The Banking Crisis – A Decade On

12 September 2018 Speech

Ed Sibley

Speaking at the Trinity College Dublin "Behind the Headlines" series

Podcast available

Good evening ladies and gentlemen.  I am grateful to Trinity College Dublin for the invitation to be here today at this important event, ten years on from the onset of the banking crisis in Ireland.1

“Those who cannot remember the past are condemned to repeat it"2 is a well-known, albeit ironically often forgotten, wisdom. There is a strong argument that the past was forgotten leading up to the crisis, including with the misconception that the great moderation had led to permanent reduction in business cycle fluctuations.  Given the costs and human misery the crisis has caused in Ireland, it would be inexcusable were this to happen again.

So today, in my brief remarks, I will spend a small amount of time on the causes of the crisis, before discussing the crisis itself, what has changed as a result and conclude by looking to the future. 

The causes of the crisis

10 years ago, almost exactly to today, New York Federal Reserve economists’ analysis of the US economy (prepared for the setting of interest rates), noted that “Falling real estate prices and rising defaults on home mortgages had caused stresses in financial markets and the economy….  However, the analysis concluded that these problems would be contained.”3 Just five days later, Lehman Brothers filed for bankruptcy.  The rest, as they say, is history. I cite this analysis not to criticise the Fed, but as just one example of the international failures to understand the strength of the gathering storm clouds, notwithstanding their proximity.  

Problems had started to crystallise in 2007, and accelerated in 2008, particularly post Lehmans, as market participants retreated towards safe assets. This tendency was intensified by the complexity and lack of transparency in the financial system. In other words, due to the complexity of the market, participants could not establish with confidence which risks would end up with whom and how they might be exposed to the ultimate holder of certain types of risks. Consequently, the market moved away from many higher risks. This included Irish banks with their large property exposures. The move away from the higher risks due to concerns about solvency took the form of a withdrawal of short-term liquidity, leading to the failure of the Irish banks and many others.

It is obvious now that these storm clouds did not arise from a single cause. A recent Bank of England Staff Working paper4 noted the following as contributing factors to the global financial crisis: 

  • Inadequate or flawed regulation, supervision or both;
  • Underestimation of the riskiness of securities created with financial engineering;
  • Bad incentives;
  • Excessive funding of long term assets with short-term liabilities;
  • Ratings agencies failures;
  • Flawed assumptions regarding house prices;
  • Elevated household debt;
  • A belief by bankers that their institutions were too big to fail;
  • Global imbalances;
  • Financial assets accounting; and
  • Too loose monetary policy.

It is a long list!  Across the globe, “financial innovation….vastly outpaced regulation and supervision…. Financial institutions went on a frenzy of reckless risk-taking…. [relying] more on short-term funding, dramatically lowering lending standards5.”

Many of these failures are clearly relevant to the catastrophic impacts of the financial crisis in Ireland too. They have been well covered in various reports6, including the Honohan Report7, which presented five root causes specific to the collapse of the Irish banking system:

  • Macroeconomic and budgetary policies that contributed to the economic overheating, and which relied to an unsustainable extent on the construction sector and other transient sources for Government revenue;
  • Comprehensive failure of bank management to maintain safe and sound banking practices;
  • A regulatory approach which was too deferential and accommodating, insufficiently challenging and not persistent enough;
  • An under-resourced approach to bank supervision that, by relying on governance and risk-management procedures, neglected quantitative assessment and the need to ensure there was sufficient capital to absorb the growing property-related risks; and
  • An unwillingness to take on board sufficiently the real risk of a looming problem and act with sufficient decisiveness and force to head it off in time.

In other words, a myriad of decisions both large and small, including decisions not to act (inaction bias), all underpinned by implicit and explicit assumptions, led to the global financial crisis and how catastrophic its effects were in Ireland.

There is also a strong argument that Ireland, as a small and open economy, was and is more susceptible to the economic cycle and is likely to experience more froth in the good times and is at greater risk of severe downturns than larger, less open economies. 

Results of the crisis

As is well documented, Ireland experienced a credit-fuelled property bubble, which inflated both residential and commercial prices, driving individual indebtedness, reducing overall competitiveness and encouraging an over reliance on property based taxes in the years from 2003 to 20088.  Asset bubbles are at least 380 years old. This time, in Ireland (and elsewhere) it happened to be property related and not tulips, trade in the ‘South Seas’ or companies.  

When confidence evaporated, there was no soft landing. The effects of plummeting property prices quickly spilled over to the rest of the economy, caused distress and difficulties for borrowers, hurt savers (including through the loss of value of supposedly ‘safe’ bank shares), and severely weakened the country’s fiscal position.

In response, a process of fiscal adjustment and banking sector support began in 2008. This included a guarantee of liabilities in the banking sector; a series of pro-cyclical, contractionary budgets; and capital injections into the domestic banks. By late 2010, Ireland’s fiscal position was unsustainable. The government agreed to enter an EU IMF support programme, with a total size of €85bn. The resulting adjustment programme brought additional contractionary budgets (albeit they would have had to be even more austere if Ireland had had to raise market-based funding, compared to cheaper, three-year official funding), public sector pay cuts, further banking stress tests, and more capital injections.

The associated human cost of this was immense. The effects are still being felt today by too many – such as those still directly affected by high levels of personal debt and indirectly by, for example, the dysfunction that still exists in the housing market. The emergency brake applied to housing construction in 2008, is still being felt acutely today. Post the onset of the crisis, the housing stock in Ireland flat-lined at just over two million homes9. That is an increase of just 8,800 between 2011 and 2016, compared to an increase of more than half a million in the first decade of the 2000s, and c.74,000 in 2007 alone10. The consequences of this particular effect of the lack of resilience in the system took a while to work through; non-performing loans emerged much more quickly.

10 years on from the onset of the crisis, and five years since their peak, non-performing loans (NPLs) in Ireland are still a cause of considerable distress to borrowers and vulnerabilities in the banking system. But that is not to say that there has not been considerable progress.

NPLs in the Irish retail banking system declined from €85bn in 2013 to c.€25bn by the end of 201711. Importantly, loan “cure” (the return of previously defaulted balances to performing loan status) has been the key driver of NPL reduction in the residential mortgage segment, particularly for owner occupier mortgages, where loan restructuring has played such a pivotal role.12 In contrast, liquidations, write-offs, and sales account for a large majority of the NPL reduction in the commercial real estate segment.

This is illustrated by approximately one in six (116,010)13 of all owner-occupier loans currently in existence having had some form of restructure.  87% of these loans were meeting the terms of this restructure, and 79% of them are no longer in arrears. This has been achieved through the hard work and sacrifices of those borrowers in distress that have engaged, the pressure and the requirements of the Central Bank (including that repossession can only be pursued as a last resort), and the actions of the banks, all underpinned by economic growth.

Unfortunately, significant problems remain. At the end of June 2018, c.46,000 accounts were in arrears greater than 90 days. The outstanding balance on all lenders’ owner occupier mortgage accounts in arrears of more than 90 days was €9.5 billion, equivalent to c.10 per cent of the total outstanding balance on all owner occupier mortgage accounts.  Accounts over 720 days past due constituted 42 per cent of all accounts in arrears, and at €2.5 billion, represent 91 per cent of arrears balances outstanding.

This matters because behind each account in arrears there is distress and, in the case of PDH loans, the vulnerability of borrowers at risk of losing their home.  This is why there are a significant number of protections and supports for consumers facing mortgage arrears. 

Within the remit of the Central Bank’s responsibilities, the approach to mortgage arrears resolution is focused on ensuring the fair treatment of borrowers through a strong consumer protection framework while ensuring banks are sufficiently capitalised, hold appropriately conservative provisions, and have appropriate arrears resolution strategies and operations in place.

This includes regulatory requirements such as the Consumer Protection Code and The Code of Conduct on Mortgage Arrears (CCMA), which govern how lenders interact with retail borrowers that are in distress. Firms must also follow the Mortgage Arrears Resolution Process (MARP) when dealing with borrowers facing arrears. The MARP process also requires alternative options if a restructure cannot be agreed or was not appropriate. 

There is a range of advice and support for those in arrears, including the Money Advice and Budgeting Service (MABS), the national mortgage Arrears Resolution Service (Abhaile) and a Court Mentor service to assist debtors faced with court proceedings.  The Government also introduced the Personal Insolvency Act 2012, which introduced three debt resolution processes: A Debt Relief Notice, A Debt Settlement Arrangement and a Personal Insolvency Arrangement. The Insolvency Service was established in March 2013 as an independent statutory body. The Government also introduced a national mortgage-to-rent scheme. There are also a number of other non-statutory organisations providing assistance to those in mortgage arrears. 

However, these protections, safety nets and restructuring options are not going to address the underlying problems if distressed borrowers do not engage with their lenders or the supports that are in place.  In this context, it is noteworthy that:

  • c.44% of PDH loans that are more than two years past due are more than five years past due;
  • c.40% of borrowers of PDH loans that are more two years past due are not engaging with their lenders;
  • more than half of loans that were more than four years past due at the end of 2016, were more than five years past due at the end of 2017; and
  • more than 70% of mortgages in arrears over five years are not engaging with their banks. 14 15

Recognising the individual distress these numbers represent, I would again urge anyone in arrears to engage with their lenders and/ or the supports that are available.  Nonetheless, if we, economically and societally, want a functioning mortgage market, where secured lending is priced in a different way to unsecured, then that security must mean something16 and banks, from both a commercial and a financial stability perspective must have a way of dealing with NPLs.  

Sales have proved to be controversial, particularly for owner occupier loans.  They are a legitimate and necessary approach for banks to address non-performing mortgage loans.  To date, there has been no material difference in the number of legal proceedings issued between banks and non-banks, as a percentage of total number of accounts in arrears17. It is also important to emphasise that the Central Bank has, with the support of the Oireachtas, ensured that the protections of our Codes of Conduct, including the CCMA, travel with the loans – it is the activity that is regulated, not the ownership.

The regulatory response to the crisis

Domestically, much of the decade since 2008 was spent in crisis management mode, including as part of the Troika programme.  A veritable ‘alphabet soup’ of regulations, implementing technical standards, accounting and supervisory policies have been implemented in recent years. The list of acronyms is long and confusing – CRD/ CRR, BRRD, SSMR, MiFID II, EMIR, IFRS9, GDPR, TRIM, PSD2, etc.18, but the extensive international regulatory response reflects the international nature of the crisis.

The Central Bank has also undergone significant organisational change in terms of culture, structure, and process. Following the onset of the crisis, including through the enactment of Central Bank Reform Act of 2010, we introduced significant changes to the regulatory framework and our supervisory culture and approach.

This has delivered a more assertive, risk based, outcomes-focused and analytical approach to supervision. Enforcement is now an important tool to effect deterrence, achieve compliance and promote the behaviours we expect. We are committed to the continued and necessary evolution of our regulatory framework, our supervisory approach and our ability to resolve firms as the industry and the risks it presents evolve.

From a prudential perspective, we are aiming for the Irish financial system to be resilient, to be trustworthy and to sustainably serve the needs of the economy and its customers over the long term – including through the economic cycle.  This is delivered by driving regulated financial services firms to:

  1. Have sufficient financial resources, including under a plausible but severe stress;
  2. Have sustainable business models over the long-term;
  3. Be well governed, have appropriate cultures, with effective risk management and control arrangements, and
  4. Be able to recover if they get into difficulty, and if they cannot, be resolvable in an orderly manner without significant externalities or taxpayer costs.

Where these objectives are not being met, we are taking action to remediate them.  This remediation continues to address legacy issue, drive improvements and build greater resilience into the system.

The creation of the Single Supervisory Mechanism (SSM) in 2014 is also fundamentally important. Its creation reflected the European nature of banking failures and has centralised banking supervision across the Eurozone under the authority of the European Central Bank. It has driven greater consistency and intensity of banking supervision and has started to rebuild confidence in the European banking system – albeit there remains some fragility to it.

If we look back to the ‘light touch’ regulation of yesteryear, while there was an extensive rulebook, there was a catastrophic over reliance on banks, their management and the market as whole to act in a certain way that would avoid the build-up of undue risk. Now 10 years since the onset of the crisis, rules are used to support a more assertive, intrusive, and outcome focused approach to supervision. The construction of today’s system has clearly been driven by the lessons of these last 10 years.

Regulation has been successful in improving the safety and soundness of banks and other financial services firms, governance and risk management arrangements, and reducing (but not yet removing) contingent taxpayer liabilities connected with failure. To take one important and tangible example of what this means in the case of the Irish banks, we have demanded increases in the quantity, and quality of capital that banks hold to protect against losses that could be triggered by a negative shock. Overall, the system has over 3.5 times the level of equity to risk weighted assets compared with 200619  Regulation has also driven greater transparency, disclosure and consistency in reporting, and product design and literature.

Notwithstanding the progress, legacy issues remain.  These include the remaining high level of non-performing loans that I referred to earlier, remaining vulnerabilities in some banks’ business models, and capability issues in critical areas, including in IT risk management.  Furthermore, resolution plans are not yet fully implementable, meaning that there is further work to be done to remove implicit taxpayer support for the larger banks.  As we have recently reported, banks also have work to do to improve their cultures and the levels of diversity at senior levels20.

This is why we remain focused on both addressing these legacy issues, and building resilience into the system during good times to mitigate the effects of the inevitable future downturns and shocks.  This is particularly important in the context of Ireland’s vulnerability as a small, open, economy, and the risks of which I referred to earlier.

We have also taken forward-looking macro prudential actions, which include:

  • The activation of the counter-cyclical capital buffer, which is designed to build resilience during times of growth, and so that it can be released during a downturn, avoiding the pro-cyclical withdrawal of credit supply;
  • The introduction of borrower-based measures that limit loan-to-value and loan-to-income ratios, which increase the resilience in the system and reduce the risk credit-fuelled property bubbles from over-borrowing and over-lending; and
  • Advice to government regarding the need for running budget surpluses as a precondition for the running of stabilising, counter-cyclical deficits in a future downturn. If surpluses are not built up we increase the risk of requiring pro-cyclical fiscal austerity policies.21

My colleague, Sharon Donnery, Deputy Governor, Central Banking, will speak more on these macro prudential measures in her remarks at the Dublin Economics Workshop on 14 September, so I will not discuss in detail here today.

We are also seeking to ensure that the financial services industry pays the direct costs of these necessary enhancements to regulation and supervision. Over the coming years we are aiming to increase the levy paid by regulated entities to 100% of our regulatory costs.

Looking ahead

The Irish financial system, and the banking system specifically, is undoubtedly stronger and more resilient that it was 10 years ago.  After many years of contraction, it is also growing again, becoming increasingly complex in nature and increasingly internationally focused, in part due to Brexit.  

So, there are grounds for optimism.  However, as I have touched on, weaknesses remain, and as I have also referenced, it is likely that we are closer to the next downturn than we are to the start of the last one. Even as the sun is shining today, there are clouds on the horizon. One can hope that future storms are not as severe as the last one, but storms there undoubtedly will be. 

In this context, it is noteworthy that both globally and domestically the levels of debt that characterised the build up to the crisis has not disappeared. In fact, global debt levels are higher now than they were in 2009, reaching the record peak of $164 trillion in 2016 – equivalent to 225 per cent of global GDP. Debt in advanced economies is at levels not seen since the Second World War22.  Closer to home, Ireland’s household and public debt levels, which are key indicators of the economy’s resilience to shocks remain very elevated (Irish households are the fourth most indebted in the European Union23).  This level of debt clearly increases the vulnerabilities to shocks (including interest rate risk) and limits authorities’ policy options for dealing with them in a counter cyclical way.

These macro risks are significant for Ireland, given the openness of the economy.  These could obviously affect the domestic retail banking system, notwithstanding its increased resilience. Risks also arise at a micro level.  One such risk arises from the increasing dependency on and interconnectedness of information technology, and the need to enhance resilience and mitigate cyber security risks in this area.  I have not discussed in detail today, but it is an area of concern that I will cover in more detail in a speech next month. 


Building resilience now, for individuals, governments and banks will serve us well for future downturns, whatever the cause may be.  It will also help with the pressing need to restore trust in the system, a foundation of which is that financial institutions are trustworthy.

All of the actions taken during the crisis, and the changes made in its aftermath are to support the vital, public interest mission of the Central Bank. That is to safeguard the stability of the financial services system and protect consumers.

My energies are focused on ensuring that the Central Bank has a robust, fit for purpose regulatory framework, and is delivering an effective, intrusive, analytical and outcomes-focused approach to supervision, which remembers the lessons of the past, anticipates future risks, and continuously improves. 

My hope is that in 10 years’ time, we will still remember the banking crisis and the lessons from it. But that between now and then we will have navigated the many challenges ahead and more successfully weathered the inevitable shocks that will occur, than we did 10 years ago.  This will require continued diligence, building resilience, a willingness to listen to different voices and challenge the assumptions we are relying on – it is inevitable that not all of them will hold.

With that said, I look forward to hearing the perspectives of my distinguished fellow speakers, answering your questions and hearing your views.

1 With thanks to Michael McInerney and Martin Maloney for their assistance in drafting these remarks.

2 George Santayana (1863-1952), Reason in Common Sense, The Life of Reason, Vol.1

3 Lawrence M Ball, The Fed and Lehman Brothers

5 Christine Lagarde, “Ten Years After Lehman – Lessons Learned and Challenges Ahead.”

9 See Central Statistics Office - Census of Population 2016 - Profile 1 Housing in Ireland

11 See Sharon Donnery, Trevor Fitzpatrick, Darren Greaney, Fergal McCann, and Mícheál O’Keeffe “Resolving Non-Performing Loans in Ireland: 2010-2018”in Quarterly Bulletin 2018 Q2.

12 See “Cures and Exits: An investigation of the drivers of NPL resolution in Ireland since 2012”, Fergal McCann and Niall McGeever

13 See Central Bank Mortgage Arrears Statistics

14 See Enda Keenan and Martin O’Brien (2018) “New Mortgage Lending Activity in a Comparative Context”, Central Bank of Ireland Economic Letter, Vol 2018 No 5.

15 Measured as whether a borrower has submitted a Standard Financial Statement in accordance with the Mortgage Arrears Resolution Process.

16 See Ed Sibley, The Irish Mortgage Market – 2018 and beyond for a more detailed discussion on this

17 Figures derived from analysis of legal proceeding issued for loans in arrears as at 31 December 2017 by statistics division of the Central Bank

18 The Capital Requirements Regulation (EU) No. 575/2013 (CRR) and Directive 2013/36/EU (CRD IV) came into effect on 1 January 2014. The CRR is directly applicable to all Member States while CRD IV has been transposed into Irish national law via Statutory Instruments 158/2014 [European Union (Capital Requirements) Regulations 2014] & 159/2014 [European Union (Capital Requirements) (No.2) Regulations 2014]. Further information available here.

The Bank Recovery and Resolution Directive (BRRD) provides national resolution authorities with comprehensive and effective powers for dealing with failing banks. This framework and related legislation enhances both the resilience and the resolvability of EU institutions and in-scope investment firms, which will be better prepared to deal with, and recover from, a crisis situation. Moreover, in the event that an institution does fail, the impact associated with that failure should be minimised. See here.

MiFID II contains new EU-wide rules governing investment firms, credit institutions, trading venues and market structures, as well as third-country firms providing investment services or activities in the EU. It significantly broadens the scope of MiFID by bringing new activities and firms into scope, removing or narrowing exemptions and covering new financial instruments and products. See here.

Regulation 648/2012 on OTC Derivatives, Central Counterparties and Trade Repositories (“EMIR”) implements increased transparency in respect of derivatives by imposing requirements concerning reporting of all derivative contracts (including exchange traded derivatives) to Trade Repositories (TRs); clearing those OTC derivatives subject to the mandatory clearing obligation; risk mitigation techniques for non–centrally cleared derivatives, and setting out requirements for both Central Counterparties (CCPs) and TRs. See here.

IFRS 9 is a new accounting standard that will be effective from 1 January 2018. It introduces an expected credit losses model (ECL) which may lead to higher impairment provisions and more volatile impairment charges with a consequent impact on capital ratios.

The SSM has committed to undertake a targeted review of internal models (TRIM) for banking groups. See here.

The General Data Protection Regulation (GDPR) is the first major reform of European data protection legislation in 20 years, comes into force on 25 May 2018. All companies who collect and process personal data in the EU will need to work out what data they hold on their customers, where they hold it, if they have permission to do so, whether it is stored safely, and how they can extract it in an easily “portable “ form or delete it if requested.

The Payments Services Directive 2 (PSD2) will come into force in the EU in January 2018. Banks will be obligated to share their customer information with third parties; such as payments firms. See here.

19 Source: SNL Financial Statement Data for AIB, BOI PTSB and UBID

20 See Central Bank of Ireland report on Behaviour and Culture of the Irish Retail Banks

22 Eurofi regulatory update, September 2018