Singapore Airlines

[Squirrel]

I think the use of the 'Mandatory' in MCB throws a curveball but on further reading, it seems that MCB is basically just a zero-coupon Convertible Bond with a Mandatory clause.

In a CB, the investor buys a 0 bond and a call option. So the investor is happy if coy does well albeit the coupons are normally lower.

In this case, for the first 9.5yrs of the MCB, if redeemed, u get a 4-6% p.a. coupon, dep when. Personally i think this sucks. In 2019, SIA issued a retail bond tranche paying 3.03%. The credit profile of SIA is MUCH worse than 2019, but the coupon is "only" 4-6%. In someways, u can say SIA is going bankrupt yet paying just 4-6%.

In yr 9.5, SIA will just need to look at what's happening. 
a. If shares > 4.84 massively, then obviously it redeems (investor dun get upside.)
b. shares are slightly >4.84, well then maybe it converts. makes no diff.
c. Shares below $4.84, then obvious convert.

As well, for the next 10yrs, if SIA shares go on a parabola i.e. start at 5, goes to 50 and back to 5 (maybe COVID-30 strikes and same thing happens), the ord share holder could sell at 50, but the MCB is just a bond (there might be some uplift due to better prospects but the correlation will not be 1)

So actually this MCB will behave very diff from a normal CB. An investor who sees the words CB and thinks "Oh, lower coupon but i get upside" is going to be very surprised eventually.

Don't really agree with your way of viewing it in terms of yield. All the while, there has been an implicit guarantee from the government to come to the rescue for SIA. So prior you are paid 3% yield for holding SIA debt (which is implied to be backed by the government/Temasek). Now you are paid 4-6% to be still backed by the government/Temasek. Risk profile is the same from that point of view regardless of how much worse the credit profile of SIA might seem.

The issue is with the mandatory conversion structure of this zero coupon bond. All that yield can go up in smoke in ten years time due to that construct as illustrated by my post earlier. That's the real pain point.