I'm building out a calibrated USD MultiCurve set, focusing on Fed Funds & 3M LIBOR for the purpose of trading in Fed Funds Futures and Eurodollar Futures (for now ...).

I had always assumed that the OIS/Fed Funds rate (perhaps with a turn-of-calendar spread adjustment) was used as the discounting curve itself. In reading Darbyshire's excellent Pricing and Trading Interest Rate Derivatives, I came across another rate (the Interest on Excess Reserves) that seems to have some merit as the true discounting basis.

  • It is truly a risk-free rate, being the interest on deposits held at the Fed rather than unsecured overnight lending.
  • It is, I believe, the interest offered on margin at the CME.

On the other hand, due to the fact that the Fed Funds rate reflects lower-than-IOER loans made from banks unable to earn interest on reserves to US Branches of Foreign Banks, the Effective Fed Funds rate is often lower than the IOER.

Question: given my interest in trading futures, which is the more theoretically sound USD MultiCurve setup?

  • OIS/Fed Funds forecasting + 3M Libor forecasting, with an EFFR vs. IOER spread to the discounting curve, or
  • OIS/Fed Funds (forecasting + discounting) + 3M Libor forecasting.
  • $\begingroup$ A practical problem is that there are no observable/liquid instruments to express a view on what the IOER itself or the difference between IOER and some other benchmark will be in the future. The Fed only made IOER different from fed funds rate recently. Whatever the theories, it is much easier to tell what the market thinks about fed funds from futures and options. $\endgroup$ – Dimitri Vulis Apr 30 '20 at 13:59
  • $\begingroup$ Fair. But I could assume a deterministic spread based on history between EFFR vs. IOER, similar to using historical data to estimate turn-of-calendar effects, right? $\endgroup$ – MikeRand Apr 30 '20 at 14:13
  • $\begingroup$ It's not good accounting to price anything with a curve based on such an iffy assumption. :) $\endgroup$ – Dimitri Vulis Apr 30 '20 at 16:27

Your Answer

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

Browse other questions tagged or ask your own question.