Wednesday, January 11, 2012

Willem Buiter on an Irish Default

Citigroup’s Chief Economist Willem Buiter’s comments that Ireland will need a second bailout have been getting an inordinate amount of coverage.  There have been plenty of observers who have made the same point from as early as the beginning of the EU/IMF programme in November 2010 and even Minister Leo Varadkar admitted as much last last May.  “Bailout 2” is not news.

What is more interesting are Buiter’s comments on the prospects of an Irish default.  This following are interlaced from two media reports of his Dublin press briefing.

“Ireland needs further assistance,” said Buiter, predicting that although Portugal and Greece will need to restructure their debts and will effectively default, Ireland can avoid going down the same route by having a plan ready in the event a second bailout is needed. (1)

“Ireland is not Portugal, nor is it Greece, but it is, because of the bank debts from September 2008, in very bad fiscal shape,” said Buiter. “I think the politicians and European partners will pursue the option of more generous funding terms before they get to sovereign-debt restructuring.” (2)

One of the primary concessions Ireland should look for is the cost of the Government promissory notes that have been put into Anglo Irish Bank, now known as Irish Bank Resolution Corp, said Buiter. “What Ireland needs to do is refinance expensive debt more cheaply,” said Buiter, saying that a deal on the notes would provide the Government with “material help”. Minister for Finance Michael Noonan was reported to be attempting to convince the European Central Bank to lower the cost of the notes last month, ultimately unsuccessfully. Buiter said that although these notes carry an interest rate of 6% or 7%, the ECB could shoulder some of the burden and that in time it may decide to do so. (1)

Buiter said a refinancing of 30 billion euros ($38.2 billion) of so-called promissory notes that Ireland used to recapitalize Anglo Irish Bank Corp., recently renamed Irish Bank Resolution Corp., at a rate of about 3 percent through the euro region’s bailout fund would “be a material help.” It “would also politically show a recognition of Ireland’s extraordinary efforts to get its fiscal house in order.” (2)

So Ireland can avoid a debt restructuring (default) if it has a “plan ready”.  As agreed last July Ireland will have access to EU funds after the end of the current programme in 2013 when they said that “ We are determined to continue to provide support to countries under programmes until they have regained market access, provided they successfully implement those programmes.”  So part A is all sewn up.

Part B is a restructuring of the Promissory Notes.  Unless we can get a reduction in the €31 billion capital amount I’m not sure there are substantial savings by reducing the interest rates on the Promissory Notes.  The notes have an interest rate of up to 8% but we are not paying the interest to a third party.  The Exchequer pays the interest to the IBRC who in turn pay it to the Central Bank.  This was summed up by Lorcan Roche Kelly with this neat graphic.

This is somewhat paraphrased from the original post (apologies to Lorcan who was making a related but different point).

The interest on the current promissory note is set with reference to the 10 year Irish bond yield. This note could be set with reference to anything and it doesn’t really matter. We will either be paying the interest to a bank that we own, or to a central bank that we own. We pay them €1 bn, they make profit of €1 bn and pay that back to their shareholder – the state. The payment is circular, so the interest rate doesn’t matter, we are paying it to ourselves.

Most important is the term (the point in the future where we actually pay this back). Ireland does not need to worry about any debt roll-overs coming in the near future, so make it a 100 yr term. We will ‘promise’ to have this paid back by the 2111. Hopefully, inflation will have taken care of some of the burden by then. With this exceptionally long term ,we are not disadvantaging any of our creditors, because they will have been paid their money up front, via the nationalised banks. The drop on the ‘real’ value of the debt will not matter at all, because we owe the money to ourselves.

Buiter wants the interest rate reduced from an interest rate of 6% or 7% to about 3%.  This actually doesn’t save us anything.  What if the interest rate was more than doubled to 15%?  Would that cost us money?

We would be paying 15% interest to the IBRC (which we own) who would continue to pay the Central Bank of Ireland (which we also own) interest for the €40 billion of Emergency Liquidity Assistance that the IBRC is using.  The IBRC would have a surplus on this transaction and this money would be returned to the State as a dividend.  The IBRC would make a profit which they can return to us or maybe use to buy golf club memberships for their staff. 

The losses in the bank would be covered by the €31 billion of capital provided by the Promissory Notes.  The interest has no bearing on that.  The details of a restructuring plan for the IBRC have been released by  This says that:

The total cost to the taxpayer for IBRC under the stress case is estimated at €35.8bn

The State has provided €30.9 billion of Promissory Notes and a €4.1 billion direct cash injection into the entities that make up the IBRC.  This is €0.8 billion short of the total cost estimated under the “stress” case.  That cost of the Anglo/INBS debacle is going to be around €35 billion and we have already provided that money.  The issue is how we repay it.

To followed Buiter’s advice and to somehow convert or transfer the Promissory Notes over to the one of the EU bailout funds would actually be a mistake.  Even if this was done at 3% we would be paying the 3% to an external entity and the interest would be lost.  It is better to be paying 7% to ourselves rather than 3% to someone else.

Of course we are involved in the slow-scale transformation of the Promissory Note debt into lower interest debt through the €3.1 billion annual repayment at the end of March.  There are now €28 billion of the original Promissory Notes outstanding following the first payment last year.  To money to make the payment came from the Exchequer which is borrowing from the EU/IMF at an average rate of 3.55% to fund the deficit.

This coming March we will make an further €3.1 billion payment.  This transforms the debt from €3.1 billion of Promissory Notes owed to the IBRC to €3.1 billion of loans owed to the EU/IMF.  This does not increase our debt but instead of paying interest on this debt to the IBRC we will be paying interest to the EU or IMF.  From 2013 this process will slow considerably as the interest due on the Promissory Notes start to be accrued from then.  See here.  Over time the Promissory Note debt will be refinanced to “cheaper” debt but is this actually a good thing?

As Lorcan correctly points out it is the term that matters.  Why should we be repaying the Promissory Notes now?  The interest rate doesn’t really matter and nobody really loses if we repay them 100 years from now.  The only ‘cost’ is that there is around €30 billion of cash floating around that the Central Bank of Ireland (or the ECB more like) would like to see “put back in the vault”. 

But why do we have to do this now we the State is in a hugely distressed financial position?  Why not give the €30 billion back to the Central Bank 20, 50 or even 100 years from now as Lorcan suggests.  Prof. Karl Whelan has been excellent on this point here and here, and explains it in much clearer terms.

Willem Buiter thinks that Ireland needs a two-point plan that will enable us to avoid a sovereign default.  This is a reasonably positive diagnosis.  “The patient is sick, but he will survive” could be one way of putting it.

We will get the official funding that he (and practically everyone else) thinks we need.  We might get to “refinance expensive debt more cheaply” through a reduction in the interest rate on the Promissory Notes by transferring them to either the EFSF or EFSM.  However, rather than being of benefit to us that  could actually cost us money.   What we need is to stop repaying them until we are in a far better position to do so.


  1. Seamus,

    Good article.

    I have to admit I was under the impression that the promissory note interest (which we begin to pay next year) was actually a net drain on the exchequer.

    I have to admit I do not understand why it needs to be paid at all if the transaction is circular.

    Regarding spreading it out over a seriously long period of time (which makes more sense to me now given the previous Gov had said the notes would have a 15 year lifespan and the new one said 20-25), I assume we cannot get out of paying the 3.1bn every year and that the stretching out over a 100 year period only applies to the interest?

    1. @ Rob,

      What about this one to really bake your noodle? What if we actually increased the interest rate on the Promissory Notes? This could actually save us money! Lets see how this might work.

      We are committed to making a €3.1 billion payment each year until 2021 (the first ten years). From next year this payment will be both capital and interest so the capital amount will be falling.

      However suppose we set the interest rate so that the annual amount of interest due was €3.1 billion. There is around €28.5 billion of the notes left so we are talking about a rate of around 11%. If the full €3.1 billion payment is just interest then the capital amount will not be reduced.

      Surely a higher interest rate would cost us more money? Well, it would be a higher interest payment but we would be paying it to the IBRC (which we fully own). In the IBRC's profit and loss statement Interest Income would increase and it is likely that the IBRC would make an operating profit. This profit could be distributed to the company's shareholder (us!).

      We would be meeting the terms of the Promissory Notes (an annual payment equal to 10% of the original capital amount) but no money would be vanishing into the vault of the Central Bank. The IBRC would be paying interest for the ELA it is getting from the Central Bank but a goodly chunk of that will be returned via the Central Bank Surplus paid to the Exchequer.

      We could then revisit the payments in 2022 and look to repay the capital amounts in the years to 2031. It seems strange but it does appear that increasing the interest rate could save us money (if we could get the excess back from the IBRC).

    2. Part of the reason, I believe, for interest rates being set at 10-year IE bond maturity is so that the promissory notes count as long-term capital and so attract a low haircut. In this case, the haircut (which sets the effective interest rate) and is set by the ECB (?) matters.

      There is also an argument that IBRC will eat up the income that it gets - a drag effect on the circle. Have there been many reduncancies at the former Anglo and INBS?

    3. @ yogan,

      That is correct. Though I think the technical reason for the interest on them was so that they could be valued 'at par' on the Anglo balance sheet. Of course, this then feeds into the haircut allowed by the CBoI went taking the Promissory Notes as collateral for the ELA.

      I don't think the IBRC would "eat up" additional income. The costs of the IBRC would not (or should not!) increase if income increased. They would not take on more staff and substantial pay increases are unlikely. If the cost of winding down Anglo is X (loans losses plus operating costs) and we give the bank X + Y, then it should be in a position to return Y regardless of what size Y is; X should not be a function of Y.

    4. @Seamus

      You are a kind hearted man. I might be described as a begrudger.

      In no more than two years you will hear that Anglo are paying bonuses because they are in profit. That this profit is supplied by the state will not be mentioned. There will be a fuss, but the bonuses will be paid to 'retain talent and experience'.

  2. Seamus
    The Promissory note is all very complex but I would like to query a few items.
    Firstly let me say that I think you are correct in deducing that a reduction in the interest rate on the PN (which is payable only to IBRC and not the ICB) would not achieve any saving for the State as claimed by Mr Buiter. But a few other points or clarifications.
    "The losses in the bank would be covered by the €31 billion of capital provided by the Promissory Notes. The interest has no bearing on that."
    Is this entirely true? The interest rate on the PN is now the main source of income of IBRC. If the interest rate is increased, IBRC make a profit (as you point out above) but if the interest rate is reduced IBRC make a loss and will 'require' further subvention by the State.
    So. Yes the money flow is circular within the State in a fashion, but the interest rate originally set, was set at a high rate to effectively subsidise Anglo and reduce the PN capital that 'had to' be put up.
    The purpose was in effect to keep the reported PN as low as possible.

    The Anglo balance sheet at Dec 2010 appears to value the PN at face value . I wonder if this is correct approach to valuing a PN with a risk weighting of zero, carrying 10% interest rate.
    When IBRC take the PN to the ICB (euro system) do they get 100% cash or is it discounted or indeed increased in value for the purposes of collateral.
    The more serious issue which you allude to is Karl Whelan’s point about the PN being used to reduce the ELA to IBRC is in effect burning money. Certainly it is burning money as far as Ireland is concerned. In effect the Anglo loan losses were covered by the State PN over a period of years with the euro system providing ELA as temporary liquidity to Anglo in the interim. As the PN is paid by the State to Anglo, the ELA is paid off to the euro system and the money is burned.
    The bottom line is that ‘Ireland’ is being deleveraged by the full amount of the Anglo Loan losses.
    In effect Ireland is being screwed twice by the ECB.
    Firstly by having to pay for the Anglo losses.
    Secondly by the having the money withdrawn from the ‘Irish’ euro system and therefore the Irish economy.
    So much for European solidarity.

    1. Hi Joseph,

      I don't think the interest rate was set so as the reduce the amount of capital that had to be put in. The Promissory Notes were during 2010 when the yield on Irish government bonds was around 5%. It was really only with the final €9 billion tranche that was not paid over until January 2011 with the ten-year yields at 8.5% that the interest rate rose significantly. It was not a deliberate action to set the interest rate so high.

      I don't know.

      I think there is a small discount (< 5%).

      The money that is withdrawn the 'Irish' euro system is offset by the money that entered the system by way of the Anglo loans (assuming they were funded externally). Although there are huge losses in Anglo the money hasn't just disappeared. A good deal of it will have gone to Irish landowners, engineers, architects, construction workers etc.

  3. Charles Ponzi would be proud!

  4. Amazing stuff gents - thanks for that Seamus.

    I never thought about that, inflation and growth further reducing the real value of the 28.5bn if we can manage to just pay interest for 10 years.

    I guess, if the the transaction is 100% ciruclar (or as Yogan suggests, less than 100% due to IBRC drains) then I wonder why the Government does not seem to have incoporated extra CB surplus monies in their fiscal projections for 2012-2015.

    Also, surprised we can't get Eurostat to do us a solid and discount the PN interest for the purposes of deficit calculation in 2013.

    1. @ Rob,

      The higher interest rate on the Promissory Notes will not affect the interest charged by the CBoI for the ELA it gives to the IBRC. It will be interesting to see what Central Bank surplus is transferred to the Exchequer in March for the year to the end of 2011. In 2010 the average amount of ELA was €22 billion; in 2011 it was nearly €55 billion.

      The Central Bank Surplus transferred last year (for 2010) was €670 million. It might increase this year. Any 'extra' interest should accumulate in the IBRC.

    2. @Rob S
      I would hazard a guess that at some stage the ICB is going to be eating some losses from somewhere so there might not be any surplus that year. If losses are forced on the ECB, then my guess is that it will adopt the Fed strategy - we can't go bust because we can rely on future revenue to fill the gap, but that would eat up a fair amount of surplus from each of the ECB countries.

      A question - does the ICB pay interest on the ELA it 'owes' to Target2? (I don't dare ask Mr. Whelan publicly for fear of kicking off another does of Sinning trolls).

    3. PS Don't forget that part of the interest rate on the PNs is to cover for the interest that was not paid in the first two years.

    4. @ yogan,

      John McManus yesterday in a useful piece under the headline State would not save cent from an Anglo note rate cut says

      "It works as follows. The Government pays something like 5 per cent interest on the note – the yield is around 8 per cent – which Anglo takes to the Central Bank, which accepts it as security for a loan for which it charges around 3 per cent. Anglo pockets the interest margin.

      The Central Bank is in turn getting the money it lends to Anglo from the ECB at a much lower and not disclosed rate which is reported to be 2 per cent or less. It keeps the difference. The real cost to the State is the rate at which the ECB provides cash and it is far from penal.

      It is thus – as the ECB claims – within the Government’s own power to significantly reduce the interest rate on the promissory notes, but there would be no gain to the State."

      You'll have to ask Prof. Whelan how this is incorporated into the Target2 framework. I haven't a clue.

      The "interest holiday" is a strange thing alright. I think in Anglo's accounts the interest is registered as being part of the €3.1 billion annual payment, but in the government accounts it's all capital for 2011 and 2012. This must have seemed like a useful idea when it was felt that this was only a short-term blip and that the public finances would be back on track by 2014.

  5. Seamus
    re point D above
    "The money that is withdrawn the 'Irish' euro system is offset by the money that entered the system by way of the Anglo loans (assuming they were funded externally). Although there are huge losses in Anglo the money hasn't just disappeared. A good deal of it will have gone to Irish landowners, engineers, architects, construction workers etc."

    It seems to me that, from a Euro system point of view, when bank losses are paid by the State (in Ireland or elsewhere) into the Euro system, money is being withdrawn from the euro system.
    The original money created by the ECB is still out there. Anglo 'lost'it but somebody else gained. The money is still in 'circulation'.
    But by insisting that the bank losses are made good by the taxpayer (in any country), and paid back into the ICB/ECB, that money paid back is effectively burned, unless the ECB put it back into the system again. And even if the ECB do put the money back into the system, it is unlikely to be in Ireland.
    Therefore it seesm to me that 'Ireland' is being deleveraged by the amount of the IBRC losses?
    You are correct that the money did go into the Irish economy in the first place. But that money has been distributed and diffused throught the system. The fact that Anglo has lost it and somebody else has gained is hardly a good reason for withdrawing the full amount of the Anglo losses from circulation.

    I may be labouring this point, but I think Karl Whelan's point needs to be elaborated on. I think he is correct when he says the money is being burnt.

  6. @ Joseph,

    The only money covering bank losses that is going back into the ICB/ECB is the money going against the capital on the Promissory Notes. And this is money that the ICB just "printed" when Anglo presented the Promissory Notes to them. The money that is being "burned" is offsetting the €30 billion of new money that the ICB created.

    The rest of the money going to cover bank losses is staying in the system because it is not going to the ECB - it's going to all them damned bondholders don't ya know. It's not totally, but that is where the money is going. To help meet the non-central bank liabilities of the banks. In fact, most of the money is still on the balance sheets of the banks. They are after all "the best capitalised banks in Europe!"

    The €31 billion that will repay the capital on the Promissory Notes will be "burned" but this "deleveraging" as you put it will take place over the next 20 years. We should be trying to stretch it out over an even longer period.