Upload
duongdung
View
218
Download
0
Embed Size (px)
Citation preview
New Paradigms in Marketing: Are Speculators or the Fundamentals Driving Prices?
Scott H. Irwin
Outline of Presentation
• Role of speculation in the recent commodity price boom
• Changing fundamentals
• Convergence problems at the CBOT
A New Type of Commodity Speculator
• Commodity Index Investors– Desire portfolio exposure
to returns from a basket of commodities
– Long-only• Popular Indexes
– GSCI– Dow Jones-AIG– Reuters/Jeffries-CRB
• Investment Types– OTC index funds– Exchange-traded funds– Exchange-traded notes
Investors
Swap
Dealer
Long Futures Positions
$
$
Index
π
π$
Notional Value of Commodity Index Trading on U.S. Futures Exchanges (CFTC)
118 133 161
946
0
250
500
750
1,000
Dec-07 Mar-08 Jun-08 All FuturesJun-08
Inve
stm
ent (
bil.$
)
“Perma-longs”
Proportion of Open Interest Held by Index Traders in Grain and Livestock Futures Markets, Jan 2006-Jun 2008
5%
10%
15%
20%
25%
30%
Jan-
06
Mar
-06
May
-06
Jul-0
6
Sep-
06
Nov
-06
Jan-
07
Mar
-07
May
-07
Jul-0
7
Sep-
07
Nov
-07
Jan-
08
Mar
-08
CIT
/Tot
al O
I
Wheat
Soybeans
Corn
5%
10%
15%
20%
25%
30%
Jan-
06
Mar
-06
May
-06
Jul-0
6
Sep-
06
Nov
-06
Jan-
07
Mar
-07
May
-07
Jul-0
7
Sep-
07
Nov
-07
Jan-
08
Mar
-08
CIT
/Tot
al O
I
Feeder Cattle
Live Cattle
Lean Hogs
Source: Sanders, Irwin, and Merrin (2008a)
Index Trading in NYMEX Crude Oil Futures
12/31/2007 3/31/2008 6/30/2008
Notional Value (bil. $) 39.1 41 51
Net Long Position (# contracts) 408,000 398,000 363,000
Total Futures Equivalent Open Interest (# contracts) 2,508,971 2,885,101 2,837,447
Index Position/Total Open Interest 16% 14% 13%
Source: CFTC Staff Report on Commodity Swap Dealers and Index Traders (2008)
It Has to be a Bubble!
• A ‘titanic’ wave of money invested in commodity futures markets
• Overwhelmed ‘normal’supply and demand fundamentals
• Greatly magnified upward trend in commodity prices
• Final result: A bubble
So it’s becoming increasingly clear that there are very few people left in academia and economics-land that think that commodities were anything but a bubble. In fact it appears the only economists left attacking the bubble theory are the ones being paid by Wall Street to defend their actions.
---Accidental Hunt Brothers Blog, October 15, 2008
The World According to Mr. Masters
Conceptual Error #1: Money Flows are Not the Same as Demand
• Futures markets are zero-sum games
• If long positions of index funds are new “demand” then the short positions for the same contracts are new “supply” ?
• Simply observing that large investment has flowed into the long side of commodity futures markets at the same time that prices have risen substantially does not necessarily prove anything
“…for every long there is a short, for everyone who thinks the price is going up there is someone who thinks it is going down, and for everyone who trades with the flow of the market, there is someone trading against it.”
Tom Hieronymus
Conceptual Error #2: Index Futures Positions Distort both Cash and Futures Prices
• Futures contracts are financial transactions that only rarely involve the actual delivery of physical commodities (i.e. “side bets”)
• To impact the equilibrium price of commodities in the cash market, index investors must take delivery and/or buy quantities in the cash market and hold these inventories off the market
• Absolutely no evidence that index fund investors are taking delivery and owning stocks of commodities
Conceptual Error #2: Index Futures Positions Distort both Cash and Futures Prices
• Futures contracts are financial transactions that only rarely involve the actual delivery of physical commodities (i.e. “side bets”)
• To impact the equilibrium price of commodities in the cash market, index investors must take delivery and/or buy quantities in the cash market and hold these inventories off the market
• Absolutely no evidence that index fund investors are taking delivery and owning stocks of commodities
• Hedging and speculation are best described as a continuum
• Index funds entered a dynamic and ever-changing “game” between commercial firms and speculators with various motivations and strategies
• Commercial firms tend to have an informational advantage
• Index funds add liquidity and may improve competition in commodity futures markets
Conceptual Error #3: Hedgers are Benign Risk-Avoiders and Speculators are Harmful Risk-Seekers
Long Short Long ShortHedging Hedging Speculation Speculation
Corn2006 328,362 654,461 558,600 208,0432008 598,790 1,179,932 792,368 182,291
Change 270,428 525,471 233,768 -25,752Soybeans
2006 126,832 192,218 183,105 107,2212008 175,973 440,793 351,379 74,844
Change 49,141 248,575 168,274 -32,377Wheat
2006 57,942 213,278 251,926 92,1482008 70,084 240,864 300,880 121,578
Change 12,141 27,585 48,954 29,430
---# of contracts---
Inconsistent Fact #1: Speculation is not Excessive Compared to Hedging (2006:I-2008:I Averages)
Source: Sanders, Irwin, and Merrin (2008a)
Inconsistent Fact #2: Price Increases Did Not Occur in All Commodity Futures Markets Included in Popular
Indexes (January 3, 2006 – April 15, 2008)
-50%
0%
50%
100%
150%
200%
Corn
Soyb
eans
Soyb
ean
Oil
CBOT
Whe
atKC
BT W
heat
Cotto
nLi
ve C
attle
Feed
er C
attle
Lean
Hog
s
Jan
2006
- A
pr 2
008
Cha
nge
(%)
Inconsistent Fact #3: Price Increases Occurred in Commodity Futures Markets not Included in Popular
Indexes or Markets Without Futures
+78%$34.40/cwt.$19.30/cwt.Edible Beans (cash)
+58% $0.41/lb. $0.26/lb. Apples Fresh Use (cash)
+37%$17.29/cwt.$12.65/cwt.Nearby Fluid Milk Futures
+168%$22.17/lb.$8.27/lb.Nearby Rough Rice Futures
ChangeApril 2008January 2006
Inconsistent Fact #4: Inventories did not Increase for Storable Commodities
PE
PB
Inventory Increase
Q
S
D
Ending Stocks as a Percent of Use, 2001/02-2007/08
0
5
10
15
20
25
30
35
40
2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08En
ding
Sto
cks/
Use
(%)
Corn Soybeans Wheat
“So my challenge to people who say there’s an oil bubble is this: let’s get physical. Tell me where you think the excess supply of crude is going.”
Inconsistent Fact #5: Commodity Index Fund Trading is Predictable
• Index funds do not attempt to hide their current positions or their next move– Publish mechanical procedures for rolling to new
contract months– Indicate desired market weightings when the index is
re-balanced• Theory shows that trading must be unpredictable for
any trader group to consistently push prices away from fundamental value
• Highly unlikely that other large traders would allow index funds to push futures prices away from fundamental values for long when trades are so easily anticipated
This Time is Different!
• Facts build a persuasive case against bubble hypothesis• But, evidence is “circumstantial”• Bubble proponents can argue “this time is different”• Direct evidence on the relationship between speculator
positions and price changes is needed
Solution: Granger causality tests
We Have Been Here Before: Lessons from History
U.S. Congress: “Speculative activity in the futures markets causes such severe and unwarranted fluctuations in the price of cash onions as to require complete prohibition of onion futures trading in order to assure the orderly flow of onions in interstate commerce.”
Harry Truman: “…the cost of living in this country must not be a football to be kicked around by grain gamblers.”
Abraham Lincoln: “For my part, I wish every one of them [speculators] had his devilish head shot off.”
Vladimir Lenin: “For as long as we fail to treat speculators the way they deserve—with a bullet to the head—we will not get anywhere.”
Changing Fundamentals?
• The most basis question in fundamental analysis is the price a prospective buyer (user) is willing to pay for a unit of the commodity
• For at least the last 50 years the primary determinant of the fundamental value of corn has been the value placed on corn by livestock feeders
“The price of feed is determined by livestock feed demand, feed production, exports, and food and industrial uses. The lines of causation are from consumer demand through the livestock sector to feed prices.”“Master Model of Midwestern Agriculture,” Hieronymus, Good, and Hinton (1980)
US Corn, Ethanol for Fuel Use, 1975/76-2008/09*
0500
1,0001,5002,0002,5003,0003,5004,000
1975/76 1982/83 1989/90 1996/97 2003/04Marketing Year
Etha
nol (
mill
ion
bush
els)
Source: USDA *2008/09 Projected
US Corn, Ethanol for Fuel Share of Total Use, 1975/76-2008/09*
0
5
10
15
20
25
30
35
1975/76 1982/83 1989/90 1996/97 2003/04
Marketing Year
FSI/T
otal
Use
(%)
Source: USDA *2008/09 Projected
Weekly Ethanol and Corn Prices at Iowa Plants,January 27, 2007 – March 20, 2009
2.00
3.00
4.00
5.00
6.00
7.00
8.00
1/26/0
73/2
3/07
5/18/0
77/1
3/07
9/7/07
11/2/
0712
/28/07
2/22/0
84/1
8/08
6/13/0
88/8
/0810
/3/08
11/28
/081/2
3/09
3/20/0
9
Cor
n Pr
ice
($/b
u.)
1.25
1.50
1.75
2.00
2.25
2.50
2.75
3.00
Etha
nol P
rice
($/g
al.)
Corn (left-scale)
Ethanol (right-scale)
Relationship between Weekly Corn and Ethanol Prices at Iowa Plants, September 7, 2007 – March 20, 2009
y = 2.4212x - 0.3103R2 = 0.8999
2.00
3.00
4.00
5.00
6.00
7.00
8.00
1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00
Ethanol Price ($/bu.)
Cor
n Pr
ice
$/bu
.)
RFS Mandate for Corn-Based Ethanol, 2008/09 – 2015/16
3.6
4.1
4.44.6
4.95.1
5.3 5.4
3.0
3.5
4.0
4.5
5.0
5.5
2008
/09
2009
/10
2010
/11
2011
/12
2012
/13
2013
/14
2014
/15
2015
/16
Marketing Year
Cor
n (b
il. b
u.)
Monthly U.S. Ethanol Production + Imports vs. RFS Mandate, October 2007 - January 2009*
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Oct-07
Dec-07
Feb-08
Apr-08
Jun-08
Aug-08Oct-
08Dec
-08Feb
-09Apr-0
9Ju
n-09Aug-09Oct-
09Dec
-09Et
hano
l (bi
l. ga
l./m
o.)
Mandate
Production + Imports
Source: EIA/EPA *Dec 08 and Jan 09 Production + Imports Projected
Convergence
The pattern of cash and futures prices tending to come together, that is, basis approaching zero at the delivery market as the futures contract expires
Price
Expiration
Futures
Cash
Basis
Arbitrage and Convergence
• In theory, arbitrage in the cash and futures markets should force prices to converge (law of one price)– Futures > cash price: buy cash commodity, sell
futures, and deliver– Futures < cash price: Buy futures, stand for delivery,
and then sell cash
• The existence of delivery options and costs of arbitrage means that convergence should be thought of as a range of basis, not necessarily a zero basis– Direct costs estimated to be 6 to 8 cents per bushel
Delivery Location Basis on the First Day of Delivery for CBOT Corn Futures, Illinois River North of Peoria,
March 2000-March 2009
-200
-160
-120
-80
-40
0
40M
ar-0
0
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Bas
is (c
ents
/bu.
)
Delivery Location Basis on the First Day of Delivery for CBOT Soybean Futures, Illinois River North of Peoria,
January 2000 – March 2009
-200
-160
-120
-80
-40
0
40Ja
n-00
Aug
-00
Mar
-01
Sep-
01
May
-02
Nov
-02
Jul-0
3
Jan-
04
Aug
-04
Mar
-05
Sep-
05
May
-06
Nov
-06
Jul-0
7
Jan-
07
Aug
-08
Mar
-09
Contract Expiration Month
Bas
is (c
ents
/bu.
)
Delivery Location Basis on the First Day of Delivery for CBOT Wheat Futures, Toledo, March 2000-March 2009
-200
-160
-120
-80
-40
0
40M
ar-0
0
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Bas
is (c
ents
/bu.
)
Problems Created by Non-Convergence
• Wedge between futures and cash indicates out-of-balance contracts
– Hieronymus (1977, p. 340) warns, “When a contract is out of balance the disadvantaged side ceases trading and the contract disappears.”
• Increased basis uncertainty and loss in hedging effectiveness
– Long-run viability of markets is threatened
Major Factors Contributing to Non-Convergence
• Spreads reflecting a relatively high percent of full carry– Corn, soybeans, and wheat
• Structural issues related to the delivery process– Wheat
Full Carry and the Decoupling of Cash andFutures Markets
Bottom line: Arbitrage link between cash and futures broken
Delivery Takers Hold Certificatesand Sell Deferred Futures
Spreads Go to 100% of Full Carry
No Load Out to Cancel Certificates
% Full Cost of Carry Calculation
% = [(F2 – F1)/(Storage + Interest Costs)]*100
• F2 = Price of next nearest to expiration futures contract
• F1 = Price of nearest to expiration futures contract
• Storage = CBOT contract rate x # days• Interest = (3 mo. LIBOR rate)/365 x # days
Spread on the First Day of Delivery between Prices of the Expiring and Next-to-Expire Contracts for CBOT Corn Futures,
March 2000-March 2009
0%
20%
40%
60%
80%
100%
120%
Mar
-00
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
% o
f Ful
l Car
ry
Spread on the First Day of Delivery between Prices of the Expiring and Next-to-Expire Contracts for CBOT Soybean
Futures, January 2000 – March 2009
0%
20%
40%
60%
80%
100%
120%
Jan-
00
Aug
-00
Mar
-01
Sep-
01
May
-02
Nov
-02
Jul-0
3
Jan-
04
Aug
-04
Mar
-05
Sep-
05
May
-06
Nov
-06
Jul-0
7
Jan-
07
Aug
-08
Mar
-09
Contract Expiration Month
% o
f Ful
l Car
ry
Spread on the First Day of Delivery between Prices of the Expiring and Next-to-Expire Contracts for CBOT Wheat
Futures, March 2000-March 2009
0%
20%
40%
60%
80%
100%
120%M
ar-0
0
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
% o
f Ful
l Car
ry
Total Deliveries of CBOT Corn Futures, March 2000-March 2009
0
20
40
60
80
100
120
140
160M
ar-0
0
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Del
iver
ies
(mil.
bu.
)
Total Deliveries of CBOT Soybean Futures, January 2000 – March 2009
0
20
40
60
80
100
120
140
160Ja
n-00
Aug
-00
Mar
-01
Sep-
01
May
-02
Nov
-02
Jul-0
3
Jan-
04
Aug
-04
Mar
-05
Sep-
05
May
-06
Nov
-06
Jul-0
7
Jan-
07
Aug
-08
Mar
-09
Contract Expiration Month
Del
iver
ies
(mil.
bu.
)
Total Deliveries of CBOT Wheat Futures, March 2000-March 2009
0
20
40
60
80
100
120
140
160M
ar-0
0
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Del
iver
ies
(mil.
bu.
)
Daily Total of Registered Shipping Certificates for CBOT Corn Futures, July 2003-March 2009
0
5
10
15
20
25
30
35
40Ju
l-03
Nov
-03
Mar
-04
Jul-0
4
Nov
-04
Mar
-05
Jul-0
5
Nov
-05
Mar
-06
Jul-0
6
Nov
-06
Mar
-07
Jul-0
7
Nov
-07
Mar
-08
Jul-0
8
Nov
-08
Mar
-09
Date
Cer
tific
ates
/Rec
eipt
s (m
il. b
u.)
Daily Total of Registered Shipping Certificates for CBOT Soybean Futures, July 2003- March 2009
0
5
10
15
20
25
30
35
40Ju
l-03
Nov
-03
Mar
-04
Jul-0
4
Nov
-04
Mar
-05
Jul-0
5
Nov
-05
Mar
-06
Jul-0
6
Nov
-06
Mar
-07
Jul-0
7
Nov
-07
Mar
-08
Jul-0
8
Nov
-08
Mar
-09
Date
Cer
tific
ates
/Rec
eipt
s (m
il. b
u.)
Daily Total of Registered Shipping Certificates or Warehouse Receipts for CBOT Wheat Futures, July 2003- March 2009
0
5
10
15
20
25
30
35
40Ju
l-03
Nov
-03
Mar
-04
Jul-0
4
Nov
-04
Mar
-05
Jul-0
5
Nov
-05
Mar
-06
Jul-0
6
Nov
-06
Mar
-07
Jul-0
7
Nov
-07
Mar
-08
Jul-0
8
Nov
-08
Mar
-09
Date
Cer
tific
ates
/Rec
eipt
s (m
il. b
u.)
Basis and Percent of Full Carry on First Day of Delivery for CBOT Corn Futures, Illinois River North of Peoria, March 2000-
December 2008
-200
-160
-120
-80
-40
0
40M
ar-0
0Se
p-00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03M
ay-0
4
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Bas
is (c
ents
/bu.
)
-220%
-140%
-60%
20%
100%
% o
f Ful
l Car
ry
80%
% Carry (right scale)
Basis (left scale)
Basis and Percent of Full Carry on First Day of Delivery for CBOT Soybean Futures, Illinois River North of Peoria, March
2000-December 2008
-200
-160
-120
-80
-40
0
40
Jan-
00A
ug-0
0M
ar-0
1Se
p-01
May
-02
Nov
-02
Jul-0
3Ja
n-04
Aug
-04
Mar
-05
Sep-
05M
ay-0
6N
ov-0
6Ju
l-07
Jan-
07A
ug-0
8M
ar-0
9
Contract Expiration Month
Bas
is (c
ents
/bu.
)
-220%
-140%
-60%
20%
100%
% o
f Ful
l Car
ry
80%
% Carry (right scale)
Basis (left scale)
Basis and Percent of Full Carry on First Day of Delivery for CBOT Wheat Futures, Toledo, March 2000-December 2008
-200
-160
-120
-80
-40
0
40
Mar
-00
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
Contract Expiration Month
Bas
is (c
ents
/bu.
)
-220%
-140%
-60%
20%
100%
% o
f Ful
l Car
ry
80%
% Carry (right scale)
Basis (left scale)
Explaining the Large Carry
1. CBOT maximum storage rates below actual commercial storage costs
2. Presence of large “long-only” index funds
3. Risk premium due to increased uncertainty
4.5 cents7.1 centsWheat
4.5 cents4.6 centsSoybeans
4.5 cents4.3 centsCorn
Contract RatesCBOT Survey
Mid-2008 Comparison of Commercial Storage Costs and CBOT Contract Rates
“Goldman Roll” Effect on the Nearby Futures Spread
F2 - F1
Expiration of Contract 1
0
Beginning of Roll Window
Average Nearby Spreads for CBOT Corn Futures during the Roll Window of Long-Only Index Funds, March 1995 - March
2009 Contracts
20
37
61
90
22
42
70
93
18
37
68
90
0
20
40
60
80
100
Mar 1995 - Dec 2001
Mar 2002 - Dec 2003
Mar 2004 - Dec 2005
Mar 2006 - Mar 2009
% o
f Ful
l Car
ry
Days 1-4 Days 5-9 Days 10-13
Average Nearby Spreads for CBOT Soybean Futures during the Roll Window of Long-Only Index Funds, January 1995 - March
2009 Contracts
21
-34
30
85
21
-26
36
87
16
-32
28
82
-40
-20
0
20
40
60
80
100
Mar 1995 - Nov 2001
Jan 2002 - Nov 2003
Jan 2004 - Nov 2005
Jan 2006 - Mar 2009
% o
f Ful
l Car
ry
Days 1-4 Days 5-9 Days 10-13
Average Nearby Spreads for CBOT Wheat Futures during the Roll Window of Long-Only Index Funds, March 1995 - March
2009 Contracts
40 43
79
105
51 55
84
106
48 46
78
100
0
20
40
60
80
100
120
Mar 1995 - Dec 2001
Mar 2002 - Dec 2003
Mar 2004 - Dec 2005
Mar 2006 - Mar 2009
% o
f Ful
l Car
ry
Days 1-4 Days 5-9 Days 10-13
Spread on the First Day of Delivery between Prices of the Expiring and Next-to-Expire Contracts for CBOT Corn,
Soybean, Wheat Futures, March 2000-March 2009
0.00
0.20
0.40
0.60
0.80
1.00
1.20
Mar
-00
Sep-
00
May
-01
Dec
-01
Jul-0
2
Mar
-03
Sep-
03
May
-04
Dec
-04
Jul-0
5
Mar
-06
Sep-
06
May
-07
Dec
-07
Jul-0
8
Mar
-09
% o
f Ful
l Car
ry
Corn Soybeans Wheat
Correlations
C/S: +0.17
C/W:+0.38
S/W: +0.04
Risk Premium in the Carry – Craig Pirrong
• Positive shock to volatility of fundamental uncertainty increases the precautionary demand for grain inventories– Like increased demand for cash in uncertain times
• Leads to an increase in the expected price of storage, as reflected in the spreads between near and deferred futures – Adds a risk premium component to spreads
Spread = Storage + Interest - Convenience + Risk Premium