Swaps and MBS

April 30th, 2009 10:09 am | by John Jansen |

Swaps spreads are mostly tighter this morning. Two year spreads are one basis point wider at 56 1/2. Five year spreads are a basis point tighter at 53. Ten year spreads are 1 3/4 basis points narrower at 9. Thirty year spreads are 3 basis points tighter at NEGATIVE 44 1/2.

Mortgages are a tad weaker this morning but the weakness is not significant.

I did have an interesting conversation about MBS with a trader this morning. he is a friend of the blog from its earliest days. He read my post this morning regarding duration extension and convexity hedging and disagreed with my analysis.

The salient point which I failed to grasp inmy analysis is that the convexity of a mortgage portfolio is linked to the price of the mortgage and not the yield on the 10 year Treasury.

He notes that the prices of mortgages are at the highest prices since the dawn of history ( which began with the publication of the book Liars Poker) and unless there is appreciable price change there those who hold mortgages will not feel pressure to hedge.

An interesting point which I had not considered.

I still believe that we are drowning in a sea of Treasury supply and it will take higher rates to accomodate the refunding next week.

Be Sociable, Share!
  1. 13 Responses to “Swaps and MBS”

  2. By troy on Apr 30, 2009 | Reply

    convexity is the rate of change of duration with yield not price. on MBS higher yield equals longer duration therefore more risk.

  3. By troy on Apr 30, 2009 | Reply

    for anyone interested in reading further

    gotta run.

  4. By John Jansen on Apr 30, 2009 | Reply

    but if the dollar price of the mbs is not changing (because the federal reserve is a giant financial vacuum cleaner) how do we get duration change?

  5. By Michael Krause on Apr 30, 2009 | Reply

    Yes. The 30 yr mortgage delivered from Fanniemae is 4.45% this morning. That with a 4.03% 30 yr and a 3.12% 10 yr. The spread between the 30 yr and the mortgage has never been so narrow. I guess mortgage hedging doesn’t need to take place as long as the Fed’s agents are busy buying mortgages. They probably have $800B left.

  6. By Bman on Apr 30, 2009 | Reply

    I agree that servicers are not in imminent danger of having to pay large blocks. However, spec traders have always tried to guess and “front-run” where those possible triggers could be – similar to the trade yesterday.

  7. By Wallie on Apr 30, 2009 | Reply

    So how can a small time investor best capture a trade that bets on a widening of the long bond to 30yr mtg spread? Is there an ETF combo trade that is relatively clean to this spread?

  8. By Bman on Apr 30, 2009 | Reply

    Don’t know how “clean”, but you could buy TBT vs. weighted amount of MBB or similar variation as close to DV01 neutral as possible.

  9. By vol-trader on Apr 30, 2009 | Reply

    so basically there won’t be a paying wave as long as the fed is holding up the mtg bonds, the spread will just continue to tighten?

    seems like something has to give here. the fed isn’t stupid enough to buy MBS through the 30yr rate are they?

    why wouldn’t they hold off on the MBS puchases and shift some of that money to 30yr treasuries instead?

    my head is spinning now…

  10. By Brian on Apr 30, 2009 | Reply

    Thats going to be near impossible to set up a DV01 neutral trade using ETFs…you can’t just use average duration in the calculation and therefore dont know how well that would capture the spread changes.

  11. By Bman on Apr 30, 2009 | Reply

    I know – just trying to give him a generic way of looking at it.

  12. By Brian on Apr 30, 2009 | Reply

    Yeah I gotcha. I like the idea, actually never thought of doing that, in theory it would make things a whole lot easier…

  13. By MFL on Apr 30, 2009 | Reply

    I think the original point was that the price change MBS (and so the change in the rate used to calculate its duration) has been very minimal despite the big moves in the long end treasury rates due compression in the MBS-treasury spread.

    This means that the duration is going to stay roughly flat as long as the MBS-treasury spread continues to compress.

    Vol-trader, I was curious about that. How much scopoe is left for MBS tightening given spreads in the 80s?

  14. By Wallie on Apr 30, 2009 | Reply

    Thanks all for the spread trade suggestions/comments.

Post a Comment