|
|
|
|
 |
Daily Speculations The Web Site of Victor Niederhoffer & Laurel Kenner Dedicated to the scientific method, free markets, deflating ballyhoo, creating value, and laughter; a forum for us to use our meager abilities to make the world of specinvestments a better place. |
Write to us at:
(address is not clickable)
11/26/04
Yield Curve, by James Sogi
In another NY Fed paper, authors find that the yield
curve ( spread
between 10yr and 13 wk) is highly predictive of recession
2-4 quarters
in advance, and outperforms the leading indicators, Stock
and Watson
indicators and the NYSE indicator. Knowing where in the
cycle we are
helps in judging the proper extent of making commitments.
The current
spread is 4.195-2.13= 2.063. According to the article study
this means
that there is less than a 5% possibility of a recession in
the next 2-4
quarters.
http://tinyurl.com/6po7o
Estimated Recession Probabilities for Probit Model
Using the Yield Curve Spread Four Quarters Ahead
Recession Probability Value of Spread
(Percent) (Percentage Points)
5 1.21
10 0.76
15 0.46
20 0.22
25 0.02
30 -0.17
40 -0.50
50 -0.82
60 -1.13
70 -1.46
80 -1.85
90 -2.40
The yield curve spread is defined as the spread between the
interest rates on the ten-year Treasury note and the
three-month
Treasury bill.
http://tinyurl.com/7xtb7 A Cinncinnati Fed paper looking
back 1875-1997
and taking a more statistical approach finds significant
predictive
power for yield curve 4 quarters before recession.
Look at the FOMC 12/22/98 transcript a few months after the
98 crash.
These minutes are a gold mine of ideas. The first thing they
look at: "
it is worth noting that both the 3-month and the 9-month
forward rates
continue to trade below the current 3-month deposit rate,
suggesting
some expectation of an ease coming in the first half of next
year, but
without much specificity to it. The two forward rates are
trading on top
of one another, so it does not look as if they imply a
series of
easings, but some further easing seems to be priced in here.
"
I have been exploring the yield curve using the 3mo-10yr
forward rates,
but maybe the use of the 3mo/9mo might be more predictive of
short term
moves. This is worth testing.
Governor Gramlich states,"As long as inflation is neither
accelerating
nor decelerating, we seem to be striving to maintain
existing
conditions. Partly this involves watchful waiting on
acceleration or
deceleration, not necessarily on inflation as such but on
leading
indicators of inflation such as those on the output side.
And in part
this involves aiming policy so that future growth in
aggregate demand
equals the trend growth in aggregate supply, which is
roughly 2.5%
percent under most models."
Rivlin states,"We used to think that central bankers had to
take big
leaps into the dark because our policy instrument, while
powerful in the
long run, affected the economy only over periods of 6 to 18
months or
more. But now we find ourselves with a tool that seems to
work faster.
Our actions in September through November are credited with
strengthening--or sometimes credited with overheating--the
U.S. economy
in the current quarter, not to mention the prospects for the
first half
of next year. One reason is the increased importance of
equity markets
both in financing business expansion and more importantly in
enhancing
consumer wealth...Another phenomenon that has cut the lags
is the
growing importance of consumer credit and home mortgages.
The ease of
refinancing home mortgages and home equity loans makes the
transmission
of monetary policy moves into consumer behavior more direct
and faster. "
And the reverse should be true, as rate rise, if not 30 yr
fixed
mortgages, at least credit card and equity loans and
adjustable rate
loans will reduce the lag.
and this can be seen in the 5 year curve humping up faster
than the
short or longer end.
<http://www.ustreas.gov/offices/domestic-finance/debt-management/interest-rate/y
ield-hist.html>
Fed Yield curve data for November:
http://tinyurl.com/2x8x6
November 2004
Date 1 mo
3 mo
6 mo 1 yr 2 yr 3 yr 5 yr 7 yr 10 yr 20 yr
11/01/04 1.79 1.99 2.20 2.34 2.61 2.89 3.36 3.76
4.11 4.84
11/02/04 1.86 1.97 2.19 2.33 2.60 2.86 3.34 3.75
4.10 4.84
11/03/04 1.83 1.96 2.18 2.32 2.60 2.85 3.35 3.74
4.09 4.83
11/04/04 1.85 1.98 2.19 2.34 2.63 2.89 3.37 3.76
4.10 4.82
11/05/04 1.86
2.03
2.27 2.44 2.80 3.04 3.51 3.88 4.21 4.92
11/08/04 1.88 2.07 2.30 2.47 2.80 3.08 3.51 3.88
4.22 4.95
11/09/04 1.89 2.08 2.29 2.46 2.80 3.05 3.53 3.88
4.22 4.95
11/10/04 1.88 2.08 2.29 2.47 2.82 3.07 3.56 3.89
4.25 4.97
11/12/04 1.91 2.08 2.31 2.49 2.86 3.10 3.53 3.89
4.20 4.91
11/15/04 1.92 2.12 2.34 2.53 2.89 3.12 3.53 3.89
4.20 4.91
11/16/04 1.93 2.14 2.36 2.54 2.91 3.14 3.56 3.91
4.21 4.92
11/17/04 1.90 2.13 2.33 2.50 2.85 3.07 3.47 3.83
4.14 4.85
11/18/04 1.91 2.13 2.34 2.51 2.86 3.08 3.48 3.81
4.12 4.82
11/19/04 1.98 2.15 2.36 2.56 2.95 3.17 3.57 3.91
4.20 4.89
11/22/04 1.98 2.20 2.43 2.60 2.95 3.18 3.56 3.90
4.18 4.85
11/23/04 1.99 2.19 2.41 2.60 2.98 3.20 3.58 3.92
4.19 4.85
11/24/04 1.98 2.18 2.40 2.60 3.01 3.23 3.61 3.93
4.20 4.85
11/26/04 2.01 2.21 2.40 2.61 3.03 3.25 3.64 3.98
4.24 4.90
Perhaps recognition of this cycle has led to a change of
cycle by making
the lag time shorter.
Check this against the drop in the leading indicators as
well
http://www.ny.frb.org/research/directors_charts/i-bcd_17.pdf
Money supply is lower http://tinyurl.com/52fyv
http://www.cm1.prusec.com/yararch.nsf/(Files)/msup_c.pdf/$file/msup_c.pdf
I criticized Roach's figures on debt. Here is data.
See Figures 42 and 43 for data showing that Consumer credit
is not
overly high except for mortgage refi's
http://tinyurl.com/3wmxt
James Sogi wrote on Tuesday, September 21,
2004.
> Spx first hit 1130 April 1998. Since then the mean price
is 1166.
> "Officer, do you think we'll hit 1166 soon?"... Sgt.
Friday ..."Just
> the facts Ma'am. "
It is not out of mere arrogance and perversity that I, an
individual
poor man, have taken to address your lordships... We must
remember that
in this matter we wrestle not against flesh and blood, but
against the
rulers of the darkness of this world. who may fill this
world with war
and bloodshed, but cannot themselves be overcome thereby."
(from
Machiavelli Address to Nobility)
With 1166 under our belt, Query; expanding range with
reversal or
capitulation of the losing half and the possibility of new
trend
breakout. (Query?( Republican=bulls - Democrat=bears)=new
trend?)
Query: will there be a pullback or breakout without lookback?
Revisiting Vix study of new 20 day vix lows and SP 10 days
after from
Wiz' supercharged script shows only one new observation on
10/1 since
last study. Updated data at
http://home.hawaii.rr.com/zephyr/vix.csv
Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
-97.9000 -21.1300 -0.7000 -0.9738 22.6000 77.0000
21.0000
> Recent observations
DATE Y M D W SP VIX SP10AHD
09/03/04 2004 9 3 5 1114.40 13.91 7.85
09/10/04 2004 9 10 5 1123.40 13.76 -11.90
09/13/04 2004 9 13 1 1128.25 13.17 -23.00
10/01/04 2004 10 1 5 1133.25 12.75 -25.00
11/03/04 2004 11 3 3 1145.00 14.04 ?
11/04/04 2004 11 4 4 1157.00 13.45 ?
The entries seem to be early. Last two pullbacks av.
36.pts