European leaders are toying with the idea of buying back Greek debt on the market. At first sight, the idea might look charmful: Greek bonds are currently being traded with huge discounts. If you spend one euro, you can buy Greek debt with a face value of three to four euro. Hence, a buyback could be a relatively cheap way to reduce the debt burden of the stricken country, couldn’t it?
ECB’s Jörg Asmussen does think so:
“We need a package of measures to close the financial gap that will include a substantial reduction of the interest rates and a debt buy-back by Greece.”
However, some economists like Klaus Adam, a Professor at Mannheim University, see the idea highly critical.
In a guest post for the WSJ’s Real Time Economics blog, Adam comes to the conclusion that a buyback involves “a huge wealth transfer from the borrower to the lender” and would be “hurting Greece”.
Norbert Häring, my former colleague at Germany’s Handelsblatt, recently made a similar point.
The core of the criticism goes back to an old paper written by Ken Rogoff and Jeremy Bulow in 1988 entitled “The Buyback Boondoggle”.
Adam illustrates the issue using a simplified numerical example:
“Consider a country which can repay 1 euro but has outstanding debt with a face value of 2 euros, so that all debt trades at 50 cents on the euro. Now suppose that the country – through whatever operation – receives an additional 50 cents and uses it to repurchase outstanding debt with a face value of 1 euro.
(…) The additional 50 cents of resources that the country has spent have been transferred in their entirety to the country’s lenders, who now receive back 1.5 euros instead of just 1 euro.”
You can’t argue with the numbers, of course. However, I wonder if I really share the conclusion that this outcome is necessarily bad for the country. Adam compares the direct costs of a sovereign default (1 Euro) with the buyback (1.5 Euro) and concludes:
“It is hard to see that the ECB fails to understand such a simple but powerful argument, which raises the question of why it sees value in pursuing the bond buyback option.”
Maybe I’m missing something but I have a hunch that things are slightly more complicated.
First of all the argument does not take into account that the buyback – which costs 0.5 euro in the example – avoids a sovereign default which in the real world is associated with additional economic costs and repercussions, as Juan Cruces (Universidad Torcuato Di Tella, Argentina) and Christoph Trebesch (Hertie School of Governance, Berlin) pointed out last year. This is how I summarised their key results last year:
“After a significant haircut, a government will find it extremely difficult to borrow new money on the capital markets, and if they get access again, they’ll have to pay prohibitively high interest rates.”
In Adam’s example, the buyback not only avoids the default but also leads to a situation where the governments finances are sustainable again. After the buyback the public debt is exactly as big as the country’s ability to repay, as he points out himself:
“The country will then be left with 1 Euro of outstanding debt and can still pay back 1 euro, so that the remaining debt now suddenly trades at its face value.”
If you take other things into account, the argument against the buyback looks even less straightforward. For instance, a full-blown default most probably would lead to Greece leaving the euro zone (“Grexit”).
The collateral damage for the Greek banking system and real economy would be huge. There are strong arguments that this could easily trigger a complete meltdown of what’s left of the Greek economy. Even if this might be exaggerated, there are clearly additional costs that are not addressed in Adam’s example.
And then there is politics. Like it or not, but the rest of the euro zone has apparently resolved to keep Greece within the single currency. Most economists are convinced that avoiding a full-blown sovereign default is a precondition for this.
If this is the case, the “cheaper” option in Adam’s example ceases to be a viable alternative. In that case the relevant question is to find the most cost-efficient way of helping Greece to sort things out. Compared to just mutualising the non-sustainable part of the debt (1 euro), a buyback suddenly becomes significantly less expensive (0.5 euro).
A lot of open questions remain of course.
An important issue is if a cheap buyback is possible in the first place. If financial markets take the increased net value of the bonds into account before the buyback is completed, the tool becomes less efficient because bond prices rise. In fact, this seems to be precisely what’s happening at the moment, as Barclays economists point out in a paper quoted by Zerohedge:
“Previous debt buybacks show that secondary market prices rally significantly up to the actual debt buyback offer. As a result,by the time the buyback occurs, debt relief due to price action is much lower than originally expected.
As such, we think a large part of the rally in Greek bonds that has occurred since the end of the summer is due to debt buyback anticipation-related buying, which resulted in average Greek strip price appreciation of 75% since mid August. The average market value of €63bn of Greek debt was €10.2bn in mid August; however, at 35 cents on the euro, the post debt buyback market value of €30bn of nominal debt will still be around €10.5bn.
Therefore, this typical price appreciation in anticipation of the debt buybacks in most cases makes debt buybacks a creditor-subsidising experience.”
However, there still are big uncertainties attached to the Greek situation.
For instance, while the Grexit has become less likely it still is possible. This definitely will dampen the market’s appetite for Greek bonds even if a buyback programme would be started. Investores still face the risk that they pay euros for a Greek bond but will be paid back in a new Drachma (if at all).
Additionally, in the real world Greece’s ability to serve its debt is not an objective fact that is known to every player on the markets. Traders have to guess, and their assessment might differ. While the announcement of a buyback would definitely make Greek bonds more expensive, both factors might curb the increases.
Another thing that is not completely clear to me is how the funds used for the buyback programme would affect the level of Greek debt. If the money was just lent to the country, in Adam’s example the level of debt after the buyback would still be 1.5 Euro rather than 1 Euro. In that case, the debt situation would still not be sustainable. The potential benefits of the buyback would be smaller.
On the other hand, the interest rates Greek would have to pay for the new loans probably would be subsidised and hence be lower than the interest payments for the old bonds.
I also don’t fully understand how interest rate dynamics and compound interest come into play. Adam told me that the interest rate burden does not has to explicitly addressed in the example because it was mirrored in the country’s ability to repay. However, if this is the case, doesn’t this mean that the ability to repay increases if the level of debt is reduced either by 1 (if the money for the buyback is a gift) or at least by the interest-rate differential (if the money is a loan)?
As I’ve said before, I’m probably missing something. A bond buyback certainly isn’t the silver bullet that would be the end of the Greek debt problem. However, I’m tempted to think that the whole idea is not as daft as simple models suggest.