Monday, July 22, 2013

Leveraged Super Seniors of the pre credit crisis

By Jasvin Josen
This article appeared in The Edge, Malaysia on Feb 4, 2013

December 2012 wormed out another account of an interesting innovation in the financial world with the “Leveraged Super Senior” product. Deutsche Bank is being blamed of not having valued these products correctly in its financial statements, that otherwise would have recognised a loss about USD12 billion .

I would like to enlighten readers, as plainly as possible, on what a Leveraged Super Senior is and the complexities that could follow.

First, the Synthetic CDO
A Leveraged Super Senior is a credit derivative product. It is a part of a Collateralised Debt Obligation (“CDO”). A CDO is simply a basket of credit products (bonds and loans) where investors share the returns (i.e. the coupons and interest income) but also take hits when any of the credit products default. Not all investors share the losses proportionately, as the CDO basket is divided (or “tranched”) to cater for very risky investors, the not so risky investors and so on.

It is worth noting at this point, that there are also synthetic CDOs which operate in the same way, except that the basket now consist of credit default swaps (“CDS”). CDSs are traded widely in global markets, they are said to be even more liquid than the bond markets and their spreads provide a more accurate estimate of default risk in entities compared to bonds. [Readers can examine CDSs  further in my previous article (Jan 11, 2010) available on this blog]

Chart 1: Example of a Synthetic CDO capital structure

In Chart 1, let us say an issuer (in most cases, an investment bank) creates a 5-year Synthetic CDO, backed by credit default swaps of 100 reference entities of investment grade status, roughly BBB rating. The total notional of the CDSs amounts to $1billion.

Based on the analysis of the portfolio loss intensity, the issuer divides the portfolio into several tranches and calls them the equity tranche, the junior tranche, etc., as in the chart. The issuer then sells these tranches to investors as credit linked notes. A reasonably risky investor who likes the junior tranche may take up the whole tranche and pay $30m as if he was buying a 5-year corporate bond. The issuer will pay him a regular premium (or spread) for five years. Let us say, in the third year, there were some big defaults in many individual CDSs that amounted to losses of $50m in the portfolio. The losses will first be borne by the Equity Investors up to $40m. The balance of the $10m loss will be borne by the junior investor. This means that the junior investor will not get back $10m of his initial principal paid.
On the other hand, a risk-averse investor may only be interested in the senior note that is rated AAA, which pays a lower spread, but it is almost default-free, as nobody perceives losses in the portfolio to creep above 10% of the portfolio.

The Leveraged Super Senior
Now, what about the super senior tranche? This tranche is viewed as “super safe”, even safer than the AAA rated senior tranche. The issuer is only prepared to pay a tiny spread (say 0.1%) to the investor. In all likelihood, no investor will be interested. Innovation takes place to make the super senior tranche more appealing. Investors who buy this tranche will only need to pay one tenth of the notional but receive the full spread. How does this happen? In Chart 1 we see that although the super senior tranche’s notional is $850m, investors effectively only fund $85m of it. This means that the investor only needs to pay $85m principal but receive a full spread on the $850m notional. The tranche is now known as a Leveraged Super Senior.

The table shows a numerical example of the investors’ return, which is now improved greatly, from 0.1% to 1.0%.

The Trigger Points
As long as there are no defaults that crawl beyond 15% of the $100m portfolio, the leveraged super senior is out of trouble. However, there is still a remote risk that defaults could crawl above the 15% point. This may be possible for a highly correlated CDS portfolio where one default will trigger several other defaults.
Let us say that the highly correlated CDSs had begun to default and losses now stand at $300m. This means that $150m of the losses has crept into the super senior tranche. But we know that the investors only paid $85m for the tranche! The bank stands to lose $65m.

To avoid this from happening, triggers are put in place for the leveraged super senior. For example if the losses of the CDS pool touches $120m, the leveraged super senior investors will be called either to put in more collateral or unwind the structure. This is called a Loss Trigger. Another trigger called a Spread Trigger is based on the collective spreads of the CDS portfolio. If the spreads move beyond a certain point, the trigger will be breached and again super senior investors will have the option of either putting in more collateral or unwinding the investment. In essence the triggers are aimed to govern the overall health of the super senior tranche.

Gap Risk
Even with the trigger in place, there is a lingering risk for the issuer. Consider a situation where the spreads on the CDSs in the portfolio shoot up but have not breached any of the triggers of the Leveraged Super Senior trades yet. The issuer marks-to-market the leveraged super senior trades in his accounts daily. If his valuations show that the trade has a marked-to-market profit, how sure is the issuer that the collateral adequately covers this amount? This is gap risk.
In the next article I focus on the mark to market valuation of leveraged super senior trades and explain how gap risk plays an important role.


  1. Hey Jasvinder, can you elaborate on "unwinding" the structure? How does that happen, who stands to incur losses on the unwinding and doesn't the bank still have the CDS in its portfolio?

    Love your blog by the way, keep on going please :-)

    1. Thanks for your comments, Julius. My apologies for the very late reply. Unwinding the structure means that the parties agree to terminate the deal. Both parties will look at the market value of the stands to profit and the other to lose. So, say the transaction is market to market at +$1.2 billion, this may mean that the protection buyer gets $1.2 billion from the protection seller. Valuing this deal is mentioned in my next article, Part II.

      Who loses, depends on the future cash flows. In my next article (Part II) I show as the CDS spears jumps in a credit crisis, the bank who is the Protection Buyer will have a positive market value...and the investor a negative market value. Off course the bank has to make sure that the investor has posts enough collateral to cover this loss.

      Yes the bank has the CDS in its portfolio, but remember the CDS is a contract, not an asset like a bond. The bank is on the long side, i.e.. it is buying protection, hence it needs to pay out a fixed coupons every year to the protection seller. When a wind up occurs, either the bank gets a payout from the investor (if the bank stands to profit)n and cancels the CDS contract...or it finds another counterpart to stand in the shoes of the old counter party.

  2. Equity benchmarks opened higher after weakness in previous session, following better-than-expected Q3 GDP data and positive Asian cues. capitalstars