Deutsche Bank’s credit strategists presented a useful analysis piece on Friday whereby they asked what the impact would be if one of the peripheral countries experienced a “credit event” on credit default swaps that reference them, thereby triggering payouts on the derivatives. Their conclusion from the exercise being that it’s far more important to mind the cash exposures to the European periphery than the derivatives ones.
To allay fears on the CDS front, the strategists discuss the potential fallout from credit event being declared by Isda’s determinations committee. The size of said fallout depends on:
1. The quantity of net positions on CDS outstanding.
2. Recovery, hence final payout amount, as determined by the usual auction process for these things.
3. Impact of margins on existing CDS positions.
In its current form, the Greek bond swap is still “voluntary” so it wouldn’t trigger a credit event. However, it looks increasingly likely that the swap won’t stay in its current form, raising the possibility of a CDS trigger. However, regardless of this, Deutsche emphasize at the outset that it’s the cash exposures that are worth watching, rather than these derivatives. Hence one should mind Italian bond yields in particular, alongside the sovereign’s funding needs:
In terms of potential Italian funding hotspots, bills dominate Italian funding requirements in the next 3 months (€51bn) although we are less worried about them due to Italian banks’ ability to use the ECB facilities. The big worry, however, is the €90bn of bond redemptions between February and April.
So at least there’s a chance we’ll get to have Christmas holidays without being glued to our terminals watching Italian bond yields all day.
Getting back to the CDS though, FT Alphaville has already explained how gross and net notionals really work so we won’t repeat, except to note that if no counterparties to the trades fail, then the maximum amount that can exchange hands if a credit event is declared for a sovereign is given by the net notional amount. That’s the column all the way to the right in the below chart that also gives a sense of scale by putting the outstanding bonds on the left hand side:
This, of course, gives no information on the maturity profile or cost of the debt, but it illustrates well the relative dinkiness of the CDS contracts. Deutsche on this point, comparing to Lehman (emphasis ours):
…the gross notional on Lehman before the CDS event was c.$360bn with a net notional of c.$7.2bn on the day of the credit event, as shown in the FT dated 11 October 2008. Of this $7.2bn only $4.92bn came up for physical settlement according to ISDA Credit Fixings, which although large compared to any other CDS auction, was still small compared to c.$138bn senior debt outstanding, of which ISDA announced c.$73bn was deliverable. So even with Lehman, the net flows on the credit event were not as scary as they initially looked to be, when looking at the gross notional outstanding.
It is also the case that European sovereign exposures have, in general, been decreasing:
According to Deutsche, much of the new volume lately has come from counterparties entering into offsetting positions, i.e. trades that reduce their net exposure.
Ifanyone is still feeling tempted by a Lehman analogy, part of the point is that the market is leading up to this. Participants have been positioning themselves over the last months. Lehman, by contrast, was very sudden.
And the final reason not to get worried about CDS, according to the Deutsche team, is the margining that goes on between counterparties. Provided the mark-to-market is a daily calculation and margin flows accordingly, there isn’t nearly as much counterparty risk, hence smaller potential fallout from a credit event.
Dealer faces sovereign, dealer hedges – is dealer hedged?
When it comes to sovereigns and counterparty risk directly to them (rather than between banks that have a trade referencing the sovereign), it’s worth remembering that sovereigns themselves do not typically post collateral on their own trades with banks, e.g. interest rate swaps. This means that some part of a given European bank, most likely the CVA desk, will probably be buying CDS protection in order to mitigate a shortfall of collateral.
So what does it mean if there is a credit event for a sovereign that hasn’t posted collateral? It probably means that banks will be hoping really hard that their CDS are made whole. If those CDS do work as expected (i.e. without any bond swap engineering by politicians), what do you think the basis is between the payout on the CDS and the loss on the swap portfolios between the banks and the sovereign in question if said sovereign is not posting collateral?
While we are here, does anyone here know anything about the size of the derivatives portfolios that banks have with sovereigns that don’t post collateral?
We’re adding that one to FT Alphaville’s Wall of Worry. It’ll go below cash exposures to the periphery though.
Related links:
Sovereign CDS posterchildren – FT Alphaville
Irish debt office agrees to collateralise derivatives – Risk
Why Italian sovereign CDS has left the banks behind – FT Alphaville