1
$\begingroup$

A colleague mentioned the following: She wanted to look at some bonds in Reuters/Bloomberg to see if they if they "correctly break-out the swap curve and z-spread".

Would anyone understand what she is trying to say/do? Would be much appreciated

$\endgroup$
2
$\begingroup$

OK I don't understand who "they" are supposed to be. Does "they" refer to Bloomberg/Refinitiv (your colleague is questioning whether Bloomberg/Refinitiv are right for this bond), or "they" refers to some other analytics, perhaps internal, that she wants to compare to Bloomberg/Refinitiv to see if they're in line?

Generally, a Z-spread calculation is pretty straightforward: one has to project the cash flows of the bond and shift the swap curve up (in parallel) until the present value of the cash flows discounted with this shifted curve matches the observable (dirty) price of the bond.

Things that can possibly go wrong with a Z-spread calculation include:

  • the bond's cash flows are not projected correctly because its indicative data is not set up correctly. For example, the bond might be amortizing, but the analytics don't know this and project cash flows as if it were a bullet bond. Or the coupon rate or the maturity etc might be wrong. (This happens on Bloomberg surprisingly often.)

  • the bond has some features that make Z-spread not the right calculation. For example, if the bond is callable and the call is not absurdly far from being at the money, then instead of Z-spread, you're better off looking at option adjusted spread (OAS).

  • The calculation uses a wrong swap curve. If the bond has less than 2 years to maturity, and you use swap curve built from swap rates for your Z-spread calculation, you'll get absurly low (possibly negative) Z-spread. This is the default setting on Bloomberg.You should change it to a swap curve built from futures instead.

As an added bonus, note that "eurobonds" are not necessarily denominated in EUR.

$\endgroup$
5
  • $\begingroup$ Yes, sorry for not clarifying enough, "they" in this case meant to the terminals (Reuters/BBG) $\endgroup$
    – F0l0w
    Mar 8 at 13:48
  • $\begingroup$ These are interesting points, Dimitri. Reading your reply, I think what confused me is that I thought of the z-spread as the spread that the bond earns above a treasury spot curve (i.e. the "swap curve" portion confused me a little bit). Pardon my ignorance, would you interpret the statement as the "swap curve" simply as a swap fixed rate curve? (rather than the "Treasury spot curve" or some risk-free curve) $\endgroup$
    – F0l0w
    Mar 8 at 13:51
  • $\begingroup$ Z-spread is on top of swap curve, not treasury curve. She asked a very reasonable question - usually Bloomberg Z-spreads are correct, but sometimes they're not, e.g. because they did not set up bond's indicative data right, but can be easily fixed just by telling Bloomberg support about it. Also ensure that the swap curve choice setting points to the futures curve, not the the swap rates curve. $\endgroup$ Mar 8 at 13:59
  • $\begingroup$ Thank you very much! (as always very appreciated) $\endgroup$
    – F0l0w
    Mar 8 at 14:00
  • 1
    $\begingroup$ Might be good to clarify that the swap curve is the zero coupon (i.e. stripped) swap curve rather than the par curve quoted by dealers $\endgroup$
    – oronimbus
    Mar 8 at 22:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.