Description
This paper seeks to study the impact of bio-fuel policies on oil and food futures prices from
December 6, 2004 to August 1, 2008.
Journal of Financial Economic Policy
Do bio-fuel policies lead to speculative behavior?
Chia-Hsing Huang Liang-Chun Ho
Article information:
To cite this document:
Chia-Hsing Huang Liang-Chun Ho, (2011),"Do bio-fuel policies lead to speculative behavior?", J ournal of
Financial Economic Policy, Vol. 3 Iss 2 pp. 161 - 174
Permanent link to this document:http://dx.doi.org/10.1108/17576381111133624
Downloaded on: 24 January 2016, At: 21:41 (PT)
References: this document contains references to 32 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 282 times since 2011*
Users who downloaded this article also downloaded:
Steve Swidler, (2011),"Homeownership: yesterday, today and tomorrow", J ournal of Financial Economic
Policy, Vol. 3 Iss 1 pp. 5-11http://dx.doi.org/10.1108/17576381111116786
Beverley Searle, (2011),"Recession and housing wealth", J ournal of Financial Economic Policy, Vol. 3 Iss 1
pp. 33-48http://dx.doi.org/10.1108/17576381111116867
Thomas D. Willett, (2012),"The role of defective mental models in generating the global financial crisis",
J ournal of Financial Economic Policy, Vol. 4 Iss 1 pp. 41-57http://dx.doi.org/10.1108/17576381211206479
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Do bio-fuel policies lead
to speculative behavior?
Chia-Hsing Huang
SolBridge International School of Business, Daejeon, South Korea, and
Liang-Chun Ho
Department of Finance, Hsiuping Institute of Technology, Taichung, Taiwan
Abstract
Purpose – This paper seeks to study the impact of bio-fuel policies on oil and food futures prices from
December 6, 2004 to August 1, 2008.
Design/methodology/approach – The daily closing prices of brent crude oil, light sweet crude oil,
corn, wheat, soybeans, and rough rice futures from December 6, 2004 to August 1, 2008 are used in this
research. The vector error correction model is applied in order to study the impact of bio-fuel policies on
oil and agricultural futures prices.
Findings – Unit root and cointegration tests showthat the brent crude oil, light sweet crude oil, wheat,
corn, soybeans, and rough rice futures are stationary and have a long-run equilibrium relationship.
Granger causality tests of the four periods shows that the causality relationship between oil futures and
food futures changes over time. The ?rst period result shows many Granger causes on several variables
at a 5 percent signi?cance level. The second period has more Granger causes at the 5 percent signi?cance
level. However, the Granger causalityrelationships become fewer andfewer inthe thirdandfourthperiod.
Originality/value – This is the ?rst paper to study the impact of the four major bio-fuel policies of
Brazil, the European Union, and the USA.
Keywords Fuels, Futures markets, Oils, Brazil, European union, United States of America
Paper type Research paper
1. Introduction
This research studies the impact of the four major bio-fuel policies of Brazil, the
European Union, and the USA. Based on the four bio-fuel policies, research time is
dividedinto four periods. December 6, 2004 to August 7, 2005 is the ?rst period, August 8,
2005 to March 7, 2007 is the second period, March 8, 2007 to December 17, 2007 is the
third period, and from December 18, 2007 to August 1, 2008 is the fourth period.
The major sources to make biofuels are corn, soybean, and wheat. Rice is still not used
as a raw material to make the biofuel. However, most of the countries that experienced
food shortage riots have rice as their major daily food source. Will the world’s markets
encourage bio-fuel supply to the rich countries by converting food crops to biofuel, while,
at the same time, poor countries not having a suf?cient foodsupplyandfacingstarvation?
This has become a major issue among politicians, economists, and environmentalists.
Including rice futures can help in the understanding of the relationship between the food
crisis and biofuels.
Implementing the Biodiesel Act of December 6, 2004, Brazil started to use biodiesels
which is a combination of conventional diesel and 5 percent vegetable oil ester. With this
biodiesel policy, there will be more farmers growing soybeans and sun?ower for the fuel
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – G0, O13
Bio-fuel policies
161
Journal of Financial Economic Policy
Vol. 3 No. 2, 2011
pp. 161-174
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381111133624
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
production since biodiesel can be derived from soybeans and sun?ower seeds. Ideally,
it will encourage both big and small farmers to enter this market and increase their
incomes.
In early 2007, the integrated energy and climate change package was proposed by the
EUCommission. This package was approved by the EUHeads of State in March 8, 2007
meeting of the European Council. The purpose of this package is to reduce global
warming. The goals of this package are to use 20 percent renewable energy by the year
2020, to reduce greenhouse gas emissions in 2020 by 20 percent compared with that of
1990, and with a 10 percent required minimumtarget for biofuel in 2020. It is also hoped
that all of the biofuel are produced in Europe, which will increase employment and
European GDP. The major sources of bio-fuel production in the EUare rapeseed, wheat,
and sugar beets.
There are two major bio-fuel policies inthe USA, the EnergyPolicy Act of 2005 andthe
Energy Independence and Security Act of 2007. In order to promote investment in energy
conservation and ef?ciency, President Bush signed the Energy Policy Act on August 8,
2005. This act stipulated that 4-billion gallons of renewable fuel were to be used in 2006
and increased to 7.5-billion gallons by 2012. One of the purposes of this act is to diversify
the energy supply with renewable sources. This act gives tax credits for the use of
renewable energy sources such as wind, solar, and bio-energy. It also gives tax incentives
to producers for ethanol and biodiesel research and expenses for development and
production. Normally, ethanol is mixed with gasoline in a 10 percent ethanol blend and is
called gasohol. The President of the USA signed the Energy Independence and Security
Act on December 18, 2007. The idea of this act is to take away the tax credit from the oil
and gas companies and move the money to the companies that develop and produce
alternative fuels. The expanded renewable fuels standard, with annual requirements for
the amount of renewable fuels produced and used in motor vehicles, requires 9-billion
gallons of renewable fuels in 2008 and rises to 36-billion gallons by 2022. The USA and
Brazil are the two major suppliers of ethanol to the world market. Brazil uses sugar cane
as the main source to produce ethanol. While in the USA, fuel ethanol is based mostly on
corn, switch grass, and soybeans.
The question of whether the bio-fuel policies cause higher food prices world-wide and
hunger in the developing countries has been widely debated. Based on a New York
Times report, those who support bio-fuel policies said that oil prices have increased so
much that an alternative is needed. One solution to high oil prices is biofuel. Biofuel is
onlyresponsible for 2-3 percent of the world’s foodprice increases while at the same time,
there is a reduction of crude oil demand by 1-million barrels a day. The food prices
increases are because of temporary problems such as droughts. Higher fertilizer and
crop transportation costs due to higher oil prices makes food more expensive. Countries,
like China and India, have increased incomes which enable the consumption of more
food. This is another driver for higher food prices. Critics, however, argue that a good
substitute of oil would bring the world a whole lot of bene?ts while at the same time have
only a minor impact on the world’s economy and environment. The negative impact of
biofuels is that farmers will growmore corn for biofuels at the expense of feeder corn and
other food crops. Unfortunately, increased food prices will have a greater impact on
low-income developing countries (The New York Times, May 30, 2008).
In addition to the aforementioned potential factors that affect food prices, the main
driver of higher food prices might be speculation in the ?nancial markets. When there
JFEP
3,2
162
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
is speculative behavior in ?nancial markets, the volatilities of agriculture futures and oil
futures prices may increase. The United States Commodity Futures Trading
Commission (CFTC) Chairman Chilton calls some of the non-traditional futures
speculator “massive passives”. These speculators have a long futures only strategy.
CFTC (2010a, b) proposed four different hard cap mandatory speculative position limits
on energy futures contracts. The purpose is to restrain the behavior that potentially can
distress markets.
Some farmers growagriculture products for bio-fuel production. Soybean, sun?ower,
and sugarcane are the main sources to make biofuel in Brazil. In the USA, farmers grow
corn, switchgrass, and soybean for biofuels. Rapeseed, wheat, and sugar beet are the
major inputs for biofuels in the EU. The Biodiesel Act of Brazil, Energy Policy Act of the
USA, integrated energy and climate change packages of EU, and the Energy
Independence and Security Act of the USA are the world’s four most important bio-fuel
policies. Brazil, the USA, and the EU are the major bio-fuel producers in the world.
Implementing these policies will encourage farmers to produce agriculture products for
biofuel and less for consumer food.
Empirical researches show that futures contracts provide price discovery function,
which is to use futures price as an indication of spot prices. Garbade and Silber (1983)
introduced a simultaneous dynamic price model to show the price discovery function of
futures. Most of the researches usedvector error correction(VEC) model to studythe price
discovery function of futures contracts, i.e. the lead-lag relationship between spot prices
and futures prices. Bopp and Sitzer (1987) showed that heating oil futures price is a good
predictor of spot prices. Moosa (2002) found that the daily crude oil futures contract
performs 60 percent of the price discovery function. Futures prices are determined by the
market trading andmayre?ect the expected future spot price of the participants. The two
groups of futures market speculators can be demarcated as the leading speculators and
the trend followers. The trend followers are the massive passives that do not undertake
intensive researchbefore makingfutures tradingdecisions. The leadingspeculators work
on fundamental and technical analysis in order to make trading decisions. New bio-fuel
policy could be one of the factors that the leading speculators may have studied and this
can in?uence the expected future spot prices, which in turn will affect the commodity
futures contracts prices.
Both theoretical and empirical researches demonstrate the existence of herd behavior
in the ?nancial markets (Froot et al., 1992; Hirshleifer et al., 1994; Barberis et al., 2005;
GreenandHwang, 2009). The prices inthe ?nancial markets cannot be fullyexplainedby
economic fundamental alone. Herdingbehavior is one of the factors that canin?uence the
prices. Investors are in?uenced by other people when making food futures transactions.
Under the in?uence of the herding behavior, investors even reverse their transaction
decision when observing the trends in the futures market. When most of the traders have
the same decision to have long position of a futures contract, the futures price will be
affected by this herding behavior. For example, if the leading speculator predicts that the
food prices will increase and buy the food futures, the followers will also buy the food
futures. The herding behavior phenomenon intensi?es in the bull market and diminishes
in the bear market (Bowe and Domuta, 2004).
The relationships of spot prices and futures prices also provide important information
for futures trading speculators. Studies all showthe gradually increasing relationships of
oil spot prices and oil futures prices in the world markets (Gulen, 1999; Sadorsky, 2000;
Bio-fuel policies
163
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hammoudeh et al., 2003). Food spot price transmission in the international markets is
studied by Balcombe et al. (2007). Balcombe and Rapsomanikis (2008) showed the
relationships of Brazilian sugar, oil, and ethanol prices. Tang and Xiong (2010) found that
the correlations of commodity returns and the world equity index, and the correlations of
commodityreturns andthe oil returnhave increasedsince the early2000s. Duringthe 2008
?nancial crisis, the spillover effects made a signi?cant contribution to the increased
commodity price volatility. These spillover effects are the results of both demand and
supply of physical commodities and fundamental ?nancialization processes of the
commodities markets.
Commodity price speculation, demand, limited supply, OPEC monopoly pricing, and
scarcity rent are the factors that may have caused 2008 high oil prices in the study of
Hamilton (2009). The results show that the low price elasticity of demand, strong world
demand, and limited supply are the three main reasons to have high oil prices in the
beginning. These fundamental reasons trigger speculative behavior. And the scarcity
rent may become an important factor in the future. Khan (2009) analyzed many
fundamental factors on the oil prices from 2003 and showed that in addition to the
fundamentals, speculation precipitatedanoil price bubble inthe ?rst half of 2008. If there
was no speculation, the oil prices would probably have been in the range of $80-$90 a
barrel. Medlock and Jaffe (2009) showed that the open interest of noncommercial players
in the oil futures market becomes a leading indicator of prices since January 2006. During
the time when speculators have net long positions and the market price moves upward.
If the following argument is true, then it may be deduced that the use of biofuel is one
contributor to increasing food prices. Only, if supply were (totally) inelastic and did not
respond to higher prices, the demand for biofuel would increase when the oil price
increases. Increasing demand for bio-fuel fosters the demand for bio-fuel sources, such
as corn, soybean, and wheat. Prices of corn, soybean, and wheat will soar when demand
for corn, soybean, and wheat intensi?es. When at the same time, the supply of corn,
soybean, and wheat for food decreases. Decreased supply and higher price of food may
cause unrest in the low-income developing countries. If this is not the case, then it may be
extrapolated that biofuel is not a major source of higher food prices.
Futures prices re?ect expectations of the future supply and demand of a commodity.
A supply or demand shock from a bio-fuel policy will affect food prices in the future.
If investors feel that a variable, e.g. the price of oil futures can be a predictor of another
variable, e.g. wheat futures prices and speculate on the futures market, then futures
prices will be in?uenced by this speculative behavior. If investors believe that when
wheat prices increase, because of the bio-fuel policy, corn prices will also increase; they
may speculate on wheat and corn futures. Two variables can be examed, e.g. wheat
futures price and corn futures price, to see whether there is any causality between them.
If bio-fuel policies have a long-term effect on commodity prices, then it will have an
impact on the commodity futures prices fromperiod one to period four in this study. If an
investor believes that the government’s bio-fuel policy will increase the demand of a
commodity, e.g. wheat, and speculate on the commodity futures, wheat futures, and the
substitute commodities futures, corn futures, then we can ?nd some evidence of this
speculative behavior in the futures market. However, if bio-fuel policies only increase
food prices a little in the short run and does not increase food prices in the long run, and
also if most investors believe that the futures market is not a good indication of future
JFEP
3,2
164
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
commodity price, then the speculative behavior will only happen in the short-term, not
from period one to period four in this study.
2. Data and research method
2.1 Data
The daily closing prices of brent crude oil, light sweet crude oil, corn, wheat, soybeans,
and rough rice futures of Chicago Mercantile Exchange Group fromDecember 6, 2004 to
August 1, 2008 are used in this research. Daily futures closing prices are taken fromthe
Data Streamdatabase. The natural log of the daily futures closing prices are used in the
analysis.
This paper will analyze the impact of the four major bio-fuel policies of Brazil, the
European Union, and America. Research time is divided into four periods. December 6,
2004 to August 7, 2005 is the ?rst period; August 8, 2005 to March 7, 2007 is the second
period; March 8, 2007 to December 17, 2007 is the third period; and from December 18,
2007 to August 1, 2008 is the fourth period. The sample sizes of the four periods are 168,
397, 197, and 157, respectively. The futures price series are shown in Figure 1.
2.2 Research method
2.2.1 Long-run equilibrium relationship. When data are close to the unit root,
Augmented Dickey-Fuller (ADF) test may not be able to reject the null hypothesis (Sims,
1988). In addition to using the ADF test for the null hypothesis test, the Phillips-Perron
(PP) test, Ng-Perron (NP) test, and Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test for
the unit root tests are also used. The null hypothesis asserts that there exists a unit root
in the ADF, PP, and NP tests. Unlike the null hypothesis of the other common unit root
tests, the null hypothesis of KPSS test is that the series is stationary. The KPSS test can
be used under the null of either trend stationarity or level stationarity. Therefore,
the KPSS can be used to test the possibility of fractionally integrated series as a
complementary method to the ADF, PP, and NP tests (Soytas and Sari, 2007).
Two or more time series are said to be cointegrated when the relationship between
themare non-stationarybut a linear combinationof themis stationary. The cointegrating
vector can be tested in order to see whether there is a statistically signi?cant connection
between the two time series. If there is a low order of integration of this cointegrating
vector, they are cointegrated. The Johansen cointegration test ( Johansen, 1991, 1992;
Figure 1.
Time series of futures
prices
2004/12/6 2005/8/8 2007/3/8
Rice
Soyabeans
Wheat
Corn
Light crude oil
Brent crude oil
2007/12/18 08/2008
Bio-fuel policies
165
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Johansen and Juselius, 1990) is used to examine the cointegration for more
than two variables. The Johansen cointegration test uses maximum eigenvalue and
trace in order to investigate the number of cointegrating vectors. Higher number of
cointegrating vector represents a long-run equilibrium relationship (Hammoudeh and
Li, 2004).
2.2.2 Short-run relationships. The concept of exogeneity is discussed by Engle et al.
(1983). Conway et al. (1984) criticized the methodology of the causality test literature.
Ericsson et al. (1998) discussed the concepts of exogeneity, cointegraion, and causality
and their applications to the economic policy analysis. Based on the premise that the
effect cannot happen before the cause, Granger (1969) de?ned Granger causality in
terms of predictability. An information set, X Granger causes Y if the past values
of X and Y can be used to predict Y more accurately than only using the past values of
Y. Granger causality tests can be used in a single equation with two variables and their
lags, multivariate causality tests with more variables, and Granger causality tests with a
vector autoregression (VAR) and VEC.
The VEC model by Johansen (1988) is used to explore the short-run relationships
among the variables (Granger, 1988). If the variables are stationary, the VAR model
introduced by Sims (1980) to analyze the short-run relationships among variables
can be applied. However, under the non-stationary variable condition, the VEC model
is used to study the short-run relationships among variables. The VEC model is
constructed thus:
C
t
¼ X
t21
· u þ
X
s
i51
f
i
· C
t2i
þj
t
Aset of six endogenous variables, percentage changes of rice futures price, corn futures
price, wheat futures price, soybean futures price, light oil price, and Brent oil futures
price, are collected in a 6 £ 1 vector C
t
. X is the error correction matrix. j
t
is the white
noise vector. This model means that the changes of endogenous variables are affected by
past changes and the adjustment to the long-term equilibrium.
3. Empirical result
3.1 Unit root test
Most of the economic and ?nancial time series are not stationary and need to be tested to
ascertain if they are stationary. If the data are not stationary, then a spurious regression
result may occur. ADF, PP, NP, and KPSS unit root tests are used on these six series in
different time periods. All of these six series are not stationary in level. The null
hypothesis of a unit root could not be rejected in the ADF, PP, and NP tests, and the null
hypothesis of stationaritywas rejectedinthe KPSS test. Table I shows that, after the ?rst
difference, all of the series are signi?cant at a 5 percent signi?cant level on the ADF, PP,
and NP tests, and not signi?cant on the KPSS test. All of these series are stationary after
the ?rst difference.
3.2 Cointegration tests
To study whether the variables move together, cointegration is used to simultaneously
model long-runpersistence andcomovement. The likelihood ratio trace statistics and the
maximum eigenvalue statistics for the cointegration test is used for evaluation. Both
the maximum eigenvalue and trace tests show the same result as shown in Table II.
JFEP
3,2
166
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
B
r
e
n
t
c
r
u
d
e
o
i
l
L
i
g
h
t
c
r
u
d
e
o
i
l
C
o
r
n
W
h
e
a
t
S
o
y
b
e
a
n
R
i
c
e
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
I
n
t
e
r
c
e
p
t
A
D
F
2
6
.
0
5
4
a
2
6
.
2
2
4
a
2
1
1
.
4
2
0
a
2
1
2
.
9
4
3
a
2
1
2
.
4
0
6
a
2
1
1
.
3
8
8
a
P
P
2
1
3
.
8
4
0
a
2
1
3
.
5
9
5
a
2
1
1
.
4
0
7
a
2
1
2
.
9
4
2
a
2
1
2
.
4
2
3
a
2
1
1
.
3
8
8
a
N
P
(
M
Z
a
)
2
5
0
.
2
9
3
a
2
2
3
.
8
1
1
a
2
3
0
.
2
1
4
a
2
1
2
.
8
1
6
a
2
4
0
.
8
0
3
a
2
3
3
.
2
7
4
a
K
P
S
S
0
.
0
6
2
0
.
0
4
7
0
.
0
6
8
0
.
0
5
7
0
.
0
8
2
0
.
0
8
5
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
6
.
0
8
6
a
2
6
.
1
9
3
a
2
1
1
.
3
8
1
a
2
1
2
.
9
1
6
a
2
1
2
.
3
9
6
a
2
1
1
.
3
5
0
a
P
P
2
1
3
.
8
3
4
a
2
1
3
.
5
5
7
a
2
1
1
.
3
5
6
a
2
1
2
.
9
1
5
a
2
1
2
.
4
1
1
a
2
1
1
.
3
5
0
a
N
P
(
M
Z
a
)
2
9
8
.
0
4
2
a
2
4
3
.
7
5
0
a
2
5
5
.
5
9
4
a
2
5
0
.
7
5
1
a
2
5
3
.
4
8
4
a
2
5
9
.
7
4
9
a
K
P
S
S
0
.
0
5
1
0
.
0
4
8
0
.
0
6
5
0
.
0
5
0
0
.
0
6
1
0
.
0
8
3
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
I
n
t
e
r
c
e
p
t
A
D
F
2
2
1
.
2
8
5
a
2
2
0
.
9
5
1
a
2
1
9
.
3
8
4
a
2
1
9
.
4
3
5
a
2
1
9
.
9
7
6
a
2
1
6
.
0
0
3
a
P
P
2
2
1
.
2
8
1
a
2
2
1
.
0
4
1
a
2
1
9
.
3
8
1
a
2
1
9
.
4
4
1
a
2
2
0
.
0
0
1
a
2
1
9
.
4
3
3
a
N
P
(
M
Z
a
)
2
3
2
.
7
9
8
a
2
4
3
.
1
2
7
a
2
1
2
2
.
5
0
1
a
2
6
8
.
0
6
3
a
2
4
5
.
0
2
6
a
2
1
6
1
.
6
0
9
a
K
P
S
S
0
.
0
7
4
0
.
0
6
7
0
.
3
4
6
0
.
0
4
1
0
.
4
3
2
0
.
0
2
3
2
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
2
1
.
2
5
9
a
2
2
0
.
9
2
7
a
2
1
9
.
4
9
6
a
2
1
9
.
4
1
1
a
2
2
0
.
1
3
7
a
2
1
5
.
9
8
4
a
P
P
2
2
1
.
2
5
6
a
2
2
1
.
0
1
6
a
2
1
9
.
4
9
8
a
2
1
9
.
4
1
8
a
2
2
0
.
1
9
4
a
2
1
9
.
4
0
7
a
N
P
(
M
Z
a
)
2
3
9
.
6
4
7
a
2
5
3
.
8
9
3
a
2
1
7
7
.
5
0
5
a
2
7
8
.
2
7
5
a
2
1
1
2
.
8
0
3
a
2
1
8
6
.
7
9
5
a
K
P
S
S
0
.
0
6
8
0
.
0
5
9
0
.
0
4
3
0
.
0
3
7
0
.
0
4
5
0
.
0
1
7
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
I
n
t
e
r
c
e
p
t
A
D
F
2
1
5
.
3
6
1
a
2
1
5
.
1
0
6
a
2
1
3
.
6
3
8
a
2
6
.
0
8
5
a
2
1
4
.
2
3
0
a
2
1
2
.
9
3
9
a
P
P
2
1
5
.
3
5
6
a
2
1
5
.
0
7
4
a
2
1
3
.
6
3
6
a
2
1
2
.
5
5
0
a
2
1
4
.
2
2
9
a
2
1
2
.
9
0
3
a
N
P
(
M
Z
a
)
2
3
6
.
0
9
8
a
2
3
0
.
4
8
5
a
2
7
3
.
7
2
8
a
2
9
.
4
4
9
a
2
6
4
.
4
7
7
a
2
8
8
.
6
4
8
a
K
P
S
S
0
.
0
4
3
0
.
0
4
5
0
.
3
4
3
0
.
1
0
0
0
.
1
5
3
0
.
3
1
5
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
1
5
.
3
1
8
a
2
1
5
.
0
6
5
a
2
1
3
.
8
3
2
a
2
6
.
0
7
6
a
2
1
4
.
2
9
2
a
2
1
3
.
0
8
7
a
P
P
2
1
5
.
3
1
4
a
2
1
5
.
0
3
4
a
2
1
3
.
8
3
2
a
2
1
2
.
5
3
1
a
2
1
4
.
2
9
7
a
2
1
3
.
1
2
3
a
N
P
(
M
Z
a
)
2
5
9
.
7
3
6
a
2
5
4
.
9
1
6
a
2
9
6
.
7
2
0
a
2
8
.
9
3
6
a
2
6
8
.
8
8
9
a
2
9
5
.
5
7
2
a
K
P
S
S
0
.
0
3
9
0
.
0
4
2
0
.
0
3
4
3
0
.
0
9
3
0
.
0
3
0
0
.
0
3
5
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
I
n
t
e
r
c
e
p
t
A
D
F
2
1
3
.
2
3
7
a
2
9
.
8
0
0
a
2
5
.
9
2
9
a
2
9
.
9
6
6
a
2
1
1
.
3
7
8
a
2
1
1
.
2
4
8
a
P
P
2
1
3
.
2
3
9
a
2
1
2
.
4
3
7
a
2
1
0
.
4
4
8
a
2
1
0
.
7
0
7
a
2
1
1
.
3
4
4
a
2
1
1
.
2
8
5
a
N
P
(
M
Z
a
)
2
6
4
.
2
9
4
a
2
7
4
.
0
4
0
a
2
9
.
5
2
9
a
2
2
.
4
3
3
2
1
1
.
8
1
6
a
2
9
.
2
3
0
a
K
P
S
S
0
.
1
1
8
0
.
1
1
4
0
.
3
1
5
0
.
1
0
6
0
.
1
5
3
0
.
7
1
8
a
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
1
3
.
2
2
3
a
2
9
.
8
1
2
a
2
6
.
1
8
2
a
2
9
.
9
4
6
a
2
1
1
.
4
2
7
a
2
1
1
.
6
7
5
a
P
P
2
1
3
.
2
2
5
a
2
1
2
.
4
2
8
a
2
1
0
.
6
2
2
a
2
1
0
.
6
7
4
a
2
1
1
.
3
9
4
a
2
1
1
.
6
5
9
a
N
P
(
M
Z
a
)
2
7
4
.
1
3
1
a
2
7
6
.
9
9
2
a
2
1
7
.
5
9
4
a
2
7
0
.
6
1
9
a
2
1
3
.
6
5
4
2
7
4
.
9
0
5
a
K
P
S
S
0
.
0
8
9
0
.
0
7
8
0
.
0
8
5
0
.
0
8
7
0
.
0
7
4
0
.
0
7
5
N
o
t
e
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
n
u
l
l
h
y
p
o
t
h
e
s
i
s
(
H
0
)
a
t
t
h
e
0
.
0
5
l
e
v
e
l
Table I.
First difference unit root
test results
Bio-fuel policies
167
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
E
i
g
e
n
v
a
l
u
e
T
r
a
c
e
M
a
x
-
E
i
g
e
n
E
i
g
e
n
v
a
l
u
e
T
r
a
c
e
M
a
x
-
E
i
g
e
n
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
N
o
n
e
a
0
.
6
3
2
5
2
4
.
4
5
7
1
6
5
.
7
7
3
0
.
4
4
3
1
,
0
6
6
.
4
6
5
2
3
1
.
3
7
1
A
t
m
o
s
t
1
a
0
.
4
6
9
3
5
8
.
6
8
4
1
0
5
.
1
5
3
0
.
3
9
9
8
3
5
.
0
9
4
2
0
0
.
8
7
7
A
t
m
o
s
t
2
a
0
.
3
8
6
2
5
3
.
5
3
1
8
0
.
9
1
0
0
.
3
8
1
6
3
4
.
2
1
7
1
8
9
.
2
4
6
A
t
m
o
s
t
3
a
0
.
3
5
2
1
7
2
.
6
2
1
7
1
.
9
4
0
0
.
3
5
0
4
4
4
.
9
7
1
1
7
0
.
2
5
6
A
t
m
o
s
t
4
a
0
.
2
8
8
1
0
0
.
6
8
1
5
6
.
3
2
8
0
.
3
0
3
2
7
4
.
7
1
5
1
4
2
.
3
0
1
A
t
m
o
s
t
5
a
0
.
2
3
4
4
4
.
3
5
2
4
4
.
3
5
2
0
.
2
8
5
1
3
2
.
4
1
4
1
3
2
.
4
1
4
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
N
o
n
e
a
0
.
3
9
2
4
7
3
.
7
0
2
9
6
.
9
9
7
0
.
4
3
6
3
9
6
.
3
8
5
8
8
.
8
3
0
A
t
m
o
s
t
1
a
0
.
3
8
5
3
7
6
.
7
0
5
9
4
.
7
5
1
0
.
3
8
0
3
0
7
.
5
5
5
7
4
.
1
9
0
A
t
m
o
s
t
2
a
0
.
3
5
5
2
8
1
.
9
5
4
8
5
.
5
6
7
0
.
3
7
8
2
3
3
.
3
6
5
7
3
.
6
4
2
A
t
m
o
s
t
3
a
0
.
2
9
7
1
9
6
.
3
8
6
6
8
.
5
9
3
0
.
3
1
5
1
5
9
.
7
2
4
5
8
.
5
8
4
A
t
m
o
s
t
4
a
0
.
2
8
3
1
2
7
.
7
9
3
6
4
.
9
8
7
0
.
3
0
0
1
0
1
.
1
3
9
5
5
.
3
7
5
A
t
m
o
s
t
5
a
0
.
2
7
5
6
2
.
8
0
6
6
2
.
8
0
6
0
.
2
5
6
4
5
.
7
6
5
4
5
.
7
6
5
N
o
t
e
s
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
0
5
l
e
v
e
l
;
H
0
–
n
o
.
o
f
C
E
(
s
)
;
t
h
e
?
r
s
t
d
i
f
f
e
r
e
n
c
e
o
f
v
a
r
i
a
n
c
e
s
i
s
u
s
e
d
i
n
J
o
h
a
n
s
e
n
c
o
i
n
t
e
g
r
a
t
i
o
n
t
e
s
t
Table II.
Johansen cointegration
test results
JFEP
3,2
168
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
All of the tests are signi?cant and reject the null hypothesis, which means
all of the variables have cointegrating vectors. Under the 5 percent signi?cant level,
there are ?ve cointegration vectors both from the maximum eigenvalue and trace tests.
Brent oil futures, light oil futures, corn futures, wheat futures, soybeans futures, and rice
futures all have signi?cant long-run comovement relationships. This result shows that
one variable can be used to predict movement of another variable.
3.3 Granger causality test
To study whether one variable is a leading indicator of another variable, the Granger
causality test is used in order to explore the causalities among the variables. Table III
shows the Granger causality test results of the four periods.
FromDecember 6, 2004 to August 8, 2005, brent crude oil futures Granger causes rice
futures. Light crude oil futures Granger causes wheat futures. Corn futures Granger
causes rice futures. Soybean futures Granger causes brent crude oil futures and light
crude oil futures. Wheat futures Granger causes corn futures.
The result indicates that after implementingthe Biodiesel Act inBrazil inDecember 6,
2004, the change of the soybeans futures price is a leading indicator of the change of
brent crude oil futures price and light crude oil futures price. Soybean is the major source
to produce biodiesel. Investors in the futures market expect that the price changes of
soybeans futures will lead the price changes of the oil futures. The investors in the
futures market seem to perceive this is a high food price signal sent by the Biodiesel
policy. In addition to speculation on the situation that the soybean futures Granger
causes brent crude oil futures and light crude oil futures, investors also speculate on
many Granger causality relationships, such as the brent crude oil futures to the rice
futures, light crude oil futures to the wheat futures, corn futures to the rice futures, and
wheat futures to the corn futures. If most of these Granger causality relationships persist
in the following periods, we may argue that the bio-fuel policy could be one of the factors
for higher food prices.
The Granger causality test results from August 8, 2005 to March 7, 2007 show that
brent crude oil futures and rice futures have two-way feedback relationships. Light
crude oil futures and rice futures also have two-way feedback relationships. Wheat
futures Granger causes corn futures, soybeans futures, and rice futures.
Corn, switch grass, and soybean are the major sources for bio-fuel production in the
USA. However, the investors in the futures market do not use this chance to speculate on
the Granger causality relationships of corn futures on oil futures, soybeans futures on oil
futures, oil futures on corn futures, and oil futures on soybeans futures. Investors
speculate on the Granger causality relationships of Brent oil futures on rice futures,
light oil futures on rice futures, wheat futures on corn futures, soybeans futures, and
rice futures. The result shows that the investors in the futures market expect that the
implementation of the Energy Policy Act on August 8, 2005 in the USA will encourage
more farmers to grow the corn and soybeans for bio-fuel production. And some of the
new corn or soybean farmers may have previously grown wheat or rice. The change in
wheat futures price then can be a leading indicator of the prices of corn futures, soybeans
futures, and rice futures.
The Granger causality relationships of soybeans futures on brent oil futures and on
light oil futures in the ?rst period do not continue to exist in the second period. The corn
futures Granger causes the rice futures in the ?rst period, but corn futures does
Bio-fuel policies
169
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
D
e
p
e
n
d
e
n
t
v
a
r
i
a
b
l
e
B
r
e
n
t
c
r
u
d
e
o
i
l
L
i
g
h
t
c
r
u
d
e
o
i
l
C
o
r
n
S
o
y
b
e
a
n
W
h
e
a
t
R
i
c
e
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
2
0
.
1
7
4
0
.
8
0
1
0
.
2
6
6
0
.
1
0
5
0
.
0
6
7
a
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
1
1
6
–
0
.
8
3
2
0
.
1
0
2
0
.
0
3
9
b
0
.
1
2
4
C
o
r
n
0
.
5
7
7
0
.
5
9
7
–
0
.
6
0
5
0
.
1
5
5
0
.
0
3
8
b
S
o
y
b
e
a
n
0
.
0
1
6
b
0
.
0
2
9
b
0
.
4
6
2
–
0
.
3
8
0
0
.
8
2
7
W
h
e
a
t
0
.
3
9
5
0
.
2
4
0
0
.
0
8
0
a
0
.
1
7
6
–
0
.
4
9
7
R
i
c
e
0
.
4
7
8
0
.
7
0
6
0
.
2
4
3
0
.
3
4
1
0
.
5
6
2
–
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
0
0
1
b
0
.
9
2
1
0
.
4
4
6
0
.
5
7
0
0
.
0
3
7
b
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
0
0
7
b
–
0
.
5
1
9
0
.
7
4
7
0
.
7
8
6
0
.
0
1
8
b
C
o
r
n
0
.
3
2
0
0
.
3
3
9
–
0
.
3
3
4
0
.
2
2
0
0
.
4
9
6
S
o
y
b
e
a
n
0
.
4
3
5
0
.
5
0
1
2
0
.
9
4
0
–
0
.
3
2
9
0
.
9
8
8
W
h
e
a
t
0
.
9
4
8
0
.
9
4
1
0
.
0
3
1
a
0
.
0
2
5
b
–
0
.
0
8
5
a
R
i
c
e
0
.
0
0
9
b
0
.
0
5
6
a
0
.
6
6
9
0
.
1
8
9
0
.
9
5
6
–
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
1
1
3
0
.
1
3
1
0
.
7
1
3
0
.
2
5
9
0
.
1
2
6
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
6
0
1
–
0
.
5
1
5
0
.
4
7
6
0
.
5
6
1
0
.
8
4
0
C
o
r
n
0
.
8
4
7
0
.
9
8
5
–
0
.
4
5
9
0
.
0
0
8
b
0
.
5
2
3
S
o
y
b
e
a
n
0
.
2
5
2
0
.
2
6
1
0
.
2
1
0
–
0
.
0
4
3
b
0
.
4
2
4
W
h
e
a
t
0
.
1
8
9
0
.
3
3
8
0
.
1
1
8
0
.
0
6
4
a
–
0
.
2
9
1
R
i
c
e
0
.
9
4
4
0
.
9
8
1
0
.
8
0
4
0
.
5
4
7
0
.
4
9
0
–
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
0
4
2
b
0
.
9
2
5
0
.
8
9
9
0
.
5
7
8
0
.
6
4
0
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
1
8
1
–
0
.
6
0
2
0
.
3
9
1
0
.
2
7
3
0
.
6
6
2
C
o
r
n
0
.
5
5
8
0
.
7
8
1
–
0
.
1
2
7
0
.
4
8
7
0
.
0
6
6
a
S
o
y
b
e
a
n
0
.
4
2
8
0
.
2
4
1
0
.
8
9
6
–
0
.
5
0
9
0
.
8
0
5
W
h
e
a
t
0
.
1
3
2
0
.
1
9
1
0
.
3
5
2
0
.
9
5
1
–
0
.
6
1
6
R
i
c
e
0
.
8
0
1
0
.
9
3
0
0
.
7
4
1
0
.
5
6
5
0
.
2
4
9
–
N
o
t
e
s
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
1
l
e
v
e
l
;
b
d
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
0
5
l
e
v
e
l
Table III.
VEC Granger
causality/block
exogeneity Wald tests
JFEP
3,2
170
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
not Granger cause rice futures in the second. The result shows that there exists
speculative behavior in the corn futures, soybeans futures, and rice futures markets and
the bio-fuel policies do not have a permanent effect on the oil futures and food futures
price relationships.
The Granger causality test results from March 8, 2007 to December 17, 2007 show
that wheat futures and soybeans futures have two-way feedback relationships. Corn
futures Granger causes wheat futures. In Europe, rapeseed, wheat, and sugar beets are
the major sources of the production of biofuel. The results showthat the implementation
of the integrated energy and climate change package, on March 8, 2007 in the EU, lead
futures investors to believe that this bio-fuel policy may attract more farmers to grow
wheat for bio-fuel production. Farmers that used to grow other crops may switch to
growing wheat. Achange in corn futures price and soybeans futures price then can have
a signi?cant impact on wheat futures price.
During the second period, there are nine signi?cant Granger causalitypairs. It reduces
to only three signi?cant Granger causality pairs in the third period. Most of the Granger
causality relationships in the previous periods cease to exist in the third period. The
bio-fuel policies only have a temporary effect on food futures and oil futures Granger
causality relationships. In the third period, investors only speculate on the Granger
causality relationships of corn futures on wheat futures and the two-way feedback
relationship of soybeans futures and wheat futures.
The Granger causality test results fromDecember 18, 2007 to August 1, 2008 showthat
brent crude oil futures Granger causes light oil futures. Corn futures Granger causes rice
futures. There are only two signi?cant Granger causality pairs in the fourth period. Some
of the Granger causality relationships during the early periods do not exist in the fourth
period. It seems most of the speculative behavior inthe futures market disappears infourth
periods of this research. The Energy Independence and Security Act of December 18, 2007
in the USA did not cause most of the variables to have signi?cant Granger causality
relationships. The decrease in speculative behavior may be due to the world’s ?nancial
problems originating fromthe subprime crisis in the USA. In July 2007, the default rate on
subprime loans was too high which resulted in some of the mortgage-backed securities
issuers having their credit ratings downgraded. For example, Bear Stearns, one of the
market’s big players, suffering from subprime loans problems closed two funds with
important subprime investments in July 2007. The subprime loan crisis spread to other
parts of the world. UBS of Switzerland announced a loss of $10 billion in the US subprime
market. We can see most of the Granger causality relationships are not consistent during
the four periods. If biofuel is the major driver for higher food price, these Granger causality
relationships would have existed in the long term.
4. Conclusion
Since December 6, 2004, Brazil’s Government have implemented the Biodiesel policy.
The USA and the EU followed with implementation of the Energy Policy Act, the
Integrated Energy and Climate Change Package, and ?nally, the Energy Independence
and Security Act of August 2005, March 2007, and December 2007, respectively. Bio-fuel
policies are intended to reduce pressure from high oil prices and increase farmers’
income from producing crops for biofuel. However, at the same time, food prices also
increase and some countries experience food shortages as well.
Bio-fuel policies
171
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The VEC model is used to study the impact of bio-fuel policies on oil and agricultural
futures prices. If the main driver of higher food prices is bio-fuel policy, oil futures prices
should have a consistently strong impact on agricultural futures prices. If this is not the
case, then the inconsistent result of the futures prices’ relationship in every period is due
to market speculation.
Unit root and cointegration tests show that the brent oil futures, light oil futures,
wheat futures, corn futures, soybeans futures, and rice futures are non-stationary in
levels and stationary once they are ?rst differenced, cointegrated, and have a long-run
equilibrium relationship. Granger causality tests of the four period shows that the
causality relationship between oil futures and food futures changes over time. The ?rst
periodresult shows manyGranger causes onseveral variables at the 5 percent signi?cant
level. Light oil futures Granger causes wheat futures. Corn futures Granger causes rice
futures. Soybeans futures Granger causes brent oil futures and light oil futures.
The second period has more Granger causes at the 5 percent signi?cant level. Brent
oil futures and light oil futures have two-way feedback relationships. Brent oil futures
and rice futures have two-way feedback relationships. Light oil futures Granger causes
rice futures. Wheat futures Granger causes soybeans futures. However, the Granger
causality relationships become progressively scarcer in the third and fourth period.
In the third period, only corn futures Granger causes wheat and soybeans futures
Granger causes wheat futures at the 5 percent signi?cant level. Only one 5 percent
signi?cant Granger causality is left in the fourth period. In the fourth period, brent oil
futures Granger causes light oil futures.
The fundamental market demand and supply are the major forces for the high oil and
food prices. Excess demand of the world oil and food and not enough supply cause the
expectation of higher oil and food prices. However, the oil and food spot prices and futures
prices cannot be explained by the economic fundamental alone. In addition to the
fundamental demand and supply, bio-fuel policy, speculation and investor herding can
alsoaffect oil andfoodspot prices andfutures prices. Higher expectedoil andfoodprices in
the market drawthe attention of futures speculators. Speculators predict the oil and food
spot prices will increase andspeculate onhigher oil andfoodfutures prices. Inaddition, the
speculators use one oil/food futures price change to predict another oil/food futures price
change. Therefore, an oil/food futures price is used to be the predictor variable of another
oil/food futures price. Investors in the futures markets herd together. Herding behavior of
followers in the oil and food futures markets intensi?es the causalities of oil and food
futures prices. In the early time periods, followers have a passive long strategy and drive
the oil and food futures prices upwards. This phenomenon is more serious in the ?rst
period and gradually fades away in the following periods.
If it is the bio-fuel policies that cause higher oil and food prices, there should be more
and more Granger causality relationships in these four periods. No evidence is found to
show that the bio-fuel policies cause higher oil and food prices. It is possible that the
investor speculation behavior in the market drives oil futures and food futures prices in
the short-term. The speculations and herding behavior only increase the numbers of
Granger causalities in the ?rst time period in this study. Similar to the result of investor
herding behavior literatures, this research shows that the oil and food futures have more
Granger causalities in the bull market than in the bear market. The speculations
and herdingbehavior only increase the oil and food futures prices inthe short-termnot in
all of the four time periods.
JFEP
3,2
172
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
References
Balcombe, K. and Rapsomanikis, G. (2008), “Bayesian estimation and selection of nonlinear
vector error correction models: the case of the sugar-ethanol-oil nexus in Brazil”, American
Journal of Agricultural Economics, Vol. 90, pp. 658-68.
Balcombe, K., Bailey, A. and Brooks, J. (2007), “Threshold effects in price transmission: the case
of Brazilian wheat, maize, and soya prices”, American Journal of Agricultural Economics,
Vol. 89, pp. 308-23.
Barberis, N., Shleifer, A. and Wurgler, J. (2005), “Comovement”, Journal of Financial Economics,
Vol. 75, pp. 283-317.
Bopp, A. and Sitzer, S. (1987), “Are petroleum futures prices good predictors of cash value?”,
Journal of Futures Markets, Vol. 7, pp. 705-19.
Bowe, M. and Domuta, D. (2004), “Investor herding during ?nancial crisis: a clinical study of the
Jakarta stock exchange”, Paci?c-Basin Finance Journal, Vol. 12, pp. 387-418.
CFTC (2010a), Statement by Chairman Chilton at the Electric Power Supply Association Annual
Meeting, January 30, 2010, Commodity Futures Trading Commission, available at: www.
cftc.gov
CFTC (2010b), “Statement by Chairman Chilton regarding the notice of proposed rulemaking for
speculative position limits for referenced energy contract, January 14, 2010”, Commodity
Futures Trading Commission, available at: www.cftc.gov
Conway, R., Swamy, P., Yanagida, J. and von zur Muehlen, P. (1984), “The impossibility of
causality testing”, Agricultural Economics Research, Vol. 36, pp. 1-19.
Engle, R., Hendry, D. and Richard, J. (1983), “Exogeneity”, Econometrica, Vol. 51, pp. 277-304.
Ericsson, N., Hendry, D. and Mizon, G. (1998), “Exogeneity, cointrgration, and economic policy
analysis”, Journal of Business and Economic Statistics, Vol. 16, pp. 370-87.
Froot, K., Scharfstein, D. and Stein, J. (1992), “Herd on the street: informational inef?ciencies in a
market with short-term speculation”, Journal of Finance, Vol. 47, pp. 1461-84.
Garbade, K. and Silber, W. (1983), “Price movement and price discovery in futures and cash
markets”, Review of Economics and Statistics, Vol. 65, pp. 289-97.
Granger, C. (1969), “Investigating causal relations by economic models and cross spectral
methods”, Econometrica, Vol. 37, pp. 424-38.
Granger, C. (1988), “Causality, cointegration and control”, Journal of Economics and Control,
Vol. 12, pp. 551-9.
Green, C. and Hwang, B. (2009), “Price-based return comovement”, Journal of Financial
Economics, Vol. 93, pp. 37-50.
Gulen, S. (1999), “Regionalization in world crude oil markets: further evidence”, Energy Journal,
Vol. 20, pp. 125-39.
Hamilton, J. (2009), “Understanding crude oil prices”, The Energy Journal, Vol. 30, pp. 179-206.
Hammoudeh, S. and Li, H. (2004), “The impact of the Asian crisis on the behavior of US and
international petroleum prices”, Energy Economics, Vol. 26, pp. 135-60.
Hammoudeh, S., Li, H. and Jeon, B. (2003), “Causality and volatility spillover among petroleum
price of WTI, gasoline and heating oil in different locations”, North American Journal of
Economics and Finance, Vol. 14, pp. 89-114.
Hirshleifer, D., Subrahmanyam, A. and Titman, S. (1994), “Security analysis and trading patterns
when some investors receive information before others”, Journal of Finance, Vol. 49,
pp. 1665-98.
Bio-fuel policies
173
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Johansen, S. (1988), “Statistical analysis of cointegrating vectors”, Journal of Economic Dynamics
and Control, Vol. 12, pp. 231-54.
Johansen, S. (1991), “Estimation and hypothesis testing of cointegration vectors in Gaussian
vector autoregressive models”, Econometrica, Vol. 59, pp. 1551-80.
Johansen, S. (1992), “A determination of co-integration rank in the presence of a linear trend?”,
Oxford Bulletin of Economics and Statistics, Vol. 54, pp. 383-97.
Johansen, S. and Juselius, K. (1990), “Maximum likelihood estimation and inference on
cointegration with applications to the demand for money”, Oxford Bulletin of Economics
and Statistics, Vol. 52, pp. 169-210.
Khan, M. (2009), “The 2008 oil price bubble”, working paper, Peterson Institute for International
Economics, available at: www.iie.com
Medlock, K. and Jaffe, A. (2009), “Who is in the oil futures market and how has it changed?”,
working paper, Rice University, available at: www.bakerinstitute.org
Moosa, I. (2002), “Price discovery and risk transfer in the crude oil futures market: some
structural time series evidence”, Economic Notes by Banca Monte dei Paschi di Siena SpA,
Vol. 31, pp. 155-65.
Sadorsky, P. (2000), “The empirical relationship between energy futures prices and exchange
rates”, Energy Economics, Vol. 22, pp. 253-66.
Sims, C. (1980), “Macroeconomics and reality”, Econometrica, Vol. 48, pp. 1-48.
Sims, C. (1988), “Bayesian skepticism on unit root econometrics”, Journal of Economic Dynamics
and Control, Vol. 12, pp. 463-74.
Soytas, U. and Sari, R. (2007), “The relationship between energy and production: evidence from
Turkish manufacturing industry”, Energy Economics, Vol. 29, pp. 1151-65.
Tang, K. and Xiong, W. (2010), “Index investing and the ?nancialization of commodities”,
working paper, Princeton University, Princeton, NJ.
Corresponding author
Chia-Hsing Huang can be contacted at: [email protected]
JFEP
3,2
174
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_197006112.pdf
This paper seeks to study the impact of bio-fuel policies on oil and food futures prices from
December 6, 2004 to August 1, 2008.
Journal of Financial Economic Policy
Do bio-fuel policies lead to speculative behavior?
Chia-Hsing Huang Liang-Chun Ho
Article information:
To cite this document:
Chia-Hsing Huang Liang-Chun Ho, (2011),"Do bio-fuel policies lead to speculative behavior?", J ournal of
Financial Economic Policy, Vol. 3 Iss 2 pp. 161 - 174
Permanent link to this document:http://dx.doi.org/10.1108/17576381111133624
Downloaded on: 24 January 2016, At: 21:41 (PT)
References: this document contains references to 32 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 282 times since 2011*
Users who downloaded this article also downloaded:
Steve Swidler, (2011),"Homeownership: yesterday, today and tomorrow", J ournal of Financial Economic
Policy, Vol. 3 Iss 1 pp. 5-11http://dx.doi.org/10.1108/17576381111116786
Beverley Searle, (2011),"Recession and housing wealth", J ournal of Financial Economic Policy, Vol. 3 Iss 1
pp. 33-48http://dx.doi.org/10.1108/17576381111116867
Thomas D. Willett, (2012),"The role of defective mental models in generating the global financial crisis",
J ournal of Financial Economic Policy, Vol. 4 Iss 1 pp. 41-57http://dx.doi.org/10.1108/17576381211206479
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Do bio-fuel policies lead
to speculative behavior?
Chia-Hsing Huang
SolBridge International School of Business, Daejeon, South Korea, and
Liang-Chun Ho
Department of Finance, Hsiuping Institute of Technology, Taichung, Taiwan
Abstract
Purpose – This paper seeks to study the impact of bio-fuel policies on oil and food futures prices from
December 6, 2004 to August 1, 2008.
Design/methodology/approach – The daily closing prices of brent crude oil, light sweet crude oil,
corn, wheat, soybeans, and rough rice futures from December 6, 2004 to August 1, 2008 are used in this
research. The vector error correction model is applied in order to study the impact of bio-fuel policies on
oil and agricultural futures prices.
Findings – Unit root and cointegration tests showthat the brent crude oil, light sweet crude oil, wheat,
corn, soybeans, and rough rice futures are stationary and have a long-run equilibrium relationship.
Granger causality tests of the four periods shows that the causality relationship between oil futures and
food futures changes over time. The ?rst period result shows many Granger causes on several variables
at a 5 percent signi?cance level. The second period has more Granger causes at the 5 percent signi?cance
level. However, the Granger causalityrelationships become fewer andfewer inthe thirdandfourthperiod.
Originality/value – This is the ?rst paper to study the impact of the four major bio-fuel policies of
Brazil, the European Union, and the USA.
Keywords Fuels, Futures markets, Oils, Brazil, European union, United States of America
Paper type Research paper
1. Introduction
This research studies the impact of the four major bio-fuel policies of Brazil, the
European Union, and the USA. Based on the four bio-fuel policies, research time is
dividedinto four periods. December 6, 2004 to August 7, 2005 is the ?rst period, August 8,
2005 to March 7, 2007 is the second period, March 8, 2007 to December 17, 2007 is the
third period, and from December 18, 2007 to August 1, 2008 is the fourth period.
The major sources to make biofuels are corn, soybean, and wheat. Rice is still not used
as a raw material to make the biofuel. However, most of the countries that experienced
food shortage riots have rice as their major daily food source. Will the world’s markets
encourage bio-fuel supply to the rich countries by converting food crops to biofuel, while,
at the same time, poor countries not having a suf?cient foodsupplyandfacingstarvation?
This has become a major issue among politicians, economists, and environmentalists.
Including rice futures can help in the understanding of the relationship between the food
crisis and biofuels.
Implementing the Biodiesel Act of December 6, 2004, Brazil started to use biodiesels
which is a combination of conventional diesel and 5 percent vegetable oil ester. With this
biodiesel policy, there will be more farmers growing soybeans and sun?ower for the fuel
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – G0, O13
Bio-fuel policies
161
Journal of Financial Economic Policy
Vol. 3 No. 2, 2011
pp. 161-174
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381111133624
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
production since biodiesel can be derived from soybeans and sun?ower seeds. Ideally,
it will encourage both big and small farmers to enter this market and increase their
incomes.
In early 2007, the integrated energy and climate change package was proposed by the
EUCommission. This package was approved by the EUHeads of State in March 8, 2007
meeting of the European Council. The purpose of this package is to reduce global
warming. The goals of this package are to use 20 percent renewable energy by the year
2020, to reduce greenhouse gas emissions in 2020 by 20 percent compared with that of
1990, and with a 10 percent required minimumtarget for biofuel in 2020. It is also hoped
that all of the biofuel are produced in Europe, which will increase employment and
European GDP. The major sources of bio-fuel production in the EUare rapeseed, wheat,
and sugar beets.
There are two major bio-fuel policies inthe USA, the EnergyPolicy Act of 2005 andthe
Energy Independence and Security Act of 2007. In order to promote investment in energy
conservation and ef?ciency, President Bush signed the Energy Policy Act on August 8,
2005. This act stipulated that 4-billion gallons of renewable fuel were to be used in 2006
and increased to 7.5-billion gallons by 2012. One of the purposes of this act is to diversify
the energy supply with renewable sources. This act gives tax credits for the use of
renewable energy sources such as wind, solar, and bio-energy. It also gives tax incentives
to producers for ethanol and biodiesel research and expenses for development and
production. Normally, ethanol is mixed with gasoline in a 10 percent ethanol blend and is
called gasohol. The President of the USA signed the Energy Independence and Security
Act on December 18, 2007. The idea of this act is to take away the tax credit from the oil
and gas companies and move the money to the companies that develop and produce
alternative fuels. The expanded renewable fuels standard, with annual requirements for
the amount of renewable fuels produced and used in motor vehicles, requires 9-billion
gallons of renewable fuels in 2008 and rises to 36-billion gallons by 2022. The USA and
Brazil are the two major suppliers of ethanol to the world market. Brazil uses sugar cane
as the main source to produce ethanol. While in the USA, fuel ethanol is based mostly on
corn, switch grass, and soybeans.
The question of whether the bio-fuel policies cause higher food prices world-wide and
hunger in the developing countries has been widely debated. Based on a New York
Times report, those who support bio-fuel policies said that oil prices have increased so
much that an alternative is needed. One solution to high oil prices is biofuel. Biofuel is
onlyresponsible for 2-3 percent of the world’s foodprice increases while at the same time,
there is a reduction of crude oil demand by 1-million barrels a day. The food prices
increases are because of temporary problems such as droughts. Higher fertilizer and
crop transportation costs due to higher oil prices makes food more expensive. Countries,
like China and India, have increased incomes which enable the consumption of more
food. This is another driver for higher food prices. Critics, however, argue that a good
substitute of oil would bring the world a whole lot of bene?ts while at the same time have
only a minor impact on the world’s economy and environment. The negative impact of
biofuels is that farmers will growmore corn for biofuels at the expense of feeder corn and
other food crops. Unfortunately, increased food prices will have a greater impact on
low-income developing countries (The New York Times, May 30, 2008).
In addition to the aforementioned potential factors that affect food prices, the main
driver of higher food prices might be speculation in the ?nancial markets. When there
JFEP
3,2
162
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
is speculative behavior in ?nancial markets, the volatilities of agriculture futures and oil
futures prices may increase. The United States Commodity Futures Trading
Commission (CFTC) Chairman Chilton calls some of the non-traditional futures
speculator “massive passives”. These speculators have a long futures only strategy.
CFTC (2010a, b) proposed four different hard cap mandatory speculative position limits
on energy futures contracts. The purpose is to restrain the behavior that potentially can
distress markets.
Some farmers growagriculture products for bio-fuel production. Soybean, sun?ower,
and sugarcane are the main sources to make biofuel in Brazil. In the USA, farmers grow
corn, switchgrass, and soybean for biofuels. Rapeseed, wheat, and sugar beet are the
major inputs for biofuels in the EU. The Biodiesel Act of Brazil, Energy Policy Act of the
USA, integrated energy and climate change packages of EU, and the Energy
Independence and Security Act of the USA are the world’s four most important bio-fuel
policies. Brazil, the USA, and the EU are the major bio-fuel producers in the world.
Implementing these policies will encourage farmers to produce agriculture products for
biofuel and less for consumer food.
Empirical researches show that futures contracts provide price discovery function,
which is to use futures price as an indication of spot prices. Garbade and Silber (1983)
introduced a simultaneous dynamic price model to show the price discovery function of
futures. Most of the researches usedvector error correction(VEC) model to studythe price
discovery function of futures contracts, i.e. the lead-lag relationship between spot prices
and futures prices. Bopp and Sitzer (1987) showed that heating oil futures price is a good
predictor of spot prices. Moosa (2002) found that the daily crude oil futures contract
performs 60 percent of the price discovery function. Futures prices are determined by the
market trading andmayre?ect the expected future spot price of the participants. The two
groups of futures market speculators can be demarcated as the leading speculators and
the trend followers. The trend followers are the massive passives that do not undertake
intensive researchbefore makingfutures tradingdecisions. The leadingspeculators work
on fundamental and technical analysis in order to make trading decisions. New bio-fuel
policy could be one of the factors that the leading speculators may have studied and this
can in?uence the expected future spot prices, which in turn will affect the commodity
futures contracts prices.
Both theoretical and empirical researches demonstrate the existence of herd behavior
in the ?nancial markets (Froot et al., 1992; Hirshleifer et al., 1994; Barberis et al., 2005;
GreenandHwang, 2009). The prices inthe ?nancial markets cannot be fullyexplainedby
economic fundamental alone. Herdingbehavior is one of the factors that canin?uence the
prices. Investors are in?uenced by other people when making food futures transactions.
Under the in?uence of the herding behavior, investors even reverse their transaction
decision when observing the trends in the futures market. When most of the traders have
the same decision to have long position of a futures contract, the futures price will be
affected by this herding behavior. For example, if the leading speculator predicts that the
food prices will increase and buy the food futures, the followers will also buy the food
futures. The herding behavior phenomenon intensi?es in the bull market and diminishes
in the bear market (Bowe and Domuta, 2004).
The relationships of spot prices and futures prices also provide important information
for futures trading speculators. Studies all showthe gradually increasing relationships of
oil spot prices and oil futures prices in the world markets (Gulen, 1999; Sadorsky, 2000;
Bio-fuel policies
163
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hammoudeh et al., 2003). Food spot price transmission in the international markets is
studied by Balcombe et al. (2007). Balcombe and Rapsomanikis (2008) showed the
relationships of Brazilian sugar, oil, and ethanol prices. Tang and Xiong (2010) found that
the correlations of commodity returns and the world equity index, and the correlations of
commodityreturns andthe oil returnhave increasedsince the early2000s. Duringthe 2008
?nancial crisis, the spillover effects made a signi?cant contribution to the increased
commodity price volatility. These spillover effects are the results of both demand and
supply of physical commodities and fundamental ?nancialization processes of the
commodities markets.
Commodity price speculation, demand, limited supply, OPEC monopoly pricing, and
scarcity rent are the factors that may have caused 2008 high oil prices in the study of
Hamilton (2009). The results show that the low price elasticity of demand, strong world
demand, and limited supply are the three main reasons to have high oil prices in the
beginning. These fundamental reasons trigger speculative behavior. And the scarcity
rent may become an important factor in the future. Khan (2009) analyzed many
fundamental factors on the oil prices from 2003 and showed that in addition to the
fundamentals, speculation precipitatedanoil price bubble inthe ?rst half of 2008. If there
was no speculation, the oil prices would probably have been in the range of $80-$90 a
barrel. Medlock and Jaffe (2009) showed that the open interest of noncommercial players
in the oil futures market becomes a leading indicator of prices since January 2006. During
the time when speculators have net long positions and the market price moves upward.
If the following argument is true, then it may be deduced that the use of biofuel is one
contributor to increasing food prices. Only, if supply were (totally) inelastic and did not
respond to higher prices, the demand for biofuel would increase when the oil price
increases. Increasing demand for bio-fuel fosters the demand for bio-fuel sources, such
as corn, soybean, and wheat. Prices of corn, soybean, and wheat will soar when demand
for corn, soybean, and wheat intensi?es. When at the same time, the supply of corn,
soybean, and wheat for food decreases. Decreased supply and higher price of food may
cause unrest in the low-income developing countries. If this is not the case, then it may be
extrapolated that biofuel is not a major source of higher food prices.
Futures prices re?ect expectations of the future supply and demand of a commodity.
A supply or demand shock from a bio-fuel policy will affect food prices in the future.
If investors feel that a variable, e.g. the price of oil futures can be a predictor of another
variable, e.g. wheat futures prices and speculate on the futures market, then futures
prices will be in?uenced by this speculative behavior. If investors believe that when
wheat prices increase, because of the bio-fuel policy, corn prices will also increase; they
may speculate on wheat and corn futures. Two variables can be examed, e.g. wheat
futures price and corn futures price, to see whether there is any causality between them.
If bio-fuel policies have a long-term effect on commodity prices, then it will have an
impact on the commodity futures prices fromperiod one to period four in this study. If an
investor believes that the government’s bio-fuel policy will increase the demand of a
commodity, e.g. wheat, and speculate on the commodity futures, wheat futures, and the
substitute commodities futures, corn futures, then we can ?nd some evidence of this
speculative behavior in the futures market. However, if bio-fuel policies only increase
food prices a little in the short run and does not increase food prices in the long run, and
also if most investors believe that the futures market is not a good indication of future
JFEP
3,2
164
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
commodity price, then the speculative behavior will only happen in the short-term, not
from period one to period four in this study.
2. Data and research method
2.1 Data
The daily closing prices of brent crude oil, light sweet crude oil, corn, wheat, soybeans,
and rough rice futures of Chicago Mercantile Exchange Group fromDecember 6, 2004 to
August 1, 2008 are used in this research. Daily futures closing prices are taken fromthe
Data Streamdatabase. The natural log of the daily futures closing prices are used in the
analysis.
This paper will analyze the impact of the four major bio-fuel policies of Brazil, the
European Union, and America. Research time is divided into four periods. December 6,
2004 to August 7, 2005 is the ?rst period; August 8, 2005 to March 7, 2007 is the second
period; March 8, 2007 to December 17, 2007 is the third period; and from December 18,
2007 to August 1, 2008 is the fourth period. The sample sizes of the four periods are 168,
397, 197, and 157, respectively. The futures price series are shown in Figure 1.
2.2 Research method
2.2.1 Long-run equilibrium relationship. When data are close to the unit root,
Augmented Dickey-Fuller (ADF) test may not be able to reject the null hypothesis (Sims,
1988). In addition to using the ADF test for the null hypothesis test, the Phillips-Perron
(PP) test, Ng-Perron (NP) test, and Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test for
the unit root tests are also used. The null hypothesis asserts that there exists a unit root
in the ADF, PP, and NP tests. Unlike the null hypothesis of the other common unit root
tests, the null hypothesis of KPSS test is that the series is stationary. The KPSS test can
be used under the null of either trend stationarity or level stationarity. Therefore,
the KPSS can be used to test the possibility of fractionally integrated series as a
complementary method to the ADF, PP, and NP tests (Soytas and Sari, 2007).
Two or more time series are said to be cointegrated when the relationship between
themare non-stationarybut a linear combinationof themis stationary. The cointegrating
vector can be tested in order to see whether there is a statistically signi?cant connection
between the two time series. If there is a low order of integration of this cointegrating
vector, they are cointegrated. The Johansen cointegration test ( Johansen, 1991, 1992;
Figure 1.
Time series of futures
prices
2004/12/6 2005/8/8 2007/3/8
Rice
Soyabeans
Wheat
Corn
Light crude oil
Brent crude oil
2007/12/18 08/2008
Bio-fuel policies
165
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Johansen and Juselius, 1990) is used to examine the cointegration for more
than two variables. The Johansen cointegration test uses maximum eigenvalue and
trace in order to investigate the number of cointegrating vectors. Higher number of
cointegrating vector represents a long-run equilibrium relationship (Hammoudeh and
Li, 2004).
2.2.2 Short-run relationships. The concept of exogeneity is discussed by Engle et al.
(1983). Conway et al. (1984) criticized the methodology of the causality test literature.
Ericsson et al. (1998) discussed the concepts of exogeneity, cointegraion, and causality
and their applications to the economic policy analysis. Based on the premise that the
effect cannot happen before the cause, Granger (1969) de?ned Granger causality in
terms of predictability. An information set, X Granger causes Y if the past values
of X and Y can be used to predict Y more accurately than only using the past values of
Y. Granger causality tests can be used in a single equation with two variables and their
lags, multivariate causality tests with more variables, and Granger causality tests with a
vector autoregression (VAR) and VEC.
The VEC model by Johansen (1988) is used to explore the short-run relationships
among the variables (Granger, 1988). If the variables are stationary, the VAR model
introduced by Sims (1980) to analyze the short-run relationships among variables
can be applied. However, under the non-stationary variable condition, the VEC model
is used to study the short-run relationships among variables. The VEC model is
constructed thus:
C
t
¼ X
t21
· u þ
X
s
i51
f
i
· C
t2i
þj
t
Aset of six endogenous variables, percentage changes of rice futures price, corn futures
price, wheat futures price, soybean futures price, light oil price, and Brent oil futures
price, are collected in a 6 £ 1 vector C
t
. X is the error correction matrix. j
t
is the white
noise vector. This model means that the changes of endogenous variables are affected by
past changes and the adjustment to the long-term equilibrium.
3. Empirical result
3.1 Unit root test
Most of the economic and ?nancial time series are not stationary and need to be tested to
ascertain if they are stationary. If the data are not stationary, then a spurious regression
result may occur. ADF, PP, NP, and KPSS unit root tests are used on these six series in
different time periods. All of these six series are not stationary in level. The null
hypothesis of a unit root could not be rejected in the ADF, PP, and NP tests, and the null
hypothesis of stationaritywas rejectedinthe KPSS test. Table I shows that, after the ?rst
difference, all of the series are signi?cant at a 5 percent signi?cant level on the ADF, PP,
and NP tests, and not signi?cant on the KPSS test. All of these series are stationary after
the ?rst difference.
3.2 Cointegration tests
To study whether the variables move together, cointegration is used to simultaneously
model long-runpersistence andcomovement. The likelihood ratio trace statistics and the
maximum eigenvalue statistics for the cointegration test is used for evaluation. Both
the maximum eigenvalue and trace tests show the same result as shown in Table II.
JFEP
3,2
166
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
B
r
e
n
t
c
r
u
d
e
o
i
l
L
i
g
h
t
c
r
u
d
e
o
i
l
C
o
r
n
W
h
e
a
t
S
o
y
b
e
a
n
R
i
c
e
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
I
n
t
e
r
c
e
p
t
A
D
F
2
6
.
0
5
4
a
2
6
.
2
2
4
a
2
1
1
.
4
2
0
a
2
1
2
.
9
4
3
a
2
1
2
.
4
0
6
a
2
1
1
.
3
8
8
a
P
P
2
1
3
.
8
4
0
a
2
1
3
.
5
9
5
a
2
1
1
.
4
0
7
a
2
1
2
.
9
4
2
a
2
1
2
.
4
2
3
a
2
1
1
.
3
8
8
a
N
P
(
M
Z
a
)
2
5
0
.
2
9
3
a
2
2
3
.
8
1
1
a
2
3
0
.
2
1
4
a
2
1
2
.
8
1
6
a
2
4
0
.
8
0
3
a
2
3
3
.
2
7
4
a
K
P
S
S
0
.
0
6
2
0
.
0
4
7
0
.
0
6
8
0
.
0
5
7
0
.
0
8
2
0
.
0
8
5
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
6
.
0
8
6
a
2
6
.
1
9
3
a
2
1
1
.
3
8
1
a
2
1
2
.
9
1
6
a
2
1
2
.
3
9
6
a
2
1
1
.
3
5
0
a
P
P
2
1
3
.
8
3
4
a
2
1
3
.
5
5
7
a
2
1
1
.
3
5
6
a
2
1
2
.
9
1
5
a
2
1
2
.
4
1
1
a
2
1
1
.
3
5
0
a
N
P
(
M
Z
a
)
2
9
8
.
0
4
2
a
2
4
3
.
7
5
0
a
2
5
5
.
5
9
4
a
2
5
0
.
7
5
1
a
2
5
3
.
4
8
4
a
2
5
9
.
7
4
9
a
K
P
S
S
0
.
0
5
1
0
.
0
4
8
0
.
0
6
5
0
.
0
5
0
0
.
0
6
1
0
.
0
8
3
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
I
n
t
e
r
c
e
p
t
A
D
F
2
2
1
.
2
8
5
a
2
2
0
.
9
5
1
a
2
1
9
.
3
8
4
a
2
1
9
.
4
3
5
a
2
1
9
.
9
7
6
a
2
1
6
.
0
0
3
a
P
P
2
2
1
.
2
8
1
a
2
2
1
.
0
4
1
a
2
1
9
.
3
8
1
a
2
1
9
.
4
4
1
a
2
2
0
.
0
0
1
a
2
1
9
.
4
3
3
a
N
P
(
M
Z
a
)
2
3
2
.
7
9
8
a
2
4
3
.
1
2
7
a
2
1
2
2
.
5
0
1
a
2
6
8
.
0
6
3
a
2
4
5
.
0
2
6
a
2
1
6
1
.
6
0
9
a
K
P
S
S
0
.
0
7
4
0
.
0
6
7
0
.
3
4
6
0
.
0
4
1
0
.
4
3
2
0
.
0
2
3
2
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
2
1
.
2
5
9
a
2
2
0
.
9
2
7
a
2
1
9
.
4
9
6
a
2
1
9
.
4
1
1
a
2
2
0
.
1
3
7
a
2
1
5
.
9
8
4
a
P
P
2
2
1
.
2
5
6
a
2
2
1
.
0
1
6
a
2
1
9
.
4
9
8
a
2
1
9
.
4
1
8
a
2
2
0
.
1
9
4
a
2
1
9
.
4
0
7
a
N
P
(
M
Z
a
)
2
3
9
.
6
4
7
a
2
5
3
.
8
9
3
a
2
1
7
7
.
5
0
5
a
2
7
8
.
2
7
5
a
2
1
1
2
.
8
0
3
a
2
1
8
6
.
7
9
5
a
K
P
S
S
0
.
0
6
8
0
.
0
5
9
0
.
0
4
3
0
.
0
3
7
0
.
0
4
5
0
.
0
1
7
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
I
n
t
e
r
c
e
p
t
A
D
F
2
1
5
.
3
6
1
a
2
1
5
.
1
0
6
a
2
1
3
.
6
3
8
a
2
6
.
0
8
5
a
2
1
4
.
2
3
0
a
2
1
2
.
9
3
9
a
P
P
2
1
5
.
3
5
6
a
2
1
5
.
0
7
4
a
2
1
3
.
6
3
6
a
2
1
2
.
5
5
0
a
2
1
4
.
2
2
9
a
2
1
2
.
9
0
3
a
N
P
(
M
Z
a
)
2
3
6
.
0
9
8
a
2
3
0
.
4
8
5
a
2
7
3
.
7
2
8
a
2
9
.
4
4
9
a
2
6
4
.
4
7
7
a
2
8
8
.
6
4
8
a
K
P
S
S
0
.
0
4
3
0
.
0
4
5
0
.
3
4
3
0
.
1
0
0
0
.
1
5
3
0
.
3
1
5
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
1
5
.
3
1
8
a
2
1
5
.
0
6
5
a
2
1
3
.
8
3
2
a
2
6
.
0
7
6
a
2
1
4
.
2
9
2
a
2
1
3
.
0
8
7
a
P
P
2
1
5
.
3
1
4
a
2
1
5
.
0
3
4
a
2
1
3
.
8
3
2
a
2
1
2
.
5
3
1
a
2
1
4
.
2
9
7
a
2
1
3
.
1
2
3
a
N
P
(
M
Z
a
)
2
5
9
.
7
3
6
a
2
5
4
.
9
1
6
a
2
9
6
.
7
2
0
a
2
8
.
9
3
6
a
2
6
8
.
8
8
9
a
2
9
5
.
5
7
2
a
K
P
S
S
0
.
0
3
9
0
.
0
4
2
0
.
0
3
4
3
0
.
0
9
3
0
.
0
3
0
0
.
0
3
5
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
I
n
t
e
r
c
e
p
t
A
D
F
2
1
3
.
2
3
7
a
2
9
.
8
0
0
a
2
5
.
9
2
9
a
2
9
.
9
6
6
a
2
1
1
.
3
7
8
a
2
1
1
.
2
4
8
a
P
P
2
1
3
.
2
3
9
a
2
1
2
.
4
3
7
a
2
1
0
.
4
4
8
a
2
1
0
.
7
0
7
a
2
1
1
.
3
4
4
a
2
1
1
.
2
8
5
a
N
P
(
M
Z
a
)
2
6
4
.
2
9
4
a
2
7
4
.
0
4
0
a
2
9
.
5
2
9
a
2
2
.
4
3
3
2
1
1
.
8
1
6
a
2
9
.
2
3
0
a
K
P
S
S
0
.
1
1
8
0
.
1
1
4
0
.
3
1
5
0
.
1
0
6
0
.
1
5
3
0
.
7
1
8
a
I
n
t
e
r
c
e
p
t
a
n
d
t
r
e
n
d
A
D
F
2
1
3
.
2
2
3
a
2
9
.
8
1
2
a
2
6
.
1
8
2
a
2
9
.
9
4
6
a
2
1
1
.
4
2
7
a
2
1
1
.
6
7
5
a
P
P
2
1
3
.
2
2
5
a
2
1
2
.
4
2
8
a
2
1
0
.
6
2
2
a
2
1
0
.
6
7
4
a
2
1
1
.
3
9
4
a
2
1
1
.
6
5
9
a
N
P
(
M
Z
a
)
2
7
4
.
1
3
1
a
2
7
6
.
9
9
2
a
2
1
7
.
5
9
4
a
2
7
0
.
6
1
9
a
2
1
3
.
6
5
4
2
7
4
.
9
0
5
a
K
P
S
S
0
.
0
8
9
0
.
0
7
8
0
.
0
8
5
0
.
0
8
7
0
.
0
7
4
0
.
0
7
5
N
o
t
e
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
n
u
l
l
h
y
p
o
t
h
e
s
i
s
(
H
0
)
a
t
t
h
e
0
.
0
5
l
e
v
e
l
Table I.
First difference unit root
test results
Bio-fuel policies
167
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
E
i
g
e
n
v
a
l
u
e
T
r
a
c
e
M
a
x
-
E
i
g
e
n
E
i
g
e
n
v
a
l
u
e
T
r
a
c
e
M
a
x
-
E
i
g
e
n
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
N
o
n
e
a
0
.
6
3
2
5
2
4
.
4
5
7
1
6
5
.
7
7
3
0
.
4
4
3
1
,
0
6
6
.
4
6
5
2
3
1
.
3
7
1
A
t
m
o
s
t
1
a
0
.
4
6
9
3
5
8
.
6
8
4
1
0
5
.
1
5
3
0
.
3
9
9
8
3
5
.
0
9
4
2
0
0
.
8
7
7
A
t
m
o
s
t
2
a
0
.
3
8
6
2
5
3
.
5
3
1
8
0
.
9
1
0
0
.
3
8
1
6
3
4
.
2
1
7
1
8
9
.
2
4
6
A
t
m
o
s
t
3
a
0
.
3
5
2
1
7
2
.
6
2
1
7
1
.
9
4
0
0
.
3
5
0
4
4
4
.
9
7
1
1
7
0
.
2
5
6
A
t
m
o
s
t
4
a
0
.
2
8
8
1
0
0
.
6
8
1
5
6
.
3
2
8
0
.
3
0
3
2
7
4
.
7
1
5
1
4
2
.
3
0
1
A
t
m
o
s
t
5
a
0
.
2
3
4
4
4
.
3
5
2
4
4
.
3
5
2
0
.
2
8
5
1
3
2
.
4
1
4
1
3
2
.
4
1
4
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
N
o
n
e
a
0
.
3
9
2
4
7
3
.
7
0
2
9
6
.
9
9
7
0
.
4
3
6
3
9
6
.
3
8
5
8
8
.
8
3
0
A
t
m
o
s
t
1
a
0
.
3
8
5
3
7
6
.
7
0
5
9
4
.
7
5
1
0
.
3
8
0
3
0
7
.
5
5
5
7
4
.
1
9
0
A
t
m
o
s
t
2
a
0
.
3
5
5
2
8
1
.
9
5
4
8
5
.
5
6
7
0
.
3
7
8
2
3
3
.
3
6
5
7
3
.
6
4
2
A
t
m
o
s
t
3
a
0
.
2
9
7
1
9
6
.
3
8
6
6
8
.
5
9
3
0
.
3
1
5
1
5
9
.
7
2
4
5
8
.
5
8
4
A
t
m
o
s
t
4
a
0
.
2
8
3
1
2
7
.
7
9
3
6
4
.
9
8
7
0
.
3
0
0
1
0
1
.
1
3
9
5
5
.
3
7
5
A
t
m
o
s
t
5
a
0
.
2
7
5
6
2
.
8
0
6
6
2
.
8
0
6
0
.
2
5
6
4
5
.
7
6
5
4
5
.
7
6
5
N
o
t
e
s
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
0
5
l
e
v
e
l
;
H
0
–
n
o
.
o
f
C
E
(
s
)
;
t
h
e
?
r
s
t
d
i
f
f
e
r
e
n
c
e
o
f
v
a
r
i
a
n
c
e
s
i
s
u
s
e
d
i
n
J
o
h
a
n
s
e
n
c
o
i
n
t
e
g
r
a
t
i
o
n
t
e
s
t
Table II.
Johansen cointegration
test results
JFEP
3,2
168
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
All of the tests are signi?cant and reject the null hypothesis, which means
all of the variables have cointegrating vectors. Under the 5 percent signi?cant level,
there are ?ve cointegration vectors both from the maximum eigenvalue and trace tests.
Brent oil futures, light oil futures, corn futures, wheat futures, soybeans futures, and rice
futures all have signi?cant long-run comovement relationships. This result shows that
one variable can be used to predict movement of another variable.
3.3 Granger causality test
To study whether one variable is a leading indicator of another variable, the Granger
causality test is used in order to explore the causalities among the variables. Table III
shows the Granger causality test results of the four periods.
FromDecember 6, 2004 to August 8, 2005, brent crude oil futures Granger causes rice
futures. Light crude oil futures Granger causes wheat futures. Corn futures Granger
causes rice futures. Soybean futures Granger causes brent crude oil futures and light
crude oil futures. Wheat futures Granger causes corn futures.
The result indicates that after implementingthe Biodiesel Act inBrazil inDecember 6,
2004, the change of the soybeans futures price is a leading indicator of the change of
brent crude oil futures price and light crude oil futures price. Soybean is the major source
to produce biodiesel. Investors in the futures market expect that the price changes of
soybeans futures will lead the price changes of the oil futures. The investors in the
futures market seem to perceive this is a high food price signal sent by the Biodiesel
policy. In addition to speculation on the situation that the soybean futures Granger
causes brent crude oil futures and light crude oil futures, investors also speculate on
many Granger causality relationships, such as the brent crude oil futures to the rice
futures, light crude oil futures to the wheat futures, corn futures to the rice futures, and
wheat futures to the corn futures. If most of these Granger causality relationships persist
in the following periods, we may argue that the bio-fuel policy could be one of the factors
for higher food prices.
The Granger causality test results from August 8, 2005 to March 7, 2007 show that
brent crude oil futures and rice futures have two-way feedback relationships. Light
crude oil futures and rice futures also have two-way feedback relationships. Wheat
futures Granger causes corn futures, soybeans futures, and rice futures.
Corn, switch grass, and soybean are the major sources for bio-fuel production in the
USA. However, the investors in the futures market do not use this chance to speculate on
the Granger causality relationships of corn futures on oil futures, soybeans futures on oil
futures, oil futures on corn futures, and oil futures on soybeans futures. Investors
speculate on the Granger causality relationships of Brent oil futures on rice futures,
light oil futures on rice futures, wheat futures on corn futures, soybeans futures, and
rice futures. The result shows that the investors in the futures market expect that the
implementation of the Energy Policy Act on August 8, 2005 in the USA will encourage
more farmers to grow the corn and soybeans for bio-fuel production. And some of the
new corn or soybean farmers may have previously grown wheat or rice. The change in
wheat futures price then can be a leading indicator of the prices of corn futures, soybeans
futures, and rice futures.
The Granger causality relationships of soybeans futures on brent oil futures and on
light oil futures in the ?rst period do not continue to exist in the second period. The corn
futures Granger causes the rice futures in the ?rst period, but corn futures does
Bio-fuel policies
169
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
D
e
p
e
n
d
e
n
t
v
a
r
i
a
b
l
e
B
r
e
n
t
c
r
u
d
e
o
i
l
L
i
g
h
t
c
r
u
d
e
o
i
l
C
o
r
n
S
o
y
b
e
a
n
W
h
e
a
t
R
i
c
e
D
e
c
e
m
b
e
r
6
,
2
0
0
4
-
A
u
g
u
s
t
7
,
2
0
0
5
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
2
0
.
1
7
4
0
.
8
0
1
0
.
2
6
6
0
.
1
0
5
0
.
0
6
7
a
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
1
1
6
–
0
.
8
3
2
0
.
1
0
2
0
.
0
3
9
b
0
.
1
2
4
C
o
r
n
0
.
5
7
7
0
.
5
9
7
–
0
.
6
0
5
0
.
1
5
5
0
.
0
3
8
b
S
o
y
b
e
a
n
0
.
0
1
6
b
0
.
0
2
9
b
0
.
4
6
2
–
0
.
3
8
0
0
.
8
2
7
W
h
e
a
t
0
.
3
9
5
0
.
2
4
0
0
.
0
8
0
a
0
.
1
7
6
–
0
.
4
9
7
R
i
c
e
0
.
4
7
8
0
.
7
0
6
0
.
2
4
3
0
.
3
4
1
0
.
5
6
2
–
A
u
g
u
s
t
8
,
2
0
0
5
-
M
a
r
c
h
7
,
2
0
0
7
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
0
0
1
b
0
.
9
2
1
0
.
4
4
6
0
.
5
7
0
0
.
0
3
7
b
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
0
0
7
b
–
0
.
5
1
9
0
.
7
4
7
0
.
7
8
6
0
.
0
1
8
b
C
o
r
n
0
.
3
2
0
0
.
3
3
9
–
0
.
3
3
4
0
.
2
2
0
0
.
4
9
6
S
o
y
b
e
a
n
0
.
4
3
5
0
.
5
0
1
2
0
.
9
4
0
–
0
.
3
2
9
0
.
9
8
8
W
h
e
a
t
0
.
9
4
8
0
.
9
4
1
0
.
0
3
1
a
0
.
0
2
5
b
–
0
.
0
8
5
a
R
i
c
e
0
.
0
0
9
b
0
.
0
5
6
a
0
.
6
6
9
0
.
1
8
9
0
.
9
5
6
–
M
a
r
c
h
8
,
2
0
0
7
-
D
e
c
e
m
b
e
r
1
7
,
2
0
0
7
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
1
1
3
0
.
1
3
1
0
.
7
1
3
0
.
2
5
9
0
.
1
2
6
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
6
0
1
–
0
.
5
1
5
0
.
4
7
6
0
.
5
6
1
0
.
8
4
0
C
o
r
n
0
.
8
4
7
0
.
9
8
5
–
0
.
4
5
9
0
.
0
0
8
b
0
.
5
2
3
S
o
y
b
e
a
n
0
.
2
5
2
0
.
2
6
1
0
.
2
1
0
–
0
.
0
4
3
b
0
.
4
2
4
W
h
e
a
t
0
.
1
8
9
0
.
3
3
8
0
.
1
1
8
0
.
0
6
4
a
–
0
.
2
9
1
R
i
c
e
0
.
9
4
4
0
.
9
8
1
0
.
8
0
4
0
.
5
4
7
0
.
4
9
0
–
D
e
c
e
m
b
e
r
1
8
,
2
0
0
7
-
A
u
g
u
s
t
1
,
2
0
0
8
P
r
o
b
.
B
r
e
n
t
c
r
u
d
e
o
i
l
–
0
.
0
4
2
b
0
.
9
2
5
0
.
8
9
9
0
.
5
7
8
0
.
6
4
0
L
i
g
h
t
c
r
u
d
e
o
i
l
0
.
1
8
1
–
0
.
6
0
2
0
.
3
9
1
0
.
2
7
3
0
.
6
6
2
C
o
r
n
0
.
5
5
8
0
.
7
8
1
–
0
.
1
2
7
0
.
4
8
7
0
.
0
6
6
a
S
o
y
b
e
a
n
0
.
4
2
8
0
.
2
4
1
0
.
8
9
6
–
0
.
5
0
9
0
.
8
0
5
W
h
e
a
t
0
.
1
3
2
0
.
1
9
1
0
.
3
5
2
0
.
9
5
1
–
0
.
6
1
6
R
i
c
e
0
.
8
0
1
0
.
9
3
0
0
.
7
4
1
0
.
5
6
5
0
.
2
4
9
–
N
o
t
e
s
:
a
D
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
1
l
e
v
e
l
;
b
d
e
n
o
t
e
s
r
e
j
e
c
t
i
o
n
o
f
t
h
e
H
0
a
t
t
h
e
0
.
0
5
l
e
v
e
l
Table III.
VEC Granger
causality/block
exogeneity Wald tests
JFEP
3,2
170
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
not Granger cause rice futures in the second. The result shows that there exists
speculative behavior in the corn futures, soybeans futures, and rice futures markets and
the bio-fuel policies do not have a permanent effect on the oil futures and food futures
price relationships.
The Granger causality test results from March 8, 2007 to December 17, 2007 show
that wheat futures and soybeans futures have two-way feedback relationships. Corn
futures Granger causes wheat futures. In Europe, rapeseed, wheat, and sugar beets are
the major sources of the production of biofuel. The results showthat the implementation
of the integrated energy and climate change package, on March 8, 2007 in the EU, lead
futures investors to believe that this bio-fuel policy may attract more farmers to grow
wheat for bio-fuel production. Farmers that used to grow other crops may switch to
growing wheat. Achange in corn futures price and soybeans futures price then can have
a signi?cant impact on wheat futures price.
During the second period, there are nine signi?cant Granger causalitypairs. It reduces
to only three signi?cant Granger causality pairs in the third period. Most of the Granger
causality relationships in the previous periods cease to exist in the third period. The
bio-fuel policies only have a temporary effect on food futures and oil futures Granger
causality relationships. In the third period, investors only speculate on the Granger
causality relationships of corn futures on wheat futures and the two-way feedback
relationship of soybeans futures and wheat futures.
The Granger causality test results fromDecember 18, 2007 to August 1, 2008 showthat
brent crude oil futures Granger causes light oil futures. Corn futures Granger causes rice
futures. There are only two signi?cant Granger causality pairs in the fourth period. Some
of the Granger causality relationships during the early periods do not exist in the fourth
period. It seems most of the speculative behavior inthe futures market disappears infourth
periods of this research. The Energy Independence and Security Act of December 18, 2007
in the USA did not cause most of the variables to have signi?cant Granger causality
relationships. The decrease in speculative behavior may be due to the world’s ?nancial
problems originating fromthe subprime crisis in the USA. In July 2007, the default rate on
subprime loans was too high which resulted in some of the mortgage-backed securities
issuers having their credit ratings downgraded. For example, Bear Stearns, one of the
market’s big players, suffering from subprime loans problems closed two funds with
important subprime investments in July 2007. The subprime loan crisis spread to other
parts of the world. UBS of Switzerland announced a loss of $10 billion in the US subprime
market. We can see most of the Granger causality relationships are not consistent during
the four periods. If biofuel is the major driver for higher food price, these Granger causality
relationships would have existed in the long term.
4. Conclusion
Since December 6, 2004, Brazil’s Government have implemented the Biodiesel policy.
The USA and the EU followed with implementation of the Energy Policy Act, the
Integrated Energy and Climate Change Package, and ?nally, the Energy Independence
and Security Act of August 2005, March 2007, and December 2007, respectively. Bio-fuel
policies are intended to reduce pressure from high oil prices and increase farmers’
income from producing crops for biofuel. However, at the same time, food prices also
increase and some countries experience food shortages as well.
Bio-fuel policies
171
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The VEC model is used to study the impact of bio-fuel policies on oil and agricultural
futures prices. If the main driver of higher food prices is bio-fuel policy, oil futures prices
should have a consistently strong impact on agricultural futures prices. If this is not the
case, then the inconsistent result of the futures prices’ relationship in every period is due
to market speculation.
Unit root and cointegration tests show that the brent oil futures, light oil futures,
wheat futures, corn futures, soybeans futures, and rice futures are non-stationary in
levels and stationary once they are ?rst differenced, cointegrated, and have a long-run
equilibrium relationship. Granger causality tests of the four period shows that the
causality relationship between oil futures and food futures changes over time. The ?rst
periodresult shows manyGranger causes onseveral variables at the 5 percent signi?cant
level. Light oil futures Granger causes wheat futures. Corn futures Granger causes rice
futures. Soybeans futures Granger causes brent oil futures and light oil futures.
The second period has more Granger causes at the 5 percent signi?cant level. Brent
oil futures and light oil futures have two-way feedback relationships. Brent oil futures
and rice futures have two-way feedback relationships. Light oil futures Granger causes
rice futures. Wheat futures Granger causes soybeans futures. However, the Granger
causality relationships become progressively scarcer in the third and fourth period.
In the third period, only corn futures Granger causes wheat and soybeans futures
Granger causes wheat futures at the 5 percent signi?cant level. Only one 5 percent
signi?cant Granger causality is left in the fourth period. In the fourth period, brent oil
futures Granger causes light oil futures.
The fundamental market demand and supply are the major forces for the high oil and
food prices. Excess demand of the world oil and food and not enough supply cause the
expectation of higher oil and food prices. However, the oil and food spot prices and futures
prices cannot be explained by the economic fundamental alone. In addition to the
fundamental demand and supply, bio-fuel policy, speculation and investor herding can
alsoaffect oil andfoodspot prices andfutures prices. Higher expectedoil andfoodprices in
the market drawthe attention of futures speculators. Speculators predict the oil and food
spot prices will increase andspeculate onhigher oil andfoodfutures prices. Inaddition, the
speculators use one oil/food futures price change to predict another oil/food futures price
change. Therefore, an oil/food futures price is used to be the predictor variable of another
oil/food futures price. Investors in the futures markets herd together. Herding behavior of
followers in the oil and food futures markets intensi?es the causalities of oil and food
futures prices. In the early time periods, followers have a passive long strategy and drive
the oil and food futures prices upwards. This phenomenon is more serious in the ?rst
period and gradually fades away in the following periods.
If it is the bio-fuel policies that cause higher oil and food prices, there should be more
and more Granger causality relationships in these four periods. No evidence is found to
show that the bio-fuel policies cause higher oil and food prices. It is possible that the
investor speculation behavior in the market drives oil futures and food futures prices in
the short-term. The speculations and herding behavior only increase the numbers of
Granger causalities in the ?rst time period in this study. Similar to the result of investor
herding behavior literatures, this research shows that the oil and food futures have more
Granger causalities in the bull market than in the bear market. The speculations
and herdingbehavior only increase the oil and food futures prices inthe short-termnot in
all of the four time periods.
JFEP
3,2
172
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
References
Balcombe, K. and Rapsomanikis, G. (2008), “Bayesian estimation and selection of nonlinear
vector error correction models: the case of the sugar-ethanol-oil nexus in Brazil”, American
Journal of Agricultural Economics, Vol. 90, pp. 658-68.
Balcombe, K., Bailey, A. and Brooks, J. (2007), “Threshold effects in price transmission: the case
of Brazilian wheat, maize, and soya prices”, American Journal of Agricultural Economics,
Vol. 89, pp. 308-23.
Barberis, N., Shleifer, A. and Wurgler, J. (2005), “Comovement”, Journal of Financial Economics,
Vol. 75, pp. 283-317.
Bopp, A. and Sitzer, S. (1987), “Are petroleum futures prices good predictors of cash value?”,
Journal of Futures Markets, Vol. 7, pp. 705-19.
Bowe, M. and Domuta, D. (2004), “Investor herding during ?nancial crisis: a clinical study of the
Jakarta stock exchange”, Paci?c-Basin Finance Journal, Vol. 12, pp. 387-418.
CFTC (2010a), Statement by Chairman Chilton at the Electric Power Supply Association Annual
Meeting, January 30, 2010, Commodity Futures Trading Commission, available at: www.
cftc.gov
CFTC (2010b), “Statement by Chairman Chilton regarding the notice of proposed rulemaking for
speculative position limits for referenced energy contract, January 14, 2010”, Commodity
Futures Trading Commission, available at: www.cftc.gov
Conway, R., Swamy, P., Yanagida, J. and von zur Muehlen, P. (1984), “The impossibility of
causality testing”, Agricultural Economics Research, Vol. 36, pp. 1-19.
Engle, R., Hendry, D. and Richard, J. (1983), “Exogeneity”, Econometrica, Vol. 51, pp. 277-304.
Ericsson, N., Hendry, D. and Mizon, G. (1998), “Exogeneity, cointrgration, and economic policy
analysis”, Journal of Business and Economic Statistics, Vol. 16, pp. 370-87.
Froot, K., Scharfstein, D. and Stein, J. (1992), “Herd on the street: informational inef?ciencies in a
market with short-term speculation”, Journal of Finance, Vol. 47, pp. 1461-84.
Garbade, K. and Silber, W. (1983), “Price movement and price discovery in futures and cash
markets”, Review of Economics and Statistics, Vol. 65, pp. 289-97.
Granger, C. (1969), “Investigating causal relations by economic models and cross spectral
methods”, Econometrica, Vol. 37, pp. 424-38.
Granger, C. (1988), “Causality, cointegration and control”, Journal of Economics and Control,
Vol. 12, pp. 551-9.
Green, C. and Hwang, B. (2009), “Price-based return comovement”, Journal of Financial
Economics, Vol. 93, pp. 37-50.
Gulen, S. (1999), “Regionalization in world crude oil markets: further evidence”, Energy Journal,
Vol. 20, pp. 125-39.
Hamilton, J. (2009), “Understanding crude oil prices”, The Energy Journal, Vol. 30, pp. 179-206.
Hammoudeh, S. and Li, H. (2004), “The impact of the Asian crisis on the behavior of US and
international petroleum prices”, Energy Economics, Vol. 26, pp. 135-60.
Hammoudeh, S., Li, H. and Jeon, B. (2003), “Causality and volatility spillover among petroleum
price of WTI, gasoline and heating oil in different locations”, North American Journal of
Economics and Finance, Vol. 14, pp. 89-114.
Hirshleifer, D., Subrahmanyam, A. and Titman, S. (1994), “Security analysis and trading patterns
when some investors receive information before others”, Journal of Finance, Vol. 49,
pp. 1665-98.
Bio-fuel policies
173
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Johansen, S. (1988), “Statistical analysis of cointegrating vectors”, Journal of Economic Dynamics
and Control, Vol. 12, pp. 231-54.
Johansen, S. (1991), “Estimation and hypothesis testing of cointegration vectors in Gaussian
vector autoregressive models”, Econometrica, Vol. 59, pp. 1551-80.
Johansen, S. (1992), “A determination of co-integration rank in the presence of a linear trend?”,
Oxford Bulletin of Economics and Statistics, Vol. 54, pp. 383-97.
Johansen, S. and Juselius, K. (1990), “Maximum likelihood estimation and inference on
cointegration with applications to the demand for money”, Oxford Bulletin of Economics
and Statistics, Vol. 52, pp. 169-210.
Khan, M. (2009), “The 2008 oil price bubble”, working paper, Peterson Institute for International
Economics, available at: www.iie.com
Medlock, K. and Jaffe, A. (2009), “Who is in the oil futures market and how has it changed?”,
working paper, Rice University, available at: www.bakerinstitute.org
Moosa, I. (2002), “Price discovery and risk transfer in the crude oil futures market: some
structural time series evidence”, Economic Notes by Banca Monte dei Paschi di Siena SpA,
Vol. 31, pp. 155-65.
Sadorsky, P. (2000), “The empirical relationship between energy futures prices and exchange
rates”, Energy Economics, Vol. 22, pp. 253-66.
Sims, C. (1980), “Macroeconomics and reality”, Econometrica, Vol. 48, pp. 1-48.
Sims, C. (1988), “Bayesian skepticism on unit root econometrics”, Journal of Economic Dynamics
and Control, Vol. 12, pp. 463-74.
Soytas, U. and Sari, R. (2007), “The relationship between energy and production: evidence from
Turkish manufacturing industry”, Energy Economics, Vol. 29, pp. 1151-65.
Tang, K. and Xiong, W. (2010), “Index investing and the ?nancialization of commodities”,
working paper, Princeton University, Princeton, NJ.
Corresponding author
Chia-Hsing Huang can be contacted at: [email protected]
JFEP
3,2
174
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
1
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_197006112.pdf