The Pricing Model of Rice: Evidence from Thailand
PITHAK SRISUKSAI1, VIMUT VANITCHAREARNTHUM2
1School of Economics, Sukhothai Thammathirat Open University
9/9 Changwattana Road, Pakkret District, Nonthaburi, 11120,
THAILAND
2Faculty of Commerce and Accountancy, Chulalongkorn University
254 Phayathai Road, Pathumwan, Bangkok, 10330
THAILAND
Abstract: Although the rice market in Thailand is linked tightly with the world market, it is fraught with
government intervention policies from time to time. In 2011 the Yingluck Shinawatra government embarked
the rice pledging scheme that aimed to bid up domestic paddy prices above the global market. Such scheme set
the premium of 40-50 percent higher than the world price. Nevertheless, the program was discontinued in
2014. This paper attempts to shed light on how the price structure in Thailand was affected by government
intervention policies. In particular, we conducted a field survey to unearth the transmission mechanism that
channel information about market factors on prices in various levels, ranging from world rice price to the
wholesale paddy prices. The findings show that the paddy price was mainly determined by the rice mill, the
central paddy market, the middleman, and the exporters. The government intervention policies could influence
the paddy price temporarily, i.e., when the scheme was in place. In the long-run equilibrium, the paddy price is
determined by a combination of the price at which the mills are willing to buy and the world price. Our
empirical investigation, based on the Engel-Granger Cointegration test and the Error-correction model,
confirms this conjecture. Moreover, the price expectation embedded in the millers’ offer price played a
significant role in the price discovery process. The causality test revealed that the expected future price causes
the spot wholesale price in the Granger sense. It is also found that the causation is bi-directional implying that
the flows of information between markets are essential in determining the equilibrium price in the rice market.
Key-Words: Rice Price, Pricing Model, Pledging, Error Correction Model
Received: September 16, 2021. Revised: May 21, 2022. Accepted: June 13, 2022. Published: July 22, 2022.
1 Introduction
In 2011 the Yingluck Shinawatra government
embarked the rice pledging scheme that aimed to
bid up domestic paddy prices. Such scheme set the
pledging price 40-50 percent higher than the global
price. Despite being among the world major rice
exporter, the price intervention policy hardly had
any impact on the world price. However, whether
such scheme has a long-lasting effect on domestic
price, and consequently farmers’ income, remain to
be explored. To be able to uncover such effect, one
needs to understand the structure of rice prices in
Thailand first. In particular, one needs to
disentangle the transmission mechanism that
channel information about market factors on prices
in various levels, ranging from world rice price to
the wholesale paddy prices.
From 2001 to 2009, approximately 70 percent of
rice production was consumed in Thailand. The
domestic paddy production averages 32 million tons
per year, while the production of milled rice
averages 20 million tons per year. The processing,
distribution and retailing all involve many parties,
starting with farmers and progressing to rice mills,
central markets, exporters, wholesalers and retailers,
and finally to domestic and foreign consumers. The
supply chain of rice also creates a linkage of prices
in various layers.
In the past, the government implemented
measures to control the price of various types of
paddy. Isavilanon (2010a) categorized the subsidy
schemes as follows
1) Price support for paddy. The principle of this
measure is minimum pricing by the government to
create artificial demand in the rice market. This
measure has been in effect since 1955 with the
establishment of the Warehouse Organization to go
out to buy and collect paddy from farmers. After
that, in 1974 the government established the
Farmers Aid Fund and the Marketing Organization
for Farmers which is an agency that purchases
paddy from farmers.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1245
Volume 19, 2022
2) Paddy Pledge. The operation started from the
1981/82 planting year. The government assigned
this duty to the Bank for Agriculture and
Agricultural Cooperatives (BAAC), but in the
1991/92 planting year, the paddy pledge was under-
funded by the agricultural fund.
3) The Insurance Program of Rice Farmers’
Income was an idea that came from the risk
insurance principle due to the price volatility. It was
a contractual agreement between the government
and the farmers of which BAAC acts as a state
representative. The farmers had to inform the
BAAC about the amount of area, quantity of
production and target price. As a result, at the
harvest time, if the current quote was lower than the
target price, the government paid the difference
between the target price and the quote price.
However, the price pledging scheme of the
Yingluck government in 2011 pushed the subsidy
beyond all past measures. It set the pledging price
about 50 percent above the market price and open to
all offers without quantity limit. It is obvious that
the pricing structure in Thailand was distorted
during the period when the scheme was in place.
The price pledging scheme was discontinued in May
2014 when the military staged the coup and
abolished the program.
Though the program was short-lived but it worths
the while to investigate how the price structure of
rice market was affected by the scheme. Therefore,
this study attempts to learn about the changing price
structure in the past years, especially after the rice
pledging scheme of the government was introduced.
It also attempts to develop a model that explains
how prices in the supply chain are determined, the
role of the futures market in reducing price
fluctuation, and the price discovery process. In the
end, the study can deliver policy recommendations
regarding the price intervention scheme based on
empirical evidence.
2 Literature Review
The price structure of Thai rice has changed
dramatically. Recently, Thailand has had to face
important competitors, namely Vietnam, India,
Pakistan, and the United States of America. A study
by Chaokul et al. (2011) found that the proportion of
rice exported and used in Thailand in the form of
direct consumption and processed food was quite
similar; about 45 percent of total production. It
indicated that the Thai rice economy noticeably had
to rely more on the world rice market. Consistent
with the study of Isavilanon et al. (2009), it was
found that the rice price level in the 2007/08
production year increased significantly from the
1987/88 production year, but the production cost
inflated more than such price. As a result, most
farmers in the rainwater fields suffered losses from
rice production. However, the research results of
Puapongsakorn and Jarupong (2010) indicated that
the 5.24 million tons of the paddy project in 2005/06
caused the implicit costs and a loss of 19,130
million baht (597.81 million US$). The paddy
pledge has led to changes in social welfare because
this project transferred resources from taxpayers and
consumers to producers. That is, welfare costs
equaled between -16,609.94 million baht (-519.06
million US$) to -17,720.99 million baht (-553.78
million US$), which was lower than the loss. The
results of a study on intervention in rice prices in
Thailand differed from those in Vietnam. Ghosh and
Whalley (2004) studied rice price controls in
Vietnam and explained that the price control always
generated income for the government. Once the
state controlled the prices by setting the selling price
of farmers’ rice to be lower than the market price,
this would increase the state's income. In addition,
the control of rice prices was an appropriate
measure in the situation of foreign market
uncertainty. As a result, the price controls could
protect the domestic market from any external
fluctuations, as well as reducing the trading
costs. Puapongsakorn et al. (2013) found that
the amount of global rice consumption would
increase slowly. Consequently, they proposed
three solutions to the problem, namely 1)
reducing production costs and/or increasing
yield per rai
1
(0.4 acre); 2) increasing yield per
farmer; and 3) producing quality rice.
That is why Isavilanon (2010b) concluded
that the state policies on rice shifted to
subsidizing the paddy producers or rice farmers,
such as the paddy price support policies, the
paddy pledge policy, the rice farmers’ income
insurance policy, the rice insurance program,
and the rice farmers' cost grant program. In
addition, Duangthip (2010) showed that the farmers
benefited more directly from the income insurance
scheme than the rice pledging scheme. More
importantly, the actual price in the rice market was
severely distorted by the rice pledging scheme
because the state set the pledging price much higher
than the market price. In addition, there was a
proposal for the government to adjust the price of
1
One rai is approximately 0.4 acre.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1246
Volume 19, 2022
the pledged mortgage to be more in line with the
market price and take the pledge at the right time.
This is consistent with the proposal of the Thailand
Development Research Institute (TDRI) (2009).
This study recommended that the government
should insure the price risk by requiring the
announcement of the insurance price at the
beginning of the season. The declared insurance
price must cover at least the cost of producing rice
for the farmers. Then, the farmers have the right to
buy insurance from the government.
In addition, Chanthaphong and Sirikanchanarak
(2012) stated that the Thai government's
intervention policy in the rice market was not able
to provide the farmers with higher incomes, widely
distribute benefits, or reduce the poverty in rural
areas. In terms of policy proposal, the government
should reduce intervention in the rice market. In
fact, the support price should not be set higher than
the price that farmers can sell in the fields.
Furthermore, the study on the utilization of the
agricultural futures market by Chaokul et al. (2011)
indicated that although the export price, wholesale
price and farm price of rice highly correlated with
the one-month or two-month future price of rice, the
amount of rice traded through this futures market
was only 2.7 percent of the total rice production. As
a result, the futures prices did not reflect the actual
supply and demand of rice. The futures market price
could not be used as a reference price for the actual
market price either. However, Taylor et al. (1996)
found that rice prices in the global market had no
long-term relationship with rice prices in the futures
market at the Chicago Rice and Cotton Exchange
(CRCE), the rice prices in the spot market in Texas,
and the 100% white rice prices from Thailand.
Consistent with John (2013), this paper explained
that the causality tests between the export prices of
rice and the domestic prices of rice regarding the
Thai rice market were not entirely clear.
3 Research Methodology
3.1 The Model
The economic analysis through mathematical
models reflects the relationship between the key
elements in the rice market, comprising farmers,
millers, merchants, rice traders and governments.
Such model will help towards an understanding of
the mechanism of price transmission between
upstream and downstream in this industry. It also
provides a theoretical framework for evaluating the
rice price support policy. In this section, the model
presentation is divided into farmer, rice mill and
merchant sections. It starts with the first part
showing the interaction between the farmer and the
mill which is considered the upstream part of the
rice industry. The second part is a simulation of the
interaction between the rice mill and the rice
merchant which is the mainstream of the rice
industry.
3.1.1 Farmer and Mills
Suppose that there are many people working as
farmers in this economy. Therefore, the number of
peasants in the economy is a very valuable but finite
positive real number. Farmers invest in the rice
cultivation with production costs, that are equal to
00c
. This cost includes the accounting cost and
economic opportunity cost. Once the farmers
harvest the paddy, they will sell their produce to the
buyers where there are two alternatives. The first
one is selling their paddy to the mill for which the
price depends on the negotiation between the
farmers and the mills. The second one is to sell it to
the government scheme. In this analysis, the price
that the state buys from the farmer is assumed to be
equal to
p
. As a result, if a farmer decides to sell
his produce to government projects the price he
receives is
p
. Therefore, the decision of the
peasants can be expressed in the form of a
mathematical equation as follows:
max[ , ]pp
(1)
That is, the farmer will compare the price offer
p
which is received from the mill with the price that
will be obtained from selling to government
projects. If either option gives a higher value, the
farmers will choose that option.
Fig. 1: The policy price from the government
project
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1247
Volume 19, 2022
As can be seen from the figure 1, if the price is
higher than the policy price from the government
project, the farmer accepts the price quoted from the
mills. Therefore, we write a price expression that
farmers can receive as follows:
i
pp
(2)
where
0
i
,
0
i
when
*
pp
,
0
i
when
*
pp
With regard to the problems that the mill faces,
this study assumes that the mill perceives a “signal”
which is the information about the market trends,
weather etc. at the beginning of the period. The
received signal affects the generation of predictions
about the price of rice in that period, then the mill
recognizes the signal and generates a prediction of
the rice price. He further calculates it as the highest
price that it will be purchased from the farmer.
e
p
is
given that the price of rice which the mill is
expected to sell. This modeled economy also
assumes that the price of paddy that the mill is
willing to pay the most for is equal to
e
p
. The
mill acts as an intermediary between the farmer and
middleman (Yong). The mill will buy the paddy to
keep and will reproduce the paddy when he receives
the definite orders from the middleman. Thus, the
rice-mill will bear the cost of collecting the paddy
while waiting for these orders. This required the
creation of a forecast of the selling price. The future
selling price will determine the purchase price of
paddy
()
e
p
again. Once the mill and the farmer
come to an agreement on the rice price, then the
bargaining effect is based on Nash's bargaining
model. In fact, the benefit required by both parties
to compete is
e
pp
. Given
e
A p p
and
further assuming that
0
. Such a parameter
explains the bargaining power of the agreement
made between the mill and the farmer. The problem
of haggling between the mill and the farmer can be
displayed in the form of the calculation
i
that
maximizes the product of the benefit of both parties
as follow.
1
max ( )
iii
A
(3)
The first parenthesis in the above equation shows
the benefits that the mill will gain from this bargain.
The outcome of the mill is to buy the paddy at a
price
i
p
from the farmer
i
which is lower than
the desired floor price
()
e
p
, and
i
shows the
share of the total benefit the farmer receives from
the bargain. This share is weighted by the amount
1
. As a result, the equilibrium price at this stage
of the supply chain is in the range
[ , ]
e
pp
. As the
mill quoted the price which is below the lower
boundary, the farmer will certainly reject this offer,
and he will sell his paddy to the government project
instead. In addition, the highest price that the mill is
regularly willing to buy will be no higher than the
upper boundary price in the range. This reflects the
expected selling price of the mill. The equilibrium
price that the mill and the farmer are willing to trade
is as follows:
(1 ) e
i
p p p
(4)
That is, if the mill has more bargaining power
( 1)
, the price the farmer receives will be
close to the price of government insurance
()
i
pp
. Still, if the peasant has more bargaining
power
( 0)
, he will be able to sell rice at a
price close to the highest price that the mill is
willing to buy
()
e
i
pp
.
3.1.2 Rice-Mill and Merchant
The next part is considered the interaction between
the rice-mill and merchant. This section assumes
that the rice merchant receives the orders and
forwards those to the middleman (Yong) providing
the required amount of rice from the mills. These
orders might come from an international market. or
the domestic market. Thus, the price of the rice
which is set to be purchased from the rice merchant
is equal to
.
W
p
Such a model also supposes that
there are many mills in the economy system. Each
mill buys the rice from local farmers at different
prices, then they also offer the price to the
middlemen at different prices. Therefore, the price
of rice is distributed, which can be expressed by the
distribution function.
( ) 0Fp
, which indicates
the proportion of mills in the market that offer a
lower selling price
p
. The middleman is aware of
this distribution function; in other words, he knows
that there are different mills that offer different
prices in the rice market. If someone offers the
middleman a certain level of selling price, he will
know the chances of getting a lower offer price.
Since the rice merchant receives the price of the
final sale of rice equal to
W
p
, they would not want
to buy rice from the mill at a higher price than
.
W
p
At the same time, the mill has a minimum
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1248
Volume 19, 2022
selling price to be willing to offer
ˆ
()p
; indeed, he
buys the paddy from farmers to keep it for milling.
The processing cost consists of the interest rate,
converting paddy into rice, etc. We assume that the
total cost of the mill concerning the rice purchased
from the farmer
i
is proportional to the purchase
price
()pi
. Hence, the minimum price for a mill to
be willing to offer rice (selling price) to the
middleman would be:
ˆˆ
(1 )
i
p p r
(5)
where
ˆ
r
consists of the interest rate and unit cost of
rice processing. Defining the interaction between
the mill and the rice merchant in this middle market
as a model of Burdett and Judd (1983), so the
middleman would randomly ask the mill for the
selling price and will choose to buy from the mills
who offer the lowest prices. The mill is aware of
middleman's strategy. However, the middleman
makes a bid decision without knowing what offer
the mill has received. Another, such model assumes
that the mill sets the selling price in order to
maximize the expected profit which can be
expressed in a mathematical equation as follows:
1
1
ˆ
( ) (1 ( )) ,
() 0,
kW
k
k
W
p p q k F p if p p
pif p p
(6)
Given that
k
q
represents the probability, the
middleman will randomly ask for a price from the
number of mills
k
. Then the profit the mill expects
to receive from setting the selling price at
p
is
derived from the product of two components. The
first part is the difference between the selling price
and the minimum price
ˆ
()pp
. It means that if the
price
p
is set too high, the difference is greater.
The second part comes from the probability that the
buyer will actually buy rice from the mill. It implies
that if the mill sets the price
p
too high, the chances
of the middleman getting a higher asking price are
much greater. Consequently, the chances of selling
rice are smaller. It means that the higher the price,
the less opportunities to sell rice. and causing the
expected profit to decrease in this second part. The
study of Burdett and Judd (1983) shows that if the
buyer (or middleman) has the cost of randomly
selecting the asking price in equilibrium, the buyer
will randomly choose to ask for prices from no more
than two sellers. More importantly, there is a price
range in equilibrium that gives the seller the same
expected profit. Accordingly, this causes the
distribution of the selling price, and the rice prices
in equilibrium are distributed as follows:
0, ( )
( ) 1 , ( )
ˆ2(1 )
1,
L
WW
L
W
if p p q
p p q
F p if p q p p
p p q
if p p
(7)
Where
ˆ ˆ ˆ
( ) (1 )
W
L
p q q p q p
ˆ2
q
qq
It can be seen that the selling price, arising from
the rice trading between the mill and the
middleman, is in the range
[ ( ), ]
W
L
p q p
because the
price in this range gives the profit expectation.
Therefore, the mill can choose such a price and
always receive the predicted profit. The estimated
price at which the mill will sell rice is equal to
()
W
L
p
p
Ep pdF p
3.2 Price Transmission Process and Policy
Implications
This section analyzes the process of transmission of
prices. arising from external factors and how it
passes the impact onto other levels of prices,
whether it is sent downstream or sent back to the
upstream. The analysis of transmission is considered
in two case studies: the price of rice in the final
stage of supply chain changes; and the case of
increasing the purchase price of paddy by the
government policy.
3.2.1 The Increase in the Price of Rice in the
Final Stage of Supply Chain
Suppose that the price of rice in the world market
has increased. It is obvious that this change affects
the distribution of the price traded between the
middleman and the mill. Once the world price of
rice increases, it causes the distribution function
()Fp
to drop down to the original range. The upper
boundary of selling price will move up accordingly
W
p
and its lower boundary will move up
accordingly
()
L
pq
. Consequently, the width of price
range will increase from the original because the
bottom edge changes only
ˆW
qp
, but the top edge
moves up
W
p
. Thus, the support of these scatter
function changes will result in
Ep
that goes up
significantly. The mill is willing to buy rice from
farmers in a wider price range than before, when the
price at which the mill had expected to sell to the
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1249
Volume 19, 2022
middleman has increased. Therefore, the upper
boundary of the acceptable price range of rice has
moved upwards
w FOB
t t t
p p e
also. The
above analysis shows the transmission of the
downstream price back to the upstream paddy price.
The essential conclusions are consistent with what
appears in the Thai rice market. That is, when the
price of rice in the world market is high in the
production season, the purchase price at which the
mill buys paddy from the farmer will also be pulled
higher. Such a model in this section describes the
mechanism that results in an increase in paddy
prices:
3.2.2 The Government Increases the Purchase
Price of Paddy
This section analyzes the effects of government
policies that intervened in the price mechanism in
the paddy market. Initially, the government
announced that they would buy paddy from farmers
at a higher price. This also assumes that a price
increase is still below the highest price level at
which the mills will be willing to buy paddy from
farmers. Once the government increases the price
support, the price of paddy will increase in
p
. As a
result, the price of trading between the mill and the
farmer goes up accordingly. It also affects the
prices in other parts of the supply chain, namely
ˆ
p
and
()
L
pq
, respectively. In contrast, it may not
affect the price at which the mill sells the rice to the
middleman, if the selling price at which rice is sold
to them is still in the range
[ ', ]
W
L
pp
. In fact, the
mill sells rice to the rice traders at a price
p
at the
beginning, when
[ ( ), ]
W
L
p p q p
. Then, the
government raises the purchase price of rice to
'p
that has an influence on prices in the supply chain to
ˆ
p
and
( )'
L
pq
. However, as the price
p
is still in
the range
[ ( )', ]
W
L
p q p
, it means that the sale of rice
at the same price still makes the anticipated profit.
unchanged. Such a model indicates that the price
characteristics in the midstream have more price
stickiness than in the downstream. Therefore, the
policy implication of this model is: if the
government increases the purchase price of rice
from farmers where the price is lower than
e
p
, it
does not cause the middle stage of trading price to
fall out of the price range
[ ', ]
W
L
pp
. Such a price-
support policy does not affect the export price, and
does not undermine the competitiveness of rice
exporters or cause trouble for the consumers. As
happened in the case of the pledging policy, the
government bought whole grain rice at a higher
price than the world price. Accordingly, the support
price
p
was higher than
W
p
and
e
p
causing the
mills to be unable to buy rice from the farmers.
Finally, the rice supply chain received the negative
impact until the trading process in the free market
failed.
3.3 Rice Support Policy
This section considers the government’s policies to
help the farmers with the method of purchasing
paddy directly or accepting paddy pledges from
farmers. Both policies aim to enable farmers to sell
rice at higher prices. As described in the previous
section, that assumes that the farmer produces rice
at cost
00c
. This section adds more detail to the
model; that is, requiring the farmer to have two
options after harvesting the crop: the first is to keep
the rice in the barnyard to be sold later, and the
second is to sell the rice to the mills immediately.
The further assumption is that the storage cost of
rice kept for sale is
10c
. The farmer with an urgent
need for money will choose to sell immediately,
whereas the farmer who can wait to sell will choose
to store the rice in the barnyard to sell it at a better
price. As a result, the farmer who keeps their rice
has their preference correspond to the following
conditions:
1
0
() 1
pc
U p U r
(8)
1
1
() 1
pc
U p U r
(9)
However, the second group of farmers have
different preferences as in equation 9, which leads
them to choose to sell rice immediately. The
function
()U
represents the utility of the two groups;
0
and
1
represent the different discount factors
between the two groups of farmers. Both equations
want to reflect a situation in which each farmer has
a different urgency to sell their produce. Some
farmers want to sell rice quickly because they need
money urgently, while others can wait to sell their
rice.
p
can be defined as the selling price that the
farmer will receive when the rice is sold later, given
that
q
is the price that the farmer will receive if the
rice is sold immediately. In addition, the future
price will be worth the storage cost and opportunity
cost, such that:
1
(1 )p q r c
(10)
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1250
Volume 19, 2022
With each type of farmer, there is a different
urgency in consumption. Moreover, the price of rice
that each type of farmer will be willing to sell as
soon as the harvest is harvested is not lower than
i
p
, which can be obtained from this condition:
1
( ) , 0,1
1
ii
pc
U p U i
r
(11)
Equation 11 shows that
0
p
is the minimum
price for the type of farmer who wants to sell
immediately and
1
p
is the lowest price for the type
of farmer who can wait to sell it. Hence, the
minimum price of the first group will be lower than
the minimum price of the second group with the
above condition, (
01
pp
).
In actual fact, some farmers have an urgent
need to spend money, and do not want to keep their
product for a long time. They try to instantly sell
rice in exchange for money for shopping; therefore,
these farmers are willing to sell rice at low prices.
This can be seen from the above equation that
shows the factors affecting the selling price of
paddy, which the farmer receives from the sale to
the mill. Whether it is higher than the cost, such a
price depends on how much bargaining power the
farmers have. As the bargaining power of farmers is
low, the mill will be able to force the purchase price
of paddy down in order to be close to the production
cost.
Therefore, the government tries to use a policy to
support the selling price of paddy, or find a way for
the farmers to keep their rice in the barn and sell it
later when the price of rice increases. That is, the
price at which the government will buy must be
higher than
0
p
, and at such a level that farmers who
are not in a hurry would immediately sell their
product.
4 Empirical Results
We presented findings from both primary and
secondary data sources. The primary data used in
this section is from our field survey covering 330
farmers in 11 provinces. The secondary data were
collected from various sources, including the Office
of Agricultural Economics, the Department of
Agriculture, the Department of Agricultural
Extension, the Rice Research Center, the National
Statistical Office, and the Bank of Thailand (BOT).
4.1 Findings from Field Survey
We conducted a field survey using quesionaires to
gain insights about farmers’ characteristics, rice
production, paddy stock management, product cost,
farmer revenue, and rice prices, etc. in key rice-
growing provinces
2
all over Thailand. There are 11
provinces in our servey and we sampled 30 farmers
from each province to cover every sub-district,
resulting in 330 farmers in total.
Regarding the cost of rice production, the total
cost in the 2013 production year averaged 4,124.44
baht per rai (128.89 US$), an increase of 30.77%
compared to the total cost in the production year
2008, which averaged 3,153.88 baht per rai (98.56
US$). The structure of rice cultivation cost in both
years are similar. That is, the chemical fertilizer
costs, rental costs, harvest costs, rice seed cost, and
soil preparation costs occupied around 70 percent of
the total cost.
Meanwhile, the rice farmers were able to sell rice
in the 2013 production year at an average price of
9,304.71 baht per rai (290.77 US$). Compared to
the average selling price of 5,466.76 baht per rai
(170.84 US$) in 2008, the paddy price in 2013
increased by 70.21 %. On average the farmer
earned more revenue, net of cost, per rai around
5,180.27 baht in 2013, compared to merely 2,312.88
baht per rai (72.28 US$) in 2008.
The survey also found that the paddy price in
each location is primarily determined by the rice
mills, the central paddy markets, and the local
middlemen. This reflects the bargaining power of
the local traders in paddy price determination. This
influence remained intact even in the period when
the government implemented the paddy pledging
scheme. The interview with the local middlemen
reveals that the offer price is determined solely by
the export price or world price. Thus, neither cost
of production or the government-price-floor has
significant influence on the wholesale paddy price.
4.2 Econometrics Model of Rice Price
Determination
In this section, we applied Engel-Granger
Cointegration test to investigate the long-term
relation between the domestic paddy price and the
world rice price, represented by the FOB export
price, during 2002-2015. Once we are able to
2
Provinces in our sample include Pathumthani, Singburi,
Chainat, and Suphanburi from the Central region,
Nakhonsawan, Kamphaengphet, Phisanulok, and Phichit
from the North, Surin from the South , Khonkaen, and
Roiet from the North-east.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1251
Volume 19, 2022
establish the long-run relationship between prices,
we construct the error-correction model to analyze
the short-term movement of the domestic and the
world price.
To test the long-run relationship between two
time series data, one needs to check first if both
series are non-stationary, or contain unit root.
Therefore, we applied the Augmented Dickey-Fuller
method to test whether the domestic paddy price and
the world price are non-stationary. The test method
is performed by estimating the coefficients in the
characteristic equation as shown below.
11
I
t t t i t
i
y y y
(12)
That is, the coefficient of the lagged value of
variable of interest,
, is estimated and test whether
it is significantly different from zero. If we are
unable to reject such null hypothesis, then that time-
series is non-stationary. Our econometric test found
that the estimated coefficients are not significantly
different from zero for the wholesale price as well
as the world price. Therefore, both series are non-
stationary.
The next step is to test whether the two series
have a long-run relationshop, or test whether they
are cointegrated. The Engle-Granger Cointegration
test applied the Augmented Dickey-Fuller method to
test the stationarity property of the OLS residuals
from the following regression.
w FOB
t t t
p p e
(13)
If it is found that
t
e
is stationary, then the
wholesale paddy price
()
w
t
p
is cointegrated with
the world price
()
FOB
t
p
. In other words, both prices
have a long-term relationship, thus; they tend to
move together.
Table 1 below reports the estimation of the
cointegration equation 12. The coefficient of the
domestic paddy prices moves in line with FOB
export prices, as shown in the positive coefficient,
and are significantly different from zero.
Table 1. The result of coefficient estimation of
regression model of the paddy price
Constant
FOB Price
Paddy price
56.4
0.48
SE
(1.88)
(0.008)
N = 144
20.96R
F(1,142) = 359.06
Source: Calculation
The Augmented-Dickey-Fuller test on the
residual series,
t
e
, showed that the series is
stationary, which implies that the domestic price and
the world price have long-term relationship. Thus,
the observed comovement of the two price series are
not spurious.
We are able to construct a model describing
short-term movement of cointegrated variables. In
case that one has interest in predicting the short-
term variation of the wholesale paddy price, one can
construct an error-correction model as follows:
01
11
ˆ
( ) ( ) ( )
nn
w w FOB e
t j t j h t h t t
jh
p p p e p
(14)
The error-correction model for the wholesale
price states that the determinants of the short-term
movement can be decomposed into three parts, i.e.,
its own past
()
w
tj
p
, past values of the world price,
()
FOB
th
p
, and the adjustment towards the long-run
equilibrium,
1
ˆt
e
. It should be noted here that 𝛾
indicates the speed of adjustment towards long-term
equilibrium so that its value is negative. The result
of estimating the coefficients in the above equation
are shown in Table 2.
Table 2. The result of the error-correction model of
the world rice price.
Estimated
coefficient
Standard
Error
t-stat
1
ˆt
e
-0.327
0.816
-4.00
1
w
t
p
0.298
0.13
2.30
1
FOB
t
p
0.09
0.11
0.84
0
0.002
0.004
0.48
Source: Calculation
The error-correction model has brought
adaptation to return to long-term equilibrium,
helping to explain the adjustment in domestic paddy
prices. As can be seen from the estimated
coefficient
1
ˆt
e
, the deviations from long-term
relationships that occur in previous periods are
corrected and gradually adjusted to long-term
equilibrium, with one-third (-0.327) of that
deviation eliminated based on the coefficient of
1
ˆt
e
. Comparing the two domestic paddy prices, the
finding displays that the rice price obtained from the
error-correction model is close to the actual data.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1252
Volume 19, 2022
This shows an appropriate model to explain price
changes.
4.3 Expectation and Price Discovery in
Thailand White Rice Market
Our model in the previous section shows that the
equilibrium paddy price is determined in part by the
expected selling price of the millers. To investigate
such relationship, we chose the option price of white
rice 5% (BWR5) in the Agricultural Futures
Exchange of Thailand (AFET) , both put and call
prices, as a representation of
e
p
in equation (4).
Under the efficient market hypothesis, the future
market plays a crucial role in price discovery
process since it digests all the available information
about future price and reflects what market
participants foresee about the future paddy prices.
In our empirical study here, we use the wholesale
price of white rice 5% in the Bangkok market
(WP5) for
p
and the price of white rice 5% in the
agricultural futures market (BWR5) for
e
p
in
equation (4). Both series are tested for stationary
property by using the daily price data between April
2, 2007 and September 14, 2011. The unit root test
results show that both series contain unit root. A
further investigation reveals that the first-difference
of both series are stationary. Thus, the Granger
Causality test is applied to find the causal
relationship between the changes in wholesale price
of white rice 5% in the Bangkok market and the
changes in the corresponging option prices in the
AFET.
The test results reject that the null hypothesis that
the price of white rice 5% with both options in the
AFET does not Granger cause on the wholesale
price of white rice 5% when the 1-day, 2-day, 3-day
and 5-day lag are included. This result supported
the conjecture of our model in Section 3 that the
equilibrium wholesale price is determined partly by
the expectation of the millers’ selling price.
However, the causality test in the opposite direction
also found that the option prices in the AFET
Granger causes the wholesale white rice price.
Therefore, we concluded that information flow back
and forth between markets in such a way that the
prices in spot and future market has bi-directional
influence on each other.
Despite the bi-directional relation, we found that
the spot and future prices have a long-term
relationship. This finding is obtained from the
cointegration test reported in Table 3. We ran the
OLS regression between WP5 and BWR5 and
conducted the ADF test on the residual of the
regrssion. It is found that the residual series is
stationary, implying the spot and futures prices of
white rice are cointegrated.
Table 3. The result of coefficient estimation of
regression model of wholesale price of white rice
5%
Constant
Price of white
rice 5% with
both options
Wholesale price of
white rice 5% (WP5)
0.4083
0.9448
Std. Error
(0.0599)
(0.0037)
t-Statistic
N = 1051
[6.8179]
[254.4294]
20.98R
F(1,142)
=4734.30
Durbin-Watson
stat= 0.4071
Source: Calculation
An additional step for the price analysis of rice is
to find what exactly determines the rice price in
Thai market as follows:
01
11
ˆ
55
t i t i l t l t t
il
WP WP BWR u
(15)
Equation 15 presents an error correction model
which is applied to explain the short-term
relationship between both types of prices. The result
shows that when the spot price of white rice 5%
(WP5) temporarily deviates from the long-term
equilibrium, the adjustment to restore the
equilibrium is swift, as seen in the estimation of
.
Table 4 below shows that the error-correction
parameter is estimated to be -12.81, implying
that 12.81 percent of deviation from L-R
equilibrium is eliminated within a day.
Table 4. The result of the error-correction model of
the wholesale price of white rice 5%.
Estimated
coefficient
Standard
Error
t-stat
1
ˆt
e
-0.1281
0.0206
-6.2274
1
5t
WP
0.2117
0.0376
5.6228
2
5t
WP
-0.200
0.0365
0.5483
1
5t
BWR
0.1360
0.0315
4.3149
2
5t
BWP
-0.0488
0.0312
-1.5658
0
0.0037
0.0082
0.4459
Source: Calculation
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1253
Volume 19, 2022
5 Conclusion
Though the rice market in Thailand is linked tightly
with the world market, various governments
intervened the domestic market from time to time.
The most notable intervention policy in recent times
in the 2011 rice price-pledging scheme, of which
offered to buy unlimited paddy at the premium of
40-50 percent above the world price. The program
successfully raised the domestic paddy price above
the would-be equilibrium level under lessez faire.
However, the distortionary scheme could possibly
left its mark on the price structure of Thailand for a
long while. This paper attempts to investigate
whether the Yingluck Shinawatra government
policy could affect the subsequent price structure in
the rice market by using data from both primary and
secondary sources. The findings from our field
survey show that the paddy price was mainly
determined by the rice mill, the central paddy
market, the middleman, and the exporters. The
government intervention policies could influence
the paddy price temporarily, i.e., when the scheme
was in place. In the long-run equilibrium, the paddy
price is determined by a combination of the price at
which the mills are willing to buy and the world
price.
Using long time-series on the wholesale level and
the FOB price allows us to conduct a cointegration
test, to see whether there exists a long-run
relationship between the wholesale paddy price and
the FOB export price. The empirical findings
confirm the field survey finding, i.e., there is a long-
run relationship between the domestic wholesale
price and the world white rice price. Moreover, the
price expectation embeded in the millers’ offer price
played significant role in the price discovery
process. The causality test revealed that the
expected future price causes the spot wholesale
price in the Granger sense. It is also found that the
causation is bi-directional implying that the flows of
information between markets are essential in
determining the equilibrium price in the rice market.
This study has unearth the linkage of rice price in
various layers and found that distortionary price
support policies did not alter the determination of
rice price in the longer term.
References:
[1] Burdett, K. & Judd, K. L. (1983). Equilibrium price
dispersion. Econometrica, 51(4), 955-969.
[2] Chaowakul, M., Bamrungthai, N., Saenthaweesuk,
K., & Phahuyuth, D. (2011). Review of the structure
of the Thai rice market (Research report). Bangkok:
Thailand Science Research and Innovation.
[3] Chanthaphong, S., & Sirikanchanarak, D. (2015).
Insight into Thai rice management in terms of
income distribution and production efficiency.
Bangkok: Bank of Thailand.
[4] Engle, R. F., & Granger, C. W. (1987). Co-
integration and error correction: representation,
estimation, and testing. Econometrica, 251-276.
[5] Ghosh, M., Whalley, J. (2004). Are price control
necessarily bad? The case of rice in Vietnam.
Journal of Development Economics, 73, 215-232.
[6] Iswilanon, S.,& Kaoian, S. (2009). The dynamics of
the production economy of Thai rice and its forward
looking. Bangkok: Department of Agricultural and
Resource Economics. Faculty of Economics
Kasetsart University.
[7] Isavilanon, S. (2010a). Economic Dynamics of Thai
Rice. Bangkok: Department of Economics and
Resources, Faculty of Economics, Kasetsart
University.
[8] Isawanon, S. (2010b). Public Policy on Rice, Chiang
Mai: The Institute of Public Policy Studies. Chiang
Mai University.
[9] John, A. (2013). Price relations between export and
domestic rice markets in Thailand. Food Policy, 42,
48-57.
[10] Puapongsakorn, N., & Jarupong, J. (2010).
Measures of Rice Market Intervention for Corruption
Prevention: Seeking Surplus Compensation and
Political Economy of Paddy Pledge Project.
(Research Report). Bangkok:Thailand Development
Research Institute Foundation.
[11] Puapongsakorn, N., Thitaphiwattanakul, B.,
Tokritsana, R., Sirisuphalak,P., Nititanpraphas, I.,
[12] Naphasinthuvong, O., Suphanchat, W., Saenglertsai,
S., Udomsophakit, S., Latthapipat, D.,
Anuchitworawong, C., Srianan, N., Methasurarak,
S., Puntakua, K., & Petsrichuang W. (2013). Thai
rice strategy: Thai rice research, development and
look forward (Research Report). Bangkok: Thailand
Development Research Institute.
[13] Sirikanchanarak, D. (2011). Preventive measures for
the price of paddy in Thailand. Bangkok: Economic
Research Department Bank of Thailand.
[14] Taylor, E. L., Bessler, D. A., Waller, M. L., &
Rister. M. E. (1996). Dynamic relationships between
US and Thai rice prices. Agricultural Economics, 14,
123-133.
[15] Thailand Development Research Institute. (2009).
Guidelines for Risk Insurance of Paddy Price.
Bangkok.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.110
Pithak Srisuksai, Vimut Vanitcharearnthum
E-ISSN: 2224-2899
1254
Volume 19, 2022
