Forecasting The Fed With Fed Funds Futures
No single
institution has greater more sway over U.S. financial markets
than the Federal Reserve. And the Fed’s most powerful weapon is the federal
funds interest rate.
The fed funds rate sets the interest
charged by banks for overnight loans to other banks in need of more cash to meet reserve
requirements. The rate is decided by the Federal Open Market Committee, which decides monetary policy of the nation’s central
bank.
By raising the rate, the FOMC
restrains bank lending downstream to businesses and consumers. This translates
into less cash to chase securities as well as stimulate the economy. By
lowering the rate, the committee makes credit cheaper. That boosts lending activity, funneling more money into the markets and the economy.
Obviously, it behooves us as traders
to stay alert to the likely future direction of this pivotal interest rate.
Fortunately, we have a handy, historically accurate tool for predicting changes in the rate — the Federal Funds
Futures contract.
With a bit of arithmetic, you can
determine from the contract where the market thinks the fed funds rate is headed. Even better, you can figure out
a probability of that forecast. For
instance, on a given trading day, the contract price might imply, say, a 75%
chance of a quarter-point increase in the fed funds rate to 6.75%.
The Chicago Board of Trade established
the fed funds market in 1988 as means for “commercials” — banks,
other companies and repo traders — to hedge against movements in the federal
funds interest rate.
The hedgers dwarf the speculators. According to the Commodities Futures Trading Commission,
commercials accounted for 32,000 of the 46,000 contracts
comprising the open interest in all fed funds futures contracts as of May 30, 2000. The remaining 14,000
contracts were held by non-commercials — mostly locals and small speculators.
The contract trades from 8:30 a.m. ET
to 3 p.m. ET on the CBOT. It also trades overnight on the Globex electronic
network. But trading on the Globex is very light compared to volume on the CBOT.
That makes Globex pricing too volatile for use in interest-rate forecasting.
Chicago’s fed funds pit has proven a trustworthy oracle.
Take the seven FOMC meetings from June
1999 through May 2000. Over this period, the committee raised the fed funds rate
from 4.75% to 6.5%. In five meetings, the FOMC raised the rate by quarter-point
increments. In another meeting, the committee left the rate unchanged. In one
meeting, the committee raised the rate a half point.
In six meetings, the fed
funds contract price implied the correct outcome, with an implied confidence
level of 70% or higher in the forecast, a week ahead of the FOMC’s decision. The
implied probability rose as the meeting day neared.
The only Fed decision that gave the
Chicago handicappers some trouble came on Nov. 16, 1999, when the FOMC raised the
fed funds target by 25 basis points. A week earlier, the November futures
contract put the odds of a quarter-point hike at a rather hesitant 55%. The clouds cleared over the pit a few days later. By Nov. 15, the probability implied in the November contract had climbed to 80%.
This level of forecast accuracy isn’t
surprising. Under Fed Chairman Alan Greenspan, the nation’s money chiefs have
tried to prepare the financial markets in advance of interest-rate decisions. As
decision times draw near, they tend to drop broad hints telegraphing their
intentions, particularly when rate increases are at hand.
When we talk about the fed funds rate
set by the Fed, we’re really talking about a target rate. The actual rates on
interbank loans oscillate around the target from day to day but average out to match bogey over time. Each monthly fed funds contract amounts to a bet on the
average effective fed funds rate for that month, as reported by the Federal
Reserve Bank of New York.
You can find end-of-day closing
futures prices at the CBOT Web site. For the fed funds contract, go to www.cbot.com/mplex/quotes,
go to “Futures Settlement Prices,” then select “Financials.”
TradingMarkets.com members can
pull up charts as well as quotes on the fed funds futures as well as any other
futures contract. Go to the Futures section of the Web site. Click on the Futures
tab near the top of the page. Then hit the Launch
Futures Charts button. That will bring up your quote and chart
display. Just enter the ticker symbol for the contract that interests you.
Ticker symbols for futures typically consist of a letter pair designating the market (wheat, crude oil, bonds, etc.) followed by a letter designating the contract month and a number designating the year. Market designation for the fed fund
futures contract is FF, followed by the letter and number designating the month
of the contract that interests you. Here are the month symbols for the CBOT-traded
contract: January F; February G; March H; April J; May K; June M; July N; August
Q; September U; October V; November X; December Z. The year symbol is simply the
last digit of the year. For instance, the year 2000 would be 0.
For example, the ticker for the August
2000 fed funds futures contract would be FFQ0. The ticker for the January 2001
contract would be FFQ1.
The price for fed funds futures is
based on settlement at 100 minus the effective average fed funds rate. So
futures prices move inversely with changes in the expected fed funds interest
rate in the settlement month.
Now, let’s run through an example of
how to use the contract to forecast the probability of a rate change by the
Federal Open Market Committee. I’ll use closing prices throughout this lesson.
Implied Rate
Let’s say today’s date is June 13,
2000. The federal funds target rate stands at 6.50%.You want to know
the probability that the FOMC will raise the rate by 25 basis points at its June
28 meeting.
The Fed will not meet again until Aug.
22. This makes the July contract perfect for forecasting the June 28 meeting as
there is no scheduled meeting in July to affect our calculations.
Your first step, then, is to figure
out the average effective fed funds interest rate for the month of July as
implied in the July contract price. The July futures closed at 93.420 on that
day. Subtract the contract price from 100 to obtain the implied rate, which in
this case comes to 6.58%.
Probability
Just by eyeballing the implied rate of
6.58%, we can tell that the market on June 13 expected the FOMC to leave the
target rate unchanged at 6.50% at the June 22 meeting. Otherwise, the July
contract would have priced in an implied rate much closer to 6.75%, the target
rate if the Fed were to tighten by its usual increment of 25 basis points per
meeting.
With one more calculation, you can
assign a percentage probability to the likelihood of a rate hike. The implied
rate of 6.58% is eight basis points above the current target. In other words,
the market has only priced in eight points of a 25 basis point hike. To obtain a
probability, just divide the tightening priced into the contract by the possible
rate hike.
So divide eight by 25. The resulting
quotient of 0.32 indicates a 32% probability that the FOMC will raise the fed
funds target by 25 basis, or a 68% probability that the committee will stand
pat.
Weighted Average
Unfortunately, not all rate decisions
meetings come at the end of the month with no FOMC meetings set for the
following month. If you lack a “clean” month after the expected rate
decision, you’ll need to use the contract for same month in which the decision
occurs.
As a result, you must use a weighted
average that blends the old rate with the new rate that you are forecasting. The
FOMC meeting on May 16, 2000 is a good example. The meeting fell square in the
middle of the month, and the committee convened again in June, ruling out the
June contract for forecasting the May 16 rate decision.
Performing a May forecast will help us
deal with another scenario. On May 16, as we know now, the Fed departed from its
usual approach of raising rates by quarter-point moves and jumped the fed funds
target by a half point.
Let’s assume today’s date is April 28.
On that day, the May fed funds futures contract closed at 93.790, giving you an implied
interest rate of 6.21%. At the time, the target rate stood at 6.00%.
So you saw that the market had priced
in an increase of 21 basis points. Write down that number. We’ll come back to it near the end of this exercise.
Now the question is, how many basis
points needed to be priced in to forecast an increase of 50 basis points?
May has 31 days. Assume that the 6.00% target prevailed every day of the month up to and including the day
of the FOMC meeting. We include the meeting day because the bulk of the contract
trades occur early in the day before the rate decision is announced.
So you will compute a weighted average
that assigns the first 16 days of the month to the old rate of 6.00% and the
remaining 15 days to the new rate of 6.50%.
Divide the number of days in the first
part of the month by total days in the month. Then multiply the quotient by the
old rate.
(16/31) * 6 = 3.097
Next divide the number of remaining
days in the month by total days in the month. Then multiply the quotient by the
forecast rate.
(15/31) * 6.5 = 3.145
Now add the two products together to
obtain an implied average effective rate of 6.24%
for the month of May.
This weighted average implied rate
represents the implied rate of the May contract if the market fully priced in a
target rate increase of 50 basis points. In other words, the contract must price
in 24 basis points above the target rate to yield a 100% probability of a
half-point rate increase on May 16.
Last Step
As we determined earlier, the May
contract closed at 93.790 on April 28, yielding an implied rate of 6.21%, or 21
basis points above the target rate of 6.00%.
There’s 24 basis points required to
price in a half-point increase. The contract has actually priced in 21 basis
points. Twenty-one divided by 24 equals 0.875, or 87.5%. As of April 28, your
calculation shows that the fed fund futures market discounted an 87.5%
probability that the FOMC would raise the fed funds target by 50 basis points on
May 16. As we know, the FOMC later proved the market right.
A Final Word
This technique works best when
forecasting forthcoming rate decisions, not decisions several months or more into
the future. Obviously, the further you project into the future, the more
unforeseen developments can intervene and disappoint your forecast.
Another problem stems from trading
volume in the fed funds futures contract. Open interest drops off sharply for
contracts expiring in months further out. You need substantial trading volume before
trying to draw conclusions from any security’s price. So these contracts tell
you little about the market’s built-in assumptions on interest rates.
In a future lesson, I will show you a
way to use the market for Eurodollars,
U.S. dollars held in banks outside the United States, to get a fix on how
the financial markets are handicapping the fed funds rate in the more distant
future.