Let’s start by saying you’re a bondholder mulling Greece’s PSI offer this weekend. (Or you’re Maynard, after a hellish week, reflecting on the offer that you helped to create.)
Remind yourself…
a) You’ve read on the front page of the FT that eurozone creditors are turning Greece (still an OECD state!) into an economic protectorate. To unlock ‘structural’ growth, etc.
b) A protectorate built on… sand. All growth forecasts are stabs in the dark, the Greek body politic no longer responds to orders from its brain, and the weight of history is against successful reform. You know the movie by now.
c) But you have numbers to play with, and some debt relief for Greece.
d) Although… now you also see that a post-default sovereign’s vital stats on inflation, and thus real GDP growth, are a laughing stock in the pages of the Economist. Years after the default. Ghost of Greece’s future? Your future if you hold on to this stuff?
In short, the big picture beyond the highly technical bond math of PSI. Growth.
____________
Well, that’s why we find valuing the GDP-linked securities included in the Greek PSI an interesting task. It says a lot about the general tenor of the enterprise, even though the warrants only form one section of the three-part deal… and the warrants are by far the least valuable according to the analyses we’ve seen so far. Value-less, almost, as Chris Whittall of IFR reported this week.
First, you get new 30-year Greek debt with various coupons that go up over time, and which have a face value of 31.5 cents in the euro of your old bonds. The bonds amortise (pay off principal) over time. Right now you have to work out the ‘exit yield’ for them. How risky do they remain, if Greece still bears too much debt for its growth?
Second, you have some magic beans (short-term paper) from the EFSF. They’re worth some 15 per cent of old face value and will probably keep it.
Last, these warrants.
(OK, there’s more in between. The parts are all English law. This is a price consideration given it makes it harder for Greece to restructure you again. Also Greece is paying some accrued interest left over on some old bonds, which remunerates you if you’d held them for some time. It’s all a bit tangential to our point here though. Check the usual place.)
Here’s how the warrants work…
Greece issues them attached to the new bonds but you can detach them for trading around. They have the same notional value (31.5 per cent). Payments come if, and only if, Greece’s nominal GDP level is above “a defined threshold” AND the GDP growth rate continues above a “specified target”. The formal PSI release contains neither definition nor specification. So far. Your mind drifts back to Argentine official inflation stats…
You, the bondholder, are also an emerging markets vet, so you’ve now butted in thinking, ‘right, I trade Argentina’s 2005 GDP warrants and they are stinkingly profitable, so is Greece that different?’
We’ll cut to the PSI calculations of Cagdas Aksu of Barclays Capital, who dropped the critical feature of these Greek warrants amidst the Argentine comparison:
The main difference between the Argentina and Greece GDP-linked warrants is the determination of the payments: floating inflation-linked in the case of Argentina versus fixed nominal for Greece. The Argentine GDP-linked warrants make payments related to GDP performance and indexed to inflation in each year over the life of the instrument. On the contrary, the Greek warrant payments will only be a nominal 1% of the face value provided that a condition on GDP growth is met. This has important implications; in the case of the Argentinean warrant, the cash flows could be front loaded (as they actually have been given the strong nominal GDP performance). Furthermore, in the Argentine case, if a payment is not ‘triggered’ in a certain year, the payment would be pushed further out the stream of payments. On the contrary, in the Greek warrant, the missed payment would be lost.
Upside — capped. We’ll now cut further to Aksu’s overall valuation for the PSI bonds, including a minuscule price for the warrants (reflecting the payment cap and risk of a failure to trigger at all) and an exit yield on the new paper. Chart via BarCap:
Overall, the NPV of all three components is around 26.6 points with a 12% exit yield on the Greek risk. Given that IIF officials have mentioned an NPV loss on the deal of 73-74%, we believe the underlying exit yield assumption on their valuation is about 12%.
Which, as Aksu writes, serves to show how small a part of the deal the warrants are.
We don’t have to stop there though. There were different and interesting calculations from Pavan Wadhwa of JPMorgan earlier in the week, which tried to fill in the growth ‘targets’ implicit in the warrants by using IMF forecasts, and in the simplest terms, several thousand rolls of the dice.
1) We assume EFSF bonds are priced at par, although if zero-coupons are offered in the exchange the price could be modestly lower (e.g. 1-2 points, or 15-30 cents lower package value).
2) We price new Greek bonds at a flat yield assumption of 17% based on interpolation shown in Exhibit 3 above [right]. We then add €5 for the estimated premium due to English law vs. domestic law bonds (See Overview, Global Fixed Income Markets Weekly, 3 February 2012).
3) We run 10,000 Monte Carlo simulations to price the Greek GDP warrants based on the following assumptions:
a) yearly real GDP growth is drawn from a normal IID distribution with a mean growth rate that matches the target IMF growth rate in 2012-2013 and which matches the 30Y historical average growth rate from 2014 onwards (1.6%); the standard deviation matches the 30Y historical average (3%);
b) inflation matches the target IMF GDP deflator each year;
c) 1% of notional payouts begin after 3 years subject to meeting nominal GDP level and real growth rate baseline targets per the IMF’s latest debt sustainability analysis (shown in Exhibit 4 above); [below] and
d) flat yield curve of 17%.
It’s not the place here to go into why the end yield is different (but essentially, JPM have worked backwards from existing yields) though the overall haircut relative to Greece’s old debt comes out similar — 76 per cent. Again, the warrants are minimal.
Mostly it’s the thorough Monte Carlo modelling. Not the models per se but the data applied and the way GDP warrants are effectively valued as a series of GDP scenarios. And, where else do you go but the IMF targets and forecasts. It’s the most probable data-set that will be used by Greece itself when it fills in the remaining blanks on this PSI offer.
The thing is though, in general, IMF forecasts for Greece are under a cloud at the moment, given their specific application in the workings of the bailout. The IMF’s weaknesses here are probably inevitable given the uncertainty, the debt overhang still facing the country, and the past mistakes that have now made Greece a failed political economy.
Small as they are, Greece’s GDP warrants seem to epitomise that.
And for a more on how these pricing quirks and potential illiquidity of Greek GDP warrantsmight influence accounting for Greece’s new bonds on bank balance sheets — we highly recommend Chris Whittall’s recent piece in IFR.
Related links:
Greece’s bond exchange: it’s official – Felix Salmon
(More foggy forecasts…) Why we can’t believe the Fed – CFR
(And even more foggy forecasts…) So many choices – Deus Ex Macchiato