United Kingdom
Financial System Stability Assessment

In recent years, the IMF has released a growing number of reports and other documents covering economic and financial developments and trends in member countries. Each report, prepared by a staff team after discussions with government officials, is published at the option of the member country.


In recent years, the IMF has released a growing number of reports and other documents covering economic and financial developments and trends in member countries. Each report, prepared by a staff team after discussions with government officials, is published at the option of the member country.

I. Introduction

1. The United Kingdom is both home and host to large domestic and international financial institutions. International foreign bank subsidiaries and branches hold half of U.K. banking assets, while U.K.-owned banks have over half of their assets outside the country. Overall, these institutions hold about £11 trillion of assets in the United Kingdom and globally, equal to about nine times GDP, of which U.K. banks account for assets equivalent to five times GDP. The U.K. insurance industry is the third largest in the world, after the United States and Japan. The United Kingdom is an important host to the global funds industry and to derivatives business and plays a significant role in cross-border trading and clearing (Table 2). Hence, U.K. financial institutions and market infrastructure are widely exposed to stresses originating from elsewhere in the world and also act as a conduit of shocks domestically and to the global financial system.

Table 1.

United Kingdom FSAP Update: Main Recommendations

article image

“Immediate” is within one year; “near-term” is 1-3 years; “medium-term” is 3-5 years.

Table 2.

United Kingdom: Global Market Share in Financial Services, as at June 2011

article image
Source: TheCityUK.

2. The U.K. banking system evolved intensively after 2000 in ways that made it more vulnerable to adverse shocks. The banks became larger, highly complex and leveraged, and relied increasingly on short-term wholesale funding (Figure 1). The trading book displaced loans as the most important balance sheet asset, reducing the importance of net interest income (Box 1). Financial activity was concentrated in a small set of institutions highly interconnected to other developed financial systems. Actual leverage was greater than was apparent, in part because regulatory requirements did not capture key risks.

Figure 1.
Figure 1.

United Kingdom: Bank Lending and Funding

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Bank of England (BoE), FSA calculations; CB Richard Ellis, De Montfort University; Dealogic DCM Analysis; financial statements of major U.K. banks; ICB; Office of National Statistics (ONS); and IMF staff calculations.

3. The crisis revealed major fault lines in the U.K. financial sector and its regulatory framework. The heavy reliance on market discipline and the assumption of a wide dispersion of risks proved to be inadequate. Substantial risks posed by large, complex, and interconnected financial institutions crystallized, exposing weaknesses in the oversight framework that had enabled the expansion of their reach and complexity, both domestically and internationally (Box 2).

4. Against this background, this report concentrates on two interrelated themes:

  • Crisis implications, the current soundness of the financial system, and potential risks and vulnerabilities; and

  • Necessary improvements to the financial sector oversight framework and addressing the too-important-to-fail (TITF) problem.

5. The report is structured as follows. It first discusses the post-crisis financial conditions and analyzes the risks and vulnerabilities. It then evaluates the existing regulatory, supervisory, and crisis management frameworks and provides recommendations on how to address the shortcomings in the context of the ongoing deliberations on global financial sector reforms.

II. Soundness of the Financial System and Potential Risks

A. The Crisis

6. The impact of the crisis on the U.K. banking sector was highly polarized. Some institutions emerged from the crisis virtually unscathed while others, including two of the six largest banks and several building societies, required extensive official support. This differentiation reflected their disparate business models and pre-crisis exposures (Figure 2 and Table 3).1 In particular, the reliance on wholesale funding ranged substantially, reflecting wide differences in loan-to-deposit ratios among institutions.

Figure 2.
Figure 2.

Major U.K. Banks: Differentiated Geographic and Business Models

(As at end-2010, in percent of individual banks’ revenues)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Bloomberg; individual banks’ Annual Reports; and IMF staff estimates.
Table 3.

United Kingdom: Composition of the Banking System

article image
Sources: FSA; and BoE.Note: The core stress test sample of the seven major U.K. banks represents 71.1 percent of banking sector assets and 89 percent of loans and advances to customers.

Non-trading book.

7. The insurance sector weathered the crisis relatively well. The introduction of risk-sensitive capital requirements ahead of Solvency II and the related adjustments to capital and investment in the aftermath of the 2003 insurance sector crisis resulted in a relatively solvent industry. While the impact of the global financial crisis reverberated through the industry, especially in the life insurance sector, the industry as a whole remained largely resilient.

8. The interbank and foreign exchange markets came under severe pressure during the crisis, necessitating policy interventions. The ability of U.K. banks to access interbank and wholesale funding markets depended on their underlying conditions—and perceptions of the counterparty risks they represented—but were also affected by the disruptions to the functioning of markets.

9. The U.K. authorities acted decisively and introduced policy measures to stabilize the financial system (Appendix II). These comprise liquidity support, including the Special Liquidity Scheme (SLS), which allows banks to swap their high-quality but illiquid mortgage-backed and other securities for U.K. treasury bills; the Credit Guarantee Scheme (CGS), which provides additional funding support; as well as an asset protection scheme (APS) against future credit losses; and an Asset Purchase Facility (APF) to purchase high-quality assets from a broad range of corporates. Although banks are ahead of schedule in repaying official liquidity support, the government retains significant stakes in the financial sector after having injected a cumulative £70 billion of capital into several banks (including taking ownership of 41 percent and 83 percent of the total share capital of LBG and RBS, respectively). As a result, the net direct cost of fiscal support to the U.K. financial sector relative to GDP is one of the largest in the EU, after Ireland and Germany (Table 4).

Table 4.

Selected Advanced Economies: Recovery of Outlays and Net Cost of Financial Sector Support 1/

(Latest available data as at end-March 2011; in percent of 2010 GDP)

article image
Source: Fiscal Monitor, IMF.Note: Fiscal outlays of the central government. In addition, some countries may have supported financial institutions via fiscal outlays at the subnational level or through other public sector institutions. For example, in Germany, capital injections from the Laender and KfW (development bank) amount to 1.1 percent of GDP; in Belgium, financial sector support from regional government amount to 1.6 percent of GDP.

Cumulative since the beginning of the crisis based on the latest available data (ranging between end-December 2010 and end-March 2011).

Direct support does not include asset purchases by the National Asset Management Agency, as these are not financed directly through the general government but with government-guaranteed bonds.

Direct support includes an estimated amount of €240 billion (9½ percent of GDP) for asset purchases.

B. Financial Sector Conditions in the Aftermath of the Crisis

10. U.K. banks have made progress in repairing their balance sheets (Table 5). Deleveraging was pursued through write-offs, capital-raising efforts, asset sales, reduced lending growth, and retrenchment of cross-border exposures, including to vulnerable European countries (Figure 3). The combined total assets of the seven major U.K. banks fell 20 percent from 2008 to 2010 and risk-weighted assets by 7 percent (Figure 3). With deposits growing faster than loans since 2009, the funding gap—albeit still large—is declining and the use of wholesale funding has been reduced significantly (Figure 4). The Core Tier 1 ratios of all the major U.K. banks are in double-digit territory (Figure 5), which compares well with most European peers.

Table 5.

United Kingdom: Financial Soundness Indicators 1/

(In percent)

article image
article image
Source: IMF.

The indicators in the table are only available collectively from December 2008 onwards.

Figure 3.
Figure 3.

U.K. Banking System: Composition of Assets and Risk-Weighted Assets

(In trillions of pound sterling)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Source: BoE.
Figure 4.
Figure 4.

United Kingdom: Funding Issuance by Banks and Their Maturity Profiles

(In billions of U.S. dollars)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Source: Dealogic.
Figure 5.
Figure 5.

United Kingdom: Financial Soundness of Major Banks 1/

(In percent)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Annual reports of individual banks; and IMF staff estimates.1/ Sample comprises Barclays, HSBC, LBG, Nationwide, RBS, SCB and Santander U.K.

11. However, some loan portfolios remain fragile and further provisioning is necessary. Banks’ profitability in 2010 was largely attributable to the significant drop in provisions (Figure 6). This partly reflects model-based approaches used by some banks for provisions that may have relied on overoptimistic assumptions about interest rates, inflation, unemployment, and house prices. With coverage ratios trending down close to 50 percent on average for the sector amid a wide dispersion across banks, there is limited cushion to absorb further losses that may affect banks, should the economy take a turn for the worse.

Figure 6.
Figure 6.

United Kingdom: Banks’ Asset Quality

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: BoE; Bloomberg; Council of Mortgage Lenders; FSA, Prudential Risk Outlook; Haver Analytics; Land Registry; Office of National Statistics; and IMF staff estimates.1/Sample includes Barclays, HSBC, LBG, Nationwide, RBS, SCB and Santander U.K.

12. Furthermore, lender forbearance—confirmed in discussions with supervisors and banks—and poor data quality may mask and delay recognition of the true extent of risks.

This problem is of particular concern in the real estate sector. While some degree of restructuring is acceptable for particular assets and debtors for a reasonable period of time, it is not yet clear whether this approach is masking true impairment and whether restructured assets are properly accounted for and monitored. The FSA recently conducted a thematic review to determine the extent of forbearance among banks and introduced formal guidance regarding reporting and classification of restructured loans and provisioning.

13. Funding costs are likely to remain elevated for some banks. Deleveraging and deposit mobilization are reducing the funding gap, but banks have substantial funding needs, amounting to about £400 billion through end 2012 (see Figure 4),2 including the expiration of official funding facilities. A number of short-term and structural factors can be expected to put upward pressure on the cost of wholesale funding for U.K. banks: (i) the rollover of cheaper debt (in particular official funding facilities) into more expensive new debt at a time when sovereign debt issuance will be elevated; (ii) regulatory uncertainties regarding the status of senior secured debt holders in a “bail in” and possible structural changes as a result of the ongoing Independent Commission on Banking (ICB) deliberations, as well as overseas regulatory changes limiting some traditional sources of funding (U.S. money market mutual funds); and (iii) potential changes in institutional investor preferences (e.g., pressure on traditional longer-term investors such as insurance companies and pension funds to shorten durations) due to expectations of increase in sovereign yields and regulatory changes (such as Solvency II for the insurance sector).

14. The two large institutions with government stakes made progress in the implementation of their restructuring programs, but challenges remain. These banks, RBS and LBG, are now in the third year of a five-year restructuring plan and are ahead of the targets set by the U.K. authorities and the European Commission (EC) state aid rules. Through a combination of run-offs and sales, the two banks shed a combined £235 billion of assets since 2008 and have repaid a substantial amount of their use of official funding facilities. They have also reduced their reliance on wholesale funding and lengthened the duration of their liabilities. However, they still face significant internal execution risks and are particularly vulnerable to circumstances that may derail planned further asset reductions or disrupt wholesale funding markets. Given the substantial amount of government stakes in these two institutions, divestment by the state will take time and will depend critically on the continued implementation of their restructuring programs.

15. Separately, the financial condition of the insurance sector has generally improved post-crisis. Life insurance activity continues to recover following a period of weakness during the crisis and is back to the levels of production seen before the crisis. The nonlife insurance is recently reporting a slight premium decrease, mainly due to the economic downturn. About 10 percent of life insurers and 25 percent of nonlife insurers remain under regulatory observation, with Pillar II solvency ratios less than 130 percent.

C. Macroeconomic and Financial Risks

16. The IMF’s central scenario projects a moderate economic recovery and a gradual return to the inflation target. A soft housing market, headwinds from fiscal consolidation, and the ongoing process of household and bank balance sheet repair are expected to continue to weigh on growth. Led by private sector investment and net exports, real GDP growth is expected to accelerate gradually to around 2.5 percent over the medium term. Inflation is projected to remain above 4 percent due to commodity price and tax hikes for most of 2011 and return to the 2 percent target by end-2012 as these transitory factors dissipate. The current mix of tight fiscal and accommodative monetary policies is seen as appropriate. The central scenario incorporates a gradual tightening of monetary policy with the resumption of growth (Table 6).

Table 6.

United Kingdom: Selected Economic Indicators

article image
Sources: ONS; HMT; BoE; International Financial Statistics; INS; World Development Indicators; and IMF staff estimates and projections.

ILO unemployment; based on Labor Force Survey data.

The fiscal year begins in April. Debt stock data refers to the end of the fiscal year using centered-GDP as a denominator.

2011: Actual data through April.

2011: Actual data through March.

Average. An increase denotes an appreciation.

Based on relative consumer prices.

17. However, the central scenario is susceptible to a number of macroeconomic and financial risks, with implications for the banking system:

  • Sluggish growth. There is uncertainty around the medium-term growth outlook, as headwinds from balance sheet repair by households, banks, and the public sector could turn out to be stronger than expected.

  • Interest rate increases. Higher-than-expected inflation could force faster policy rate hikes than envisaged in the baseline. Households remain heavily indebted and bank losses could surge if asset quality deteriorates as borrowers increasingly struggle to service variable rate loans. The FSAP’s analysis shows that low-income households are most sensitive to real interest rate and income shocks (Box 3).

  • Reduced fiscal space to cope with future adverse shocks. The financial crisis brought about a sharp increase in government debt.

  • A new sharp downturn in the residential housing market. The price-to-income ratio remains 30 percent above its historical average (based on OECD data), leaving household balance sheets vulnerable to further declines in house prices.

  • Financial distress in the commercial real estate (CRE) sector. Prices remain substantially below their peak, leaving a number of borrowers with negative equity and in breach of loan-to-value covenants. The FSAP’s analysis shows that macroeconomic shocks would disproportionately affect this sector (Box 3 ).

  • Spillovers from external financial market shocks. Sustained disruptions in wholesale funding markets due to external shocks could create funding shortages in the system.

  • Depending on their geographic footprint, banks are exposed to realization of risks in other regions such as the Eurozone (where risks are rising sharply), Asia, and Latin America. Direct and indirect exposures to the private and public sector in the three Eurozone countries with EU/IMF-supported programs (Greece, Ireland, and Portugal) amount to $173 billion (approximately £111 billion; 2.8 percent of total assets), with the bulk of this exposure being to Ireland at $135 billion (£86 billion). Additionally, exposures to Spain amount to $107 billion (£69 billion). As documented in the BoE’s June 2011 Financial Stability Report, major U.K. banks’ claims on Ireland and Spain, primarily to the nonbank private sector, is equivalent to about 50 percent of their Core Tier 1 capital (split evenly between the two countries), while lending to Italy accounts for a significantly smaller share of capital (about 8.5 percent). Further, some banks in core European countries may be vulnerable to shocks emanating from these countries, which in turn may affect U.K. banks due to their substantial exposures to those banks.

18. The analysis of banking sector conditions and stress tests aim to capture the most salient risks to the U.K. banking system (Figure 7). Solvency stress tests have been conducted on the major U.K. banks against the Basel III hurdle rates, while the FSA Interim Capital Regime requirements (see discussion on banking supervision below) are also discussed. Separately, liquidity stress tests were undertaken for a larger set of institutions, including foreign subsidiaries of retail and investment banks as well as a broader group of building societies, incorporating quasi-Basel III measures.

Figure 7.
Figure 7.

Overview of the U.K. FSAP Update Stress Testing Exercise

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

19. The findings support the current supervisory focus in the United Kingdom to require banks to build up adequate capital and liquidity buffers. They suggest that the banking system would be resilient to most of the macro-financial risks described above, but that there are pockets of vulnerabilities that need to be addressed:3

  • Banks’ solid capital buffers have reduced solvency concerns to a low-probability confluence of adverse macroeconomic developments. The analysis suggests that the substantial recapitalization of the U.K. banking sector, including by the state, and the de-risking of balance sheets have been critical in boosting the solvency of the system. However, an important caveat is that lender forbearance—which could not be accurately estimated in the stress tests due to lack of data—may have masked the extent of credit risk in some institutions, especially given the high indebtedness of the household and commercial real estate sectors.4

  • The banking system, in aggregate, has adequate liquidity buffers to withstand short-lived stresses, but remains vulnerable to sustained disruptions in funding markets. The overall liquidity shortfall remains contained even in a very severe 30-day stress test. However, sustained disruptions to wholesale funding markets, coupled with persistent declines in asset values could expose vulnerabilities. Funding sources are not sufficiently stable (see discussion on banks’ Net Stable Funding Ratio below) and banks, including the major ones, have mismatched assets and liabilities at six-month maturities and longer. The substantial funding needs and the interconnectedness of the U.K. banks to the global financial system amplify the risks. The new discount window facility and pre-positioning of collateral aimed at facilitating quick use of the facility are important in managing such extreme stress events.

20. The FSA’s stress tests of the insurance sector indicate resilience, but the low interest rate environment could create adverse incentives. Risk weightings under Solvency II may encourage insurers to allocate their portfolios as insurance companies seek to meet nominal return objectives and offset competition in their underwriting businesses. Supervisors need to remain vigilant regarding the quality of the investment portfolio and, if necessary, apply Pillar II requirements under Solvency II.

Banking stress test analysis

21. Stress tests were conducted for the seven major U.K. banks to assess solvency risk under a baseline and three adverse scenarios. These banks account for 71 percent of the assets in the system (80 percent of loans) and 90 percent of retail deposits. The stress-test horizon is from 2011-2015 using end-2010 financial data. The baseline projections are consistent with the IMF’s World Economic Outlook, while the adverse scenarios comprise: (i) a “double-dip” recession scenario of one standard deviation of real GDP growth from the baseline growth trend over the first two years of a five-year horizon with positive adjustment dynamics during the subsequent three years (“DD mild”);5 (ii) a double-dip recession scenario of two standard deviations of the same (“DD severe”); and (iii) a “prolonged slow growth” scenario with a cumulative negative deviation of over 7 percentage points from baseline growth, or an average annual growth rate of almost 0.9 percent over five years (“SG”). Related key macro and financial variables are projected using the FSA’s macro models (Figure 8 and Appendix III).

Figure 8.
Figure 8.

Overview of the U.K. FSAP Update Stress Test Scenarios

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: BoE; FSA; and IMF staff calculations.

22. Bottom-up (BU) solvency stress tests represent the core element of the analysis and are cross-validated by top-down (TD) tests. Prescriptive assumptions covering areas such as risk, behavioral adjustments, and regulatory changes are also applied as consistently as possible across the three approaches (Appendix IV). The results are necessarily subject to wide confidence bands and can be summarized as follows:

  • BU results show that banks have robust capital buffers that could absorb losses from severe stresses. The distribution of Core Tier 1 ratios for individual banks remain over the regulatory minima at any time over the 2011-15 period, per the Basel III schedule (Figure 9). The weighted-average capital ratios comfortably exceed the regulatory minima; and the average leverage ratio remains over 4.5 percent. Capital buffers comfortably absorb credit losses arising from a sharp recession, with retail-focused banks hardest hit. Buffers are also sufficient to absorb market-risk charges due to exchange rate shocks to major currencies and higher risk weights on securitization and counterparty risk exposures under revised regulatory standards. Exposures to sovereign and bank debt do not seem to represent a major source of risk;6 aggregate losses of around 0.2 percent of GDP (£3.5 billion) over the five-year horizon are estimated for the former, which represents the greater risk, unless stresses lead to severe disruptions in wholesale funding markets (see below).

  • The authorities’ TD tests using the Bank of England’s Risk Assessment Model for Systemic Institutions (RAMSI) show similar broad trends under each scenario.7 In contrast to the BU findings, however, shocks to sovereign and bank debt appear to have greater effect on capital adequacy, of up to 0.9 percentage points on core Tier 1 capital, which remains well above the relevant hurdle rates nonetheless (Figure 10).

  • The FSAP’s TD approach, the Systemic Contingent Claims Analysis (Systemic CCA), is applied to estimate systemic solvency risk.8 The estimated joint potential capital shortfall suggests that markets are broadly comfortable that the banks have sufficient buffers to withstand the prescribed shocks (Figure 11 and Table 7). Under (median) baseline conditions, the impact on market-implied capital from the realization of declining profitability amounts to an average 0.03 percent of 2010 GDP (£0.4 billion) between 2011-15. Under the severe double-dip recession scenario, market-implied capital levels could decline by up to an average 0.12 percent of 2010 GDP (£1.8 billion) over the next five years. If multiple banks experience a dramatic escalation of losses under a “tail of the tail risk” scenario (0.1 percent probability event), current market-implied capital levels over the five-year stress test horizon could potentially be eroded by between 2.7-3.4 percent of 2010 GDP (£37-£50 billion) on average, albeit still well below the peaks seen during the Lehman crisis (Figure 11 and Table 7). Under this scenario, an average capital shortfall of between 1.3-1.6 percent of 2010 GDP could materialize relative to the Basel III Tier 1 hurdle rates (and 1.6-1.8 percent of 2010 GDP relative to the FSA Interim Capital Regime Tier 1 requirements).

Figure 9.
Figure 9.

Distribution of Core Tier 1 Capital Ratios from the U.K. FSAP Update Bottom-Up Stress Tests, 2011-15

(In percent)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: FSA; major U.K. banks; and IMF staff calculations.
Figure 10.
Figure 10.

Weighted-Average Core Tier 1 Capital Ratios from the U.K. FSAP Update Bottom-Up and BoE RAMSI Top-Down Stress Tests, 2011-15

(In percentage points)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: BoE; FSA; individual banks; and IMF staff calculations.Note: The definitions of capital are as follows: For the BU exercise, the starting point is in line with FSA definitions as laid out in the FSA Handbook and the definition of capital in the FSA Interim Capital Regime; for the RAMSI, the Basel II-consistent definition of capital is used.
Figure 11.
Figure 11.

Systemic CCA Estimates of the Market-Implied Joint Capital Losses from the U.K. FSAP Update Top-Down Stress Tests, Historical and Expected (with IMF Satellite Model)

(In billions of pound sterling)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Source: IMF staff calculations.
Table 7.

Systemic CCA Estimates of the Market-Implied Joint Expected Capital Loss and Resulting Shortfall from the U.K. FSAP Update Top-Down Stress Tests, 2011-15

(Average over time period, in billions of pound sterling unless stated otherwise)

article image
Source: IMF staff calculations.Note: The estimations show the joint capital requirements for maintaining the market value of Tier 1 capital, with a gradual increase of the hurdle rate from 2013 onwards consistent with the Basel III proposal as at December 2010.

The IMF satellite model uses a set of macroeconomic variables (short-term interest rates, long-term interest rates, real GDP growth, and unemployment) as well as income elements specific to each bank (operating profit, net interest income) to project expected losses generated by the CCA methodology.

As an alternative, projected operating profit based on RAMSI model results is integrated in the CCA framework by adjusting implied bank assets, which increase expected losses via an option pricing approach. The treatment of losses from haircuts on holdings of sovereign and bank debt differs between both satellite model approaches. In the case of the former, these losses are calculated each year and added to the estimated overall expected losses. In contrast, for the alternative satellite model, losses from these debt holdings are subtracted from the RAMSI-model projected operating profit each quarter.

The tail risk at the 95th percentile represents the average probability density beyond the 95th percentile as a threshold level.

23. An important caveat to the haircuts applied to banks’ debt holdings of all non-AAA rated sovereigns and banks in those countries is that their severity and dynamics are informed by the forward term structure of 5-year sovereign credit default swap (CDS) spreads as at end-2010. Since then, the sovereign CDS spreads for select debtor countries of U.K. banks have increased commensurately with the rising risks to the economic outlook (Figure 12), which are not reflected in the haircuts, given the cut-off point for the stress tests. This means that U.K. banks could be affected by additional losses—well beyond the prescribed haircuts—in the event that shocks lead to extreme stresses in the private sector in those countries or in core European banks to which they have large exposures.

Figure 12.
Figure 12.

Selected EU Countries: CDS Spreads

(In basis points)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Source: Markit.

24. The liquidity tests assess the short-term resilience of the system to sudden, sizeable withdrawals of funding. The sample includes 16 institutions—the 7 major banks, selected foreign investment banks and building societies. Two implied cash-flow tests—based on the Quantitative Impact Study (QIS-6) assumptions—simulate a gradual outflow of funding over five consecutive days and over a 30-day time horizon in keeping with a reverse stress test rationale (Appendix V). The FSAP team’s interpretation of the proposed Basel III liquidity risk measures are also applied—the Liquidity Coverage Ratio (LCR) measures the sufficiency of the stock of high quality liquid assets to meet up to 30 days of institution-specific and systemic stress and the Net Stable Funding Ratio (NSFR) provides insights into the quality of liquid assets and the maturity profile and stability of funding in the system. We refer to these as quasi-measures, given that they are not yet definite. The findings are summarized below (Tables 8 and 9):

Table 8.

U.K. FSAP Update Liquidity Stress Tests—5- and 30-Day Implied Cash Flow Analysis

article image
Sources: FSA; and IMF staff calculations.

Cumulative outflows are weighted across different types of deposits and sources of wholesale funding, whose relative magnitude differs across sample banks. Note that the implied cash flow analysis applies outflow assumptions to aggregate values of funding maturities of up to one month irrespective of callability. The FSA does not complete tests that are directly comparable to these results. However, the survival day metric monitored by the FSA to assess vulnerabilities to wholesale funding is broadly similar to the 30-day test.

Table 9.

U.K. FSAP Update Liquidity Stress Tests—Quasi-Basel III and Maturity Mismatch Analysis

(In percent of total sample assets included in the liquidity tests)

article image
Sources: FSA; and IMF staff calculations.1/ The maturity mismatch of each bank is calculated as the difference of the value of assets and liabilities in each of six “maturity buckets” determined by their tenor and callability, i.e., (i) on demand; (ii) up to one month; (iii) between one month and three months; (iv) between three months and six months; (v) between 6 months and 12 months, and (v) greater than 12 months. For a given “maturity bucket,” the total assets of all banks that exhibit mismatch are added and divided by the total assets of all banks in the sample. The differences between the FSAP and FSA assumptions lie largely in the interpretation of the proposed Basel III measures.
  • Banks have sufficient holdings of highly-liquid assets to offset outflows during short-lived (five days) moderately severe cash-flow shocks, assuming that deposits remain stable. In the case of deposit outflows, the aggregate shortfall would equal 0.03 percent of assets for the sample of 16 banks. The results highlight the importance of retail deposits as a key source of funding for the banking system, while intra-group funding does not appear to play an important role for the foreign subsidiaries in the sample.

  • The overall liquidity shortfall remains largely contained even when deposit outflows and the run-off of liabilities are doubled in a very severe 30-day scenario. The FSAP’s assumed cumulative outflows of more than 40 percent of both wholesale and deposit funding under this test exceed current supervisory requirements and are greater than those experienced by Northern Rock during the crisis. Given the severity of the underlying assumptions, this test would, in effect, help identify institutions’ “point of failure” in the nature of a reverse stress test. The aggregate shortfall amounts to about 6 percent of assets for 14 out of the sample of 16 banks, with significant impact concentrated in a few institutions.

  • Sources of funding in the banking system are not sufficiently stable. The quasi-NSFR test shows that banks accounting for more than 95 percent of liquid assets do not have sufficient stable funding (i.e., NSFR ratio of less than unity). In the 6-12 months maturity bucket, assets that are mismatched amount to about 70 percent of the total assets of all banks in the sample, and the proportion goes up to around 90 percent in the 12-plus months maturity bucket. For some banks, this vulnerability would be amplified by their reliance on wholesale funding.

Insurance stress test analysis

25. The quality of the insurance stress testing carried out by the FSA is high and consistent with the sophistication of the market. While the Individual Capital Assessment (ICA) regime has required stress testing at an individual firm level since its introduction in 2005, industry-wide stress testing only began in 2008 when the FSA identified the need to gain a better understanding of the resilience of the market in the wake of the crisis. Since then, the FSA has carried out, at least on a yearly basis, different types of stress testing, including firm-specific reverse stress tests that capture the extreme situations leading to the failure of an insurer.

26. Insurance stress tests undertaken by the FSA in 2010 targeted the six largest insurance groups based on end-December 2009 balance sheets. The exercise, part of the recently established Core Prudential Program Insurance (CPPi), consisted of a series of stress scenarios, including extreme tail-risk scenarios, to test regulatory surplus capital and group liquidity. The firms passed all severe tests at group and solo levels. For life insurers, credit risk remains the largest driver of risks to solvency; market and property risks are also significant. The results led to a series of supervisory recommendations, including changes to the investment portfolios and strengthening of interest rate exposure management. Based on the results of the extreme tests, a series of actions were imposed on firms to further improve their resilience beyond regulatory requirements.

III. Toward a More Effective Oversight Framework

27. The systemic importance, size, and complexity of the U.K. financial markets necessitate the highest quality regulatory and supervisory framework. Domestic and global shocks transmit through the United Kingdom and may be amplified in transmission. Thus, proportionally intense supervision—commensurate with the importance of the U.K.’s global role—is necessary to ensure domestic and global financial stability.9

28. In the aftermath of the crisis, the U.K. government decided to overhaul the domestic institutional framework for regulation and supervision. Accordingly, the existing tripartite regulatory regime will be dismantled; the FSA will cease to exist in its current form; and a “three peak” structure will be put in place by end-2012; specifically:

  • A new prudential regulator, the Prudential Regulation Authority (PRA), a subsidiary of the BoE, will carry out the prudential regulation of financial firms, including banks, investment banks, building societies, and insurance companies;

  • An independent Financial Policy Committee (FPC) will be established at the BoE, with responsibility for regulation of stability of the financial system as a whole;

  • A new Financial Conduct Authority (FCA) will regulate the conduct of authorized financial firms providing services to consumers. It will also be responsible for ensuring the good conduct of business in U.K. retail and wholesale financial services, market regulation, and prudential regulation of smaller entities.

29. A key risk is that the transition to the new framework diverts resources from efforts to enhance supervision of the financial sector, which is still in recovery mode. It is critical to build on the progress made on strengthening supervision of the banking and insurance sector; this may require additional supervisory resources. The reform proposal to give explicit and clear mandate for financial stability to the prudential regulator is welcome. However, the broad remit of the new conduct authority may dilute the focus on market regulation and supervision, and prudential supervision of smaller entities. This risks recreating the conditions that resulted in broadening of the FSA’s objectives and thus diluting its focus in the lead-up to the crisis. Thus, explicit clarity in legislation is needed.

30. Another concern relates to the challenges in implementing the new governance arrangements to achieve the desired results. While the proposals have built in a variety of cooperation and coordination mechanisms, the experience from other countries suggests the need for one forum for formal and/or informal relationship building, cooperation, and coordination among authorities. These relationships will be necessary for effective governance of the framework. There is no formal crisis management committee established in legislation, but coordination mechanisms are in place. A clear locus for all regulatory agencies to interact might also assist the United Kingdom in interacting effectively with the new European Financial Stability Board (EFSB) and with the European Supervisory Authorities (ESAs).

31. The establishment of the FPC is an important step in developing mechanisms to mitigate systemic risk. For the FPC to be credible, it will be important to set realistic expectations of what macroprudential regulation can achieve, especially in its early stages of implementation. A range of macroprudential tools should be considered, given uncertainties regarding their effects and the nature of future risks.

A. Dealing with Too-Important-to-Fail and the International Dimension

32. Progress has been made in addressing the too-important-to-fail (TITF) problem, but more needs to be done. Regulatory ratios have been strengthened, and a bank levy on wholesale funding has been introduced. Ring fencing of retail operations (a form of retail subsidiarization) and establishment of depositor preference, as proposed by the Independent Commission on Banking, should help improve resolvability of the retail entity. However, ring fencing must be weighed against the costs of such an approach and does not necessarily improve resolvability of the whole entity unless complemented by measures that improve loss-absorption capacity (capital and liquidity surcharges, contingent capital, debt subject to bail-in), recovery and resolution plans, and cross-border resolution arrangements. International collaboration will be critical for progress in these areas and the U.K. authorities should continue exercising leadership on these matters. If such progress is not achieved, there will be a strong push toward more ring fencing and stand-alone subsidiarization which have efficiency and, possibly, financial stability implications.

33. Central Counter Parties (CCPs) should be subject to robust standards. CCPs enhance transparency and simplify the complex web of counterparty exposures through multilateral netting and they stand between each counterparty. As a result, they centralize and concentrate counterparty credit risk. Thus, it is vital that they maintain robust prudential and risk-management standards, including financial soundness requirements for access by their members, initial and variation margining, and levying contributions to a default fund. Given their systemic nature, a clear ex-ante framework and contingency plans, including the role of the surviving clearing members, are needed to deal with the implications of their failure. And given their global nature, cross-border coordination is required to avoid regulatory arbitrage and to mitigate systemic risk and adverse spillovers across countries.

34. Oversight of cross-border institutions needs to be improved further. The United Kingdom is a global hub and the authorities’ powers vary depending on whether incoming institutions are branches or subsidiaries and whether the parent entity is in the EU or not. While the United Kingdom has strong powers over subsidiaries, it has very limited powers (liquidity supervision only) in relation to EU branches and cannot prevent an incoming EU branch from establishing itself. Equally, the United Kingdom is a major home-state jurisdiction and there will be many host jurisdictions for which U.K. firms play a systemic role. Gaps in this domain must be and can only be addressed through international cooperation, in particular across the key financial centers. Better collaboration between home and host authorities, including joint inspections and decision-making processes is needed, as well as enhanced sharing of information using both formalized channels such as colleges and well-developed bilateral contacts. Important issues in cross-border crisis management and resolution remain unresolved, notably on burden sharing and on harmonizing legal frameworks.

35. Investment banking is a particular challenge to cross-border oversight, given the international nature of wholesale and trading activities. Without intensive supervision of investment bank risk taking, domestic and global financial stability cannot be assured. The United Kingdom is one of the major international markets and, as most if not all key players are active in London, the authorities are well placed to have oversight of investment banking activities if efforts at cross-border coordination are taken, so that developing risks in the wholesale markets are not overlooked.

36. The U.K. authorities should continue to work toward an ambitious international package of regulatory reform. The authorities are rightly advocating for European legislation under the EU Capital Requirements Directive (CRD4) that enables the establishment of strong standards that (i) exceed the Basel III minima, including by setting ambitiously high capital requirements together with significantly topped-up capital demands on systemically important financial institutions; and (ii) allows flexibility for national authorities to introduce macroprudential tools, including adjusting capital and liquidity requirements or varying risk weights to address emerging financial and systemic risks. Collaboration with other macroprudential bodies will be important to ensure home-host coordination and reciprocity where appropriate.

B. Improving the Supervision of Banks

37. Serious deficiencies in risk measurement were revealed by the crisis, as were the shortcomings in supervisors’ ability to identify and rectify those deficiencies. Following the Turner Report, the FSA embarked on a program of supervisory enhancement.10 This has been articulated as the “intensive and intrusive” approach and “close and continuous supervision” for the larger banks.

38. The current supervisory framework was assessed according to the revised Core Principles for Banking Supervision (2006). This is significant for two reasons: (i) the revised principles have a heightened focus on risk management and its practice by supervised institutions, and its assessment by the supervisory authority; and (ii) the standards are evaluated in the context of a financial system’s sophistication and complexity vis-à-vis the capacity and resources for adequate oversight of that particular system. These factors should be taken into account in comparing assessments over time and across countries.

39. The authorities need to continue to address shortcomings in supervisory processes. These include clarifying the prudential mandate, improving data quality and the supervisory approach, and techniques and reporting. The Supervisory Enhancement Program (SEP) and the Core Prudential Program (CPP) have recently been put in place by the FSA to overcome these shortcomings. However, there has been insufficient time for these new programs to become adequately embedded. Furthermore, they need to be enhanced with more detailed verification on a proactive basis than appears to be currently planned, better integration of specialist work into the supervision program, and more thematic peer analysis. These efforts should focus on identifying actual or potential risks and addressing risk-management deficiencies.

40. The regulatory authority should be provided with enforcement powers at the holding company level.11 The FSA’s present powers to take supervisory or remedial actions are targeted at the regulated entities that fall within the consolidated group. In other words, the FSA can only impose consolidated obligations by acting through the U.K.-regulated entities, which may not be effective when those entities in the group are small relative to the non-U.K.-regulated entities over which the supervisor has no direct control. Legislation should be amended to provide holding company level enforcement powers.

41. The FSA has been effective in enhancing the quantity and quality of banks’ capital. Capital rules are compliant with Basel and European requirements, including Basel III trading book capital charges, according to the agreed gradual phase-in schedule, in order to balance the potential adverse implications on lending. The FSA further strengthened capital regulations by promulgating an Interim Capital Regime starting in 2009, based on capital expectations in normal and stressed conditions, which ensures that U.K. banks are well placed to meet the Basel III capital requirements, and approval of dividends and variable remuneration is linked to the outcome of stress tests. The robust stress-testing process for major banks is instrumental in allowing the FSA to set further capital buffers for major banks, above the normal individual capital guidance. Banks are operating with material cushions above the guidance and buffers that appear to be able to withstand severe shocks as prescribed in the FSAP stress tests (see above). The FSA recently introduced formal guidance regarding reporting, classification, and provisioning for loans subject to lender forbearance.

42. The FSA has appropriately required regulated banks to build up liquidity buffers, but liquidity supervision of branches should be improved. Extensive monitoring, stress testing, and assessment of liquidity positions, including the setting of individual bank liquidity guidance, are in place and will be further enhanced. Liquidity positions of major banks have improved. There is a plan for these detailed assessments to be extended to a wider range of high-impact banks, and this needs to proceed expeditiously. However, the FSA does not collect sufficiently detailed liquidity information, even on the large foreign branches that may have systemic relevance for the U.K. financial system. The FSA applied liquidity stress tests to foreign banks’ group data before granting “whole firm modifications,” whereby it delegated responsibility for liquidity supervision over foreign branches located in the United Kingdom, even though it has powers under EU rules to do so. As a result, the FSA does not have specific information on the activities of U.K. branches of foreign banks, e.g., on rehypothecation, an area in which foreign branches dominate.

43. Achieving the highest standards on home-host relationships is paramount for the United Kingdom, given its central role in the global financial system. The appropriate legal authority to preserve confidentiality is in place. However, there is a need to improve the information sharing between the United Kingdom and other countries on both home and host basis.

44. Additional supervisory resources—a combination of enhanced skills and additional persons—may be needed. Resources in supervision have been increased, as have specialist support resources in various risk areas. The authorities need to agree on a supervisory operating model, as well as the extent of the resource requirements for the new macro-prudential overlay in order to effectively determine further resource needs.

45. The plans to introduce a staged intervention scheme are welcome. The planned scheme will include indicators which, when triggered, will lead to gradually more-intrusive actions by the FSA. The scheme should include an appropriate degree of discretion, but at the same time ensure that appropriate action is taken well before insolvency. It should also include triggers for contacts and coordination actions with other authorities, including the resolution authority and the Financial Services Compensation Scheme (FSCS).12

46. The United Kingdom lags behind many other countries in standards for the public disclosure of bank data. Such disclosures are important for enabling market discipline—analysts and bank counterparties are now limited in their assessment of the risks being assumed by the banks and of the banks’ developments in general. The FSA should enhance publication of regular and comparable data on a bank-by-bank basis, including data from prudential returns, whenever appropriate.

C. Strengthening Insurance Supervision

47. The new intrusive risk-based approach to supervision applied by the FSA is still in its early stages of implementation. This approach should be extended to a wider range of insurers with careful adjustment to ensure it is applied proportionately to firms of various risk levels, including firms represented across a range of key financial risks. Greater use of in-depth analysis should be applied to general insurers. Carrying out more frequent transaction examinations (in relevant areas such as underwriting, reinsurance, claims, and even investments) for both the largest insurers and some of the smaller insurers on a random basis can help reveal problems.

48. Implementation of Solvency II, on target for 2013, may necessitate additional resources. The approval of internal models is a crucial step for the insurance industry in this highly specialized and sophisticated market. This will require sufficient resources and expertise to assess and verify models and the adequacy of the resulting capital levels to support a solvent industry.

49. There is a need to improve the scope and frequency of prudential data as well as its disclosure. The published standard return information has not been substantially reviewed for a number of years, in anticipation of Solvency II, which is now on target to be adopted in 2013. Solvency II will lead to a new basis for both qualitative and quantitative reporting, including much greater public disclosure.

D. Effective Regulation of Securities Markets

50. Robust market surveillance systems are in place and the FSA has full powers to address market abuse. The FSA has put in place an enhanced enforcement program, beginning in 2007, and this appears to have yielded results in creating a more credible deterrence to improper behavior.

51. Investment firms and managers are subject to risk-based supervision, which appears adequate for firms identified as high impact. Greater attention should be paid to mid-sized firms (across all securities-related activities) and, further, there is a greater need to implement “bottom-up” analysis of firm operations using on-site examinations to supplement the “top-down” risk analysis. While many good supervisory initiatives have been put in place, such as the CPP, these are still taking hold.

E. Strengthening the Regulation of Financial Market Infrastructures13

52. Financial market infrastructures (FMIs) are governed by a consistent set of laws, regulations, and contractual arrangements. The role and responsibilities for oversight are clearly defined and transparent. Euroclear U.K. & Ireland Limited, as operator of the securities settlement system, CREST, and recognized CCPs are regulated and supervised by the FSA. Interbank payment systems including the Clearing House Automated Payment System (CHAPS); CLS; Bankers’ Automated Clearing Services (BACS); Faster Payment Services (FPS); and those operated as part of CREST, LCH, and ICE are overseen by the BoE.

53. The oversight of FMIs is risk-based. Both the BoE and FSA have taken steps to strengthen supervision since the crisis. For example, the FSA is now analyzing CCPs’ internal risk-management models in detail, which had not been the case previously. The FSA is also placing greater emphasis on the quality of close and continuous supervision and the use of regulatory judgment; the need to anticipate and explore issues in a proactive manner; and ongoing supervisory programs around the key issues.

54. The BoE operates the real time gross settlement (RTGS) infrastructure for CHAPS and CREST, as well as settlement of other payment schemes. The RTGS infrastructure also facilitates intraday liquidity transfers, reserve account transfers, transfers in respect of the note circulation system, and transfers that can occur outside of CHAPS operating hours. The BoE assesses the RTGS infrastructure against the CPSS Core Principles through its oversight of recognized systems that rely on the RTGS infrastructure for settlement. This approach is indirect and fragmented and may not include all transactions.

55. The BoE does not agree with the FSAP’s recommendation to undertake a direct, unified assessment of the RTGS infrastructure. It does not consider the RTGS to be an interbank payment system but rather an accounting infrastructure that supports some payment systems. It does, however, accept the recommendation to assess the finality of a small number of transfers on RTGS accounts made independently of schemes designated under the Settlement Finality Directive and that these transfers should meet equivalent risk standards. Separately, the BoE intends to conduct in 2011 a unified assessment of the RTGS infrastructure based on its existing internal risk assessment.

56. The U.K. payment and settlement systems are highly tiered, posing vulnerabilities. If a major clearer were to become unavailable, this would likely impose significant liquidity pressures. CHAPS payment flows are very concentrated, with the five most active banks accounting for approximately 80 percent of payment value and the two most active accounting for half of the total value. Consequently, while CHAPS does settle in central bank money, a significant amount of large-value payment activity in the United Kingdom settles in commercial bank money on the books of the main CHAPS members. Exposures to commercial settlement banks are also significantly concentrated in CREST, since two of them account for around 50 percent of the settlement value in sterling. The authorities are well aware of the risks associated with tiering and have made an effort to make banks aware of the risks while encouraging banks with significant activity in the second tier to become direct members. They should institute close monitoring of concentration of payment and settlements activities in banks.

57. Most cash settlement of U.K. CCP transactions (mainly margin calls) is done through commercial bank books (in commercial bank money), which exposes CCPs to intraday credit risk. CCPs should aim to reduce their cash-settlement exposures through settlement in central bank money when possible and practicable.

58. The authorities should work with the CCPs to ensure that other safe and reliable funding options are in place. These should include committed credit lines subject only to presentment, and could also include other options like mutualization of liquidity risk among the CCPs’ membership. For moral hazard reasons, the BoE does not want to provide emergency access to overnight central bank liquidity on a pre-arranged basis. However, there is no legal impediment for the BoE to provide emergency liquidity facilities, if deemed necessary, on a discretionary basis. The authorities should also identify in advance what resolution arrangements will apply to CCPs and in what manner and then ensure that all the necessary legal and operational means are implemented.

F. Crisis Management and Resolution

59. The discount window facility, introduced after the crisis, is filling a useful role and the pre-positioning of collateral facilitates access. The perceived risk of stigmatization of this facility remains an issue—banks avoid using extraordinary funding facilities, so they are not seen by the market as “being in need.” The BoE has undertaken to mitigate this effect—for instance, a broad group of institutions has been invited to use the facility on a regular but low-level basis.

60. A permanent liquidity facility was introduced by BoE in June 2010, allowing the use of wider pools of collateral. Counterparties can simultaneously bid for funds against varying collateral sets—a narrow set which comprises high-quality debt and is expected to remain liquid in all but the most exceptional circumstances, and a wider set which comprises high-quality debt that is expected to remain liquid in most circumstances. This facility is also a useful diagnostic tool, as increasing demand for liquidity against the wider pool of collateral could act as an indicator of growing market tension or funding difficulty.

61. The Special Resolution Regime (SRR) provides the authorities with a range of legal tools for the resolution of distressed banks. Of perhaps most significance are the partial property transfer powers (and accompanying creditor safeguards) that: protect depositors; facilitate resolution through transfers of assets and liabilities; permit loss allocation to shareholders and other creditors; and also ensure that all creditors are left no worse off than in a liquidation.

62. The authorities should consider establishing resolution tools for other potentially systemically important firms not covered by the SRR. This may include investment banks, insurance firms, and FMIs and perhaps, as needed, money market funds and hedge funds, if of systemic importance (Appendix VI). While the present modalities of the SRR may not fit such firms exactly, certain features of the SRR (particularly the private sector transfer arrangements) could be useful in the resolution of these types of firms. If some sort of equivalent to the SRR were developed for certain designated nonbanks, checks and balances would be needed to address potential moral hazard issues that might stem from designation, and sufficiently strong supervision of designated firms would be imperative.

63. Powers of the authorities under the SRR to suspend contractual termination provisions do not currently extend to derivatives contracts. This limitation on SRR powers is due to the strictures of applicable European legislation protecting close-out netting provisions. The authorities noted that the EC is aware of the issue and that EU-wide reform on this technical but important point is being considered.

64. The mandate of the deposit insurance scheme, Financial Services Compensation Scheme (FSCS), is appropriate and operational reforms designed to expedite payouts are commendable. The scheme has been improved after the crisis in a number of ways. In October 2007 in the immediate wake of Northern Rock, the co-insurance aspect of the scheme (whereby the scheme provided only 90 percent coverage beyond the first £2,000 of any claim) was eliminated. In 2010, further enhancements were made to the FSCS in line with emerging European standards, including: an increase in coverage levels to £85,000; faster payout rules (targeting payout within 7 days for most claims and within 20 days for the remainder); and gross payouts (ring fencing the deposits of customers holding savings and loans with the same institution). Since the adoption of the Banking Act 2009, the scheme may also—if required to do so by HMT—participate in resolution transactions by providing funding to support transfers of assets and liabilities to a private sector purchaser or a “bridge bank.” Further improvements, such as moving to a risk-based premia, need to be considered.

65. FSCS funding is raised ex post, creating survivor bias, and the scheme has standing access to HMT finance. FSCS coverage levels are in line with EU Directives and FSCS is working with U.K. banks to develop operational systems that provide a “single depositor view” that should permit more rapid satisfaction of depositor claims.

66. The recently introduced bank levy is to be regarded as a tax on banks and is not intended to build any fund to finance future bank resolution costs. The levy carries similar features as the IMF proposal to the G20 for a financial stability contribution.14 It is based on U.K. banks’ consolidated accounts—on global balance sheets, not solely on their U.K. operations—as well as the U.K. operations of banks from other countries. The levy is designed to target risky funding models and there are a number of features consistent with this, including the exclusion of tier 1 capital and stable deposits from the liability base subject to the levy. It is, however, not linked to a credible and effective resolution mechanism as proposed by the IMF. Rather than act as a fund for the direct costs of resolving any individual bank failure, the levy is intended to ensure that banks make a contribution with respect to the potential risks they pose to the U.K. financial system and wider economy. The levy should be further refined to reflect individual institutions’ riskiness and contributions to systemic risk—such as those related to size, interconnectedness, and substitutability—and variations in overall risk over time.

G. Anti-Money Laundering and Combating the Financing of Terrorism Framework

67. The anti-money laundering and combating the financing of terrorism (AML/CFT) framework of the United Kingdom was evaluated by the Financial Action Task Force (FATF) in 2006/2007. The FATF June 2007 mutual evaluation report (MER) indicates that, overall, the United Kingdom had a comprehensive AML/CFT system in place, but that further improvements were nevertheless needed; weaknesses were notably identified with respect to customer due diligence (CDD) measures in the financial sector, and supervision of small financial institutions.15

68. The United Kingdom updated its AML CFT legislation in 2007 with the issuance of the new Money Laundering Regulations (MLRs). These regulations strengthened the legal arrangements within the United Kingdom in order to further improve compliance with the FATF Recommendations and implement, in part, the Third EU Money Laundering Directive.16 Since the evaluation, the progress made by the U.K. authorities in addressing the most significant deficiencies identified in the MER was analyzed by the FATF in two follow-up reports, respectively, in June and October 2009. These reports indicate that, overall, significant progress had been made. In particular, the deficiencies noted with respect to CDD have been addressed. In addition, the FSA commenced a project looking at AML/CFT systems and controls in small firms: 159 small firms were selected across the wholesale and retail sectors and a report was published in May 2010. The FSA’s Financial Crime Operations Team now routinely visits small firms as part of its case work and ongoing thematic work, and has visited a total of 337 firms from 2007 to 2010.

Did Trading Activities Increase the Vulnerability of Banks to Financial Stress?1/

The increasing involvement by banks in leveraged trading activities in the lead-up to the crisis is largely seen to have played an important role in the financial stress many experienced. A survey by the FSA of 10 investment banks suggests that some $240 billion of losses on trading positions between January 2007 and March 2009 were attributable to structured credit products, mostly linked to the U.S. housing market.2/ Work done at the Fund in this area includes empirically testing the hypothesis that banks with significant trading businesses may have been potentially more vulnerable to shocks:

  • The sample comprises 79 banks across Asia, Europe, and the United States, and includes commercial, investment, and universal banks. Asian banks—which have remained resilient through the crisis—are used as a control group to determine if the hypothesis holds.

  • The proportion of total trading income to total revenue is first calculated for each bank to determine the extent of the bank’s overall revenue source, which may be susceptible to market fluctuations.3/

  • A filter rule test then is applied to identify potentially vulnerable banks—any bank whose share of trading income to total revenue ratio in 2008 exceeds the defined filters is deemed to fall in this category.4/

  • These institutions are then compared against those that received official support in 2008-09.

  • Two separate filter rules—one and two standard deviation (SD) filters—are applied in the tests (Box Table 1):

  • When the 1*SD filter is applied, 72 percent of European LCFIs that are identified as having been potentially vulnerable actually required official assistance. In this group, the vulnerability of four out of seven U.K. banks are highlighted, and three of them did eventually receive government support. The proportion for U.S. banks is similar at 67 percent, but is very different for Asia, where only one out of the eight banks (13 percent) identified by the rule actually received official support.5/

  • When the 2*SD filter is imposed to capture banks with most significant trading businesses, the predictive ability for Europe in general, and the United Kingdom in particular, remains the same but improves for both U.S. and Asian banks to 80 percent and 33 percent, respectively.

The results provide some support for the importance of trading activities in increasing the vulnerability of banks during the recent crisis. Given that the findings for Asian banks are less supportive of the hypothesis, other factors clearly play an important role as well, e.g.:

  • Economic fundamentals—Asian economies were more resilient during the crisis, possibly because they were in a different business cycle relative to Europe and the United States, which underpinned the overall health of the region’s banking systems.

  • The quality of assets and earnings—Asian banks typically had low exposure to toxic assets such as subprime mortgages, residential mortgage-backed securities (RMBS), and their derivatives, which may have helped them weather the shocks.

In theory, trading activity—when undertaken with proper risk management and monitoring—should promote efficient price discovery and enhance the breadth and depth of securities, derivatives, and structured products markets, and thus provides a valuable service. Thus, any tightening in prudential regulations against banks’ trading activities should consider the following issues: (i) potential vulnerabilities arising from trading for market-making, pure proprietary trading, hedging or rehypothecation activities; (ii) improving transparency by mandating the reporting of trading income by types of transaction; and (iii) the possibility of imposing capital charges based on business lines and the nature and riskiness of their respective activities.

Box Table 1:

Results of Filter Rules Analysis for Asian, European (and U.K.) and U.S. Banks

article image
Sources: Bloomberg; and IMF staff calculations.Note: Sample taken from 1999-2007. United States, 15 banks; Europe (including United Kingdom), 46 banks; Asia (including Australia & Japan), 18 banks; United Kingdom, 7 banks.
1/ Prepared by Julian Chow (IMF/MCM).2/ See Financial Services Authority, 2011, Prudential Risk Outlook (London, March).3/ The majority of banks do not distinguish between proprietary trading and hedging activities in trading income, which means that the reported “total trading income” figures comprise revenues from valuation of securities held in the “Trading Book”, net realized gains/losses from proprietary trading activities, and disposal of “Available-for-Sale” securities and mark-to-market valuation of derivatives for “hedging.”4/ The methodology requires the computation of means and standard deviations (SD) of trading income-to-total revenue ratios for the period between 1999-2007 with the filter rule defined as: Mean + k*SD where k=1 in and k=2.5/ The U.K. banks sample consists of HSBC Group, Barclays, RBS, LBG, Standard Chartered Bank (SCB), HBOS, Nationwide.

Systemic Risk and the U.K. Banking System 1/

The United Kingdom, as one of the world’s biggest and most open financial centers, dominates international financial market activity in many areas and is also an important home and host country to some of the biggest financial institutions in the world. The very high volume of financial transactions that take place in the United Kingdom, combined with the increasing inter-linkages across both the local and international financial systems, mean that distress affecting the U.K. businesses of any large financial institution could be transmitted both locally and overseas. Conversely, the United Kingdom’s role as a center for global finance raises the possibility that it may potentially be exposed to significant spillovers or “contagion,” from stresses originating elsewhere in the global financial system.

We estimate the market’s perception of systemic risks associated with the U.K. banking system by constructing indicators with CDS and equity data.2/ The methodology is based on a nonparametric empirical copula, which measures coincidence of dispersion (“entropy”) of a time series of default probability observations across multiple entities. It captures both linear and nonlinear dependence between financial institutions and the changes through the economic cycle, which reflect increasing dependence during periods of stress.3/ Specifically, we calculate: (i) the Joint Probability of Distress (JPoD), which is the probability that all banks in the sample experience distress simultaneously (i.e., the “tail risk”); and (ii) the Conditional Probability of Distress (CoPoD), which presents the conditional probability of distress of an institution, given that another institution becomes distressed.

Our sample consists of major U.K. banks and foreign Large Complex Financial Institutions (LCFIs). To be consistent with the BoE’s Financial Stability Report, we define the major U.K. banks peer group (“U.K. banks”) based on the provision of customer services in the United Kingdom—regardless of the country of ownership—and the LCFIs to include the world’s largest financial institutions from other European countries (“European LCFIs”) and the United States (“U.S. LCFIs”) that carry out a diverse and complex range of activities in major financial centers.4/ The market-price based results suggest that:

  • Systemic tail risks within the U.K. banking system—which we define as the market’s estimated probability that all U.K. banks would default at the same time—increased during the crisis but have receded significantly since. These risks peaked in 2009 Q1 at the height of interventions by HMT, through the Asset Protection Scheme (APS) and the recapitalization of LBG, and when funding markets beyond three months were closed to all major U.K. banks. Fears of sovereign debt crises among some EU states also increased during that period and then intensified again around mid-2010 when the crisis in Greece came to a head and Ireland experienced significant stress, but fell sharply in subsequent months (Box Figure 1). Risk perceptions improved after Greece, and then Ireland, sought loans from the EU and IMF.

  • Individual U.K. banks appear to have become more susceptible to distress spillovers from other major financial institutions since the onset of the crisis, but their exposures have moderated somewhat since 2010 Q2. On average, the CoPoD of a particular U.K. bank, given that another major financial institution in the United Kingdom, Europe, or the United States becomes distressed, has decreased in recent months but remain elevated compared to pre-crisis levels (Box Figure 2). Notwithstanding the fact that actual exposures among U.K. banks have fallen, they continue to be seen as being more exposed to one another and to European LCFIs, than to U.S. LCFIs.

  • The implied spillover threat posed by U.K. banks to foreign LCFIs rose sharply over the course of the crisis, but has fallen markedly back down to almost pre-crisis levels. Interestingly, U.S. LCFIs appear to have been more exposed to U.K. banks, on average, during the crisis, compared to their European counterparts (Box Figure 3). Since then, the average derived probability that a default in a U.K. bank would have systemic consequences for a foreign LCFI has declined to around 20 percent for both European and U.S. LCFIs, from 40 and 50 percent, respectively, at the height of the crisis.

Box Figure 1.
Box Figure 1.

Probability that All U.K. Banks in the Sample Default at the Same Time

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Bloomberg; Markit; and IMF staff calculations.
Box Figure 2.
Box Figure 2.

Probability of Default of a Particular U.K. Bank Conditional on the Default of Another U.K. Bank, a U.S. or European LCFI

(Averaged across sample)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Bloomberg; Markit; and IMF staff calculations.
Box Figure 3.
Box Figure 3.

Probability of Default of a U.K. Bank, U.S. or European LCFI Conditional on the Default of a Particular U.K. Bank

(Averaged across sample)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: Bloomberg; Markit; and IMF staff calculations.

The results point to heightened spillover risks arising from interconnectedness between the U.K. banks and European and U.S. LCFIs during periods of stress. The findings emphasize the importance of ensuring that an adequate crisis management and resolution framework is in place to deal with any problems that may arise from shocks to the domestic or international financial system; and reveal the greater exposure of the U.K. banking system to the European banks, underscoring the need to continually improve collaboration and coordination with other European countries.5/

1/ Prepared by Hiroko Oura and Ryan Scuzzarella (both IMF/MCM).2/ The analysis is based on information contained in market prices of CDS and equities, rather than any actual information on the actual business inter-linkages among banks.3/ See Segoviano, Miguel and Charles Goodhart, 2009, “Banking Stability Measures,” IMF Working Paper No. 09/04 (Washington: International Monetary Fund).4/ This U.K. banks peer group comprises Banco Santander, Bank of Ireland, Barclays, Hong Kong and Shanghai Banking Corporation (HSBC) Holdings, Lloyds Banking Group (LBG), National Australia Bank, Nationwide Building Society, Northern Rock and Royal Bank of Scotland (RBS); Co-operative Financial Services is omitted as CDS spreads are unavailable, while the FTSE100 share price index is used as proxy for Nationwide’s share price. The LCFIs comprise: BNP Paribas, Credit Suisse, Deutsche Bank, Societé Générale and UBS (all Europe); and Bank of America, Citigroup, Goldman Sachs, JPMorgan Chase & Co., Morgan Stanley (all United States).5/ It should be noted that the construction of distress dependence metrics such as JPoDs and CoPoDs do not account for the stochastic nature of actual recovery rates and the price of risk (moving from risk-neutral to real world probabilities of default). This would suggest that probabilities of distress could be over-estimated on some occasions but under-estimated at other times.

Vulnerabilities of U.K. Household and Corporate Balance Sheets and Risks to the Financial Sector 1/

The financial crisis had a significant effect on U.K. household and corporate sector balance sheets. Although the process of balance sheet repair is well underway, vulnerabilities remain in both sectors, particularly in commercial real estate, unsecured household debt, and some segments of the mortgage debt market. A key concern is that any stress in these segments—which could be triggered by a rapid increase in interest rates, a sustained rise in unemployment, a sharp decline in real estate prices, or a combination thereof—could have a significant impact on the health of the banking sector, both directly and indirectly through feedback loops between the real and financial sectors.

In the household sector, pockets of vulnerability exist in unsecured household debt and some segments of the mortgage market:

  • Household debt relative to disposable income remains high by both international and U.K. historical standards, leaving households vulnerable to income and interest rate shocks. While unsecured debt is a relatively small share of bank lending to U.K. resident sectors (5 percent), absolute losses from unsecured debt have so far surpassed losses from mortgage defaults, which have remained low relative to previous crises. Since low-income households account for just over a quarter of banks’ unsecured lending to households, banks could sustain some losses if interest rates were to normalize rapidly to or above pre-crisis levels (Box Figure 4)

  • With the ratio of house prices to income still significantly above the historical average, further sharp declines in house prices could impact household balance sheets. Although house price shocks do not directly affect household debt service capacity, the empirical evidence shows that arrears in the North and in Wales, which have experienced large house price declines and a slower housing market recovery, are higher than in other regions. Thus, a combined shock that would affect both household debt service capacity—such as an increase in unemployment—and the collateral value of secured debt could have a material impact on banks’ asset quality.

  • Risky borrowers, such as those with high loan-to-value (LTV) or buy-to-let loans, have tended to post higher arrears and represent yet another source of credit risk for banks.

Box Figure 4.
Box Figure 4.

United Kingdom: Breakdown of Bank Lending to Residents by Borrower

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: BoE; 2010 NMG Consulting survey; and IMF staff estimates.

Our sensitivity analysis of increases in interest rates and income shocks suggests that loan losses from household mortgages for the banking sector would be manageable, but that a house price shock could result in significant threat to banks’ asset quality. 2/ That said, any knock-on effect from the manifestation of these shocks on banks, while likely to be nonnegligible, would be mitigated to a large degree given that (i) most household bank debt is secured; and (ii) low-income households, which are most sensitive to interest rate and income shocks, account for a relatively small proportion of bank mortgage lending (Box Figure 5).

Box Figure 5.
Box Figure 5.

United Kingdom: Share of Mortgage Debt-at-Risk by Income Group

(In percent)

Citation: IMF Staff Country Reports 2011, 222; 10.5089/9781462336531.002.A001

Sources: 2010 NMG Consulting Survey; and IMF staff estimates.