I know these are newbie questions, but I have a couple of doubts which I can´t solve. Why is the difference between Sept-Dec ZN so small and the Dec-March ZN so big in comparison? And why is the roll return higher for the Bund if the yield curve is flatter for the German bonds? Thank you
As to ZN, I think you're a wee bit confused with the whole decimal vs 32s prices. The Sep-Dec spread is arnd 27/32s, whereas the Dec-Mar spread is arnd 31/32s, which, given the non-existent liquidity of the 3rd contract, means the two rolls are identical. In general, liquidity only exists in the front-2nd contract roll and that is only meaningful when the roll season is upon us. That's why I wouldn't read too much into the Dec-Mar bund roll at all, 'cause it's not a meaningful price. In theory, the rolls are a function of the choice of CTD and the implied repo to delivery.
Thank you for the answer, Martinghoul. You were right, I got confused by the 32s. Forgetting the Dec-March spread and focusing only in the Sept-Dec spread of ZN and GBL, the latter is clearly higher (1.46 vs 28/32 at the moment of writing this) and I understand that as you said both depend on the CTD and its implied repo rate, but in the long term with the current interest rates, roll after roll, shouldn´t be on average the ZN roll higher? If I open at a determined level a spread between ZN and GBL, shorting the second and adjusted for the â¬/$, of course, it´s important for me to know if the roll is going to work against me or not.
I don't see any particular reason why the ZN roll should necessarily be higher, roll after roll. Every roll is different, 'cause the two variables change from roll to roll. I don't really see a good way to look at the rolls other than actually looking at the cheapness/expensiveness of the contracts and the fair value of the roll. The actual prices of the spreads themselves don't really have a lot of meaning to me, 'cause they can vary for all the wrong reasons, so to speak.