CDS pricing question!

[no_ming]

Hi, all , I see there is a question in 2014 which is about the pricing of CDS. The question asks that the pricing of CDS changed from bid price to bid-ask mid point price and how s the consequences.

The answer is the CDS issuer is required to post more collateral related to CDS.

Does anyone know the reason for post additional collateral?

 

[David Harper CFA FRM]

Hi @no_ming That sounds like a good question, but I can't find it, can you source it more specifically? In any case, please note:

• The ask price is higher than the bid (see https://www.sec.gov/answers/ask.htm), so changing from bid price to the bid-ask midpoint implies an increase in the price. For example, if bid-ask is 10-12 (i.e., bid = 10, ask = 12, and midpoint = 11), that's an increase from 10 to 11. The ask is higher because the ask is the lowest price for which a market maker will sell the security, and the market maker (typically) wants to sell for more than it will pay (i.e., the bid). The market maker buys at the lower bid and sells at the higher ask, in order to pocket the spread.
• An increase in the price of the CDS implies the protection seller will post collateral. Directionally, the CDS price increase corresponds to credit quality deterioration: if the credit reference deteriorates, the value of the market-to-market CDS increases and the protection seller needs to post additional collateral. (CDS prices move inversely with bond prices. We can imagine with hedge: if you buy the bond and purchase CDS--i..e., buy credit protection--and the bond price decreases, then your CDS value must increase to hedge). I hope that helps!

 

[no_ming]

Hello, Mr. Harper, thanks for your help.

The question asks that the trader of bank is forced to change the pricing from bid price to mid point price in order to generate more income. But from my points of view, changing pricing from bid price to mid point price does not mean that the credit reference deteriorates and the credit risk of CDS buyer is unchanged, its just a choose of pricing method, although the price of CDS actually increase.

So I am confused why the CDS seller is required to post additional collateral.

 

[David Harper CFA FRM]

Hi @no_ming

Yes, I agree that changing to mid-point does not impact the quality of the credit reference (of course!). I introduced the credit quality point because, candidly, I have to remind myself sometimes what is the implication of a CDS price increase; a price increase corresponds to deterioration of the credit reference which naturally would imply that the credit protection seller needs to post additional variation margin; aka, mark-to-market margin.

They key point is that the change (from bid to mid point point) is a price increase and that collateral is posted based on the price of the position. Like all prices, we expect the price to be a function firstly of fundamental factors (in this case, the primary fundamental factor is the credit quality of the reference) but prices are also secondarily a function of technical factors. Just like we do not expect the bond price to trade at exactly its discounted cash flow price, we are not surprised to see bonds trade rich or trade cheap due to technical factors. So, in this question a technical factor increases the price, but collateral posts based on the price. I hope that clarifies. I really like this question!

 

[no_ming]

Hi, Mr. Harper,

In another words, the position changed from bid price to mid point price is similar to the increase in the number of CDS issued by the protection seller in this case, so protection seller is required to post additional collateral~it that right?

 

[David Harper CFA FRM]

Hi @no_ming No, I consider that different because in your example there is a fundamental change in the position; although, yes, it's true that if the protection seller wrote additional protection, they would like post additional capital. In this example, neither the fundamental position nor the referenced credit change, only the price changes. Here is an analogy: say I write you an OTC call option on a traded stock with strike X and I post collateral based on our mutually-agreed-upon option pricing model using the underlying stock's bid price. At then end of today, my collateral posted (due to your credit exposure) is based on the stock's bid price. But say we switch tonight to the stock's bid-ask midpoint: the call option itself is unchanged, the ultimate exposure is really unchanged, but the option pricing model returns a slightly higher price to the option value due to the slightly higher asset price input, so I am posting slightly more collateral than if we used the bid price. So the better analogy (imo) would be to a change in the pricing model given the same underlying position. When you write "the position changed from bid price to mid point price is similar to the increase in the number of CDS," then I think that is correct if you just mean: it is similar in the sense that both directionally increase the price of the CDS.

Here is another way to look at this: if the CDS were not (daily) marked-to-market, this question couldn't really be asked. To mark the CDS is to assign a price, but there is no single "perfect" price. Why re-price the CDS daily? It could be done internally for exposure purposes, but here it is done to determine the collateral posted. I hope that helps,

 

[ami44]

Isn't the method to calculate Collateral most likely specified in some contract between the counterparties (e.g. CSA) and independent of any internal marking used by risk management?

Also, I'm always confused by this kind of questions that talk about marking by a trading desk. From my experience risk management and accounting have to mark positions independently of the trading desk and that is something regulators are enforcing. But maybe my experiences are not that universal.

 

[no_ming]

Hi @no_ming No, I consider that different because in your example there is a fundamental change in the position; although, yes, it's true that if the protection seller wrote additional protection, they would like post additional capital. In this example, neither the fundamental position nor the referenced credit change, only the price changes. Here is an analogy: say I write you an OTC call option on a traded stock with strike X and I post collateral based on our mutually-agreed-upon option pricing model using the underlying stock's bid price. At then end of today, my collateral posted (due to your credit exposure) is based on the stock's bid price. But say we switch tonight to the stock's bid-ask midpoint: the call option itself is unchanged, the ultimate exposure is really unchanged, but the option pricing model returns a slightly higher price to the option value due to the slightly higher asset price input, so I am posting slightly more collateral than if we used the bid price. So the better analogy (imo) would be to a change in the pricing model given the same underlying position. When you write "the position changed from bid price to mid point price is similar to the increase in the number of CDS," then I think that is correct if you just mean: it is similar in the sense that both directionally increase the price of the CDS.

Here is another way to look at this: if the CDS were not (daily) marked-to-market, this question couldn't really be asked. To mark the CDS is to assign a price, but there is no single "perfect" price. Why re-price the CDS daily? It could be done internally for exposure purposes, but here it is done to determine the collateral posted. I hope that helps,

Mr. Harper, i get the point that the increase in price is simply caused by a change of pricing model and its a technical issue rather than change of position of CDS. Is that right?

Moreover, I have a question on HQLA, should account receivables; gold & inventory be classified as high quality liquidity assets?

 

[David Harper CFA FRM]

Hi @no_ming Yes, on one level, that's correct: the question implies an increase in price and, if we naively assume that the price determines the collateral posted (as the question implies), then the price increase implies additional collateral is posted. As discussed, most of the price movement should be fundamental (i.e., changes in the reference's credit quality) but, as is always the case, there can be price movements due to technical factors (e.g., change the model assumptions). However, I was not frankly thinking about @ami44 's excellent point: I am not sure the question contemplates the implications of an internal CRO versus trading desk marks. Thanks,