures aimed at recapitalising the insurer so it can continue to be able to write new
business, or resolution measures aimed at providing continuity in order to facilitate substitutability
over time. Furthermore, the FSB also recognises that the identification of critical functions involves
making a broader political and economic judgement, e.g. it might be necessary to preserve the
continuity of functions that are material to the real economy, e.g. in the event of disruption of cover
or payments to a significant number of policyholders, in particular in the life business (FSB 2016b).
It may be appropriate to consider critical functions from a broader perspective. Some
elements of the previous definition have not proved very helpful in identifying critical functions in the
insurance sector. It should be noted, especially in the current environment of low yields, that the
failure of an insurer does not occur suddenly, but gradually. Slow-burning distress can materially
impact the financial system and the real economy, undermining confidence. This type of failure
should therefore be taken into account in the RR framework. Furthermore, there may be other
important functions which may be worth examining in the context of insurance resolution, including
pension fund management and asset management. Lastly, it might be of interest to safeguard the
viable parts of an insurer.
Furthermore, the industry-wide definition of critical functions might prove more relevant in
the LIR environment. The FSB focuses on criticality stemming from the failure of a single entity.
As pointed out by the IMF, such an approach covers “domino” systemic effects, where the failure of
a single entity has systemic consequences for the broader economy (IMF 2016). Another
contribution to systemic risk is, however, through common exposures across firms that may
endanger financial intermediation in the system as a whole in the event of an adverse shock,
referred to as “tsunami” systemic effects. In the LIR environment, which raises the possibility of an
extreme severe double-hit scenario, no individual failing insurer might have a significant impact on
financial stability on its own, but if several insurers were to fail simultaneously they might
collectively produce a material systemic effect.
For the purpose of this report, a broader perspective is taken in respect of criticality, and the
report advocates further discussion of critical functions in the insurance sector. Within the
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Systemic risks and the case for an effective recovery and resolution framework for insurers 23
spirit of the FSB statement that criticality is not a binary concept and that there is a spectrum of
criticality (FSB 2013), the report goes beyond a narrow interpretation of the FSB definition of critical
functions in the insurance sector. A critical function, in the context of this report, is understood as
any function of an insurer or a group of insurers, which (if not provided) might have a significant
impact on the financial system or the real economy. On the back of these considerations, it is
recognised that the definition and the scope of critical functions merit further discussion, both at
European and global level, i.e. when a harmonised EU-wide RR framework is being developed, in
order to adequately cover all critical functions worth preserving, from both an individual-firm and an
industry perspective.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 24
3.1 Introduction
45. This section provides an overview of recent initiatives in both a global and a European
context. It describes the work of the IAIS and the FSB, with a special focus on the approach
to G-SIIs. It shows how some of the ideas developed for G-SIIs are being integrated into the
IAIS Insurance Core Principles (ICPs) for all insurers. Moreover, the section describes the
initiatives at European level, in particular by the COM and EIOPA. It would appear that, in the
absence of a harmonised approach, several national initiatives aimed at reinforcing national
RR frameworks are increasingly under consideration.
3.2 Global initiatives
46. At a global level, the IAIS and the FSB are the main driving forces establishing the
principles of recovery and resolution (RR) in the insurance sector. The IAIS is the global
standard-setter for the insurers. In addition, it participates in a global initiative, under the
purview of the FSB and the G20, to identify global systemically important insurers (G-SIIs),
whose failure or distress, it has been recognised, would have financial stability implications
due to their size, market importance or global interconnectedness. The IAIS developed the
initial assessment methodology as well as policy measures to be applied to G-SIIs. The
designation of G-SIIs, however, is the responsibility of the FSB.
47. The RR standards envisaged by the IAIS are currently structured in three layers. They
differ according to the importance of insurers in terms of their size, the intensity of their cross-
border activities and their systemic importance (Table 1). The additional layers dealing with
complex undertakings were developed following the 2008 financial crisis. The near collapse of
a few large financial conglomerates (including cross-border insurers) during the global
financial crisis led the IAIS to publish a proposal for the resolution of cross-border entities and
groups in 2011, which is currently being discussed as part of ComFrame,
29
the framework for
internationally active insurance groups (IAIG). Moreover, the G-SII package was released in
2013 and provides a broader list of policy measures for RR, although these are applicable to
designated G-SIIs only.
48. The work to harmonise the approaches to the G-SII standards is ongoing. Both the
relevant IAIS Insurance Core Principle for RR, namely ICP 10
30
(for recovery) and ICP 12
31
(for resolution) are currently under revision. Moreover, work is ongoing to integrate ComFrame
for IAIG into ICPs, including the M3E3 element (for both recovery and resolution).
32
As such,
29
The Common Framework for the Supervision of Internationally Active Insurance Groups (ComFrame). It is also expected
that the ComFrame will later be integrated into the ICPs.
30
ICP 10 on preventive and corrective measures.
31
ICP 12 on winding-up and exiting from the market. This was originally referred to as ICP 16 and was first published in 2006.
32
In August 2016, IAIS released a draft Module 3 Element 3 (so called “M3E3”) part of ComFrame, which deals with RR.
Section 3
International and European initiatives on recovery and
resolution for insurers
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 25
the M3E3 will no longer exist as a standalone standard for IAIG, although the considerations
with regard to IAIG will be kept. The current ICP 12 requires the national authorities to have
the liquidation option available only for failing insurers. Moreover, this standard applies
primarily to individual insurers and does not address an insurer’s failure from a cross-border or
group perspective. The revised ICP 12 standard is expected to be approved by the end of
2017 and should include new aspects of the RR framework, thereby making it more consistent
with the approach for G-SIIs.
Table 1
International standards on RR
Individual Insurer (group)
IAIG
G-SII
ICPs
ICP 10 (for recovery) and ICP 12 (for resolution, both ICPs are under review)
ComFrame
M3E3 (under review)
G-SII package
G-SII measures
Source: ESRB.
49. Financial stability considerations have caused a broadening of RR standards. In
general, the protection of policyholders is viewed as a priority in winding-up. In ICP 12 (as it
currently stands), the winding-up procedure is understood as a tool that minimises disruption
to the timely provision of benefits to policyholders. Since the crisis, however, focus has shifted
more towards financial stability considerations, in the recognition that the distress or failure of
some insurers might cause a dislocation in the global financial system, with adverse economic
consequences across a range of countries.
33
50. Winding-up and run-off is still appropriate for traditional business lines with limited
cross-border implications, but changes in market structure call for new policy tools to
be considered. According to ICP 12, all jurisdictions should introduce necessary procedures
into their legislation for the exit of an insurer from the market, giving the supervisory authority
the power to close an insurer. However, experience of the recent crisis has led the IAIS to
note that the tools available to insurance supervisors to resolve insurers with cross-border
business activity have not kept pace with the evolution of the groups themselves in terms of
their complexity and their geographical and cross-sectoral interconnectedness (IAIS 2011b).
For this reason, the future ICP 12 will be more consistent with the approach taken in respect
of G-SIIs. The standard will stipulate that a series of tools – such as run-off, portfolio transfer,
writing down or restructuring liabilities including insurance liabilities – should be available to
the resolution authority, provided their use respects the no-creditor-worse-off (NCWO)
principle.
51. With so little global guidance, several issues might arise in dealing with a cross-border
resolution. The IAIS notes that there is currently no international insolvency framework for
insurers (IAIS 2011b). Moreover, relations in the resolution process between viable and non-
viable entities, as well as relations between insurance and non-insurance entities, including
cross-sector regulated entities and non-regulated entities (such as holding companies), raise
33
In line with the FSB definition of G-SIIs.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 26
further issues for dealing with a resolution in a group-wide context. As a reaction, the future
ICP 12 will also comprise standards on (international) cooperation in resolution.
52. The work on G-SIIs builds upon the FSB Key Attributes (KAs) of effective resolution
regimes. First published in October 2011, the KAs were updated in 2014 with additional
guidance, including insurance sector-specific guidance (Annex II). The KAs set out the core
elements of an effective national resolution regime for systemically important financial
institutions (SIFIs) that does not rely on taxpayers’ funds. In June 2016, the FSB provided
further guidance on developing resolution strategies for systemically important insurers,
specifically in respect of determining a preferred resolution strategy based on an analysis of
insurers’ business models, the criticality of insurers’ functions and policyholder protection
arrangements, as well as elements for making the resolution strategy feasible and credible.
These include, among other things, the capacity to absorb losses in resolution (FSB 2016,
§ 3.6).
53. The policy measures for G-SIIs include enhanced supervision, effective resolution and
higher loss absorption capacity. The IAIS is working towards applying these measures to
G-SIIs, reflecting the greater risk that these institutions pose to the global financial system.
The measures are designed to reduce moral hazard and the negative externalities that would
stem from the potential disorderly failure of a G-SII. In particular, they aim at reducing the
probability and impact of a failure of G-SIIs, incentivising G-SIIs to become less systemically
important, and giving non-G-SIIs strong disincentives to become G-SIIs. Effective resolution of
G-SIIs includes: the establishment of crisis management groups (CMGs); the preparation of
RR plans, including a liquidity risk management plan and a systemic risk management plan;
the carrying out of resolvability assessments; and the adoption of institution-specific cross-
border cooperation agreements. Greater loss absorption capacity is intended to internalise
some of the costs to the financial system and the overall economy, make G-SIIs more resilient
to low-probability but high-severity events, and act as a disincentive to carrying out activities
that pose a threat to the financial system.
54. Five insurers headquartered in four EU Member States have been identified as global
systemically important insurers (G-SIIs). G-SIIs are designated on an annual basis by the
FSB, which has identified nine insurers as G-SIIs since 2013, of which five are headquartered
in the EU (Table 2). Jurisdictions with G-SIIs are expected to apply the agreed policy
measures (outlined in the previous paragraph) to their respective G-SIIs.
Table 2
List of European G-SIIs
Aegon N.V. (NL, as of 2015) Allianz SE (DE)
Aviva plc (UK) Axa S.A. (FR)
Prudential plc. (UK)
Source: FSB.
Note: Another European insurer ‒ Assicurazioni Generali S.p.A. (IT) ‒ was designated as a G-SII in 2013 and 2014.
55. The IAIS updated its methodology for the identification of G-SIIs in mid-2016. While the
assessment methodology used to identify G-SIIs also applies to reinsurers, all global
reinsurers have so far been below the threshold designated by the IAIS. The IAIS released an
updated G-SII assessment methodology in mid-2016, which will be used by the FSB to
identify insurers as G-SIIs as of 2017, using end-2016 data (IAIS 2016c). The new
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 27
methodology aims at better capturing the heterogeneity across the industry by the structured
use of both quantitative and qualitative elements, including the use of absolute reference
values for the reinsurance-related indicator, in order to effectively measure the global systemic
importance of reinsurance activities and related risks. Furthermore, a reinsurance supplement
assessment has been introduced to provide an in-depth analysis of insurers with significant
third-party reinsurance activities. Moreover, as is the case in the banking sector, domestic SIIs
could be established in individual jurisdictions under domestic regulatory arrangements, and
may follow different definitions and timelines, although this approach has not yet been
adopted in the EU.
56. Insurance guarantee schemes (IGSs), also known as policyholder protection schemes,
provide a minimum level of compensation to policyholders and the IAIS has recognised
that they can also contribute to an orderly resolution. The IAIS published a paper in 2013
summarising the main guidance and core principles relating to this issue (IAIS 2013). It
recognised that IGSs could contribute to the objectives of resolution by facilitating the
continuing provision of insurance, providing financial support to an entity intending to
purchase an insolvent insurer (or to which insurance policies would be transferred), or by
working as a bridge institution. As such, IGSs contribute to maintaining public confidence and
stability in the insurance sector and in the financial system in general.
Box 3
Insurance RR in light of other sectoral RR initiatives
The KAs have been used as the main reference for RR frameworks covering financial
institutions of all types whose failure could be systemic. They were first published in 2011 and
updated in 2014 to accommodate sector-specific considerations. Annex 2 provides an example
dealing with the insurance sector. The sector-specific guidance recognises that not all KAs are
equally relevant and might need to be adapted for effective implementation in a particular sector.
The sector-specific guidance therefore sets out how KAs should be understood in a sector-specific
context.
Global standards for RR are being implemented at different speeds. The banking sector is a
frontrunner, with the policy framework being implemented in most FSB member states, in particular
in G-SIBs’ home jurisdictions. The first G-SIIs were identified in 2014, and the policy framework
was slowly implemented throughout FSB member states. Recently, the focus of the work has
shifted towards developing RR standards for central counterparties (CCPs).
Furthermore, the EU approach to RR for insurers differs from that taken in other sectors.
The KAs were developed for G-SIFIs, yet the EU has expanded the RR framework to all credit
institutions in the EU, including third-country subsidiaries. The EU RR framework for the banking
industry has been fully operational since 2015, and includes a single resolution authority. Similarly,
the EU recognises all EU CCPs as being systemically important and, as such, the European
Commission’s proposal regarding the RR of CCPs, which was published at the end of 2016, has
been expanded to all CCPs domiciled in the EU. In both cases the EU has been seen as a first
mover, releasing its first proposals for RR frameworks while the global standards were still being
developed. In contrast to the banking sector and financial market infrastructures, no legislative
initiative at EU level has so far been undertaken in respect of RR standards for insurers.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 28
3.3 EU initiatives
57. The European Commission (COM) and EIOPA are the main driving forces in respect of
the insurance sectors at EU level. Crisis prevention and crisis management are two of
EIOPA’s key responsibilities.
58. The COM initiated a discussion on insurance RR, but no concrete proposals have been
followed up yet. In 2012, the COM launched a consultation on the possible framework for the
RR of non-bank financial institutions, including primary insurers and reinsurers (European
Commission 2012). The aim of this consultation was to gather the views of stakeholders as to
the possibility of developing an RR framework in Europe. In 2013, the Expert Group on
Banking, Payments and Insurance (EGBPI) was set up as a consultative body whose role was
to help the COM prepare draft delegated acts in the area of banking, payments and
insurance.
34
An exchange of views on insurance RR took place at the EGBPI meeting in
March 2015. At that time there seemed to be no consensus among EU Member States on the
need for an EU-level regulatory initiative on this matter (European Commission 2015 and
EGBPI 2015). Furthermore, in 2016, Jonathan Hill, then Commissioner for Financial Stability,
Financial Services and the Capital Markets Union, stated that new EU regulation on that
matter was not considered to be a priority at that time.
35
59. In November 2014 EIOPA adopted a set of Sound Principles to help National Competent
Authorities (NCAs) in developing their crisis prevention, management and resolution
frameworks (EIOPA 2014a). The stated overarching goal of EIOPA’s initiative was to support
the stability of the EU financial system and the protection of policyholders. The principles are
consistent with the FSB’s KAs and represent a repository of sound principles in the field of
crisis prevention, management and resolution with regard to insurance, arguing, for example,
that RR plans should be drafted, at least for systemically important insurers, in a pre-emptive
and proportionate way.
60. The analyses, conducted by both the COM and EIOPA, have noted a remarkably high
degree of heterogeneity among EU Member States regarding their RR frameworks and
IGSs, which could inhibit cross-border cooperation in the event of a failure. The series
of surveys were conducted with the aim of reaching a better understanding of the situation in
EU Member States. In 2010, the COM published a White Paper on IGSs, accompanied by an
impact assessment. This was followed by an EIOPA report on cross-border cooperation
mechanisms between IGSs in the EU, published in 2011 (EIOPA 2011). With regard to crisis
management, EIOPA conducted its first survey on crisis prevention, management and the
resolution preparedness of NSAs in 2013 (EIOPA 2013) and, more recently, it published a
discussion paper on potential harmonisation of recovery and resolution frameworks for
insurers (EIOPA 2016b), followed by an EIOPA opinion on the same subject (EIOPA 2017),
which both included a comprehensive overview of national RR frameworks for insurers.
61. The ESRB and EIOPA have pointed out that EU improvements plus harmonisation of
the RR and IGS frameworks for the insurance sector could help, also in light of the risk
of a prolonged LIR environment. The ESRB stated in early 2016 that IGSs and RR
34
The EGBPI is a consultative entity composed of experts appointed by EU Member States in order to provide advice and
expertise to the Commission and its services in the preparation of draft delegated acts in the area of banking, payments
and insurance
35
European Commission – SPEECH/16/274, Extract from speech by Commissioner Jonathan Hill
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 29
arrangements at national level may not be sufficient to handle episodes of widespread
financial stress involving the simultaneous failures of life insurers in several countries (ESRB
2015). In its strategy paper on macroprudential policy beyond banking, the ESRB stressed
that a proposal for an RR framework for insurers should be a priority from a macroprudential
perspective (ESRB 2016a). Recently, against the background of the protracted LIR
environment, the ESRB has established that further improvements and harmonisation of RR
frameworks for insurers could help facilitate an orderly exit from the market for insurers who
are unable to adapt their business models (ESRB 2016b). Furthermore, EIOPA‘s discussion
paper provides, inter alia, a rationale for harmonisation and puts forward preliminary ideas for
the building blocks of a minimum harmonised RR framework for insurers in the EU (EIOPA
2016b), which were further elaborated in EIOPA’s opinion (EIOPA 2017).
Box 4
RR framework and financial conglomerates
In the EU, a Directive governing the supervision of financial conglomerates supplements the
relevant sectoral legislation. A typical large conglomerate has over 400 licences over several
jurisdictions and several sectors, while the biggest conglomerates may have over 900 legal entities
or licences (European Commission 2010c). It had already been recognised by 1999 that large and
complex groups, which combine licences in various sectors of the financial system, need additional
supervision so they can effectively monitor, ring fence, limit and influence activities which may
spread risks between the entities of a financial conglomerate (European Commission 2010c). The
objective of this supplementary supervision was to control for the risk of double gearing (i.e.
multiple use of capital) and risks which are not sector specific, such as the risk of contagion,
management complexity, concentration and conflicts of interest which arise when licences for
different financial services are combined in a single conglomerate. The Financial Conglomerates
Directive (FICOD)
36
was the EU’s response to this need, supplementing sectoral legislation.
The recent crisis has shown the difficulty of resolving a troubled conglomerate. The near
failure and bailout of the American International Group (AIG), a US-based insurance-led
conglomerate, is the most prominent example. The near failures of the ING Group, the Aegon
Group and SNS Reaal, all financial conglomerates based in the Netherlands, are also worth
mentioning. All three provided bancassurance, ING being a bank-led, Ageon an insurance-led and
SNS Reaal a symmetric financial conglomerate (IAIS 2011a). In 2004 AIG was the largest insurer
in the world (IAIS 2011a), but it came under extreme pressure during the financial crisis due to its
involvement in non-traditional non-insurance (NTNI) activities, primarily through a non-insurance
subsidiary specialising in over-the-counter derivatives trades. Simultaneously, AIG’s life insurance
subsidiaries also experienced significant liquidity and capital stress as a result of the massive
losses it incurred in the securities lending program and from intra-group investments. With regard to
the Dutch financial conglomerates, bank entities faced acute liquidity problems due to the overall
loss of confidence in financial markets, while insurance entities suffered large losses on their
investment portfolios. Moreover, there were also signs of reputational effects impacting the groups.
36
Directive 2002/87/EC of the European Parliament and of the Council of 16 December 2002 on the supplementary
supervision of credit institutions, insurers and investment firms in a financial conglomerate (FICOD).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 30
Both the US and the NL governments concluded that the potential failure of their respective
conglomerates would create systemic risks and decided to bail out the groups instead.
37
No regulation specifically addresses the resolution of financial conglomerates as a whole in
the EU. Instead, sectoral legislation is applied if a financial conglomerate fails. The BRRD applies
fully to bank-led conglomerates, and in those cases the group-level resolution authority would be a
banking resolution authority.
38
This authority, together with the resolution authorities of subsidiaries
within the scope of the BRRD, would have the power to draft the group resolution plan. The group
recovery plan should, inter alia, include a detailed description of all financial, operational and
commercial interconnections within the conglomerate and should suggest options to ensure the
continuity of shared services, e.g. between the bank entity and the insurer entity. Moreover, it would
need to take into account the specificities of the different types of business lines covered. So far,
only the banking RR regulation has been harmonised across the EU, while national RR regulations
apply to the insurance entities of a financial conglomerate. For internationally active insurance-led
conglomerates, the creation of a harmonised RR framework for insurers in the EU would have the
advantage of enhancing the consistency of resolution regimes across sectors and jurisdictions.
3.4 Current status and recent initiatives at national level
62. There are significant differences across the EU. Based on information provided by 30
NSAs, EIOPA’s results reveal that there are major differences in respect of the RR measures
available across EU Member States (EIOPA 2016b and EIOPA 2017). Besides taking stock of
the available powers and tools, it also captured different stages of crisis management and it
included questions on cross-border cooperation and coordination and IGSs. The main results
are summarised in this section.
63. In general, the authorities make limited use of pre-emptive RR plans as there are
constraints on the early intervention powers available to national authorities. The use of
recovery and resolution planning and resolvability assessments is currently essentially limited
to designated G-SIIs – their expansion to IAIGs is at an early stage. Some early intervention
powers are common throughout the EU, such as the power to require insurers to make
additional provisioning and the power to require the removal of members of the management
body, directors or managers (EIOPA 2016b and EIOPA 2017). Other, more intrusive powers,
such as the power to require the sale of subsidiaries and the power to require the transfer of
financing operations to the parent company, are much less common. Furthermore, in many
cases the exercise of the powers is subject to a variety of conditions or prerequisites (e.g. a
breach of the Solvency Capital Requirement (SCR)), which may inhibit supervisors’ ability to
intervene at a sufficiently early stage to prevent the disruption of an insurer.
37
It should be noted that the insurer subsidiaries of bank-led conglomerates were among the beneficiaries of the bailout,
receiving around 30% of state support in the case of AIG and ING and 60% in the case of SNS Reaal (IAIS 2011a). It is a
positive outcome that all insurers repaired their state loans, however the indirect costs related to this exercise have not yet
been calculated (e.g. in terms of the costs of personnel or the costs of the winding-down of some parts of these companies,
which is still ongoing).
38
Articles 12.1 and 3 of the BRRD.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 31
Chart 4
Resolution tools in the insurance sector
Source: EIOPA (2017)
64. Significant differences were also identified in the resolution approaches in the different
EU Member States. These differences refer to the issues below.
• Resolution authorities. Only two EU Member States have a designated resolution
authority for insurers. The relevant NSA and/or a relevant ministry are usually in charge
of the winding-down of insurers.
• Objectives of the resolution. In most EU Member States the national framework does not
set out specific objectives for resolution, but specifies instead more general objectives
for the NSA to pursue. In most countries the main objective is the protection of
policyholders, followed by maintaining financial stability.
• Entry into resolution. In most EU Member States there are no specific conditions for
entry into resolution. In general, the approach is based on the conditions for winding-
up/liquidation and/or those related to a breach of the Solvency II requirements (EIOPA
2016b and EIOPA 2017).
• Resolution powers. Traditional tools, such as the withdrawal of authorisation, putting the
company into run-off or transferring the portfolio, are widely available. However, other
resolution powers (such as the transfer of reinsurance contracts, measures affecting
policyholders’ rights in relation to insurance contracts or the restructuring of
(re)insurance liabilities), are not common in the EU (see Chart 4).
2
2
3
2
3
3
5
6
7
12
15
12
21
30
1
1
3
3
4
3
3
6
2
2
10
3
25
25
25
23
22
20
20
19
15
14
11
7
4
write down liabilities (other than (re)insurance liabilities) and to convert
them into equity (‘bail-in’ of shareholders and creditors)
impose separation of toxic assets from good assets by establishing a
vehicle for managing and orderly running-down those assets
create/operate a bridge institution to which the business of a failing
insurer is transferred
restrict/suspend (reinsurers) rights in relation to reinsurance contracts to
terminate or not to reinstate reinsurance cover
vary/reduce value of contracts transferred in a portfolio transfer
impose a temporary stay on early termination rights exercisable under
financial contracts
restructure, limit or write down (re)insurance liabilities, and allocate
losses to policyholders
restrict/suspend (policyholder) rights in relation to insurance contracts to
withdraw from contracts
transfer reinsurance associated with transferred policies
request court to appoint liquidator to manage liquidation of insurer
appoint liquidator to manage liquidation of insurer
transfer insurer’s business to private purchaser and override any
transfer restrictions (e.g. policyholder consent)
direct insurer to refrain from writing new business and require fulfilment
obligations for in-force business (run-off)
withdraw authorisation of insurer
available
available but restrictions apply
not available
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 32
• Safeguards. The resolution of insurers is subject to the no-creditor-worse-off (NCWO)
39
principle in ten EU Member States, and that four EU Member States’ national authorities
also make reference to the ability to depart from the pari passu principle in order to
maximise value for all creditors, including policyholders, as a whole (EIOPA 2016b and
EIOPA 2017).
65. There is also significant heterogeneity across the EU in respect of resolution funding
and insurance guarantee schemes (IGSs). Only Romania has set up a resolution fund.
Moreover, while bank depositors are covered by up to €100,000 in the event of the failure of a
bank, according to the EU Directive on deposit guarantee schemes
40
, European policyholders
do not have a similar harmonised level of protection. According to the COM’s White Paper on
IGSs, only 12 EU-EEA countries operate one (or more) general IGSs, leaving around 26% of
all life insurance policies and 56% of all non-life insurance policies unprotected (European
Commission 2010a). More recently, EIOPA showed that this number has increased, although
many of these have very limited coverage (EIOPA 2016b and EIOPA 2017). The IGSs also
differ in scope, operational procedures, funding arrangements and main functions. In addition
to the compensation of policyholders for losses in the event of liquidation, some could also
fund the transfer of an insurer’s portfolio to a bridge institution or other insurer (thus staying
closer to the functions of a resolution fund). For further analysis, please refer to Section 5.
66. The resilience of IGS frameworks in the EU has not been tested for the failure of large
insurers. The ESRB notes that current schemes in the EU are appropriate in the event of the
failure of small insurers partly due to the long-term nature of insurers’ winding-up proceedings
given their long-term liabilities (ESRB 2015). However, the resilience of current IGSs has not
been tested by the failure of a large insurer, or by the simultaneous failure of several large
insurers.
67. There are generally no specific cross-border coordination and cooperation
arrangements for crisis situations. Apart from the designated G-SIIs, there are often no
(formal) crisis management groups or other equivalent arrangements in place, and the
existing supervisory college structures act as a partial substitute. In addition, in a crisis a
distinction is made between domestic and cross-border business. For example, the way
existing IGSs are designed means that domestic insurance business activity is better
protected than cross-border business activity via branches. The impact assessment
accompanying the Commission White Paper found that while a third of domestic activity is not
protected, the proportion rises to just under half for cross-border activity (European
Commission 2010b).
68. Some NSAs reported gaps and shortcomings in their existing frameworks. Almost half
of the NSAs reported shortcomings in the existing legal framework, including resolution
powers and early intervention powers (EIOPA 2016b and EIOPA 2017). A few NSAs reported
that they plan to reinforce their national RR frameworks. The initiatives are at different
stages – some EU Member States, such as the Netherlands and France, have already
formulated a specific proposal for reforming the national framework, and Romania has already
39
From the FSB Key Attributes: “Creditors should have the right to compensation where they do not receive at at least what
they would have received in the liquidation of the firm under the applicable insolvency regime”.
40
Directive 2014/49/EU of the European Parliament and of the Council of 16 April 2014 on deposit guarantee schemes.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 33
introduced a RR framework for insurers. These three examples of national initiatives are
shown in Box 5.
69. So far there has been little or no consensus or political support for harmonising the
frameworks in the EU. The Commission’s White Paper recognised that most of the problems
stemming from the existence of different national legal frameworks could be resolved by the
establishment of a single EU-wide IGS framework covering all life and non-life policies written
and purchased within the EU. It also acknowledged that harmonising the RR framework
throughout the EU would not only increase the availability of RR tools, but would also provide
a common basis for dealing with any cross-border failure. So far, however, this idea has not
received the necessary political support (EGBPI 2015).
Box 5
A comparison between RR frameworks in Romania, the Netherlands and France
The Romanian RR framework for insurers is the first national framework of its kind in
Europe, while France and the Netherlands are in the process of strengthening their
respective RR frameworks. The Romanian RR framework was set up in 2015 through an
amendment of the prevailing national law concerning insurance and the insurance guarantee fund.
The framework is fully operational and developed to secondary legislation. In France, the national
resolution authority for insurers has been designated,
41
and the RR legislative framework for
insurers is expected to be adopted in 2017. In addition the Netherlands is finalising a legislative
proposal which seeks to improve the resolution framework for insurers, as well as the rules
governing insolvency and winding-up proceedings.
These three countries decided that, in the event of the failure of an insurer, the existing
tools and procedures would not be sufficient. In Romania, the need for an RR framework was
triggered by the difficulties faced by two large insurers in 2014 and 2015, while both the
Netherlands and France experienced near-to-failure cases of financial conglomerates during the
global financial crisis.
42
Their experience showed that in a situation of widespread financial stress
the scope for recovery is limited, the court-led winding-up procedures are not always efficient, and
the transfer of portfolios is problematic so, as a result, government intervention is required.
Moreover, policyholders are affected by the cessation of payouts during resolution, as is particularly
the case in long winding-up proceedings. Large insurers pose additional problems, as they cannot
be easily absorbed by other insurers and their portfolios cannot be readily transferred. Reinforcing
RR planning therefore seemed appropriate for these entities. Added to the economic background of
a low interest rate, low yield environment, and coupled with decreasing premium income from new
individual life insurance business, the two authorities decided to reinforce their RR frameworks for
insurers.
The protection of policyholders and maintaining financial stability are the objectives of the
respective RR frameworks in all three countries, although there are differences in the design
of the frameworks, the resolution tools and funding arrangements. All three frameworks
envisage strengthening the RR planning and resolution process, although the Romanian framework
41
Autorité de contrôle prudentiel et de résolution (ACPR) has been designated as the national resolution authority for
insurers, following its designation as the resolution authority for banks.
42
E.g. ING Group, AEGON Group and SNS Reaal in the case of the Netherlands and Dexia in the case of France.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
International and European initiatives on recovery and resolution for insurers 34
also includes changes to early intervention powers and the Dutch proposal foresees improved
insolvency and winding-up proceedings. In terms of resolution tools, sale of activity, portfolio
transfer and a bridge institution are common to all three frameworks. In Romania, the resolution
authority also has an asset separation tool at its disposal, while the Dutch proposal foresees the
possibility of a bail-in without portfolio transfer, as well as the option to amend policyholder rights in
all scenarios. The French proposal seeks to expand administrative policy measures and determine
the conditions of remuneration of managers. It does not include any measures amending
policyholder rights beyond those already included in current national legislation, i.e. the possibility
of reducing life technical provisions when transferring a portfolio. In terms of resolution funding, the
Romanian regime relies on the combination of an IGS and a resolution fund (RF), which is collected
ex ante by the industry. The proposed Dutch framework would use an RF financed ex post, which
would serve the more limited purpose of setting up a bridge institution and compensating
policyholders if they are deemed to be worse off than they would have been under insolvency. The
French regime does not plan to introduce an RF, although it has left the existing IGS untouched.
Table 3 below summarises the comparison.
Table 3
Comparison of Romanian (existing), Dutch and French (both proposed) RR frameworks
Romania
The Netherlands
France
Structure
• RR planning
• Early intervention
• Resolution
• RR planning
• Resolution
• Improved insolvency and
winding-up proceedings
• RR planning
• Resolution
Resolution tools
• Sale of activity and portfolio
• Bridge institution
• Transfer to asset management
vehicles (asset separation tool)
• Sale of activity and portfolio
• Bridge institution
• Bail-in without portfolio transfer
• Amendment of policyholder
rights
• Sale of activity and
portfolio
43
• Bridge institution
• Administrative policy
measures
• Determination of the
conditions of renumeration of
the managers
Funding
• Resolution fund (ex ante until
threshold
• IGS
• Resolution fund (ex post)
• IGS
Source: ESRB.
It is difficult to address the problem of cross-border resolution in the absence of a European
regime. A resolution measure available in one of the three EU Member States cannot be applied to
a group subsidiary in another EU Member State if it is not available in the host authority. The
possibility of group resolution is therefore limited to national regimes, although the French RR
regime includes some measures in this respect. Moreover, none of the three frameworks extends
its powers to the incoming branches of insurers from other jurisdictions, and frameworks do not
specifically address the protection of policyholders outside their jurisdictions who hold contracts
with domestic insurers.
44
All in all, this could still lead to disorderly failures and could also have
financial stability implications.
43
The current French Law (Code des Assurances - Article L423-2) already allows the ACPR to amend policyholders’ rights in
the case of a transfer of portfolios before the withdrawal of the insurer’s license.
44
All national frameworks should reflect the principle of Article 160 of the Solvency II Directive of equal treatment of contracts
underwritten under the freedom to provide services or through branches and with no discrimination against foreign insured
persons and beneficiaries.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 35
4.1 Introduction
70. This section provides an overview of the possible recovery and resolution (RR) tools
available to the insurance sector. Recovery, early intervention and resolution tools are
covered by this section, which also discusses the different plans that could be developed to
increase awareness of the RR measures that may be adopted once a situation has
deteriorated. It is worth noting that most of these tools and pre-emptive plans were developed
during work on global systemically important institutions (G-SIIs), including internationally
active insurance entities. Unlike in the banking sector, there is no harmonised framework to
prevent and manage failures of insurers in the EU, and different tools are currently available at
national level. This section discusses the advantages, implications and challenges of different
tools and plans in respect of the insurance sector.
71. The specificities of the insurance sector need to be clearly reflected by the design of an
RR framework. First, the probability of a run on an insurer is lower than it is of a run on a
bank. Resolving an insurer is therefore less time-critical than resolving a credit institution.
45
Second, there is generally a high level of substitutability in the insurance sector, although the
situation could be different if several insurers were to face difficulties simultaneously (see also
the discussion of critical functions in Box 2). Third, the protection of policyholders is a
recognised objective of the RR framework for insurers, although this does not exclude the
possibility that policyholders might absorb some losses (FSB 2014a). An effective insurance
guarantee scheme could address this issue.
72. In the RR cycle, identifying triggering events is important. Insurers (for recovery) and
competent authorities (for early intervention and resolution) should be required to apply the
respective recovery, early intervention and resolution tools based on a triggering event. For
example, resolution tools could only be applied if an insurer is no longer viable or likely to be
no longer viable, and has no reasonable prospect of becoming so.
46
The issue of triggers for
entering recovery, early intervention or resolution is not harmonised at this time. This leads to
different approaches across the EU, which may create uncertainty, particularly in respect of
cross-border insurers (EIOPA 2016b and EIOPA 2017). Harmonising the RR framework with
the ladder of intervention under Solvency II is therefore important.
4.2 Recovery and resolution planning
73. The timely application of recovery, early intervention and resolution measures requires
planning and the existence of appropriate legal powers. Even though time may be less
critical for the insurance sector than it is for the banking sector, it is still essential to address
45
The terms for banks and credit institutions are used interchangeably throughout this report.
46
Based on EIOPA’s Discussion Paper, this trigger point usually refers to a significant breach of the Solvency II requirements
as a trigger initiate the resolution of an insurer (EIOPA 2016b).
Section 4
Recovery and resolution powers and tools
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 36
potentially severe financial problems at an early stage in order to be able to reduce the
likelihood of failure.
74. Recovery and resolution plans provide a guide if the financial position of an insurer
deteriorates. A recovery plan sets up a range of viable options the insurer could take if
financial stress occurs, although the insurer would not yet meet the conditions for resolution. If
the insurer implements these recovery measures there should be a reasonable probability of
recovery.
47
The resolution plan recommends the effective use of resolution tools by the
resolution authority in order to resolve the failing insurer, while protecting policyholders and
preventing severe disruption to the financial system and the real economy. The recovery plan
is drawn up by the insurer and assessed by the supervisor, whereas the resolution plan is
drafted by the resolution authority.
75. Ideally, both recovery and resolution plans should be drafted ex ante. The RR plans
could be pre-emptive (ex ante) or ex post. Pre-emptive RR plans have been introduced in
global standards for G-SIIs and it is likely that this requirement will be extended to IAIGs in
2020. Moreover, pre-emptive RR plans are also seen as having sound principles by EIOPA
(EIOPA 2014a). Pre-emptive RR plans are currently required in only a small minority of EU
Member States – mostly those where G-SIIs are located. Ex post recovery plans are required
under Solvency II when an insurer no longer complies with the Solvency Capital Requirement
(SCR). In contrast to global standards, which are of a “soft” law nature, Solvency II rules are
binding across the whole of the EU.
76. A requirement to develop and maintain a credible pre-emptive recovery plan would
raise awareness of possible adverse scenarios before they actually occur. A plan of this
type would complement the Own Risk and Solvency Assessment (ORSA) prepared under
Solvency II. While the ORSA also takes a forward-looking perspective and considers the
future capital needs of an insurer in light of its business strategy and risk profile, it does not
necessarily identify detailed possible adverse scenarios and available recovery options,
including governance arrangements and appropriate recovery indicators that act as triggers.
48
During the pre-emptive recovery planning process, insurers are able to gauge their recovery
capacity in terms of their ability to withstand a range of adverse scenarios of an idiosyncratic
or a systemic nature. Developing and maintaining credible and consistent recovery plans
would force insurers not only to analyse possible adverse scenarios in detail, but also to
ascertain what kind of recovery measures are available.
77. It would also raise supervisors’ awareness of potential herding behaviour in respect of
common recovery strategies which could cause systemic risks. The supervisory
authority would review the plans as part of the supervisory process that assesses their
credibility. By reviewing plans across the sector, supervisors would become aware of insurers
being herded into strategies that might be optimal when considered in isolation, but could
become a systemic risk if pursued by many insurers at the same time. For example, multiple
insurers triggering their recovery options by divesting the same assets simultaneously could
have a much greater impact than if this action were undertaken by just one insurer.
Cooperation with macroprudential authorities might be relevant to addressing such findings.
47
FSB Key Attributes, I-Annex 4 – Essential Elements of Recovery and Resolution Plans.
48
FSB Key Attributes, 11 – Recovery and resolution planning.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 37
78. Similarly, pre-emptive resolution plans could be used as effective tools by authorities
in charge of resolution. These plans are tailored to insurers and take all their specificities
into account, including their complexity, interconnectedness, level of substitutability and size.
They also address any potential impediments to an effective and timely resolution, and identify
responsibilities across authorities. From a macroprudential perspective, resolution planning
assists the early identification of the likely impact of an applied resolution tool on the wider
market, including its impact on the stakeholders who will bear the costs of resolution. It also
raises awareness of possible implications for the real economy and the wider insurance
sector.
79. This measure should be applied in accordance with the principle of proportionality.
Supervisory authorities should be given flexibility in respect of the scope, content and level of
detail of the pre-emptive plans, in accordance with the principle of proportionality. This would
avoid excessive burdens, e.g. in terms of the costs of implementation, on both insurers and
supervisors.
4.3 Recovery measures and early intervention tools
80. Recovery measures and early intervention tools are intended to be implemented when
institutions face financial stress, to avoid having to resort to resolution. The rationale for
this is that both financial stability and consumer protection are normally best served if the
situation can be restored without resorting to resolution.
49
81. Recovery measures would be implemented by the insurer in line with the agreed
recovery plan. Examples of recovery measures include, for instance, the sale of subsidiaries
to reduce the risk profile, the cancellation of dividends, and a capital increase (e.g. rights
issue).
82. On the other hand, early intervention measures would be taken by the supervisory
authority responsible. In light of this, the supervisory authority should have the appropriate
legal powers to intervene prior to putting an individual insurer into resolution. Examples of
early intervention measures include: (i) requiring changes to the business strategy, or the
legal or operational structure; (ii) requiring the removal of members of the senior management
and management body; (iii) requiring the company to create additional provisioning or
reserves, and (iv) appointing a temporary administrator.
50
83. Many early intervention measures for the insurance sector are already available in EU
Member States, although in practice this availability is subject to restrictions. As a
general rule, under the Solvency II framework supervisors are required to take all measures
necessary to safeguard the interests of policyholders if the financial situation continues to
deteriorate following non-compliance with the SCR.
51
EIOPA shows that, although early
intervention powers are widely available across the EU, even before a breach of the SCR has
49
Applicable in cases where resolution is considered necessary in respect of the public interest, otherwise normal insolvency
procedures are applicable.
50
For the banking sector, the BRRD explicitly requires eight early intervention measures that supervisory authorites at least,
should have at their disposal (BRRD, Article 27).
51
Solvency II directive, Article 141.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 38
occurred, their availability is subject to restrictions (EIOPA 2016b and EIOPA 2017). Granting
supervisors further explicit powers would enable them to intervene early.
84. The LIR environment emphasises the need for the effective use of pre-emptive
measures. A prolonged LIR environment may reduce some insurers’ solvency ratios until
eventually they no longer have enough capital to pay claims in full. A significant proportion of
the industry might be affected by such a capital shortfall. However, this is a “slow burn” issue,
giving authorities an opportunity to undertake significant RR planning as well as to apply early
intervention tools. These tools will be more effective if they can be used before the SCR has
been breached, especially when a capital shortfall in the future might be forecast in view of the
current trend of the insurer’s solvency position.
52
The tools will be key to proactively
minimising the capital shortfall, and reducing the likelihood of conditions deteriorating to the
extent that it becomes necessary to use resolution tools.
4.4 Resolution tools
85. A resolution authority should have an effective resolution toolkit at its disposal,
providing it with the flexibility to apply the least disruptive measure under any given
circumstances. Moreover, the use of resolution tools is not generally mutually exclusive and
a combination of several tools might produce the best result. It should be noted, however, that
no measure is without side effects, e.g. in terms of the time needed to implement it or in terms
of the impact it would have on consumer confidence. When deciding which resolution tool to
use, the resolution authority should consider the relevant circumstances, i.e. the ability of
different groups to bear losses and the resulting contagion risk.
86. The existing resolution toolkit is fairly limited across the EU. Since the crisis, additional
resolution tools for insurers have been developed. Many of these were inspired by the
resolution tools in the banking sector, although some new insurance-specific tools have also
been developed. As Table 4 shows, they have been proposed as part of the G-SIIs
framework, although the IAIS is currently amending the ICP 12 standard, which recommends
making resolution tools available in their respective jurisdictions. Most of the new tools are not
currently available in the majority of EU Member States (EIOPA 2016b and EIOPA 2017).
87. Moreover, resolution tools can be applied at different levels within an insurance group.
The FSB stipulates that resolution tools can be applied at the holding or parent company level
of a group, at the level of a sub-holding, and/or at the level of subsidiary entities. Resolution
strategies for insurance groups may be based on entry into resolution at the level of individual
operating entities or at the level of a non-operating holding company, or a combination of both,
depending on the characteristics of the group (FSB 2016). The choice of points of entry
should be guided by the group structure of the insurer and the way that its activities are
organised within that structure. The availability and coverage levels of IGSs may also be a
relevant consideration.
52
For example, the solvency position of an insurer, which is subject to transitional measures activated under Solvency II,
might not improve given the LIR environment. As a result, the insurer could breach its SCR in the future, once the
transitional measures would behave been phased out over time.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 39
Table 4
Resolution tools
Resolution Tools
Sector
Bank
Insurance
Liquidation
Withdrawal of permission to write new business / Run-off
Bridge institution
KAs
Asset separation
KAs
Bail-in / Restructuring of liabilities
KAs
Portfolio transfer
KAs
Temporary stay on early termination rights
KAs
Moratorium / Suspension of surrenders and pay outs
KAs
Source: ESRB.
Note: The blue background colour refers to tools which are broadly available in the EU Member States, ‘KAs’ refers to tools introduced by the FSB
KAs (Annex 2). Only tools discussed in this section are covered in this table.
4.4.1 Commonly used resolution tools
4.4.1.1 Liquidation
88. Liquidation constitutes the main part of ordinary insolvency proceedings. It involves a
withdrawal of authorisation from the insurer. The insurer is declared insolvent by a court or
ministry of its competent jurisdiction and is then placed under legal supervision and controlled
by an appointed liquidator. The liquidator controls the gathering of the insurer’s assets and
determines its liabilities. During the liquidation process, all policies are cancelled and
premiums are returned to policyholders. All claims against the insurer are identified,
quantified, and adjudicated. Also, applicable insurance guarantee funds are triggered, and all
assets are generally liquidated. The insurer’s assets and securities are ranked and distributed
in the manner set forth by law as being the most effective way to discharge the debts that are
owed to various creditors, and policyholders may rank equally with certain creditors.
53
To this
end, the liquidator develops a plan to distribute the insurer’s assets according to established
law and submits this plan to the relevant court and/or competent authority for approval.
Finally, the insurer’s assets are equitably distributed to its creditors and the proceedings are
closed.
89. In contrast to other resolution tools, liquidation might not guarantee continuity of
critical functions and is quite a lengthy process. As the experience of the global financial
crisis suggests, a bankruptcy filing by a financial institution can have a negative systemic
impact on the financial system. Moreover, the process can take several years and is also
associated with delays in the settlement of outstanding claims, possibly undermining the trust
of the wider public in the EU insurance sector as a whole.
90. Nevertheless, liquidation remains an applicable tool within an effective resolution
toolkit. It remains appropriate to use liquidation for insurers with no critical functions. It could
53
Article 275 of the Solvency II Directive in general affords policyholders a certain priority.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 40
also be applied together with another resolution tool. In a case of this type, where the
resolution authority is applying another resolution tool partially (e.g. the sale of business or
asset separation), the remaining (non-critical) part of the institution could be wound up under
ordinary insolvency proceedings.
4.4.1.2 Run-off
91. Run-off has long been a distinct resolution tool in the insurance sector. Its use across
the EU is longstanding and widespread. The current level of non-life insurance liabilities for
insurers in run-off is approximately €247 billion (PwC 2016), and the insurance business
model lends itself to the use of run-off as a resolution tool. First, there are usually penalties
involved when an insurance contract is cancelled. In contrast, depositors incur almost no
costs when withdrawing funds from a bank. Second, insurance is basically driven by the
liability side of the balance sheet. As a result, it does not have the maturity transformation role
of a bank.
54
92. Run-off contributes to the stability of the sector, particularly if it is combined with other
resolution tools. Run-off in the context of resolution is triggered by the decision of the
competent authority to withdraw an insurer’s licence to write new business in the case of
financial distress or non-compliance with Solvency II.
55
The rationale of its use is that insurers
may not be able to afford to continue to expand and to write new business, but they can afford
to honour contractual obligations related to the existing stock of policies, while avoiding much
of the cost and disruption involved in the normal liquidation procedure. Moreover, preventing
new business from being written reduces an insurer’s capital requirements and the expenses
associated with new business (European Commission 2012). However, a run-off in the
context of resolution usually means that the insurer is only able to partially honour its
contractual obligations and that a number of claims might not be paid in full. It is, therefore, an
insolvent run-off. An accompanying resolution tool (such as a bail-in or the restructuring of
liabilities) or some other injection of capital
56
may be used to allow the insurer to enter solvent
rather than insolvent run-off. Otherwise, a resolution tool other than run-off might be more
appropriate. It should, however, be noted that a solvent run-off may become an insolvent run-
off over time, e.g. if changes to the value of the insurer’s assets and liabilities lead to a
quicker-than-expected run down of its capital buffer (FSB 2011).
93. A run-off can be compatible with continuity in cover for existing policyholders. To the
extent that policyholders can purchase insurance from other providers on renewal, run-off is
compatible with continuity of cover for existing policyholders. The lack of a substitute on the
renewal of the cover provided by the EU insurer in run-off could give rise to systemic
disruption or undermine confidence in the EU insurance sector. In such a case, resolution
tools other than run-off might be more appropriate.
54
Inspired by the insurance sector, some regulators have asked banks to prepare plans for solvent wind-down, which would
be similar to run-off but would largely focus on a bank’s trading book, rather than on the institution as a whole. See, for
example, Bank of England (2015)).
55
A run-off could also happen on a voluntary basis by an insurer’s decision to stop the sale of new policies, if, for example,
the insurer changes its business model. This is also referred to as a ‘solvent’ run-off and the insurer is able to make
payments in full without altering arrangements with policyholders and other creditors. Voluntary run-off is, however, not
considered within the scope of this section, as it cannot be viewed as a resolution tool.
56
For example, an IGS may have the power to inject capital into an insurer instead of paying claims if the insurer fails.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 41
94. The success of a run-off also requires policyholders to believe that their claims will be
honoured. If this is not the case policyholders might begin to cancel their contracts. Given the
increased use of a special termination clause in the insurance sector, in particular for
reinsurance contracts, insurers seem to be more vulnerable to early policy cancellations than
in the past (Paulson et al 2014). Cancellation is more likely for highly insurable, low-risk
individuals, although this also applies to institutional investors. So, an insurer might be left with
a portfolio of higher-risk contracts with less insurable policyholders or economically
unattractive contracts, e.g. life policies with high guaranteed returns. There may not be
adequate capital for this type of portfolio shift. As a result, policyholders with claims that arise
earlier might be better off than policyholders with claims that arise later. In addition, the
maturity profile of the insurer’s asset portfolio may not provide sufficient liquidity to meet
demand in the event of a spate of cancellations. This scenario may be averted by ring fencing
the assets that correspond to certain insurance liabilities (European Commission 2012).
4.4.2 Resolution powers and tools that allow the business to be transferred or
separated
4.4.2.1 Portfolio transfer
57
95. Portfolio transfer involves the transfer of policyholder obligations, along with
corresponding assets, from a distressed insurer to one or more third parties. A third
party may be another insurer(s) or a bridge institution if a willing buyer (or buyers) cannot be
found. Portfolio transfer has been used in both the banking and the insurance sectors,
although the processes are not identical. In contrast to the banking sector, where assets (loan
books) are transferred in exchange for other assets, portfolio transfer in the insurance sector
consists of transferring liabilities (the portfolio of insurance contracts) and corresponding
assets, which are received to cover the liabilities the third party has taken over.
96. Portfolio transfer ensures continuity of cover and payments while the distressed
insurer is wound down without disruption to the transferred policyholders. It allows
claims to be paid as they fall due and reduces disruption and possible value destruction. The
key advantage of portfolio transfer is that continuity of coverage can be maintained for certain
parts of a distressed insurer’s business with limited use of policyholder protection or other
guarantee schemes. The portfolio transfer is expected to provide continuity of cover, as
policyholders should be able to purchase insurance from the insurer who has taken over the
portfolio.
97. An effective portfolio transfer in resolution might require the availability of other
resolution tools. This is particularly true when there is a capital shortfall. For example, an
assets and liabilities separation tool could be used in conjunction with a portfolio transfer,
since a potential buyer would only be willing to take on a viable part of the distressed insurer’s
portfolio. Some countries allow the policyholders’ liabilities to be restructured in order to make
a transfer possible or, alternatively, the IGS could be used to facilitate a portfolio transfer. The
IGS could be involved with the justification that a portfolio transfer might be the most cost-
57
This tool is referred to as a sale of business tool in a banking context.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 42
effective way to protect the rights of the current policyholders. Powers to transfer related
reinsurance without having to obtain the reinsurer’s consent may also be required since the
continuity of reinsurance is often critical to ensuring that the new insurer remains sufficiently
capitalised and that policyholders are adequately protected. Furthermore, if some of the
distressed insurer’s liabilities have been left behind additional resolution tools may be
required, e.g. run-off or liquidation. If only the remnant of the distressed insurer is liquidated, it
should be ensured that compensation is no worse than the amount that would have been
received if the insurer had been liquidated as a whole (FSB 2016b).
98. One significant challenge to enacting portfolio transfer is finding a willing buyer in a
suitable timeframe. Evaluating the policyholder liabilities and assets involved in a transfer
takes time and a bridge institution might need to be set up to run the insurance business
temporarily (FSB 2011). Moreover, operational difficulties could slow down the process – for
instance, IT systems used to service insurance products may be difficult to transfer (EIOPA
2016b and EIOPA 2017). Portfolio transfer might be a less viable option in times of systemic
disruption. The problem of finding a willing and appropriate buyer for parts of a distressed
insurer’s business is exacerbated when other insurers are also subject to similar
vulnerabilities.
4.4.2.2 Bridge insurer
99. A bridge insurer takes the failed insurer over for a limited period of time, thereby
protecting the continuity of critical functions, including the fulfilment of existing
obligations to policyholders. The bridge insurer is entrusted with continuing the operations
of the insolvent insurer (or some of these, i.e. critical functions and viable operations) until the
portfolio has been transferred to a private purchaser (see Figure 1 for the process). This
transaction provides sufficient time for prospective purchasers to carry out the appropriate due
diligence on the failed insurer (FSB 2016a). It could also be particularly beneficial for
managing crisis situations, where willing buyers are less likely to be found at short notice. It
allows the resolution authority to react promptly
58
by maintaining the value of the insurer (or its
viable part), acting as a stopgap until the portfolio can be sold to a third party. For
policyholders, this means that the continuity of their cover and payments (in whole or in part)
is ensured. As such, this tool helps to preserve stability and trust in the insurance sector.
100. If a bridge insurer is created by the resolution authority, it should have an insurance
authorisation, comply with regulatory requirements and operate in a conservative
manner. Moreover, it may be recapitalised by a resolution fund or an IGS or, where
applicable, by a forced debt to equity conversion. A bridge institution may operate for a limited
period of time and may be subsequently liquidated. This tool was created with a single insurer
in mind, although it could also be possible to create a bridge insurer that could pool the viable
business of different insurers (i.e. those which had failed simultaneously). The bridge insurer
would be controlled and (partially) owned by a public authority, which would then operate the
transferred business of a failing insurer. The resolution authority would therefore have a
number of powers including (i) deciding on the management and corporate governance of the
58
It should be noted that a prompt reaction is only possible if the bridge insurer is founded a priori or if the insurance RR
framework stipulates special conditions for the licencing of such an institution.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 43
bridge insurer; (ii) setting the terms and conditions under which the bridge insurer must
operate as a going concern; and (iii) arranging the sale or winding-down of the bridge insurer,
or the sale of some or all of its assets and liabilities, to a third party. In addition, the bridge
insurer could permit options to be exercised under existing insurance contracts, such as the
surrender or withdrawal of contract cash value and the payment of further premiums provided
for under the existing contracts. The continuation of reinsurance coverage could also be
considered.
101. The tool has not been actively used in the EU in the past. A bridge insurer is under
consideration in the global resolution regime for G-SIIs. However, in contrast to the banking
sector, the tool has not been used frequently across the EU in the insurance sector in the
past. Its use is stipulated in the Romanian RR framework and its introduction is also planned
in France and the Netherlands.
Figure 1
Bridge insurer
Source: ESRB.
102. Creating a bridge insurer also poses some challenges. First, running a bridge institution is
a labour-intensive process which includes, for example, identifying the relevant business
which needs to be preserved. Second, funding for the bridge insurer needs to be clarified,
particularly in relation to a possible recapitalisation by a resolution fund or an IGS. Moreover,
finding a private purchaser could prove challenging. If the sale to a private owner cannot be
executed at an acceptable price, the cost of operating the bridge insurer may exceed the cost
of liquidation. Challenges may also arise if the resolution of the failed insurer triggers the
cancellation of services provided by the insurance group, in which case additional powers may
be needed imposing temporary stays on the termination of contracts (FSB 2016b).
Resolution Authority
Insurer under
resolution
Temporary
Bridge Insurer
Sale in the market
Viable business
Critical functions
Total or partial
transfer
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 44
4.4.2.3 Assets and liabilities separation
59
103. This tool enables authorities to relieve the balance sheet of an insurer under resolution
from certain risks, while maintaining the continuity of critical functions. The resolution
authority would establish a liabilities and assets separation scheme, which would allow the
liabilities and related assets and rights of an insurer under resolution (or those of a bridge
insurer) to be transferred – at market value – to a separate vehicle. This separate vehicle
would operate as a legal entity, with the resolution authority exercising its influence via its
appointment of the manager of this separate vehicle.
104. The separate vehicle should maximise the value of the policies. It would either make
legal arrangements to terminate the policies in return for compensation from the sale of the
assets, or run them off over a longer time period. This is different from a special vehicle in the
banking sector, which aims at maximising the value of the instruments through their sale or by
winding these activities down in an orderly manner.
105. The emphasis of this tool in the insurance sector is on the separation of liabilities,
together with the assets held for these policies. In the banking sector, the relief to banks’
balance sheets stems from separating depreciated assets and related risks from viable asset
classes. The main purpose in the insurance sector would be to allow the authorities to
separate the policies (liabilities) of an insurer under resolution (or those of a bridge insurer)
which might cause distress, from viable policies (liabilities). The tool would involve the
identification of these liabilities, together with the assets held for these policies, and their
transfer to a separate vehicle. It might, however, be difficult in practice to identify precisely
which assets in the insurer’s asset portfolio are held for a subset of policies. This identification
is performed for larger classes of policies, but not for a subset of policies, with the exception of
separate account assets and liabilities (such as unit-linked contracts).
106. The use of the tool should be subject to safeguards in order to prevent any undue
competitive advantage accruing to the failing insurer. The tool may therefore only be used
in conjunction with other tools in the resolution framework.
60
It may only be used if conditions
in the particular market are such that the liquidation of the liabilities and related assets of the
failing insurer under normal insolvency proceedings could have an adverse effect on one or
more financial markets, or where the transfer is necessary to ensure the proper functioning of
the insurer under resolution (or the bridge institution), or to maximise liquidation proceeds.
107. The assets and liabilities separation tool is less likely to be used in the insurance
sector. Although it is included in the global standard for G-SIIs, it has not been implemented
in relevant national legislation for insurers in the EU. Its use is not very likely given the long-
term perspective taken by insurers’ balance sheets. One major advantage of insurance is that
insurers’ assets do not need to be liquidated until claims or benefits under the policies have to
be paid. The application of the assets and liabilities separation tool involves the disposal of
assets and liabilities at market value. Losses could be realised if sufficient time is not allowed
for the assets to return to their long-term economic value, leading to a destruction of value for
policyholders.
59
The FSB refers to this instrument as an asset separation tool. Given the strong link between assets and liabilities in the
insurance sector, the tool is referred as an “assets and liabilities separation tool” throughout this report.
60
This is the condition for its use in the banking sector (see Article 42(5) of the 2014/59/EU).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 45
4.4.3 Resolution tools affecting contractual rights
4.4.3.1 Imposing restrictions on the termination of contracts
108. The power to impose temporary restrictions on the termination of insurance contracts
gives the resolution authority more time to deal with a distressed insurer in an orderly
fashion. The resolution authority would have the power to temporarily suspend the
termination rights of any party to a qualifying contract, including reinsurers, in order to protect
the insurers’ risk mitigation programme and thus reduce the risk of unnecessary destruction of
value. The suspension would be on the condition that contractual obligations would continue
to be met. This would avoid any deterioration in the financial position and any subsequent
market instability resulting from the counterparties closing out their transactions at the same
time. The exercise of this power, its scope of application, and the duration of the stay should
be appropriate for the type of insurance product. Different treatment would be envisaged for
life and non-life primary insurers and reinsurers.
109. The power to impose restrictions might be less relevant for insurers. The FSB has
envisaged the use of this power by G-SIIs. However, it is presumed that this power might be
less relevant for insurers than for banks, e.g. policyholders may be less likely to cancel their
insurance contracts.
110. The tool comes with some costs. First, the application of this tool could face enforceability
issues in some EU Member States. Moreover, the restriction could undermine confidence in
the sector and should therefore be subject to adequate safeguards. Second, the enforceability
of the suspension or restriction is not certain, especially where third-party law governs the
contract; however, this issue has been increasingly addressed by changes to the contracts, in
particular in respect of derivatives contracts.
61
4.4.3.2 Bail-in / liabilities restructuring tool
111. A bail-in tool would enable losses to be quickly assigned to shareholders and
creditors. This tool should however be available only as a last resort in the case of
policyholders bail-in. A bail-in involves the use of techniques that restructure liabilities. This
could be done by limiting or writing down liabilities held by providers of debt and other
creditors, or by converting those liabilities into shares or other instruments of ownership. The
tool would allocate losses to shareholders and creditors in a manner consistent with the
statutory creditor hierarchy and legal framework. Where an insurer is failing and the bail-in tool
is used, shareholders and other unsecured creditors should be fully written down before
losses are imposed on policyholders. The bail-in tool would help ensure that losses are
absorbed and the insurer is recapitalised.
112. A bail-in tool would also ensure continuity of critical functions and it could address
some of the systemic risks the insurance sector may pose. Bail-in mimics the results of
insolvency, but it can do this on a going-concern basis, without unnecessary value destruction
61
The International Swaps and Derivatives Association, in conjunction with the FSB, has devised a “Stay Protocol” which
aims to achieve the cross-border recognition of suspensions and overrides, by incorporating this recognition into the
relevant financial contract between the parties opting in to the protocol.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 46
or interrupting the provision of critical functions. Losses can be assigned quickly without a long
sale process, and certainty can be restored. As such, a bail-in could address the systemic
risks posed by insurers, by internalising losses and limiting these to the creditors of the failing
insurer, and by helping to ensure the continuity of critical functions.
113. The success of bail-in in resolution depends on the availability of liabilities that can be
bailed in and the identification of the optimal proportion of liabilities to be bailed in. The
design of bail-in requirements for insurers might be challenging due to the particular structure
of their liabilities. If the bail-in tool is intended to “recapitalise” the insurer, the non-policyholder
related liabilities will be converted into equity. As Chart 5 shows, insurers make less use of
debt financing than other financial institutions and a large part of their liabilities consist of
technical reserves. However, the share of debt financing is not negligible, especially with
regard to systemically important insurers. The conversion of policyholder liabilities into equity
might be more problematic, although given insurers’ liabilities structures it might be inevitable
in the event of a substantial shortfall.
114. A write-down of policyholder liabilities could accompany a portfolio transfer, ensuring
that the buyer can afford to take on the insurance liabilities. The resolution authority
could be given the power to restructure, limit or write down liabilities, by reducing or
terminating future benefits and guarantees, reducing the value of contracts upon surrender,
and terminating or restructuring options for policyholders. It could also suspend the rights of
policyholders by temporarily restricting or suspending their right to withdraw from insurance
contracts. Policyholders could be compensated by the IGS (assuming one exists and that it
covers that particular policy) if their rights have been affected by the resolution. An IGS allows
EU Member States to identify, in advance, the policyholders and products that should receive
protection, and to decide on the appropriate level of such protection. Moreover, if the process
resulted in a situation where policyholders were worse off than they would have been under
liquidation, the RR framework should foresee compensation in line with the NCWO principle.
115. The tool could produce undesired side effects. Its potential impact, as well as its various
dimensions, should therefore be carefully assessed. First, it impacts protected contractual
rights. As a result, bondholders might be reluctant to invest in insurers if there is a risk that the
liability could be written down. Second, a write-down of insurance liabilities (or even the mere
possibility of this occurring) could create a lack of confidence in the insurance sector.
Moreover, if a bail-in were applied in a situation of widespread distress among insurers or in
financial markets, the market could suffer from the impact this would have on investors if
liabilities were written down in multiple institutions. On the other hand, the alternatives to a
bail-in might also be undesirable. For example, a bailout would be undesirable from a public
finance perspective and, furthermore, insolvency could also lead to a loss of confidence in the
insurance sector and, as a consequence, in the broader financial sector.
116. The tool could be applied, to some extent, in some EU Member States along with a
portfolio transfer tool. The FSB has established bail-in as a resolution tool for G-SIIs,
suggesting that this power should be available to all resolution authorities in jurisdictions
which are home to these insurers. In the EU, the tool has been proposed as part of the RR
framework in the Netherlands, while a few EU Member States allow the restructuring of
policyholders’ liabilities prior to use of the portfolio transfer tool. A breach of policyholders’
contractual rights has not, however, been contemplated as a possible solution in most EU
countries and, in the absence of the consent of all policyholders, the write-down of insurance
liabilities may only be executed in the context of a winding-up. Moreover, the conversion of
liabilities into equity is also not envisaged in national legislation across the EU.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 47
Chart 5
Structure of liabilities of the insurance sector (2008-2015, %)
Source: ECB - SDW
Note: Data refer to the period 2008-2015. Moreover, the data are not fully comparable across countries.
117. This report recognises that EU Member States have differing views on the use of the
bail-in tool. The benefits of the conversion into equity of bondholders is generally recognised
across the ESRB membership. Still, a few ESRB member institutions view this option as a last
resort only, e.g. when the stability of the financial system is at stake or if creditors would face
larger losses in a regular insolvency procedure. With regard to the allocation of losses to
policyholders, the majority of ESRB member institutions view this as a last resort option, while
a few ESRB member institutions are of the view that policyholders should not be affected by a
bail-in decision at all, arguing that this might undermine the trust of the general public in the
insurance sector and could therefore lead to a systemic disturbance, with mass surrenders
and cancellations of policies.
4.5 Cross-sectoral implications of resolution measures
118. The application of resolution tools may also produce some cross-sectoral spillover
effects. Although an RR framework should result in less contagion and fewer spillovers to
other sectors of the financial system than a normal insolvency procedure, spillover effects
from some resolution tools cannot be entirely avoided. Building on experience from the
banking RR framework, this section discusses the possible cross-sectoral impact of selected
resolution tools on the insurance sector.
4.5.1 Stay on termination rights tool
119. This tool provides the resolution authority with more time for the resolution process.
The resolution authority could apply a stay on termination rights and other powers to terminate
contracts included in the contractual clauses, giving the authority more time to resolve the
failing insurer. This could be useful in cases where insurers are suffering temporary runs on
their business, although runs have only been experienced exceptionally in the past (e.g. in
Hong Kong and Singapore, see IAIS 2011a as well as in the UK, see Kelliher et al. 2005).
Surrender penalties dampen incentives for policyholders to cash in their policies prematurely;
0
10
20
30
40
50
60
70
80
90
100
AT BE EE FI FR DE EL IT LU NL PT
SK
SI ES euro
area
capital
debt
technical reserves
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 48
however, the increased use of special clauses (e.g. to terminate a contract if the insurer’s
rating deteriorates) might increase the likelihood of such exceptional events in the future.
120. If applied to derivatives contracts, the tool exposes the counterparty to credit risk.
Stays may also be applied to the termination of access to financial market infrastructures or
other services (FSB 2016), and the tool could stop the spread of losses in derivatives markets.
At the same time, counterparties that are not able to terminate a derivative contract would be
forced to keep existing positions open with a potentially failing institution, and would therefore
no longer be protected from counterparty credit risk. In the case of a resolution authority
applying the stay on termination rights tool in the event of the failure of an insurer, banks
might be particularly affected due to their links with derivatives trades.
4.5.2 Bail-in tool
121. The application of the bail-in tool has a direct impact on other financial market
participants. This can be seen in the use of the bail-in tool in the banking sector, involving the
bail-in of bank bonds where losses materialise for the holders of those bonds.
62
The loss
allocation mechanism implies spillover effects for other financial market participants, which
increase in line with the size of the exposure. Moreover, the introduction of the bail-in tool in
the banking sector has been accompanied by the creation of TLAC/MREL requirements.
63
These ensure that there are sufficient liabilities available in each bank that may beexpected to
be bailed in, both to absorb losses and to provide the capital needed to facilitate a transfer or
recapitalise the company. At the same time, the TLAC/MREL requirements might have
increased interconnectedness with other sectors, including the insurance sector.
122. There are large differences across countries and sectors for bank “bail-inable” debt
holdings. Over the last two years, although banks have reduced their exposures, non-bank
sectors have continuously increased their stake (ECB 2016). However, the sectoral exposures
of non-bank sectors are relatively minor when compared with the amount of total assets held
by each sector. For example, the combined exposures of the insurers and pension funds to
bail-inable bank debt are 2.3% of total assets (see Chart 6). The increase in bail-inable debt
holdings seems to be in line with the general trend of increased risk-taking by insurance
corporations and pension funds (ICPFs) observed in their portfolio shifts towards lower-rated
debt securities (ECB 2016). However, the aggregated number by sector may hide significant
differences between the individual ICPFs (ECB 2016). Moreover, there are also differences
across countries, e.g. ICPF holdings of bail-inable bank debt have a high proportion of debt
issued by French banks (see Chart 6).
62
However, the losses could not be larger than they would have been if the institution had entered insolvency instead (no-
creditor-worse-off safeguard).
63
TLAC refers to the FSB’s Total Loss Absorbtion Capacity applicable to G-SIBs, and MREL stands for Minimum
Requirements for own funds and Eligible Liabilities, a standard applicable to all EU-domiciled banks.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Recovery and resolution powers and tools 49
123. Additional regulatory
standards may be needed to adequately
address risks and the consequences of
holding bail-inable debt. In the banking sector,
a global standard aimed at reducing the risk of
contagion from the use of the bail-in tool will be
applicable from 2019 for most G-SIBs
64
(BIS
2016). Above certain thresholds credit
institutions will be required to deduct holdings of
TLAC instruments from their own regulatory
capital, which will discourage them from
investing largely in bail-inable debt.
Furthermore, the European Commission has
proposed that a similar approach be followed for
EU banks (European Commission 2016).
Sectoral regulation, other than that applicable to
the banking sector, does not require special
treatment for bail-inable debt and standards are
unavailable for holdings of bail-inable debt. For
the insurance sector, it may be expected that
insurers’ holdings of the bail-inable debt of
banks, or that of other insurers, will increase in
the future. This is due to proposed changes in
the banking regulation related to the
introduction of the TLAC/MREL requirements
(IMF 2016a) and a possible shift in holdings of
bail-inable bank debt to non-bank entities
(including insurers) after the introduction of
standards covering holdings of bail-inable debt
in the banking sector, and in light of the possible introduction of the bail-in tool for insurers in
the EU (e.g. the proposed RR framework for insurers envisaged in the Netherlands). This
suggests that a revision of Solvency II could provide an opportunity to include rules on bail-
inable debt for insurers too.
124. Moreover, the application of the bail-in tool should be preceded by an analysis of
directly impacted investors. An analysis of the holdings of bail-inable debt at aggregated
level in the banking sector indicates that there could be possible interdependencies and
spillover effects. The resolution authority, possibly in cooperation with the macroprudential
authority, could therefore benefit from assessing the financial stability or contagion impact of a
bail-in on other supervised entities, sectors or regions.
125. The IAIS has decided against the establishment of a TLAC/MREL requirement for G-SIIs
at this juncture (IAIS 2016a). The use of the bail-in tool in the insurance sector would seem
to be more challenging, given the particular structure of insurers’ balance sheets. The IAIS
has decided not to set a minimum level of bail-inable debt. Such a requirement for insurers
would not only increase their linkages within the rest of the financial system, but would also
imply changes to the business models and funding structures of insurers. If it were introduced,
insurers would be expected to hold more bail-inable debt – typically long-term unsecured
subordinated debt (IAIS 2011).
64
The term G-SIBs refers to global systemically important banks.
Chart 6
Holdings of bail-inable bank debt securities
by euro area holding sector and by country
of issuance
(Q1 2016; EUR billions; Bail-inable bank debt as a percentage of total
assets of the holding sector)
Sources: ECB sectoral SHS and ECB calculations.
Notes: insurers are captured in ICPFs (Insurance Companies and
Pension Funds), and other institutions are: Credit Institutions (CI),
Households (HH), Investment Funds (IF) and Money Market Funds
(MMF).
The breakdowns in the chart show issuance by domicile of issuing bank
and holdings by euro area (EA) sectors. Percentages at the top of
columns show debt holdings relative to total assets (for financial sectors)
and relative to financial assets (for households)
0
100
200
300
400
500
600
CIs HHs ICPFs IFs MMFs other
1.5%
1.3%
2.3%
3.2%
8.6%
FR
DE
IT
NL
ES
other euro area
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 50
5.1 Introduction
126. There are significant operational costs associated with the failure of an insurer. One
key operational cost is the cost of providing for and maintaining practical arrangements that
guarantee continuity of cover and payments to existing policyholders. These costs are
realised at different times and by different agents depending on the type liability and related
funding arrangements. Resolution requires significant time and human resources. This cost
would, to a large extent, be proportional to the size and complexity of the insolvent insurer.
For example, the resolution of a cross-border insurer would require coordination between
different authorities across jurisdictions.
127. The use of specific instruments leads to further costs, including compensation of
policyholders. For example, there is likely to be a large one-off administration cost when a
transfer of a portfolio of insurance contracts is required. The scale of this cost varies according
to the nature of the insolvent insurer and the way the transfer is facilitated: a transfer to an
existing insurer will probably have a lower cost than a transfer to a newly-created bridge
institution. Moreover, managing bigger portfolios may be also expected to lead to economies
of scale. When a measure other than insolvency procedure is applied, the resolution process
aims at honouring the no-creditor-worse-off principle (NCWO): authorities may need to
compensate policyholders and other creditors if they find themselves in a worse situation
under resolution than they would have done under the ordinary insolvency of an insurer.
128. The market values of both the assets and the liabilities of an insurer could fall if it
enters resolution. In the case of liabilities, differences could arise between the valuation of
liabilities in a “going-concern” situation (under Solvency II) and the valuation in a “gone
concern” situation. Furthermore, and particularly in a system-wide crisis, the disruption caused
by a market-relevant insurer’s failure could cause the market value of some of its assets to
fall. It is therefore plausible that asset depreciation could be minimised by prudent market
signalling and the effective management of the resolution process, e.g. by avoiding the forced
sale of any assets unless the alternative would be even more costly. The FSB highlighted this
point in its Key Attributes (KAs), stating that resolution authorities should avoid unnecessary
destruction of value and should seek to minimise the overall costs of resolution and the losses
to creditors (FSB 2014a).
5.2 Funding sources other than public funds
129. One of the purposes of a resolution framework is to avoid a bailout of troubled
institutions. In the absence of a government bailout, the resolution costs associated with the
failure of an insurer could be borne by three groups: (i) the insurance industry, through the use
of pre-defined funding arrangements; (ii) policyholders through a write-down of insurers’
liabilities; and (iii) other creditors, through a debt write-down. Which group should bear the
costs should relate to how able it is to bear the costs, as well as the incentives created by
passing on the costs, and will be affected by the creditor hierarchy. Since the write-down of
policyholders’ liabilities and the write-down of debt have been covered in Section 4, the
remainder of this section focusses on financing from the industry.
Section 5
Resolution funding
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 51
130. Industry-financed resolution requires reliable funding sources. An effective RR
framework should be accompanied by secure, continuous and stable funding sources,
otherwise the funding arrangements cannot be relied upon by creditors, market participants
and policyholders and, as a consequence, resolution may not work as a mechanism ensuring
financial stability without the need to resort to public funds. Besides the funding costs of
resolution tools, these funding arrangements should also cover the immediate liquidity needs
of a failed insurer and the administrative costs of operating the funding arrangement.
131. A resolution funds (RF) and insurance guarantee schemes (IGS) are examples of
industry funded financing sources. As for the banking union,
65
the RF and the IGS may be
understood as two distinct but complementary pillars with different objectives. The primary
function of an RF is to finance all the various resolution actions undertaken for an insolvent
insurer, including, for example, the run-off, the sale of business, and transfer to a bridge
institution. It could also be used to compensate policyholders, although this would be limited to
ensure the NCWO safeguards. In contrast, IGSs have a primary function to settle claims for
policyholders, thus providing for policyholder protection in the event of an insurer’s failure by
covering claims not related to NCWO safeguards. Since IGSs in some EU Member States
have been given limited powers to finance resolution actions too, this report also considers
IGSs.
5.2.1 Resolution funded by a resolution fund
132. Even though an RF is envisaged in global standards for G-SIIs, it is rarely used in the
EU. According to the KAs, jurisdictions should have an industry-financed RF
66
or a funding
mechanism in place, for the costs of providing temporary financing to facilitate the resolution
of an insurer (FSB 2014a). However, only Romania has an RF for insurance, while its
introduction is under consideration in the Netherlands (see Box 6), while some EU countries
have IGSs with certain powers that can be used to fund selected resolution tools.
Box 6
Examples of national resolution funds for insurers in the EU
Experience of RFs for insurers in the EU is limited to two EU Member States. This is despite
the fact that establishing an RF is a requirement of KAs in respect of G-SIIs, which are domiciled in
four EU Member States. A RF is fully operational In Romania, which is not the home jurisdiction of
any G-SIIs, while a proposal has been made to establish an RF in the Netherlands.
67
The two frameworks differ significantly in terms of scope, with the RF in Romania used to
finance a broad range of objectives. It was established as a specialised structure and is
managed by the Romanian IGS. The RF is able to: (i) cover the needs related to the
65
Within the banking union, a resolution fund has been created to provide funding to the resolution process where and if
needed subject to conditionality, whereas deposit guarantee schemes provide funding for deposit insurance.
66
This function could be provided by an IGS, which would be entrusted with assisting the use of resolution tools, e.g. a bridge
insurer.
67
Moreover, the RF in Romania is supplemented by an IGS, which covers all life and non-life products. In the Netherlands,
the existing IGS is strictly limited to motor vehicle insurance.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 52
implementation of resolution tools; (ii) ensure the continuity of the critical functions of the insurer
under resolution; (iii) recover any reasonable expenses incurred in relation to the use of the
resolution tools or the exercising of the resolution powers; (iv) guarantee the assets or the liabilities
of the insurer under resolution; (v) finance a bridge institution or asset management vehicle, or
grant a loan to these or to the insurer under resolution; (vi) transfer the assets and the insurance
portfolio of the insurer under resolution; or (vii) pay compensation to shareholders or creditors. The
resources may not, however, be used directly to absorb the losses of the insurer. The RF supports
the resolution objectives, including the protection of public funds, by minimising reliance on public
financial support and avoiding any significant adverse effects on the financial stability of the
Romanian insurance market.
In contrast, the RF proposed in the Netherlands will have limited coverage. This will be a
small RF restricted to financing the operational costs of resolution, such as the establishment of a
bridge insurer. It should not be used for the recapitalisation of insurers or the absorption of losses,
or to guarantee risks. The proposal envisages that no-creditor-worse-off safeguard breach
compensation will be applied.
There are also significant differences in respect of funding arrangements. The Romanian RF
is pre-funded, based on gross premiums, while the RF in the Netherlands is expected to be
financed ex post by contributions from the insurance industry. In Romania contributions are
received from all authorised insurers and are calculated separately for non-life and life insurance,
through the application of a percentage share in relation to the gross premiums earned,
68
while no
risk weights are applied. If there is a deficit in the RF to cover the obligations resulting from the
implementation of resolution tools, the percentage share of the contributions may be increased up
to a certain limit.
5.2.2 Resolution funded by an insurance guarantee scheme
133. Insurance guarantee schemes (IGSs) have been set up to provide protection to
policyholders. As Solvency II, like any other framework, cannot create a zero-failure
environment for insurers, IGSs have a wider positive market impact by preserving consumer
confidence and minimising market disruption if an insurer fails (Oxera 2007). IGSs provide
last-resort protection to consumers when insurers are unable to fulfil their contractual
commitments. As such, they protect policyholders against the risk that their claim will not be
met if their insurer becomes insolvent (European Commission 2010a). However, the existence
of an IGS may also be associated with possible adverse effects on the behaviour of market
participants (e.g. moral hazard) although the alternatives, such as ex post government
interventions, could have even greater drawbacks.
134. The role of IGSs could be expanded to fund the use of resolution powers. IGSs have
been created with the objective of providing last-resort protection to policyholders, although
some IGSs in the EU could take action that facilitates the use of selected resolution powers,
i.e. portfolio transfer. This is done under certain conditions, which vary across EU Member
States. The objective of any such action would typically be to protect policyholders rather than
68
It is 0.25% of gross premiums for life insurance and 0.4% of gross premiums for non-life insurance, up to a cap of
RON 50 million (currently about EUR 11 million).
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 53
to safeguard financial stability. For instance, in the UK one of the criteria for IGS involvement
in portfolio transfer is that a transfer must be beneficial to policyholders (EIOPA 2012a). It is
therefore envisaged that the use of IGSs to fund resolution powers would be limited to the
failure of primary insurers only and that IGSs would not generally fund a reinsurer resolution.
69
135. It follows that the use of funding arrangements could be further strengthened with the
aim of increasing the credibility of the overall RR framework. Their compatibility with the
KAs could be further explored, such as allowing the funding arrangements to cover costs of
any resolution tool for both policyholder and financial stability objectives.
Box 7
Analysis of existing IGSs in the EU
The importance of ensuring the continuity of payments is well established within the EU,
albeit only with regard to certain insurance policies. Most EU Member States have an IGS in
place, although in many cases with very limited coverage of insurance policies.
70
Still, compared
with 2010,
71
there has been an observable increase in the number of IGSs in the EU.
Existing IGSs could mostly provide relief to policyholders for selective policies in the event
of a primary insurer’s failure, although the picture across the EU is somewhat fragmented.
There are different degrees of policyholder protection across different insurance policies within
individual EU Member States. Sector-wide policyholder protection is provided only in a few EU
Member States, while the coverage of IGSs in a greater number of countries is limited to specific
insurance products – mostly compulsory non-life insurance, such as motor vehicle insurance. The
existence of IGSs for motor third-party liabilities insurance may be associated with the EU directive
requiring each EU Member State to set up a fund providing policyholder protection for this
purpose.
72
No similar EU-wide requirement exists for other insurance policies. Looking across the
EU, there are big differences in the level and scope of coverage of IGSs. Moreover, the picture is
further fragmented in terms of the different loss consequences for consumers, depending on their
country of residence. For example, some IGSs provide coverage for branches operating in their
jurisdictions while others do not.
73
Data demonstrate a general preference towards ex ante rather than ex post funding of the
IGS. There are also cases where a hybrid approach is followed i.e. funding through a mix of ex ante
and ex post arrangements. This is particularly the case for countries where the IGS plays an
important role in respect of policyholder protections, when measured by breadth of coverage.
69
This is due to the nature of the parties in the reinsurance contract and the absence of any direct impact on policyholders.
70
For the purpose of this report, any funding arrangement that provides a policyholder relief in case of an insurer’s failure
(beyond the NCWO principle) has been identified as an IGS.
71
A comparison is made with data from the COM White Paper (European Commission 2010).
72
Directive 2009/103/EC of the European Parliament and of the Counil of 16 September 2009 relating to insurance against
civil liability in respect of the use of motor vehicles, and the enforcement of the obligation to insure against such a liability.
73
For example, the IGS in the UK has very broad coverage, including UK policyholders from non-UK insurers (if established
in the UK via a branch or subsidiary) as well as EU policyholders of UK insurers. The IGS in the ES, instead, applies the
home-state principle only, covering claims from insurers established in ES including EU policyholders of ES insurers.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 54
Table 5
Estimated funds available under IGS
Member
State
Available
funds 2010
(€m)
Available
funds 2015
(€m)
Market size
by total
assets (€m)
Sector
Type of
funding
Coverage
AT
8
9
NA
Non-life
Ex-post
motor third-party liabilities insurance
BE
33
802
309,992
Life
Ex-ante
life insurance with a guaranteed interest rate
0
0
12,678
Non-life
Ex-post
motor third-party liabilities insurance and worker's compensation insurance
BG
121
119
740
Life
Ex-post
all life insurance products
269
286
1,137
Non-life
Ex-post
all non-life insurance products
CY
20
20
580
Non-life
Ex-ante
motor third-party liabilities insurance
CZ
NA*
54
NA
Non-life
Hybrid
motor third-party liabilities insurance
DE
1,407
1,739
978,237
Life
Hybrid
all life insurance products
1
1
176,318
Non-life
Ex-post
property and casualty products (including health)
DK
40
58
17,729
Non-life
Ex-ante
all life and non-life insurance products
EE
0,0073
0
1,042
Life
Hybrid
mandatory funded pension (pillar 2)
1
1
638
Non-life
Ex-ante
motor third-party liabilities insurance
EL
NA
NA
NA
Life
Ex-ante
all life and health insurance products
NA
NA
NA
Non-life
Ex-ante
motor third-party liabilities insurance
ES
1,331
1,867
308,031
Life +
non-life
Ex-ante
all life and non-life insurance products
FI
25
29
14,831
Non-life
Ex-post
motor third-party liabilities insurance and patient and worker's compensation insurance
FR
346 (630)
428 (785)
1,591,366
Life
Hybrid
all life and health insurance products
250
250
1,344,129
Non-life
Hybrid
compulsory non-life insurance
HR
29
21
2,222
Non-life
Ex-ante
traffic, motor boat and aircraft third-party liabilities insurance
HU
5
10
1,790
Non-life
Hybrid
motor third-party liabilities insurance
IE
31
-862
35,105
Non-life
Ex-post
a broad coverage of non-life products
IT
486
580
358,201
Non-life
Ex-ante
motor third-party liabilities insurance and hunting accidents
LU
0,07
0
12,185,125
Non-life
Hybrid
motor third-party liabilities insurance
LT
13
21
409
Non-life
Ex-ante
motor third-party liabilities insurance
LV
2
5
173
Life
Ex-ante
all life insurance products (except health)
5
13
471
Non-life
Ex-ante
a broad coverage of non-life insurance policies
MT
1
3
3,574
Life
Ex-ante
all life and health insurance products
1
2.6
311
Non-life
Ex-ante
a broad coverage of non-life insurance products
NL
55
60
74,188
non-life
Ex-ante
motor third-party liabilities insurance
PL
39
28
43,033
life +
non-life
Ex-post
all life insurance products and compulsory non-life insurance policies
PT
587
844
3,779
Non-Life
Ex-ante
motor third-party liabilities insurance and worker's compensation insurance
RO
130
227
7,242
Life +
Non-life
Ex-ante
a broad coverage of life and non-life insurance policies
SI
9
7
2,432
Non-Life
Ex-ante
motor third-party liabilities insurance
SK
28
24
6,437
Non-life
Hybrid
motor third-party liabilities insurance
UK
(1,766)
(939)
1,571,100
Life
Ex-post
a broad coverage of life and health insurance policies
(316)
(816)
129,700
Non-life
Ex-post
a broad coverage of non-life insurance policies
NO
2 (290)
2 (335)
17,424
non-life
Hybrid
a broad coverage of non-life insurance policies
Source: ESRB, based on national sources.
Note: In order to improve readability, most figures were rounded to zero decimal places. Figures in brackets represent the amounts that could be
levied, not actual funds which were available on balance sheets of IGSs at end-dates. Figures with a minus sign represent a deficit. NA refers to data
which is not available and NA* is used when IGS did not exist in that specific period.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 55
Some IGSs have additional powers to apply resolution tools. However, a single resolution tool
is envisaged in the national legislation for IGSs, which limits flexibility in the use of the most
appropriate tool in any given circumstances. For example, in Greece the main objective of the IGS
in the life insurance sector is to facilitate the portfolio transfer of a failed insurer (or a part of it). If a
transfer is not successful, the IGS will compensate policyholders within certain limits. In Spain, in
comparison, a special winding-down procedure has been established which allows an orderly and
faster process characterised by better policyholder protection and compensation, although no
predetermined level of protection is guaranteed.
74
Similarly, a specific insolvency regime has been
envisaged in Austria.
75
The use of IGSs for resolution funding purposes cannot be viewed as a credible option at
this juncture. Table 5 provides data on the estimated funds available under national IGSs.
Although the data has some limitations it is still possible to draw a number of conclusions. First,
there are considerable variations in available funds, largely related to the difference in coverage
provided by national IGSs and the size of respective markets. Given this limited coverage of IGSs
in many EU Member States, the IGSs might have neither the legal power nor the capacity to
provide sector-wide relief to policyholders. Second, notwithstanding the fact that the financing of
portfolio transfer by IGSs is not allowed in all EU Member States, the practical application of this
resolution tool might be further hampered by the limited coverage of IGSs in many countries. It
might not be sufficient to apply the resolution tool to the limited portfolio of a failing insurer, while
the use of IGSs for a broader portfolio might not be possible given a low level of available funds
and the limited coverage of the IGSs. Lastly, the use of IGSs for resolution funding purposes, if
included in the legislation, is limited to the objective of policyholder protection. As such, it is thought
that their use might be constrained in respect of financial stability objectives and that these funds
could only be used for failing primary insurers (and not for a failing reinsurer).
5.3 Ex post and ex ante industry financing
136. The ultimate aim of a funding arrangement is to ensure that sufficient funds are in
place to meet the contractual obligations vis-à-vis policyholders in the case of an IGS
and to contribute towards resolution costs in the case of an RF. Both IGSs and RFs incur
direct and indirect costs. Direct costs include administration costs and, in the event of a
failure, the costs of policyholder coverage (for the IGSs) and the costs of resolution (for the
RF). Indirect costs, on the other hand, could be from negative market impacts (e.g. moral
hazard or possible adverse effects on market structure and competition). The indirect costs
should be weighed against the expected benefits, e.g. in the case of an IGS in terms of
consumer protection and market confidence (Oxera 2007). Moreover, operational costs
increase following a failure. In the absence of failure administration costs are minimal
(especially for ex post funded frameworks), and the guarantee costs (IGSs) and resolution
costs (RF) are non-existent. However, administrative costs would increase in the event of a
74
In the past, the protection reached up to 100% on average.
75
In AT, insurance companies have to establish a premium reserve fund (“Deckungsstock”) for life and health business. This
fund shall be administered separately from other assets and constitutes a special fund (“Sondermasse”) in case of
bankruptcy. Insurance claims shall have priority over the remaining claims in bankruptcy. “Deckungsstock” assets may only
be subject to execution for the benefit of insurance claims.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 56
failure, although they would still be small compared with the actual cost of providing the
guarantee or of financing a resolution.
76
It is also worth highlighting that different resolution
tools are associated with different cash flows.
137. The fees levied on the industry may be passed on to policyholders. Depending on the
level of competition in the insurance market, the costs of any industry-financed funding
arrangement could be passed on to policyholders through higher premiums. The EU market
seems to be, in general, competitive, which potentially limits the possibility of passing the
additional costs of the new measures on to policyholders. For example, EIOPA states that
intense competition has driven the industry to considerably lower premium rates, particularly
for non-life insurance sector items such as motor insurance (EIOPA 2015). Coupled with the
current low growth environment, the pass-through to policyholders would probably be small.
138. Funding can take different forms. The IGSs and RFs could be financed: (i) on an ex ante
basis, through regular contributions from insurers, before any insurer failure takes place;
(ii) on an ex post basis, through recoupment mechanisms, after a failure takes place; or
(iii) through a combination of the two, with a hybrid fund model holding a pre-fund, also with
the ability to post-fund. In a hybrid model, when an ex ante fund has reached a target level,
the annual contributions can be reduced or even eliminated. An example of this type of
scheme is the “Protektor” IGS for life insurers in Germany which is mainly funded ex ante, but
also has an ex post element. Furthermore, contributions could be calculated using different
sensitivities to risk weighting. The fees could be based on (i) contributions or levies calculated
in proportion to premium income, (ii) the significance and risk profile of an insurer, (iii) the size
of liabilities, or (iv) a fixed charge per contract.
139. Ex ante funding comes with several advantages, but also has some drawbacks. In terms
of liquidity, it provides cash resources within a short period of time to meet liquidity needs in
the initial phase of a failure. In terms of credibility, it demonstrates that an IGS or an RF has
the resources required to protect policyholders and/or to contribute towards funding the
resolution process, particularly when no government guarantee is provided. Moreover, ex ante
funding may be considered to be fairer to the industry, since all companies bear the cost of
failure, including the failing company. In terms of procyclicality, raising ex ante funds in
profitable times means creating a financial buffer for times of stress or failure of an insurer.
Little or no additional burden is thus placed on the surviving insurers, reducing the probability
of a further failure. As such, the cost can be absorbed and allocated more easily. Moreover, if
combined with risk-weighted or risk-sensitive contributions, insurers taking a greater risk and
thus exposed to a higher probability of failure can be charged ex ante through a higher
contribution, therefore contributing further to a fairer distribution of costs and incentivising the
insurer to adopt a modest risk profile. Ex ante funds could also be a source of income,
particularly in respect of possible investment income on raised funds, which could be used to
cover operational costs. On the other hand, ex ante RFs or IGS are associated with possible
higher administrative and operational costs for their management, e.g. in terms of collecting
contributions or managing investments. This drawback could be (at least partially) offset by
yields on investments.
76
In the past, guarantee costs in the EU were not high (generally, a fraction of 1% of gross premiums), but in the event of a
large failure, or multiple failures, the short-term costs of providing a guarantee are likely to be far higher than those
experienced previously.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 57
140. There are certain arguments favouring ex post funding, although this form of funding is
also associated with financial stability risks. First, it might guarantee a more efficient
allocation of capital. As regular contributions are not needed, insurers could use those funds
more efficiently in their operations to generate income and to meet liquidity and solvency
ratios. Second, ex post funding is associated with lower operational and administrative costs
in ‘good times’, and as such IGSs or RFs do not incur any investment management or
administrative costs. Third, it could be also argued that the cost of a post-fund may be lower
for the industry as only the actual amount required would be collected. The disadvantages are
associated with timing issues, namely an inability to raise sufficient funds, and the fact that
any levies on surviving entities could lead to possible contagion due to the additional burden
in times of market stress.
141. A balance could be struck between funding ex ante and ex post. A hybrid model could be
used to keep the level of the ex ante fund adequate to meet all immediate liquidity needs and
keep the ex post fund mechanism available for less likely but plausible failures.
Box 8
Potential impact of a large primary insurer insolvency
This analysis discusses the potential impact of capital shortfalls related to the failure of the
largest primary insurer in individual EU Member States. The analysis is based on the
assumption that alternatives to the winding-down of a large insurer would be explored, given the
impact such a failure would have on financial stability in an EU Member State. Authorities would
consider different funding sources that could be needed to apply resolution tools and to provide
policyholder compensation. The example chosen does not prejudge any large primary insurer
insolvency and serves only as an illustration. Although the impact of other possible scenarios is not
discussed (such as a scenario in which multiple insurers fail simultaneously), the findings with
regard to possible funding sources could also be applicable to other scenarios.
An example has been chosen examining a 5% and a 10% shortfall in the assets of the
largest primary insurer. The table below examines a situation where an IGS or RF has been
called upon to fund the failure of the largest primary insurer in each Member State, using an
example where a 5% or 10% shortfall in the assets of the largest insurer needs to be covered. For
each EU Member State the size of the insurance market, the size of the largest insurer, the number
of market participants, and (where data are available) the amount of surplus capital in the market is
shown. In order to increase comparability across the EU, the whole primary insurance sector is
considered as a whole.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 58
Table 6
Significance of the largest primary insurer assets in respect of national insurance market
own funds
Member
State
Largest insurer (by total assets)
Market
Capital shortfall over
Excess capital
Total
assets
(€m)
5% capital
shortfall
10% capital
shortfall
Size by total
assets (€m)
Excess
capital (€m)
No of
participants
5% CS as
% of EC
10% CS as
% of EC
AT
15,130
756
1,513
116,309
9,771
38
8
15
BE
76,627
3,831
7,663
322,670
14,878
69
26
52
CY
550
28
55
2,577
333
25
8
17
CZ
3,579
179
358
18,635
2,276
53
8
16
DE
188,142
9,407
18,814
1,154,555
131,270
298
7
14
DK
58,708
2,935
5,871
385,677
31,590
38
9
19
EE
513
26
51
1,680
220
18
12
23
EL
3,418
171
342
15,609
777
40
22
44
ES
64,632
3,232
6,463
308,031
25,109
185
13
26
FI
6,645
332
665
39,116
8,385
49
4
8
FR
887,070
44,354
88,707
2,935,495
106,203
32
42
84
HR
1,102
55
110
4,959
873
24
6
13
HU
1,266
63
127
7,876
332
29
17
33
IE
38,810
1,941
3,881
247,353
9,369
147
21
41
IT
107,323
5,366
10,732
1,177,304
115,915
124
5
9
LT
256
13
26
1,174
274
10
5
9
LU
28,926
1,446
2,893
172,811
8,054
73
18
36
LV
140
7
14
644
45
8
16
31
MT
1,744
87
174
3,885
397
8
22
44
NL
96,330
4,817
9,633
496,205
25,783
146
19
37
PL
10,122
506
1,012
43,033
8,360
60
6
12
PT
14,021
701
1,402
53,319
1,185
46
59
118
RO
707
35
71
3,621
299
32
12
24
SE
80,681
4,034
8,068
444,641
115,942
150
3
7
SI
2,708
135
271
6,629
579
15
23
47
SK
2,272
114
227
6,533
573
16
20
40
UK
344,999
17,250
34,500
1,700,800
55,225
286
31
62
NO
56,463
2,823
5,646
151,754
10,508
71
15
31
Source: ESRB based on national data.
Note: data as at end-2015, but RO used Solvency II data for 1 January 2016. Excess capital (Market own funds) expressed in terms of SII eligible
own funds whenever possible. For DK, pension funds are also covered and only non-life insurers with non-life assets worth more than EUR 500m are
covered. In EE, five branches are also covered.
The impact on IGSs of the failure of a large primary insurer could be significant, although
the overall effect is difficult to assess. IGSs provide policyholder relief and, in some specific
cases, they could be allowed to finance some resolution tools (such as a portfolio transfer).
Comparing the amount of funds available in the IGSs (see Box 7) to the amount of capital shortfall
could provide an estimate of the funds available in the IGS to cover the shortfall. There are obvious
limitations to this data – for example a shortfall of 5% (10%) in the capital of the largest primary
insurer would not translate into the same amount of IGS payouts, since not all policies are
necessarily covered by the IGSs and compensation payments could be capped. Moreover, the IGS
funding could be sufficient if long-term claims are paid only when they fall due. Also, a large capital
shortfall on protected policies could deplete available funds across the insurance sector and it could
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Resolution funding 59
take several years to fully refund the IGSs (see the example of IE with a deficit for its IGS in 2015 in
Box 7). Further work would be useful to better understand the potential impact on individual IGSs of
a significant primary insurance failure.
Moreover, given the absence of an RF in the majority of EU Member States, the authorities
would be left with no sources for funding the resolution process, even if there were
resolution tools in place. This means that a failed insurer would need to absorb the costs, which
could further exacerbate losses incurred by policyholders. At this juncture, where there are no
resolution tools other than liquidation (which might lead to severe contagion effects) and no related
sources of funding available to national authorities, a bailout by the state would seem to be the
most likely option where there are financial stability concerns caused, for example, by the failure of
a large insurer.
77
This analysis stresses that an effective RR framework should also take funding
arrangements into consideration.
Time provides an opportunity for authorities to strengthen funding arrangements. As the
analysis shows, some EU Member States would have difficulty affording the failure of their largest
insurer, without the allocation of losses to policyholders or a bailout. However, the specificity of the
insurance sector is that time might be a less pressing issue. For example, in the case of the impact
of the LIR, this would be a “slow burn” issue. Time therefore provides an opportunity for the
authorities to create new funded schemes or strengthen existing schemes, both IGSs and RFs, and
to build up their target levels ahead of an insurer’s failure.
77
It is worth noting that EU state aid rules must be complied with before a state bailout is approved by the European
Commission.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Role of the macroprudential authority 60
142. Macroprudential policies and recovery and resolution (RR) policies can complement
and reinforce each other, and there is the potential for each of them to support the
other’s objectives. A macroprudential authority can propose or take measures itself to
address sources of systemic risk in the insurance sector, thus reducing the likelihood of
failures. An effective RR framework could reduce the likelihood of failures of individual entities
and could reduce any impact on financial stability, e.g. in terms of systemic contagion, in the
event of a failure. Moreover, the resolution framework could strengthen market discipline and
reduce incentives to take excessive risks, thus mitigating the need for intervention by the
macroprudential authority. In contrast, in the absence of an effective RR framework and in the
presence of financial stability concerns, this situation might require more forceful
macroprudential action. Robust resolution frameworks improve the ability of the authorities to
deal credibly with individual weak insurers, thereby helping to guarantee a more efficiently
structured system-wide regime.
143. Furthermore, crisis management requires close coordination among all authorities,
both ex ante as well as during the crisis. The 2008 financial crisis demonstrated severe
weaknesses in the financial system’s safety net. While the deficiencies were evident almost
worldwide, the most severe problems were encountered in Europe, largely due to the weak
coordination, consultation and development of coherent strategies to deal with the crisis (COM
2010a and COM 2010b). Financial instability is made worse by uncertainty, and crisis
management and resolution are complex processes, requiring the involvement of different
types of expertise. It is therefore important to identify and agree upfront on the relevant
organisations involved in the process, including their roles and responsibilities, as well as to
set out a framework for the decision-making process for all parties involved in a crisis
situation. A clear division of responsibilities and actions to be taken is also important for
transparency and accountability, and to ultimately improve the understanding of policy actions
by the financial sector and the public at large.
144. Ongoing interaction is desirable between all the authorities involved. The supervisory,
macroprudential and resolution authorities play different roles in respect of their tasks,
functions and powers. It is therefore crucial that they cooperate closely and exchange
information for the resolution process to be efficient. Where an individual insurer’s distress is
linked to wider market turmoil or macroeconomic issues, or the insurer itself is systemically
important, the macroprudential authority may be helpful in advising on the consequences of
various measures taken by the supervisory or resolution authority.
145. Cross-border and cross-sectoral perspectives and coordination are also crucial. The
recognition of losses in a predictable and transparent manner, and their allocation across
insurers in the context of cross-border resolution, inherently requires coordination and
agreement among national authorities. Delaying this agreement could ultimately result in
resolution being disruptive when it is attempted or resolution not being attempted at all for fear
of a backlash. National initiatives, which aim at improving resolution frameworks for insurers in
individual jurisdictions, are likely to fall short of what is required to guarantee the consistent
application of rules and the viability of resolution regimes in different countries, including the
maintenance of the integrated operations of parent institutions with subsidiaries in different
countries and the ex ante coordination of loss allocation between these countries. If a planned
measure by a resolution authority could also have spillover effects, the macroprudential
authority should be involved, in order to guarantee an adequate degree of coordination and
Section 6
Role of the macroprudential authority
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Role of the macroprudential authority 61
limit possible negative spillover effects. Moreover, the macroprudential authority is also well
equipped to give an opinion on the cross-sectoral effects of measures taken by the resolution
authority.
146. Currently, considering financial stability does not always feature as one of the
objectives of national RR frameworks for insurers across the EU. Section 2 explained
that the insurance sector can pose systemic risks. Moreover, according to the IAIS, a benign
record showing little systemic risk from insurers in the past does not guarantee the absence of
systemic risk potential in the future, as empirical assessments of the systemic importance of
insurers can change over time (IAIS 2015). The current LIR environment is a good example of
how this evaluation can change. However, the resolution frameworks currently applicable in
several EU Member States do not generally recognise financial stability objectives in the
resolution process. It follows that no formal involvement of the macroprudential authority in the
RR process is envisaged. Harmonising the RR objectives across the EU with the FSB
approach, by recognising policyholder protection and financial stability as objectives in the
national RR frameworks, would also help to clarify the role of the macroprudential authority in
the RR process.
147. The macroprudential authority should be involved in the RR process in case there are
financial stability implications. Close coordination between the resolution authority and the
macroprudential authority should be ensured so that their respective actions can be
coordinated and reinforced. The resolution authority should be mandated to liaise with the
national macroprudential authority, informing it in advance of any significant actions it plans to
propose, in case there are any financial stability implications. Although the resolution
authorities are mandated to take the lead in coordinating the overall policy response, the
macroprudential authority could support the process by providing advice, based on its
assessment of the evolution of the level and sources of systemic risk. It may also use
available instruments to contain the amplification of risks and spillovers.
148. The involvement of the macroprudential authority is foreseen by the current
supervisory regime. Solvency II foresees the possible consultation of the ESRB in
“exceptional adverse situations”. Solvency II grants national authorities the power to allow
extended recovery periods for insurers that breach their capital requirements when this is
related to a wider exceptional adverse situation in the markets declared by EIOPA.
78
This is
aimed at reducing the possible procyclical effects of a breach of SCR, such as distressed
sales of assets on financial markets. Normally insurers have six months (extendable to a
maximum of nine months) to return to financial soundness in accordance with a recovery plan,
a condition which is approved by the supervisory authority. However, in the case of an
exceptional adverse situation, declared by EIOPA, and where appropriate after consultation
with the ESRB, in which the financial situation of insurers representing a significant share of
the market, or of the affected lines of business, is seriously or adversely affected, the
supervisory authorities concerned could extend the recovery period by a maximum of seven
years. This is particularly relevant in the current economic climate, as the Directive lists a
“persistent low interest rate environment” as one of the conditions for an exceptional adverse
situation.
149. However, it should be noted that only a minority of EU Member States have a dedicated
resolution authority for insurers. This could lower the chances of developing ex ante
coordination procedures with macroprudential authorities in the event of a resolution, both at
national and at cross-border level.
78
See Article 138(4) of the Solvency II Directive.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Conclusions and possible ways forward 62
150. In recent years increasing attention has been paid to recovery and resolution (RR)
frameworks for insurers. The FSB has developed its Key Attributes on effective resolution
regimes for the insurance sector, requiring these to be applied for global systemically
important insurers (G-SIIs). In a 2015 report, the ESRB concluded that the current RR
regimes in place at national level are unlikely to be fit to handle all distress scenarios related
to the current low interest rate (LIR) environment and the “double hit”, and recommended the
development and operationalisation of effective RR frameworks for insurers. EIOPA recently
published a Discussion Paper and an Opinion outlining the rationale for, and potential building
blocks of, a harmonised EU RR framework for the insurance sector. At national level, RR
frameworks have been reviewed and reinforced in Romania, France and the Netherlands.
151. The driving force behind the development of RR frameworks is the assessment that
regular insolvency procedures may not always meet resolution objectives. The key
resolution objectives are the protection of policyholders and financial stability, with the aim of
ensuring the continuity of critical functions, avoiding disorderly failures and preventing
contagion to other parts of the financial system. These objectives should also consider
possible cross-border implications for Europe. Regular insolvency procedures do not include
these specific objectives, and without an effective RR framework in place, a lack of legal
powers can constrain authorities when faced with a failing insurer.
152. The identification and protection of critical functions and the prevention of spillover
effects are relevant to macroprudential supervision. The identification of the critical
functions of an insurer depends on the impact of its failure and its substitutability at the time of
default. Examples include protection from risks in areas such as trade credit, marine and
aviation, and the provision of retirement income. Because of the interconnectedness of
insurers in the financial system and their high share of cross-border business, the spillover
effects of distress and, in the worst case, default, are possible, both within the financial system
and across borders. These should be identified and managed.
153. The current environment of low interest rates highlights the need for strengthening the
RR framework. The probability of an insurer facing financial distress or default increases in
this environment. The deterioration of the balance sheets of insurers may occur gradually,
allowing the authorities to prepare for different adverse scenarios. This highlights the
importance of pre-emptive RR plans as well as the need to strengthen early intervention
powers. If recovery and early intervention measures fail, the authorities should be prepared for
the insolvency of a large insurer and for sector-wide insolvencies. Moreover, an LIR
environment may make it difficult to use traditional resolution strategies (such as run-off and
portfolio transfer), given the potential solvency shortfalls. Given the type of business
conducted by life insurers, including the provision of retirement income, pressure on the state
to bail out an insurer in distress cannot be excluded.
154. A credible and consistent RR framework could help mitigate systemic risks. RR plans
could identify critical functions and contagion channels. Resolution and early intervention tools
could enable such functions to be protected and prevent contagion. Moreover, they could
allocate the costs of resolution to where the authorities believe they should be borne,
establishing an escalation ladder of intervention. Coordination arrangements could prevent
spillover effects across sectors and borders. EIOPA concluded recently that current national
Section 7
Conclusions and possible ways forward
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Conclusions and possible ways forward 63
RR frameworks are fragmented and do not contain all these elements (EIOPA 2016b and
EIOPA 2017).
155. A patchwork of RR frameworks at national level may increase existing fragmentation
and exacerbate cross-border contagion. The EU insurance sector has a higher proportion
of cross-border activity, relative to its total activity, than the banking sector. National RR
frameworks do not take cross-border considerations into account in the application of
resolution tools. Moreover, a patchwork of national RR regimes and insurance guarantee
schemes (IGS) would not contribute to financial integration in Europe and might create
uncertainty for market participants over the level of policyholder protection and RR tools that
could be applied in the event of an insurer failing.
156. This calls for the development of a harmonised RR framework within the EU. As the
risks of a default of an insurer depend on time and context, the authorities should have at their
disposal a broad and flexible set of tools, which could be used in parallel. Several resolution
scenarios should therefore be feasible, including a run-off, a portfolio transfer, a restart after
bail-in, and a liquidation. In light of current challenges stemming from, for example, the LIR
environment, the framework should have broad coverage, without any prejudice to the
principle of proportionality. This process would also result in the harmonisation of RR
objectives and the definition of triggers across the EU, as well as ensure the consistency of
RR regimes for financial conglomerates.
157. The discussion on the RR framework should also encompass how resolution is funded.
The application of any resolution tool involves certain costs, such as the costs related to
portfolio transfer or to setting up a bridge insurer. In order to have in place a reliable
framework for dealing with the failure of an insurer without needing to use public funds, the
resolution authority must have available resources at its disposal. Setting up a resolution fund
or expanding the role of IGSs to contribute towards the use of resolution tools are factors
which could be seen as part of the harmonised RR framework.
158. Losses incurred through the failure of an insurer should be allocated first to
shareholders and creditors other than policyholders. An RR framework for insurers
should allocate losses to the risk-bearing creditors of insurers. This would improve market
discipline and protect critical functions. Given the business model of insurers, however, there
might not be sufficient debt allocated to fully protect policyholders. Although requiring insurers
to hold a minimum level of loss absorption capacity might protect policyholders, this might
create more downsides than benefits. In particular, it would require changes to the existing
business models and the funding structure, and would also increase interconnectedness
within the financial system. Losses should only be allocated to policyholders as a last resort
when all other resources have been exhausted and if the stability of the financial system is at
stake or in case they would face larger losses than under the regular insolvency procedure.
159. Policyholder protection, in the form of compensation by an IGS for losses incurred, is
beyond the scope of this report, but needs to be considered further. Policyholders would
be protected by the NCWO principle from incurring losses greater than those resulting from
ordinary insolvency procedures. Whether and to what degree they would be protected from
these losses would depend on the existence and scale of an IGS. IGSs help create
confidence in the insurance sector and some IGS exist at national level, albeit with differing
scope and coverage. In contrast to the situation for bank depositors, there is no harmonised
level of protection for policyholders in the event of the failure of an insurer, which results in a
variable degree of consumer protection across the EU. The issue deserves further analysis.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Conclusions and possible ways forward 64
160. Against this background, the report advocates the development of a harmonised
effective RR framework for insurers across the EU. This includes the following:
• Existing RR frameworks should be evaluated and, if appropriate, enhanced and
harmonised at EU level. Furthermore, efforts should be made to ensure their consistent
implementation.
• The existing RR toolkit should be expanded. A majority of ESRB member institutions
take the view that this should include giving resolution authorities the power to modify
the terms of existing contracts as a measure of last resort and subject to adequate
safeguards.
• The RR framework should cover the whole insurance sector, while allowing for
proportionality.
• The financial stability objectives of the RR framework should be recognised, with a
majority of ESRB member institutions taking the view that it should be put on an equal
footing with the objective of policyholder protection. In addition, the interactions of the
resolution authority with the macroprudential authorities should also be clarified.
• Work on RR frameworks should go hand-in-hand with discussion of resolution should be
funded.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
References 65
Abad, J., Aldasoro, I., Aymanns, C., D’Errico, M., Fache Rousova, L., Hoffmann, P., Langfield, S.,
Neychev, M., and Roukny, T. (2016), “Shedding light on dark markets: First insights from the new
EU-wide OTC derivatives dataset”, ESRB Occasional Paper, No 11, September.
Alves, I., Brinkhoff, J., Georgiev, S., Heam, J-C., Moldovan, I., and Scotto di Carlo, M. (2015),
“Network analysis of the EU insurance sector”, ESRB Occasional Paper, No 7, July.
A.M. Best (2010), Global Reinsurance – Segment Review, September.
A.M. Best (2014), Global Reinsurance – Segment Review, September.
A.M. Best (2015), Global Reinsurance – Segment Review, September.
Baluch, F., Mutenga, S., and Parsons, C. (2011), “Insurance, Systemic Risk and the Financial
Crisis”, The Geneva Papers, No 36, pp. 126-163.
Bank of England (2015), “Implementing the Bank Recovery and Resolution Directive – Response to
CP 13/14”, Policy Statement PS1/15, January.
Bank of England and Procyclicality Working Group (2014), Procyclicality and structural trends in
investment allocation by insurers and pension funds: A discussion paper, July.
Bijlsma, M and Vermeulen, R (2015), “Insurance companies’ trading behaviour during the
European sovereign debt crisis: Flight home or flight to quality?”, De Nederlandsche Bank Working
Paper, No 468.
BIS (2013), “Global systemically important banks: updated assessment methodology and the
higher loss absorbency requirement”, Basel Committee on Banking Supervision, July.
BIS (2016), “Standard TLAC holdings. Amendments to the Basel III standard on the definition of
capital”, Basel Committee on Banking Supervision, October.
Commission Delegated Regulation (EU) 2015/2303 of 28 July 2015 supplementing Directive
2002/87/EC of the European Parliament and of the Council with regard to regulatory technical
standards specifying the definitions and coordinating the supplementary supervision of risk
concentration and intra-group transactions, Official Journal of the EU, OJ L 326, 11.12.2015,
pp. 34-38.
Cummins, J.D. and Weiss, M.A. (2014), “Systemic Risk and the U.S. Insurance Sector”, The
Journal of Risk and Insurance, Vol. 81, No 3, September, pp. 489-528.
Directive 2014/59/EU of the European Parliament and of the Council of 15 May 2014 establishing a
framework for the recovery and resolution of credit institutions and investment firms and amending
Council Directive 82/891/EEC, and Directives 2001/24/EC, 2002/47/EC, 2004/25/EC, 2005/56/EC,
2007/36/EC, 2011/35/EU, 2012/30/EU and 2013/36/EU, and Regulations (EU) No 1093/2010 and
(EU) No 648/2012, of the European Parliament and of the Council (BRRD), Official Journal of the
EU, OJ L 173, 12.6.2014, pp. 190–348.
Directive 2002/87/EC of the European Parliament and of the Council of 16 December 2002 on the
supplementary supervision of credit institutions, insurers and investment firms in a financial
conglomerate and amending Council Directives 73/239/EEC, 79/267/EEC, 92/49/EEC, 92/96/EEC,
93/6/EEC and 93/22/EEC, and Directives 98/78/EC and 2000/12/EC of the European Parliament
and of the Council (FICOD), Official Journal of the EU, OJ L 35, 11.2.2003, pp. 1–27.
References
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
References 66
Directive 2009/138/EC of the European Parliament and of the Council of 25 November 2009 on the
taking up and pursuit of the business of Insurance and Reinsurance (Solvency II), Official Journal of
the EU, OJ L 335, 17.12.2009, pp. 153-307.
Directive 2009/103/EC of the European Parliament and of the Council of 16 September 2009
relating to insurance against civil liability in respect of the use of motor vehicles, and the
enforcement of the obligation to insure against such liability, Official Journal of the EU, OJ L 263,
7.10.2009, pp.11-31.
Dungey, M., Luciani, M. and Veredas, D. (2014), “The Emergence of Systemically Important
Insurers”, CIFR paper No 38, September.
ECB (2016), Financial Stability Review, November.
EGBPI (2015), Minutes of the EGBPI meeting on 5 March 2015.
EIOPA (2011), Report on cross-border cooperation mechanisms between Insurance Guarantee
Schemes in the EU, EIOPA-TFIGS-11/007, June.
EIOPA (2012a), Report on the Role of Insurance Guarantee Schemes in the Winding-Up
Procedures of Insolvent Undertakings in the EU/EEA, EIOPA-TFIGS-12/007, May.
EIOPA (2012b), Consultation on a possible recovery and resolution framework for financial
institutions other than banks - EIOPA Response, December.
EIOPA (2013), Survey on Crisis Prevention, Management and Resolution. Preparedness of NCAs,
November.
EIOPA (2014a), Opinion of 24 November 2014 on Sound principles for Crisis Prevention,
Management and Resolution preparedness of NCAs, EIOPA-BoS-14/120, November.
EIOPA (2014b), Guidelines on operational functioning of colleges, EIOPA-BoS-14/146.
EIOPA (2015), Financial Stability Report, EIOPA-FSC-15-67, May.
EIOPA (2016a), Report on long-term guarantees measures and measures on equity risk,
EIOPA-BoS-16/279.
EIOPA (2016b), Discussion paper on potential harmonisation of recovery and resolution
frameworks for insurers, EIOPA-CP-16/009, December.
EIOPA (2016c), 2016 EIOPA Insurance Stress Test Report, EIOPA-16/302, December.
EIOPA (2017), Opinion to Institutions of the European Union on the Harmonisation of Recovery and
Resolution Frameworks for (Re)insurers across the Member States, EIOPA-BoS/17-148, July.
ESRB (2015), Report on systemic risks in the EU insurance sector, December.
ESRB (2016a), ESRB policy beyond banking: an ESRB strategy paper, July.
ESRB (2016b), Macroprudential policy issues arising from low interest rates and structural change
in the EU financial system, November.
ESRB (2017), ESRB Secretariat staff response to the EIOPA Discussion Paper on the potential
harmonisation of recovery and resolution frameworks for insurers (EIOPA-CP-16-009), February.
European Commission (2010a), White paper on Insurance Guarantee Schemes, COM(2010) 370.
European Commission (2010b), Commission Staff Working Document Impact Assessment –
Accompanying document to the White Paper on Insurance Guarantee Schemes
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
References 67
European Commission (2010c), Commission Staff Working Document Impact Assessment –
Accompanying document to the Proposal for a Directive of the European Parliament and of the
Council amending Directives 98/78/EC, 2002/87/EC and 2006/48/EC as regards the supplementary
supervision of financial entities in a financial conglomerate, SEC/2010/979, August.
European Commission (2012), Consultation on a possible recovery and resolution framework for
financial institutions other than banks.
European Commission (2016), Frequently Asked Questions: Capital requirements (CRR/CRD IV)
and resolution framework (BRRD/SRM) amendments, dated 23 November 2016.
French, A., Vital, M. and Minot, D. (2015), “Insurance and financial stability”, in Bank of England
(2015), Quarterly Bulletin, Vol. 55 No 3, pp. 242-258.
FSB (2011), Key Attributes of Effective Resolution Regimes for Financial Institutions.
FSB (2014a), Key Attributes of Effective Resolution Regimes for Financial Institutions, October.
FSB (2014b), “Recovery and Resolution Planning for Systemically Important Insurers: Guidance on
Identification of Critical Functions and Critical Shared Services”, Consultative Document, October.
FSB (2013), Global systemically important insurers (G-SIIs) and the policy measures that will apply
to them, July.
FSB (2014c), 2014 update of list of global systemically important insurers (G-SIIs), November.
FSB (2015), 2015 update of list of global systemically important insurers (G-SIIs), November.
FSB (2016a), 2016 list of global systemically important insurers (G-SIIs), November.
FSB (2016b), Developing Effective Resolution Strategies and Plans for Systemically Important
Insurers, June.
IAIS (2012a), Reinsurance and Financial Stability, July.
IAIS (2012b), Global Insurance Market Report, October.
IAIS (2011), Insurance and Financial Stability, November.
IAIS (2011b), Issues Paper on Resolution of Cross-Border Insurance Legal Entities and Groups,
June.
IAIS (2013), Issues Paper on Policyholder Protection Schemes, October.
IAIS (2015), Frequently Asked Questions for IAIS Financial Stability and Macroprudential Policy
and Surveillance (MPS) Activities, updated 25 June 2015.
IAIS (2016a), IAIS Newsletter, Issue 57, October 2016.
IAIS (2016b), Systemic Risk from Insurance Product Features, June.
IAIS (2016c), Global Systemically Important Insurers: Updated Assessment Methodology, June.
IMF (2016a), Global Financial Stability Report: Potential Policies for a Successful Normalization,
April.
IMF (2016b), Global Financial Stability Report: Fostering Stability in a Low-Growth, Low-Rate Era,
October.
Joint Committee of the European Supervisory Authorities (ESAs) (2016), List of Identified Financial
Conglomerates, as per 31 December 2014 figures, JC-2016-077, October.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
References 68
Kelliher, P.O.J., Barlett, D.L., Chaplin, M., Dowd, K. and O’Brien, C. (2005), Liquidity
Management in UK Life Insurance: A Discussion Paper, Faculty and Institute of Actuaries, April.
Oxera (2007), Insurance guarantee schemes in the EU: Comparative analysis of existing schemes,
analysis of problems and evaluation of options, November.
Paulson, A., Plestis, T., Rosen, R., McMenamin, R. and Mohey-Deen, Z. (2014), “Assessing the
Vulnerability of the U.S. Life Insurance Industry”, in Biggs, J.H. and Richardson, M. (eds.),
Modernizing Insurance Regulation, New Jersey, Wiley. pp. 61-84.
PWC (2016), “Unlocking value in run-off”, A Survey of Discontinued Insurance Business in Europe,
10th edition, September.
Stringa, M. and Monks, A. (2007), “Inter-industry contagion between UK life insurers and UK banks:
an event study”, Bank of England Working Paper, No 325, May.
Schoenmaker, D. and Sass, J. (2014), “Cross-border Insurance in Europe”, DSF Policy Paper,
No 45, November.
The Geneva Association (2012), Insurance and Resolution in Light of the Systemic Risk Debate: A
contribution to the financial stability discussion in insurance, February.
Timmer, Y. (2016), “Cyclical investment behavior across financial institutions”, ESRB Working
Paper, No 18, July.
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Abbreviations 69
ART Alternative risk transfer
BCBS Basel Committee on Banking Supervision
BRRD Bank Recovery and Resolution Directive
CCP Central counterparty
CDS Credit default swaps
CMG Crisis Management Group
COM European Commission
ComFrame Common Framework for the Supervision of Internationally Active Insurance Groups
EEA European Economic Area
EGBPI Expert Group on Banking, Payments and Insurance
EIOPA European Insurance and Occupational Pensions Authority
ESRB European Systemic Risk Board
EU European Union
FICOD Financial Conglomerates Directive
FSB Financial Stability Board
G-SIB Global systemically important bank
G-SII Global systemically important insurer
IAIG Internationally Active Insurance Group
IAIS International Association of Insurance Supervisors
ICP Insurance Core Principle
ICPF Insurance corporations and pension funds
IGS Insurance guarantee scheme
IMF International Monetary Fund
KAs FSB Key Attributes
LIR Low interest rate
M3E3 Module 3 Element 3
MOF Market own funds
MREL Minimum requirement for own funds and eligible liabilities
NCA National competent authority
NCWO No creditor worse off
NTNI Non-traditional Non-insurance activities
OECD Organisation for Economic Co-operation and Development
ORSA Own risk and solvency assessment
RF Resolution fund
RMBS Residential mortgage-backed securities
RR Recovery and resolution
SCR / MCR Solvency Capital Requirement / Minimum Capital Requirement
SIFI Systemically important financial institution
SRMR Single Resolution Mechanism Regulation
TLAC Total Loss-Absorbing Capacity
Abbreviations
ESRB
Recovery and resolution for the EU insurance sector: a macroprudential perspective August 2017
Members of the drafting team
Alexia Yiangou
Central Bank of Cyprus
Lucia Orszaghova
ESRB Secretariat
Andrea Austin
Bank of England
Magnus Strömgren
Finansinspektionen (Sweden)
Andreas Baumann
ECB
Mara Aquilani
IVASS (Italy)
Casey Murphy
Bank of England
Matthias Becker
BaFin
Faidra Frygana
ESRB Secretariat
Miriam Brosnan
Central Bank of Ireland
Iulia Moldovan
ECB
Nicolas Strypstein
Nationale Bank van België/Banque Nationale de Belgique
Jeroen Brinkhoff
De Nederlandsche Bank
Simon Pittaway
Bank of England
Juan Zschiesche
EIOPA
Tim O’Hanrahan
Central Bank of Ireland
Lázaro Cuesta
EIOPA
Members of the drafting team
Imprint
© European Systemic Risk Board, 2017
Postal address 60640 Frankfurt am Main, Germany
Telephone +49 69 1344 0
Website www.esrb.europa.eu
All rights reserved. Reproduction for educational and non-commercial purposes is permitted provided that the
source is acknowledged.
The cut-off date for the data included in this report was 15 July 2017.
ISBN 978-92-95210-66-0 (pdf)
DOI 10.2849/779705 (pdf)
EU catalogue No DT-01-17-797-EN-N (pdf)
