5 ms·
I'm not sure what comparison you're trying to make. These derivatives are typically used modify the cashflows of those instruments for different issuer/investor
by ddeck 10y ago
I'm not sure what comparison you're trying to make. These derivatives are typically used modify the cashflows of those instruments for different issuer/investor preferences.
The vast majority (>90%) of that OTC derivative figure is interest rate andFX swaps and the gross market value of those swaps are less than $15 trillion. The aggregate notional principal has also been declining for the last several years.
Here's an example trade:
I'm a European company and I've determined there's demand for my USD fixed rate bonds in America, but I need EUR to invest in my business and my profits are linked to interest rates so it's less risky for me to pay a floating interest rate.
So to reduce my risk, I need to swap the bond cashflows into EUR and the fixed rate payments into floating rate payments.
I issue $100 of bonds to US investors, and then I execute a $100 notional fixed-floating FX swap with a bank. That bank then executes a $100 notional fixed-fixed FX swap with another counterparty and a $100 notional (EUR equiv) rate swap with a different counterparty.
So we have $300 notional value of OTC swaps and a $100 bond. And if we take out the central counterparty, the three swaps actually net into a single $100 swap.
- TheOtherHobbes 10y agoWhat happens if any part of the chain breaks?
- ddeck 10y agoParts of the chain break regularly. Individuals, companies, and even banks go bankrupt. These outcomes are expected and swap agreements are protected by collateral under an ISMA agreement to mitigate the potential losses to counterparties should the other counterparty default. Now it's reasonable to argue about the amount and type of collateral required (which is broadly the topic of the article), but aggregating all of the notional doesn't produce a meaningful number, because it doesn't reflect the underlying risk. Even without the sort of hedging I described, it's impossible to draw any meaningful understanding of the risk from the notional number. Consider the following two contracts: Party A sells $100 protection on a corporate to party B (i.e. CDS) Party A enters a $100 1Y USD interest rate swap with party B, paying floating. If the underlying corporate defaults with no recovery in the first case, it's quite possible party A will owe $100 to party B In the second case, there is a realistic ceiling on what can happen to USD interest rates within the next year. Let's imagine the impossibly unlikely scenario where the fed decided to raise rates to 5% overnight one day after the contract was signed. Even in this situation, the value of the swap (i.e. loss to party A) is still likely only around 5%.