SERVING THE QUANTITATIVE FINANCE COMMUNITY

HTFB
Topic Author
Posts: 148
Joined: February 17th, 2004, 12:47 pm

Can anyone explain why 30yr swap spreads are negative? They've been negative in both USD and EUR for a month or two now.

Martinghoul
Posts: 3256
Joined: July 18th, 2006, 5:49 am

Many reasons, but here are the main ones I can think of (btw, in EURland convention is the other way, so 30y spreads are +ve)... In EUR, it's because Depfa/Dexia and their ilk are gone, so the usual, carry-driven buyers of long-dated govt assets have disappeared. Obviously, the root cause there is the lack of funding/balance sheet.Also, there's been lots of flattening in swap land, in USD and EUR both, due to big flows from exotic gamma hedgers.Fundamentally, though, given that the govts everywhere are bailing people out left and right (UST issuance for next year is $1.5 -$2trn), why would you own govt bonds here? It makes sense to me.
Last edited by Martinghoul on December 15th, 2008, 11:00 pm, edited 1 time in total.

freddiemac
Posts: 557
Joined: July 17th, 2006, 8:29 am

Hi! liquidity is very low in the ultra long en d of swaps. It is driven mainly by hedging of exotic derivatives (eg non inversion notes). PM me if you want more details and I can send you some research.

HTFB
Topic Author
Posts: 148
Joined: February 17th, 2004, 12:47 pm

Freddiemac... so what you're saying is that there are a bunch of exotic desks that need to receive 30yr, and there is very little liquidity... so they've driven swap rates way down (vs govt's) What will cause these desks to need to pay 30yr? How illiquid are 30yr swaps? What's the bid/ask for 10M EUR? 10M USD?

AndrewDSmith8
Posts: 1
Joined: December 18th, 2008, 2:22 pm

I wrote a detailed analysis of negative GBP swap spreads at Deloitte.com. Grateful for any comments or insights.This is a big deal for insurance companies. Traditionally they've had to value their liaiblities using government bond yields. There's a long-running campaign from the big boys (the CFO forum) to shift to a swaps basis. See for example CFO forum letter to European regulators section 6.6.2. The stated reasons are consistency with banks and with market values of liability hedging instruments. The unstated reasons were lower liabilities and spoiling the smooth earnings of smaller competitors whose assets are mostly in government bonds.At year end 2008 some regulators caved in. The big insurers got what they asked for but they don't want it any more.

Martinghoul
Posts: 3256
Joined: July 18th, 2006, 5:49 am

Of course they don't... And, guess what, with the desperate straits that PFs and insurance co's are in, no regulator worth their salt is going to stand in their way, if/when they want to switch the discounting methodology. All you have to do is look at how expeditiously the Dutch were able to switch from swaps to bonds. It's the same everywhere: when the sh*te really hits the fan, nobody cares about regulations.

amophis
Posts: 1
Joined: December 30th, 2007, 6:49 pm

Seeing the same thing in canadian swaps too......the neg spread starts from CDSW5Anyone has any thoughts about that? Freddiemac.........could you please send me some research to amophis108@hotmail.comjust curious what those desks do with the 30yr.thanks

himalayan
Posts: 13
Joined: September 1st, 2005, 10:23 am

The point on the gamma hedging is an interesting one and has caused enormous volatility in the EUR long end. As an example 50 year EUR swaps did almost a 90bp move intraday earlier this month (somehwere around 5th Dec)The exotic desks have written all of these non-inversion notes (CMS spread). These notes can have a built in digital floor which pays when the curve is steep and doesn't pay anything when the curve inverts. Consequently the size and sign of the gamma has the potential to change very quickly, which is exactly what has been happening. The exotics traders have +ve gamma when the curve is steep, this they hedge by putting on a curve steepener trade. Unfortunatley for these guys the curve has flattened and they are suddenly short gamma. Now the curve is already flattening but they have to take off their steepener trades and put on flatteners which just speeds up the flattening. Also as the curve starts to steepen again they will have to put steepeners back on which pushes it back the other way. Also the point about liquidity....long dated swaps aren't too bad, this week is illiquid obviously but ordinarily could get a 100k PV01 trade at 1-1.5bps from mid.

HTFB
Topic Author
Posts: 148
Joined: February 17th, 2004, 12:47 pm

Thanks himalayan that's great info.Now... how to buy the 30yr swap spread for my personal account? hmmmm

Martinghoul
Posts: 3256
Joined: July 18th, 2006, 5:49 am

Do you have a pension? If so, I am sure you're already the wrong way around

edult
Posts: 49
Joined: October 31st, 2007, 8:24 pm

It is interesting that this negative swap spreads still exist in many markets http://www.ipe.com/news/Pension_funds_a ... _31717.php

mathmarc
Posts: 202
Joined: March 18th, 2003, 6:50 am

QuoteOriginally posted by: HTFBCan anyone explain why 30yr swap spreads are negative? Why should they be positive? The swap spread reflects the relation between the 30Y government credit and the "average over 30Y of (the average of (good bank 6m credit))" (this is for EUR, for USD, replace 6m by 3m). Personnaly I have no intuition on which one should be higher. Swap spread is certainly a credit issue, but not the difference between gouvernement and bank credit. The bank credit is rolled on short periods (6m) over a long period. At each roll only the "good" banks are in the sample; only the banks meeting a certain criterion are included. The exact rules to select which bank is "good" is decided by BBA (or its EUR equivalent). The credit is over different periods, there is not obvious reason one should be higher that the other.On the very short end (6m) the government credit is often higher (hence a positive spread); on the long end (30y) the natural selection in the swap rates (survival of the fittest, last to default) create a positive bias for the swap. Which one of the initial positive spread in government favor and the positive bias in favor of swap is dominating over the long run may vary.

Posts: 1
Joined: January 31st, 2009, 6:50 pm

mathmarc's explanation really makes sense...and its basis on fundamentals only (as opposed to market-oriented explanations such as exotics hedging, pension funds hedging, etc) is pretty attractive...but i dont understand why they've turned negative just now... i dont have any long-dated series of 30yr swap spreads, but i dont think they ever remained negative for such a long period..

HTFB
Topic Author
Posts: 148
Joined: February 17th, 2004, 12:47 pm

With respect to Mathmarc's rationale...if this is correct, than why were swap spreads positive until 2008? What he's saying has always been true.I have looked at a long time-series, and this is indeed the first time they have gone negative.I have heard freddiemac's explanation from some other people too, so I'm inclined to believe that is the real reason. mathmarc... one problem with your explanation is that a swap counterparty doesn't have any benefit from survival of the fittest/last to default of a swap... he has c/p risk to whomever he/she did the trade with on day one.

freddiemac
Posts: 557
Joined: July 17th, 2006, 8:29 am

Quotemathmarc... one problem with your explanation is that a swap counterparty doesn't have any benefit from survival of the fittest/last to default of a swap... he has c/p risk to whomever he/she did the trade with on day one.This is true but it is usually ignored as it has no direct impact on swap pricing. Counterparty risk do exist in an interest rate swap and is bigger the longer the term of the swap. Credit risk under an interest rate swap start at zero since the value of the swap is zero at initiation (a par swap). It then increases as time passes and reaches a maximum about half the term of the swap and then declines. However, different techniques exist to reduce the counterparty risk of a swap (collateral posting, rating trigers, possibility to end the swap eg every year if counterparty is not suitable). Hence, the cptyrisk is small and cannot serves as an explanation for the negative swap spreads. HTH