« The New Chinese Exchange Regime Explained | Main | Are American Consumers Responsible For The Large Trade Deficits? Some Further Thoughts »
July 25, 2005
Fed Funds Probabilities: The Early November Results Are In
I'm a little late posting these today because I have been attending a workshop on -- using market data to uncover expectations about future federal funds rates. As a bonus, though, we have our first pass at the probabilities implied by options on the November contract on fed funds futures, which will include the November 1 meeting of the Federal Open Market Committee. It looks like the market doesn't expect 3.75% to be the end of the trail.
The pictures:
And the data:
Download october_pdfs_072505.xls
Download imp_pdf_slides_for_blog_072205.ppt
Download imp_pdf_slides_for_nov.ppt
July 25, 2005 in Fed Funds Futures | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef00d834842b5269e2
Listed below are links to blogs that reference Fed Funds Probabilities: The Early November Results Are In:
Comments
Posted by:
sjonas |
July 25, 2005 at 11:46 PM
sjonas --
I don't disagree that the probability may be higher than what is indicated by these calculations. There may be some bias in the methodology because the contract is actually written on the average effective funds rate, not the target. The effective rate of late has been running above the target rate for a variety of technical reasons, and that may be causing our methodology to load some probability into the higher rate that that shouldn't be there. On the other hand, there may be term premia considerations that could distort things in the opposite direction. We're working --- or the people who created these things are working, I should say -- on improving our approach. Maybe your suggestion can help us find our way to some of those improvements. Thanks for the input.
Posted by:
Dave Altig |
July 26, 2005 at 07:20 AM
The effective rates' running over the target is a function of the certainty of the FED's moves when the FOMC meeting occurs early in the maintence period..
This is by now a well known aribitrage and requires a small adjustment that everyone in the marketplace is aware of and takes into account..
That of course is how martingales get formed..
As to the term premia I don't even know what the "term" means.. Which direction do you think it lies in??
Either the FED goes or it doesn't unless you assume that for some reason there are more borrowers than lenders, or that in some Lucasian sense that higher FED FUNDS rates are correlated with a good or bad state of nature..
Posted by:
sjonas |
July 26, 2005 at 10:41 PM
The difference between the 100 percent probability calculation by Mr. Jonas and the roughly 65 percent calculation in the chart probably relates to two factors.
1. The chart is from July 22, and prices moved just a little bit from July 22 to July 25.
2. The calculation by Mr. Jonas essentially uses a single option price to solve for the probability of the FOMC moving to 4.0 in November. The calculations in the chart use several options from November and consider more than just two possible outcomes for the November meeting.
I do think it is a slight overstatement that the method proposed by Mr. Jonas would yield a 100 percent probability of a move to 4.0 percent in November. Rather, the probability from this method is more like 80 percent, still above the roughly 65 percent in the chart.
Fundamentally, it is clear that the market is becoming increasingly convinced that the FOMC will move to 4.0 percent in November.
Posted by:
Will Melick |
July 28, 2005 at 04:29 PM
Actually the numbers are higher than you're methodology indicates.
Essentially the market is pricing almost full certainty of an unconditional 25 basis point move in November, irrespective of what happens before.
All one needs to do is compare the October 96.25 call to the November 96.1875 calls which are the most actively traded in the FED FUNDS options.
If the November is about the same price as the October.. this is a zero cost contingent bet on the FED "not going in November".. put call parity then impies that the probability of a FED miss is approximately zero, hence the probability of a go 100-0, or 100%...