State would have to repay billions to depositors in risky Anglo wind-down, by Alan Dukes.

See full piece here.

41 replies on “State would have to repay billions to depositors in risky Anglo wind-down, by Alan Dukes.”

“Alan Dukes was appointed to the board of Anglo Irish Bank in December under the Bank Guarantee Scheme. The contents of this article are the personal views of the author.”

‘nuf said.

I am very disappointed in that article. What is going on in AngIB, that it has to be covered up by both main political parties?

This article ignores the main points concerning the Anglo case:

1) Operating a failed bank as a going concern only makes sense if the bank has a working business plan. Anglo does not. Its model was to lend wildly to property buchaneers, and when this bubble burst, it was left with nothing but debt.

2) The liabilities Mr Dukes presents as inevitable are a result of the State’s guarantee scheme. This scheme was a horrible mistake, of course, but it is not irrevocable. Liabilities held by Anglo will fall at a discounting rate if the government credibly announces a plan to allow the BGS to expire.

3) Under no circumstances can it credibly be argued that Anglo is systemically important. At most, it could be argued that honouring its liabilities is systemically important (some of them should be honoured, but we have EU legislation for that…), but that is not the same thing as keeping it running as a going concern.

4) There is, of course, one cadre of people who clearly stand to benefit from keeping Anglo running as a going concern. See Marcus’ comment above.

Alans a great guy. But his article (which, I suspect, is in response to mine) is pretty ropy
It ignores the key point : there is not enough in the asset base at any feasible realistic rate of discount to pay the liabilities of the bank. Not. Enough.
So, in a winddown or zombie-life-maybe-itll-get-better-somehow-look-over-there-its-a-green-shoot scenario, we are on the hook. My question is : why throw good money after bad?

Alans article , summarised is as follows. One sentence per paragraph

the government said anglo was important recently.
It must be imporatnt as its big
And it must be important as it lent a load of money to developers
The markets look at anglo.
Brian Lucey and Richard Bruton and so on say close it and save 4b
But the central bank is making us keep to regulatory minima on capital
The retail deposits would stay
the corporate and internbank deposits might or might not stay
Anyhow its not enough to offer 7times the market rate for money and a government guarantee, is it?
Plus the bondholders, what about them, eh?
sure they have lost loads of money as well
and we cannot negotiate with them any more, sure they lost loads poor lambs
Anglo owes a boatload to the ECB which would have to be repaid
And if we wind it down we would have to repay that
Trust us – we have a plan
Its to keep anglo going.

Now, thats not, to my mind, convincing. We need a few facts here , enough opinion (including mine) has been aired
a) Will the ECB supply interbank funding to anglo while it winds down?
b) will the Central Bank take its licence away as soon as it says its winding down?
c) Will corp treasurers keep deposits when the bank is paying significantly over the odds and is guaranteed even if its going out of business?
d) What guarantee do we have that 4b will be enough to keep anglo going?

On a related note –

To what extent is the key issue (bondholder pain before taxpayer ‘investment’) already being played out?

In my mind irish banks buying back non-guaranteed debt is putting the (junior) bondholders first and it can be seen that this is being done with taxpayers money (via recapitalisation).
This doesn’t tally with my understanding of the solvency supporting deal that said ‘we will give you x funds provided you go raised y funds’.

What about the derivative book? 192 bn of traded derivatives (i.e. not held for hedging purposes). Both unwinding and keeping going scenarios have to address this issue.

BTW – given where Ireland Inc is, I believe the very high quality debate on this website makes it very important.

The derivatives book is something that has begun to poke its head over the surface.
How much would be crystallied in a windup?

Who knows. But I really believe the Mr. Dukes let slip that it was many billions and that is what he initially referred to. This business of deposits or secured bond holders is smoke and daggers.


I asked you in the discussion on you article what would the consequences be if your opinion, that a large proportion the deposits would not leave if a wind up was announced, were wrong. How would we fund it when the abnks assets are illiquid? Can you please address that point?

Alan Dukes article is short and easy to understand. Most of the paragraphs are one sentence long and none of the paragraphs are longer than two sentences. Not wanting to be confrontational, I suggest that people read the article itself.

First, dont pester. Im actually off sick at present so gimme a mf break, alright? Plus, AFAIK I dont answer to you directly.
Sheesssh…..I actually have other things to do apart from save the goddam banking system, y’know. like get well.
As you ask so nicely, I will break off my recuperation regime and tell you: You asked what if I was wrong. Then, oops…..This aint physics, so my (informed) opinion is that. Opinion. For what its worth, i cant see the ECB/EC letting us go down the pan for what is shortterm liquidity reasons. But I could be wrong.

Alan’s article is short and snappy. But mine is shorter and snappier.

Decided to take a closer look at Anglo account. A simple balance sheet looks like this:

Assets Liabilities
Deriviative fin instruments +4.7b -4.0b
Loans & advances/deposits +66.6b -34.1b
Loans & advances banks +6.7b -30.5b
Available for sale fin assets +7.8b
Debt securities -14.2b
Other bits & pieces +2.7b -5.0b

88.5b -88.5b

In a liquidation it is the €66.6b that has the main question mark over its value.

According to the accounts the loan book is made up as follows:
Good 39b
Satisfactory 4b
Lower quality but not past due 5b
Past due but not impaired 13b
Impaired 11b
Total 72b

Prov 5b

In the chairman’s statement he says if accounting standards allowed they would have provided €7.5b at 31/3 and there was risk of a further €1.5b to €3.5b under stress testing. Say €11b in total. Past due but not impaired includes €2.5b over 90 days overdue.

In theory then you wind up the bank pay all the creditors (exc debt securities €14.2b) €74.3b and collect as much of the asset side as possible. Allowing impairment to double to €22b you collect €66.5b, net shortfall €7.8b.

The first step is sell the good loans of €48b to another bank/s. That covers your non bank depositers. Then work on the rest. Of the €30.5b bank deposits €23.5b is from central banks and won’t vanish overnight.

Probably missing something but it could be done for say €10b.

Excellent work. Main issue however is whether you would get 48b, or 35b, or any other figure for the loans.
We hear today that a 20% haircut is being suggested for the banks.

In the US the govt kept the Savings and Loan banks alive by guaranteeing high interest rates for a year. They stayed alive, then crashed when the guarantees ended — deposits insured by the govt, of course. An S&L association is a financial institution that accepts savings deposits and makes mortgage loans.

After the collapse, the govt acquire the assets at nearly full value, then had auctions where it sold everything at whatever it could get, which meant anything from a half a percent on up.

They sold off credit card debt at .05% and speculators collected the money from the debtor for 20% and up.

They sold real estate for anything. Condos in Massachusetts, Florida and Texas went for $10,000.

The govt economists’ estimate was that it would only cost $10 billion to “resolve the crisis.” The ultimate cost is estimated to have totaled around $160.1 billion, about $124.6 billion of which was directly paid for by the U.S. government—that is, the U.S. taxpayer, either directly or through charges on their savings and loan accounts – which contributed to the large budget deficits of the early 1990s.

From 1986 to 1995, the number of US federally insured savings and loans in the United States declined from 3,234 to 1,645. This was primarily, but not exclusively, due to unsound real estate lending.

The U.S. government ultimately appropriated 105 billion dollars to resolve the crisis. After banks repaid loans through various procedures, there was a net loss to taxpayers of approximately $124 billion dollars by the end of 1999.

Of course, that could never happen in Ireland. It’s different here, right?

If Dublin property is down 50%, why does anyone think a 20% “haircut discount” at Anglo-Irish will be attractive to investors? “The value of anything is what someone will pay for it.” – Ben Dunne

Good work Stuart and Henry.

My overall feeling is that the value of a company is nearly always greater as a going concern than on a “break up” basis.

Profits in Anglo are 295 million before the (admittedly enormous) provision of 4.1 billion. If we get through this there may be a viable business and we will have saved 1000 jobs (again I admit at a very high price).

However, I find Alan Dukes’s points convincing. The existing policy of keeping the bank going is unpalatable, but the alternative of a wind down is a far worse alternative.

@ John M
So if we spend 10b plus on saving 1000 jobs thats a good outcome? Jeepers, that would make FAS look like a bargain


It’s a bargain compared to the cost of winding up the bank.

I don’t accept your figure of 10bn plus to recapitalise. The bank is looking for another 4bn with a risk of another 2.6bn. As I’ve said from what I have seen these are very reasonable and conservative estimates.

Actually the bank wanted to make a €7.5b provision but accounting rules prevented it. They were not allowed to take into account events that occurred post 31/3/09. (See Chairman’s report). They also say under stress testing another €1.5b to €3.5b may be needed.

The bank itself are therefore saying another €2.5b – €6b as a going concern on top of the €5b.

The difference is in the liquidation you don’t pay the bondholders €14.2b. They took a risk, they lost. (As I did as a shareholder!). That gets my liquidation cost of €7.8b – the difference between the write downs on loans and amount due to bondholders (By the way my €7.8b includes an extra €22b of impairment/wind up costs. Collecting €66.5b vs €88.5b!) If we feel we should pay the bondholders something to get them to cooperate then throw in another €2.2b for them and you get €10b. At least then we’re in the clear, no future shocks.

In my scenario you sell the good debt to another (probably foreign) bank who operate here under their own name as a wholesale style bank. A lot of the jobs will be saved and we get a bank that will actually be doing business. (Actually on the jobs what surprised me was no big change in head count. Surely there must be a lot of people twiddling their thumbs if they’re not actually doing any new business.)

This is just back of the envelope stuff by me. What am I missing? Why is the wind down a far worse alternative?

Anglo is NAMA 1

If you chant this long enough I believe that the answers to our current crisis become clearer:
1) No NAMA 2,
2) The idea of transferring loans (with friction, decade long court cases, etc) from NAMA 1 to a NAMA 2 is ludicrous.

What about the original plan; support Anglo through the crisis, then as the economy picks up (as it will) let them pay back what has been loaned by the taxpayers with a suitable bonus on top and then pass it back to the original shareholders which incudes many innocent people who are just as much victims as anyone else


Just to clarify my figures. My understanding is that the Bank is looking for an additional €4bn from the State. This is with a provision of €4.9bn (4.1bn plus 0.8bn from previous periods). On the assumption of deteriorating market conditions they will need another €2.6bn. This will bring the provision up to €7.5bn. This is being conservative and prudent as accountants are supposed to be. The additional €1.5 bn to €3.5bn is worst case scenario. It is not based on what we know but what might happen. So the worst that can happen is an €11bn provision or extra funding of €10.1bn.

I have to sign off until this evening when I will hopefully look at your figures in more detail. But my instant reaction is that the liabilities will crystallise much more quickly than the realisation of the assets if the company is wound down. Also, the value of assets realised will be far less than what can be achieved as a going concern.


I think I can follow your figures. As I said the big problem (at least per Alan Dukes) is that the liabilities are likely to crystallise almost immediately whereas the assets may take far longer to realise. So the 7.8 billion (or 10 bn) loss might be the final outcome after a number of years but there might be an immediate outflow of over 50bn.

Your extra 22 billion liquidation cost seems reasonable although Henry Barth might think it is too little. The quicker the liquidator needs the money the greater the discount required. So in this fire sale situation I imagine we will be close to Brian Lucey’s view of a €40 bn write down of the €72 bn loan book.

Reading Alan Dukes he thinks that we can write off the 4.9 bn subordinated debt (see page 21 of the acconts) and not the 14.2 bn bondholders debt which you have included (this 14.9 bn may be covered by the State guarantee?). If this is the case you will have to increase your loss figure by 9.3 bn.

There are other intangibles such as the credibility of the banking system and the credibility of the State which owns Anglo if the bank is dissolved.

Finally, regarding Charles’s point. There is absolutely no way that the shareholders should receive anything. The bank would be dead and buried if it were not for the State. You have my sympathy Charles, but that’s the risk shareholders take.

Personally not whingeing as a shareholder as having bought in 1997 didnt lose out but I do feel for the recent buyers. The fact is that the directors of the bank i.e. Fitzpatrick , Drumm, McAteer and the ones still in situ behaved in the most unbelievable manner in taking the loans they did and concealing them from the shareholders and the auditors (I think its called fraud but not sure how to spell it). They also took deposits from other institutions with their full complicity at the year end and paid these loans back immediately after to bolster the figures (I think its called fraud etc) so the shareholders have been the victims of fraud so they should receive something whether it is from the phoenixed bank or from the ex directors or the bank/auditors insurance companies.


The fact that the shareholders were cheated does not equate to a moral obligation on the part of the taxpayer to foot the bill. By all means, if fraud can be proven, go after the personal assets of those responsible.

But that is not the main issue. The main issue is whether we can let this bank fail without ruining the entire banking system, and whether it is feasible to do so. Certainly it is moral to do so.

Think you have answered your own point. Allow the bank to fail and confidence will be lost in all Irish financial institutions; nurse it through and back to health and there is a positive message to outside lenders and depositors.

“Allow the bank to fail and confidence will be lost in all Irish financial institutions; nurse it through and back to health and there is a positive message to outside lenders and depositors.”

Several questions all unanswered by the powers that be.
1. The price of that positive message for the Irish taxpayer. First it must be the €14.2b to the bondholders who I am presuming get nought in a liquidation.
2. Do we have another mini Anglo out there (e.g. Irish Nationwide) and do we rescue that too?
3. Can Anglo actually be nursed back to health? I still suspect the end result will be to sell off what’s left i.e the good loans to someone else. It’ll just take longer and yes you may get more but balance that against the interim cost. (PS John – you don’t pay the depositers until you get some cash in. That’s how all liquidations work even for creditors who’s liabilities are guaranteed)

I suspect the confidence in Irish institutions is pretty shot to pieces anyway in an investment world where Ireland might get 5 minutes look in and will take years to recover regardless.

All I am asking is for someone in the know to answer the above questions and not adopt the “trust us to know what’s best” approach.

Looking at the share prices of AIB and BoI over the past few weeks it looks as though confidence in these institutions at least is already on the way back and part of that confidence quite definitely relates to the “rescue” of Anglo; if it had been allowed to fail would confidence in the other two be the same? I think not


I agree that in a normal liquidation creditors have to wait in line until assets are liquidated. However, I don’t think you are right that that logic applies for creditors with a State guarantee. The bank may be insolvent but the State is not insolvent. The State will be legally obliged to pay these creditors immediately.

Of course, if the State can’t meet these obligations then we really are in trouble. I prefer not to provoke such a situation by winding down Anglo immediately.

Incidentally, I am still under the impression that the €14.2 b bond holders are covered by the guarantee. The creditors who are not covered are the €4.9b in subordinated debt. That is the only sense I can make of Dukes’s article. It is only this €4.9b that we (i.e. the State and us taxpayers) can avoid paying in the event of liquidation of the Bank.

I think JM is correct that only certain undated subordinated bonds are not covered by the guarantee and these are about €4.9B.

If Anglo is nursed back to health it will be after transformation into a non-systemic bank. Once that transformation is achieved the Govt can let it survive or fail. Once the govt can let it survive or fail the non-guaranteed bondholders are in a considerably weaker bargaining position.

Perhaps the people buying shares in AIB and BOI don’t realise they still risk losing everything. The only people getting stung by nationalisation/NAMA are shareholders and the taxpayers. The possibility still remains that the state will end up owning a very high % of the banks once NAMA runs its course and the state has to put a further investment in. By the way they are very good speculative shares if you’re prepared for the rollercoaster and the possible end result.

It will look like Nationalisation in all but name if the state owns 80%+ of the banks.

You may be right about the €14.2b. I don’t know because there is very little clarity on the liabilities and that appears to be deliberate on the part of the government. It would be quite easy for them to do my calculation and explain exactly what the position is. But they won’t.

Finally – anyone know what the story is with Irish Nationwide?


Dated subordinate debts (€2,125m)are covered by the guarantee.
Undated subordinated debts (€2,820m) are not covered by the guarantee.

This is confirmed on p.13 of the report you linked to. The breakdown is on p. 48.

Apologies JM – perhaps you had spotted that already and were just pointing out the co-incidence.

However the bank does not intend paying out fully on all sub=debt. See statement on p.5:

“Furthermore, the Bank intends to offer to repurchase certain outstanding subordinated bonds at a significant discount to par value which will generate profit and additional capital for the Bank. This is a major step towards placing the Bank on a firmer footing for which we are most
grateful to the Minister.”

Thanks for that Zhou. I hadn’t realised that the dated subordinated debt was covered by the guarantee. I was also relying on the Dukes article in which he says, as far as I recall, €5b odd not covered by the guarantee.

All of this means that the case for continuing to run Anglo as a going concern is much stronger since on liquidation the only amount that the State can walk away from is €2.8b.

Given what we read lately in the newspapers, there seems to have been a great deal of fraud in the Irish banking system, from mere lying on loan apps to outright theft.

Yet I see no action being taken against the perpetrators.

Consider the US: Madoff was discovered as a fraudster, arrested, his assets seized, and sentenced to prison all within three months.

Why doesn’t this happen with criminals in Ireland? Where is the CAB?

I could never understand why Maureen Haughey was allowed to keep the house in Kinsealy after CJ passed to his reward. The CAB should have seized the house and contents and auctioned it all, instead of discounting the taxes owed and reaching an accommodation with the family.

CJ’s money was made illegally. Do we need a Trubunal to know that?

Does anyone think Seanie Fitz will have anything seized, or go to prison, even years after his fraud was discovered?

Lack of punishment is part of the Irish problem.

I agree with you that it is all part of the Irish psyche; good on you lads if you can get away with it and its been going on for as long as I have been alive (too fecking long as some would say) but the blessed Charlie must have been one of the greatest and what happened about him? Diddly squat. After the first few weeks I havent had a word from anyone, least of all the press, about “Seannie” and his hundreds of millions. Presumably if he cant pay back the debt to the bank (and of course everything will be offshore or in his family’s names) then he will be made bankrupt and disappear into the Spanish sunset giving himself a hernia laughing

Comments are closed.