You've sunk my battleship! And seized my carrier...

There is a widely-held view that sovereign bonds don't contain the optimal terms but are slow to incorporate better ones. Right or wrong, that view has prompted many government-sponsored initiatives to reform bond contracts, such as the current plan to mandate the use of standardized collective action clauses in all euro area government bonds. These reform initiatives often fail, and the view persists that sovereign bond contracts could use some improvement.

Why do I mention this? My last post discussed how the sovereign immunity waiver in its bonds got Argentina into trouble, allowing jilted bondholders to convince a court in Ghana to help them seize an Argentine navy ship. Perhaps this was consistent with what Argentina agreed to in the bond, but the country predictably objected when the seizure occurred. The ensuing diplomatic kerfuffle highlights why enforcing jurisdictions (like Ghana, in this case) might be better off forbidding their courts to help private creditors seize a foreign country's military assets.

Below the jump, you'll find two figures showing how sovereign bonds have addressed the subject of sovereign immunity over the past two decades. As you'll see, Argentina is no outlier; plenty of bonds include waivers that are just as broadly-written.

To make sense of the figures, keep a few things in mind. First, they cover the period 1990-2010 and focus only on bonds that were (a) issued in foreign markets and (b) governed by New York or English law. (More details on the dataset here.) Also:

• Countries are immune from suit in foreign courts and their assets are immune from execution (i.e., being seized and sold).
• Creditors can sometimes overcome these immunities, but it's easier to do so if the sovereign has waived them in advance.
• Immunity from suit and immunity from execution must be separately waived. A creditor who has obtained only the first kind of waiver will have to find some other way to overcome the sovereign's immunity from execution if it wants to enforce a judgment.
• A broadly-worded waiver of immunity from execution, like Argentina's, might be read to allow seizure of all kinds of sensitive property - e.g., military, diplomatic, etc. But...
• The law might forbid the seizure of such assets notwithstanding the waiver. Thus, even if Argentina's bonds say that creditors can seize its military assets, other countries don't have to help creditors do this.
• Still, to be safe, a country that doesn't want to put sensitive assets at risk should exempt them from its waiver of immunity. Argentina didn't, and that got it into trouble.

The first figure shows that, over the past twenty years, virtually all issuers have waived their immunity from suit, and a sizeable majority have also waived their immunity from execution.
These latter waivers are generally pretty broad, like Argentina's, and could in theory be interpreted to allow creditors to seize military, diplomatic, and other sensitive assets. (Note: the law may distinguish between these asset types. For example, there are multilateral treaties that protect diplomatic and consular property.)

The next figure focuses on exemptions for military and diplomatic assets. (Some bonds include other exemptions - such as for central bank assets - but the figure omits those.) As the figure shows, bonds have increasingly exempted military and diplomatic assets from the issuing country's waiver of immunity from execution, but these exemptions remain relatively rare. Many bonds, that is, contain immunity waivers roughly similar to Argentina's.

I have written about these varied contracting practices elsewhere, but until recently I have been skeptical that the differences in sovereign immunity waivers amounted to much. Really, how likely is it that a country would find itself in a fight over whether its military assets could be seized? Er... not that unlikely, it seems. Which makes me wonder whether we'll start to see changes in drafting practices with respect to sovereign immunity waivers. If not, perhaps other bond issuers can look forward to one day playing BattleshipTM with Death.