FT Alphaville » The “autocracy risk premia” of frontier market bonds

In responding to the direct action of protestors, investors (indirectly) may never before have attached so much weight to democratic credentials, and we suspect there have been few such moments in history when a small number of institutional characteristics can explain so much variation in bond prices.

That’s in a new paper by frontier market investment shop Exotix (full paper in the usual place; hat tip Sid Verma at FT Tilt), which has applied “a model-based approach that focuses only on variables directly related to the underpinnings of unrest” and found that “democratic credentials have been a key determinant of relative frontier bond performance so far in 2011.”

There were other variables that have also punished frontier sovereigns, of course — poor labour market participation (especially amongst the young) and higher inflation chief among them.

And there were some flaws in the model:

Of course, [democratic credentials] cannot explain everything; numerous country-specific factors (and plain old volatility) are not captured: for example the model could not be expected to explain the recent rallies in Ecuador or Serbia. But as chart 1 shows, the model correctly predicts the >3% declines in bond prices in each of Egypt, Jordan, Tunisia, Morocco, Iraq and Saudi Arabia.

Some further conclusions from Exotix, beginning with a hopeful message for Egypt …

(1) We think some MENA sovereigns will bounce back once reforms are under way: Our model suggests Egypt has upside potential, and, with prices currently 11% below year-end and significant reform appearing inevitable, we retain our HOLD, but think it is a good buy for those willing to carry short-term risks. Recall that as recently as October the finance ministry was sufficiently confident to consider issuing a 100-year bond! Our model also points to Tunisia, a sovereign with broadly similar potential and short-term risk, as undervalued.

(2) What if thirst for increased democracy spreads outside the region? (There were even demonstrations in North Korea on 24 February!) Some sovereigns with relatively strong democracies have underperformed in 2011 according to the model: These include Argentina, Costa Rica and Panama (on which we have BUYs; and Ghana and Gabon (on which we have HOLDs). Conversely, our model identifies others (with relatively weak democracies) as having overperformed so far in 2011: within MENA, these include Kuwait and Dubai. Outside the region, this includes Kazakhstan, Venezuela, Vietnam, Georgia and Ecuador.

And ending with the obligatory mention of oil prices …

(3) We recognise factors beyond the scope of the model. There is no automatic link between the model results and Exotix recommendations. A prolonged spike in oil prices would mean all frontier sovereigns are affected by a global economic slowdown; and global wealth and risk will be redistributed between oil consumers and producers. In that regard, we reiterate our BUY on Iraq Paris Club loans. But elsewhere in the region, extra oil reveues and spending have been less obviously successful in placating activists (Bahrain and Saudi Arabia). We also recognise risks throughout the region that political transformation will not be smooth.