Resolution of Uncertainty - Comments on yesterday FED's meeting

The blue dot chart is continuing the seemingly inexorable decline in Treasury yields, with the median prediction of the most likely level for the Fed Fund rates for the end of 2019 down to 1.6% from 1.9% in September – no surprises here since the year end is virtually upon us –, down to 1.9% (from 2.1%) for the end of 2021, and down to 2.1% (from 2.4%) for the end of 2022. The very-long-term prediction has remained unchanged.

We have become accustomed to these downward revisions – even if we should not, because every time the blue dots are released, they should reflect an unbiased prediction. What is a more surprising is however the high degree of consensus for the level of rate at the end of next year: last September’s range of 1.625% to 2.375% for December 2020 has shrunk to 1.625% to 1.875%, with only four members at 1.625% now timidly breaking rank from their thirteen colleagues at 1.875%. No member sees the Fed Fund rates touching 2% by the end of next year.

What about the Fed Funds futures? The Dec 2020 contract sees an almost even probability (neglecting risk premium) for rates between 1.5% and 1.75% and between 1.25% and 1.50%. Once again, the market seems to have a more dovish outlook than the Fed, with the tail of the Futures-implied distribution totally skewed on the downside.  What is going on? Is it the case that the market ‘does not believe the Fed’?

Perhaps not: the blue dots reflect the most likely level of rates (the mode); the Fed futures their expected level (their expectation). Especially at the short end of the maturity spectrum, the rates distribution is extremely skewed, because of the asymmetry between short-term good and bad surprises, and this can easily account for the ‘wedge’ between the Fed blue dots and the message from the futures.

What about the very long end of the blue dot plot and the yields of long-dated Treasuries, for which the asymmetry of surprises should even out? Recalling that (neglecting convexity) yields are the average of the forward rates, the 1-year forward in 10-years’ time is under 2.00%. Once you factor back in a smidgen for convexity, you are left with a 10-year risk premium that is at best zero, certainly small, and probably negative.

Riccardo Rebonato is Professor of Finance at EDHEC Business School. He was previously Global Head of Rates and FX Research at PIMCO. He also served as Head of Front Office Risk Management and Head of Clients Analytics, Global Head of Market Risk and Global Head of Quantitative Research at Royal Bank of Scotland (RBS). Prior joining RBS, he was Head of Complex IR Derivatives Trading and Head of Head of Derivatives Research at Barclays Capital. Riccardo Rebonato has served on the Board of ISDA (2002-2011), and has been on the Board of GARP since 2001. He was a visiting lecturer in Mathematical Finance at Oxford University (2001-2015). He is the author of several books, in particular having published extensively on interest rate modelling, risk management, and most notably books on SABR/LIBOR Market Model pricing of interest rate derivatives, as well as on the use of Bayesian nets for stress testing and asset allocation. He has published articles in international academic journals such as Quantitative Finance, the Journal of Derivatives and the Journal of Investment Management, and has made frequent presentations at academic and practitioner conferences. He holds a doctorate in Nuclear Engineering (Universita' di Milano) and a PhD in Science of Materials (Condensed Matter Physics, Stony Brook University, NY).