Debt Buybacks

Commenter zhou_enlai raised the possibility earlier today of a debt buyback from the ECB. Such a buyback provides an interesting third way between the problematic bailout strategy (new lending in the hope that some combination of growth/limited bank losses/primary budget surpluses will restore solvency) and the problematic bail-in strategy (restoring solvency through default).

There is an interesting literature on debt buybacks, with classic contributions following the Latin American debt crisis by Jeremy Bulow and Kenneth Rogoff (see, e.g., Bulow and Rogoff, 1988). They refer to buybacks as being a boondoggle for creditors. A key part of the argument is that when there is a high probability that creditors would not have got their hands on resources used for the buyback otherwise, the buyback provides an expected net transfer from debtor to creditors. However, others have argued that it is more likely that both sides can gain where there are large costs to default for the debtor. (See here for an excellent survey on the implications of sovereign default.)

In a potential Irish buyback, I think it is reasonable to suppose that creditors would have eventually got their hands the assets that would be used for a buyback (say the NPRF and the NTMAs liquid reserves). But I think it is also reasonable to suppose that there would be a large cost to Irelands highly open economy from a default on ELG guaranteed bank debt or State bonds. Using the IMFs fiscal space model, market participants have concluded that recent developments on growth and banking losses have eroded Irelands fiscal space. However, it might not take a large reduction in outstanding debt to restore sufficient space, significantly reducing the risk of a messy restructuring. Thus it is still possible that both sides could gain from a buyback despite creditors access to Irish resources.

One would think that the mechanics of a buyback could be reasonably straightforward with the ECB holding a significant fraction of Irelands outstanding debt. (Someone might be able to fill in an estimate of this number.) Just for illustration, a buyback of debt equal to 30 percent of GDP at a one-third discount would reduce the net debt to GDP ratio by 10 percentage points, possibly enough to restore enough fiscal space to ease creditworthiness fears.

There are of course potential legal issues to the extent bonds have covenants against such buybacks. But where buybacks are seen to benefit all parties, these restrictions do not appear to have been deal stoppers in the past. It would be good if those with knowledge of the legalities and other practicalities could weigh in. It certainly seems worth some debate.

Bulow, Jeremy, and Kenneth Rogoff. 1988. The Buyback Boondoggle, Brookings Papers on Economic Activity, No. 2, pp. 675-704.

31 replies on “Debt Buybacks”

Indeed and not the first to say it. It makes sense to let our yields blow out, keep talking up default and not actually do it. Meanwhile the ECB provide enough liquidity to soak up any sales on the supposition that we buy back/refinance the debt at the due date at the price they bought it rather than at par value.

Our market yields do not matter for the next three years. Rant away about default. Get the country’s credit rating down to junk. It does not matter with the funds that are available.

Meanwhile, get the deficit to a surplus as quickly as possible… ‘over-loss’ the banks so they have massive reserves… but without telling anyone…

Anybody care to explain that trick in plain english – I have no idea what its about – or is that as simple as it gets?

@John McHale

On a figure? On the IMF/EC/ECB talks I thought we had a reasonably strong case to ‘park at least €50 billion of banking debt’ within an ECB SPV as our position on the curve, and explosive particularities, threatened the core of the european banking system. We are in this case unfortnatley ahead of a Central Solution – as proposed on EuroIntelligence and elsewhere – e_bond, restructure banks & bonds, + a lesser hit for the European citizen in periphery and core mediium and long term. Europe needs to get REAL. Not only did we not park the 50billion, but we had our back-pockets picked on the way out. This game is ongoing ….

Can’t do the sums, beyond my don’t-receive-dole allowance, but it goes something like this.

Yields rise because prices of bonds are falling.
Prices are determined by buyers vs sellers. An excess of one = a direction in price.
Buyers of sovereign bonds are generally influenced by the rating of the bonds they are buying.
Irish bonds are 3 (?) steps from junk.
The ECB are currently providing liquidity by buying when no-one else will, so they are not so much supporting price as preventing a blow-out in price (to the downside); also known as providing liquidity.
So the ECB are buying at below face value (i.e. the value that will be repaid when the bond matures).

So, we keep making wailing and gnashing of teeth noises, get lots of people to flood to the exits and buy the bonds up on the QT (with the help of our friendly ECB which allows a jump-drop before stepping in).

Do the same for bank bonds.

There is a floor to price – the level that hedge funds and other speculative buyers will jump in, but it may be a moveable feast…


I did make it sound more complicated than it is by getting into the debate about who gains from such buybacks.

The basic idea is very simple: Buyback the debt at market value and retire the face value of the debt. The outstanding net debt falls by the difference between the face value and the buyback price. It has essentially then same effect as a debt restructuring, but without a direct default.

There are still legal and reputational issues, but it is an option that should be thrown into the mix, especially as the restructuring idea gains momentum.

Would there be any legal issues if we just made an open tender offer?
At say 85c in the €.

The advantage of this strategy is it could also push the bond price towards this price decreasing the yields making it easier for us to borrow in the future.

The only problem is it can’t be done with borrowed money otherwise I think you get into a bit of loop. Where you have to over 85c in the € for the old debt but we would only get 85c for each € of new debt. Of course EU/IMF money could be used for this purpose.


I don’t fully get why there would be legal issues.

In order for there to be legal problems, a party must object. If the ECB buys the debt on the secondary market and sells it back to the Irish government over time, who would object to the legality of it (presuming the ECB would has full EU backing for this course of action)? As the remaining private bondholders would then stand a better chance of getting paid in full, I can’t see who would lose out.

I also cannot see why there would be reputational damage, as again, everyone who wants to hold onto their bonds until maturity stands a better chance of getting paid.

Furthermore, as the bailout money will have to be use to roll-over sov debt anyway, this would provide a more cost-effective way of rolling over the debt. So, wouldn’t it make sense to use some of the bail-out money for this purpose.

The main problem that I can see is the decreasing effectiveness of the mechanism – as the ECB buys up the debt, the yields will fall and perhaps overshoot on the expectation of further buybacks, making further buybacks less effective.

@ DE (and all)

as you suggest, if we have the cash to buy it back at 85 cents, then we probably aren’t worried about defaulting, right?

So the key is to find a liquidity mechanism, via the ECB/EU/IMF, to do it on a large scale. You could argue that the ECB is already doing this via their SMP buying, but nothing formal has been suggested as to where the “profits” go – at the moment it would appear the ECB pockets it.

This buyback was also suggested in the much-read Tremonti/Juncker article in the FT on Monday, albeit via a Eurobond-for-nationalbond swap.

@ Johnny McH

guestimate would be ECB bought 15-20bn of Irish govvies at this point. (est 30bn Greece, 15-20bn in both Ireland and Portugal, 5bn in Spain. Total at moment is 69bn)

@John McHale

Thanks for posting links to additional papers. It really is great to get direction to these resources.

Without having yet read the papers, I expect that the biggest legal difficulties arise from the legal relationships between Ireland and the ECB (or such other body as may suppy the funds for the re-purchase of bonds). However, it may be possible to adjust these relationships accordingly. For instance, if the EFSF is lending to Ireland at 6% and the ECB buys Irish debt on the secondary market at such a price to give an effective yield of 10% then shouldn’t the ECB be allowed to reduce the nominal capital value of the bond to give an effective yield of 6% on the price paid by the ECB?

More generally, if the market dictates that a state only has so much capacity to repay debt and values debt accordingly then it is a pity that if this this market mechanism cannot be used to formalise the position. In some regards this market evaluation is more effective than inflation and QE which change the base and make it more difficult to assess the true position. It is also more effective than restructuring which similarly implies the rules being changed for investors which leads to uncertainty and instability. Clearly some legal innovations are needed to embed CACs and in this regard one can see where Merkel was coming from. If the ECB or another EU organ steps into the breach to allow bonds to be repurchased then that will skew the market and defeat the purpose.

Perhaps we need the assistance of private actors (pensions, irish companies) who will take the action on our behalf.

Or perhaps we need to set up another NPRF style fund based on a compulsory discrete tax that gives each taxpayer a discrete share in the said fund, with such fund being dedicated to puchasing Irish debt and restructuring such debt in the joint interests of the state and the part-owners of the fund. That’s it, problem solved!

In the meantime, if the markets re really behaving irrationally as many Finance Ministers have said then surely this is the best time to buy back debt?

@ Hogan

I see a flaw in your cunning plan. Downgrades trigger unwanted reactions at insurers and banks.

I note that Ireland had a AAA rating in 2009 and once the death grip of the banking albatross kicked in it there has been no brake to the fall.

It would be a flaw if we still have a serious investment grade rating. As it is, the trajectory means that we have de-facto junk status. By not letting us arrange some default of bank debt, the institutions of Europe have ensured a ratings default of Irish sovereign debt.

You should be clearer about:
1) Who is doing the buying the ECB, Irish Central bank, some other entitiy?

2) What debt is being bought, Sovereign debt, commercial debt, Credit Card accounts?

In the beginning there was a liquidity crisis? Although I’d say that the liquidity crisis was caused by a solvency crisis.

This ‘liquidity’ crisis caused Irish businesses & consumers to have a shortage of funding for investments. NAMA was introduced to address the ‘liquidity’ crisis. Liquid bonds was traded for illiquid difficult to value assets. The liquid bonds were supposedly to provide liquidity for Irish businesses & consumers.

Not surprisingly this liquidity went to pay back bonds leading to little to no improvement in the liquidity for Irish businesses & consumers.

Now we have a plan to address the solvency crisis by using the liquidity still in banks for debt buy-backs?

Some might see this debt buy-back as something similar as a government sucking up all available capital thus starving the rest of the economy of capital….

I have to admit that the solvency situation is likely to improve & I suppose in the long-term this will improve liquidity. However, in the short term I have to ask: I have difficulty seeing these actions providing any benefit to the wider economy, what economic function do these banks fulfill?

Is the economy in place to serve the banks or are the banks in place to serve the economy?

“Is the economy in place to serve the banks or are the banks in place to serve the economy?”

Neither. The banks are in place to serve the bankers.


Such a buyback provides an interesting “third way” between the problematic bail-out strategy (new lending in the hope that some combination of growth/limited bank losses/primary budget surpluses will restore solvency) and the problematic bail-in strategy (restoring solvency through default).

But unless I’m much misinformed, a great deal of the “untouchable” Irish debt (the sovereign and bank senior bonds) is not freely for sale at market price. It’s sitting on the books of financial institutions across Europe, marked to fantasy valuations that play an important role in keeping the holders adequately-capitalised on paper, yes? Is this not why the Troika won’t accept restructuring of senior bank debt? It’s not because they’re concerned about the propriety of reneging on the ELG or whatever. So, even leaving aside the question of further haircuts foregone, I’d tentatively guess that there’s not that much to be saved by the Irish government simply offering to buy “untouchable” debt from willing sellers at market price, and that’s before you consider the interest rates at which it would have to borrow to fund this. (Or the fact that the new debt is a bilateral sovereign obligation, because sure won’t the lads be seeing us right.) Given that our government is eager to negotiate with senior bondholders somehow, and never has any qualms about jeopardising our long-term position in the interests of immediate survival, I presume that if buybacks on this basis were an immediate winner they would have happened already.

Presumably such a buyback could be made more effective if some of the European banks sitting on “untouchable” debt marked as par could be induced to participate, through collective action or whatever. The problem is that this solution lies on the far side of Europe’s bank-restructuring pons asinorum, just as restructuring of the “untouchable” debt does. Similarly, all the buyback schemes that involve the ECB or the Commission or etc. buying the “untouchable” debt at face value (or whatever value) and then writing it down lie on the far side of the fiscal-transfer pons asinorum alongside EU grants for Irish bank restructuring. When one or both of these bridges is crossed, then the matter of grants versus Eurobond swaps or buybacks versus restructuring and so on is just detail. Until one of those bridges is crossed, Europe’s position is “What part of ‘No’ don’t you understand?”

Crucially, the Eurosystem is effectively one of those banks which can’t afford to write down any of its Irish bank or sovereign debt, though mainly for political-legal reasons in its case. And there’s little hope of getting around this by stealth, partly because any significant Eurosystem writedown would soon be loudly telegraphed as the ECB printed money or begged capital to repair its precarious balance sheet.

However, it might not take a large reduction in outstanding debt to restore sufficient space, significantly reducing the risk of a messy restructuring.

Barely enough debt relief to keep Ireland out of default? Be careful what you wish for, we might just get it. What’s the easy way to quietly give Ireland some breathing room without alerting the German voter (much), without giving Spain any ideas, and certainly without putting any stress on the EU banks? What method has just been employed in the case of Greece? Quiet maturity extensions on our EFSF/EFSM/IMF debt. The metaphor of a half-hanging comes to mind: the noose can be opened a bit as soon as we start to lose consciousness, then allowed to tighten again. This process could potentially be repeated for a very long time; surely a good five years is perfectly possible. If you want a more economic metaphor, this would be a process of rackrenting, where our landlords capture all the surplus and slowly grind us down. Of course this belies the pro-bailoutist argument that as soon as our debts become unmanageable, if not before, the EU will have to allow us to participate in some kind of Great Jubilee.

@ Hogan

It is really grim innit. What are likely to be the unintended consequences of ratings drift to junk ?

John, unless you can show you have the support of the people to put up with the fallout, everyone will know it is an empty threat.

Well for one, the guaranteed bonds also drift to junk making them also candidates for buyback at a discount… Mind you, we know that the other banks bought these, so maybe there is not much benefit to this…

Borrowing for other state agencies (e.g. the HFA) and the semi-states will become more expensive/less possible.

If we did have some natural disaster we would be stuck, unlike, for example, Chile which was able to finance reconstruction following its recent earthquake.

Is it a bad thing that new debt is limited?

It’s a good post, but it’s not great… if nobody is selling, then how are the ECB holding 20% of Irish sovereign debt having bought it in OMO…


It’s a good post, but it’s not great…

I’d never deny that either. 🙂 I certainly didn’t suggest that nobody is selling. The 20% does suggest that there’s a significant amount of debt still at the horse fair. Though again, I can’t see how the ECB itself can accept significant losses on its holdings. At a wild guess I’d also presume that the proportion of Irish senior bank debt in the wild is smaller than that of sov. debt, since it was the banks that were borrowing pre-2008?

In terms of bond debt, yes, I’d say their is less in the way of bank debt. Still, buying it in secondary rather than through a formal exchange offer is preferable (cf Moody’s calling BoI sub. buyback a default).

How much of the sovereign was direct secondary of primary issues is interesting. I don’t know the answer. Certainly the rumours were that this was happening, but there is no breakdown on this. I think Central Banks in general buy debt as a matter of course, but my understanding is that the 20% is over and above this.

In terms of losses, it’s not that the ECB would be making losses. The Irish 2025 is trading about 20% discount to par at the moment. The ECB buys at 80, we ‘buy’ from the ECB at 81 and so on across the curve.

In essence the original seller would take the loss – that’s why I call it a “ratings default” – by the rating on the debt being so low, the value is lowered substantially particularly for longer dated debt. Because of the reflexive nature of markets, the ratings downgrade because the future debt level is too high has the effect of lowering the value of the future debt (implied probability of loss). This only works if the longer dated debt is actually bought back, though…


Yes, the ECB paid a discounted price for our sov. debt, but apparently the current market value is lower still. I’m assuming that it would at least be awkward for the Irish government to pay significantly more than the current price to the ECB, though maybe I’m being too trusting again. It is only a billion or so after all.

Comments are closed.