In response to the growing violence in Syria, CGD’s Kim Elliot and Owen Barder have revived the idea (here and here), not just as a method for preventing excessive debt, but as a way of putting more pressure on the current regime – if Assad’s remaining lines of credit are dropped, then we might expect his position to crumble even faster.
Let’s break the idea down a little further. There seems to be an implicit model working in the background here, and I think it looks something like this: a rational investor knows the best deal it can get from the Galactic Empire is a rate of return of r¬†on a loan. Based on this, the investor decides how much to loan, conditional on its outside options. The investor knows that, even if the Empire falls to the Rebel Alliance, it is still guaranteed its return¬†r.
Now, assume that an outside party has declared all future debt incurred by the Empire as¬†unenforceable. If the Rebel Alliance wins (with probability P), then the investor won’t receive any return at all, and now there is also a smaller chance the Empire will pay them back even if it remains in power (this probability of repayment¬†E). The investor’s expected rate of return is now significantly lower [(1-P)Er < r], so it will disinclined to lend as much – and might even choose to disengage completely.
This is all pretty intuitive and sensible – although there a few reasons why it might not be that straight forward. These should be thought more as possible (general) caveats rather than criticisms:
- Expectations over the value of P¬†are important here – we can expect the impact of an odious debt declaration to be stronger as it looks more and more likely the Rebels will win. Fortunately, the Syrian army might¬†already¬†be signalling its views on P.
- It’s very likely that the probability the Empire stays in power is a function of the amount of money lent to it – that is, potential investors know that they can directly affect¬†P¬†when choosing how much to lend. While most assumptions over the functional form of¬†P¬†would lead us to the same conclusion, I’m still a bit worried that there could be identifiable thresholds here (if we lend them enough for a Death Star, the probability of winning converges to 1). This is why it is really important that odious debt declarations only affect future contracts (as the CGD suggests) – we don’t want investors given the incentive to prop up bad governments so they can avoid sunk costs.
- Another assumption is that there is symmetry in the way contracts are enforced: in reality, the financial and legal institutions responsible for¬†international; contract enforcement are much more likely to be binding when a¬†legitimate¬†government takes over and wants to rid itself of odious debt, but the same is not¬†necessarily¬†true if the illegitimate government wins. When dodgy governments (or firms) trade with others of the same ilk, different, more informal mechanisms are likely¬†underlying¬†enforcement. These types of governments are inherently riskier anyway, so potential investors might already be relying on these alternative mechanisms. Again, these wouldn’t reverse the expected results, but they might dampen it.
- Related to this point – we assume that countries that do business with awful regimes are only interested in profit. Once you add in preferences for that regime staying in power (democratic governments don’t usually buy as many weapons + it’s good to have another evil dictator in the¬†neighborhood), the picture gets even murkier.
Caveats aside, I’m excited that this idea is finally getting the attention it deserves.