Evaluating Treasury Futures Basis Returns Against Balance Sheet Usage

Summary

The document discusses comparing long basis positions in Treasury futures, using the TU and FV contracts as examples. The questioner considers ignoring option value, compares quoted cheapness in basis points, and asks whether expected profit on a fixed notional is the right way to assess relative value when capital and balance sheet are constrained. The question also asks why returns differ when measured against DV01.

The reply interprets the positions as long the cheapest-to-deliver bond and short the futures contract. Under the assumption that the delivery option is negligible, it frames the return relative to balance sheet usage and notes that the assumed delivery date affects the estimate. This is a brief, conditional answer rather than a general valuation method: it does not derive the calculation, analyze financing or margin requirements, or explain the persistent DV01-based return difference in detail.

Key ideas

• A Treasury futures basis position can pair a long cheapest-to-deliver bond with a short futures contract.
• Comparing basis trades by profit on notional differs from evaluating returns against DV01 or balance sheet usage.
• Ignoring delivery-option value is an assumption that may affect the relative-value comparison.
• The assumed delivery date influences the estimated basis return, and the document does not provide a complete explanation of DV01 return differences.

Tags

# Carry and Return on Basis Trades and Asset Swaps


# Carry and Return on Basis Trades and Asset Swaps












I have been looking into the return of the treasury bonds basis trade. I am trying to analyze the relative value between the different contracts. If one were to ignore any option values in the contracts, for example, in the front contracts TU and FV the switch option is almost worthless, TU basis is about 1.4bps cheap and FV is about 0.6bps cheap. Is it correct to look at it from a dollar amount earned on notional? For instance, if I purchased 100mil notional of either, my return would be about 22K over the life of the contract. In this way I am considering balance sheet/capital constraints.

If this is not correct, why is it that there is a persistent difference in the return across contracts for DV01?

## Answer by dm63 (score 1)

https://quant.stackexchange.com/a/84123

I think I understand the question. You are looking at long basis positions in the TU and FV for December delivery ? Meaning, long the ctd and short the futures. Indeed, if the delivery option is worthless, then one typically sees returns of around 15-20bp on the balance sheet usage. For 1.5 months that amounts to about 22k. One should use the optimal delivery date (I assumed beginning of month).

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.