Secured bond trading at the same level as unsecured bonds
Collateral is very valuable - need draconian assumptions to result in an impairment for the unsecured bonds
12 month gross return potential of up to 15%
You can construct a relative value trade (long secured / short unsecured) if you don't have a strong view on the credit
Transocean and Shell have entered into a 10 year contract whereby Transocean is contracting a drill ship to Shell for a fixed day-rate. Shell is using the vessel for its Stones project in the Gulf of Mexico. According to Wood Mackenzie this project has OpEx of ~$30/boe and peak production in 2021 (production commenced in 2016). Given the upfront capex has been spent a large and sustained reduction in oil prices would be required for the project to become uneconomic.
In October 2016 Transocean issued $600m 7.75% senior secured bonds through an SPV called Transocean Phoenix 2 Ltd. The bonds currently trade at ~106% of par. Key features of the bonds:
The secured bonds are guaranteed on an unsecured basis by TransOcean Ltd. & Transocean Inc. (so recovery in the event of default can functionally never be worse than Transocean unsecured credit);
The Collateral the bond is secured by includes:
(1) the Deepwater Thalassa: a 7th generation drill ship with a build cost of $890m (~1.5x the bond principal); and
(2) a 10 year drilling contract with Shell which commenced in Feb'16 and expires in Feb'26. Based on the contract's day-rate the life-of-contract revenues from this contract are ~$1.7bn (~2.8x the bond principal)
The bond has scheduled mandatory amortisation: 5% of the outstanding principal is repaid to holders every 6 months (i.e. an investor's capital at risk decreases & coverage from collateral increases every 6 months)
There are other creditor protections including: the existence of a debt service reserve, prevention of distributions to the parent in the event the parent's leverage exceeds a predetermined level (i.e. cash is trapped at the SPV for the benefit of secured bond holders) and the immediate full repayment of the bonds in certain circumstances
In summary, the 7.75% secured bonds cannot have worse recovery than unsecured credit in default, will almost certainly have higher recovery (unless you believe both the vessel and the Shell contract are worthless) and (unlike Transocean's unsecured bonds) amortise. As such, I contend the secured bonds are superior to Transocean unsecured credit.
Despite the characteristics described above, the 7.75% secured bonds trade at an equivalent valuation to Transocean unsecured credit. Specifically, if one discounts the cash flows (principal + interest) for the secured bond using the credit spread for the Transocean unsecured credit it results in a bond price of ~106% of par. Why? I think because the market looks at this as a "2024" bond and as such compares it to longer dated unsecured instruments (e.g. the 7.75% secured bond trades at a lower spread than RIG's 2023 bonds as one example). However (because of the principal amortisation described above) the duration of the bond is only 4 years. Transocean short-dated credit trades at significantly lower spreads than long dated credit (the curve is very steep) because short term liquidity for the company is reasonably good. This is very important as far as the 7.75% bonds are concerned because an investor receives cash (Principal + Interest) equivalent to ~9% of face value every six months. This is a fundamental aspect of this instrument that I think is overlooked.
What would recovery for the secured bond be in default?
Perhaps one way to think about this is what would have to happen in order for recovery to be less than Par:
(1) Default would likely need to occur quickly (because of the principal amortisation described above); and
(2) Transocean would need to default; and
(3) Recovery on Transocean unsecured credit would need to be less than Par (because otherwise the secured bonds would also recover par due to the guarantee); and
(4) Shell would need to default on its contract and recovery from Shell would need to be very low; and
(5) The value of the vessel (the Deepwater Thalassa) would need to be very low
Putting some theoretical numbers around each of these:
(1) Transocean defaults in 1 year (note: 10% of the principal has been repaid + interest equivalent to 7.6% of par has been received in that year);
(2)+(3) Say recovery on default of Transocean unsecured is 10% of par (note: unsecured bonds out to 2021 are all currently trading at >100% of par);
(4) Say recovery on default from Shell on the contracted payments is 20% of remaining amounts due (note: creditors would have a legal contractual claim to Shell for 100% of payments owing);
(5) Say recovery on the Deepwater Thalassa is 20% of book value
In the scenario described above (which I think is quite draconian) recovery on the 7.75% unsecured bonds would be approximately 80% of par (of which ~17% would be returned in cash payments before default in the form of principal and interest payments). For the abundance of clarity, in this scenario Transocean unsecured credit is recovering 10% of par (important for the relative value construct described above). By way of sensitivity: (i) if you push the default date out by 1 year recovery improves by ~10% of par (i.e. to 90%); and (ii) every 10% increase in recovery assumption for either the vessel collateral or Shell recovery is also equal to a ~10% increase in recovery assumption.
** As an aside, Transocean currently has liquidity (in the form of cash) to meet bond payments for at least the next 2-3 years (note: cash actually exceeds maturities out to 2020 but there will likely be negative free cash flow over the coming years which is why I say 2-3 years). In any event a default in the next 12-24 months seems unlikely (2 year CDS implies <5% probability for 12 months and <10% probability for 24 months).
What's fair value / Return Potential?
I would argue that this is really Shell risk in drag (i.e. if Shell keeps paying on the contract the SPV will remain current on its payments to bond holders and the principal amount will amortise regardless of what happens to Transocean). As such, I contend the pricing for this bond should be Shell credit risk plus a spread. If you discount the cash flows for this bond at Shell credit + 150bps it implies a 1 year forward price of ~$115. You also get $7.75 of interest over 12 months for a total gross 12 month return of 15% versus the current price. If you think the right spread is Shell +250bps then its 11% gross return. At Shell +350bps its 8% gross return.
Alternatively, if you want to anchor to Transocean unsecured credit then I would argue that this bond must trade at a negative spread to Transocean unsecured. If you discount the cash flows at 200bps inside Transocean unsecured then the one year forward return its 13% gross.
However you frame it the returns should be decent from current levels, especially in light of the limited downside risk. If the trade construction is long secured vs. short unsecured then a deterioration in the credit should result in a material divergence in the relative prices of secured and unsecured prices. A deterioration in transocean credit will be very positive for that trade construct.
I do not hold a position with the issuer such as employment, directorship, or consultancy. I and/or others I advise hold a material investment in the issuer's securities.
Market recognition that this is not a "2024" bond & pricing it appropriately
Cash flows received by investors (amortisation payments in particular)
Default or credit deterioration - if you construct this as a relative value trade a default will cause (long) secured to materially outperform (short) unsecured