Treasury Futures Basis, Carry, and Delivery Options

Summary

The document explores how carry and delivery optionality can affect the basis between a Treasury futures contract and its cheapest-to-deliver bond near contract expiry. It considers a long basis position under both upward-sloping and downward-sloping yield curves. With positive carry, the writer reasons that delaying delivery may preserve carry and the end-of-month option; with negative carry, delivering early may be preferable if the carry cost outweighs the option value.

The central puzzle is whether these choices imply a positive or negative basis around the last trading date, and whether an apparently negative basis would invite cash-and-carry arbitrage. The text gives no answer or market evidence, so its scenario analysis remains tentative. Actual conclusions depend on precise contract delivery rules, conversion factors, financing and repo costs, accrued interest, and the value and timing of delivery options; the proposed arbitrage comparison does not assess those details.

Key ideas

• The choice of delivery timing reflects a trade-off between bond carry and the value of delivery options.
• Positive carry can favor delaying delivery, while sufficiently costly negative carry can favor earlier delivery.
• The document questions whether the basis can remain negative when a cash-and-carry trade appears possible.
• No resolution is provided, and a complete analysis would need to account for contract mechanics and financing costs.

Tags

# Futures basis (Bond) optimal delivery


# Futures basis (Bond) optimal delivery












i have a confusion regarding how the basis converges in a couple of scenarios. Lets assume I am long UST CTD Basis

- Say the curve is upward sloping: optimally, i would choose to make delivery of the bond at the last notice date (say last trade date is 1week before the last notice date). I would choose not to deliver at the last trading date as I can benefit from +ve carry for the next 1 week (+ don't want to squander my end of month option). Aside from the fact that the futures price ~ (Spot - 1w carry - EOM option PV)/CF, at the last trading date, can i say anything else about the basis (net of carry)? At this point it should probably just be positive a few ticks?

- Say the yield curve is downward sloping: There is a trade off between the carry and the end of month (EOM) optionality. So there are 2 scenarios, say: a). -ve Carry outweights the optionality lost. My optimal strategy here would be to save the -ve carry and deliver as soon as possible. Say this date is the last trading date of the contract. In this scenario as the negative 1w Carry more than offsets the EOM option PV => the converted futures price > spot price (therefore, basis is -ve) at the last trading date. But how can basis be -ve? As I can arbitrage this if I had no position by buying the CTD and selling a future and immediately delivering into this (this it self would push the basis back to 0). b). -ve Carry doesn't outweight the optionality lost. This would revert to something similar to point 1.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.