Serving the Quantitative Finance Community

 
User avatar
sme123
Topic Author
Posts: 0
Joined: October 5th, 2010, 7:34 pm

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 1:08 am

We want to terminate a 10yr interest rate swap that has a negative MTM to us (we pay fixed vs 3M Libor). The swap sales guy told us that we have to pay an additional 5 bps over the mid-market rate because the bank discounts swap terminations at the FF curve rather than the Libor curve. This makes absolutely no sense to me especially since the bank is receiving the cash termination value which seems like a funding benefit and not an additional cost. Is there any legitimacy to the the dealer's argument about needing to discount a cash payment to them at below Libor or am I being misled?About 2 years ago we terminated a swap that had a positive MTM to us and the dealer said he had to reduce this amount because of the bank's funding cost. This argument made sense to me but when the bank is receiving cash I don't see why they need to discount below Libor.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 7:34 am

Yes, welcome to the wonderful world of discounting/collateral/etc. Essentially, while some banks will use this as a way to extract bid/offer (i.e. you get charged regardless of whether PV is positive or negative), the "correct" procedure IS, in fact, to discount unwind PVs using OIS. As far as I am concerned, everyone I speak to is doing this.
 
User avatar
sme123
Topic Author
Posts: 0
Joined: October 5th, 2010, 7:34 pm

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 12:22 pm

Conceptually, I don't understand it though. What is the rationale for discounting a termination payment at OIS? The bank is receives the cash termination payment from me today that it otherwise would have received in the future. If anything the bank should have a benefit because its funding costs (the bank borrows at ~ L + 100) have gone down due to the receipt of cash. I do understand why a bank would need to discount at OIS if was posting collateral where it only earns Fed Funds. But this is not the case.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 12:57 pm

I think this discussion has taken place in many threads here, so, if you don't mind, I'd rather not revisit it. Let me just say that, as far as I am aware, this is the current mkt convention. Again you will find all sorts of discussions of these subjects here (incl stuff on exchanges opting for OIS discounting).
 
User avatar
sme123
Topic Author
Posts: 0
Joined: October 5th, 2010, 7:34 pm

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 7:02 pm

I honestly think the argument that the bank needs to discount the incoming swap termination payment at FF is complete BS. Thanks very much for your thoughts though - I'll try to find something about this on other threads.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 7:19 pm

Ha, you just have to make sure you never ever lose money on swaps and swap legs of any trades you decide to do... Let me know if you have trouble locating the relevant threads.
 
User avatar
TinMan
Posts: 21
Joined: September 21st, 2006, 9:42 am

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 8:51 pm

It makes sense if you consider the swap is in the money so you usually have to place the MTM as collateral with the bank and they pay you FF on that.So they can realise the value over the term left remaining costing them FF, or today.So you discount with that rate.If you're not a CSA counterparty then that opens a whole other can of worms.
 
User avatar
sme123
Topic Author
Posts: 0
Joined: October 5th, 2010, 7:34 pm

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 10:43 pm

There is not a collateral posting requirement under this swap. The way I see it, when the hedge is terminated the bank gets the cash and offsets his original market hedge by doing an offsetting swap in the market. Getting cash upfront and paying it off over time (via the offsetting swap) is effectively a loan at Libor flat which represents about 100 bps of savings relative to where the bank can normally borrow. So it seems like the bank gets a benefit from the swap termination, not an additional cost. The can of worms is opening I guess...Thanks for any additional thoughts or insights. Very frustrating.
 
User avatar
TinMan
Posts: 21
Joined: September 21st, 2006, 9:42 am

Discounting swap termination below Libor - does this make any sense?

January 13th, 2011, 11:27 pm

Well you could theoretically justify a lot of things by what the bank could be doing on the other side.Maybe they should look at what your collateral situation is and see they're not getting any funds at FF from you and adjust accordingly.But nobody can realistically view their book with that kind of granularity.Your cashflows come out in the wash with everyone elses and presumably most of their counterparties are CSA so the policy would be to mark against OIS.If they were to change the discount rate on termination there'd be a hole in P&L to be explained.Also, if they are cutting out a swap on the other side they will have to cough up the OIS discounted amount.So charging you the same is appropriate from that point of view.The way I would look at it is that the market rate you see is for CSA counterparties really.If you are not collateralized you're a risky counterparty which means you should also be charged a credit spread.My comment about the can of worms is that potentially banks will start differentiating between customers who post collateral and those who don't.You weren't charged it when the deal was done but the world has changed now.Think about it this way, your position is underwater but you haven't had to post any collateral.So that's a factor in your favour, but by tearing it up you're forfeiting that advantage.An alternative might be to enter into an offsetting swap rather than terminating.The usual caveat is that you'd be posting collateral, but if that doesn't arise it might work out cheaper.Like Martinghoul says there have been a lot of threads on these issues, especially a couple of years ago when it was all up in the air.These are two that spring to mind, but do a search on things like 'basis' and user 'cpulman' and you'll find more.Thread 1Thread 2
Last edited by TinMan on January 13th, 2011, 11:00 pm, edited 1 time in total.
 
User avatar
Blazes
Posts: 0
Joined: September 6th, 2004, 7:36 am

Discounting swap termination below Libor - does this make any sense?

November 14th, 2011, 2:50 pm

I have been thinking about this and have tried to clarify my thoughts by breaking it down into different pieces. If we assume that the bank's hedging options for the non-collateralised trade are IB OIS and LIBOR swaps which require collateral to be posted and such collateral attracts the OIS rate in the ccy in which the trade is denominated (LCH model as I understand it) and fixed rate loans/deposits with its "treasury desk", it will initially have a theoretical hedge in each tenor point for all 3 curves up to and including the maturity of the trade. In this situation the price to the c/p will be the 1. the PV of the forward Libors (adjusted for OIS discounting) discounted using the funding discount curve divided by the sum of funding curve discount factors for the fixed rate payment dates.2. plus or minus an adjustment for expected funding costs this. Should be relatively minor and will depend on the direction of the trade, the assumed dynamics underlying the interest rates' evolution (including the correlation between the three different curves) and the level of interest rate volatility3. plus or minus CVA depending on whether the bank is receiving or paying. (I ignore DVA)When it comes to a buyout the approach should be similar except that now @ 1 we are solving for a PV and @ 3 the CVA should always be in the clients favour. Departures from this using OIS discounting mean that the bank will have a profit in economic terms (that it discounts NC trades using OIS is irrelevant and an error except by coincidence) Anyone have any further thoughts on this? BTW I am hoping to put an Excel/VBA example together (subject to corrections as a result of responses to this post!) if anyone would like a copy please PM me.
 
User avatar
Gmike2000
Posts: 0
Joined: September 25th, 2003, 9:49 pm

Discounting swap termination below Libor - does this make any sense?

November 15th, 2011, 2:41 pm

QuoteOriginally posted by: sme123We want to terminate a 10yr interest rate swap that has a negative MTM to us (we pay fixed vs 3M Libor). The swap sales guy told us that we have to pay an additional 5 bps over the mid-market rate because the bank discounts swap terminations at the FF curve rather than the Libor curve. This makes absolutely no sense to me especially since the bank is receiving the cash termination value which seems like a funding benefit and not an additional cost. Is there any legitimacy to the the dealer's argument about needing to discount a cash payment to them at below Libor or am I being misled?About 2 years ago we terminated a swap that had a positive MTM to us and the dealer said he had to reduce this amount because of the bank's funding cost. This argument made sense to me but when the bank is receiving cash I don't see why they need to discount below Libor.Deal with it. It is the way the swap market works now. Sorry to sound so harsh, but a swap can no longer be discounted naively at libor. Where have you guys been trading the last 2 years, on Mars?