Skip to content
All library documents

CVA Netting Sets Depend on Legal Counterparty Agreements

Article Quant Q&A · Author: eddiewould

Summary

The document discusses how trades are grouped into netting sets for calculating credit valuation adjustment. A netting set consists of trades that can legally be offset against one another if a default occurs, so the applicable agreements must permit the trades to be netted together.

The response says netting sets are generally supplied as inputs rather than generated by a universal algorithm. If they must be inferred from trade data, grouping by counterparty is a possible starting point, with the counterparty understood as the relevant legal entity. Trades involving separate entities should not be assumed to offset merely because the entities are related or participate in different legs of a transaction. The answer is brief and caveated: it is not legal advice, and it does not provide a full procedure for determining enforceable netting rights. In practice, legal agreements and entity identity are central inputs to the grouping.

Key ideas

  • A netting set groups trades that can legally be offset following a counterparty default.
  • The governing netting agreements must be compatible across trades in a set.
  • Netting sets are usually provided as inputs to CVA calculations rather than derived by a general algorithm.
  • Counterparty legal entity can guide initial grouping, but enforceability requires legal review.

Tags

Full text
# How are netting sets determined for CVA calculation?


# How are netting sets determined for CVA calculation?












In his book, Gregory describes a netting set as

> a set of trades that can be legally netted together in the event of a default

Obviously, the netting agreements (as per ISDA master agreement) need to be compatible.

But are there other requirements?

Is there a generally-accepted algorithm for producing netting sets from a portfolio?

## Answer by SRKX (score 1, accepted)

https://quant.stackexchange.com/a/19474

I believe netting sets are usually provided as inputs to the algorithm in most-cases.

If you were to kind of "guess" netting sets given different trades (in general) data, you could start by grouping them by counterparty. I'm not a legal specialist but my understanding is that counterparty here is to be understood as "legal entity" or something like that, and the fact that one entity defaults doesn't necessarily entitle you not to pay what you may own to some other entity on another leg of the deal.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.