« One Thoroughly Unscientific Poll On Inflation Expectations, And Some More Reliable Ones | Main | A Little Of This, A Little Of That From The Forecasting Pros »
September 12, 2005
Funds Rate Probabilities: September 25, November A Toss-Up
The options markets on federal funds futures told a pretty clear story last week. After falling off precipitously in the immediate aftermath of Katrina, the probability of 25 basis increase at next week's FOMC is the clear market guess.
From there, the picture gets fuzzier. The Carlson-Craig-Melick estimates make it about even up on the probability of either another 25 basis points or a pause (over at least one meeting):
Certainly Katrina complicated things -- and, by the looks of these pictures, had a huge impact on market expectations. In light of the weight being put on the possibility of a pause at 3.75 percent, the following picture, which appears in the most recent edition of the Federal Reserve Bank of Cleveland's Economic Trends publication, is kind of interesting:
In essence, the picture says that a central bank targeting a 1.5 percent rate of inflation (measured by the PCE price index excluding food and energy) and an output gap as estimated by the Congressional Budget Office (through the second quarter), would find the neighborhood of a 3.75 percent federal funds rate reasonably pleasing. (The upper and lower bounds in the picture correspond to 1% and 2% inflation targets, respectively.)
For the hardcore fans:
-- The estimated probabilities are calculated using the constraints described in last week's post:
Any single probability must be nonzero.
The estimates must sum to one.
The mean funds rate implied by the estimates must equal the implied rate from the futures market.
-- The market for the January contract is looking a bit thin, so I am not reporting a set of December probabilities this week. We'll keep the monitor on.
-- Here's the data:
Download implied_pdf_october_090905.xls
Download implied_pdf_november_090905.xls
September 12, 2005 in Exchange Rates and the Dollar, Fed Funds Futures | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef00d834590bdc69e2
Listed below are links to blogs that reference Funds Rate Probabilities: September 25, November A Toss-Up:
» Hurricane? What hurricane? from Econbrowser
Amazingly, gasoline futures on the New York Mercantile Exchange today ended back where they were before Hurricane Katrina struck with all its fury. Retail prices will likely follow that lead. B... [Read More]
Tracked on Sep 12, 2005 11:16:07 PM
» Monetary policy and Newcomb's problem from Stumbling and Mumbling
Macroblog reports that the markets are unusually uncertain about where the fed funds rate is heading. Could it be that one reason for this is that the Fed faces a version of Newcomb's problem?To remind you, the problem is as [Read More]
Tracked on Sep 13, 2005 11:35:07 AM
Comments
Posted by:
knzn |
September 12, 2005 at 05:41 PM
Over Greenspan's term the Fed Funds rate has averaged about +200bp over the 5 year CPI; the Standard Deviation is also about 200bp. So with the 5 year CPI at 2.50% (where it has been for the last 13 years) if the Fed only pushes the FF to the avg it will be 4.50%. Also, when in a tightening mode the Greenspan Fed has not stopped before FF hit the avg +100bp which would mean 5.50%. So, what is all of this about the Fed stopping at 4.00%? Logic or wishful thinking?
Posted by:
Norman |
September 12, 2005 at 06:22 PM
Norman, By my calculations the funds rate has averaged about +175 bps over the 5-year CPI inflation rate during Greenspan’s term. Before Katrina hit, it was reasonable to expect the CPI inflation rate to come down, because much of the inflation was from rising energy prices that, futures markets suggested, were not expected to continue. So a reasonable guess of 2.25% for the CPI going forward would imply 4% for an average federal funds rate. And up until a few months ago, there looked to be enough slack in the economy that the Fed would not want to push the rate above the average. That’s the logic.
Of course the logic has been thrown all awry by recent developments. Now it seems you can make a case for any rate you want depending on whether you assume the Fed is more worried about recession or inflation. In the former case, they shouldn’t even go up to the average; in the latter case, they should go above the average. And perhaps they’re equally worried about both, in which case maybe the average is a good guess (but now maybe the average is higher than 4%).
Posted by:
knzn |
September 12, 2005 at 07:40 PM
Wishful thinking OR talking one's book? I understand the argument, McCulley said it quite clearly. When it comes down to it, it's not about right or wrong, or even what's best for the economy; it's about whether AG has what it takes to go out the door standing up. For McCulley, the answer seems clear: it's very unlikely.
I've been vocally anti-AG for 14 years, but I believe this time he'll do what's right. I believe AG's first, foremost & forever an Ayn Rand groupie, and I think he's had enough of bailing out tactical managers who've made fortunes playing him for the fool.
Posted by:
bailey |
September 12, 2005 at 07:52 PM
Who's risk premium?
For the liability manager, or for that matter the typical pension fund and/or insurance company the risk is that we have a lower flatter yield curve...
Tell me what FED FUNDs are going to be for the next two years.. and I'll tell you today with certainty what the Two Year Note will yield.. anyother price will produce a certain aribitrage profit through the Repo market..
Tell me that that rates are lower by another 150 basis points and I'll tell you that GM's pension fund is truly defunct..
Posted by:
sjonas |
September 13, 2005 at 12:17 AM
sjonas: I realize it is theoretically possible for the normal yield curve to slope downward, or to be flat, but we have many decades of data showing a consistent tendency for the upward slope to predominate. That seems to indicate that the risk premium is determined primarily by short horizon players (e.g. banks facing uncertain withdrawals, managers who must mark to market). Of course, if we knew for sure what the funds rate would be, there would be no risk premium, and the normal yield curve would be flat. As it is, with a normally upward-sloping yield curve and the possibility of arbitrage, future bets on the funds rate should be higher than the mean of what is actually expected.
Posted by:
knzn |
September 13, 2005 at 10:17 AM
knzn -- The estimate's gurus do indeed have some versions that incorporate a term premium. It's a pretty simple-minded adjustment -- basically a deterministic trend -- so it not surprisingly turns out that the general picture that emerges is not much changed by its inclusion. Your point is well-taken, though, and it is one of the reasons that we offer these estimates more as a glimpse of the direction in which sentiment changes on economic news than a set of point estimates that ought to be taken to the bank.
Posted by:
Dave Altig |
September 14, 2005 at 11:01 AM
knzn
Your analysis sounds like something out of an old fashioned economics 101 textbood..
Uncertain withdrawls? Bank liquidity risk?
More to the point would be the impact of a flattening yield curve on their mortgage portfolio... now that's a risk feeding directly into pre payment and the lack of assets just when they need them..
Short term players?... In the curent environment that's the Trillion Dollar leverage Hedgfund both in the US and Euroland..
Their short term preferences are if anything ecclectic..
Telling me that on average the yield curve is positive is akin to saying that most of the time options expire worthless ...
Averages are meaningless in terms of risk..
A quick recall of the last two weeks or so.. should ocnvince you that anyone playing the averages would have been eliminated from the population pool as the "certainty of the FED's likely tightening was replace by the possibility that it might actually ease..
How an "arbitrage" as you put it turns out is totally irrelevant.. the worst type data mining.. suggest you take a look at Robert Barro's recent paper Rare Events and the Equity Premium* for a better way to look at averages and event risk..
Posted by:
sjonas |
September 16, 2005 at 12:07 AM
“The mean funds rate implied by the estimates must equal the implied rate from the futures market.” Might not this requirement give an upward bias to the estimates, given the fact that yield curves are normally upward-sloping?
For example, if the funds rate were expected to be constant, we would expect that the yield curve for securities priced off the funds rate (if there are such things) would be upward-sloping, because the longer-dated securities would require a higher term risk premium, in accordance with what is usually observed. Therefore the implied forward curve for such securities would be even more upward-sloping. Arbitrage conditions should then require that the more distant funds rate contracts must imply higher interest rates, even though, by assumption, the funds rate is expected to be constant.
The paper linked at the top states that “there is no evidence of bias when the two-months-ahead futures price is used to forecast the actual fed funds effective rate.” I’ll take their word for that as an empirical fact, but it strikes me as something of an anomaly (possibly because there aren’t yet enough data to identify the bias).