Interest Rates on Promissory Notes Not the Key Issue

I am now planning to talk at Friday’s conference about promissory notes, ELA and all that. I will post a link to a detailed presentation when it’s finished, so I don’t want to spend a lot of time on this now.

However, I do want make a brief comment on the recent media commentary on the promissory note issue. Most of this commentary has motivated the issue in the same terms as this article in today’s Irish Times by Arthur Beesley:

State support for the bank is being financed with expensive promissory notes which carry a comparatively high interest rate of some 8.6 per cent.

This is considerably in excess of the prevailing rates for stability facility loans, leading the Government to explore whether it is feasible to draw down additional stability fund aid to replace the promissory note scheme.

Arthur is a fine journalist but I’m afraid this is not a good way to think about this issue. The interest on the promissory notes is going from one part of the state (central exchequer funds) to another (the IBRC). Since the interest rate on these notes is higher than the average interest rate on IBRC’s liabilities, the additional margin can be retained inside IBRC and handed back to the state at a later date. 

So the key issue in relation to the burden on the taxpayer of the IBRC is the amount of liabilities that need to be paid out to bondholders and central banks, and the timing of these repayments, not the interest rate on the promissory note.

I’d note that Arthur’s colleague, John McManus, correctly explains this aspect of the promissory note issue in this article (though other parts of the article are not correct, such as the claim that the Central Bank of Ireland had to borrow the ELA funds from the ECB and that the ELA needs to be collateralised by marketable assets.) The true interest cost of the promissory notes is the interest on the €3.1 billion a year being borrowed from the EU and IMF to hand over to the IBRC, not the notional interest rate on the promissory notes.

Promissory Note Campaign: A Quiet Downgrading

From the Irish Times:

THE GOVERNMENT has quietly downgraded its campaign to persuade the European Central Bank to change the terms of the €30 billion of promissory notes it issued to bail out Anglo Irish Bank, according to an authoritative Government source.

The efforts by Minister for Finance Michael Noonan to seek a reduction from the ECB in the 8.2 per cent interest rates being charged on the notes or extend the term of the loan has not really worked, said the source.

I suspect most of us can think of other euphemisms for “quiet downgrading”.

Time for a Deal on ELA

Whatever happens, there’s going to be a lot of Euro summitry in the coming months. It seems clear that Germany is pushing for a swift Treaty change to introduce all sorts of legal limits on debt and deficits as the solution to the debt crisis. (You could argue it’s a bit like a flood defense plan that relies on banning rain.) In return for this, the ECB will agree to provide funds to bail out Italy and others, perhaps via turning EFSF into a bank.

Personally, I still think the economics and politics of the “Debt Treaty” approach are terrible. But it’s probably going to happen.

Given that, what should Ireland’s government do? Most likely, with the EU threatening to pull fiscal and bank funding if they don’t co-operate, our leaders will just agree to sign the dotted line at the relevant EU Council meeting and then see if they can get away with not having a referendum. (Unlikely — an Irish referendum will be one of many banana skins the process could encounter).

So here’s one thing that I think they can do. If the ECB is going to move into uncharted territory, then it’s time to ask for a small favour that will barely register as relevant when compared with a huge sovereign bond purchase scheme: Delaying repayment of the IBRC’s ELA debts. While unimportant in the European scheme of things, it would give Enda Kenny a big political win if he could announce the cancellation of the €3.1 billion March 31 promissory note payment.

If you want to read more about this, here‘s a column I’ve written for Business and Finance.

Burning Ourselves?

On tonight’s edition of The Frontline on RTE, Gavin Blessing, Head of Bond Research at Collins Stewart made some comments about repayments of ELA liabilities by the IBRC (i.e. Anglo-INBS) that I’d like to elaborate on. Gavin pointed out that IBRC’s major liabilities are to the Central Bank of Ireland. Indeed, I estimate that IBRC now owes about €42 billion in ELA to the Central Bank.

Gavin then followed this up by saying that we would be “burning ourselves” if we cancelled these payments to the Central Bank. This is a complicated business and I fully understand Gavin Blessing expressing the situation in this way. However, I would like to emphasise that it is my understanding that there is no offsetting financial gain to the Irish state from the IBRC’s repayment of Emergency Liquidity Assistance to the Central Bank.

The details are below but I can summarise this issue as follows: Channelling taxpayer funds towards repayment of ELA is equivalent to burning public money.

Let me start by describing the information communicated by a central bank balance sheet, such as this one for the Central Bank of Ireland. Central banks could create money by following Milton Friedman’s analogy and dropping it from a helicopter. However, helicopter drops are neither efficient nor fair. So the long-standing tradition has been for central banks to issue money by acquiring assets via open market operations.

Central bank balance sheets thus show you the assets that a central bank has accumulated via its money issuance. At some point in time, somebody decided it was a good idea to place the money that was issued to acquire these assets on the “liability” side of this balance sheet. I’m not sure this was such a great idea as central bank balance sheets can cause a lot of confusion. Suffice to say, however, these liabilities are somewhat theoretical. If someone brings a banknote to the Central Bank, the only thing they can exchange it for is other banknotes that the cost the Bank almost nothing to print.

That over with, the accounting treatment for Central Bank’s issuance of ELA can be described as follows.

1. The Central Bank provided ELA by crediting, for example, Anglo’s reserve account that it holds with the Central Bank. This was just the Central Bank creating electronic money out of nowhere and this new money was counted as a liability on the Bank’s balance sheet.  In particular, this shows up in “Other Liabilities” on the CBI’s balance sheet.

2. On the other side of the balance sheet, the money that Anglo then owed back to the CBI as a result of the ELA is counted as an interest-bearing asset for the CBI.

Now consider the repayment of part of the ELA by the IBRC. For example, consider repayments funded by IBRC’s annual receipt of €3.1 billion in promissory note payments. One could imagine two possibilities for what happens next.

One possibility is that the following happens. A €3.1 billion repayment gets taken in by the CBI who can then, for example, buy German bonds with it and ultimately use the interest payments on these to pay money back the government when they make profits.  In this case, the amount of money created from the original operation doesn’t change and the Central Bank’s ELA asset gradually turns over time into other, more tangible, financial assets. It is likely that this is what Gavin Blessing thinks is happening.

The alternative possibility is less attractive. The Central Bank takes in the €3.1 billion repayment and then deducts this from the value of its ELA asset. On the liability side it reduces “other liabilities”—the idea is that taking in this €3.1 billion is effectively siphoning off part of the money that was created in the original ELA operation.  In this case, no new securities are purchased by the Bank. The €3.1 billion is effectively being burned.

The available evidence indicates that the latter, less attractive, mechanism is what occurs.

Earlier this year, the Irish government deposited a large amount of money in the Irish banks; this money was later converted from a deposit liability into equity when the banks were recapitalised. When the banks obtained these funds, they reduced their ELA debts to the Central Bank of Ireland.

A quick look at the Central Bank’s balance sheet shows that “other assets”  (which we know is mainly ELA) are down by €17 billion since February. Other liabilities are also down by €19 billion. There is no sign of any jump in the Central Bank’s holdings of other securities as a result of the ELA repayments. There is no hidden positive story at the end of the ELA rainbow.

So why repay it at all? Well, if we don’t repay this money, the Central Bank’s ELA operation will have been equivalent to flying a helicopter over the IBRC, dropping €40 billion and not asking for it back. A jolly good wheeze for the bondholders and depositors who got paid back but possibly not a good precedent for the Euro area. If every Euro area country could do that with their troubled banks, there would be no banking problems but there would probably be a decent amount of inflation.

So our European partners would consider failure to repay ELA to be bad form. But that still seems to leave the pace of repayment, and the funding of this repayment, as very much an open question. In the meantime, let’s not kid ourselves about hidden benefits from these payments.

How Would a Greek-Style Haircut Affect Ireland?

Someone asked me today how a Greek-style haircut for private bondholders would impact on the Irish debt situation if applied here. Without any claim that this is a prediction for what could happen to Ireland, or a policy recommendation, here are the calculations.

While the figure grabbing the headlines is the 50%-60% haircut for private holders of Greek sovereign bonds, it appears that the bonds bought by the ECB will not be written down, nor will the IMF loans. FT Alphaville discuss a UBS report that calculates that a 50% haircut for private bondholders actually implies a 22% reduction in total debt.

In Ireland’s case, the latest EU Commission report estimates (page eight) that our year-end general government debt will be €172.5 billion or about 110 percent of GDP. The report also estimates that by the end of this year, we will owe €38.2 billion to the EU and IMF.  (Table 4 on page 23).

We don’t know how much Irish sovereign debt the ECB own but it’s believed to be a large amount. I do remember a report from Barclay’s claiming they owned €18 billion by June 2010. Let’s say ECB owns €22 billion of Irish debt (that’s just a guess, I really don’t know). Combine that with €38 billion from EU-IMF and you have €60 billion in debt that wouldn’t be getting a haircut. Better guesses of ECB holdings of Irish sovereign debt are welcome.

Now apply a 50% haircut to the remaining €92.5 billion of our debt and you reduce the debt by €46.25 billion, or 29 percent of GDP, getting the debt ratio down to 81 percent. (Of course, we’d still be running large deficits, so it would start increasing again.)

So that’s the answer. Perhaps worth noting, however, is that an alternative method of writing down Ireland’s debt by close to 30 percent of GDP without haircutting private bondholders at all would be to have Anglo’s ELA debt to the Central Bank of Ireland written off.

According to its interim report Anglo owed €28.1 billion in ELA at the end of 2010 but this had risen to €38.1 billion by the end of June. This is because Anglo transferred €12.2 billion in NAMA senior bonds to AIB in February to back the deposits that were being moved out of the bank.

On July 1, Anglo was merged with Irish Nationwide Building Society (INBS) to form what is now called the Irish Bank Resolution Corporation (IBRC). As of the end of 2010, INBS had €7.3 billion in loans from the ECB. However, €3.7 billion of this was backed by NAMA bonds and other assets that were transferred to Irish Life and Permanent. INBS has been in receipt of ELA since February to replace this lost funding. While this has been admitted by a Department of Finance official (see this story) the exact figure has not been released. I assume it is about €4 billion.

So my estimate is that the IBRC now owes about €42 billion in Emergency Liquidity Assistance to the Central Bank of Ireland. If the European authorities ever decide they like the idea of haircuts for Irish debt, it would be fair to ask which of a fifty percent haircut or a write-off of ELA would be more likely to damage Ireland’s reputation or cause financial market contagion.