Patrick Honohan: Major banking decisions in Ireland during the financial crisis

Introductory statement by Mr Patrick Honohan, Governor of the Central Bank of Ireland, to the Oireachtas (National Parliament) Banking Inquiry, Dublin, 25 June 2015.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
21 July 2015

"Nexus phase"

I have judged it best to concentrate in this statement mainly on crisis management aspects. These were the chief focus of the first years of my term of office from late September 2009 to the end of 2013. Besides, my previous appearances before the Inquiry, and the separate documentation submitted by the Central Bank have already covered much of the remaining ground.1

Crisis management aspects

1. Appropriateness of bank guarantee decision

Much of the discussion on the appropriateness of the bank guarantee decision has focused on what information was available to the political decision makers on the night of September 29th, 2008, on what should have been understood by their advisers, given the information that was available, and on the alternative short-term courses of action which were available and which might have achieved smaller losses for Irish taxpayers, whether that was to involve a sharing with other EU governments or a bail-in of some creditors.2

After all the ink that has been spilled over this issue, nothing has altered the conclusion of my May 2010 report which suggested some time should have been bought through emergency liquidity assistance (ELA) in order to allow for consultation with the ECB and other EU Governments, and to obtain a clearer picture of the potential exposure involved in blanket guarantee. Probably some extensive form of guarantee would still have to have been granted, but it should not have included subordinated debt nor existing senior term debt. Given the evident failure of their business models - as was clearly indicated in the documents prepare by the investment bank consultants, Anglo and INBS should probably have been dealt with separately from the others through immediate nationalization and not included in a general guarantee. To what extent additional bail-in of creditors could have been fully effected over and above what was subsequently achieved is hard to say. As there was later, there would have been strong pressure from EU partners not to bail-in senior debt, and it is unclear whether any quid-pro-quo could have been secured in return.

All in all, the focus on the night of the guarantee has been excessive. As explained below, much of the losses and the costs for the Irish people had already been ineradicably incurred.

2. Decision to recapitalize banks and alternatives

In what follows I focus on the qualitative and policy aspects of this issue as it was dealt with during 2010.3

Given the existence of a watertight guarantee enacted through the CIFS legislation of October 2008,4 in designing policy with regard to bank capitalization I was guided by a fixed principle, informed by the Central Bank's statutory mandate, that any event of default by a bank (as would trigger a large cash call on the State) was to be avoided in view of the financial instability that this would cause. In practice, this implied that the Central Bank had to ensure that all of the guaranteed banks would continue to maintain regulatory capital compliant with international standards. This would help ensure that they had access to the ECB's standard liquidity facilities, as well as, where necessary, ELA.5

The second principle applied by the Central Bank was that, for the purpose of defining eligibility for access to Central Bank financing, regulatory capital adequacy would continue to be measured in line with conventional accounting principles.6 Anglo was the first bank to encounter problems in meeting regulatory capital minima; their management advised in December 2009 that they were likely to breach regulatory minima at end-year following a reassessment of provisioning needs. This information triggered a request to the Minister for Finance for an injection of capital sufficient to maintain capital at the regulatory minimum. The Minister undertook to do so and this led to the creation of a Promissory Note (PN) which was injected into Anglo in March 2010. The amount of the PN was increased several times during 2010 when additional provisioning needs were identified.

While the injection of State capital was essential if the eligibility of Anglo for Central Bank financing was to be retained, it was evident that this might complicate subsequent efforts to impose burden-sharing on Anglo's creditors after the ending of the initial guarantee.

NAMA first tranche and the March 2010 capital adequacy review (PCAR)

When they became available in late March 2010, the initial NAMA valuations (based only on the handful of exposures that had been fully valued in time) implied much higher losses than had been generally expected by industry specialists.

The Central Bank had decided to take the announcement of the valuations of the first tranche of NAMA loan purchases as the occasion to the necessary capital infusions for Allied (AIB), Bank of Ireland and EBS - the going concern guaranteed banks selling loans into NAMA.7 The requirements announced by the Bank at the end of March 2010 (in what was known as PCAR2010) assumed that all of the NAMA purchases would involve haircuts as large as thefirst tranche, and also took account of forward-looking loss estimates in a base case and a stress scenario of the non-NAMA books. For example, a flat 5% loan-loss estimate - much higher than that assumed by the banks' managements themselves - was applied to the residential mortgage portfolios. These banks were also required to reach capital adequacy standards above the existing international regulatory minima - thereby reflecting trends in international regulatory thinking. The banks were given a 9-month period in which to raise the additional capital needed. Bank of Ireland proceeded immediately to raise capital from the market; Allied (AIB) started to dispose of some of its noncore assets in order to reduce the additional cash capital needed.

Although PCAR2010 increased projected capital requirements, there was, given the prevailing uncertainty surrounding these badly-impaired portfolios, still a large margin of error which might eventually consume the apparent surplus above international minima. Early hopes in the Central Bank that comfortable over-capitalization could be safely engineered were soon replaced by a concern that there was insufficient information to back what risked being a destabilizingly large need of State capital (when taken into account with the steady deterioration of the remaining fiscal prospects). The situation was complicated by the extended interval over which partial information about the NAMA haircuts would unfold. How could bank solvency be assured throughout this period of uncertainty? The "choreography" of this interval necessarily involved a risky choice between two options.

The first option was to require banks to build an unquestionably ample capital buffer; the problem with this option was the very large potential margin of error in estimates of the NAMA haircuts. To announce a sum that the Government would have to inject so that the banks would have sufficient capital to cover all eventualities would have triggered an immediate loss of market confidence in the Government's finances (putting paid, for example, to the chance of the Government being able to avoid an IMF programme).

An alternative, less exigent option was therefore chosen. This required additional capital sufficient to cover NAMA haircuts on the assumption that the first tranche was representative, but without a further buffer in case later tranches were worse than the first. These haircuts were much more severe than that generally mandated for the European CEBS stress test later that summer, but they proved insufficient when information about the second and third NAMA tranches became available in August and September. Additional capital requirements had to be announced in September 2010 for Allied (AIB) and also for Anglo, because of Anglo management's estimate of additional costs imposed by the split and wind- down plan that the Government had already announced earlier in the same month.

The September 2010 bank funding cliff

Faced with the steep cliff of bank liability maturities in September, I gave some consideration to possible alternative courses of action which might be recommended to Government.8 The only safe way forward seemed to continue to rely on ELA while also pursuing all possible steps to rebuild confidence-and recognizing that recourse to an IMF Programme would be the fall-back position.

It is important to note that the increase in ELA during September 2010 and through to the end of the year did not create a new contingent liability for the State, but merely shifted that liability from one set of creditors to another. Mar a deir an seanfhocal, "Cibé cé ólfhas 's é Domhnall a íocfhas".

3. Programme of assistance

International assistance mooted

By September 2010 most informed international observers began to factor in a likely need for Ireland to enter a programme of official financial assistance. The prospective rapid increase in ELA (due to the funding cliff at the end of the initial guarantee period) and the growing evidence that the Government's multi-year fiscal plan would not stabilize the debt dynamics were the main influences here. The drip-feed of news about bank recapitalization needs did not help.

Realizing that ELA that was now likely to increase considerably over the following couple of months, ECB officials increasingly looked to official borrowing by the Government as a way of getting the ELA repaid promptly. Given the additional interest burden that borrowing from the IMF or EFSF would impose, the Government was not interested in following this line of approach to dealing with the ELA.

However, the growing yield spreads on Irish Government debt made recourse to official borrowing essential to fund the rest of the Government deficit, which was also veering wider than had been projected in the Government's 2009 plan - a plan whose spending and taxation policies had been adhered to, but which had seen a wider than projected deficit largely because of the weaker than projected external demand position. The fact that there were cash balances sufficient to cover the deficit for several months did not alter this perspective.

The move to the EU-IMF Programme

By the time that spreads on 10-year bonds exceeded 500 basis points on November 4, with accelerating depositor outflows also reflecting a loss of confidence, it was clear to me that application for a programme could no longer safely be deferred. The Minister agreed to exploratory discussions with Troika officials in Brussels.

Delayed until November 14 (apparently because of the need for the Troika to agree a strategy among themselves), these discussions, attended by about twenty Irish officials, confirmed that there was a basis for full negotiations on a programme which, with the approval of the Minister, began in Dublin on November 18. These meetings took place in an increasingly tense situation of national and international media speculation, market tensions (which had brought the yield on Irish Government 10-year bonds above 8 per cent)9 and considerable international official alarm at the highest levels.

In an attempt to restore depositor and public confidence, I was anxious to make a credible public statement especially as depositor outflows were accelerating.10 Indeed, the previous day had seen an outflow of €900 million of retail deposits - by far the largest daily figure before or since. The start of negotiations provided a basis for convincing reassurance, which was clearly needed, and I gave a radio interview to that effect on the morning of November 18.

Meanwhile ECB officials made public and private overtures to the Irish Authorities to encourage application for a programme. Given the views of Irish officials, already preparing the ground for a programme, these overtures were behind the curve and as such can be considered at best as having been largely gratuitous. This includes the letter of the ECB President to the Minister for Finance on November 19.

The Programme structure

The Programme finally agreed with the Troika largely reflected the 4-year budgetary plan published by the Government at the outset of the negotiations. Indeed, the target date for reaching the threshold deficit level of 3 per cent of GDP was pushed back in the negotiations to 2015. But would the implied degree of "fiscal austerity" be sufficient to bring the public finances back on to a sustainable path? Here the remaining question-mark over the banking system was key.11 (11)

The Programme included the requirement for a new and much more comprehensive stress test of the banks, capitalization of the banks up to an even higher standard than before, and a programme of rapid deleveraging of the banks.

Provision was made in the Programme's financial envelope for up to €35 billion in cash to be borrowed by the Government and injected into the banks if it proved necessary following the stress test. The Central Bank shared the view of other Irish officials that, if €35 billion were really needed for that purpose, debt sustainability of the Irish government would be likely unattainable.12

Indeed, the Central Bank found the financial terms of the Programme unsatisfactory especially in regard to the high interest rates and the lack of an insurance mechanism against tail risks in the banks.13 Nevertheless, having set out these concerns, I advised the Government in writing that it should proceed on the basis of the Programme, given that the alternative of struggling forward without access to market finance would have been more economically damaging. If the Programme proved unable to deliver a sustainable debt path (as seemed possible, even likely14) then it could be renegotiated. The spending and tax adjustments that would have been needed if Programme funding was not available would be much more severe. Mar a deir an seanfhocal "Ní hé lá na gaoithe lá na scolb."

Better financing conditions help make the Programme a success

Fortunately, there were subsequent considerable improvements in the terms of financing, not least through the reduction in EU interest rates (from an initial level of about 5.8 per cent per annum) and lengthening of the loan maturities, and subsequently through the financial arrangements around the liquidation of IBRC.

These improvements, combined with close adherence to the budgetary targets in the following three years, allowed the Government to take the decision to exit the Programme on schedule at the end of 2013.

Despite the risks and the pressures involved at the outset, recourse to the EU-IMF loan proved to be a successful policy move for Ireland, limiting the need for fiscal austerity (because funding was made available in amounts and at costs far below what the market would have offered), and underpinning a relatively rapid return to confidence. Successive Governments' adherence to the Programme, in difficult political circumstances, helped restore Ireland's international reputation for disciplined macroeconomic management, a reputation essential for sustained recovery of employment and incomes.15

4. Promissory Notes

The origin of the PNs (for Anglo Irish Bank, INBS and EBS) has been explained in section 2 above. The PNs were created and provided to these banks in order to ensure that they complied with regulatory capital requirements. As liquidity drained from the system in the last months of 2010, these PNs became part of the collateral pledged by Anglo to the Central Bank of Ireland in return for the growing amounts of ELA that were being provided.

Both before and after entry into the EU-IMF Programme, the question of finding a long-term solution for the huge stock of ELA was increasingly pressing. ECB rules specify that ELA should be short-term; but there was no obvious source of an early repayment. A cash call on the Government by the Central Bank would have clearly been highly destabilizing. Besides, the Government's contingent liability here had been assumed (as a result of the guarantee of September 2008) in the interest of financial stability and in part on the recommendation of the Central Bank. It would be strange indeed for the Central Bank to then turn, in a financially destabilising manner, on the State, which had only stepped into the situation in order to stem a financial stability problem traceable to imprudent bank lending.

Squaring this circle was very challenging, especially considering the Treaty prohibition of monetary lending to the Government (TFEU Article 123). Only if the ELA could be extinguished and replaced by some long-term financing facility could stability be protected. A variety of financial engineering approaches was suggested, but in the end, only one seemed both compliant with law and acceptable to all involved. It required a liquidation of IBRC under special legislation.16, 17

After the liquidation, the Central Bank was in possession of a PN in the amount of €25 billion (as well as about €3.5bn of Government bonds and almost €14bn of NAMA bonds, most of the latter obtained in return for the Central Bank's claim on the remaining assets of IBRC). These large acquisitions meant that the Central Bank's net financial asset holdings were well above the normal agreed ceilings.

The PN, being not freely transferable or marketable, was not an ideal instrument for the Central Bank to hold, especially considering that it would need to be disposed of. Accordingly the Central Bank had arranged, as part of the liquidation plan, to exchange the PN for marketable bonds. The Government agreed to this on the basis that the bonds to be exchanged would be very long-term and carry floating rate interest. Thus all parties were satisfied with the exchange: an end to ELA, acceptable assets on the Central Bank's books and a satisfactory improvement in the terms and conditions of Irish Government debt.

The ECB continues to monitor the disposal of the bond portfolio with a view to ensuring compliance with Article 123. These disposals are being made as quickly as possible subject to financial stability considerations, and have been running well ahead of the minimum schedule announced in February 2013.

5. Cost of crisis

The question: "What was the cost of the crisis?" is far from unambiguous. Cost of what aspects? Cost to whom? Compared to what? Direct costs only or also including the indirect costs related to the wider adjustment of the economy and of the cost of borrowing? Including all gross outlays, or net of subsequent recoveries? While my earlier appearances before the Inquiry have touched on some of these questions, it is worth adding some further discussion.

  • Net cost of Government assumption of banking liabilities. With so many non- performing and other questionable bank loans still awaiting cure or collection, with a block of NAMA assets still awaiting sale, and with a large part of the ownership of the banking system still in the hands of the Government, it is too soon even to have a definitive estimate of even some of the most straightforward versions of the question, namely those related to the fiscal costs of making the creditors of the banks whole. In January I provided the Inquiry with an estimate of about €40 bn or a little less for this dimension of the costs.18 To arrive at that figure one can take the gross fiscal outlay, of about €64 bn, subtract the guarantee fees paid by the banks, and the cash recoveries through sale of capital instruments, and the NTMA's estimate of the value of the remaining Government investments in the banks. An adjustment - in practice small - should also be made for the time value of money, in order to bring each outlay and receipt to a common date. This figure will change from time to time as the investments are revalued or realized. Any profit of NAMA, or from the IBRC liquidation will also serve to lower the estimate.19 It is noteworthy that the net figure is about the same as the aggregate haircut on the NAMA purchases of property- related loans: had there been no losses on those, the net fiscal cost of the assumption of banking liabilities would have been small. In other words, the main banks had enough pre-crisis capital (€28 bn between the top two) to absorb the actual and prospective losses on residential mortgages.
  • Loan losses of the banks. This is another potential cost figure. Banks normally incur only very small loan losses. In this crisis the losses of Irish banks and foreign banks operating in Ireland greatly exceeded the figure noted under (a) insofar as they burned through all of their equity capital (thereby imposing sizable losses on private shareholders) and much of their subordinated debt. As I have already mentioned in public several times, the overall figure here is well in excess of €100bn. (To be sure, the banks had made unsustainable profits in the boom years).
  • Overall economic cost to Ireland of the banking boom and bust. Of course (a) and (b) are only partial views of the overall costs. The boom brought some transitory prosperity even to those not directly involved in construction or property; the bust threw many out of work and imposed losses on many Irish people other than those that came through the losses on bank loans. And of course a sizable part of the economic downturn 2008-12 was not home-grown but attributable to the global financial crisis. In order to get an overall cost to Ireland of the Ireland-specific banking boom and bust we would need to model how the economy might have performed in the presence of prudent pre-crisis banking practice. A precise quantification of this is impossible, but it is clear that the net cost is well in excess of €100 billion and still growing (cf. Honohan, 2015b).
  • Cost of the guarantee decision relative to the best alternative available on the night. Finally we come to the least important and yet most over-analyzed question in this area, namely what was the net economic cost to Ireland imposed by the guarantee decision, relative to the best alternative available on the night. What if an alternative path had been tried, following prior consultation with EU partners during an interval covered by ELA, involving a more limited guarantee, excluding some or all existing senior bonds and subordinated debt, and in conjunction with an attempt to get agreement that Anglo and INBS bondholders would be bailed-out only on the basis of cost-sharing with EU partners? How much (in net national economic terms) might have thereby been saved. This is not a question susceptible of any precise quantification: could a cost-sharing deal have been successfully negotiated? And if not, what then? All in all, a possible net economic saving in the area of €2-€10bn could be imagined, but surely no more than that.20 If so, by September 2008, well over 90 per cent of the net economic cost to Ireland of the boom and bust had become unavoidable.

The other areas of the Inquiry's focus

As mentioned, this Statement does not dwell on the other areas highlighted to me by the Inquiry, namely regulatory, supervisory and Government, and banking aspects.

The very extensive changes in regulatory and supervisory staffing, methodology and culture 2009-2013 brought into the Central Bank have been documented in papers submitted by the Central Bank to the Inquiry. I believe that the smooth transition of the prudential aspects of Ireland's supervision of banks to the ECB's Single Supervisory Mechanism in November 2014 testifies to the credibility of the transformation that was achieved in those years.

On the rehabilitation and restructuring of banking in Ireland - aside from the recapitalization aspects - in the period 2009-13, I have spoken repeatedly to Oireachtas committees about the disappointingly slow progress in dealing with non-performing loans, especially mortgage arrears. The banks have now got reasonably effective systems in place to deal with collection and restructuring on the scale needed, but they did not introduce adequate systems until chivvied by the Central Bank from late 2011. The mortgage arrears targets introduced in early 2013 simplified a quantitative assessment of progress being made by the banks. An increasing number of loan restructurings, designed to be fully sustainable, have been agreed. Compliance with the restructured payments schedule is high, though deeper restructuring would arguably increase assurance that the borrowers would not fall again into arrears and would reduce debt overhang in the economy at large. However, despite all of the efforts, including the initiation of court repossession proceedings, there is still insufficient engagement between borrower and lender in a large number of cases. The resulting lengthy delays leave the borrowers in a limbo of over-indebtedness and hamper both the borrowers' and the banks' ability to put the crisis behind them.


Honohan, Patrick. 2012. "Recapitalisation of Failed Banks: Some Lessons from the Irish Experience". Lecture given to the 44th Annual Money, Macro and Finance Conference, Trinity College, Dublin, 7 September 2012.

Honohan, Patrick. 2014a. "Ireland's EU-IMF Programme: Delivering what it said on the tin". Address to the European Commission DG-ECFIN Conference (in collaboration with Trinity College Dublin) Future Directions for the Irish Economy. 10 January 2014.

Honohan, Patrick. 2014b. "Brian Lenihan and the Nation's Finances" in Brian Murphy, Mary O'Rourke and Noel Whelan, eds., Brian Lenihan: His Times and Trials (Dublin: Irish Academic Press), pp. 64-82.

Honohan, Patrick. 2015a. "Some Lessons Learnt from the EU-IMF Programme", Speech at the IMF-CEPR-Central Bank of Ireland Conference Ireland-Lessons from Its Recovery from the Bank-Sovereign Loop, Dublin Castle, 19 January 2015.

Honohan, Patrick. 2015b. "Where did all the money channelled into property-backed lending go? Irish Times, 31 January 2015.

Honohan, Patrick. 2015c. "Reconciling Diverse Interests in an International Stabilisation Programme: Some Thoughts about the EU-IMF Programme for Ireland." Speech at Institute for New Economic Thinking (INET) Annual Conference, Paris, 9 April 2015.

Schadler, Susan. 2013. "Unsustainable debt and the political economy of lending: Constraining the IMF's role in Sovereign Debt Crises." CIGI Papers 19, Center for International Governance Innovation, Washington DC.

1 Taken together with the present statement, those previous pieces of written evidence cover all 31 lines of Inquiry proposed in the Direction.

2 These matters were discussed at my previous appearances before the Inquiry on 15 January and 11 March 2015. My letter to the Inquiry of 12 February 2015 also refers.

3 Further details and quantification are included in Honohan (2012).

4 In order to allow the banks to continue to fund themselves as deposits and bonds matured during the two-year period of the initial guarantee, a further more limited guarantee (known as ELG) on new deposits and bonds was provided by the Government from late 2009.

5 Having run out of eligible assets for normal ECB liquidity operations, Anglo requested ELA in February 2009 shortly after its nationalization. The amount of ELA it drew fluctuated up to about €15 billion between then and late September 2010.

6 Of course there was the known potential of a large haircut on the purchase of property-related loans by NAMA, but the prospective losses might not fully affect accounting measures of capital until the losses were incurred. This approach sufficiently balanced the need to ensure adequate capital for prudential purposes with the need to avoid needless and damaging acceleration of the crystallization of the costs of the Government guarantee. It also took account of the existence of the Government guarantee as a form of quasi-capital support.

7 A capital adequacy assessment of Anglo and INBS was deferred. PTSB too was not included in that special capital adequacy exercise as it was not affected by NAMA purchases.

8 One such possibility was for the Government to abrogate the bank guarantee (in respect of Anglo and INBS) in August or September 2010 and to bail-in the bond holders and potentially large depositors. While, if successful, such a measure would have reduced the amount paid out to bank creditors by the State, it would also have represented a very large and conspicuous default of the State on a class of creditors which it had created by virtue of the guarantee and confirmed by the CIFS legislation of October 2008. It seemed doubtful whether it could even be successfully accomplished: for example, pressure from the ECB might well have resulted in any such decision having to be reversed in a way that resulted in the State resuming its obligations in regard to these liabilities while still suffering the loss of market credibility implied by a default. The inevitable funding difficulties which would ensue, both for the banks and for the sovereign, would have been regarded by European partners as having been wholly self-imposed. For such reasons, this course of action (never contemplated to my knowledge elsewhere in the Government services) was not recommended by the Central Bank.

9 Implying a significant probability of sovereign default perceived by market participants at that time.

10 The Minister for Finance shared this concern and, in a phone call earlier that month, had urged me to make a statement of reassurance to depositors; lacking any convincing basis for such reassurance, I demurred at that time. Indeed, as late as November 17, senior ECB officials rebuffed my suggestion that I should publicly declare that the ECB would continue to support the Irish banks.

11 The discussions that took place in the margin of the Programme negotiations about possible bail-in of bank senior debt have been described by several participants (cf. Honohan, 2014b).

12 In the event, despite the Troika's insistence on higher capital standards and more rapid deleveraging of the banks, the additional capital required as a result of the aggressive PCAR2011 stress test of March 2011 was met with less than a half of the pencilled-in figure, thanks in part to the bail-in of subordinated debt holders.

13 This could, for example, have been alleviated through the purchase by a European entity of equity capital in the banks.

14 Unable to provide the normal assurance to their decision-making bodies that the prospective debt profile was, with high probability, sustainable, the IMF staff invoked a "systemic risk waiver" that had been created only a few months earlier in the context of the Greek loan (cf. Schadler, 2013), in their recommendation that the IMF provide this large loan to Ireland.

15 The effectiveness of the Programme and the initial design shortcomings are discussed at greater length in Honohan (2014a and 2015a and b).

16 A different set of financial engineering ideas centred around the idea of "Tap and Swap". This would have involved exchanging the PNs for Government bonds followed by a transaction between the continuing banks and IBRC in which IBRC would receive low-yielding mortgage loans in exchange for the Government bonds. After detailed consideration no available variant of Tap and Swap was found that would yield overall net benefit to the State and therefore this was not pursued.

17 One of the measures pressed on the Irish authorities by the Troika was the merger of Anglo with INBS, and that of EBS with AIB. From my perspective this measure served no strategic purpose, and (although, by concentrating more of the ELA - and far fewer deposits - in the new entity being wound down, IBRC, it allowed some technical simplification of the subsequent liquidation) seemed a largely pointless cosmetic exercise.

18 It is worth noting that this sum is much smaller than the total amount of fiscal belt-tightening that has occurred since the end of the boom. Comparing tax and spending policies now with those in 2008, cumulative "austerity" savings of about €150 billion have been put in place. This sum continues to grow at close to €30 billion per annum (cf. the chart in Honohan, 2015c).

19 The large crisis and post-crisis period profits of the Central Bank (€8 bn between 2009 and 2014) could also be considered an additional offset to the net fiscal costs.

20 For example, the lower figure might correspond to a scenario where all attempts to get some cost-sharing failed and it was decided to go ahead nevertheless with covering all senior creditors; the higher sum might possibly have been attained if a very favourable risk-sharing deal had been negotiated. Failing that, if Anglo and INBS were put into liquidation, over €30 billion in guarantee-related outlays might have been saved, but these fiscal savings would have been substantially offset by the wider economic disruption costs which would have been triggered and which could, to take a phrase from my 2010 Report (albeit there relating to a slightly different scenario), "have run into tens of billions of euros".