Looking at the spread between 3 month commercial paper and the 3 month bill (using say, Fred), how/why (economically speaking) can this spread be negative? Are there some mechanics in the commercial paper markets at play that could make this happen? Or are there some idiosyncrasies with how these data-series/rates are sampled/constructed?

In general I would assume short term lending to private companies to be riskier than to the US government (except of course in a few extremely special cases, e.g. J&J and Microsoft).

Fred Commercial Paper Spreads

  • 1
    $\begingroup$ I think the spread between highest quality CP and bills is more like a liquidity premium rather than credit risk. But yes you would expect this to be positive. Perhaps some market segmentation and bill supply effects explain this. $\endgroup$
    – fes
    Jan 10, 2023 at 9:51
  • $\begingroup$ Thanks @fes - could you quickly explain what you mean in terms of market segmentation/bill supply? Is that just general illiquidity in the bill market/idiosyncracies of trading them? (I have very little experience with the mechanics of bond trading) $\endgroup$ Jan 10, 2023 at 17:41
  • $\begingroup$ E.g. if there is a spike in bill supply, the bill yield should increase for this to be absorbed by the market. Given some segmentation this might not affect CP yields or might do so at a lag. $\endgroup$
    – fes
    Jan 10, 2023 at 20:05
  • 3
    $\begingroup$ This chart looks way too noisy to be representative of a direct comparison of the quoted products with good quality synchronised and consistent data. I expect it might fall down in all those respects. $\endgroup$
    – Attack68
    Jan 10, 2023 at 21:21
  • $\begingroup$ compared this with another index from a commercial data provider that has no negative spreads except on two occasions $\endgroup$
    – oronimbus
    Jan 11, 2023 at 12:59

1 Answer 1


It could also be due to a mismatch in the frequencies which they measure the yield of commercial paper. So perhaps that cp index number is "pinned" for a bit while treasuries move.

  • $\begingroup$ That's very true - the FRED CP data is missing a ton of observations. $\endgroup$ Jan 10, 2023 at 21:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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