Address by Matthew Elderfield, Head of Financial Regulation, at the Thomson Reuters News Event, London

06 April 2011 Speech

Ireland: Reshaping the Regulatory and Banking System

Good morning ladies and gentlemen. Thank you Jodie for your kind introduction. My thanks also to Tom Gilbert, Huw Jones and colleagues at Reuters for giving me this opportunity to talk about reshaping Ireland’s regulatory framework and its’ bank system.

I know that there is a huge amount of interest in the financial position of the Irish banks, our recent stress tests and the plans for restructuring the banking system. I propose to make that the principal focus of my remarks today by explaining the stress test results and their design in some detail. However, I would like to start by taking a bit of time to talk about the reforms that we have been implementing on the regulatory front more generally. Not only has the banking crisis resulted in the need to refinance and restructure the Irish banks, but it has also brought into sharp focus serious failures and gaps in the regulatory framework. Since I arrived to the Central Bank at the start of last year, reshaping the regulatory framework, along with the banking situation, has been my top priority, so, let me take a few moments to outline the principal changes that are taking place.

Post crisis it is common to hear regulators make a distinction between capacity to act, and will to act. There have been a number of reports on the Irish banking crisis, including one by the Governor of the Central Bank of Ireland, Patrick Honohan, and another by Max Watson and Klaus Regling, now head of the EFSF, which indicate that Ireland had weakness both in terms of capacity and will. Another official investigation into the crisis due to be published shortly, this time by Peter Nyberg, will no doubt make similar points. Making changes on both fronts is therefore central to the agenda.

Making sure we have the right resources to do supervision effectively is crucial to having the capacity to act. We are mid-way through a programme of doubling our numbers to address the very weak staffing levels at the Irish Regulator in almost all areas, but most notably front line supervisors and enforcement. Pre-crisis there was only one dedicated supervisor for AIB and Irish Nationwide combined, for example, which was clearly unacceptable. With the help of Oliver Wyman consultants, we have done a benchmarking exercise to establish our staffing numbers in line with our peer regulators taking account of the size of the markets we supervise. These recruits are a mix of experienced industry hands, special risk advisers with senior management experience and talented graduates. We have also increased our investment in training.

Capacity to act is not just about staffing numbers, but also adequate powers, and ensuring an appropriate set of rules is in place. For example, we now have new statutory fitness and probity powers across all sectors which will allow the Central Bank to vet individuals who wish to enter the financial service sector. It also provides us with stronger powers to investigate and, if appropriate, suspend or prohibit individuals who are in controlled functions. We are consulting on the detailed operation of these new powers, which are due to come into force later this year.

Incumbent staff will be grandfathered under the new regime. However, as I announced recently in Dublin, we plan to review the fitness and probity of all existing executive and nonexecutive board members at the Irish banks who have received government assistance. We will assess the incumbent directors against the new statutory Fitness and Probity Standards, including, where it is relevant, their competence and track record in the period leading up to the financial crisis. We will use our new investigative powers, where appropriate, to ensure that the people in those positions meet the required level of fitness and probity. Where they fall short of the required Standards, we will not just remove individuals, but we will also, where appropriate, issue notices to prohibit individuals from continuing as directors. I don’t underestimate the legal challenges we might face in using our new powers, but we are prepared to make difficult judgments on fitness and probity, and it is right that we should start with this group. We will be writing to all bank directors to advise them that this process will apply to anyone that plans to be in office as of 1 January 2012 to allow them an opportunity to make their plans accordingly.

The new fitness and probity powers are a welcome first step in strengthening our capacity to act as a supervisor but more is needed and, indeed, planned. The government intends to bring forward legislation to provide for a general strengthening of the legislative framework governing regulation in Ireland and discussions are well advanced as to the nature of those new powers. We will, for example, be seeking the ability to required skilled persons reports comparable to the FSMA powers available to the UK FSA and are requesting a significant increase in fines available for enforcement cases against individuals or companies. We are also seeking broad special resolution powers, similar to those introduced in the UK Banking Act 2009, which are currently awaiting legal enactment.

Improving Ireland’s capacity to act also means having the right rules in place to govern financial services activity. In many instances this will be a case of ensuring that Irish standards match prevailing international requirements, for example promptly implementing the new Basel III framework. However, Ireland’s financial crisis was not solely caused by weakness in international standards of bank capital or liquidity, although they certainly played a big part. There were home grown elements too, where it is important that the Central Bank is prepared to act with more rigorous standards than prevail internationally.

We have taken a conscious decision to do this in the area of corporate governance. There are some notable examples of Irish financial companies who have suffered from over dominant CEOs, or ineffective challenge in the Board room. Poor governance at the Irish banks was exacerbated by the concentrated nature of corporate life in Ireland, with challenge and awkwardness in the board room perhaps blunted by the social constraints of working and living in a small business community in a small country.

These changes speak to the capacity of the Central Bank to act as a supervisor in a reformed regulatory system, but it is also important to have the will to act. Our objective is to development a framework of assertive risk-based supervision underpinned by a credible threat of enforcement action. In the past year we have set a record fine for a domestic financial institution and have also shown a willingness to take decisive supervisory action regarding serious and persistent solvency breaches at one of our largest domestic insurance companies. We are developing a risk assessment framework that will allow a more systematic review of the firms that we supervise based on their impact ranking and which will require mitigation plans for areas where we identify risks. As supervisors, we should be prepared to discuss the adequacy of management’s proposed mitigation actions but, where we are not satisfied, be prepared to take an assertive approach, and insist on action.

These changes will help build a better regulatory framework for the future but the immediate challenge is to implement a plan to rebuild the banking system. This plan involves a number of interconnected exercises under the Financial Measures Programme agreed with the IMF, ECB and European Commission, the results of which were announced at the end of last week. These are:

  • First, an independent loan loss forecasting exercise, carried out by consultants BlackRock Solutions the results of which have helped to establish the capital
  • requirements of the banks under the Prudential Capital Assessment Review (PCAR).
    Second, the PCAR itself, which is an annual stress test of the capital resources of the domestic banks within a given stress scenario to calculate the capital required to meet Central Bank imposed requirements.
  • Third, a liquidity review – the Prudential Liquidity Assessment Review (PLAR) – to establish quantitative and qualitative targets for banks participating in the PCAR. These targets are aimed at reducing the banks’ reliance on short-term funding and ensuring convergence to Basel III liquidity standards over time, and,
  • Fourth, a review of the structure of the domestic banking system and the viability of individual banks. Undertaken together with the NTMA and Department of Finance, this review looked at the conditions for the maintenance of a sustainable domestic banking system. It is the basis for decisions on both the composition of this system and the structure of its constituent institutions.

Let me take the first of these two exercises together and explain the approach we have adopted to loan loss forecasts and the capital stress test. We know that these exercises are taking place in a fundamentally different context to a year ago. We now face a severe crisis in the euro sovereign debt markets, a weaker Irish economy subject to further fiscal consolidation, a formal EU-IMF programme for Ireland, and, above all, a lack of market confidence in the Irish banking system, reflected in the stressed wholesale funding position of the banks, and, concerns about future loan losses under adverse scenarios.

Given this context, the Central Bank took a more conservative approach in the latest exercises, and built in more transparency and stronger external validation into the exercises.

I want to talk about conservatism in a fair bit of detail in a moment but let me briefly mention validation and transparency.

In terms of validation, the stress tests were informed not only by the banks’ own views, but also a fully independent assessment of portfolio loan losses conducted by Black Rock Solutions. The data inputs for this exercise were themselves subject to an assurance process by a number of accounting and legal firms acting as BlackRock subcontractors to make sure that the information was in fact a reliable basis for the calculation. As I will explain, the BlackRock assessment itself was in fact determinative for the capital calculation for the stress test. In addition to this, we used Boston Consulting to provide quality assurance over the capital calculation and loan loss methodology. And, as you would imagine our process and detailed findings were also subject to close scrutiny by large teams from the IMF, ECB and European Commission. By any measure, then, this has been a very thorough process with very extensive external validation.

In terms of transparency, I would encourage you to take an opportunity later today to go on-line and download our Financial Measures Programme report. This 80 page report provides loan loss information from BlackRock on a life time and three year basis under a base and stress scenario, as well as base and stress calculations by the banks, broken down by residential mortgage, corporate, commercial real estate, consumer and other portfolios, all set out for each bank individually. We also provide details of the deleveraging plans of each of the banks, including the amount of assets required for sale to meet target loan to deposit ratios. We provide considerable detail as to the macroeconomic assumptions, including detailed house price assumptions, underpinning the stress test. And, of course, we provide detail as to the capital calculation.

The goal of this transparency is straightforward: to address market scepticism of the banks’ financial position by being completely open about the results of this process so that market participants can make an independent judgement about the conclusions of our exercise.

Most importantly of all we have approached this exercise from a perspective of conservatism.

This conservatism has a number of elements. Of course we are starting with an adverse scenario to begin with. For example, the stress scenario assumes a 60% peak-to-trough drop in Irish residential house prices and no recovery at all in these prices for a decade. The macroeconomic stress scenario was developed principally by the ECB and is on a comparable basis to that which will inform other European stress tests.

However, given the parlous nature of the Irish banks and the level of market scepticism regarding their financial strength, we recognised that further conservatism was required. I would highlight four elements:

  • The target stress core tier 1 capital requirement is set at 6%, rather than the 4% we used last year or the 5% that is contemplated for other exercises;
  • The stress period was three years, rather than the two year norm;
  • We relied on the BlackRock loan loss estimates as being determinative of the capital calculation; and
  • We applied an additional capital buffer on top of the results of the three year stress test.

The application of an additional capital buffer of €5.3bn in aggregate for the system is principally designed to take account of possible losses beyond our three year assessment. I don’t know of any supervisor who would propose to capitalise their banks now for remote hypothetical loan losses over the full lifetime of the assets. Such losses are inherently uncertain due to their long term nature and only emerge over decades. In addition offsetting income would potentially arise and offset losses. However, we decided to publish both stress and base lifetime loan loss projections by BlackRock for all the banks and all the portfolios assessed as part our transparent approach. And we also decided to apply the capital buffer to take account of potential losses in the immediate period after the three year assessment horizon, without any recognition of offsetting income.

The use of BlackRock as being determinative in the setting of loan loss estimates, rather than the banks, was another important element of conservatism: the difference between the two estimates resulted in about a third of the overall additional capital required by our exercise. That is not to criticise the banks approach, which were an entirely plausible methodology for considering losses in their portfolios. The BlackRock methodology took what I might describe as a more hard-nosed capital markets approach to assessing the loan losses, focusing on the intrinsic value of the loans in light of the value of collateral and putting aside the possible ameliorative impact of forbearance techniques.

The banks’ residential mortgage portfolios are a good case in point. Understandably, the banks own internal provisioning models, generally speaking, place considerable weight on the rate of unemployment in the economy and see greater potential for rescheduled loans to move out of arrears as employment improves over time. In contrast, the BlackRock model places greater weight on loan-to-value in assessing likely delinquency levels and takes a more US-view of the likely roll rate from delinquency into foreclosure. This would, based on the stress assumptions about flat house prices for a decade, hypothetically lead to unheard of levels of repossession in Ireland. We recognise that is a very unlikely scenario, but we have in fact required the banks to be capitalised to that level to address market scepticism about their financial position.

As you well know, loan losses and credit quality is only half the story in the resuscitation of the Irish banking system. The other half of the story is the funding position of the banks which in turn is linked to their structure.

During the period from 2003-2007 the assets of the domestic Irish banks rose dramatically on the back of a booming property and construction sector. This asset growth was funded in large part by short term wholesale and interbank borrowing. As a result, the banks experienced rapid balance sheet growth. However, between September-December 2010, the 6 Irish banks lost over €100 billion in funding as debt instruments failed to roll over and deposits were withdrawn. This outflow came mainly from the retreat of wholesale providers of funds from outside the euro area. These counterparts had placed funds with the Irish banks when the Sovereign and the principal banks had high credit ratings. This loss of confidence, punctuated by downgrades of the Sovereign, and of the banks, reflected increased market uncertainty.

The PLAR was specifically designed to rebalance the funding mix of the bank’s Balance Sheets through a combination of agreed deleveraging transactions or asset disposals, and a set of liquidity metrics to be achieved by December 2013. The four institutions, AIB, BOI, EBS and ILP, must execute the PLAR plans agreed with the Central Bank.

Three key target funding ratios have been set under the PLAR:

  • A loan to deposit ratio;
  • A Net Stable Funding Ratio (“NSFR”); and
  • A Liquidity Coverage Ratio (“LCR”).

We have set a target Loan to deposit ratio of 122.5% that each of the each of the four banks must meet by December 2013. Each bank must identify non-core assets and establish plans to deleverage their balance sheets by disposing of these assets by 2013. The purpose of the 122.5% target is to begin the Irish banks’ return to a more appropriately leveraged and more stable funding position. The main driver of this ratio for each bank will be sales of asset. In this respect the PLAR and PCAR process interact and we have required the banks to set aside sufficient capital to fund their deleveraging plans.

The Net Stable Funding Ratio targets are aimed at ensuring that banks create more medium and long term funding of their activities and encouraging them over time to build more stable deposit bases, issue longer term wholesale funding and reduce their Central Bank borrowing. Each bank has been set an individual target for end 2013 and interim targets for end 2011 and end 2012. These are to ensure that convergence to Basel III standards can be readily met.

The Liquidity Coverage Ratio target is aimed at ensuring that banks have sufficient high-quality liquid assets to survive a significant, one-month liquidity stress scenario. Setting this target will encourage banks to strengthen their liquidity buffer once their reliance on central bank borrowing has been reduced to a sustainable level. Each bank has been set a target level for LCR for end 2013 that is designed to ensure that convergence to Basel III standards can be readily met.

We will monitor progress towards achieving all these three key targets. Banks will have to submit a detailed point-in-time liquidity profile every quarter beginning at end-Q2 2011. Funding ratios will be calculated and compared against the interim targets established as part of PLAR 2011. In addition, the Central Bank and other public authorities will interact more regularly with banks to monitor progress towards the disposal of non-core assets. Further full Prudential Liquidity Assessment Reviews will be carried out in 2012 and 2013 in accordance with the terms of the MOU. Plans and targets may be updated as a consequence of these reviews.

Deleveraging at individual banks is therefore a critical part of restructuring the Irish banking system. This Deleveraging Review started with the submission by the banks of formal deleveraging plans for assessment by the Central Bank and its advisors. This has resulted in the proposed sub-division of assets on a core and non-core basis, with the latter cohort representing the assets to be deleveraged over time, a significant portion of which should be achieved by end 2013. The core components of the banks will focus on supporting the Irish economy. This core/non-core approach is similar to that adopted by RBS and Citigroup.

As noted above, we have ensured that capital is set aside to support the deleveraging process. But there is common cause between the Central bank, Government and the banks that deleveraging should not come at any price and that fire sales should not be adopted for the deleveraging process. We therefore plan to monitor the process closely and take a pragmatic approach as the banks sequence their non-core disposals to take account of prevailing market conditions.

We recognise that in the meantime the banks will continue to be reliant on Central Bank and Eurosystem support for their funding. The Eurosystem has already provided considerable liquidity support to the Irish banking system, which is very welcome. The ECB noted in its press release welcoming the stress test results that the stress tests provide the banks with a sound capital basis and that, against this background, the Eurosystem will continue to provide liquidity support to banks in Ireland.

This progress on deleveraging comes in the context of a broader government strategy on restructuring the banking system. The strategy is the creation of two pillar banks to support the economy going forward, comprising Bank of Ireland and a merged AIB and EBS. The fourth bank in the stress test exercise, ILP, will also require government assistance and will hive off its profitable insurance arm as it slims down and rebuilds its capital position. The government has taken a clear decision that action will not be taken against the senior bond holders of these four institutions, as they represent the basis of the restructured banking system for the future. There has been considerable debate about this point but the markets now have certainty about government policy.

What of the implications of all this for the Irish sovereign finances? I fear that is a topic for another speech in its own right. Let me just make two points.

First, that the €24bn total from the stress test exercise is the gross cost to the government and will be offset by asset sales and liability management exercises involving subordinated bond holders. Also, €3bn of this total is in the form of contingent capital instruments which will be able to be redeemed by the state after a period of time if the stress case does not materialise.

Second, even at a gross cost of €24bn this is still within the envelope of €35bn made available under the IMF/EU Programme for the banks, which was assessed by the lenders to be a sustainable level of additional borrowing.

Let me conclude by acknowledging that it will take some time before a reformed regulatory system and a restructured banking system are in place in Ireland. The financial crisis was rapid in its damage to the banking system and its exposure of the weakness of the regulatory framework. Addressing these challenges and rebuilding market confidence in the banks is going to take considerable time. But the process of improving regulation in Ireland is well underway. And the conservative, transparent and externally validated stress tests concluded last week are an important milestone in the process of rebuilding a well capitalised and right-sized Irish banking system.

Thank you for your attention.