<%BANNER%>

Commodity Index Investment and Wheat Futures Market

Permanent Link: http://ufdc.ufl.edu/UFE0042262/00001

Material Information

Title: Commodity Index Investment and Wheat Futures Market
Physical Description: 1 online resource (69 p.)
Language: english
Creator: Meng, Fang
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The turmoil in futures market in recent years becomes intense concern to the industry, the exchanges and the Commodity Futures Trading Commission (CFTC). In my thesis, I examine the role of speculation in the wheat markets and the effective regulatory system to prevent excessive speculation. First, I summarize wheat market performances during the 2006 to 2010. Second, I develop methods to examine how the activities of speculators in aggregate, have led to the wheat market disruption. The quantity methods find that firstly, the volatility of wheat futures has preformed strong heterosckedascity and the volume and open interest provide the powerful explanation of volatility when enter separately. Second, the speculator trading activities have led to the change of wheat futures return while the hedger activities show a comparative steady and little impact on the price changes. Third, futures and cash prices do not cointegrated in long term and the futures price does not contain much information to reflect the cash price. In addition, the derivatives trading through over-the-counter (OTC) market become more dynamic and dominate. Commodity index trader through purchase of commodity index could directly affect the futures price fluctuation. In my thesis, I also analyze the impact of index trader activity on the wheat futures market. The empirical tests show that the injection of index funds does cause the recent volatile in wheat futures market.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Fang Meng.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Weldon, Richard N.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042262:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042262/00001

Material Information

Title: Commodity Index Investment and Wheat Futures Market
Physical Description: 1 online resource (69 p.)
Language: english
Creator: Meng, Fang
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The turmoil in futures market in recent years becomes intense concern to the industry, the exchanges and the Commodity Futures Trading Commission (CFTC). In my thesis, I examine the role of speculation in the wheat markets and the effective regulatory system to prevent excessive speculation. First, I summarize wheat market performances during the 2006 to 2010. Second, I develop methods to examine how the activities of speculators in aggregate, have led to the wheat market disruption. The quantity methods find that firstly, the volatility of wheat futures has preformed strong heterosckedascity and the volume and open interest provide the powerful explanation of volatility when enter separately. Second, the speculator trading activities have led to the change of wheat futures return while the hedger activities show a comparative steady and little impact on the price changes. Third, futures and cash prices do not cointegrated in long term and the futures price does not contain much information to reflect the cash price. In addition, the derivatives trading through over-the-counter (OTC) market become more dynamic and dominate. Commodity index trader through purchase of commodity index could directly affect the futures price fluctuation. In my thesis, I also analyze the impact of index trader activity on the wheat futures market. The empirical tests show that the injection of index funds does cause the recent volatile in wheat futures market.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Fang Meng.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Weldon, Richard N.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042262:00001


This item has the following downloads:


Full Text





COMMODITY INDEX INVESTMENT AND WHEAT FUTURES MARKET


By

FANG MENG















A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2010





































2010 Fang Meng




























To mom and dad, the most important persons in my life









ACKNOWLEDGMENTS

The experience of two years study in University of Florida is the most precious

treasury in my life. During the past two years, I got so many help from my professors,

classmates and family. First, I would like to thank Dr. Weldon, chair of my supervisory

committee for ongoing support. His enlightened and inspiring guidance lead me to finish

my thesis and complete my masters program. Second, I would like to thank Dr.

Livingston to give me a chance to do research with him. Although the culture and

language difference, he is always patient and supportive, instructing me and correcting

my language problems. In addition, his knowledgeable made me have a fervent passion

towards finance studying. I also feel grateful to Dr. Winner. I took his three statistics

courses during my masters studying. He taught me the statistical theory and statistics

packages, which made it possible for me to apply the statistically analytical methods to

the whole research process.

Here, I also want to give my deeply gratitude to my parents. Throughout these

years, their love and encourage bring me much courage to overcome difficulties. I

should also thank Mr. Weiyi,Ni. Without his constantly urging me on to study hard, I will

not be at the point in my life. There are so many people I should express my thanks, this

thesis is my reward for all your help.









TABLE OF CONTENTS

page

A C K N O W LE D G M E N T S ......... ............................ ................................ .... ............... 4

LIST O F TA BLES .............. ......... ..................................................................... 7

LIS T O F F IG U R E S .................................................................. 8

A B S T R A C T ............... ........................................................................ ..... 9

CHAPTER

1 IN T R O D U C T IO N .................................................................................................... 1 1

2 THE DESCRIPTION OF WHEAT FUTURES MARKET .................................. 13

3 PROBLEMS OF THE WHEAT FUTURES MARKET ................................ 17

Price Change and Volatility............................ ......... 17
Open Interest and Volume ........... ...... ......................... 19
Convergence Problem ............................................................ ......... ........ ..........20
Speculations in the Wheat Futures Market ...... ........ ...... ............................ 24

4 RELATIONSHIP AMONG VOLATILITY, VOLUME AND OPEN INTEREST .......... 26

Literature Review ......... .. ................ ......... ........................................... 26
M methodology .................................................................................. ...... .... 26
ARC H-LM Test for A RCH Effect ...... ......... .................................... ... ............... 28
Relations between Return Volatility and Open Interest............................... 29
Relations between Return Volatility and Volume ............ ................................... 30
Relations between Return Volatility, Volume and Open Interest.......................... 31

5 SPECULATION BEHAVIORS IN THE WHEAT FUTURES MARKET ................ 33

Literature Review ......... .. ................ ......... ........................................... 33
M ethodology and D ata ........................................................................ ......... 35

6 THE CASH MARKET AND FUTURES MARKET ...................... ...... ............... 40

7 COMMODITY INDEX AND INDEX SPECULATION ........................................ 43

Standard Poor's-Goldman Sachs Commodity Index................ ............... 43
W heat M market and S & P G S C I........................................................ ... .. ............... 45

8 REGULATORY SYSTEM AND POLICY RECOMMENDATION........................... 50

Development of Futures Regulation ...... .... ............................. 50









Speculative Position Limits ............... .... ......... .................... 53
Reporting System ....................... ......... ..... .. ......... 55
Commitment of Trader Report ............................. .... .................... 55
Commodity Index Trade Supplement.......................... ................... 55
Special Calls ...... ...................... ........ ......... 56
Enron Loophole ................ ......... ....................... 56
Swaps Loophole .................. ...... .................. 57
Policy Recommendation ...................... ... ............. 59
Improve Transparency and Data Accuracy ......................... 60
Improve Regulation on Index Trading Activity ...................... ...... ......... 61

9 C O N C L U S IO N ............... ... ...................... ......................... 62

LIST O F REFERENCES ......... ..... ........ .......... ............... ............... 64

B IO G RA PH ICA L S K ETC H ............. ...................................................... ............... 69



































6









LIST OF TABLES


Table page

4-1 Summary Statistics for Daily Return ................ ....... ..... ................... ............. 27

4-2 Summary Statistics for Futures Returns ........................ ....................... 29

4-3 GARCH (1,1) Estimate for Volatility and Open Interest ................................... 30

4-4 GARCH (1,1) Estimate for Volatility and Volume......................... ......... ..... 31

4-5 GARCH (1,1) Estimate for Volatility, Open Interest and Volume..................... 31

5-1 S um m ary S statistics .......... ....... ......... ............... ................ ........... 36

5-2 Unit Root Test and ARCH Affect Test .............. ............ ............... ............ 38

5-3 Granger-Causality Wald Test ...... ..................... .................. 38

5-4 W hite Noise Diagnostics....................... ..... ............................. 38

5-5 C orrelation A nalysis............................................................... 39

6 -1 U n it R o ot T e st................... ........................... ...................... 4 1

6-2 Cointegration Rank Test ................................................ 42

7-1 Pearson Correlation Coefficients Prob > Irl under HO: Rho=0.......................... 46

7-2 Summary Statistics and Unit Root Test................. ........ ................ 48

7-3 C ausality W ald Test................... .............................................. 48

7-4 Model Parameter Estimates .......................... .................... 48

8-1 Current Practices in Setting Speculative Position Limits and Position
Accountability Levels In selected Futures Contract Market as of July 24, 2009
Source: Federal R regulation ......... ................. ........................... ............... 54

8-2 Speculative Position Limit (2010) in Contract Units Source:Federal
Regulation .............................. .................. ........ ....... ......... 54









LIST OF FIGURES

Figure age

2-1 Actively Trade Future &Option Contracts 1997-2009................................. 14

2-2 Growth in Volume of Futures& Option Contracts Traded, 1998-2008 ................ 14

2-3 U.S. Wheat Production. Types and amounts of wheat grown in the United
States in 2009-2010. .............. ....... .. .................................. 15

3-1 Continuous Futures Price. .......... ..................... ................ ... 18

3-2 V volatility of C hicago W heat Futures.............................................. ... .................. 19

3-3 Volume and Open interest Growth-Chicago Wheat............ ............... 20

3-4 March Wheat Basis (Toledo, OH cash less wheat CBOT futures).................... 22

3-5 May Wheat Futures Basis (Toledo, OH cash less wheat CBOT Futures) .......... 22

3-6 July Wheat Basis (Toledo, OH cash less Wheat CBOT futures) ...... ........ 23

3-7 September Wheat Basis (Illinois Cash Less Wheat CBOT Futures) ................ 23

3-8 December wheat Basis (Toledo, OH cash less Wheat CBOT futures) .............. 24

3-9 Percentage of Open Interest for Chicago Wheat Market Held by Participants
(% ). .............. ......................................................................................................... 2 5

5-1 Open interest: Wheat in Chicago Board of Trade ...... ................................. 33

5-2 Speculators Position(% ) ....................... ............... ............... .............. 37

5-3 H edgers Position(% ) ............................................... .............. 37

5-4 Daily Return(% ) ................................. ..... ........................ 37

7-1 S&P GSCI Index Weight Composition.................... .... .................. 44

7-2 Daily Return of GSCI Futures(% ) ............... ................. ................................. 46

7-3 Daily Return of Wheat Futures(%)......................... ... .......................... 47









ABSTRACT OF THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE


COMMODITY INDEX INVESTMENT AND WHEAT FUTURES PRICES
By

Fang Meng

August 2010

Chair: Richard Weldon
Major: Food and Resource Economics

The turmoil in futures market in recent years becomes intense concern to the

industry, the exchanges and the Commodity Futures Trading Commission (CFTC). In

my thesis, I examine the role of speculation in the wheat markets and the effective

regulatory system to prevent excessive speculation. First, I summarize wheat market

performances during the 2006 to 2010. Second, I develop methods to examine how the

activities of speculators in aggregate, have led to the wheat market disruption.

The quantity methods find that firstly, the volatility of wheat futures has preformed

strong heterosckedascity and the volume and open interest provide the powerful

explanation of volatility when enter separately. Second, the speculator trading activities

have led to the change of wheat futures return while the hedger activities show a

comparative steady and little impact on the price changes. Third, futures and cash

prices do not cointegrated in long term and the futures price does not contain much

information to reflect the cash price.

In addition, the derivatives trading through over-the-counter (OTC) market become

more dynamic and dominate. Commodity index trader through purchase of commodity

index could directly affect the futures price fluctuation. In my thesis, I also analyze the









impact of index trader activity on the wheat futures market. The empirical tests show

that the injection of index funds does cause the recent volatile in wheat futures market.









CHAPTER 1
INTRODUCTION

The futures market in the United Stated experienced an extremely volatile period

in recent years. The wheat futures traded on Chicago Mercantile Exchange (CME), for

example, increased from $3.00 per bushel in 2006 to $11.00 per bushel in the mid-2008,

and afterwards declined sharply back to $3.00 per bushel at the end of 2008. Moreover,

many other commodities endured the similar performances. First, trading volume and

open interest increased tremendously. The prices of commodities, such as oil, wheat

and soybean have reached unprecedented levels. Second, the cash and futures prices

failed to converge during the expiration month. Third, large amount of index have been

observed in the commodities derivative markets.

The unexpected changes in commodity prices has led to intensive investigations

by the Commodity Futures Trading Commission (CFTC) and has been widely debated

by academia. Many recent testimonies and Congress reports show that the distortion of

the futures market and even the financial recession are because of the failures of the

regulatory structure and agencies. Since 2000, the U.S. congress has continuously

deregulated the over-the-counter (OTC) market, in order to make the U.S. futures

market more flexible and competitive in the global market. The commodity index traders

are exempt from position limits when purchase commodity futures contracts to offset

their business risk.

In this thesis, firstly, it provides a comparative description of the wheat futures

market, price change and volatility, market depth and liquidity, cash and futures prices

and trader position. After market analysis, this thesis will go in detail to investigate the

speculation behaviors in wheat futures market during extreme volatility period. I am









trying to find evidence to examine whether the speculators activates cause the distortion

of wheat futures market.

In addition, I differentiate the speculators into index speculators and traditional

speculators, since there are somewhat different from each other. I tried to examine the

role of the commodity index trading and investigate casualty relationship between

commodity index trade and wheat futures price.









CHAPTER 2
THE DESCRIPTION OF WHEAT FUTURES MARKET

The first organized futures market in U.S. Chicago Board of Trade was established

in 1848. At that time, because of its unique location and access to the Great Lakes,

Chicago became the transportation and distribution pivot for agriculture commodities.

Initially, it severed as a market place for farmers to sell their products, including grain,

wool, pork, beef and lumber before delivering to lock in a price. At the beginning, the

trading was a type of forward contract. In 1865, standardized futures contracts were

introduced. The original trading activates were carried on the floors, known as "floor

trading". The traders stood on the floor, shouting the price they were willing to pay or

accept. They simply reached an agreement on the price and number of contracts. The

clearinghouse, which established in the 1920s and guaranteed against default, is a

significant development for the modern futures market.

Since its establishment, the futures market in U.S. has achieved great

development in less than 200 years. The Chicago Board of Trade, the New York

Mercantile Exchange, Kansas Board of Trade and Intercontinental Exchange developed

into the world's leading futures exchanges. Meanwhile, the futures market exceeded the

agricultural commodities range. Numerous new futures and options based on such

things as foreign exchange rates, interest rates, equity index and even credit ratings are

listed on the exchanges. The number of future contracts expanded from 266 in 2000 to

1730 in 2009 (Figure 2-1).Since 2000, volume growth on futures market has increased

sevenfold (Figure 2-2).



















1200- 1135

0
1000- 906
9

800-

Z 600- 538


250 286 251 266 250 278
200-



1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Fiscal Year



Figure 2-1. Actively Trade Future and Option Contracts 1997-2009 Source: FY 2009
President's Budget & Performance Plan


Fiscal Year


Figure 2-2. Growth in Volume of Futures and Option Contracts Traded, 1998-2008
Source: FY2009 President's Budget & Performance Plan









There are five basic types of wheat growing in the United States: hard red winter,

hard red spring, soft red winter, durum and white. Figure 2-3 illustrates the percentage

of production for each type of wheat during 2009. Nowadays, in the United States, three

major futures exchanges specialize in a particular type of wheat. These wheat futures

contracts specify the type and quality of the wheat to be delivered, period for delivery,

the possible locations for delivery and the required price at the time of delivery. The soft

red winter wheat is traded on the Chicago Mercantile Exchange (CME), hard red winter

wheat is traded on the Kansas City Board of Trade (KCBT) and hard red spring wheat is

traded on the Minneapolis Grain Exchange (MGE).






/Soft red
winter
18%

Hard red spring
25%





Figure 2-3. U.S. Wheat Production. Types and amounts of wheat grown in the United
States in 2009-2010.Source: USDA, Economic Research Service and Wheat
Data: Yearbook Tables, Table 6.

The CME soft red winter wheat contract provides sale and delivery of #2 soft red

winter wheat. Currently, there is one warehouse in Chicago and two facilities in Toledo.

Delivery usually takes a multi-step process several days prior to the expiration of the

contract. Wheat futures contracts usually expire on the first business day prior to the

15th calendar day of the contract month. About one month prior to the expiration, an









approved firm gives notice of the plan to make a delivery. On next day, the exchange

will examine the outstanding long open interest and selects those positions that have

been held for the longest period to accept delivery. On that day, the holder of the long

position may either accept the delivery or pass it on to another buyer. The following day

is called delivery day, when both the long and short open interests are closed.

Delivering wheat under contract from an approved warehouse does not require

the physical movement. Instead, it is through exchanging the right in the form of a

shipping certificate. Under the request by certification holder, the warehouse should

have to load the grain within a period specified by the rules of exchange.

Through this delivery process, traders do not necessarily physically handle any

wheat. Then traders may store grain in grain elevators, in the form of shipping

certificates. They could hold these shipping certificates, pay the storage fees, and sell

the wheat when price increases.









CHAPTER 3
PROBLEMS OF THE WHEAT FUTURES MARKET

In this chapter, we will first evaluate wheat futures market performances from the

following perspectives: price change and volatility, market depth and liquidity, cash and

futures prices and trader position

Price Change and Volatility

To look at the trend of futures price, we have to create a series of futures price.

Since at any particular date, more than five futures contracts on wheat may be traded.

For example, futures with expiration in July this year and July next year may be traded

simultaneously. I took the price of nearby expiration contract as the reference price and

created a continuous futures price for each contract. Figure 3-9 shows the trend of

wheat futures prices from 2006 to 2010.During the past three years, the wheat futures

price has experienced the spike and collapse. First, wheat futures price continue to

surge high form $4.83 per bushel to $12.99 per bushel in a year. It reached the peak at

$12.99 per bushel in the February 2008. Then the wheat contract price fell from $12.99

per bushel to $5.12 per bushel.









1,400

1,200

1,000

800

600

400

200

0


1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

Figure 3-1. Continuous Futures Price, Source: Bloomberg.

In addition, the volatility of wheat futures price has reached unprecedented level.

Data are from Jan 1, 2006 to Feb 28, 2010. Return volatility is calculated by the

standard deviation of daily return.

Volatility=(Xt X)2

Where Xt= l001n ( )

Daily futures return is defined as 100xln(Ft/Ft) where Ft is the futures


settlement price at a given day t. The logarithm transformation improves the statistical

properties, because it provides a clearer tendency of the variable than untransformed

series. Most macroeconomics time series data are estimated based on logarithmic

scale over time. Being a squared quantity, volatility calculated will be high in periods

when there are big changes in the prices and comparatively small when there are

modest changes in price compared to the previous year, the volatility of price has

increased drastically since 2008 (Figure3-2).


I I I I


_ -------










1.2000

1.0000

0.8000

0.6000

0.4000

0.2000

0.0000
1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010


Figure 3-2. Volatility of Chicago Wheat Futures. Source: Bloomberg

Traditionally, when a farmer, or a grain elevators uses the futures contract to

hedge their business risks, they are required to maintain working capital as margin and.

If they fail to make a margin call as needed, their futures positions are immediately

liquidated. During the extremely volatility period, hedgers has incurred a large margin

calls on a daily basis, which require hedger to have much larger working capital than

ever before for their futures positions. According to National Grain and Feed Association,

the hedging cost has tripled since 2008.

Open Interest and Volume

Open interest refers to the total number of contracts that are not exercised, offset

or delivered and an appropriate measure of participant activities. For every long or short

trade, there must be a counterpart trade, or the contract is considered "open". An

increasing open interest usually indicates new money flowing into the market.

Volume measures market intensity. It represents the total amount of trading

activity in a given day. The greater the amount of trading activities, the higher will be

volume. Figure 3-3 shows the trading volume and open interest in the Chicago wheat










market from 2001 to 2010. Before 2005, both volume and open interest maintained a

lower move. However, after 2005, the drastic increase in volume and open interest have

been observed.

600000 250000


500000
200000

400000
150000
300000
100000
200000

50000
100000


0 .... 0
1/3/2001 1/3/2002 1/3/2003 1/3/2004 1/3/2005 1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

-FUT AGGTE OPEN INT --FUT AGGTE VOL


Figure 3-3. Volume and Open interest Growth-Chicago Wheat. Source: Bloomberg

Convergence Problem

Basis is the current cash price of a given commodity at a location minus the price

of a particular futures contract for the same commodity. Several features of this need to

be explained. First, basis depends upon a cash price of a commodity at a specific

location. The cash price of wheat, for example, might differ between Kansas City and

Chicago, so that the basis for those two locations will be different. If wheat had two

different prices in two locations, a speculator could potentially buy the commodity in the

cheaper market and sell it in the market with the higher price until arbitrage profit

disappear. However, prices for wheat in Chicago and Kansas City can base on the

expense of transporting wheat from one market to another. If wheat is grown near

Chicago, then we might reasonably expect the price of wheat in Chicago to be lower









than the price of wheat in Kansas City. For CBOT wheat, the key delivery area is in

Toledo, Ohio. The cash price I use to analyze is based on the wheat price reported in

Toledo, Ohio.

Usually, when a futures contract gets closer to expiration, the futures price will

converge to the price in the cash market. However, over the past three years, the

futures prices for wheat were abnormally high. They also failed to converge with the

cash price as the futures contract neared expiration. The CBOT wheat futures have five

expiration months: March, May, July, September, and December. From Figures3-4 to

Figure 3-8 ,they suggest that the basis for five delivery months of wheat future contract.

The data of futures price were obtained from Bloomberg and the spot prices are from

the USDA, Agricultural Market Service. The last observation for each contract every

year is the expiration day, usually around the 15th of the calendar month. The figures

suggest that the poor convergences exist in all five wheat contracts. If a futures contract

fails to converge at the expiration, it will severely impair the ability of the farmers, grain

elevators or other merchants use the futures markets to hedge their exposure to price

risk over time.











$1.5
$1.0
$0.5
$0.0 ..... ... ...... ...
$0.5

$1.0
$1.5
$2.0 t
$2.5
$3.0
14/01/04 28/05/05 10/10/06 22/02/08 6/07/09 18/11/10



Figure 3-4. March Wheat Basis (Toledo, OH cash less wheat CBOT futures)

$1.5

$1.0

$0.5 -

$0.0 .. .. ... ,. ,. .. .. .. ..

$0.5 -

$1.0 -I f

$1.5 -

$2.0 ;

$2.5 -

$3.0 -
17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10


Figure 3-5. May Wheat Futures Basis (Toledo, OH cash less wheat CBOT Futures)










$3.0


$2.0 -i






$0.0 .
$1.0
$2.0



$3.0
17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10


Figure 3-6. July Wheat Basis (Toledo, OH cash less Wheat CBOT futures)

$3.0

$2.0 -

$1.0

$0.0 ..... ..

$1.0

$2.0 -

$3.0o

$4.0
17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10

Figure 3-7. September Wheat Basis (Illinois Cash Less Wheat CBOT Futures)










$2.0
$1.5
$1.0 -
S'S
$0.5O
$0.0
$0.5 "
$1.0
$1.5 r
$2.0
$2.5
5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10 10/08/10

Figure 3-8. December wheat Basis (Toledo, OH cash less Wheat CBOT futures)

Speculations in the Wheat Futures Market

Figure 3-9 illustrates the relative trader's position in the Chicago wheat futures

market. The data were obtained for Commitment of Traders (COT) on the CFTC

website. The Commitment of Traders (COT) report provides a breakdown of each

Tuesday's open interest for markets with 20 or more traders' positions equal to or more

than the reporting level set by the CFTC. The CFTC collects data from the

clearinghouse and classifies traders into either commercial or non-commercial. The

commercial category represents traders using futures market to hedge. The non-

commercial category generally includes pension funds, hedge funds and other type of

fund who profit from the prices changes in the futures market.

Historically, the demand by the hedger has exceeded the demand by speculator in

the futures market. Since there is difference in the number of hedgers and speculators,

hedgers have to make an attractive offer to get the speculators to take the other side of

the transaction. However, in Figure 3-9 it obviously shows that speculators have








dominated the futures market. They have taken more than 50 percent of open interest in

the futures market over the past three years. Michael Master in his one testimony has

proposed that the appropriate open interest taken by speculator should be in the range

of 25%-35%. It believed that this excess speculation has undermined the integrity of

market.


80
70
60
50
-
40
30
20
10
0
V/


-PctofOl hedgersLong All


Pct of Ol_Speculators_Long_All


Figure 3-9. Percentage of Open Interest for Chicago Wheat Market Held by Participants
(%)









CHAPTER 4
RELATIONSHIP AMONG VOLATILITY, VOLUME AND OPEN INTEREST

Literature Review

In the Chapter 3, we have found several problems in the wheat market. From the

Chapter 4 to Chapter 7, we will continue to analyze these problems separately by using

quantity methods. We begin with investigating the relations between among volume,

open interest and volatility to uncover the source of variability in the wheat futures prices.

The trading volume and open interest are two important indices of performance

measurement for futures market. Their relationship are utmost importance for one

reason, that is a well understanding the link could contribute to a superior speculation

and hedge strategies. Most empirical researches found the positive relationship

between trading volume and price volatility. Conell (1981) Tauchen Pitts (1983) and

Karpoff (1987) find the positive relations between trading volume and price volatility.

Bessembinder and Seguin (1993) study the impacts of volume and open interest on

price volatility. They differentiated the volume and open interest into expected and

unexpected categories. They find that both of them have positive effects on price

volatility. However, the unexpected volume tends to have a larger impact on volatility

than expected volume. Girma and Mougoue (2001) investigate the relation between the

petroleum futures spread variability, trading volume and open interest. Their study

shows the contemporaneous weighted average of volume and open interest provide

significant explanation for futures spread volatility.

Methodology

Table 4-1 provides summary statistics for daily return of wheat futures. The mean

of 0.036 is very close to 0 and Standard deviation is 2.57. The skewness and kurtosis









are 0.1796 and 2.6398 respectively. The kurtosis is less than three, the distribution has

thicker tails and a lower peak compared to a normal distribution. However, these

descriptive statistics do not provide conclusive information about normality. SAS

provides four different statistics for testing normality. Shapiro-Wilk W of .979 (P-

value<0.0001), which rejects the null hypothesis of normality distribution. Similarly,

Kolmogorov-Smirnov, Cramer-von Mises, and Anderson-Darling tests reject the null

hypothesis. Therefore, we can conclude that the variable is not normally distributed.

Table 4-1. Summary Statistics for Daily Return
Cramer-
Std Shapiro- Kolmogorov- Cr r- Anderson
Mean Dev newness Wik Smirnov -Darling
Mises
Daily 1023 0.036 2.57 0.1796 2.6398 <0.0001 <0.0100 <0.0050 <0.0050
Reutn
The return series on wheat futures in Table 4-1 shows that it is characterized by

excess kurtosis and exhibit heteroscedasticity (Akgiray 1991, Hall 1989, Harvey and

Siddique 1999).Therefore, the Generalized ARCH model is appropriate to employ.

Engle first introduced the autoregressive conditionally heteroscedastic (ARCH) model in

1982 and Bollerslev developed a more general structure GARCH model in 1986.The

variance in the model is assumed to follow an autoregressive model. The approach to

GARCH (p, q) models is to set up an error term Etin terms of white noise.

Et = JTh e, et-(0,1) where ht is assumed to follow an autoregressive model:

ht = ao + =1jht- + 1Yii2

In the papers of Lamoureux and Lastrpes 1990 and Najand 1991, the GARCH (1,1)

have been proved to be adequate model to test the relations between price volatility and

volume. Since then, the GARCH and ARCH models have been widely used in financial

forecasting and derivatives pricing. Bollerslev (1986) using the GARCH model more

accurately describes the phenomenon of volatility clustering and related effects.









Foster(1995) uses the GARCH(1,1) and GMM model to investigate the relationship

between trading volume and price volatility in crude oil futures market. He finds the

contemporaneous volume is positively related to the price volatility. Fujihara and

Mougoue(1997) use the GARCH(1,1) model to study the crude oil, heating oil and

unleaded gasoline futures market. They conclude that the volume is a significant

explanatory in explaining the price volatility.

Volume is the number of buy and sell of futures contract for the Chicago wheat.

While the trading volume also represents the daily activities of hedger and long-time

trader, which is more transitory (Chang, Chou and Nelling2002). I analyze daily data

from January 1, 2006 through February 28, 2010. The sample consists of 1040

observations. Data on futures price, open interest and volume come from Bloomberg

DataStream. The returns are calculated as 100 times the natural logarithm of the first

differences of daily futures settlement prices Rt = 100 x In (Ft/Ft_1).The open interest

and volume are not adjusted.

ARCH-LM Test for ARCH Effect

A method proposed by Engle (1982) to test the lag length of ARCH errors is the

Lagrange Multiplier test. This procedure is to first estimate the best fitted AR(q) to obtain

the squares of the errors, Et and regress them on a constant and lagged values in the

following equation:

q
E 2 = w + ai S2_
i=l1

The null hypothesis is that ai = 0 for all i=1,2,3...n. The alternative hypotheses is

that at least one of the estimated ai coefficients is not equal to zero. Under the null










hypothesis of no ARCH errors, the test statistic TR2 follows X2 distribution

with q degrees of freedom. If TR2 is greater than the Chi-square table value,

we reject the null hypothesis and conclude the presence of an ARCH effect in the AR

model. Otherwise, we fail to reject the null hypothesis. Table 4-1 illustrates the

descriptive statistics for daily return. These tests strongly indicate that return series is

characterized by significant heteroscedasticity, with p < 0.0001 for all lags.

Table 4-2. Summary Statistics for Futures Returns
Q and LM Tests for ARCH Disturbance
Daily return Order Q Pr > Q LM Pr > LM

N 1040 1 25.5900 <.0001 25.6121 <.0001
Mean 0.02974 2 37.8803 <.0001 33.1802 <.0001
Sum Weights 1040 3 58.4281 <.0001 46.4873 <.0001
Sum 34.9258 4 71.8117 <.0001 51.6899 <.0001
Observations
Observations 6.0539 5 86.7485 <.0001 57.6692 <.0001
Variance
Variance 1.09832 6 89.7813 <.0001 57.6698 <.0001
Kurtosis
Kurtosis 2.4605 7 99.7232 <.0001 61.1580 <.0001
Std Deviation
Std Deviation -0.0690 8 102.6048 <.0001 61.1621 <.0001
Skewness
Skewness 6290.8752 9 108.1520 <.0001 62.6076 <.0001
Uncorrected SS
Uncorrected SS 8274.2457 10 111.2476 <.0001 62.7274 <.0001
Coeff Variation
0.07630 11 114.9901 <.0001 63.4837 <.0001
Std Error Mean 12 120.2228 <.0001 64.4779 <.0001



Relations between Return Volatility and Open Interest

The following model is a GARCH(1,1) model when open interest is entered. The

origins of analyzed technique can be traced back to Paul Girma and Mbodja Mougoue

(2002).

[ = u + Et


At = +2ht et

ht = ro + aE2_- + a2,ht_ + -9,OPINIt









Where rt is the futures return, OP_INI are open interest. Table 4-3 shows the result

of GARCH (1,1), when the open interest is used as an explanatory variable in the

conditional variation equation. The data suggests that the coefficient of open interest in

the equation is statistically significant at 5% level. Furthermore, the persistence shock is

a, + a2=0.9897, which remain highly volatility when the open interest was calculated in

the model. Besides, the estimator of a2(0.9570) is larger than that of a1(0.0327)

represents long persistence of volatility.

Table 4-3. GARCH(1,1) Estimate for Volatility and Open Interest
Standard
Variable DF Estimate t Value Pr > Itl
Error
Intercept 1 0.0622 0.0716 0.87 0.3846
ao 1 1.0537E-8 5.924E-11 177.85 <.0001
a, 1 0.0327 0.007866 4.16 <.0001
a2 1 0.9570 0.009734 98.31 <.0001
OP INI 1 1.6216E-7 6.1293E-8 2.65 0.0082
Relations between Return Volatility and Volume

The following model is a GARCH(1,1) model when volume is entered.

[t = P + Et

Et = /-ht et

ht = ao + ajEt21 + a2ht-1 + O1Vt

V represents the daily volume for wheat futures. Table 4-4 shows the result of

GARCH(1,1) when the volume is used as explanatory variable in the conditional

variation equation. The data shows that the volume is significant power explanatory for

explaining the return volatility. Furthermore, the persistence shock decrease from

0.9897 to a, + a2=0.9752, which a slight lower than the model 1. However, the

persistence shock maintains a high level.










Table 4-4. GARCH (1,1) Estimate for Volatility and Volume

Standard Approx
Variable
Variable DF Estimate Error t Value Pr > Itl

Intercept 1 0.0506 0.0719 0.70 0.4819
ao 1 1.0541E-8 5.718E-10 18.44 <.0001
a1 1 0.0336 0.008998 3.74 0.0002
az 1 0.9416 0.0134 70.22 <.0001
Volume 1 2.0028E-6 6.2309E-7 3.21 0.0013
Relations between Return Volatility, Volume and Open Interest

The following model is a GARCH (1,1) model when open interest and volume are

entered simultaneously.

t = Pu + Et

Et = /ht et

ht = ao + + a2ht-1 + 01Vt + O10P_INIt

Table 4-5 shows the result of GARCH (1,1), which the volume and open interest

have the significant effect on price volatility. The persistence of volatility reduced to

a, + a2 =0.8615.


Table 4-5. GARCH (1,1) Estimate for


Volatility, Open Interest and Volume


Standard Approx
Variable DF Estimate Error t Value Pr > Itl
/P 1 0.0193 0.0728 0.26 0.7912
ao 1 0.9336 0.4369 2.14 0.0326
aC 1 0.0666 0.0229 2.91 0.0036
a2 1 0.7949 0.0597 13.32 <.0001
Volume 1 9.5364E-6 2.8907E-6 3.30 0.0010
OP INI 1 -2.158E-6 8.4214E-7 -2.56 0.0104
Through this chapter, we could find that the contemporaneous volume and open

interest provide the power explanations for the volatility when enter separately. The

coefficients of open interest to volatility are negative and statistically significant. The









result confirm the Bessembinder and Seguin (1993) conclusion that greater market

depth is lower the volatility, given a trading volume. Besides open interest has a relative

little impact on volatility than a trade incurs a high trading volume without a

corresponding changing in open interest.









CHAPTER 5
SPECULATION BEHAVIORS IN THE WHEAT FUTURES MARKET

Since the open interest and trading volume, to certain extent, have led to the price

volatility, this chapter will go further to analyze the different participants on the price

movement. In Figure 3-9, it shows that the non-commercial long position takes a much

larger part than the commercial position. Meanwhile, Figure 5-1 indicates that the

commercial position maintains a slower and more stable change than non-commercial

position during the sampling period. The commercial long positions increased steadily,

however, not so much volatile as the non-commercial positions did.

600000

500000 -

400000 -

300000 -

200000

100000 *

0

/24/19912/6/1994/19/20091/2002/14/200/220010/1Q020202/20006209 2010


Open_lnterest_All NonComm_Positions_Long_AllI Comm_Positions_Long_All


Figure 5-1. Open interest: Wheat in Chicago Board of Trade

Literature Review

The relationship between price volatility and types of market participants has been

of interest to both scholars and regulators. Peck (1981) finds the relationship between

speculators and price volatility in three agriculture commodities. His study finds the

negative relationship between speculation activities and price volatility. Bryant, Bessler









and Haigh (2006) investigate the activities of specific types of traders as causes of price

volatility. They analyze eight futures: corn, crude oil, Eurodollar deposits, gold,

Japanese Yen, coffee, live cattle, and S&P 500 through vector autoregression (VAR).

They also estimate the causal relationships among return, volume, large hedgers'

activity, large speculators activity, and small hedger activity, small speculator activity.

They find that activity of particular types of traders in futures market did not affect levels

of price volatility, either positively or negatively.

Shalen develops the noisy rational expectations (NRE) model in 1993. He divides

the traders into two categories. He postulates that informed traders have private

information while the uniformed speculators have no private information. Therefore, the

uninformed traders will attempt to extract price signals from observed futures price

changes. His study shows that the uninformed speculators cause the increased price

volatility.

Daigler and Wiley (1999) examine various financial futures markets. They

conclude that trading activities taken on the trading floor are associated with the

decreased price volatility. However, the trading activity through electronic platform are

associated with increased volatility. Chang, Chou and Nelling (2000) investigate the

S&P 500 futures market. Their empirical study suggests that the large hedging activity

causes the increased volatility. They conclude that the large hedging trading activities

are correlated with volatility. The demand side of hedging increased, the volatility

increased as well.

Swanson and Granger(1997) study the causal relationship among the variables in

a vector autoregression. Reale and Wilson (2001) and Moneta (2004) research on









monetary policy issues by using a VAR model. Akleman, Bessler and Burton (1999)

investigate causal relationships among corn exports and exchange rates also using this

model. Haigh and Bessler (2004) research on price discovery in cash grain markets and

a related transportation market, again using this model. Bryant,Bessler and Haigh

(2006)tests causal hypotheses emanating from theories of futures markets through VAR

model

Methodology and Data

This study explores nonlinearities in the response of speculators' trading activity to

price changes in wheat futures markets. I analyze weekly data from January 1, 2006

through February 28, 2010. The sample consists of 216 observations. Data on futures

price come from Bloomberg DataStream. The returns are calculated as 100 times the

natural logarithm of the first differences of weekly futures settlement prices. The trader

position data are from Commitments of Traders (COT) reported on the official website of

Commodity Futures Trading Commission (CFTC). All of the traders' reported futures

positions are classified as commercial if a trader uses futures contracts for hedging or

as noncommercial if a trader uses the futures contract for profit.

Rothig and Chiarella (2007) defined changes in hedger and speculator's position

as 100 times the natural logarithm of the first differences of position respectively. This

method could effectively reflect the changes of position. Thus, I use VAR model to

measure the changes in hedger and speculators long positions.

Table 5-1 shows the summary statistics for the return and changes in speculator

and hedger positions. The distributions of speculators indicate significant positive

skewness. Furthermore, the speculator shows evidence of slightly excess kurtosis









compare to the normal distribution. Harvery and Siddique (1999) shows that the excess

skewness and kurtosis potentially indicate the presence of heteroscedasticity.

Table 5-1. Summary Statistics

Cramer-
Shapiro- Kolmogorov- Anderson-
N Mean Std Skewneww Kurtosis irVon Darling
Wilk Smirnov Mises Darling
Mises

Speculator 216 0.1175 6.5002 0.1141 1.9833 0001 <0.0100 <0.0050 <0.0050
0.0001
Hedger 216 0.1342 2.6175 0.1377 0.2140 0.3116 >0.1500 0.2290 0.1861
Return 216 0.1627 5.3436 -0.0240 0.02101 0.9182 >0.1500 >0.2500 >0.2500
To avoid the spurious regression problem, I took the stationarity test for the three

categories. The Dickey and Fuller (1979) have developed a method to test the

stationary series. The method has become very popular over the past thirty years. The

method is also known as Augmented Dickey-Fuller (ADF) test. The ADF test consists of

estimating the following equation;

n
AXt = bo + blXt-1 + Yj(AXt-j) + Et
j=1

Where AXt is the first difference value of X, Xt_, is the first-lagged valued of X.

AXt_i is the jth lagged of first difference of values of X. The null hypothesis of non-

stationary is b, = 0. Dickey and Fuller have shown that the t value for the b, follows the

T statistics. The Table 5-2 shows that the T statistics are -10.38 for speculator, -8.63 for

hedger and -9.97 for return. All of them are significant at 5% level. Thus the speculator

position changes, hedge position changes and return display stationary fluctuations.











20
10
0
-10
-20
-30


Figure 5-2. Speculators Position(%)




10-

5

0-



-10 -


Figure 5-3. Hedgers Position(%)

20
15 -
10
5
0
-5 $ I V0 RIti5 "I 9
-10 Ln r n r m
-15
-20


Figure 5-4. Daily Return(%)


II AI


I1hl h i Al hA A


r 0t1 1 r "r j r ( T) m m r raT) (T)) (T) rl (' r' ( (T) (T) ITv) I
SL n r m Ln r, ( r Ln r Ln. r, (m i










Table 5-2. Unit Root Test and ARCH Affect Test

Lag Tau Pr < Tau Q Pr > Q LM Pr > LM

Speculator 1 -10.38 <0.0001 9.8737 0.0017 9.7982 0.001
2 -9.33 <0.0001 10.3734 0.0056 9.7998 0.0074
Hedger 1 -8.63 <.0001 0.1818 0.6699 0.1417 0.7066
2 -7.25 <.0001 0.4774 0.7877 0.4613 0.7940
Return 1 -9.97 <.0001 8.2822 0.0040 8.2903 0.0040
2 -8.42 <.0001 8.2986 0.0158 8.4636 0.0145


The results of the Granger causality test in my study are presented in the

Table 5-4.The no causality null hypothesis can be rejected for changes in commercial

hedger long positions and return. The hedging strategies did not affect the changes in

futures prices. These results support the theory that hedgers hold futures positions

corresponding to their spot positions. On the other hand, the test shows the significant

evidence that the speculators affect the fluctuation of return. Therefore, it could

conclude that the extreme volatility is caused primarily by speculators.

Table 5-3. Granger-Causality Wald Test
Test DF Chi-Square Pr > ChiSq


1 1 0.13 0.7136

2 1 4.97 0.0258

Test 1: Group 1 Variables: price Group 2 Variables: Commercial
Test 2: Group 1 Variables: price Group 2 Variables: Noncommercial
The Table 5-4 presents the model diagnostic analysis, autocorrelation, normality

and ARCH effect test. The normality shows that the residual has a normal distribution

and no ARCH effect.

Table 5-4. White Noise Diagnostics
Variable Durbin Watson Normality ARCH
Chi-Square Pr > ChiSq F -Value Pr > F
Return 2.03170 2.54 0.2807 2.68 0.1030









In an effort to improve the transparency of the futures market, in 2006, the CFTC

removed the swaps position from the hedger position to measure the trading activates

through the OTC market. Therefore, by using the daily data, we calculated the

correlations between market volatility and the presence of each trader type, including

hedger, swaps dealers, speculators and others. The market presence represents the

percentage of all open interest held by a trade group.

Volatility appears positively related to the presence of two of the four large trader

groups, but is negatively related to the presence of hedger and other traders.

Speculators and swaps dealers tend to move together with return.

Table 5-5. Correlation Analysis
Volatility Hedger Others Swaps dealer speculators
Volatility 1.00000
Hedger -0.15506 1.00000
Others -0.08836 0.07001 1.00000
Swaps dealer 0.24209 0.00001 -0.09483 1.00000
Speculators 0.36956 -0.18371 -0.07241 0.10347 1.00000









CHAPTER 6
THE CASH MARKET AND FUTURES MARKET

In this Chapter 6, the convergence problem will be discussed. Futures market and

cash market are always interacted. Under the efficient market, there exists a long-term

equilibrium between cash and futures prices. In addition, the futures market has price

discovery function, which helps the farmers, grain elevators and other merchants better

forecast the trend of cash prices.

However, in recent years, is there still a long-term equilibrium between cash and

futures market? Is there the bidirectional casualty between cash price and futures price?

This chapter tries to find empirical evidence and answer the above questions.

Cointegration is considered as a necessary condition for market efficiency according Lai

and Lai, 1991. However, to conclude efficiency, we should also examine whether

futures contracts are unbiased predictors of spot markets. If the wheat spot and futures

contract prices are cointegrated, then a long-run relationship must exist between these

two series.

Cointegration is an appropriate indicator of long-term co-movement. Among all

cointegration tests, such as the Engle and Granger (1987) method and the Stock and

Watson (1988) test, Johansen's test has a number of desirable properties, including the

fact that all test variables are treated as endogenous variables.

First of all, the Augmented-Dick-fuller test has used to whether data series for

wheat futures and spot are stationary. Second, through cointegration, I study whether

wheat futures and cash prices have a long-term equilibrium. If both of them are

cointegrated, the futures price should be the unbiased estimator of cash prices, as

shown in the following equation:









InSt=a + fllnFt + pt

The data series for cash price are obtained from USDA and future prices from

Bloomberg. The futures prices used in the test are the March Wheat futures prices. I

created two data series from January 1, 2006 to February 2, 2010. There are total 1038

observations. Figure 6-1 shows that both of In Ft and InSt are nonstationary, but the first

differences are stationary.

Table 6-1. Unit Root Test
Pr <
Type Lags Rho P< Tau Pr < Tau F Pr> F
Rho
Spot 1 -4.0151 0.5376 -1.54 0.5119 1.25 0.7498
Futures 1 -4.8218 0.4526 -1.87 0.3458 1.88 0.5898
FirstdifSpot 1 -1060.42 0.0001 -22.99 <.0001 264.18 0.0010
First dif Futures 1 -1114.51 0.0001 -23.57 <.0001 277.71 0.0010
Johansen's (1998) approach is used to test for co-integration. We consider a general

VAR model of order k,

k-1
A Yt = D + HYt-k + i Yt-i + Et
i=1

Where A Yt = Yt Yt-1 and D is a deterministic term, Hand ri are matrices of

coefficient. The cointergration relationship is examined by looking at the rank of the

coefficient of matrixfl, If 1=0, there is no cointegration vector, hence no cointegration

relationship. If H=1, the two series are cointegrated (Johanse and Juseliues, 1990) The

trace numbers are in the Table 6-2, which shows that there is no a long-term

relationship between cash and futures prices. Neither of the series could predict its

counterpart.










Table 6-2. Cointegration Rank Test
5%
HO: H1: Drift Drift in
Eigenvalue Trace Critical
Rank=r Rank>r in ECM Process
Value
0 0 0.0062 6.4447 12.21
NOINT Constant
1 1 0.0003 0.0451 4.14

The result of cointegration test shows no long-term relationship between cash and

futures market. It is believed that the great volatility in wheat futures market has

deteriorated the efficient of futures market









CHAPTER 7
COMMODITY INDEX AND INDEX SPECULATION

Another phenomenon in driving up the commodity futures was deregulation of

index trading activates. The U.S. Senate Permanent Subcommittee in 2009 report

suggested that Index speculators had contributed to the unreasonable price fluctuation

in the energy and agriculture commodities and distorted the function of futures market,

which could not reflect the accurate information of supply and demand in the physical

market.

The index traders, to the extent, are like speculators. However, they focused on

long-side exposure to duplicate an index of multiply commodities. Their trading goals

are to diversify the investment portfolio. Because the returns between commodity and

other securities such as stocks and bond may be negatively correlated, commodity

investment may be optimal strategy for hedging purpose. Traders are commonly called

the long-term investors, since they seek exposure to commodities through passive long-

term investment. This strategy could be implemented via swaps agreement or mutual

funds, exchange traded funds. Usually they enter into a swap agreement for underlying

asset of a specific commodity index. The counterparts swap dealers, who have the

hedge exemption, enter the futures market and take long positions in commodity index.

There are two important indices: Standard Poor's-Goldman Sachs Commodity

Index(S&P GSCI) and the Dow Jones-AIG Commodity Index (DJAIGCI).

Standard Poor's-Goldman Sachs Commodity Index

The Figure 7-1 following shows the weight composition of each sector. One of the

famous commodities indices, GSCI was created by Goldman Sachs in July 1992 and

acquired by Standard &Poor's in 2007. It includes 24 commodity futures contracts










traded on U.S. exchanges: six energy products, five industrial metals, eight agricultural

products, three livestock products and two precious metals. Futures contracts on the

GSCI are traded at the Chicago Mercantile Exchange (CME).There is no limit on the

number of contracts that may be included in the S&P GSCI. Its returns are calculated

based on the arithmetic average of long positions in futures contracts (S&P GSCI Index

Methodology 2010 edition). Goldman Sachs has calculated the historical value of the

GSCI and normalized to a value of 100 on January 2, 1970, in order to permit

comparisons of GSCI values over time.


S&P GSCI Index Weight Composition
Sectors
Energy
SLivestock
Agriculture
Precious Metals
Livestock I Industrial Metals
6.07%



1.99%




Precious Metals
3.51%
Industrial Metals
6.83%


Figure 7-1. S&P GSCI Index Weight Composition

The contract value of each GSCI futures position is $250 times the GSCI index.

When futures contracts in the GSCI are close to expiration, they are rolled forward to

the next nearest contracts at the beginning of their expiration months. The rollover

period is from the fifth through the ninth business day of each month. During this period,

the GSCI portfolio is shifted from the existing to the next nearby futures contract at a

rate of 20% per day. This means that on the fifth business day, the GSCI is adjusted to









consist of 20% of the second nearby contracts and 80% of the existing contracts.

Likewise, on the sixth business day, the GSCI include 40% of second nearby contracts

and 60% of the existing contracts(S&P GSCI Index Methodology 2010 edition). On the

last day of the rolling period, the GSCI only include the second nearby contract.

Therefore, the GSCI will always have contracts with two maturities for each of the 10

commodities between the fifth and ninth business day of each month. For other

commodities with fewer futures, such as livestock, they are rolled forward less

frequently.

Wheat Market and S&P GSCI

Related search includes Stoll and Whaley 2009, they analyze the price co-

movements of index commodities and Granger-Causality between index inflow and

wheat future recent. They found that "commodity index rolls have little futures price

impact"," the failure of the wheat futures price to converge has not undermined the

futures contract's effectiveness as a risk management tool" and correlation of

commodities price levels that include in the commodity index of are quite low. Another

research is carried by Irwin, Sanders and Merrin 2009. They also use the Granger-

Causality Test to analyze the commodity index and price change and come up with the

same results as Stoll and Whaley. What is more, they also conclude that when price

low, seller complains the speculators, while when price are high, buyer complain the

speculators. Actually, "it is the historical pattern of attacks upon speculation".

In 2009, Congress published staff report titled "Excessive Speculation in the

Wheat Market". The report shows in detail about how commodity index traders affected

the price of wheat contracts traded on the Chicago Mercantile Exchange over the past

three years. The report pointed out around one-third and one-half of all of the










outstanding wheat futures contracts purchased by index traders offsetting part of their

exposure to commodity index instruments sold to third parties. The report also finds that

there are significant evidences that commodity index traders were one of the major

causes of "unwarranted changes" in futures market.

Through correlation analysis, it is worthy noticing that in the Table 7-1 the

significant relationship between S&P GSCI and wheat futures.

Table 7-1. Pearson Correlation Coefficients Prob > Irl under HO: Rho=0
Wheat S&P GSCI
Wheat 1.00000 0.43409
(<.0001)
S&P GSCI 0.43409 1.00000
(<.0001)
Figure 7-2 and Figure 7-3 are the daily return of S&P GSCI futures and wheat

futures traded on Chicago Mercantile Exchange (CME). The returns are measure by

100*ln(Ft/Ft-_). We observed the increased return variability in the recent years. Date

series are obtained from the Bloomberg.



8.00
6.00
4.00
2.00
0.00
-2.e3/ r 0
-4.00
-6.00
-8.00
-10.00



Figure 7-2. Daily Return of GSCI Futures(%)









10.00

5.00

0.00
1/3/ 00 0
-5.00

-10.00

-15.00



Figure 7-3. Daily Return of Wheat Futures(%)

Using the same method from the Chapter 6, I analyze the casualty between S&P

GSCI futures contract and Wheat futures contract. The model can be written in matrix

form as:


(YGSCI,t C1 b1,1, b1,2 (YGscI,t-1 + (ei,t)
ywheat,tj \C2 \b2,1 b2,2 \Ywheat,t-l/ \e2,t/

Where the e is stochastic error term, call shocks in the VAR model. Before

estimate the equation, we should judge the lag length. Including too many lagged terms

will consume degrees of freedom and potentially induce multicollinearity. One way of

deciding the lag length is to use a criterion such as Akaike and Schwarz rule. Using

option=ESACF ,MINIC and SCAN, SAS program provides us the optimal lag length. In

this study, the lag length one is appropriate suggested by SAS output. The ADF test in

the Table 7-2 suggests both of the variables are stationary.

The Table 7-4 presents the VAR results based on one lag of each variable. The

parameters tests given in the table are to test the hypothesis that collectively the various

lagged coefficients are zero. The test shows that the first equation is not statistically

different from zero, but the second equation is significant. Based on the result the wheat










is dependent on GSCI fluctuations. Next, the granger-causality confirms the VAR result.

The no causality null hypothesis is rejected for GSCI and Wheat futures. The changes

of wheat futures did not affect the GSCI fluctuation. However, the GSCI significantly

influence the wheat price fluctuation.


Table 7-2. Summary Statistics and Unit Root Test
Variable N Mean Std.Dev Skewness Kurtosis Q(6) Pr>ChiSq ADF Pr <
Tau
GSCI 1044 0.0142 1.8819 -0.2479 2.0730 6.57 0.3625 -23.68 <.0001
Futures 1044 0.0333 2.4618 -0.0663 1.0890 5.23 0.05143 -23.27 <.0001


Table 7-3. Causality Wald Test
Test DF Chi-Square Pr > ChiSq
1 1 0.04 0.8322
2 1 11.24 0.0008
Test 1: Group 1 Variables: GSCI Test 2: Group 1 Variables: wheat
Group 2 Variables: wheat Group 2 Variables: GSCI


Table 7-4. Model Parameter Estimates
Test DF Chi-Square Pr > ChiSq
1 1 0.04 0.8322
2 1 11.24 0.0008
Test 1: Group 1 Variables: GSCI Test 2: Group 1 Variables: wheat
Group 2 Variables: wheat Group 2 Variables: GSCI
Equation Parameter Estimate Error t Value Pr> Itl Variable
GSCI CONST1 0.01497 0.05827 0.26 0.7973 1
AR1_1_1 -0.05999 0.03437 -1.75 0.0812 GSCI(t-1)
AR1_1_2 0.00557 0.02627 0.21 0.8322 wheat(t-1)
wheat CONST2 0.03514 0.07589 0.46 0.6435 1
AR1_2_1 -0.15011 0.04477 -3.35 0.0008 GSCI(t-1)
AR1_2_2 0.01522 0.03422 0.44 0.6566 wheat(t-1)
Testing of the Parameters Correlations of Residuals
Test DF Chi-Square Pr > ChiSq Variable/
1 2 3.41 0.1816 Lag 0 1 2 3
2 2 12.49 0.0019 GSCI ++....










Information Criteria wheat ++.
AICC 2.857007 SBC 2.885449 + is > 2*std error, is <
HQC 2.867775 AIC 2.856974 2*std error, is between
FPEC 17.40877









CHAPTER 8
REGULATORY SYSTEM AND POLICY RECOMMENDATION

Development of Futures Regulation

Through the empirical analysis, we have found that the speculators, especially

index speculators have tremendously influenced wheat futures market. Now we turn our

focus to the federal regulatory system. We will look at the evolvement of futures

regulation and research on whether the current regulatory system is effective to prevent

excessive speculation.

The first complete regulatory framework for futures market in U.S. history is

Commodity Exchange Act (CEA), which passed in 1936. The Commodity Exchange Act

(CEA) required that all regulated agriculture commodities had to be traded only on the

federal designated markets. Also, the Act authorized the Commodity Exchange

Commission as an agency in the Department of Agriculture to oversight of the futures

market. In addition, the Commodity Exchange Act required Commodity Exchange

Commission to set up speculative position limits to prevent excessive speculation.

In 1974, the Commodity Futures Trading Commission (CFTC) was founded as an

independent regulatory agency to oversee futures market. The CFTC was empowered

more jurisdiction than ever before. Previously, the Commission could only regulate

agricultural commodities listed in the Commodity Exchange Act, whereas CFTC have

authorized to regulate all types of futures trading in all commodities.

The 1980s saw a rapid development of derivatives market. Financial derivatives

became common and served as effective risk-management tools. Numerous types of

financial derivatives had been created, such as foreign exchange, interest rate and even

credit rating. At the same time, regulatory burdens were blamed for retarding the









progress of the OTC derivatives market. Particularly, in 1990s, the legal uncertainty has

been a great problem when financial institutions developed a new instrument, which

was potential to reduce the flexibility and competitiveness of U.S. financial markets. To

keep the competitive advantage of U.S futures markets and ensure integrity and

efficiency of the market, the Congress passed the Commodity Futures Modernization

Act (CFMA) in 2000. The CFMA of 2000 incorporated much more flexible standards into

the act and greater legal certainty in the OTC market. The CFTC established a new

regulatory framework with the guidance of innovation of the OTC market and elimination

of excessive burden on sophisticated institutions or wealthy participants. Later on, the

CFMA has caused a series of serious consequence for OTC market. Since the CFMA

clarified that derivatives transactions was outside of its jurisdiction, which has been

believed to contribute to the turmoil in futures market.

The CFMA created three categories of commodities: agriculture commodities,

excluded commodities and exempt commodities. The excluded commodities are

defined as interest rates, exchange rates, currencies, credit risk, debt instruments,

measures of inflation, or other macroeconomic index. The exempt commodities are

neither excluded commodities nor agriculture commodities. Particularly, the energy and

metal contracts are categorized as exempt commodities.

Exempt commodities have two specific provisions. First, a contract or agreement

enters into between eligible contracts participants and not traded on trading facilities is

outside of the CFTC jurisdiction. Second, the contract or transaction entered between

the eligible contracts participants traded or executed on an electronic trading facility is

also exempt from the CFTC regulation, which is also known as Enron Loophole.









Before 2000, there only existed the Designated Contract Market (DCM). The

CFMA in 2000 established another two regulated markets: Derivatives Transaction

Execution Facilities (DTEF) and Exempt Market. The three types of exchanges, based

on the nature of futures traded and knowledgeable participants, are subject to the

varying levels of regulation.

The Designated Commercial Markets open to all types of market participants and

any type commodity and has the highest level of regulatory oversight. The exchange

should report to the CFTC the larger trader positions on daily basis. Besides, the CFTC

was required to continuously keep this kind of market under surveillance. The Chicago

Board of Trade (CBOT), Chicago Mercantile Exchange (CME) and New York Mercantile

Exchange (NYMEX) are Designated Commercial Markets.

The Derivatives Transaction Execution Facilities receives an intermediate level of

regulations. This kind of market is divided into another two categories, Regular and

Commercial. This trading facility is limited to eligible commercial entities. Any

commodity other than an agriculture commodity can be traded. However, some

agriculture commodities could be trade if approved by the commission. In addition, the

Derivatives Transaction Execution Facilities must comply with some requirements, for

example, establishing the compliance and surveillance, monitoring of trading activates

and having a record keeping system.

Unlike the Designated Commercial Market and Derivatives Transaction Execution

Facilities, the third Exempt Market is exempted from most provisions of CEA. There are

two types of exempt markets, Exempt Board of Trade (EBOT) and Exempt Commercial









Market (ECM). The commodities eligible to be traded on an Exempt Board of Trade

(EBOT) are excluded commodities.

The CFMA established separated exemptions for Exempt Commercial Market

(ECM). To be qualified for these exemptions, the contracts and transactions must be

traded or executed on electronic trading facilities between eligible commercial entities.

In addition, the commodities traded on the Exempt Commercial Market are restricted to

the exempt commodity. This trading facility operating as an Exempt Commercial Market

(ECM) must limit trading to eligible commercial entities

Speculative Position Limits

To prevent excessive speculation, an effective regulatory system is the guarantee.

There are two important and effective practices for futures market regulations: position

limit and daily reporting system.

The speculative position limits have been believed to be an effect tool for the

commission to impede the manipulation of U.S. futures markets and prevent

unreasonable price fluctuations. From its inception, this rule was little changed until

1981 that the CFTC required exchanges to adopt position limits for all the contracts not

listed on the federal position limits

Currently, the position limits are regulated under two frameworks (Table 8-1). One

is the federal position limits for agricultural commodities in futures and option market,

which are enumerated in the federal regulations (Table 8-2). The other is exchange-set

speculative position limits. Both the CFTC and the Designated Commercial Market take

on the enforcement of the exchange-set speculative position limits.










Table 8-1. Current Practices in Setting Speculative Position Limits and Position
Accountability Levels In selected Futures Contract Market as of July 24, 2009
Source: Federal Regulation
Category Sample Contracts Speculative Position Limits(q) or Position
Accountability(PA)
Spot Month Single Month All Months
Combined
CFTC Speculative Position Limits


Federal Limits Corn, oats, soybeans,
Soybean oil, Soybean
meal, wheat and Cotton

Limits and Levels Set by Exchanges
Other CME Frozen Pork Bellies;
Agricultural ICE U.S. Frozen
Concentrated Orange Juice
CME Livestock and Milk
ICE Cocoa, Coffee and /
World Sugar
Energy NYMEX Crude Oil, Natural PA PA
Gas, Heating Oil, and
Gasoline; ICE U.K.WTI

Metal COMEX Gold and Silver PA PA


Table 8-2. Speculative
Regulation


Position Limit (2010) in Contract Units Source:Federal


Contract Spot month Single month All months
Chicago Board of Trade
Corn and Mini-Corn 600 13,500 22,000
Oats 600 1,400 2,000
Soybeans and Mini- 600 6,500 10,000
Soybeans
Wheat and Mini-Wheat 600 5,000 6,500

Soybean Oil 540 5,000 6,500
Soybean Meal 720 5,000 6,500
Minneapolis Grain Exchange
Hard Red Spring Wheat 600 5,000 6,500
New York Board of Trade
Cotton No. 2 300 3,500 5,000
Kansas City Board of Trade









Hard Winter Wheat 600 5,000 6,500
Reporting System

The daily report system could be traced back to early 1920s. It required the

clearing member on each exchange to report the large trader positions. This system is

an important part of regulation and is still used by the CFTC today.

Commitment of Trader Report

The Commitment of Trader Report is published weekly by the CFTC and contains

more detailed information about position holdings, spreading, open interest by category,

and numbers of traders. The categories are:

1) producer/merchant/processor/user, who engages in the production, processing,

packing or handling of a physical commodity and uses the futures markets to manage or

hedge risks associated with those activities

2) speculator, who is a registered commodity trading advisor (CTA), a registered

commodity pool operator (CPO). These traders are engaged in operating a fund for a

commodity pool, that is, "an enterprise in which funds contributed by a number of

persons are combined for the purpose of trading futures contracts or commodity options,

or to invest in another commodity pool." (National Futures Association definition).

3) other reportable, who are not falling in any of the other two categories.

Commodity Index Trade Supplement

The CFTC also published weekly Commodity Index Trade (CIT) Supplement on its

website. CIT report contains 12 agricultural commodities, providing more detailed

information about index trader activities, for example, long/short position, percentage

open interest and number of trader and changes in their position.









Special Calls

The special calls provision could be traced back to the 1970s when the CFTC

encountered the problems of regulating the individuals located outside the U.S. and

trade in U.S. market. For the people in the domestic market, the large traders are

required to file reports of their daily positions to the CFTC. This requirement would

make it easy for the CFTC to inspect that whether the unreasonable price movement

was caused by the large positions. However, the CFTC could not obtain the large

positions hold by the foreign brokers' customers and their financial interests. To solve

the imbalance of regulations between foreign and domestic traders, the CFTC adopted

the special calls provision in which anyone receiving the special call must provide the

information the CFTC asked for, such as a firm's trading and delivery activity. Although

the foreign large traders do not have daily position responsibility, but they have to

provide the business activities and daily positions if the CFTC issue the special calls to

get those information

Enron Loophole

The CFMA allowed exempt commodities such as oil to be traded on Exempt

Commercial markets, free from the most of requirement by the CFTC and speculative

position limits, which are accommodated industry interests and fostered anticompetitive

behaviors and market manipulations. Particularly, Enron has been reported to

purposefully, improperly influence energy prices in western markets. Enron, through its

Enron Online (EOL) launched in 1999, provided web-based electronic trading platform

for wholesale energy, swaps, and other commodities. The daily average transactions on

the platform were $2.5 billion. 2,100 products were offered to traders across four









continents based on 15 different currencies. Between 1999 and 2001, Enron Online

reported the profit by derivatives trades as $920 million.

Swaps Loophole

The swaps market has grown rapidly since 1980, because it provides the

participants more flexible ways to hedge their exposure risks. According to the Bank for

International Settlements, the amounts of outstanding of global over-the-counter (OTC)

derivatives are estimated to be $614 trillion in December 2009. In essence, a swap

contract is an agreement between two parties that exchange a series of cash flows in

the future. Compare to other financial market, the swaps market has it unique

characteristics: 1) It affords the privacy for the swaps trading .Only two counterparties

knows the swap trading. 2) The swaps market could escape the government regulation.

In the past, the CFTC believed that the swaps dealers had no incentive to

manipulate futures prices. Just like bona fide physical hedger, swaps dealers used the

futures market to hedge their transactions. Therefore, the CFMA of 2000 greatly swept

away the legal uncertainty by significantly expanding the exemptions of swaps

transactions on financial instruments and physical commodities. The CFMA allowed the

swaps transactions based on the excluded commodity and exempt commodity to be

exempted from the regulations of the CFTC and Securities Exchange Commission.

Nowadays, the agriculture commodities could be traded on the multiple exchanges,

including the OTC market. Since swaps dealers who have hedge exemptions could

enter the futures market with unlimited quantities, institutional investors investing in

commodity futures do not directly trade through futures exchange. They used the swaps

contracts to hold an excessive amount of futures contracts without any constrain from

speculative position limits. In absence of speculative position limits, the commodities









derivatives market are exposure to excessive speculation. It was estimated that, in 2006,

Goldman Sachs and Morgan Stanley had earned billions of dollars in energy trading for

two years.

In the end of 2007, the CFTC issued special calls to 32 entities and their sub

entities to get a comprehensive knowledge of the quantity of commodity index trading in

OTC and exchange markets. The special calls of 2007 required the entities who

engaged in the index trading activity to provide the notional value of their index business

including in both domestic market and foreign markets. The index commodity funds

should classify the positions they hold directly in the futures market and positions

through swaps dealers. The data collected by the CFTC from the special calls

demonstrated the conclusion that positions held by the swap clients exceed the federal

speculate limits. For wheat futures, the total notional value of open contracts on June 30,

2008 was estimated to be $19 billion and the net notional index value took

approximately 47 percent of this total.

The increased commodity index trading has made the CFTC reconsider whether it

is appropriate to put the swap dealer trading in the commercial category, because

swaps dealers are using the future market to hedge risk in the OTC market, but their

client might be a speculator. In an effort to improve the transparency of the futures

market, the commission publishes the commitments of trader report, which remove

Swap Dealer from Commercial Category and Create New Swap Dealer Classification

since Jan 1, 2006. The swap dealer in the Commitment of Trader defines as the people

who deal primarily in swaps for a commodity and uses the futures markets to hedge the









risk associated with those swaps transactions. However, the swap dealer's

counterparties may be speculator, like mutual funds, or traditional commercial hedger,

In addition,there is a little bit difference between index traders' position in the

supplement and swap dealer position in the Commitment of Trade (COT). The swap

dealer category includes some position that has nothing with commodity index business.

On the other hand, the index trader category in the CIT supplement, including the

institutional investors who do the index trading directly in the futures market rather than

go to OTC market. The institutional investor positions are classified as managed money

or other reportable in the Commitment of Trade (COT). Therefore, the index trading

activities through the swap dealers still could not be precisely measured.

What is more, the CFTC published the Quarterly Index Investment Data on its

website since 2008. The data was selected from the special calls issued to the swap

dealers and index traders. The report includes the national values and the equivalent

number of futures contracts for all U.S. markets with more than $0.5 billion of reported

net notional value of index investment at the end of any one quarter.

Policy Recommendation

Today, the CFTC jurisdiction scope expands much larger than ever before. It

regulates the activities of about 3000 commodity exchange members, 360 public

brokerage houses and 2,500 commodity trading advisers and commodity pool operators.

In addition, off-exchange transactions involving instruments with similar feature to

futures contracts are also within CFTC jurisdiction. Futures contracts are so diversified

that the underlying asset have expanded to precious metals, raw materials, foreign

currencies, commercial interest rates, U.S. governments and so on. A number of

contracts have begun to be traded on multiple exchanges, including exchanges outside









the United States. Therefore, in order to keep the pace with development of futures

market, the CFTC need some amendments to maintain the integrity of the market and

hedger's benefit.

The CFTC reauthorization Act of 2008 (Farm Bill 2008) expands the CFTC

authority by creating a new regulatory category, which is the contract with the significant

price discovery function traded on the exempt commercial market. Price discovery is an

important function for futures market, since it helps people have more accurate

forecasts and a better investment plan. The futures prices cannot be known with

certainty today, the high level of relationship between cash and futures markets make it

possible for people to estimate the future price better. The Title XIII of the Farm Bill

provides standards applicable to significant price discovery contracts. Once the CFTC

finds that a contract traded on exempt commercial market performs significant price-

discovery function, the contract will be subject to Commodity Exchange Act and

commission regulations.

Improve Transparency and Data Accuracy

In this study, we find that the index trading activities through OTC market are still

not enough to judge and measure. The data collected though the special calls will take

a comparative long time. The new Commitment of Trader since 2006 removes the

swaps dealers from the commercial categories. However, the swaps dealer's clients

might be hedge and speculators. In order to enhance the transparency of futures and

options market, the Commission has to improve the weekly commitment of Trader

reports by including more detail trader classification.









Improve Regulation on Index Trading Activity

The special calls of 2007 illustrated that the positions held by the many swaps

dealers' client has exceed the speculative position limits. A question is raised by

whether swaps dealers should receive hedger exemption from speculative position

limits in consideration of the mix of commercial and noncommercial activity. In order to

keep the regulatory consistency and market integrity, the CFTC should improve the

evaluation of the noncommercial activity by swaps dealers and prevent the

noncommercial positions from exceeding position limits.









CHAPTER 9
CONCLUSION

This thesis describes the wheat futures market and speculation behaviors. The

main conclusions are

The futures market experienced a great volatile in recent years. Through the basis

analysis, the large discrepancy between spot and futures market during the expiration

month has been a major problem in the wheat market.

The volatility of wheat futures shows strong heterosckedascity and autocorrelation.

Based on the analysis of relationship between spot and futures price, they are not

cointegrated and the futures price did not contain much information to reflect the spot

price and verse well. Therefore, the larger volatility of wheat futures market has lower

the efficiency of price discovery functions

Based on the GARCH model, the volume and open interest provides a power

explanation for the volatility when enter separately. The coefficients of open interest to

volatility are negative and statistically significant. In addition, the test found that the

greater market depth, the lower the volatility, given a trading volume. Besides open

interest has a relative small impact on volatility than a trade incurred a high trading

volume without a corresponding changing in open interest.

The granger-casualty test indicates that the speculator trading activates have led

to the change of wheat return. The hedger activates comparatively shows steady and

little impact on the volatility of wheat futures. Since the data are not available for the

index traders and swaps dealers. That the CFTC issuing the special calls to swaps

dealers and then collecting them need a long time. Therefore, casualty-test to measure









the index trader and volatility is hard to carry on. However, through special calls, there

still a clear information that a large fund flow into the wheat futures.

Since 2000, the CFTC has ongoing deregulated over-the-counter market. This

regulatory system makes the excessive speculation possible. Many participants through

the OTC market hold more position than position limits. In addition, the Farm Bill still did

not stop the index trader through OTC market to influence the futures price.

Future research may have two directions: first, the sampling period of this study is

during the financial crisis, there must be some endogenous variables that might distort

the futures market or even deteriorate the economy. Understanding these inner links will

be better for find out the reason of extremely volatile in futures market.

Second, in order to actually measure the index traders activates, improve the

futures market transparence, I will expand my to research scope to other commodities,

such as soybean, cotton and grain. After acquiring the relative data about index traders'

business activities, further research on this topic will be moved on.









LIST OF REFERENCES


Akleman, D.G., Bessler, D.A.,&Burton, D.(1999). Modeling Corn Exports and Exchange
Rates with Directed Graphs and Statistical Loss Functions, in Clark Glymour and
Gregory Cooper editors Computation, Causation, and Discovery Cambridge, MA:
AAAI/MIT Press, 497-520.

Alexia B.,& Meredith S,.(2007). Beyond Enron: Regulation in Energy Derivatives
Trading, Northwestern Journal of International Law &Bussiness 27,665

Abbott, P.C., Christopher H.& Wallace E.T.(2008). What's Driving Food Prices? A Farm
Foundation Issue Report, July.

Babula, R.,Bessler, D.A.& Payne,W.S.(2004). Dynamic Relationships among U.S.
Wheat-Related Markets: Applying Directed Acyclic Graphs to a Time Series Model.
Journal of Agricultural and Applied Economics, 36,1-22.

Bessembinder,H.,& Seguin,P.J.(1993). Price volatility, trading volume, and market
depth: Evidence from futures market. Journal of Financial and Quantitative
Analysis, 28, 21-39.

Black, F. & Myron S. (1973). The Pricing of Options and Corporate Liabilities, Journal of
Political Economy 81,637-659.

Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity.Journal
of Econometrics, 31, 307-327.

Bollerslev,T. (1987). A conditional heteroskedastic time series model for speculative
prices and rates of return. Review of Economics and Statistics, 9,542-547.

Bryant, H. L., Bessler, D. A.& Haigh M.S.(2006). Causality in futures markets,Journal of
Futures Market, 26,1039-1057

Carlton, D. (1984).Future Markets: Their Purpose, Their History, Their Growth, Their
Successes and Failures. The Journal of Future Markets 4(3), 237-271.

Chang, E. C. (1985). Returns to speculators and the theory of normal backwardation.
Journal of Finance, 40,193-208.

Chang, E. C., Chou, R. Y.,& Nelling, E. F. (2000). Market volatility and the demand for
hedging in stock index futures. Journal of Futures Markets, 20,105-125.

Commodity Futures Trading Commission (2008). Agricultural Forum, April 22.

Commodity Futures Trading Commission (2008), Office of the Chief Economist.
Personal Communication with Dr. Jeff Harris, July.









Commodity Futures Trading Commission (2008). CFTC Approves Amendments to the
CBOT's Wheat Futures Contract, Press Release No. 5579-08, December 4.

Commodity Futures Trading Commission (2008). Staff Report on Commodity Swap
Dealers & Index Traders with Commission Recommendations, September

Commodity Futures Modernization Act of 2000

Commodity Future Trading Commission, Fiscal Year 2002 Annual Report.

Cornell, B. (1981). The relation between volume and price variability in futures markets.
Journal of Futures Markets, 1, 303-316.

Cooper, R.N., & Lawrence. R.Z. (1975) The 1972-75 Commodity Boom,Brookings
Papers on Economic Activity 3,671-723.

Daigler, R. T., & Wiley, M. K. (1999). The impact of trader type on the futures volatility-
volume relation. Journal of Finance, 54, 2297-2316.

Domanski, D, &A. Heath (2007). Financial Investors and Commodity Markets, Bank for
International Settlements Quarterly Review, 53-67.

Fortenbery, T.R,&Hauser. R.J.(1990). Investment Potential of Agricultural Futures
Contracts. American Journal of Agricultural Economics 72(3),721-726.

Fujihara, R.A.& Mougoue, M.(1997). An examination of linear and nonlinear causal
relationships between price variability and volume in petroleum futures
markets. Journal of Futures Markets 17, 385-416.

Girma,P. B.,&Paulson,A.S.(1998).Seasonality in petroleum futures spreads.Journal of
Futures Markets, 18, 581-598.

Girma,P.B.,&Mougoue,M.(2002).An empirical examination of the relation between
futures spread volatility, volume and open interest. The Journal of Futures Markets,
22(11), 1083-1102

Gorton, G., & Rouwenhorst. K.G. (2006).Facts and Fantasies about Commodity Futures.
Financial Analysts Journal 62(2),47-68.

Greenberger, M.(2007), Testimony of Energy Speculation: Is Greater Regulation
Necessary to Stop Price Manipulation?

Greenberger, M. (2008).Testimony of Energy Market manipulation and Federal
Enforcement Regimes,

Haigh, M. S. (1999).Cointegration, unbiased expectations and forecasting in the
BIFFEX freight futures market. Journal of Futures Market, 20,546-571.









Haigh, M., Nomikos,N.K.&Bessler, D.A.(2004). Integration and Causality in International
Freight Markets Modeling with Error Correction and Directed Acyclic Graphs,
Journal of Southern Economic 71,145-162.

Haigh, M. S.,& Bessler, D. A. (2004). Causality and price discovery: An application of
directed acyclic graphs. Journal of Business, 77,1099-1121.

Irwin, S.H.,Garcia, P.,Good,D.L.,& Kunda,E.L.(2008) Recent Convergence Performance
of CBOT Corn, Soybean, and Wheat Futures Contracts, Choices, American
Agricultural Economics Association, 2nd pp. 16-21.

Irwin, S.H., Sanders D.L.,& Merrin, R.P. (2009). Devil or Angel: The Role of Speculation
in the Recent Commodity Price Boom (and Bust), Invited paper at the Southern
Agricultural Economics Association Meetings, Atlanta, GA.

Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of Economic
Dynamics and Control, 12, 231-254.

Johansen, S. (1991). Estimation and hypothesis testing of cointegration vectors in
gaussian vector autoregressive models. Econometrica, 59,1511-1580.

Karpoff, J. M. (1987). The relation between price changes and trading volume: A survey.
Journal of Financial and Quantitative Analysis, 22,109-126.

Labys, W.C.,&Thomas H.C. (1975). Speculation, Hedging and Commodity Price
Behavior: An International Comparison. Applied Economics, 7,287-301.

Lamoureux, C., & Lastrpes, W. (1990). Heteroscedasticity in stock return data: Volume
versus garch effects. Journal of Finance, 45,221-229.

Lai,K.&Lai, M. (1991). A cointegration test for market efficiency, Journal of Futures
Markets, 11, 567-575

Majority and Minority staff report,(2009), Excessive Speculation in the Wheat Market.

Mark J.& Lynn J.(2008). Speculation and Energy Prices: Legislative Response,
Cunningham, Congressional Research Service Report RL34555.

Moneta, A.(2004). Identification of Monetary Policy Shocks: A graphical causal
approach. Notas Econ6micas, Faculdade de Economia, Universidade de Coimbra,
20,39-62.

Randolph, W. L., & M. Najand, (1991). A Test of Two Models in Forecasting Stock Index
Future Price Volatility, Journal of Futures Markets 11,179-190.

Park, H., Mjelde, J.&Bessler,D.A. (2008). Price Interactions and Discovery among
Natural Gas Spot Markets in North America. Energy Policy 36, 290-302.









Peck A.E. (1981). The Adequacy of Speculation on the Wheat, Corn, and Soybean
Futures Markets. Research in Domestic and International Agribusiness
Management, 2,17-29.

Rothig,A., &Chiarella,C. (2006). Investigating Nonlinear Speculation in Cattle, Corn and
Hog Futures Markets Using Logistic Smooth Transition Regression
Models, Research Paper Series 172, Quantitative Finance Research Center,
University of Technology, Sydney.

Reale, M.,& Wilson,G.T. (2001). Identification of vector AR models with recursive
structural errors using conditional independence graphs. Statistical Methods and
Applications, 10,49-65.

Sanders, D.R., S.H. Irwin, and R.P. Merrin.(2008) The Adequacy of Speculation in
Agricultural Futures Markets: Too Much of a Good Thing? Marketing and Outlook
Research Report.

Shalen, C. T. (1993). Volume, Volatility, and the Dispersion of Beliefs. Review of
Financial Studies, 6, 405-434.

Staff report. (2006). The Role of Market Speculation in Rising Oil and Gas Prices: A
Need to Put the Cop Back on the Beat.

Stock, J.S. &M. Watson (1988), Variable Trends in Economic Time Series. Journal of
Economic Perspectives, 2( 3)

Stoll, H. R., &Whaley, R. E. (1988). Volatility and futures: Message versus
messenger. Journal of Portfolio Management, 14,20-22.

Stoll, H. R., &Whaley, R. E. (1990). The dynamics of stock index and stock indexfutures
returns. Journal of Financial and Quantitative Analysis, 25,441-468.

Streeter, D.H.,&Tomek, W.G. (1992). Variability in Soybean Futures Prices: An
Integrated Framework. The Journal of Future Markets, 12(6),706-728.

Subcommittee report (2006), The Role of Market Speculation in Rising Oil and Gas
Prices: A Need to Put the Cop Back on the Beat, S. Prt, 109-165

Swanson, N. R.,& Granger, C. W. J. (1997). Impulse response functions based on a
causal approach to residual orthogonalization in vector autoregressions. Journal of
the American Statistical Association, 92,357-367.

Tauchen, G.E.,&Pitts, M. (1983). The price variability-volume relationship on speculative
markets. Econometrica, 51,485-506.

The Enron Loophole(2008). Congressional Research Service.









Wang, C. (2002). The effect of net positions by trader type on volatility in foreign
currency futures markets. Journal of Futures Markets, 22,427-450.

Wiley, M.,&Robert D.(1998). Volume relationships among types of traders in the
financial futures markets, Journal of Futures Markets, 8,91-114.

Zhang, J. E.,& Zhu, Y. (2006). VIX futures. Journal of Futures Markets, 26,521-531.









BIOGRAPHICAL SKETCH

Fang Meng earned her Bachelor of Arts in financial engineering from Shandong

University, China, in 2008. Degree of Master of Science in food and resource

economics was completed at the University of Florida during the summer of 2010.

Meng's primary research interests include agriculture commodity market.





PAGE 1

1 COMMODITY INDEX INVESTMENT AND WHEAT FUTURES MARKET By FANG MENG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Fang Meng

PAGE 3

3 To mom and dad, the most important persons in my life

PAGE 4

4 ACKNOWLEDGMENTS The experience of two years study in University of Florida is the most precious treasury in my life. During the past two years, I got so many help from my professors, classmates and family. First, I would like to thank Dr. Weldon, chair of my supervisory committee for ongoing support. His enlightened and i nspiring guidance lead me to finish my thesis and complete my masters program. Second, I would like to thank Dr. Livingston to give me a chance to do research with him. Although the culture and language difference, he is always patient and supportive, inst ructing me and correcting my language problems. In addition, his knowledgeable made me have a fervent passion towards finance studying. I also feel grateful to Dr. Winner. I took his three statistics courses during my masters studying. He taught me the statistical theory and statistics packages, which made it possible for me to apply the statistically analytical methods to the whole research process. Here, I also want to give my deeply gratitude to my parents. Throughout these years, their love and encourage bring me much courage to overcome difficulties. I should also thank Mr. Weiyi,Ni. Without his constantly urging me on to study hard, I will not be at the point in my life. There are so many people I sh ould express my thanks, this thesis is my reward for all your help.

PAGE 5

5 TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. 4 page LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 A BSTRACT ..................................................................................................................... 9 CHAPTER 1 INTRODUCTION .................................................................................................... 11 2 THE DESCRIPTION OF WHEAT FUTURES MARKET .......................................... 13 3 PROBLEMS OF THE WHEAT FUTURES MARKET .............................................. 17 Price Change and Volatility ..................................................................................... 17 Open Interest and Volume ...................................................................................... 19 Convergence Problem ............................................................................................ 20 Speculations in the Wheat Futures Market ............................................................. 24 4 RELATIONSHIP AMONG VOLATILITY, VOLUME AND OPEN INTEREST .......... 26 Literature Review .................................................................................................... 26 Methodology ........................................................................................................... 26 ARCHLM Test for ARC H Effect ............................................................................. 28 Relations between Return Volatility and Open Interest ........................................... 29 Relations between Return Volatility and Volume .................................................... 30 Relations between Return Volatility, Volume and Open Interest ............................. 31 5 SPECULATION BEHAVIORS IN THE WHEAT FUTURES MARKET .................... 33 Literature Review .................................................................................................... 33 Methodology and Data ............................................................................................ 35 6 THE CASH MARKET AND FUTURES MARKET ................................................... 40 7 COMMODITY INDEX AND INDEX SPECULATION ............................................... 43 Standard Poors Goldman Sachs Commodity Index ............................................... 43 Wheat Market and S&P GSCI ................................................................................. 45 8 REGULATORY SYSTEM AND POLICY RECOMMENDATION ............................. 50 Development of Futures Regulation ....................................................................... 50

PAGE 6

6 Speculative Position Limits ..................................................................................... 53 Reporting System ................................................................................................... 55 Commitment of Trader Report .......................................................................... 55 Commodity Index Trade Supplement ......................................................... 55 Special Calls .............................................................................................. 56 Enron Loophole ...................................................................................................... 56 Swaps Loophole ..................................................................................................... 57 Policy Recommendation ......................................................................................... 59 Improve Tr ansparency and Data Accuracy ...................................................... 60 Improve Regulation on Index Trading Activity .................................................. 61 9 CONCLUSION ........................................................................................................ 62 LIST OF REFERENCES ............................................................................................... 64 BIOGRAPHICAL SKETCH ............................................................................................ 69

PAGE 7

7 LIST OF TABLES Table page 4 1 Summary Statistics for Daily Return ................................................................... 27 4 2 Summary Statistics for Futures Returns ............................................................. 29 4 3 GARCH (1,1) Estimate for Volatility and Open Interest ...................................... 30 4 4 GARCH (1, 1) Estimate for Volatility and Volume ................................................ 31 4 5 GARCH (1,1) Estimate for Volatility, Open Interest and Volume ........................ 31 5 1 Summary Statistics ............................................................................................. 36 5 2 Unit Root Test and ARCH Affect Test ................................................................ 38 5 3 Granger Causality Wald Test ............................................................................. 38 5 4 White Noise Diagnostics ..................................................................................... 38 5 5 Correlation Analysis ............................................................................................ 39 6 1 Unit Root Test ..................................................................................................... 41 6 2 Cointegration Rank Test ..................................................................................... 42 7 1 Pearson Correlation Coefficients Prob > |r| under H0: Rho=0 ............................ 46 7 2 Summary Statistics and Unit Root Test .............................................................. 48 7 3 Causality Wald Test ............................................................................................ 48 7 4 Model Parameter Estimates ............................................................................... 48 8 1 Current Practices in Setting Speculative Position Limits and Position Accountability Levels In selected Futures Contract Market as of July 24, 2009 Source: Federal Regulation ................................................................................ 54 8 2 Speculative Position Limit (2010) in Contract Units Source:Federal Regulation .......................................................................................................... 54

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Actively Trade Future &Option Contracts 19972009 ......................................... 14 2 2 Growth in Volume of Futures& Option Contracts Traded, 19982008 ................ 14 2 3 U.S. Wheat Production. Types and amounts of wheat grown in the United States in 20092010. .......................................................................................... 15 3 1 Continuous Futures Price. .................................................................................. 18 3 2 Volatility of Chicago Wheat Futures. ................................................................... 19 3 3 Volume and Open interest GrowthChicago Wheat. ........................................... 20 3 4 March Wheat Basis (Toledo, OH cash less wheat CBOT futures) ...................... 22 3 5 May Wheat Futures Basis (Toledo, OH cash less wheat CBOT Futures) .......... 22 3 6 July Wheat Basis (Toledo, OH cash less Wheat CBOT futures) ........................ 23 3 7 September Wheat Basis (Illinois Cash Less Wheat CBOT Futures) .................. 23 3 8 December wheat Basis (Toledo, OH cash less Wheat CBOT futures) ............... 24 3 9 Percentage of Open Interest for Chicago Wheat Market Held by Participants (%) ...................................................................................................................... 25 5 1 Op en interest: Wheat in Chicago Board of Trade ............................................... 33 5 2 Speculators Position(%) ..................................................................................... 37 5 3 Hedgers Position(%) ........................................................................................... 37 5 4 Daily Return(%) .................................................................................................. 37 7 1 S&P GSCI Index Weight Composition ................................................................ 44 7 2 Daily Return of GSCI Futures(%) ....................................................................... 46 7 3 Daily Return of Wheat Futures(%) ...................................................................... 47

PAGE 9

9 ABSTRACT OF THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL F ULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE COMMODITY INDEX INVESTMENT AND WHEAT FUTURES PRICES By Fang Meng August 2010 Chair: Richard Weldon Major: Food and Resource Economics The turmoil in futures market in recent years becomes intense concern to the industry, the exchanges and the Commodity Futures Trading Commission (CFTC). In my thesis, I examine the role of speculation in the wheat markets and t he effective regulatory system to prevent excessive speculation. First, I summarize wheat market performances during the 2006 to 2010. Second, I develop methods t o examine how the activities of speculators in aggregate, have led to the wheat market disruption. The quantity methods find that firstly, t he volatility of wheat futures has preformed strong heterosckedascity and t he volume and open interest provide the powerful explanation of volatility when enter separately Second, the speculator trading activities have led to the change of wheat futures return while t he hedger activities show a comparative steady and little impact on the price changes. Third, f utures and cash prices do not cointegrated in long term and the futures price does not contain much information to reflect the cash price In addition, the derivatives trading through over thecounter (OTC) market become more dynamic and dominate. Commodit y index trader through purchase of commodity index could directly affect the futures price fluctuation. In my thesis, I also analyze the

PAGE 10

10 impact of index trader activity on the wheat futures market. The empirical tests show that the injection of index funds does cause the recent volatile in wheat futures market.

PAGE 11

11 CHAPTER 1 INTRODUCTION The futures market in the United Stated experienced an extremely volatile period in recent years. The wheat futures traded on Chicago Mercantile Exchange (CME), for example, increased from $3.00 per bushel in 2006 to $11.00 per bushel in the mid2008, and afterwards declined sharply back to $3.00 per bushel at the end of 2008. Moreover, many other commodities endured the similar performances. First, trading volume and open interest increased tremendously. The prices of commodities, such as oil, wheat and soybean have reached unprecedented levels. Second, the cash and futures prices failed to converge during the expiration month. Third, large amount of index have been observed in the commodities derivative markets. The unexpected changes in commodity prices has led to intensive investigations by the Commodity Futures Trading Commission (CFTC) and has been widely debated by academia. Many recent testimonies and Congress reports show that the distortion of the futures market and even the financial recession are because of the failures of the regulatory structure and agencies. S ince 2000, the U.S. congress has continuously deregulated the over the counter (OTC) market, in order to make the U.S. futures market more flexible and competitive in the global market. The commodity index traders are exempt from position limits when purchase commodity futures contracts to offset their business r isk. In this thesis, firstly, it provides a comparative description of the wheat futures market, price change and volatility, m arket depth and liquidity c ash and futures prices and t rader position. After market analysis, this thesis will go in detail to i nvestigate the speculation behaviors in wheat futures market during extreme volatility period. I am

PAGE 12

12 trying to find evidence to examine whether the speculators activates cause the distortion of wheat futures market. In addition, I differentiate the speculators into index speculators and traditional speculators, since there are somewhat different from each other. I tried to examine the role of the commodity index trading and investigate casualty relationship between commodity index trade and wheat futures pr ice.

PAGE 13

13 CHAPTER 2 THE DESCRIPTION OF W HEAT FUTURES MARKET The first organized futures market in U.S. Chicago Board of Trade was established in 1848. At that time, because of its unique location and access to the Great Lakes, Chicago became the transportation and distribution pivot for agriculture commodities. Initia lly, it severed as a market place for farmers to sell their products, including grain, wool, pork, beef and lumber before delivering to lock in a price. At the beginning, the trading was a type of forward contract In 1865, standardized futures contracts were introduced. The original trading activates were carried on the floors, known as floor trading. The traders stood on the floor, shouting the price they were willing to pay or accept. They simply reached an agreement on the price and number of contracts. The clearinghouse, which establi shed in the 1920s and guaranteed against default, is a significant development for the modern futures market. Since its establishment, the futures market in U.S. has achieved great development in less than 200 years. The Chicago Board of Trade, the New Yo rk Mercantile Exchange, Kansas Board of Trade and Intercontinental Exchange developed into the worlds leading futures exchanges. Meanwhile, the futures market exceeded the agricultural commodities range. Numerous new futures and options based on such thin gs as foreign exchange rates, interest rates, equity index and even credit ratings are listed on the exchanges. The number of future contracts expanded from 266 in 2000 to 1730 in 2009 (Figure 21).Since 2000, volume growth on futures market has increased sevenfold (Figure 22).

PAGE 14

14 Figure 21. Actively Trade Future and Option Contracts 19972009 Source: FY 2009 Presidents Budget & Performance Plan Figure 22. Growth in Volume of Futures and Option Contracts Traded, 1998 2008 Source: FY 2009 Presidents B udget & Performance Plan

PAGE 15

15 There are five basic types of wheat growing in the United States: hard red winter, hard red spring, soft red winter, durum and white. Figure 23 illustrates the percentage of production for each type of wheat during 2009. Nowadays in the United States, three major futures exchanges specialize in a particular type of wheat. These wheat futures contracts specify the type and quality of the wheat to be delivered, period for delivery, the possible locations for delivery and the requir ed price at the time of delivery. The soft red winter wheat is traded on the Chicago Mercantile Exchange (CME), hard red winter wheat is traded on the Kansas City Board of Trade (KCBT) and hard red spring wheat is traded on the Minneapolis Grain Exchange ( MGE). Figure 23. U.S. Wheat Production Types and amounts of wheat grown in the United States in 2009201 0. Source : USDA, Economic Research Service and Wheat Data: Yearbook Tables, Table 6. The CME soft red winter wheat contract provides sale and delivery of #2 soft red winter wheat. Currently, there is one warehouse in Chicago and two facilities in Toledo. Delivery usually takes a multi step process several days prior to the expiration of the contract. Wheat futures contracts usually expire on the first business day prior to the 15th calendar day of the contract month. About one month prior to the expiration, an Hard red winter 41% Hard red spring 25% Soft red winter 18% white 11% Durum 5%

PAGE 16

16 approved firm give s notice of the plan to make a delivery. On next day, the exchange will examine the outstanding long open interest and selects those positions that have been held for the longest period to accept delivery. On that day, the holder of the long position may ei ther accept the delivery or pass it on to another buyer. The following day is called delivery day, when both the long and short open interests are closed. Delivering wheat under contract from an approved warehouse does not require the physical movement. I nstead, it is through exchanging the right in the form of a shipping certificate. Under the request by certification holder, the warehouse should have to load the grain within a period specified by the rules of exchange. Through thi s delivery process, traders do not necessarily physically handle any wheat. Then traders may store grain in grain elevators, in the form of shipping certificates. They could hold these shipping certificates, pay the storage fees, and sell the wheat when price increases.

PAGE 17

17 CHAPTER 3 PROBLEMS OF THE WHEAT FUTURES MARKET In this chapter, we will first evaluate wheat futures market performances from the following perspectives: price change and volatility, m arket depth and liquidity c ash and futures prices and t rader position Price Ch ange and Volatility To look at the trend of futures price, we have to create a series of futures price. Since at any particular date, more than five futures contracts on wheat may be traded. For example, futures with expiration in July this year and July next year may be traded simultaneously. I took the price of nearby expiration contract as the reference price and created a continuous futures price for each contract. Figure 39 shows the trend of wheat futures prices from 2006 to 2010.During the past three years, the wheat futures price has experienced the spike and collapse. First, wheat futures price continue to surge high form $4.83 per bushel to $12.99 per bushel in a year. It reached the peak at $12.99 per bushel in the February 2008. Then the wheat c ontract price fell from $12.99 per bushel to $5.12 per bushel.

PAGE 18

18 Figure 31. Continuous Futures Price, Source: Bloomberg. In addition, the volatility of wheat futures price has reached unprecedented level. Data are from Jan 1, 2006 to Feb 28, 2010. Return volatility is calculated by the standard deviation of daily return. Volatility= ( ) Where Xt= 100 Daily futures return is defined as 100 ln( FF ) where Ft is the futures settlement price at a given day t. The logarithm transformation improves the statistical properties, because it provides a clearer tendency of the variable than untransformed series. Most macroeconomics time series data are estimated based on logarithmic scale over time. Being a squared quantity, volatility calculated will be high in periods when there are big changes in the prices and comparatively small when there are modest changes in price compared to the previous year, the volatility of price has increased drastically since 2008 (Figure32). 0 200 400 600 800 1,000 1,200 1,400 1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

PAGE 19

19 Figure 32. Volatility of Chicago Wheat Futures. Source: Bloomberg Traditionally, when a farmer, or a grain elevators uses the futures contract to hedge their business risks, they are required to m aintain working capital as margin and. If they fail to make a margin call as needed, their futures positions are immediately liquidated. During the extremely volatility period, hedgers has incurred a large margin calls on a daily basis, which require hedger to have much larger working capital than ever before for their futures positions. According to National Grain and Feed Association, the hedging cost has tripled since 2008. Open Interest and Volume Open interest refers to the total number of contracts t hat are not exercised, offset or delivered and an appropriate measure of participant activities. For every long or short trade, there must be a counterpart trade, or the contract is considered open. An increasing open interest usually indicates new money flowing into the market. Volume measures market intensity. It represents the total amount of trading activity in a given day. The greater the amount of trading activities, the higher will be volume. Figure 33 shows the trading volume and open interest in the Chicago wheat 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 1.2000 1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

PAGE 20

20 market from 2001 to 2010. Before 2005, both volume and open interest maintained a lower move. However, after 2005, the drastic increase in volume and open interest have been observed. Figure 33. Volume and Open interest GrowthChicago Wheat. Source: Bloomberg Convergence Problem Basis is the current cash price of a given commodity at a location minus the price of a particular futures contract for the same commodity. Several features of this need to be explained. First, basis depends upon a cash price of a commodity at a specific location. The cash price of wheat, for example, might differ between Kansas City and Chicago, so that the basis for those two locations will be different. If wheat had two different prices in two locations, a speculator could potentially buy the commodity in the cheaper market and sell it in the market with the higher price until arbitrage profit disappear. However, prices for wheat in Chicago and Kansas City can base on the expense of transporting wheat from one market to another. If wheat is grown near Chicago, then we might reasonably expect the price of wheat in Chicago to be lower

PAGE 21

21 than the price of wheat in Kansas City. For CBOT wheat, the key delivery area is in Toledo, Ohio. The cash price I use to analyze is based on the wheat price reported in Toledo, Ohio. Usually, when a futures contract gets closer to expiration, the futures price will converge to the price in the cash market. However, over the past three years, the futures prices for wheat were abnormally high. They also failed to converge with the cash price as the futures contract neared expiration. The CBOT wheat futures have five expiration months: March, May, July, September, and December. From Figures3 4 to Figure 38 ,they suggest that the basis for five delivery months of wheat future contract. The data of futures price were obtained from Bloomberg and the spot prices are from the USDA, Agricultural Market Service. The last observation for each contract every year is the expiration day, usually around the 15th of the calendar month. The figures suggest that the poor convergences exist in all five wheat contracts. If a futures contract fails to converge at the expiration, it will severely impair the ability of the farmers, grain elevators or other merchants use the futures markets to hedge their exposure to price risk over time.

PAGE 22

22 Figure 34. March Wheat Basis (Toledo, OH cash less wheat CBOT futures) Figure 35. May Wheat Futures Basis (Toledo, OH cash less wheat CBOT Futures) $3.0 $2.5 $2.0 $1.5 $1.0 $0.5 $0.0 $0.5 $1.0 $1.5 14/01/04 28/05/05 10/10/06 22/02/08 6/07/09 18/11/10 $3.0 $2.5 $2.0 $1.5 $1.0 $0.5 $0.0 $0.5 $1.0 $1.5 17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10

PAGE 23

23 Figure 36. July Wheat Basis (Toledo, OH cash less Wheat CBOT futures) Figure 37. September Wheat Basis (Illinois Cash Less Wheat CBOT Futures) $3.0 $2.0 $1.0 $0.0 $1.0 $2.0 $3.0 17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10 $4.0 $3.0 $2.0 $1.0 $0.0 $1.0 $2.0 $3.0 17/02/05 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10

PAGE 24

24 Figure 38. December wheat Basis (Toledo, OH cash less Wheat CBOT futures) Speculations in the Wheat Fut ures Market Figure 39 illustrates the relative traders position in the Chicago wheat futures market. The data were obtained for Commitment of Traders (COT) on the CFTC website. The Commitment of Traders (COT) report provides a breakdown of each Tuesdays open interest for markets with 20 or more traders positions equal to or more than the reporting level set by the CFTC. The CFTC collects data from the clearinghouse and classifies traders into either commercial or noncommercial. The commercial category represents traders using futures market to hedge. The noncommercial category generally includes pension funds, hedge funds and other type of fund who profit from the prices changes in the futures market. Historically, the demand by the hedger has exceeded the demand by speculator in the futures market. Since there is difference in the number of hedgers and speculators, hedgers have to make an attractive offer to get the speculators to take the other side of the transaction. However, in Figure 39 it obvio usly shows that speculators have $2.5 $2.0 $1.5 $1.0 $0.5 $0.0 $0.5 $1.0 $1.5 $2.0 5/09/05 24/03/06 10/10/06 28/04/07 14/11/07 1/06/08 18/12/08 6/07/09 22/01/10 10/08/10

PAGE 25

25 dominated the futures market. They have taken more than 50 percent of open interest in the futures market over the past three years. Michael Master in his one testimony has proposed that the appropriate open interest taken by speculator should be in the range of 25% 35%. It believed that this excess speculation has undermined the integrity of market. Figure 39. Percentage of Open Interest for Chicago Wheat Market Held by Participants (%) 0 10 20 30 40 50 60 70 80 Pct_of_OI_hedgers_Long_All Pct_of_OI_Speculators_Long_All

PAGE 26

26 CHAPTER 4 RELATIONSHIP AMONG VO LATILITY, VOLUME AND OPEN INTEREST Literature Review In the Chapter 3, we have found several problems in the wheat market. From the Chapter 4 to Chapter 7, we will continue to analyze these problems separately by using quantity methods. We begin with inves tigating the relations between among volume, open interest and volatility to uncover the source of variability in the wheat futures prices. The trading volume and open interest are two important indices of performance measurement for futures market. Their relationship are utmost importance for one reason, that is a well understanding the link could contribute to a superior speculation and hedge strategies. Most empirical researches found the positive relationship between trading volume and price volatility. Conell (1981) Tauchen Pitts (1983) and Karpoff (1987) find the positive relations between trading volume and price volatility. Bessembinder and Seguin (1993) study the impacts of volume and open interest on price volatility. They differentiated the volume and open interest into expected and unexpected categories. They find that both of them have positive effects on price volatility. However, the unexpected volume tends to have a larger impact on volatility than expected volume. Girma and Mougoue (2001) inv estigate the relation between the petroleum futures spread variability, trading volume and open interest. Their study shows the contemporaneous weighted average of volume and open interest provide significant explanation for futures spread volatility. Met hodology Table 41 provides summary statistics for daily return of wheat futures. The mean of 0.036 is very close to 0 and Standard deviation is 2.57. The skewness and kurtosis

PAGE 27

27 are 0.1796 and 2.6398 respectively. The kurtosis is less than three, the distri bution has thicker tails and a lower peak compared to a normal distribution. However, these descriptive statistics do not provide conclusive information about normality. SAS provides four different statistics for testing normality. ShapiroWilk W of .979 ( P value<0.0001), which rejects the null hypothesis of normality distribution. Similarly, Kolmogorov Smirnov, Cramervon Mises, and Anderson Darling tests reject the null hypothesis. Therefore, we can conclude that the variable is not normally distributed. Table 41. Summary Statistics for Daily Return N Mean Std Dev Skewness Kurtosis ShapiroWilk Kolmogorov Smirnov Cramer von Mises Anderson Darling Daily Reutn 1023 0.036 2.57 0.1796 2.6398 <0.0001 <0.0100 <0.0050 <0.0050 The return series on wheat futures in Table 41 shows that it is characterized by excess kurtosis and exhibit heteroscedasticity (Akgiray 1991, Hall 1989, Harvey and Siddique 1999).Therefore, the Generalized ARCH model is appropriate to employ. Engle first introduced the autoregressive conditionally heteroscedastic (ARCH) model in 1982 and Bollerslev developed a more general structure GARCH model in 1986.The variance in the model is assumed to follow an autoregressive model. The approach to GARCH (p, q) mod els is to set up an error term in terms of white noise. = ~ (0 ,1 ) where is assumed to follow an autoregressive model: = + + In the papers of Lamoureux and Lastrpes 1990 and Najand 1991, the GA RCH (1,1) have been proved to be adequate model to test the relations between price volatility and volume. Since then, the GARCH and ARCH models have been widely used in financial forecasting and derivatives pricing. Bollerslev (1986) using the GARCH model more accurately describes the phenomenon of volatility clustering and related effects.

PAGE 28

28 Foster(1995) uses the GARCH(1,1) and GMM model to investigate the relationship between trading volume and price volatility in crude oil futures market. He finds the contemporaneous volume is positively related to the price volatility. Fujihara and Mougoue(1997) use the GARCH(1,1) model to study the crude oil, heating oil and unleaded gasoline futures market. They conclude that the volume is a significant explanatory in explaining the price volatility. Volume is the number of buy and sell of futures contract for the Chicago wheat. While the trading volume also represents the daily ac tivities of hedger and longtime trader, which is more transitory (Chang, Chou and Nelling2002). I analyze daily data from January 1, 2006 through February 28, 2010. The sample consists of 1040 observations. Data on futures price, open interest and volume come from Bloomberg DataStream. The returns are calculated as 100 times the natural logarithm of the first differences of daily futures settlement prices = 100 ln (F/ F ) .The open interest and volume are not adjusted. ARCHLM Test for ARCH Effect A method proposed by Engle (1982) to test the lag length of ARCH errors is the Lagrange Multiplier test. This procedure is to first estimate the best fitted AR(q) to obtain the squares of the errors, and regress them on a constant and lagged values i n the following equation: = w + The null hypothesis is that = 0 for all i=1,2,3n. The alternative hypotheses is that at least one of the estimated coeffi cients is not equal to zero. Under the null

PAGE 29

29 hypothesis of no ARCH errors, the test statistic TR follows distribution with q degrees of freedom. If TR is greater than the Chi square table value, we reject the null hypothesis and conclude the presence of an ARCH effect in the AR model Otherwise, we fail to reject the null hypothesis. Table 41 illustrates the descriptive statistics for daily return. These tests strongly indicate that return series is characterized by significant heteroscedasticity, with p < 0.0001 for all lags. Table 42. Summary Statistics for Futures Returns Daily return Q and LM Tests for ARCH Disturbance Order Q Pr > Q LM Pr > LM N Mean Sum Weights Sum Observations Variance Kurtosis Std Deviation Skewness Uncorrected SS Coeff Variation Std Error Mean 1040 0.02974 1040 34.9258 6.0539 1.09832 2.4605 0.0690 6290.8752 8274.2457 0.07630 1 25.5900 <.0001 25.6121 <.0001 2 37.8803 <.0001 33.1802 <.0001 3 58.4281 <.0001 46.4873 <.0001 4 71.8117 <.0001 51.6899 <.0001 5 86.7485 <.0001 57.6692 <.0001 6 89.7813 <.0001 57.6698 <.0001 7 99.7232 <.0001 61.1580 <.0001 8 102.6048 <.0001 61.1621 <.0001 9 108.1520 <.0001 62.6076 <.0001 10 111.2476 < .0001 62.7274 <.0001 11 114.9901 <.0001 63.4837 <.0001 12 120.2228 <.0001 64.4779 <.0001 Relations between Return Volatility and Open Interest The following model is a GARCH(1,1) model when open interest is entered. The origins of analyzed technique can be traced back to Paul Girma and Mbodja Mougou (2002). = + = = + + +

PAGE 30

30 Where rt is the futures return, OP_INI are open interest. Table 43 shows the result of GARCH (1,1), when the open interest is used as an explanatory variable in the conditional variation equation. The data suggests that the coefficient of open interest in the equation is statistically significant at 5% level. Furthermore, the persistence shock is + =0.9897, which remain highly volatility when the open interest was calculated in the model. Besides, the estimator of (0.9570) is larger than that of (0.0327) represents long persistence of volatility. Tabl e 43. GARCH(1,1) Est imate for Volatility and Open I nterest Variable DF Estimate Standard Error t Value Pr > |t| Intercept OP_INI 1 1 1 1 1 0.0622 1.0537E 8 0.0327 0.9570 1.6216E 7 0.0716 5.924E 11 0.007866 0.009734 6.1293E 8 0.87 177.85 4.16 98.31 2.65 0.3846 <.0001 <.0001 <.0001 0.0082 Relations between Return Volatility and Volume The following model is a GARCH(1,1) model when volume is entered. = + = = + + + V represents the daily volume for wheat futures. Table 44 shows the result of GARCH(1,1) when the volume is used as explanatory variable in the conditional variation equation. The data shows that the volume is significant power explanatory for explaining the return volatility. Furthermore, the persistence shock decrease from 0.9897 to + =0.9752, which a slight lower than the model 1. However, the persistence shock maintains a high level.

PAGE 31

31 Table 44. GARCH (1, 1) Estimate for Volatility and Volume Relations between Return Volatility, Volume and Open Interest The following model is a GARCH (1,1) model when open interest and volume are entered simultaneously. = + = = + + + + Table 45 shows the result of GARCH (1, 1), which the volume and open interest have the significant effect on price volatility. The persistence of volatility reduced to + =0.8615. Table 45. GARCH (1, 1) Estimate for Volatility, Open Interest and Volume Through this chapter, we could find that the contemporaneous volume and open interest provide the power explanations for the volatility when enter separately. The coefficients of open interest to volatility are negative and statistically significant. The Variable DF Estimate Standard Error t Value Approx Pr > |t| Intercept Volume 1 1 1 1 1 0.0506 1.0541E 8 0.0336 0.9416 2.0028E 6 0.0719 5.718E 10 0.008998 0.0134 6.2309E 7 0.70 18.44 3.74 70.22 3.21 0.4819 <.0001 0.0002 <.0001 0.0013 Variable DF Estimate Standard Error t Value Approx Pr > |t| Volume OP_INI 1 1 1 1 1 1 0.0193 0.9336 0.0666 0.7949 9.5364E 6 2.158E 6 0.0728 0.4369 0.0229 0.0597 2.8907E 6 8.4214E 7 0.26 2.14 2.91 13.32 3.30 2.56 0.7912 0.0326 0.0036 <.0001 0.0010 0.0104

PAGE 32

32 r esult confirm the Bessembinder and Seguin (1993) conclusion that greater market depth is lower the volatility, given a trading volume. Besides open interest has a relative little impact on volatility than a trade incurs a high trading volume without a corr esponding changing in open interest.

PAGE 33

33 CHAPTER 5 SPECULATION BEHAVIORS IN THE WHEAT FUTUR ES MARKET Since the open interest and trading volume, to certain extent, have led to the price volatility, this chapter will go further to analyze the different participants on the price movement. In Figure 39, it shows that the noncommercial long position takes a much larger part than the commercial position. Meanwhile, Figure 51 indicates that the commercial position maintains a slower and more stable change than non commercial position during the sampling period. The commercial long positions increased steadily, however, not so much volatile as the noncommercial positions did Figure 51. Open interest: Wheat in Chicago Board of Trade Literature Review The relationship between price volatility and types of market participants has been of interest to both scholars and regulators. Peck (1981) finds the relationship between speculators and price volatility in three agriculture commodities. His study finds t he negative relationship between speculation activities and price volatility. Bryant, Bessler 0 100000 200000 300000 400000 500000 600000 Open_Interest_All NonComm_Positions_Long_All Comm_Positions_Long_All

PAGE 34

34 and Haigh (2006) investigate the activities of specific types of traders as causes of price volatility. They analyze eight futures: corn, crude oil, Eurodollar de posits, gold, Japanese Yen, coffee, live cattle, and S&P 500 through vector autoregression (VAR). They also estimate the causal relationships among return, volume, large hedgers activity, large speculators activity, and small hedger activity, small specul ator activity. They find that activity of particular types of traders in futures market did not affect levels of price volatility, either positively or negatively. Shalen develops the noisy rational expectations (NRE) model in 1993. He divides the traders into two categories. He postulates that informed traders have private information while the uniformed speculators have no private information. Therefore, the uninformed traders will attempt to extract price signals from observed futures price changes. His study shows that the uninformed speculators cause the increased price volatility. Daigler and Wiley (1999) examine various financial futures markets. They conclude that trading activities taken on the trading floor are associated with the decreased price v olatility. However, the trading activity through electronic platform are associated with increased volatility. Chang, Chou and Nelling (2000) investigate the S&P 500 futures market. Their empirical study suggests that the large hedging activity causes the increased volatility. They conclude that the large hedging trading activities are correlated with volatility. The demand side of hedging increased, the volatility increased as well. Swanson and Granger(1997) study the causal relationship among the variabl es in a vector autoregression. Reale and Wilson (2001) and Moneta (2004) research on

PAGE 35

35 monetary policy issues by using a VAR model. Akleman, Bessler and Burton (1999) investigate causal relationships among corn exports and exchange rates also using this model. Haigh and Bessler (2004) research on price discovery in cash grain markets and a related transportation market, again using this model. Bryant,Bessler and Haigh (2006)tests causal hypotheses emanating from theories of futures markets through VAR model M ethodology and Data This study explores nonlinearities in the response of speculators trading activity to price changes in wheat futures markets. I analyze weekly data from January 1, 2006 through February 28, 2010. The sample consists of 216 observations Data on futures price come from Bloomberg DataStream. The returns are calculated as 100 times the natural logarithm of the first differences of weekly futures settlement prices. The trader position data are from Commitments of Traders (COT) reported on t he official website of Commodity Futures Trading Commission (CFTC). All of the traders reported futures positions are classified as commercial if a trader uses futures contracts for hedging or as noncommercial if a trader uses the futures contract for profit. Rothig and Chiarella (2007) defined changes in hedger and speculators position as 100 times the natural logarithm of the first differences of position respectively. This method could effectively reflect the changes of position. Thus, I use VAR model to measure the changes in hedger and speculators long positions. Table 51 shows the summary statistics for the return and changes in speculator and hedger positions. The distributions of speculators indicate significant positive skewness. Furthermore, the speculator shows evidence of slightly excess kurtosis

PAGE 36

36 compare to the normal distribution. Harvery and Siddique (1999) shows that the excess skewness and kurtosis potentially indicate the presence of heteroscedasticity. Table 51. Summary Statistics To avoid the spurious regression problem, I took the stationarity test for the three categories. The Dickey and Fuller (1979) have developed a method to test the stationary series. The method has become very popular over the past thirty years. The method is also known as Augmented Dickey Fuller (ADF) test. The ADF test consists of estimating the following equation; = + + + Where is the first difference v alue of X, is the first lagged valued of X. is the t h lagged of first differenced of values of X. The null hypothesis of nonstationary is = Dickey and Fuller have shown that the t value for the follows the statistics. The Table 52 shows that the statistics are 10.38 for speculator, 8.63 for hedger and 9.97 for return. All of them are significant at 5% level. Thus the speculator position changes, hedge position changes and return display stationary fl uctuations. N Mea n Std Skewneww Kurtosis ShapiroWilk Kolmogorov Smirnov Cramer V on Mises AndersonDarling Speculator 216 0.1175 6.5002 0.1141 1.9833 < 0.0001 <0.0100 <0.0050 <0.0050 Hedger 2 16 0.1342 2.6175 0.1377 0.2140 0.3116 >0.1500 0.2290 0.1861 Return 216 0.1627 5.3436 0.0240 0.02101 0.9182 >0.1500 >0.2500 >0.2500

PAGE 37

37 Figure 52. Speculators Position(%) Figure 53. Hedgers Position(%) Figure 54. Daily Return(%) 30 20 10 0 10 20 301/3/06 3/3/06 5/3/06 7/3/06 9/3/06 11/3/06 1/3/07 3/3/07 5/3/07 7/3/07 9/3/07 11/3/07 1/3/08 3/3/08 5/3/08 7/3/08 9/3/08 11/3/08 1/3/09 3/3/09 5/3/09 7/3/09 9/3/09 11/3/09 1/3/10 10 5 0 5 10 20 15 10 5 0 5 10 15 20 1/3/06 3/3/06 5/3/06 7/3/06 9/3/06 11/3/06 1/3/07 3/3/07 5/3/07 7/3/07 9/3/07 11/3/07 1/3/08 3/3/08 5/3/08 7/3/08 9/3/08 11/3/08 1/3/09 3/3/09 5/3/09 7/3/09 9/3/09 11/3/09 1/3/10

PAGE 38

38 Table 52. Unit Root Test and ARCH Affect Test The results of the Granger causality test in my study are presented in the Table 54.The no causality null hypothesis can be rejected for changes in commercial hedger long positions and return. The hedging strategies did not affect the changes in futures prices. These results support the theory that hedgers hold futures positions corresponding to their spot positions. On the other hand, the test shows the significant evidence that the speculators affect the fluctuation of return. Therefore, it could conclude that the extreme volatility is caused primaril y by speculators. Table 53. Granger Causality Wald Test Test DF Chi Square Pr > ChiSq 1 1 0.13 0.7136 2 1 4.97 0.0258 Test 1: Group 1 Variables: price Group 2 Variables: Commercial Test 2: Group 1 Variables: price Group 2 Variables: Noncommercial The Table 54 presents the model diagnostic analysis, autocorrelation, normality and ARCH effect test. The normality shows that the residual has a normal distribution and no ARCH effect. Table 54. White Noise Diagnostics Variable Durbin Watson Normality ARCH Chi Square Pr > ChiSq F Value Pr > F Return 2.03170 2.54 0.2807 2.68 0.1030 Lag Tau Pr < Tau Q Pr > Q LM Pr > LM Speculator 1 2 10.38 <0.0001 9.33 <0.0001 9.8737 0.0017 9.7982 0.001 10.3734 0.0056 9.7998 0.0074 Hedger 1 2 8.63 <.0001 7.25 <.0001 0.1818 0.6699 0.1417 0.7066 0.4774 0.7877 0.4613 0.7940 Return 1 2 9.97 <.0001 8.42 <.0001 8.2822 0.0040 8.2903 0.0040 8.2986 0.0158 8.4636 0.0145

PAGE 39

39 In an effort to improve the transparency of the futures market, in 2006, the CFTC removed the swaps position from the hedger position to measure the trading activates through the OTC market. Therefore, by using the daily data, we calculated the correlations between market volatility and the presence of each trader type, including hedger, swaps dealers, speculators and others. The market presence repres ents the percentage of all open interest held by a trade group. Volatility appears positively related to the presence of two of the four large trader groups, but is negatively related to the presence of hedger and other traders. Speculators and swaps dealers tend to move together with return. Table 55. Correlation Analysis Volatility Hedger Others Swaps dealer speculators Volatility 1.00000 Hedger 0.15506 1.00000 Others 0.08836 0.07001 1.00000 Swaps dealer 0.24209 0.00001 0.09483 1.00000 Speculators 0.36956 0.18371 0.07241 0.10347 1.00000

PAGE 40

40 CHAPTER 6 THE CASH MARKET AND FUTURES MARKET In this Chapter 6, the convergence problem will be discussed. Futures market and cash market are always interacted. Under the efficient market, there exists a longterm equilibrium between cash and futures prices. In addition, the futures market has price discovery function, which helps the farmers, grain elevators and other merchants better forecast the trend of cash prices. However, in recent years, is there still a longterm equilibrium between cash and futures market? Is there the bidirectional casualt y between cash price and futures price? This chapter tries to find empirical evidence and answer the above questions. Cointegration is considered as a necessary condition for market efficiency according Lai and Lai, 1991. However, to conclude efficiency, w e should also examine whether futures contracts are unbiased predictors of spot markets. If the wheat spot and futures contract prices are cointegrated, then a longrun relationship must exist between these two series. Cointegration is an appropriate indic ator of longterm co movement. Among all cointegration tests, such as the Engle and Granger (1987) method and the Stock and Watson (1988) test, Johansens test has a number of desirable properties, including the fact that all test variables are treated as endogenous variables. First of all, the AugmentedDick fuller test has used to whether data series for wheat futures and spot are stationary. Second, through cointegration, I study whether wheat futures and cash prices have a longterm equilibrium. If bo th of them are cointegrated, the futures price should be the unbiased estimator of cash prices, as shown in the following equation:

PAGE 41

41 = + + The data series for cash price are obtained from USDA and future prices from Bloomberg. The futures prices used in the test are the March Wheat futures prices. I created two data series from January 1, 2006 to February 2, 2010. There are total 1038 observations. Figure 61 shows that both of ln and are nonstationary, but the first differences are stationary. Table 61. Unit Root Test Type Lags Rho Pr < Rho Tau Pr < Tau F Pr > F Spot Futures 1 1 4.0151 4.8218 0.5376 0.4526 1.54 1.87 0.5119 0.3458 1.25 1.88 0.7498 0.5898 First_dif_Spot First_dif_Futures 1 1 1060.42 1114.51 0.0001 0.0001 22.99 23.57 <.0001 <.0001 264.18 277.71 0.0010 0.0010 Johansens (1998) approach is used to test for co integration. We consider a general VAR model of order k, = + + + Where = and D is a deterministic term, and are matrices of coefficient. The cointergration relationship is examined by looking at the rank of the coefficient of matrix If =0, there is no cointegration vector, hence no cointegration relationship. If =1, the two series are cointegrated (Johanse and Juseliues,1990) The trace numbers are in the Table 62, which shows that there is no a longterm relationship between cash and futures prices. Neither of the series could predict its counterpart.

PAGE 42

42 Table 62. Cointegration Rank Test H0: Rank=r H1: Rank>r Eigenvalue Trace 5% Critical Value Drift in ECM Drift in Process 0 1 0 1 0.0062 0.0003 6.4447 0.0451 12.21 4.14 NOINT Constant The result of cointegration test shows no longterm relationship between cash and futures market. It is believed that the great volatility in wheat futures market has deteriorated the efficient of futures market

PAGE 43

43 CHAPTER 7 COMMODITY INDEX AND INDEX SPECULATION Another phenomenon in driving up the commodity futures was deregulation of index trading activates The U.S. Senate Permanent Subcommittee in 2009 report suggested that Index sp eculators had contributed to the unreasonable price fluctuation in the energy and agriculture c ommodities and distorted the function of futures market, which could not reflect the accurate information of supply and demand in the physical market. The index traders, to the extent, are like speculators. However, they focused on longside exposure to duplicate an index of mul tiply commodities. Their trading goals are to diversify the investment portfolio. Because the returns between commodity and other securities such as stocks and bond may be negatively correlated, commodity investment may be optimal strategy for hedging purpose. Traders are commonly called the longterm investors, since they seek exposure to commodities through passive longterm investment. This strategy could be implemented via swaps agreement or mutual funds, exchange traded funds. Usually they enter into a swap agreement for underlying asset of a specific commodity index. The counterparts swap dealers, who have the hedge exemption, enter the futures market and take long positions in commodity index. T here are two important indices: Standard Poors Goldman Sachs Commodity Index(S&P GSCI) and the Dow Jones AIG Commodity Index (DJAIGCI). Standard PoorsGoldman Sachs Commodity Index The Figure 71 following shows the weight composition of each sector. One of the famous commodities indices, GSCI was created by Goldman Sachs in July 1992 and acquired by Standard &Poors in 2007. It includes 24 commodity futures contracts

PAGE 44

44 traded on U.S. exchanges: six energy products, five industrial metals, eight agricultural products, three livestock products and two precious m etals. Futures contracts on the GSCI are traded at the Chicago Mercantile Exchange (CME).There is no limit on the number of contracts that may be included in the S&P GSCI. Its returns are calculated based on the arithmetic average of long positions in futures contracts (S&P GSCI Index Methodology 2010 edition). Goldman Sachs has calculated the historical value of the GSCI and normalized to a value of 100 on January 2, 1970, in order to permit comparisons of GSCI values over time. Figure 71. S&P GSCI Index Weight Composition The contract value of each GSCI futures position is $250 times the GSCI index. When futures contracts in the GSCI are close to expiration, they are rolled forward to the next nearest contracts at the beginning of their expiration m onths. The rollover period is from the fifth through the ninth business day of each month. During this period, the GSCI portfolio is shifted from the existing to the next nearby futures contract at a rate of 20% per day. This means that on the fifth business day, the GSCI is adjusted to

PAGE 45

45 consist of 20% of the second nearby contracts and 80% of the existing contracts. Likewise, on the sixth business day, the GSCI include 40% of second nearby contracts and 60% of the existing contracts(S&P GSCI Index Methodology 2010 edition). On the last day of the rolling period, the GSCI only include the second nearby contract. Therefore, the GSCI will always have contracts with two maturities for each of the 10 commodities between the fifth and ninth business day of each month. For other commodities with fewer futures, such as livestock, they are rolled forward less frequently. Wheat Market and S&P GSCI Related search includes Stoll and Whaley 2009, they analyze the price co movements of index commodities and Granger Causal ity between index inflow and wheat future recent. They found that commodity index rolls have little futures price impact the failure of the wheat futures price to converge has not undermined the futures contracts effectiveness as a risk management tool and correlation of commodities price levels that include in the commodity index of are quite low. Another research is carried by Irwin, Sanders and Merrin 2009. They also use the Granger Causality Test to analyze the commodity index and price change and come up with the same results as Stoll and Whaley. What is more, they also conclude that when price low, seller complains the speculators while when price are high, buyer complain the speculators Actually, it is the historical pattern of attacks upon speculation. In 2009, Congress published staff report titled Excessive Speculation in the Wheat Market. The report shows in detail about how commodity index traders affected the price of wheat contracts traded on the Chicago Mercantile Exchange over t he past three years. The report pointed out around onethird and onehalf of all of the

PAGE 46

46 outstanding wheat futures contracts purchased by index traders offsetting part of their exposure to commodity index instruments sold to third parties. The report also f inds that there are significant evidences that commodity index traders were one of the major causes of unwarranted changes in futures market. Through correlation analysis, it is worthy noticing that in the Table 7 1 the significant relationship between S &P GSCI and wheat futures. Table 71. Pearson Correlation Coefficients Prob > |r| under H0: Rho=0 W heat S&P GSCI W heat S&P GSCI 1.00000 0.43409 (<.0001) 0.43409 (<.0001) 1.00000 Figure 72 and Figure 73 are the daily return of S&P GSCI futures and wheat futures traded on Chicago Mercantile Exchange (CME). The returns are measure by 100*ln(Ft/Ft 1). We observed the increased return variability in the recent years. Date series are obtained from the Bloomberg. Figure 72. Daily Return of GSCI Futures(%) 10.00 8.00 6.00 4.00 2.00 0.00 2.00 4.00 6.00 8.00 1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

PAGE 47

47 Figure 73. Daily Return of Wheat Futures(%) Using the same method from the Chapter 6, I analyze the casualty between S&P GSCI futures contract and Wheat futures contract. The model can be written in matrix form as: = + ,, , + Where the e is stochastic error term, call shocks in the VAR model. Before estimate the equation, we should judge the lag length. Including too many lagged terms will consume degrees of freedom and potentially induce multicollinearity. One way of deciding the lag length is to use a criterion such as Akaike and Schwarz rule. Using option=ESACF ,MINIC and SCAN, SAS program provides us the optimal lag length. In this study, the l ag length one is appropriate suggested by SAS output. The ADF test in the Table 7 2 suggests both of the variables are stationary. The Table 74 presents the VAR results based on one lag of each variable. The parameters tests given in the table are to test the hypothesis that collectively the various lagged coefficients are zero. The test shows that the first equation is not statistically different from zero, but the second equation is significant. Based on the result the wheat 15.00 10.00 5.00 0.00 5.00 10.00 1/3/2006 1/3/2007 1/3/2008 1/3/2009 1/3/2010

PAGE 48

48 is dependent on GSCI fluctuat ions. Next, the granger causality confirms the VAR result. The no causality null hypothesis is rejected for GSCI and Wheat futures. The changes of wheat futures did not affect the GSCI fluctuation. However, the GSCI significantly influence the wheat price fluctuation. Table 73. Causality Wald Test Test DF Chi Square Pr > ChiSq 1 1 0.04 0.8322 2 1 11.24 0.0008 Test 1: Group 1 Variables: GSCI Group 2 Variables: wheat Test 2: Group 1 Variables: wheat Group 2 Variables: GSCI Table 74. Model Parameter Estimates Test DF Chi Square Pr > ChiSq 1 1 0.04 0.8322 2 1 11.24 0.0008 Test 1: Group 1 Variables: GSCI Group 2 Variables: wheat Test 2: Group 1 Variables: wheat Group 2 Variables: GSCI Equation Parameter Estimate Error t Value Pr > |t| Variable GSCI CONST1 0.01497 0.05827 0.26 0.7973 1 AR1_1_1 0.05999 0.03437 1.75 0.0812 GSCI(t 1) AR1_1_2 0.00557 0.02627 0.21 0.8322 wheat(t 1) wheat CONST2 0.03514 0.07589 0.46 0.6435 1 AR1_2_1 0.15011 0.04477 3.35 0.0008 GSCI(t 1) AR1_2_2 0.01522 0.03422 0.44 0.6566 wheat(t 1) Testing of the Parameters Test DF Chi Square Pr > ChiSq 1 2 3.41 0.1816 2 2 12.49 0.0019 Correlations of Residuals Variable/ Lag 0 1 2 3 GSCI ++ .. .. .. Table 7 2. Summary Statistics and Unit Root Test Variable N Mean Std.Dev Skewness Kurtosis Q(6) Pr>ChiSq ADF Pr < Tau GSCI Futures 1044 1044 0.0142 0.0333 1.8819 2.4618 0.2479 0.0663 2.0730 1.0890 6.57 5.23 0.3625 0.05143 23.68 23.27 <.0001 <.0001

PAGE 49

49 Information Criteria AICC 2.857007 SBC 2.885449 HQC 2.867775 AIC 2.856974 FPEC 17.40877 wheat ++ .. .. .. + is > 2*std error, is < 2*std error, is between

PAGE 50

50 CHAPTER 8 REGULATORY SYSTEM AND POLICY RECOMMENDAT ION Development of Futures Regulation Through the empirical analysis, we have found that the speculator s especially index speculator s have tremendously influence d wheat futures market. Now we turn our focus to the federal regulatory system. We will look at the evolvement of futures regulation and research on whether the current regulatory system is effective to prevent excessive speculation. The first complete regulatory framework for futures market in U.S. history is Commodity Exchange Act (CEA), which passed in 1936. The Commodity Exchange A ct (CEA) required that all regulated agriculture commodities had to be traded only on the federal designated markets. Also the Act authorized the Commodity Exchange Commission as an agency in the Department of Agriculture to oversight of the futures marke t. In addition, the Commodity Exchange Act required Commodity Exchange Commission to set up speculative position limits to prevent excessive speculation. In 1974, the Commodity Futures Trading Commission (CFTC) was founded as an independent regulatory agency to oversee futures market. The C FTC was empowered more jurisdiction than ever before. P reviously, the Commission could only regulate agricultural commodities listed in the Commodity Exchange Act, whereas CFTC have authorized to regulate all types of futures trading in all commodities. The 1980s saw a rapid development of derivatives market. Financial derivatives became common and served as effective risk management tool s. Numerous types of financial derivatives had been created, such as foreign exchange, interest rate and even credit rating. At the same time regulatory burdens were blamed for retarding the

PAGE 51

51 progress of the OTC derivatives market. Particularly, in 1990s, t he legal uncertainty has been a great problem when financial institutions developed a new instrument, which was potential to reduce the flexibility and competitiveness of U.S. financial markets. To keep the competitive advantage of U.S futures markets and ensure integrity and efficiency of the market, the Congress passed the Commodity Futures Modernization Act (CFMA) i n 2000. The CFMA of 2000 incorporated much more flexible standards into the act and greater legal certainty in the OTC market. The CFTC established a new regulatory framework with the guidance of innovation of the OTC market and elimination of excessive burden on sophisticated institutions or wealthy participants. Later on, the CFMA h as caused a series of serious consequence for OTC market. Since the CFMA clarified that derivatives transactions was outside of its jurisdiction, which has been believed to contribute to the turmoil in futures market. The CFMA created three categories of c ommodities: agriculture commodities, excluded commodities and exempt commodities. The excluded commodities are defined as interest rates, exchange rates, currencies, credit risk, debt instruments, measures of inflation, or other macroeconomic index. The ex empt commodities are neither excluded commodities nor agriculture commodities. Particularly, the energy and metal contracts are categorized as exempt commodities. Exempt commodities have two specific provisions. First, a contract or agreement enters into between eligible contracts part icipants and not traded on trading facilities is outside of the CFTC jurisdiction. Second, the contract or transaction entered between the eligible contracts participants traded or executed on an electronic trading facility is also exempt from the CFTC regulation, which is also known as Enron Loophole.

PAGE 52

52 Before 2000, there only existed the Designat ed Contract Market (DCM). The CFMA in 2000 established another two regulated markets: Derivatives Transaction Execution Facilities (DTEF) and Exempt Market. The three types of exchanges bas ed on the nature of futures traded and knowledgeable participants are subject to the varying levels of regulation. The Design ated Commercial Markets open to all types of market participants and any t ype commodity and has the highest level of regulatory oversight. The exchange should report to the CFTC the larger trader positions on daily basis. Besides, the CFTC was required to continuously keep this kind of market under surveillance. The Chicago Board of Trade (CBOT), Chicago Mercantile Exchange (CME) and New York Mercantile Exchange (NYMEX) are Designated Commercial Markets. The Derivatives Transaction Execution Facilities r eceives an intermediate level of regulations. This kind of market is divided into another two categories, Regular and Commercial. This trading facility is limited to eligible commercial entities. Any commodity other than an agriculture commodity can be traded. However, some agriculture commodities could be trade if approved by the commission. In addition, the Derivatives Transaction Execution Facilities must comply with some requirement s, for example, establishing the compliance and surveillance, monitoring of trading activa tes and having a record keeping system. Unlike the Designated Commercial Market and Derivatives Transaction Execution Facilities, the third Exempt Market is exempted from most provisions of CEA. There are two types of exempt markets, Exempt Board of Trade (EBOT) and Exempt Commercial

PAGE 53

53 Market (ECM). The commodities eligible to be traded on an Exempt Board of Trade (EBOT) are excluded commodities. The CFMA established separated exemptions for Exempt Commercial Market (ECM). To be qualified for these exemptions the contracts and transactions must be traded or executed on electronic trading facilities between eligible commercial entities In addition, the commodities traded on the Exempt Commercial Market are restricted to the exempt commodity. This trading facility operating as an Exempt Commercial Market (ECM) must limit trading to eligible commercial entities Speculative Position Limits To prevent excessive speculation, an effective regulatory system is the guarantee. There are two important and effective practices for futures market regulations: position limit and daily reporting system. The speculative position limits have been believed to be an effect tool for the commission to impede the manipulation of U.S. futures markets and prevent unreasonable price fluctuations. From its inception, this rule was little changed until 1981 that the CFTC required exchanges to adopt position limits for all t he contracts not listed on the federal position limits Currently, the position limits are regulated under two frameworks (Table 81). One is the federal position limits for agricultural commodities in futures and option market, which are enumerated in the federal regulations (Table 82). The other is exchangeset speculative position limits. Both the CFTC and the Designated Commercial Market take on the enforcement of the exchangeset speculative position limits.

PAGE 54

54 Table 81. Current Practices in Setting Spec ulative Position Limits and Position Accountability Levels In selected Futures Contract Market as of July 24, 2009 Source: Federal Regulation Category Sample Contracts Speculative Position Limits( Accountability(PA) Spot Month Single Month All Months Combined CFTC Speculative Position Limits Federal Limits Corn, oats, soybeans, Soybean oil, Soybean meal, wheat and Cotton Limits and Levels Set by Exchanges Other Agricultural CME Frozen Pork Bellies; ICE U.S. Frozen Concentrated Orange Juice CME Livestock and Milk ICE Cocoa, Coffee and World Sugar Energy NYMEX Crude Oil, Natural Gas, Heating Oil, and Gasoline; ICE U.K.WTI PA PA Metal COMEX Gold and Silver PA PA Table 82. Speculative Position Limit (2010) in Contract Units Source:Federal Regulation Contract Spot month Single month All months Chicago Board of Trade Corn and Mini Corn 600 13,500 22,000 Oats 600 1,400 2,000 Soybeans and Mini Soybeans 600 6,500 10,000 Wheat and Mini Wheat 600 5,000 6,500 Soybean Oil 540 5,000 6,500 Soybean Meal 720 5,000 6,500 Minneapolis Grain Exchange Hard Red Spring Wheat 600 5,000 6,500 New York Board of Trade Cotton No. 2 300 3,500 5,000 Kansas City Board of Trade

PAGE 55

55 Hard Winter Wheat 600 5,000 6,500 Reporting System The daily report system could be trace d back to early 1920s. It required the clearing member on each exchange to report the large trader positions. This system is an important part of regulation and is still used by the CFTC today. Commi tment of Trader Report The Commitment of Trader Report is published weekly by the CFTC and contains more detailed information about position holdings, spreading, open interest by category, and numbers of traders. The categories are: 1) producer /merchant/processor/user, who engages in the production, processing, packing or handling of a physical commodity and uses the futures markets to manage or hedge risks associated with those activities 2) speculator who is a registered commodity trading adv isor (CTA), a registered commodity pool operator (CPO). These traders are engaged in operating a fund for a commodity pool, that is, an enterprise in which funds contributed by a number of persons are combined for the purpose of trading futures contracts or commodity options, or to invest in another commodity pool. ( National Futures Association definition). 3) other reportable, who are not falling in any of the other two categories. Commodity Index Trade Supplement The CFTC also published weekly Commodity Index Trade (CIT) Supplement on its website. CIT report contains 12 agricultural commodities, providing more detailed information about index trader activities, for example, long/short position, percentage open interest and number of trader and changes in their position.

PAGE 56

56 Special Calls The special calls provision could be traced back to the 1970s when the CFTC encountered the problems of regulating t he individuals located outside the U.S. and trade in U.S. market. For the people in the domestic market, the large traders are required to file reports of their daily positions to the CFTC. This requirement would make it easy for the CFTC to inspect that w hether the unreasonable price movement was caused by the large positions. However, the CFTC could not obtain the large positions hold by the foreign brokers customers and their financial interests. To solve the imbalance of regulations between foreign and domestic traders, the CFTC adopted the special calls provision in which anyone receiving the special call must provide the information the CFTC asked for, such as a firms trading and delivery activity. Although the foreign large traders do not have daily position responsibility, but they have to provide the business activities and daily positions if the CFTC issue the special calls to get those information Enron Loophole The CFMA allowed exempt commodities such as oil to be traded on Exempt Commercial markets, free from the most of requirement by the CFTC and speculative position limits, which are accommodated industry interests and fostered anticompetitive behaviors and market manipulations. Particularly, Enron has been reported to purposefully, improperly influence energy prices in western markets. Enron, through its Enron Online (EOL) launched in 1999, provided webbased electronic trading platform for wholesale energy, swaps, and other commodities. The daily average transac tions on the platform were $2.5 billion. 2,100 products were offered to traders across four

PAGE 57

57 continents based on 15 different currencies. Between 1999 and 2001, Enron Online reported the profit by derivatives trades as $920 million. Swaps Loophole The swaps market has grown rapidly since 1980, because it provides the participants more flexible ways to hedge their exposure risks. According to the Bank for International Settlements, the amounts of outstanding of global over t hecounter (OTC) derivatives are estimated to be $614 trillion in December 2009. In essence, a swap contract is an agreement between two parties that exchange a series of cash flows in the future. Compare to other financial market, the swaps market has it unique characteristics: 1) It affords the privacy for the swaps trading .Only two counterparties knows the swap trading. 2) The swaps market could escape the government regulation. In the past, the CFTC believed that the swaps dealers had no incentive to manipulate futures prices. Just like bona fide physical hedger, swaps dealers used the futures market to hedge their transactions. Therefore, the CFMA of 2000 greatly swept away the legal uncertainty by significantly expanding the exemptions of swaps transactions on financial instruments and physical commodities. The CFMA allowed the swaps transactions based on the excluded commodity and exempt commodity to be exempted from the regulations of the CFTC and Securities Exchange Commission. Nowadays, the agriculture commodities could be traded on the multiple exchanges, including the OTC market. Since s waps dealers who have hedge exemptions could enter the futures market with unlimited quantities institutional investors investing in commodity futures do not directly trade through futures exchange. They used the swaps co ntracts to hold an excessive amount of futures contracts without any constrain from speculative position limits. In absence of speculative position limits, the commodities

PAGE 58

58 derivatives market are exposure to excessive speculation. It was estimated that, in 2006, Goldman Sachs and Morgan Stanley had earned billions of dollars in energy trading for two years. In the end of 2007, the CFTC issued special calls to 32 entities and their sub entities to get a comprehensive knowledge of the quantity of commodity index trading in OTC and exchange markets. The special calls of 2007 required the entities who engaged in the index trading activity to provide the notional value of their index business including in both domestic market and foreign markets. The index commodity funds should classify the positions they hold directly in the futures market and positions through swaps dealers. The data collected by the CFTC from the special calls demonstrated the conclusion that positions held by the swap clients exceed the feder al speculate limits. For wheat futures, the total notional value of open contracts on June 30, 2008 was estimated to be $19 billion and the net notional index value took approximately 47 percent of this total. The increased commodity index trading has made the CFTC reconsider whether it is appropriate to put the swap dealer trading in the commercial category, because swaps dealers are using the future market to hedge risk in the OTC market, but their client might be a speculator. In an effort to improve the transparency of the futures market, the commission publishes the commitments of trader report, which remove Swap Dealer from Commercial Category and Create New Swap Dealer Classification since Jan 1, 2006. The swap dealer in the Commitment of Trader defines as the people who deal primarily in swaps for a commodity and uses the futures markets to hedge the

PAGE 59

59 risk associated with those swaps transactions. However, the swap dealers counterparties may be speculator, like mutual funds, or traditional commercial hedger, In addition,t here is a little bit difference between index traders position in the supplement and swap dealer position in the Commitment of Trade (COT). The swap dealer category includes some position that has nothing with commodity index business On the other hand, the index trader category in the CIT supplement, including the institutional investors who do the index trading directly in the futures market rather than go to OTC market. The institutional investor positions are classified as managed money or other reportable in the Commitment of Trade (COT). Therefore, the index trading activities through the swap dealers still could not be precisely measured. What is more, the CFTC published the Quarterly Index Inv estment Data on its website since 2 008. The data was selected from the special calls issued to the swap dealers and index traders. The report includes the national values and the equivalent number of futures contracts for all U.S. markets with more than $0.5 billion of reported net notional value of index investment at the end of any one quarter. Policy Recommendation Today, the CFTC jurisdiction scope expands much larger than ever before. It regulates the activities of about 3000 commodity exchange members, 360 public brokerage houses and 2,500 commodity trading advisers and commodity pool operators. In addition, off exchange transactions involving instruments with similar feature to futures contracts are also within CFTC jurisdiction. Futures contracts are so diversified that the underlying asset have expanded to precious metals, raw materials, foreign currencies, commercial interest rates, U.S. governments and so on. A number of contracts have begun to be traded on multiple exchanges, including exchanges outside

PAGE 60

60 the United States. Therefor e, in order to keep the pace with development of futures market, the CFTC need some amendments to maintain the integrity of the market and hedgers benefit. The CFTC reauthorization Act of 2008 (Farm Bill 2008) expands the CFTC authority by creating a new regulatory category, which is the contract with the significant price discovery function traded on the exempt commercial market. Price discovery is an important function for futures market, since it helps people have more accurate forecasts and a better i nvestment plan. The futures prices cannot be known with certainty today, the high level of relationship between cash and futures markets make it possible for people to estimate the future price better. The Title XIII of the Farm Bill provides standards applicable to significant price discovery contracts. Once the CFTC finds that a contract traded on exempt commercial market performs significant price discovery function, the contract will be subject to Commodity Exchange Act and commission regulations. Impr ove Transparency and Data Accuracy In this study, we find that the index trading activities through OTC market are still not enough to judge and measure. The data collected though the special calls wil l take a comparative long time. The new Commitment of T rader since 2006 removes the swaps dealers from the commercial categories. However, the swaps dealers clients might be hedge and speculators. In order to enhance the transparency of futures and options market, the Commission has to improve the weekly comm itment of Trader reports by including more detail trader classification.

PAGE 61

61 Improve Regulation on Index Trading Activity The special calls of 2007 illustrated that the positions held by the many swaps dealers client has exceed the speculative position limits. A question is raised by whether swaps dealers should receive hedger exemption from speculative position limits in consideration of the mix of commercial and noncommercial activity. In order to keep the regulatory consistency and market integrity, t he CFTC should improve the evaluation of the noncommercial activity by swaps dealers and prevent the noncommercial positions from exceeding position limits.

PAGE 62

62 CHAPTER 9 CONCLUSION This thesis describes the wheat futures market and speculation behaviors. The main conclusions are The futures market experienced a great volatile in recent years. Through the basis analysis, the large discrepancy between spot and futures market during the expiration month has been a major problem in the wheat market. The volatilit y of wheat futures shows strong heterosckedascity and autocorrelation. Based on the analysis of relationship between spot and futures price, they are not cointegrated and the futures price did not contain much information to reflect the spot price and vers e well. Therefore, the larger volatility of wheat futures market has lower the efficiency of price discovery functions Based on the GARCH model, the volume and open interest provides a power explanation for the volatility when enter separately. The coeffic ients of open interest to volatility are negative and statistically significant. In addition, the test found that the greater market depth, the lower the volatility, given a trading volume. Besides open interest has a relative small impact on volatility than a trade incurred a high trading volume without a corresponding changing in open interest. The granger casualty test indicates that the speculator trading activates have led to the change of wheat return. The hedger activates comparatively shows steady and little impact on the volatility of wheat futures. Since the data are not available for the index traders and swaps dealers. That the CFTC issuing the special calls to swaps dealers and then collecting them need a long time. Therefore, casualty test to m easure

PAGE 63

63 the index trader and volatility is hard to carry on. However, through special calls, there still a clear information that a large fund flow into the wheat futures. Since 2000, the CFTC has ongoing deregulated over the counter market. This regulatory system make s the excessive speculation possible. M any participants through the OTC market hold more position than position limits. In addition, the Farm Bill still did not stop the index trader through OTC market to influence the futures price. Future res earch may have two directions: first, t he sampling period of this study is during the financial crisis, there must be some endogenous variables that m ight distort the futures market or even deteriorate the economy. Understanding these inner links will be better for find out the reason of extremely volatile in futures market. Second, i n order to actually measure the index traders activates, improve the futures market transparence, I will expand my to research scope to other commodities, such as soybean, cott on and grain. After acquiring the relative data about index traders business activities, further research on this topic will be move d on.

PAGE 64

64 LIST OF REFERENCES Akleman, D.G., Bessler, D.A. &Burton, D.(1999). Modeling Corn Exports and Exchange Rates with Directed Graphs and Statistical Loss Functions, in Clark Glymour and Gregory Cooper editors Computation, Causation, and Discovery Cambridge, MA: AAAI/MIT Press, 497 520. Alexia B.,& Meredith S,.(2007). Beyond Enron: Regulation in Energy Derivatives Trading, Northwestern Journal of International Law &Bussiness 27,665 Abbott, P.C., Christopher H.& Wallace E.T.(2008). Whats Driving Food Prices? A Farm Foundation Issue Report, July. Babula, R., Bessler, D.A.& Payne,W.S.(2004). Dynamic Relationships among U.S. Wh eat Related Markets: Applying Directed Acyclic Graphs to a Time Series Model. Journal of Agricultural and Applied Economics 36,122. Bessembinder,H.,& Seguin,P.J.(1993). Price volatility, trading volume, and market depth: Evidence from futures market. Journal of Financial and Quantitative Analysis 28, 2139. Black, F. & Myron S. (1973). The Pricing of Options and Corporate Liabilities, Journal of Political Economy 81,637659. Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity Journal of Econometrics 31, 307327. Bollerslev,T. (1987). A conditional heteroskedastic time series model for speculative prices and rates of return. Review of Economics and Statistics 9,542547. Bryant, H. L., Bessler, D. A.& Haigh M.S.(2006). Causali ty in futures markets,Journal of Futures Market 26,10391057 Carlton, D. (1984).Future Markets: Their Purpose, Their History, Their Growth, Their Successes and Failures. The Journal of Future Markets 4(3), 237271. Chang, E. C. (1985). Returns to speculators and the theory of normal backwardation. Journal of Finance 40,193208. Chang, E. C., Chou, R. Y.,& Nelling, E. F. (2000). Market volatility and the demand for hedging in stock index futures. Journal of Futures Markets 20,105125. Commodity Futures Trading Commission (2008). Agricultural Forum April 22. Commodity Futures Trading Commission (2008), Office of the Chief Economist. Personal Communication with Dr. Jeff Harris, July

PAGE 65

65 Commodity Futures Trading Commission (2008). CFTC Approves Amendments to the CBOTs Wheat Futures Contract, Press Release No. 557908, December 4. Commodity Futures Trading Commission (2008). Staff Report on Commodity Swap Dealers & Index Traders with Comm ission Recommendations September Commodity Futures Modernization Act of 2000 Commodity Future Trading Commission, Fiscal Year 2002 Annual Report. Cornell, B. (1981). The relation between volume and price variability in futures markets. Journal of Futures Markets 1, 303 316. Cooper, R.N., & Lawrence. R.Z. (1975) The 197275 Commodity Boom, Brookings Papers on Economic Activity 3,671723. Daigler, R. T., & Wiley, M. K. (1999). The impact of trader type on the futures volatility volume relation. Journal of Finance 54, 22972316. Domanski, D, &A. Heath (2007). Financial Investors and Commodity Markets, Bank for International Settl ements Quarterly Review 5367. Fortenbery, T.R,&Hauser. R.J.(1990). Investment Potential of Agricultural Futures Contracts. American Journal of Agricultural Economics 72(3),721726. Fujihara, R.A.& Mougoue, M.(1997). An examination of linear and nonlinear causal relationships between price variability and volume in petroleum futures markets. Journal of Futures Markets 17, 385 416. Girma,P.B.,&Paulson,A.S.(1998).Seasonality in petroleum futures spreads. Journal of Futures Markets 18, 581 598. Girma,P.B.,& Mougou ,M.(2002).An empirical examination of the relation between futures spread volatility, volume and open interest. The Journal of Futures Markets 22(11), 10831102 Gorton, G., & Rouwenhorst. K.G. (2006).Facts and Fantasies about Commodity Futures. Financial Analysts Journal 62(2),4768. Greenberger, M.(2007), Testimony of Energy Speculation: Is Greater Regulation Necessary to Stop Price Manipulation? Greenberger, M. (2008).Testimony of Energy Market manipulation and Federal Enforcement Regimes, Haigh, M. S. (1999).Cointegration, unbiased expectations and forecasting in the BIFFEX freight futures market. Journal of Futures Market 20,546571.

PAGE 66

66 Haigh, M., Nomikos,N.K.& Bessler,D.A.(2004). Integration and Causality in International Freight Markets Modeling with Error Correction and Directed Acyclic Graphs, Journal of Southern Economic 71,145162. Haigh, M. S.,& Bessler, D. A. (2004). Causality and price discovery: An application of directed acyclic graphs. Journal of Business 77,10991121. Irwin, S.H.,Garcia, P.,Good,D.L.,& Kunda,E.L.(2008) Recent Convergence Performance of CBOT Corn, Soybean, and Wheat Futures Contracts, Choices American Agricultural Economics Association, 2nd pp. 1621. Irwin, S.H., Sanders D.L.,& Merrin, R.P. (2009). Devil or Angel: The Role of Speculation in the Recent Commodity Price Boom (and Bust), Invited paper at the Southern Agricultural Economics Association Meetings, Atlanta, GA. Johansen, S (1988). Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control 12, 231254. Johansen, S. (1991). Estimation and hypothesis testing of cointegration vectors in gaussian vector autoregressive models. Econometrica 59,15111580. Karpoff, J. M. (1987). The relation between price changes and trading volume: A survey. Journal of Financial and Quantitative Analysis, 22,109126. Labys, W.C.,&Thomas H.C. (1975). Speculation, Hedging and Commodity Price Behavior: An International C omparison. Applied Economics, 7,287 301. Lamoureux, C., & Lastrpes, W. (1990). Heteroscedasticity in stock return data: Volume versus garch effects. Journal of Finance, 45,221 229. Lai,K.&Lai, M. (1991). A cointegration test for market efficiency, Journal of Futures Markets 11, 567 575 Majority and Minority staff report,(2009), Excessive Speculation in the Wheat Market. Mark J.& Lynn J.(2008). Speculation and Energy Prices: Legislative Response, Cunningham, Congressional Research Service Report RL34555. Moneta, A.(2004). Identification of Monetary Policy Shocks: A graphical causal approach Notas Econmicas Faculdade de Economia, U niversidade de Coimbra, 20,3962. Randolph, W. L., & M. Najand, (1991). A Test of Two Models in Forecasting Stock Index Future Price Volatility, Journal of Futures Markets 11,179190. Park, H., Mjelde, J.&Bessler,D.A. (2008). Price Interactions and Discove ry among Natural Gas Spot Markets in North America. Energy Policy 36, 290302.

PAGE 67

67 Peck A.E. (1981). The Adequacy of Speculation on the Wheat, Corn, and Soybean Futures Markets. Research in Domestic and International Agribusiness Management 2,17 29. Rthig,A., &Chiarella,C. (2006). Investigating Nonlinear Speculation in Cattle, Corn and Hog Futures Markets Using Logistic Smooth Transition Regression Models Research Paper Series 172, Quantitative Finance Research Center, University of Technology, Sydney. Reale, M.,& Wilson,G.T. (2001). Identification of vector AR models with recursive structural errors using conditional independence graphs. Statistical Methods and Applications 10,4965. Sanders, D.R., S.H. Irwin, and R.P. Merrin.(2008) The Adequacy of Speculation in Agricultural Futures Markets: Too Much of a Good Thing? Marketing and Outlook Research Report. Shalen, C. T. (1993). Volume, V olatility, and the Dispersion of Beliefs. Review of Financial Studies 6, 405434. Staff report. (2006). The Role of Market Speculation in Rising Oil and Gas Prices: A Need to Put the Cop Back on the Beat. Stock, J.S. &M. Watson (1988), Variable Trends in Economic Time Series. Journal of Economic Perspectives, 2( 3) Stoll, H. R., &Whaley, R. E. (1988). Volatility and futures: Message versus messenger. Journal of Portfolio Management 14,2022. Stoll, H. R., &Whaley, R. E. (1990). The dynamics of stock index and stock indexfutures returns. Journal of Financial and Quantitative Analysis 25,441468. Streeter, D.H.,&Tomek, W.G. (1992). Variability in Soybean Futures Prices: An Integrated Framework. The Journal of Future Markets 12(6),706728. Subcommittee repor t (2006), The Role of Market Speculation in Rising Oil and Gas Prices: A Need to Put the Cop Back on the Beat, S. Prt,109165 Swanson, N. R.,& Granger, C. W. J. (1997). Impulse response functions based on a causal approac h to residual orthogonalization in vector a utoregressions. Journal of the American Statistical Association, 92,357 367. Tauchen, G.E.,&Pitts, M. (1983). The price variability volume relationship on speculative markets. Econometrica, 51,485 506. The Enron Loophole(2008). Congressional Research Service

PAGE 68

68 Wang, C. (2002). The effect of net positions by trader type on volatility in foreign currency futures markets. Journal of Futures Markets 22,427 450. Wiley, M.,&Robert D.(1998). Volume relationships among types of traders in the financial futures markets, Journal of Futures Markets,8,91 114. Zhang, J. E.,& Zhu, Y. (2006). VIX futures. Journal of Futures Markets 26,521531.

PAGE 69

69 BIOGRAPHICAL SKETCH Fang Meng earned her Bachelor of Arts in financial engineering from Shandong U niversity, China, in 2008. D egree of Master of Science in food and resource economics was completed at the University of Florida during the summer of 2010. Mengs primary research interests include agriculture commodity market