Luxembourg
Financial System Stability Assessment
Author: International Monetary Fund. Monetary and Capital Markets Department
Publication Date:
15 May 2017
eISBN:
9781484300374
Language:
English
Keywords:
ISCR; CR; investment funds; financial system; foreign exchange; liquid asset; money market fund; investment industry; well-developed distribution platform; supervisory framework; bank-investment fund surveillance
Download PDF (3.2 MB)

This paper evaluates the stability of the financial system of Luxembourg. Financial soundness indicators for Luxembourg’s financial system, which plays a key role in the intermediation of financial capital, have remained relatively robust in recent years. Following rising asset prices and inflows, the investment fund industry has enjoyed strong growth in assets under management, while exposure to liquid assets has remained steady. An assessment of the financial system’s ability to withstand severe but plausible shocks suggests a good deal of resilience, albeit with some risks. Insurance stress test results indicate that strong initial levels of capital and low guaranteed product exposure offer insulation against market shocks.

Abstract

This paper evaluates the stability of the financial system of Luxembourg. Financial soundness indicators for Luxembourg’s financial system, which plays a key role in the intermediation of financial capital, have remained relatively robust in recent years. Following rising asset prices and inflows, the investment fund industry has enjoyed strong growth in assets under management, while exposure to liquid assets has remained steady. An assessment of the financial system’s ability to withstand severe but plausible shocks suggests a good deal of resilience, albeit with some risks. Insurance stress test results indicate that strong initial levels of capital and low guaranteed product exposure offer insulation against market shocks.

Executive Summary

Financial soundness indicators for Luxembourg’s financial system, which plays a key role in the intermediation of financial capital, have remained relatively robust in recent years. Rising asset prices and inflows have seen the investment fund industry enjoy strong growth in assets under management, while exposure to liquid assets has remained steady. The banking sector, where a relatively large share of liquidity and revenues derive from private banking and fund management activities, has maintained high levels of profitability, capital, liquidity, and asset quality. The insurance industry, which is relatively less exposed to guaranteed products than regional peers, has adjusted to the new regulatory regime, maintaining high profitability and capitalization levels. Complementing reforms at the European level, the national authorities have pursued a domestic reform agenda in recent years, including the adoption of key recommendations from the 2011 FSAP and a strengthening of the Anti-Money Laundering/Combating the Financing of Terrorism regime.

Nevertheless, the defining structural characteristics of the financial system—size and interconnectedness—as well as elevated real estate valuations, give rise to potential vulnerabilities. First, as many Luxembourg foreign bank subsidiaries aggregate liquidity from investment fund and wealth management operations and ‘upstream’ it to their parents abroad, they may be exposed to maturity and currency transformation at the parent level—where exposures can be large (relative to subsidiaries capital) and oversight by the Luxembourg authorities may be limited where parents are not regulated under the Single Supervisory Mechanism. Second, should investment fund liquidity buffers struggle to accommodate large redemption shocks, asset fire sales and a drawdown of local bank deposits could occur (though operational deposits have been broadly stable in past stress periods). Third, a shock to Clearstream, which has a central role in worldwide securities depositary and settlement services, could disrupt numerous institutions and markets it serves. Fourth, after a strong run up in prices and easing in lending standards, the real estate market, to which domestically-oriented banks are most exposed, poses the main home-grown vulnerability.

An assessment of the financial system’s ability to withstand severe but plausible shocks suggests a good deal of resilience, albeit with some risks. High starting levels of capital allow most banks to absorb a large shock in the adverse stress test scenario, while retaining substantial buffers. Bank liquidity displays broad resilience, though some banks could be exposed to wholesale funding drying up or foreign exchange (FX) funding stresses. This should be closely monitored, and the merits of a foreign currency liquidity requirement at the group level should be examined. Insurance stress test results indicate that strong initial levels of capital and low guaranteed product exposure offer insulation against market shocks. Money market fund liquidity and solvency risk appears muted, though continued monitoring of liquidity and concentration risk among select bond funds is warranted. Spillover analysis confirms material cross-border exposures for banks and investment funds, underscoring the need for ongoing vigilance.

The system of prudential oversight in Luxembourg seems to function well, though several cross-cutting themes warrant attention. A continued pivot towards risk-based supervision and further increase in resources for entities engaged in financial stability oversight would be consistent with rising industry assets and regulatory demands. A formal framework should be introduced to govern the relationship between the government and banks with state involvement. The operational independence of the Commission de Surveillance du Secteur Financier (CSSF) and Commissariat aux Assurances (CAA) should be enshrined in law. The intensity of engagement with supervisors in countries where Luxembourg’s investment funds and banks are most active should also be raised.

The new macroprudential policy framework appears to be working well in practice and could be strengthened further. The institutional framework could be enhanced by removing the potential inaction bias arising from unanimous decision making; enshrining in law the lead role of the Banque centrale du Luxembourg (BCL); and awarding it formal powers to issue recommendations. Surveillance of real estate and bank-investment fund vulnerabilities is appropriate, and closing related data gaps is a priority. The instrument toolkit should be expanded to include borrower-based limits, as further macroprudential tightening may be required.

Additional measures could also help mitigate remaining gaps in banking supervision. More frequent on-site inspections, particularly for subsidiaries availing of the waiver to large exposure limits for intragroup transactions, should be introduced, and banks should be required to periodically demonstrate their continued eligibility for the waiver. The authorities are encouraged to harmonize data reporting standards (as intended), and remain vigilant in their monitoring of real estate vulnerabilities.

The regulation and supervision of investment funds and insurers is prudent but could be strengthened. To mitigate liquidity-related vulnerabilities among investment funds, the CSSF should issue industry guidance on liquidity management tools and liquidity stress testing modalities, and develop internal stress testing capacity. Other priorities include increasing the frequency of on-site inspections; introducing comprehensive inspections; engaging with regulators in jurisdictions where delegated activities are prominent; ensuring custodian banks are independent of funds for which they have oversight functions; and issuing guidance on fund directorships. In the insurance sector, introducing a revised early warning system and increasing resources are priorities.

Clearstream (CBL) is assessed as broadly observing international standards for safe and efficient settlement and custody of securities transactions, though further risk mitigation measures are warranted. Priorities relate to reducing CBL’s deposit exposure to select commercial banks (preferably through direct links with Central Securities Depositories and central banks), tightening collateralization arrangements, and further investment in a more distant data center.

The Single Resolution Mechanism and transposition of relevant European directives have reconfigured Luxembourg’s bank resolution framework. While the authorities now have adequate legal powers to resolve failing banks, more work is required to finalize credible resolution plans. Efforts to operationalize bail-in should be complemented by developing the sale of business and bridge bank resolution tools, and by ensuring adequate liquidity funding in resolution. Policies to address intragroup exposures and the transfer of custodian functions in recovery and resolution should be addressed. Additional contingent FX funding arrangements should be pursued to complement the BCL’s emergency liquidity assistance facility.

Key Recommendations

Table 1.

Luxembourg: FSAP Update 2017: Key Recommendations

article image
Agencies: BCL = Banque centrale du Luxembourg; CRS = Comité du Risque Systémique; CSSF = Commission de Surveillance du Secteur Financier; ECB = European Central Bank; MoF = Ministry of Finance; MoJ = Ministry of Justice, SRB = Single Resolution Board. Time Frame: C = continuous; I (immediate) = within one year; NT (near term) = 1–3 years; MT (medium term) = 3–5 years.

Macrofinancial Setting

A. Background

1. The Luxembourg financial system has expanded significantly in recent decades to play a key role in the intermediation of global capital. Deep-rooted traditions of fiscal prudence, predictable regulation, a skilled multilingual labor force, and political and social stability have helped turn Luxembourg into a financial center of worldwide importance. The industry directly accounts for one quarter of GDP, 18 percent of tax revenues and 12 percent of employment.

2. Diversity and cross-border expertise in financial services provision has helped Luxembourg’s small open economy enjoy a robust recovery from the global financial crisis. With financial services exports and investment rebounding strongly, the level of real GDP surpassed the pre-crisis peak in 2011, and has continued rising strongly since (Table 2; Figure 1). Unemployment and inflation have remained relatively low and stable. The strong increase in residential real estate prices has outpaced income growth, reflected in an increase of household debt (to moderate levels; Figure 1). The fiscal position remains among the strongest in Europe, underpinning Luxembourg’s long-held AAA credit rating.

Figure 1.
Figure 1.

Luxembourg: Macroeconomic and Macrofinancial Context

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Table 2.

Luxembourg: Selected Economic Indicators, 2014–22

article image
Sources: Luxembourg authorities; IMF staff estimates and projections.

Contribution to GDP growth.

Overall economy.

Including a reclassification of investment companies from financial to non-financial institutions

3. Luxembourg’s financial landscape has been reshaped by reforms at both the European1 and domestic level in recent years, and the national authorities have adopted key recommendations from the 2011 FSAP (Annex I). The Finance Ministry is no longer responsible for granting or revoking banking licenses; the CSSF has shed its industry promotional role; its resources have been bolstered; and there has been a significant increase in the use of sanctioning powers. A new macroprudential policy regime, spearheaded by the Comité du Risque Systémique (CRS), has been introduced to safeguard macrofinancial stability. The BCL, which monitors system-wide bank liquidity (the ECB and the CSSF are the competent authorities for individual bank liquidity supervision), has strengthened its financial stability department in response to its de facto lead surveillance role in the CRS. The CAA has also increased resources to assist with the implementation of Solvency II.

B. Luxembourg Financial System Structure and Trends

4. The defining features of Luxembourg’s financial system remain its size and interconnectedness, though recent changes in industry composition have also been notable. The financial system is principally comprised of investment and money market fund industry assets under management (AuM of €3.6 trillion, 67 times GDP), bank assets (€764 billion, 14 times GDP), insurance industry assets (€219 billion, 4 times GDP), and financial market infrastructures (FMIs) that settle payments and securities transactions (Figure 2 and Table 3). Luxembourg also records a large residual category of ‘other financial intermediaries’ (OFI) in its flow of funds.2 The rising importance of Luxembourg’s nonbank financial industry over recent years—bank assets have declined by more than one-fifth from the onset of the global financial crisis while investment fund AuM have expanded strongly—reflects various factors, not limited to consolidation and business model adjustments in the regional banking landscape, and broad market trends (i.e. higher asset prices) conducive to Luxembourg’s investment industry.

Figure 2.
Figure 2.

Luxembourg: Key Components of the Luxembourg Financial System

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF Staff, BCL, and ECB.Notes: Insurance figures based on end-2015 data
Table 3.

Luxembourg: Structure of the Luxembourg Financial System

article image

As of December 2016 (Source: BCL)

As of December 2015 (Source: CAA, BCL)

As of November 2016 (Source: CSSF)

Investment funds refer to Undertakings for Collective Investment and Specialized Investment FundsSources: BCL, CSSF, and CAA.

5. The activities of Luxembourg banks are distinguished by their international or domestic orientation. The majority of banking sector activities in Luxembourg are internationally-oriented (Table 3), including depositary services for investment funds, private wealth management, and intragroup liquidity management services (Figure 3). Internationally-oriented banks are typically deposit-rich and thus net providers of liquidity to their parents. A small number of domestically-oriented banks (17 percent of industry-wide assets) focus principally on mortgage underwriting and retail and commercial banking, among which, the government has minority ownership stakes in three, in addition to the state-owned bank.

Table 4.

Luxembourg: Composition of Luxembourg Bank Assets and Liabilities

(in percent of total, December 2016)

article image
Source: BCL.Notes: Top 12 destinations, based on all counterparties, including investment funds.
Figure 3.
Figure 3.

Luxembourg: Bank Profit Sources

(in percent of pre-tax profit, as of June 2016)

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: ECB and SNL Financial.

6. Nonbank financial institutions are similarly outward facing, with investment funds and insurers experiencing robust AuM growth since the last FSAP on account of valuation effects and new inflows. Investment fund AuM have doubled since the spring of 2010 (Figure 4), and are diversified by asset class and geography (only 7 percent of exposures are domestic; Figure 5). Investment fund units (liabilities) are sold globally, a result of Luxembourg’s well-developed distribution platform and established branding under the Undertakings for Collective Investment in Transferable Securities (UCITS) regime. Luxembourg-domiciled funds typically delegate key functions such as portfolio and risk management abroad, while retaining ultimate oversight responsibilities. In the insurance industry, life insurers are predominant, accounting for 85 percent of assets and two-thirds of premiums (Figure 6; Table 3). Unit-linked investment products have become an increasingly prominent business for life insurers (71 percent of technical provisions), reflecting Luxembourg’s related expertise.

Figure 4.
Figure 4.

Luxembourg: Investment Fund Exposures by Country

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 5.
Figure 5.

Luxembourg: Investment Fund Assets Domiciled in Luxembourg

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 6.
Figure 6.

Luxembourg: Insurance Industry: Premiums and Assets

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

7. Luxembourg is also home to a globally systemic FMI, Clearstream Banking S.A., Luxembourg (CBL). As one of the world’s largest international central securities depositories (ICSDs), CBL facilitates the settlement and custody of securities globally. It accounted for a daily average settlement value of €480 billion in 2015, with the value of securities held in accounts with CSDs around €6 trillion. CBL’s international orientation is also evident in its servicing of more than 50 foreign markets and 60 central banks, and through its links to three large central counterparties (CCPs) in the European Union (EU; Figure 7).

Figure 7.
Figure 7.

Luxembourg: Clearstream Luxembourg and the FMI Ecosystem

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

8. Luxembourg’s financial soundness metrics have remained relatively robust in recent years. Bank profitability, capital, liquidity and asset quality has been high (Table 5; Figure 8). Net interest margin compression has been relatively less problematic for Luxembourg banks due in part to the fee and commission income derived from private banking and fund management activities (Figure 3). Across money market funds and the expanding investment fund industry, liquid asset holdings have remained broadly steady (Figure 9). The Luxembourg insurance industry has stayed profitable and well capitalized (Figure 10), reflecting a relatively modest asset-liability duration mismatch and the relatively modest exposure of life insurers to guaranteed rate products.

Table 5.

Luxembourg: Bank Financial Soundness Indicators (All Banks)

(all figures in percent)

article image
Source: BCL.
Figure 8.
Figure 8.

Luxembourg: Bank Financial Soundness Indicators

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: BCL, CSSF, and IMF.Notes: Luxembourg bank time series data in left panels are updated as of Q2 2016. Cross country comparisons in right panels are based on latest available data between Q4 2015 and Q2 2016. In the international comparisons, data displayed for Luxembourg banks include all banks (domestic and internationally oriented).
Figure 9.
Figure 9.

Luxembourg: Investment and Money Market Funds: Liquid Asset Holdings

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: CAA.
Figure 10.
Figure 10.

Luxembourg: The Insurance Industry: Capital and Profitability

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: BCL and IMF staff.

C. Vulnerabilities

9. Nevertheless, the unique structure—i.e. sheer size and interconnectedness—of Luxembourg’s financial system gives rise to potential vulnerabilities (Figure 11):

  • Banks—Deposits accumulated from private wealth management and investment fund activities are frequently ‘upstreamed’ to foreign parents where centralized treasury management functions are performed. Luxembourg subsidiaries may thereby be exposed to maturity and currency transformation risk at the parent level, where oversight by Luxembourg authorities may be limited (this risk materialized during the global financial crisis, since which time mitigants have been introduced).3 Some 35 percent of Luxembourg bank assets comprise intragroup claims (77 percent of which now fall under ECB supervision; Figure 12), and Luxembourg bank exposures to intragroup entities (principally EU parents) often amount to several times their capital.4 These characteristics, along with large holdings of zero-weighted sovereign bonds, help explain the relatively low risk density (ratio of risk-weighted to total assets) in Luxembourg (31 vis-à-vis 39 percent euro area average).

  • Investment Funds—should investment funds have inadequate liquid asset holdings and liquidity management tools (LMTs) prove ineffective, an unexpectedly large redemption shock could result in a drawdown of bank deposits (Figure 13, Table 6) and fire sales in markets where such funds account for a large share of tradeable securities.5

  • FMIs—a shock to the functioning of CBL could disrupt the vast number of international institutions and markets it serves. This could leave participants unable to access or trade some or all of the securities held in CBL, resulting in credit and liquidity pressures with potentially significant repercussions for the international financial system.

Figure 11.
Figure 11.

Interconnectedness: The Luxembourg and Global Financial Systems

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF.Notes: White (yellow) text denotes domestically (foreign) domiciled entities. Private wealth management activities are subsumed in the internationally oriented Luxembourg banks. The central position of Clearstream simply aids in the visual interpretation of system flows (i.e. it is not meant to imply an outsized role for Clearstream relative to other components of the financial system).
Figure 12.
Figure 12.

Luxembourg: Bank Intragroup Exposures

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: BCL.
Figure 13.
Figure 13.

Luxembourg: Bank Liabilities Accounted for by Investment and Money Market Funds

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: CSSF and BCL.
Table 6.

Luxembourg: Investment Fund Exposures to Banks

article image
Source: BCL statistical reporting form S.521 and investment fund reporting.Note: Exposures to Luxembourg banks are taken from S.251. Exposures to banks abroad are the residual of investment fund total global exposures to banks. Asset exposures include deposits, loans, equity instruments, debt instruments, and derivatives. Liabilities include loans, short sales, and derivatives.

10. The main domestically-originated vulnerability relates to residential real estate developments, to which a small number of domestically-oriented banks (and the government through its bank ownership stakes) are exposed. Following a marginal decline in 2009, real home prices have since risen 22 percent while real disposable income has been flat, thus impairing affordability with potential implications for debt servicing capacity (Figure 14; Box 1). Mortgage lending standards have eased of late, evidenced by an uptick in the share of high loan-to-value (LTV) mortgages, and household indebtedness has increased (Figure 1). Floating rate mortgages also raise the issue of borrower interest rate risk.6 Balanced against these vulnerabilities, however, are several factors—a constrained supply response, a modest share of mortgages in domestically-oriented bank assets (16 percent), the low (52 percent) industry-wide loan-to-value (LTV) ratio, below average household indebtedness,7 and high household net worth (financial assets exceed household debt by 2.4 times).

Figure 14.
Figure 14.

Luxembourg; Residential Real Estate Market Affordability Measures

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF staff.

Is the Residential Real Estate Market in Line with Fundamentals?

Luxembourg’s steadily rising real estate prices have raised questions as to whether such gains have been consistent with underlying fundamentals. Results from the estimation of a number of models (following Igan and Loungani, 2012) of annual real house price growth over the period 1981Q1–2015Q4 do not, however, point to a significant misalignment of prices with fundamentals, although housing affordability has deteriorated (Figure 14). While real house prices appeared overvalued prior to the global financial crisis, their evolution since has been more consistent with fundamentals. In spite of flat disposable income over recent years, the low interest rate environment has strengthened borrowing capacity, supply responses have been muted, and robust population growth has spurred demand. Nevertheless, model results suggest an unexpected shock to these fundamental drivers—such as population and income growth reversing due to a downturn in the financial services industry—could impart significant downward pressure on real estate valuations.

uA01fig01

Real Home Prices vs. Model-Implied Equilibrium Values

(percent deviation of actual prices from those implied by econometric models)

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF Staff
article image
*** p<0.01, ** p<0.05, * p<0.1Note: model estimated with a newey west estimator with 4 lags.

Risks, Resilience and Spillovers

11. The principal risks faced by Luxembourg’s small open economy and internationally-oriented financial system relate to interruptions to real and financial cross-border flows, and second round effects for real estate and domestically-oriented banks. As outlined in the Risk Assessment Matrix (RAM, Annex II), an acute external shock whereby financial market risk premia rise sharply and the Euro area relapses back into recession constitutes the primary risk.8 Earlier described vulnerabilities could see a seizure in cross-border flows lead to strains in Luxembourg’s internationally-oriented banks (liquidity and solvency stress), investment funds (liquidity stress) and insurers (solvency stress). Second round effects would be felt domestically through a downturn in the residential real estate market, adversely impacting household net worth and consumption, and profitability and possibly capital at domestically-oriented banks. It is in this context that stress test and spillover analysis assesses the resilience of Luxembourg’s financial system (Figure 15). 9

Figure 15.
Figure 15.

Luxembourg: Summary of Luxembourg FSAP Stress Test Sectoral Coverage

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF.

A. Stability Risks and Resilience

Bank Solvency and Liquidity Stress Tests

12. Capital resilience in the banking system was assessed through “top-down” macro-financial stress tests. Coverage included 16 banks (15 significant institutions (SIs) directly supervised by the ECB and one less significant institution (LSI)), accounting for 73 percent of industry assets.

13. Stress tests were based on fully-fledged macroeconomic scenarios comprising a baseline and one severe but plausible adverse scenario spanning a three-year period (to 2019) where:

  • The baseline scenario is based on the October 2016 World Economic Outlook projections;

  • The adverse scenario features a V-shaped GDP profile with a cumulative decline larger than that experienced following the global financial crisis (Figure 16 and 17). This outcome derives from a combination of external and domestic shocks (a surge in global financial market volatility, a renewed euro area recession, and spillover to the housing market; see RAM) transmitted through the IMF Global Macro-Financial Model. The result is a 1.5 percent per annum decline in GDP averaged over 2017–2019, leading to a cumulative deviation of 14.7 percentage points with respect to the baseline (equivalent to a 2 standard deviation shock).

Figure 16.
Figure 16.

Luxembourg: Real GDP in the Adverse Scenario

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 17.
Figure 17.

Luxembourg: FSAP Stress Test Macro Projections

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: WEO, national sources, and IMF staff estimates.

14. Stress test results suggest system-wide bank capital would display a good deal of resilience in the face of severe but plausible shocks, with some select risks. High starting levels of capital allow the banking system to absorb a large adverse shock while retaining substantial buffers (Figure 18). Real estate vulnerabilities are mitigated by high levels of collateralization and the modest share of mortgages in bank assets. Based on fully-loaded Basel III regulatory requirements, the ratio of Common Equity Tier 1 capital to total risk-weighted assets (CET1 ratio) would decline over three years from 18.6 percent to 15.1 percent. One bank would experience a decline in capital below the CET1 hurdle rate, with three banks (comprising less than 10 percent of industry assets) experiencing a decline in capital below the capital adequacy ratio hurdle,10 by 0.1 and 0.8 percent of GDP respectively. The system-wide leverage ratio (Tier 1 capital to total assets) would decline from 5.9 to 5 percent, with six banks experiencing a decline below the 3 percent hurdle rate in 2019 (after which time it becomes binding) by an amount equivalent to 1.6 percent of GDP.11

Figure 18.
Figure 18.

Luxembourg: Results of Top-Down Bank Solvency Stress Tests

All Banks: Contributions to Net Profit and Change in Capitalization Ratio

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: IMF staff calculations.

15. Owing to methodological differences, these results differ from those based on banks’ own estimates in an exercise recently coordinated by the EBA, ECB, and national authorities. A lack of historical credit loss data meant credit risk benchmarks were conservatively based on the recent Ireland financial crisis. Additionally, risk mitigants including economic hedges and financial guarantees were not incorporated as their granular characteristics could not be assessed and they could not be harmoniously applied. Growth assumptions for RWA under the adverse scenario also exceeded those estimated by the authorities and banks in recent stress tests.

16. With respect to concentration risk, sensitivity tests reveal that some Luxembourg banks could be vulnerable to the simultaneous default of their largest nonfinancial corporate exposures.12 The simultaneous default of the five largest exposures would, after application of exemptions and credit risk mitigation,13 lead three banks to be undercapitalized by 1.6 percent of GDP with regard to the Tier 1 capital ratio hurdle of 6 percent

17. Stress tests confirm the general strength in bank liquidity, though intragroup treasury activities could contribute to liquidity mismatches in a highly stressed scenario:

  • Liquidity coverage ratio (LCR)—Under the standard scenario, all banks comfortably meet the 80 percent LCR requirement currently effective (Figure 19, Table 7), and all but 4 of the 16 banks in the sample meet the 100 percent LCR requirement effective January 2018 (first column, Table 7). Separate retail and wholesale deposit withdrawal scenarios, which layer stress onto the (already stressed) LCR, show that 4 and 7 banks respectively would struggle to maintain an LCR above the 100 percent requirement Shortfalls in the LCR in USD terms, which is non-binding but reported by banks, reflect subsidiaries’ reliance on their parent to manage FX assets on their behalf.

  • Net stable funding ratio (NSFR)14—6 out of 8 banks pass the 100 percent NSFR threshold ahead of its January 2018 implementation, with modest shortfalls in the remaining two.

  • Cashflow analysis—6 out of 8 banks would face funding gaps over a seven-day shock, though all but one of the affected banks hold sufficient central bank eligible securities to cover such a shortfall (final column, Table 7).

Figure 19.
Figure 19.

Luxembourg: Bank Liquidity Coverage

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: Luxembourg authorities and IMF staff calculations.
Table 7.

Luxembourg: Bank Liquidity Stress Test Results

article image
Sources: Luxembourg authorities and IMF staff calculationsNote:

Liquidity shortfall is the amount required so that the Liq. Ratio in each bank in the system be equal to or above 100 percent; the ratio effective as of January 2018.

CC=Counterbalancing capacity

Nonbank Stress Tests

Household Solvency Stress Tests

18. Households’ solvency would be adversely impacted by a drop in income and house prices and a rise in the unemployment rate, but the share of households in default would remain modest. Utilizing micro-level household data, a stress test assessed the sensitivity of household balance sheets to the earlier described adverse macro-financial shock. The exercise points to an increase in household debt default probabilities from 2.5 to 6.4 percent (a multiplier of 2.6). The low share of households in default reflects numerous factors: many households either do not hold or have repaid part of their mortgage; high household net worth; and the large holdings of liquid assets which could be used in times of stress for debt service payments.

Insurance Solvency Stress Tests

19. The Luxembourg insurance sector displays resilience to market shocks (consistent with the findings of the 2011 FSAP), despite a notable decline in Eligible Own Funds (EOF). This is the result of a bottom-up solvency stress test (coordinated by the CAA) covering 10 life insurers and one reinsurer, both accounting for roughly three quarters of the respective markets. While aggregate EOF declines by 27 percent in the third year of the adverse scenario, each insurer in the sample would maintain its capital well above the 100 percent Solvency II requirement (Figure 20). Cross-sectionally, insurers with a relatively larger share of unit-linked business are more severely hit than peers, reflecting their exposure to reduced AuM-related fee income.

Figure 20.
Figure 20.

Luxembourg: Insurance Capital and Eligible Own Funds in the Adverse Scenario

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Source: CAA.
Money Market Fund ‘Solvency’ Stress Tests

20. Stress tests and sensitivity analysis reveal that a very large increase in risk-free interest rates and credit spreads would be needed to ‘break the buck’ (a measure of solvency risk) in Luxembourg’s Constant Net Asset Valuation (CNAV) money market fund (MMF) industry. For instance, based on a sample of 5 CNAV funds (accounting for 75 percent of CNAV MMF AuM), the combination of a 130 basis point shock to risk-free interest rates and a 150 basis point widening in credit spreads would be required for the shadow NAV (fully reflecting market fluctuations) to deviate substantially (by 0.2 percent or more) from par value (Figure 21, left panel). Market risks are mitigated by large holdings of cash and short duration, high quality assets (Figure 21, right panel).

Figure 21.
Figure 21.

Luxembourg; CNAV Money Market Fund Portfolios

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Notes: This analysis follows the methodology in IMF (2016b). Data and results are aggregated as an asset-weighted average of the 5 largest CNAV MMFs. All individual funds produced qualitatively similar results as the asset-weighted average. The shadow NAV refers to the NAV if market fluctuations were reflected directly in the NAV.
Investment Fund and MMF Liquidity Stress Tests

21. Funds exposed to asset classes with lower liquidity/higher credit risk are more vulnerable to severe but plausible redemption shocks (Table 8). Redemption shocks based on past experience (historical approach) and on macro-financial outcomes from the aforementioned adverse scenario (forward-looking simulation) are compared to fund liquidity buffers for a variety of funds covering significant portions of AuM. 15, 16 Around 70 percent of HY funds would face a shortfall in the historical and forward-looking scenarios. EM, mixed and other bond funds encounter problems only under the (conservative) assumption that liquidity comprises cash and short-term debt. Liquidity risk in MMFs appears muted.

Table 8.

Luxembourg: Money Market and Investment Fund Liquidity Stress Tests

article image
Source: BCL, CSSF, IMF Staff calculationsNotes: Italicized figures in white depict the average redemption shock for each fund (as a share of total net assets, TNA).Figures in gray panels depict the percentage of funds in the sample unable to cover a redemption shock^ The HQLA measure is not applicable for MMFs* Estimated model is not significant** Under the adverse forward-looking scenario, mixed funds experience net inflows

22. In the absence of adequate liquid security holdings and effective liquidity management tools (LMTs), select bond funds may need to draw on bank deposits. In the forward-looking adverse scenario, in which all funds are simultaneously subjected to a severe redemption shock, outflows amount to 20–24 percent of investment fund bank deposits in the sample (equivalent to €4.9–5.3 billion).17 Across the sample of depositary banks, these outflows would amount to 3 percent of total deposits, or 4 to 10 percent of wholesale or operational deposits respectively.18

23. Caveats and context are important in the interpretation of these results. The tests assume managers cannot avail of LMTs, and embed restrictive assumptions on the ability of funds to liquidate long-term securities. This is most notable for HY funds, where manager surveys recently introduced by the CSSF point to a more benign assessment of liquidity risk. (Note also that HY fund AuM account for only 5 percent of Luxembourg’s investment industry). Data limitations also require the assumption that the entire liquidity shortfall is borne by a single depositary bank in Luxembourg, while in practice fund deposits are typically spread across several banks in and outside Luxembourg.

B. Spillover and Contagion Analysis

24. Network analysis and other modelling of supervisory and market data confirm that while there is limited interconnectedness across banks and insurers domestically, it is high across borders.

  • At the domestic level, the interconnectedness within and across banks and insurers seems limited.19 Aggregate indices of contagion and vulnerability (based on balance sheet data) record benign readings (Figure 22). For instance, failure of the entity with the highest contagion index causes losses of less than 0.5 percent of counterparties’ capital. Market data-based analysis suggests Luxembourg banks are not strongly interconnected (excepting Bank 2 and 4; Figure 23).

  • By contrast, at the cross-border level, quantitative analysis confirms a high degree of interconnectedness for Luxembourg-domiciled banks and investment funds (Figure 24):20

    • Banks. Based on gross original exposures (before credit risk mitigation) and under standard model assumptions,21 an unexpected increase in the defaulted share of cross-border claims through the failure of a parent poses a threat to the capital of Luxembourg subsidiaries, reflecting large intra-group exposures (Figure 25, diagonal cells). Vulnerability and contagion risk emanating from foreign parents emanates more prominently through credit rather than funding channels (blue columns, Figure 26).

    • Investment funds. Luxembourg-domiciled investment funds appear interconnected with select global banks and EM bond markets. Mixed and bond funds are connected with U.S. and UK banks, evidenced by their central location in the global network (red/orange nodes, Figure 27). This likely reflects exposure to common factors. Additionally, some fixed income investment funds have large and concentrated holdings in select frontier/EM bond markets (beyond those dictated by capitalization-weighted benchmarks). Exposure to a single sovereign issuer is around 30 percent of assets for a number of funds (Figure 28),22 and collectively, Luxembourg bond funds hold more than 10 percent of outstanding tradable debt in six frontier/EM bond markets (Figure 29).23 Luxembourg funds could therefore act as a conduit of shocks for these markets.

Figure 22.
Figure 22.

Luxembourg: Indices of Domestic Contagion and Vulnerability—Balance Sheet Analysis

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 23.
Figure 23.

Luxembourg: Domestic Network Model—Market Data Analysis1/2/

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

1/ Results based on market data employed in the Diebold and Yilmaz (2014) model.2/ Edge thickness and darkness shows the strength of the pairwise relationship, depicted as the 12-month forecast variance of entity i due to shocks from entity j. The underlying error-variance decomposition matrix was calculated using probability of distress data (PoD). PoDs for Luxembourg banks were derived using bond spreads, and for investment funds, derived from marked-to-market return data. Node size indicates total asset size, color of nodes indicates “total connectedness to others” with dark red indicating connections with other entities in the sample. Node location is derived using ForceAtlas2 algorithm in which nodes repel each other, but strength of edges (i.e. connections) is attracting the nodes to each other.Sources: Datastream and Bloomberg for bank data; Luxemburg authorities for investment fund data; and IMF staff calculations.
Figure 24.
Figure 24.

Luxembourg: Domestic vs. Cross-border Network of Financial Institutions

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 25.
Figure 25.

Luxembourg: Network Analysis of Gross Intragroup and Cross-border Bank Exposures—Balance Sheet Analysis

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Note: Colors indicate the loss-given-default rate (λ) at which the failure of a foreign parent or other foreign bank (P1, P2, etc.) would result in the insolvency of a Luxembourg bank subsidiary (L1, L2, etc.). Cells highlighted along the diagonal (off-diagonal) refer to Luxembourg bank subsidiaries and their foreign parents (other banks). For instance, the L1 – P1 cell depicted in red refers to the situation where a Luxembourg subsidiary would have its capital position impaired by the failure of its parent assuming a loss-given-default rate of 5 percent.
Figure 26.
Figure 26.

Luxembourg: Indices of Contagion and Vulnerability—Balance Sheet Analysis

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 27.
Figure 27.

Luxembourg: Global Network Model—Market Data Analysis1/2/

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

1/ Results based on market data employed in the Diebold and Yilmaz (2014) model.2/ Edge thickness and darkness shows the strength of the pairwise relationship, depicted as the 12-month forecast variance of entity i due to shocks from entity j. The underlying error-variance decomposition matrix was calculated using probability of distress data (PoD). PoDs for banks were derived using credit spreads, and for investment funds, derived from marked-to-market return data. Node size indicates total asset size, color of nodes indicates “total connectedness to others” with dark red indicating stronger connections with other entities. Node location is derived using ForceAtlas2 algorithm in which nodes repel each other, but strength of edges (i.e. connections) is attracting the nodes to each other.Sources: Datastream and Bloomberg for bank data; Luxemburg authorities for investment fund data; and IMF staff calculations.
Figure 28.
Figure 28.

Luxembourg: Concentration Risk and Common Exposures in Luxembourg Bond Funds

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Figure 29.
Figure 29.

Luxembourg: Investment Fund Ownership Share of Sovereign Debt Markets

Citation: IMF Staff Country Reports 2017, 122; 10.5089/9781484300374.002.A001

Sources: BLC, Bloomberg LP, and IMF staff.

Prudential Oversight

A. General and Cross Cutting Themes

25. Though Luxembourg’s system of prudential oversight appears to function well, several cross-cutting themes warrant attention.

  • Risk-based supervision and resources. A continued pivot towards risk-based supervision and a further related increase in resources for the CSSF, CAA, and BCL (where planned capital increases should also proceed), would allow these agencies to more easily meet the demands imposed by new regulation and an enlarged financial system.

  • Governance. A formal framework should be agreed to govern the relationship between the government and banks with state involvement to ensure it is kept at arm’s length and that the relevant banks are free to operate on commercial terms. Additionally, while the mission encountered no evidence of political or industry interference, the operational independence and accountability of the CSSF and CAA should be enshrined in law (as recommended by international standards) in order to safeguard financial stability well into the future. CSSF and CAA board members should also be bound by codes of conduct, in line with best practice (the concurrent updating of the BCL’s code of conduct would be welcome in this regard).

  • International engagement. Luxembourg authorities should increase the intensity of engagement with supervisors in countries where Luxembourg’s financial institutions, particularly investment funds, are most active.24

  • Dampening residual risks. More consideration should be given to the imposition of supervisory measures that go beyond the minimum prescribed by European legislation, where justified by the idiosyncrasies of Luxembourg’s banking and investment industry, and where compatible with the EU Single Rule Book.25 The authorities should also take the initiative to reinforce the oversight of nonbank holding companies of banks to improve risk monitoring, while continuing to advocate for a coordinated approach at the European level.26

B. Macroprudential Policy Framework

26. The macroprudential policy framework introduced in 2015 is working well in practice, though some elements could be strengthened.27 The institutional framework could be enhanced by revoking the unanimity voting requirement in the CRS to eliminate the potential inaction bias arising from each agency’s veto power.28 The independence of each member would also strengthen the functioning of the CRS. 29 Additional measures could further strengthen the new policy framework, including by: enshrining in law the BCL’s de facto lead role in financial stability analysis; publishing the BCL’s risk reports and dashboard to increase transparency and accountability; and awarding the BCL powers to make formal policy recommendations to the CRS.

27. A macroprudential toolkit for the real estate market is in place and the authorities have demonstrated a willingness to deploy it, but the operational capacity of the CRS can be strengthened. Measures enacted have included, since 2012, higher risk weights where loan-to-value ratios exceed 80 percent; stricter stress test requirements for IRB bank mortgage books; pillar II capital add-ons; and a higher risk weight floor for IRB bank exposures to domestic real estate. In December 2016, the authorities also transposed the EU Mortgage Credit Directive into law, thereby strengthening consumer information requirements and obligations to assess creditworthiness. Further macroprudential tightening may be required should real estate price growth continue to outpace that of incomes, and lending standards continue to ease. In this regard, operational capacity would be enhanced by providing a legal basis for borrower-based macroprudential tools, notably, limits on loan-to-value and debt-service-to-income.

28. The monitoring of systemic risks by the BCL and CSSF is appropriately focused on real estate vulnerabilities and bank-investment fund linkages, but closing related remaining data gaps is a priority. Real estate-related data gaps span both the residential market (including the rental market and on the loan-to-value distribution across borrowers and banks) and the commercial market (for which there is no data besides a survey undertaken by a commercial firm every five years). Bank-investment fund surveillance has made good progress,30 but is still hampered by data gaps regarding individual investment fund exposures to depository banks, with data typically gathered at the asset management company level. The authorities are also encouraged to continue recent efforts to deepen their analysis of investment fund liquidity risk, the use of synthetic leverage and securities financing transactions, and the concentration and categorization of beneficial investors.

C. Banking Regulation and Supervision

29. Banking regulation and supervision in Luxembourg has been strengthened in recent years. The authorities have transposed into national law the EU Capital Requirements Directive (CRD IV). Following the introduction of the SSM in 2014, 61 of the 140 banks in Luxembourg are now supervised by the ECB, either as SIs or as subsidiaries or branches of foreign SIs, representing 77 percent of Luxembourg bank assets. The ECB has also replaced Luxembourg’s Finance Minister as the body responsible for bank licensing. Against this background, a risk-based examination of the regulatory and supervisory approach to identifying and mitigating vulnerabilities was undertaken, using as a reference the 2012 Basel Core Principles (BCP) for Effective Banking Supervision.

30. Notwithstanding recent advances, a number of additional measures could help to safeguard banking system stability.31 As the CSSF’s board is answerable to the Minister and includes industry representatives,32 the operational independence and accountability of the CSSF should be enshrined in law, as recommended by the BCP. More frequent on-site inspections, particularly for foreign subsidiaries availing of the waiver to large exposure limits for intragroup transactions, should be put in place, and banks should be required to periodically prove their continued eligibility for the waiver.33 Continued monitoring of mortgage lending standards is warranted, and data reporting standards for loan-to-value and debt-service-to-income ratios should be harmonized across banks. The BCL should also work closely with the ECB to ensure households are included in the euro area credit registry initiative, particularly given the recent uptick in household indebtedness.

31. Liquidity stress test results suggest risks relating to short term funding and foreign currency arrangements with parent groups should be addressed.34 Banks should lengthen the maturity of unsecured funding beyond seven-days. Additionally, implementation of an FX LCR framework at the group level would help manage the risk of FX-related liquidity mismatches for Luxembourg subsidiaries. Furthermore, the authorities should work with the EBA to close liquidity reporting gaps and expand harmonized EU bank reporting.

D. Investment Fund Regulation and Supervision

32. The CSSF applies a strong and comprehensive regulatory and supervisory framework to the investment fund industry. Following recommendations in the 2011 FSAP, the CSSF has bolstered its resources devoted to fund management oversight and gives due priority to its monitoring and oversight of the sizable UCITS sector.

33. Nevertheless, delegation practices and the concentration of fund directorships merit increased attention. Guidance on the ‘substantial presence’ threshold would be welcome given the extent to which Luxembourg funds avail of delegated activities such as portfolio and risk management, and the CSSF should take steps to engage more actively with regulators in the jurisdiction where delegated activities are frequently performed. Additionally, the CSSF should review data on fund directorships held by individuals and issue guidance on limits accordingly, framed in terms of aggregate time commitments.

34. The authorities should issue guidance on the use of LMTs and the modalities of liquidity stress tests—focused initially on fixed income funds—and develop liquidity stress testing capacity. The CSSF should leverage its extensive experience with LMTs in conducting an assessment of their effectiveness to inform industry guidance. To strengthen surveillance and management of system-wide liquidity risk (beyond that provided by existing tools including the new UCITS risk reports), the CSSF should provide industry guidance on the modalities of liquidity stress tests (e.g., frequency, fund coverage, scenarios), and develop system-wide stress testing capacity, consistent with recent Financial Stability Board (FSB) and European Systemic Risk Board guidance.

35. The CSSF should assess the impact of changes on depositary independence introduced under UCITS V, and identify whether risks remain. Appropriate safeguards should be introduced to address any residual concerns. Luxembourg benefits from a large and diverse depositary sector, which gives greater scope for intra-group depositary arrangements to be adjusted if necessary.

36. While the CSSF’s supervisory approach is sound, enhancements to its inspection program and better data access would further improve the effectiveness of risk-based supervision. Efforts should be made to further increase the number of thematic on-site inspections, and to introduce comprehensive on-site inspections as another element of the supervisory toolkit.

37. Analysis of nonbank financing and special purpose vehicles should continue. The BCL is encouraged to continue its analysis of the OFI sector and examine whether some entities should be brought under the regulatory perimeter. The authorities are also encouraged to fully participate in international for a on nonbank finance, including initiatives led by the FSB.

E. Insurance Regulation and Supervision

38. The CAA has worked diligently to ensure the insurance industry has adjusted smoothly to the introduction of Solvency II. The FSAP assessed the CAA’s regulatory and supervisory compliance with selected Insurance Core Principles (ICPs) drawn up by the International Association of the Insurance Supervisors (IAIS), notably those concerned with cooperation with other supervisory jurisdictions, risk management, internal controls, and supervisory review reporting.

39. Luxembourg’s supervisory framework demonstrates an adequate level of consistency with IAIS standards, though the CAA’s resources and governance arrangements could be enhanced. The CAA is an active participant in supervisory colleges owing to the international orientation of Luxembourg insurers. However even after a recent uptick in resources, the CAA has a modest staff of about 40 to oversee nearly 300 insurance companies with assets approaching €220 billion. A revised early warning system, calibrated to new Solvency II parameters, should be prioritized (as planned) once the requisite data become available. Additionally, the CAA’s governance structure would benefit from: (i) subjecting board members to a formal code of conduct; (ii) having board activities periodically reviewed by an independent committee; and (iii) limiting the government’s power to dismiss the Executive Committee, thus safeguarding the CAA’s operational independence.

F. Financial Markets Infrastructure—Clearstream Banking Luxembourg

40. An assessment of CBL’s risk management practices and supervisory treatment against the CPSS-IOSCO Principles for Financial Market Infrastructures (PFMI) reveals it to be in broad observance. CBL offers a relatively safe and efficient system for the settlement and custody of securities transactions. Luxembourg’s implementation of European directives provides a solid statutory basis for netting, finality of settlement, and securities lending. CBL uses an effective risk management framework to manage operational and liquidity risks, has developed a comprehensive business continuity plan, and has in place a detailed counterparty default management framework. CBL’s oversight and supervision is conducted prudently by the BCL and CSSF.

41. Nevertheless, further emphasis on risk mitigation is recommended. Priorities include:

  • Reducing the large deposit-based exposure of CBL to a small number of commercial banks—this could be mitigated through an increase in the number of contracted banks, or the establishment of direct links with local CSDs and central banks, where possible.

  • Tightening collateralizing arrangements—though CBL has a process for managing credit risks (exposure limits, partial collateralization), residual risks could be mitigated with full collateralization and applying (independently validated) haircuts to cash collateral.

  • Strengthening operational risk management—while contingency plans and back-up facilities are in place to assist operational recovery, the close proximity of such facilities, coupled with CBL’s systemic importance, warrant investment in a third (more distant) data center. A full failover test is also advised, to assess the efficacy with which remote locations can assume critical operations.

  • Enhancing supervisory cooperation arrangements—although interagency cooperation functions well domestically, formalizing modalities between the BCL and CSSF would support efficiency and accountability. The authorities should also engage with international colleagues when conducting their assessment of CBL against the PFMI. Cooperation between the Belgian and Luxembourg authorities with respect to Euroclear Bank should also be strengthened as planned.

42. Related, and notwithstanding effective national level supervision, the sheer scale and scope of its activities suggest CBL should be designated as a SI under SSM supervision, alongside Belgium’s Euroclear Bank.35 Although harmonization of national supervisory approaches is expected to increase with the implementation of the EU CSD Regulation, this does not address CBL as a bank. SSM-level supervision would ensure a level playing field and better facilitate monitoring of cross-border group level spillover risks, a prudent objective given the systemic importance of both ICSDs.

G. Anti-Money Laundering and Combating Financing of Terrorism

43. Luxembourg has strengthened its anti-money laundering and combating the financing of terrorism (AML/CFT) regime since the 2010 mutual evaluation. Notably, as of January 1, 2017, aggravated tax evasion and tax fraud constitute predicate offenses to money laundering (ML). Earlier progress includes strengthening the ML offense and the licensing and supervision of financial institutions, increasing mutual legal assistance in CFT efforts, and dematerializing bearer shares. Luxembourg is currently conducting its first national risk assessment (NRA) and seeking to increase transparency of beneficial ownership information. AML/CFT supervision of the financial and nonfinancial sectors has been strengthened, though remains relatively weak with respect to lawyers.

44. The government’s push for tax transparency constitutes material progress. Luxembourg is amongst the first countries to implement the OECDs Common Reporting Standard, which should greatly facilitate the exchange of information between tax administrations. The number of suspicious transaction reports and the exchange of financial information with foreign counterparts have increased over recent years.

45. Challenges remain, and the authorities have indicated their firm commitment to address them. Key priorities include: (i) ensuring that the ongoing NRA adequately focuses on risks related to trust and company services providers across all relevant professions; (ii) provide guidance on the identification of suspicions of tax crimes and related ML to reporting entities not supervised by the CSSF; (iii) further intensifying AML/CFT monitoring of lawyers; and (iv) continuing to ensure resources for risk-based AML/CFT supervision of the financial sector remain sufficient.

Crisis Resolution and Safety Nets

46. The landscape for crisis management in Luxembourg has changed significantly since the 2011 FSAP. As a participant in the Banking Union, the CSSF shares intervention and recovery planning competencies with the ECB, and resolution competences with the Single Resolution Board (SRB). By transposing the Bank Recovery and Resolution Directive (BRRD) and Deposit Guarantee Scheme Directive (DGSD) in late 2015, the authorities introduced a new resolution framework and public deposit insurance scheme. The Resolution Board and the Depositor Protection Council were established within the CSSF to carry out resolution work and administer the deposit insurance scheme respectively. Nevertheless, a further increase in staffing would better equip the CSSF’s new Resolution Department.

47. Recovery planning is advanced while resolution planning is at an early stage. Both would benefit from additional guidance from the ECB and SRB along various lines. Guidance on the identification of critical functions (e.g., custodian functions) and the treatment of large intragroup exposures would enhance recovery plans. Guidance on the use of the sale of business and bridge bank tools would help strengthen resolution plans, including in developing fall-back resolution strategies where bail-in is the preferred strategy. Under SRB policy, resolution planners cannot contemplate the use of the Single Resolution Fund (SRF) in their plans—and thus plans do not address how the SRF access requirements would be met; the policy should be reconsidered. The goal should be to establish credible and feasible resolution plans for the most important Luxembourg banks as soon as possible.

48. Particular characteristics of the Luxembourg financial system present challenges that resolution planning authorities should address. Ambiguities include whether excluding investment fund bank deposits from the scope of bail-in would give rise to “No Creditor Worse Off” claims should other creditors in the same creditor hierarchy class be bailed-in. Ensuring the continuity of custodial functions through the sale of business tool may necessitate advance preparations in order to overcome the operational challenges in selling such business lines. Bail-in of the large intragroup (typically foreign parent) claims held by Luxembourg subsidiaries may adversely affect their viability. These issues should be factored into resolution plans.

49. There is scope for improvement in cross-border cooperation. The recent conversion of several Luxembourg subsidiaries into branches raises questions as to how member states hosting branches that are not “significant branches” but that provide “critical functions” can be engaged in reviewing group-wide recovery and resolution plans. With respect to third country relations, there is a need for the SRB to continue, and the CSSF to seek, to establish cooperation arrangements with relevant resolution authorities.

50. The insolvency regime applying to banks should be strengthened to complement the newly established bank resolution regime. The introduction of explicit asset and liability transfer powers during winding-up proceedings may help achieve an orderly liquidation with reduced costs to the Fonds de Garantie des Dépôts Luxembourg. Additionally, the creditor hierarchy in Luxembourg is determined by different laws, making it difficult to interpret. The authorities are assessing these rules and could leverage the recent Commission proposal to harmonize aspects of creditor hierarchy across the EU.

51. New deposit insurance arrangements are in place but should be supported with back¬up funding. They target a fund balance twice that required in EU legislation, and the timetable for rapid payouts is advanced. However, no backstop funding arrangements are in place and should be pursued swiftly, with the private sector where possible. As a last resort, public funding in compliance with State Aid rules may be necessary.

52. While a contingent framework is in place for the provision of ELA, efforts to operationalize it quickly when needed should continue. BCL has access to foreign exchange from market sources and the Bank for International Settlements, but has no bilateral foreign exchange swap arrangements with relevant central banks. It should continue examining additional contingent foreign exchange funding arrangements, possibly by collaborating with the ECB. Additionally, the terms and conditions of state guarantees for the extension of ELA should be determined.

53. Arrangements for the management of a system-wide financial crisis should be put in place. The MoF should take the lead to streamline and clarify the modalities of coordination arrangements between domestic authorities and the SRB and ECB. Contingency plans should be developed and tested via crisis simulation exercises.

References

  • Banque centrale du Luxembourg (2015), “Interconnectedness between banks and market-based financing entities in Luxembourg,” Revue de Stabilité Financière, May 2015, Luxembourg.

    • Search Google Scholar
    • Export Citation

  • Diebold, Francis X., and Kamil Yilmaz (2014), “On the network topology of variance decompositions: Measuring the connectedness of financial firms,” Journal of Econometrics 182, No. 1: 119–134.

    • Search Google Scholar
    • Export Citation

  • Espinosa-Vega, Marco, and Julian Solé (2010), “Cross-border Financial Surveillance: A Network Perspective,” IMF Working Paper No. 105, Washington D.C.

    • Search Google Scholar
    • Export Citation

  • European Commission, “Financial Reforms and their Progress,” available at: https://ec.europa.eu/info/business-economy-euro/banking-and-finance/financial-reforms-and-their-progress_en.

    • Search Google Scholar
    • Export Citation

  • European Securities and Markets Authority (2015), “Measuring the Shadow banking system -a focused approach,” Trends, Risks and Vulnerabilities, No. 2, September 2015, Paris.

    • Search Google Scholar
    • Export Citation

  • European Systemic Risk Board (2016), “Risk Dashboard,” Issue 15, March 2016, Frankfurt.

  • Igan, Deniz, and Prakash Loungani (2012), “Global Housing Cycles,” IMF Working Paper No. 212, Washington D.C.

  • International Monetary Fund (2013), “Key Aspects of Macroprudential Policies,” IMF Policy Paper, Washington D.C.

  • International Monetary Fund (2016a), “Luxembourg—Staff Report for the 2016 Article IV Consultation.” IMF Country Report No. 16/119, April 2016, Washington D.C.

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund (2016b), “Technical Note—Asset Management and Financial Stability,” Ireland Financial Sector Assessment Program, IMF Country Report No. 16/312, September 2016, Washington D.C.

    • Search Google Scholar
    • Export Citation

  • IMF-FSB-BIS (2016), “Elements of Effective Macroprudential Policies: Lessons from International Experience,” August 2016, Available at: www.imf.org/external/np/g20/pdf/2016/083116.pdf

    • Search Google Scholar
    • Export Citation

  • Segoviano, M., and C. Goodhart (2009), “Banking Stability Measures,” IMF Working Paper No. 09/04, Washington D.C.

Annex I. Implementation Status of Key 2011 FSAP Recommendations

article image
article image
article image
article image