A recent paper posted on the Federal Reserve Board's "FEDS Notes" research page, entitled, "Front-End Term Premiums in Federal Funds Futures Rates and Implied Probabilities of Future Rate Hikes," discusses "term premiums" and Fed funds rates. It explains, "A few recent market commentaries have proposed simple rules of thumb that describe term premiums in money market futures and forward rates in the current interest rate environment as a linear function of horizon with a slope of about minus 1 to minus 2 basis points per month. Such term premiums can have significant effect on inferences regarding the market-implied probability of future monetary policy actions at the next few FOMC meetings, but solid empirical evidence on term premiums at such horizons has been lacking. In this note, we examine empirical evidence on term premiums at the very front end, utilizing federal funds futures data as well as responses to the Desk's sell-side survey (Survey of Primary Dealers, or PD survey) and buy-side survey (Survey of Market Participants), and discuss plausible front-end term premium assumptions that one can use to extract probabilities of a rate hike at upcoming meetings from market quotes." Authors Don Kim and Hiroatsu Tanaka continue, "We find that, for horizons over the next two to three FOMC meetings, a simple assumption that term premiums are zero produces a good agreement between federal funds futures data and the Desk survey data on the distribution of the timing of the next rate hike, while an assumption of significantly negative term premiums suggested by some market commentaries produces futures-implied probabilities of near-term rate hikes that appear generally much higher compared with those from the surveys. In addition, Desk survey responses on the distribution of the federal funds rate at year-ends are also consistent with the assessment that term premiums are small in magnitude at two-to-three-meeting horizons while more sizable (negative) further out; however, even at somewhat longer horizons of about six months, estimated term premiums are generally smaller in magnitude than some of the numbers proposed in recent commentaries. Overall, it appears that term premiums are nonlinear with respect to horizons at the front end."

Email This Article




Use a comma or a semicolon to separate

captcha image