Trading Mechanics and Notional Adjustments in Index CDS Options

# Mechanics of index CDS options


# Mechanics of index CDS options












I am looking at some documents regarding pricing approaches for index CDS options but none of them give much detail on the mechanics of trading the product. I have looked at the CDX UNTRANCHED TRANSACTIONS SWAPTION STANDARD TERMS SUPPLEMENT (published on September 22, 2014) available on the ISDA website here to get a better idea but I still have gaps in my understanding. I was hoping to ask some questions (3 questions Q1, Q2 and Q3 in bold below) in the context of a concrete example to understand the topic better.

#### Example Data

I take an option on the CDX NA HY Series 34 Version 4 5Y, traded on Fri 5 Jun 2020 with an expiry of Wed 16 Dec 2020 and a strike price of 110 (Quoting Style is Price) as an example. The various versions of the CDX NA HY Series 34 are shown in the table below. Data on the underlying reference entities that defaulted and their auctions are given in columns 6 to 11. The `starts trading` column values are deduced from the date on which I start seeing Markit spreads and Markit option prices being quoted for the relevant version. It appears to be one business day after the `auction date` of the underlying reference entity that underwent a credit event leading to the creation of the new version. Q1) I am not sure if this is a market convention. In other words, is it possible for example to trade Version 2 of the index, i.e. 2I65BRTV2, between Fri 3 Apr 2020 and Thu 7 May 2020?

| version | red code | index factor | annex date | starts trading | defaulted | default date | CERRD / EDD | auction date | cash settlement date | auction price |
| 1 | 2I65BRTL4 | 100% | Friday 27 March 2020 | Friday 27 March 2020 |  |  |  |  |  |  |
| 2 | 2I65BRTV2 | 99% | Friday 3 April 2020 | Thursday 7 May 2020 | 9FC9BF | Wednesday 1 April 2020 | Wednesday 1 April 2020 | Wednesday 6 May 2020 | Monday 11 May 2020 | 7.000% |
| 3 | 2I65BRUV0 | 98% | Tuesday 28 April 2020 | Tuesday 26 May 2020 | 2H798D | Sunday 26 April 2020 | Monday 27 April 2020 | Friday 22 May 2020 | Thursday 28 May 2020 | 7.375% |
| 4 | 2I65BRUZ1 | 97% | Friday 8 May 2020 | Monday 1 June 2020 | 6A423G | Thursday 7 May 2020 | Thursday 7 May 2020 | Friday 29 May 2020 | Wednesday 3 June 2020 | 3.000% |
| 5 | 2I65BRWE6 | 96% | Tuesday 19 May 2020 | Wednesday 10 June 2020 | UB78A0 | Friday 15 May 2020 | Monday 18 May 2020 | Tuesday 9 June 2020 | Friday 12 June 2020 | 0.125% |
| 6 | 2I65BRWX4 | 95% | Wednesday 27 May 2020 | Thursday 25 June 2020 | 46A844 | Friday 22 May 2020 | Tuesday 26 May 2020 | Wednesday 24 June 2020 | Monday 29 June 2020 | 26.375% |
| 7 | 2I65BRXL9 | 94% | Friday 12 June 2020 | Wednesday 8 July 2020 | 1E35CX | Wednesday 3 June 2020 | Wednesday 10 June 2020 | Tuesday 7 July 2020 | Friday 10 July 2020 | 1.125% |
| 8 | 2I65BRYD6 | 93% | Tuesday 30 June 2020 | Wednesday 5 August 2020 | 17B67D | Sunday 28 June 2020 | Monday 29 June 2020 | Tuesday 4 August 2020 | Friday 7 August 2020 | 3.500% |
| 9 | 2I65BRYM6 | 92% | Thursday 6 August 2020 | Friday 11 September 2020 | GMB964 | Friday 31 July 2020 | Friday 31 July 2020 | Thursday 10 September 2020 | Tuesday 15 September 2020 | 1.000% |

The payer and receiver index option quotes as of Fri 5 Jun 2020, the Trade Date, for an option on each version of the CDX NA HY Series 34 5Y with expiry of Wed 16 Dec 2020 and a strike of 110 are given in the following table.

| version | red code | payer mid (bps) | receiver mid (bps) |
| 1 | 2I65BRTL4 | 1180 | 10 |
| 2 | 2I65BRTV2 | 1091 | 13 |
| 3 | 2I65BRUV0 | 1000 | 15 |
| 4 | 2I65BRUZ1 | 905 | 20 |

If I buy a Payer option on a notional of \$1M of Version 4, i.e. 2I65BRUZ1, I will pay a premium of \$90,500 (ignoring bid-ask spread) on the third business day after the Trade Date.

At expiry, Wed 16 Dec 2020, I need to make a decision whether I exercise (assume exercise in full so that the Partial Exercise Factor is 1). As of Wed 16 Dec 2020, Version 9 of the index has a price of 108.95%. If $(1) + (2) - (3)$ in the following is greater than zero, I would exercise.



- Value of index CDS that is exercised into: value of a \$1M buy protection position in Version 9 as of Wed 16 Dec 2020.

- Settlement Payment = Strike Adjustment Amount - Accrued Amount: these are defined in the ISDA document referenced above. When Quoting Style is Price, the Strike Adjustment Amount is 1 minus strike price, i.e. 110 here, multiplied by the Adjusted Swaption Notional Amount. The Accrued Amount is the usual accrued from either the effective date or the prior fixed rate payment date to the expiry referencing the Adjusted Swaption Notional Amount.

#### Q2) Front end protection

For the example here, what are the relevant names for inclusion in the front end protection payment? If I look at the option Trade Date of Fri 5 Jun 2020 versus the Credit Event Resolution Request Date, i.e. `CERRD` column, of the various defaulted entities, I would include only the reference entities `1E35CX`, `17B67D` and `GMB964` in the front end protection calculation. These are the only entities with a `CERRD` after the option Trade Date of Fri 5 Jun 2020. Or, is it any entity that causes a payment under Version 4 and any subsequent Version up to the expiry date? This would add reference entities `UB78A0` and `46A844`, already known to have defaulted on Fri 5 Jun 2020, to the list of reference entities contributing to the front end protection.

#### Q3) Adjusted Swaption Notional Amount

From $(3)$ above, the Adjusted Swaption Notional Amount makes sense to me if it is the original option notional, i.e. \$1M, multiplied by the index factor at the expiry date, i.e. 92%, giving \$920,000. Is this the correct interpretation?

I ask because the Adjusted Swaption Notional Amount is also referenced in the ISDA document when explaining the Strike Adjustment Amount for the case of Quoting Style is Spread. In particular, the payment is $N^* (s_K - c) R(s_K;t_e,T)$, where $N^*$ is the Adjusted Swaption Notional Amount, $s_K$ is the strike spread, $c$ is index CDS running coupon and $R(s_K;t_e,T)$ is the risky annuity from expiry to index CDS maturity using a flat curve at the strike spread. I have some papers emphasise that this notional $N^*$ is the notional at the option trade date i.e. 97% * \$1M = \$970,000 here. For reference, the papers are A CDS Option Miscellany, Richard J. Martin, May 24, 2019 and ICE Credit Index Option, Sep 2018. Is this a recent change to the ISDA 2014 document referenced above that I am missing or am I misinterpreting something here?