On Fri, 25 Oct 2002, Christian Gregory wrote:
> I'm not sure I understand where you get 56% and 44%
Here's the way I was thinking about it.
I start with the idea that when the country risk premium on a bond reaches 20%, the risk of default is 50%.
The "country risk premium" is how much more a third world bond yields than the yield for a comparable US Treasury bond. So if the benchmark US Treasury is paying 5%, these are paying 25%. By buying these bonds, you are getting what you'd get for a holding a treasury bond plus this extra 20% for your troubles.
(For simplicity sake, I'm thinking of newly issued coupon bonds at par. But I assume of course that all other bonds trade until their yields are equivalent and their prices appropriate to those yields.)
I'm assuming a simple universe where your holdings are made up entirely of these sorts of bonds that pay a 20% premium over treasuries and have a default risk of 50% per year. So over the long haul I'm expecting them to default pretty close to once every two years. But over this average two year period, if all proceeds are reinvested in this same type of bonds, your "premium" investment (independently of the treasury component) has grown by 44% by compound interest: 1.2 times 1.2 = 1.44.
Now if, after default, the bonds became worthless, you'd have lost 56% on your investment. Conversely though, if the bonds after default were worth 56% of their face value, your investment over that period would be worth exactly as much as if you'd held treasuries the whole time -- you wouldn't have lost a dime. And if you were a rational investor, you'd be perfectly willing to go back into the same market again. (And naturally if some investors thought the country risk was a more or less than 50%, or could get an interest rate that was a more or less than 20%, or though the bonds post default were be more or less than 56%, then they'd each think they knew a way to make money even given these conditions, and you have all the conditions you needed for a market.)
This rough equation of 50% default risk with a 20% country risk premium I get from (what seems to be the common sense of) the financial papers. The rest of it flows from the idea that investors are being rational. The conclusions I draw are that if these bonds actually do trade after default at around this expected rate, then
(a) bondholders have already been fully completely monetarily reimbursed for the risks of default through the original rate -- and they know it;
(b) there is no reason why a sovereign risk market in which default wasn't the end of the world is any harder to imagine than a liquid junk bond market; and
(c) when we force a country not to default in such a situation, but to pay such bonds in full, we are essentially coercing a poor country to pay rich bondholders extra. Sometimes almost twice what they deserve.
(d) Whereas if they restructured their bonds at this point in the manner suggested by the pseudonymous Norman Strong in "How To Default: A Primer" <http://www.leftbusinessobserver.com/HowToDefault.html>.
the countries would be able to cut their debt by almost half, and the bankers wouldn't be losing a cent they hadn't already been reimbursed for in advance.
Michael