Content uploaded by Witsanu Attavanich
Author content
All content in this area was uploaded by Witsanu Attavanich on Dec 11, 2015
Content may be subject to copyright.
The Effect of Climate Change on Thailand’s Agriculture
Witsanu Attavanich1
Abstract
Agriculture is potentially affected by climate change especially in developing countries where
the agricultural sector plays a crucial role including Thailand. The objectives of this study are to
analyze the effect of climate change on Thailand’s agriculture and investigate implications for
greenhouse warming under future climate change scenarios using the Ricardian approach
allowing a variety of the adaptations that farmers make in response to changing economic and
climate conditions. The study finds that both temperature and precipitation significantly
determine farmland values. Summer temperature, precipitation in the early rainy and summer
season negatively affect the farmland values, while winter temperature, precipitation in the late
rainy and winter season enhance the farmland values. Overall, the projected negative impacts of
climate change on Thailand’s agriculture during 2040-2049 range from $24 to $94 billion. By
downscaling the analysis to the province level, this article finds that western, upper part of
central, and the left part of northern regions are projected to be better off, while southern, eastern
regions, lower part of central, and the right part of northern regions is projected to be worse off.
Key words: Thailand’s agriculture, climate change, Ricardian analysis, regional climate model,
farmland value
1 Witsanu Attavanich is a lecturer in the Department of Economics, Faculty of Economics,
Kasetsart University. Correspondence to be sent to Email: attavanich.witsanu@gmail.com.
1. INTRODUCTION
Recent studies, including those by the Intergovernmental Panel on Climate Change (IPCC)
(2001a; 2001b; 2007a; 2007b), indicate that greenhouse gas (GHG) emissions and resultant
atmospheric concentrations have led to changes in the world’s climate conditions, such as
increases in temperatures, extreme temperatures, droughts, and rainfall intensity. Such changes
are expected to continue and agriculture is potentially the most sensitive economic sector to
climate change, given that agricultural production is highly influenced by climatic conditions
(e.g., IPCC 2007b; Mendelsohn, Nordhaus, and Shaw 1994; Deschenes and Greenstone 2007;
McCarl, Villavicencio, and Wu 2008; Schlenker and Roberts 2009). Compared with developed
countries, developing countries are more vulnerable to climate change since they are already in a
hot climate zone, depend on labor-intensive technologies with fewer adaptation opportunities,
and a majority of people in these countries rely heavily on the agricultural sector (Mendelsohn et
al. 2001).
Thailand is one of developing countries that agriculture plays a crucial role. For example,
in 2011 the agricultural sector employed about 14.88 million people, accounting for 38.7 percent
of the Thai labor force (National Statistical Office Thailand 2012) and agricultural activities
generate about $40 billion, which contributed to 12.8 percent of the gross domestic production
(Office of the National Economic and Social Development Board 2012). Thailand is also a major
exporter for many agricultural commodities such as rice, natural rubber, and cassava. Therefore,
climate change impacts on agriculture are expected to significantly affect the economy and the
livelihood of the people in this country.
The objectives of this study are to analyze the effect of climate change on Thailand’s
agriculture and investigate implications for greenhouse warming under future climate change
scenarios on Thailand’s agricultural sector using the Ricardian approach firstly proposed by
Mendelsohn et al. (1994). Although there are many studies (Office of Environmental Policy and
Planning 2000; Buddhaboon, Kongton, and Jintrawet 2005; Pannangpetch et al. 2009;
Isvilanonda et al. 2009) analyzing the effect of climate change on Thailand’s agriculture, most
studies (except for Khamwong and Praneetvatakul 2011) use the traditional production function
approach, which potentially overestimates the damage from climate change since the model
allows little adaptation of farmers (Mendelsohn et al. 1994).
3
The current article differs from Khamwong and Praneetvatakul (2011) in several aspects.
First, it expands the scope of the previous research from the northeast region to cover entire
country. Second, the current paper uses the finer scale of dataset, a farm-level dataset, which
could improve the estimated results since it reflects the farmer’s decision regarding the climate
adaptation strategies more than the use of provincial-level data. Third, the current article
employs the land value as a dependent variable similar to the original Ricardian approach, which
could address the potential problem from the use of the annual net farm revenue. Fourth, more
constructed important explanatory variables determining the farmland values collected from
various sources are included to address the problem of endogeneity bias. Lastly, the current
article projects the impacts of climate change under climate scenarios using the unique dataset
from regional climate models.
2. LITERATURE REVIEW
In view of its importance to economic well-being, effects of climate change on agriculture have
been well research and documented, dating back at least 25 years (e.g. Smith and Tirpak 1989;
Mendelsohn et al. 1994; Adams et al. 1999; Reilly et al. 2003; McCarl et al. 2008; Attavanich et
al. 2013; and various IPCC reports). Overall, the effect of climate change on agriculture is mixed
in developed countries, but negative impacts are found in developing countries. Moreover, in a
country, the damage is heterogeneous across regions.
Using an agricultural sector model, Adams et al. (1999) find that agricultural welfare
strictly increases in the United States (U.S.) with a 1.5°C warming and further warming could
decrease this benefit at an increasing rate. The welfare gain from a 1.5°C warming with 7 percent
precipitation is $55 billion in 2060. Further warming by 2.5°C could reduce these benefits to $47
billion. With similar approach, Reilly et al. (2003) estimated the net effect in terms of economic
welfare of the combined changes in crop yields including adaptation and CO2 fertilization
effects, water supply, irrigation demand, pesticide expenditures, and livestock effects was
generally positive. The increase in economic welfare was ranged from $0.8-$7.8 billion in 2030
and $3.2-$12.2 billion in 2090. U.S. producers generally suffered income losses due to lower
commodity prices while consumers gained from these lower prices.
Using the Ricardian analysis, Mendelsohn et al. (1994) find that higher temperatures in
all seasons except autumn reduce average U.S. farm values, while more precipitation outside of
4
autumn increases farm values. They estimate that a climate change induced loss in U.S. farmland
value ranging from -$141 to $34.8 billion. Schlenker, Hanemann, and Fisher (2005) do a similar
study and find an annual loss in U.S. farmland value in the range of $5-$5.3 billion for dryland
non-urban counties. Mendelsohn and Reinsborogh (2007) find that U.S. farms are much more
sensitive to higher temperature than Canadian farms and but are less sensitive to precipitation
increases. Deschenes and Greenstone (2007) find that climate change will lead to a long run
increase of $1.3 billion (2002$) in agricultural land values. They indicate that agricultural land
values in California, Nebraska, and North Carolina will be lowered substantially by climate
change, while South Dakota and Georgia will have the biggest increases.
For developing countries, Seo and Mendelsohn (2008) find that in South America climate
change will decrease farmland values except for irrigated farms. Moreover, they find small farms
are more vulnerable to the increase in temperature, while large farms are more vulnerable to
increases in precipitation. Mendelsohn, Arellano-Gonzalez, and Christensen (2010) project that,
on average, higher temperatures decrease Mexican land values by 4 to 6 thousand pesos per
degree Celsius amounting to cropland value reductions of 42-54% by 2100. Wang et al. (2009)
find that in China an increase in temperature is likely to harm rain-fed farms but benefit irrigated
farms. A small value loss is found in the Southeast China farms, while the largest damage is
discovered in the Northeast and Northwest farms (Wang et al. 2009).
In Thailand, several studies have investigated the effect of climate change on agriculture.
By using the Crop Environment Resource Synthesis (CERES) model, Office of Environmental
Policy and Planning (2000) reveals that rice grown under rainfed conditions was found to be
highly vulnerable to climate change. Moreover, yields of rice and maize are projected to decline
as much as 57 and 44 percent as compared to the baseline, respectively, although their
magnitudes vary depending on climate conditions, soil types, and crop practices. Their results are
different from Pannangpetch et al. (2009) who employ the Decision Support System for Agro
Technology Transfer (DSSAT) model to analyze the impacts of global warming on rice,
sugarcane, cassava, and maize production in Thailand. They find a little impact of rising
atmospheric CO2 concentration and temperature on the rice, sugarcane and maize production.
However, cassava production may drop as much as 43 percent as compared to the baseline.
Buddhaboon, Kongton, and Jintrawet (2005) simulate the effect of climate change on
KDML 105 rice yield in Tung Kula paddy field by direct seeding method and set CO2
5
concentration at 1.5 and 2.0 times of year 1980-1989 (normal year) in the period of 2040-2049
and 2066-2075, respectively. They reveal that climate change likely enhances overall rice yield.
Similar finding is founded in Isvilanonda et al. (2009) who conclude that climate change
enhances KDML 105 rice yield in the north-eastern and the northern regions using CropDSS
simulation model. Using changes in yields data, they find that the total production of KDML 105
is project to increase approximately 1.4 million ton, which is equivalent about 14,195 million
baht. However, Isvilanonda et al. (2009) project that climate change could adversely affect
Suphan Buri 1 rice yield in central plain with a reduction of about 0.249 million ton, creating a
loss in value approximately 2,029 million baht. Unlike previous studies, Khamwong and
Praneetvatakul (2011) apply the Ricardian model with province-level data to analyze the impacts
of climate change on agriculture in northeast region. They find that rising temperature in summer
and early rainy season and increased rainfall at the end of the rainy season decrease net farm
revenue. On the other hand, increased rainfall in summer and early rainy season increases net
farm revenue.
According to the above studies, we can classify the model used to analyze the impacts of
climate change on agriculture into three categories (Mendelsohn et al. 1994; Wang et al. 2009)
consisting of: 1) Traditional production function approach (e.g., Smith and Tirpak 1989); 2)
Ricardian approach (e.g., Mendelsohn et al. 1994); and 3) Agro-economic approach (e.g., Adams
et al. 1999). Mendelsohn et al. (1994) criticize that the traditional production function approach
has a serious drawback since the model tends to overestimate the damage from climate change
by omitting a variety of adaptations that farmers can make in response to changing economic and
environmental conditions. While the agro-economic approach incorporates the climate change
adaptation of farmers, they are difficult to build especially in the developing countries due to
data availability and complexity of the model.
Mendelsohn et al. (1994) then introduce the Ricardian approach to bridge the gap
between the traditional production function approach and the agro-economic approach. Recently
the Ricardian approach is gaining popularity. This approach is applied to both developed
countries such as U.S. and Canada (e.g., Mendelsohn et al. 1994, 2001; Reinsborough 2003) and
developing countries such as Brazil, India, Sri Lanka, and China (Dinar et al. 1998; Kumar and
Parikn 2001; Mendelsohn et al. 2001; Seo, Mendelsohn, and Munasinghe 2005; Wang et al.
2009).
6
3. THEORETICAL FRAMEWORK
The Ricardian approach developed by Mendelsohn, Nordhaus, and Shaw (1994) is the primary
method that we use in this paper. In contrast to the traditional production function approach, the
Ricardian approach allows a variety of the adaptations that farmers make in response to changing
economic and climate conditions. By not permitting a complete range of adjustments, previous
studies could overestimate damages from climate change. Instead of studying yields of specific
crops in the traditional production function approach, the Ricardian approach examines how
climate in different locations affects the net rent or value of farmland. By directly measuring
farmland values, the approach account for the direct impacts of climate on yields of different
crops as well as the indirect substitution of different inputs, introduction of different activities,
and other potential adaptations to different climates.
The Ricardian approach assumes that each farmer maximizes income subject to the
exogenous conditions of their farms. Specifically, the farmer chooses the inputs and the
combination of crop and/or livestock, indexed by j, which maximizes net revenue for each unit
of land:
ij
Jj ijiiiijij
Jj ijiXPSHCXQPMax ),,(
(1)
where
i
is the net revenue of farm i,
ij
P
is a vector of input and output prices,
ij
Q
is the
production function for each crop or livestock j,
ij
X
is a vector of endogenous input choices
such as seeds, fertilizer, pesticides, irrigation, hired labor and capital,
i
C
is a vector of climate
variables,
i
H
is a vector of economic control variables and
i
S
is a vector of soil characteristics.
Differentiating equation (1) with respect to each input identifies the set of inputs that
maximizes net farm revenue. The resulting locus of net revenues for each set of exogenous
variables is the Ricardian function shown in equation (2). It describes how net revenue will
change as exogenous variable change.
),,(
*ijiii PSHC
(2)
7
If land is traded in the perfectly competitive market, the land value (
V
) will be equal to
the present value of the net revenue of each farm shown in equation (3).
0
*dteV rt
i
(3)
where r is the interest rate.
The welfare impact (W) of climate change is calculated by computing the difference
between the value of farmland under the new climate (B) and the value of farmland under the
current climate (A) as illustrated in equation (4).
it
iAitBittLCVCVW )()(
(4)
where
it
L
is the amount of land at period t of farm i.
4. METHODOLOGY
Empirical Estimation
To answer the first objective and capture the expected non-linear relationship between the
farmland value and climate, this article specifies the following model to examine the impacts of
climate change on farmland values in Thailand:
kkkssssssss eZdPPTTV 2
43
2
210
(5)
where the dependent variable, V, is the farmland value (dollar) per rai2, T and P represent a
vector of seasonal temperature and precipitation variables, s is season including: winter
(November-January); summer (February-April); early rainy (May-July); and late rainy (August-
October), Z is a vector of relevant control variables capturing characteristics of principal operator
(level of education), farm characteristics (irrigation status, whether the farm has main total sales
from crop or livestock, whether the farm has the flood problem, and whether the farm has the
problem of steep slope), soil conditions (whether the farm has the problem of soil salinity and
sandy soil) and location characteristics (district-level population density, whether the farm
2 1 rai is equal to about 0.395 acre.
8
locates in the plain area, Euclidian distance of the farm to the city of the province in which farm
is located, and percent of agricultural land to total land area for province in which farm is
located) , and e is an error term.
Uncertainty of Climate Change Impacts
To answer the second objective of the study, this study investigates the implication for
greenhouse warming on Thailand’s agriculture by employing our estimated coefficients in
equation (5) together with future climate projections during 2040-2049 simulated by PRECIS
(Providing REgional Climates for Impacts Studies) regional climate model and used Global
Circulation Model (GCM) ECHAM4 dataset as initial data for calculation. The simulation covers
IPCC emission scenarios A2 and B23, which could account for the upper and lower limit of
climate change impacts on Thailand’s agriculture.
5. DATA
Data used in this study are collected from various sources. For the most part, the data are from
the 2011/2012 national agricultural household socio economics survey at the farm level with
6,701 completed farms sampled across 76 provinces from Office of Agricultural Economics. 331
out of 6,701 farms, or about 5%, have been removed from calculation to address the outliner
problem4 and incomplete data on farmland values. In total, we have 6,370 farms. The gathered
data consist of: the estimated current market value of farmland including building expressed in
dollars per rai; education level of the principal operator; soil conditions; whether the farm has the
problem of steep slope and flood problem; irrigation status; whether the farm has main total sales
from crop.
3 A2 scenario is characterized by: a world of independently operating, self-reliant nations; continuously increasing
population; and regionally oriented economic development. B2 scenario are more ecologically friendly. The B2
scenario is characterized by: continuously increasing population, but at a slower rate than in A2; emphasis on local
rather than global solutions to economic, social and environmental stability; and intermediate levels of economic
development (IPCC 2007a).
4 We have found that several farms located in the urban area, especially in large city such as Bangkok, Nonthaburi,
and Chiangmai provinces have very high land prices per rai with small income generated from agricultural activities.
Including these farms in the estimation could bias the impacts of climate change on overall Thailand’s agriculture.
9
Climate data are obtained from Thailand Meteorology Department, which gathers data
from 76 meteorological stations throughout Thailand. The data include information on monthly
temperature and precipitation from 1971 through 2012. Since the purpose of this study is to
forecast the impacts of climate changes on agriculture, we focus on the long-run impacts of
temperature and precipitation on agriculture, not year-to-year variation weather. We
consequently analyze the “normal” climatological variables—the 42-year average of each
climate variable for every station during 1971-2012. To capture seasonal effects of climate on
agriculture, we construct the seasonal climate variables divided into four seasons including:
winter (November-January); summer (February-April); early rainy (May-July); and late rainy
(August-December). In order to link the agricultural data which are organized in the farm-level
and the climate data which are organized by station, we assign these climate variables to each
farm that locates close to the climate station using the nearest distance criterion even though the
farm locates in different province from climate station.
To account for location characteristics and potential of land for non-agricultural
development, we collect several variables including: district-level population density; whether
the farm locates in the plain area; Euclidian distance of the farm to the city of the province in
which farm is located; and percent of agricultural land to total land area for province in which
farm is located from various sources mainly from the National Statistical Office, Ministry of
Agriculture and Cooperatives, Ministry of Interior, and Google Earth. Lastly, data of climate
projections mentioned in the methodology section are collected from Center of Excellence for
Climate Change Knowledge Management, Chulalongkorn University.
Table 1 summarizes variables used in the estimation, their definitions, and summary
statistics across the full sample of farms. For example, on average the farmland in Thailand has
its value equal to $2,945 per rai. The monthly temperatures averaged during 1971-2012 in the
early rainy, late rainy, and summer seasons are around 27.693-28.844 degree Celsius, while in
the winter the month temperature drop to 24.778 degree Celsius. Late rainy season has the
highest level of monthly precipitation equal to 211.923 millimeters, while winter season has the
lowest level of monthly precipitation equal to 44.591 millimeters.
10
Table 1. Description of variables and summary statistics
Variable
Definition of Variables
Mean
Std. Dev.
(N=6,370)
farm value
Estimate of the current market value of farmland
2,945
2,948
including building (dollars per rai)
early rainy temperature
Normal monthly mean temperature (°C) from 1971
28.844
0.568
to 2012 during the early rainy season (May-July)
late rainy temperature
Normal monthly mean temperature (°C) from 1971
27.693
0.530
to 2012 during the late rainy season (August-October)
summer temperature
Normal monthly mean temperature (°C) from 1971
28.358
1.003
to 2012 during the summer season (February-April)
winter temperature
Normal monthly mean temperature (°C) from 1971
24.778
1.428
to 2012 during the winter season (November-January)
early rainy precipitation
Normal monthly precipitation temperature (mm) from
182.673
71.772
1971 to 2012 during the early rainy season (May-July)
late rainy precipitation
Normal monthly precipitation temperature (mm) from
211.923
59.194
1971 to 2012 during the late rainy season (August-October)
summer precipitation
Normal monthly precipitation temperature (mm) from
49.436
18.742
1971 to 2012 during the summer season (February-April)
winter precipitation
Normal monthly precipitation temperature (mm) from
44.591
91.192
1971 to 2012 during the winter season (November-January)
education
Whether the principal operator graduated at least grade 9
0.110
0.312
(equal to 1 if yes)
salt soil
Whether the farm has the problem of soil salinity
0.011
0.104
(equal to 1 if yes)
sandy soil
Whether the farm has the problem with sandy soil
0.018
0.133
(equal to 1 if yes)
steep slope
Whether the farm has the problem of steep slope
0.013
0.115
(equal to 1 if yes)
flood problem
Whether the farm has the flood problem (equal to 1 if yes)
0.117
0.321
irrigate
Whether the farm is the irrigated farm (equal to 1 if yes)
0.254
0.435
main sale from crop
Whether the farm has main total sales from crop
0.789
0.408
(equal to 1 if yes)
plain
Whether the farm locates in the plain area
0.712
0.453
(equal to 1 if yes)
distance
Euclidian distance, in kilometers, of the farm to the city of
42.112
31.153
the province in which farm is located
population density
Population density per square kilometer for district in
170.852
195.706
which farm is located
agricultural land
Percent of agricultural land to total land area for province
54.271
18.084
in which farm is located
Note: Values in Baht are converted with the exchange rate of 32 Baht/US.
11
6. EMPIRICAL RESULTS
As the Ricardian approach estimates the importance of climate and other variables on farmland
values. Table 2 provides the regression results by regressing farmland values on variables of
climate, soil, operator characteristics, farm characteristics, and location characteristics to
estimate the best-value function across different farms in Thailand. There are 6,370 cross-
sectional observations. In order to give a sense of the importance of nonfarm variables in the
model, we begin with a model that contains only climate variables. Specification 1 in Table 2 is a
quadratic model that includes the eight measures of climate. For each variable, linear and
quadratic terms are included to reflect the nonlinearities that are apparent from field studies. For
Specification 2, we include variables capturing characteristics of operator, soil, farm, and
location to control for other factors influencing farmland values.
Overall we find that all climate variables statistically affect farmland values and their
squared terms are significant, implying that the observed relationships are non-linear as found in
the field studies. However, some of the squared terms are positive (i.e., early rainy temperature,
winter temperature, early rainy precipitation, and summer precipitation) implying that there is a
minimally productive level of temperature and precipitation in that season and that either more or
less temperature and precipitation will raise farmland values. The negative coefficient of squared
terms implies that there is an optimal level of a climate variable from which the value function
decreases in both directions.
The overall impact of climate as measured by the marginal impacts evaluated at the mean
level of each variable is provided in Table 3. In general, we discover that higher summer
temperatures and higher early rainy and summer precipitation are harmful for crops, while higher
winter temperatures and higher late rainy and winter precipitation are beneficial for crops. The
higher summer temperatures by 1°C decrease the farmland values equal to $479 per rai, while
higher winter temperatures by 1°C increase the farmland values equal to $299 per rai. For
precipitation, higher early rainy and summer precipitation by 1 millimeter decrease the farmland
values equal to $7 and $28 per rai, respectively. On the other hand, the farmland values will be
increased $11 and $18 per rai, as late rainy and winter precipitation increase by 1 millimeter,
respectively.
12
Table 2. Regression models explaining farm values
Specification 1
Specification 2
farm value
Coef.
Std. Err.
Coef.
Std. Err.
Constant
-471,422.50***
129,626.90
-556,252.10***
136,264.00
early rainy temperature
-23,327.83**
11,453.95
-36,941.65***
11,141.28
early rainy temperature^2
404.93**
198.99
634.68***
193.69
late rainy temperature
56,029.68***
16,266.61
79,825.58***
16,507.53
late rainy temperature^2
-1,014.24***
293.46
-1,440.79***
297.42
summer temperature
15,099.00***
2,819.34
7,444.66**
2,982.07
summer temperature^2
-274.63***
49.14
-139.94***
52.05
winter temperature
-14,180.45***
2,237.70
-9,004.27***
2,292.01
winter temperature^2
292.93***
45.18
188.91***
46.21
early rainy precipitation
-13.78***
4.22
-11.59***
4.18
early rainy precipitation^2
0.02**
0.01
0.01*
0.01
late rainy precipitation
25.58***
6.44
21.61***
6.37
late rainy precipitation^2
-0.03***
0.01
-0.03***
0.01
summer precipitation
-55.08***
11.73
-50.94***
11.57
summer precipitation^2
0.22***
0.09
0.23***
0.08
winter precipitation
25.64***
3.27
21.80***
3.27
winter precipitation^2
-0.05***
0.01
-0.05***
0.01
education
449.55***
116.90
soil salt
-521.69**
237.05
sandy soil
-88.86
251.31
steep slope
-900.84***
251.58
flood problem
-183.36*
106.27
irrigate
223.40**
93.32
main sale from crop
-1,010.18***
95.29
plain
-251.42**
103.08
distance
-5.73***
1.13
population density
1.51***
0.38
agricultural land
1.84
3.01
Adj. R squared
0.1473
0.1851
Note: Standard Errors are calculated using the Huber/White/sandwich estimator. ***,**,* are
significant at 1, 5 and 10 percent, respectively.
13
Table 3. Marginal effect of climate variables
farm value
Coef.
Std. Err.
early rainy temperature
-388.80
359.32
late rainy temperature
173.81
359.46
summer temperature
-478.78***
182.60
winter temperature
298.64**
151.29
early rainy precipitation
-6.74***
2.43
late rainy precipitation
10.68***
3.44
summer precipitation
-28.38***
5.14
winter precipitation
18.47***
2.81
Note: ***,**,* are significant at 1, 5 and 10 percent, respectively.
7. IMPLICATIONS FOR GREENHOUSE WARMING
This section investigates the implications for greenhouse warming during 2040-2049 on
Thailand’s agriculture. To project the climate change impacts, the estimated coefficients from
Specification 2 in Table 3 are used together with the climate projections from the regional
climate model mentioned in the previous sections. In brief, future climate projections in both
scenarios shows trend of increasing temperature throughout Thailand, especially in the central
plain of Chao Phraya river basin and lower part of northeastern region. Total annual precipitation
likely fluctuates in the early part of the century. Since scenarios A2 is assumed to have more
carbon dioxide emissions than scenario B2, projected temperatures from scenario A2 is higher
than those from scenario B2 while projected precipitation from scenario A2 is slightly lower than
those from scenario B2.
By substituting climate projections, this study finds that during 2040-2049 farmland
values per rai are projected to decrease from $2,703 per rai to $2,068 and $2,538 per rai in
climate scenarios A2 and B2, respectively. By multiplying the farmland values per rai to the total
farmland area in Thailand (149 million rai), climate change are projected to adversely affect
Thailand’s agriculture range from $24 billion to $94 billion as shown in Table 4. By downscaling
the analysis to the province level, this article finds that western, upper part of central, and the left
part of northern regions are projected to be better off, while southern, eastern regions, lower part
of central, and the right part of northern regions is projected to be worse off under both climate
scenarios as illustrated in Figure 1. As expected, scenario A2 projects higher negative impacts of
climate change on Thailand’s agriculture more than scenario B2.
14
Table 4. Implications for greenhouse warming (national level)
farmland value/rai
total land values
total change
($)
($1,000)
($billions)
Baseline
2,703
403,361,918
-
A2
2,068
308,674,640
-94.69
B2
2,538
378,838,371
-24.52
Note: Agricultural land is equal to 149,246,428 rai in 2011.
Figure 1. Implications for greenhouse warming (projection of 2040-2049)
Top ten provinces that are projected to adversely affect under both climate scenarios are
Surat Thani, Chiang Mai, Chumphon, Rayong, Chachoengsao, Songkhla, Chanthaburi, Nakhon
Si Thammarat, Trang, Suphanburi, respectively, with the impacts ranging from $3.48- $19.43
billion. On the other hand, Kamphaeng Phet, Udon Thani, Chaiyaphum, Phetchabun, Nakhon
Ratchasima, Nongbua Lamphu, Buriram, Bangkok, Khon Kaen, Sukhothai are top ten provinces
that are projected to be better off with the values ranging from $0.27 - $7.80 billion.
15
8. CONCLUSION
Agricultural sector is potentially the most sensitive economic sector to climate change and
Thailand is one of developing countries that agricultural sector plays an important role. This
study utilizes the Ricardian approach to analyze the effect of climate change on Thailand’s
agriculture and investigate implications for greenhouse warming under future climate change
scenarios during 2040-2049. A unique farm-level dataset is constructed using data from several
sources mainly from the 2011/2012 national agricultural household socio economics survey. The
normal climatological variables during 1971-2012 are constructed using climate data from
Thailand Meteorology Department. Future climate projections are simulated by PRECIS regional
climate model.
The article finds that both temperature and precipitation significantly determine farmland
values. Overall, greenhouse warming is projected to adversely affect Thailand’s agriculture
ranging from $24 billion to $94 billion. For the analysis in the province level, western, upper part
of central, and the left part of northern regions are projected to be better off, while southern,
eastern regions, lower part of central, and the right part of northern regions is projected to be
worse off. Surat Thani, Chiang Mai, Chumphon, Rayong, Chachoengsao, Songkhla,
Chanthaburi, Nakhon Si Thammarat, Trang, Suphanburi are top ten provinces adversely affected
by climate change, while Kamphaeng Phet, Udon Thani, Chaiyaphum, Phetchabun, Nakhon
Ratchasima, Nongbua Lamphu, Buriram, Bangkok, Khon Kaen, Sukhothai are top ten provinces
that are beneficial under climate change.
Governmental organizations related to the agricultural sectors should support farmers on
several ways such as providing knowledge to farmers regarding adequate cropping techniques,
agricultural resource management and encourage farmers to create adaptation plan to reduce the
damage from climate change. At the same time, policy makers should develop plans or programs
ahead to relief those who are projected to adversely affect by climate change in provinces across
Thailand.
16
9. REFERENCES
Adams, R. M., B. A. McCarl, K. Segerson, C. Rosenzweig, K. J. Bryant, B. L. Dixon, J. R.
Conner, R. E. Evenson, and D. S. Ojima. 1999. The economic effects of climate change on
u.s. agriculture. In The impact of climate change on U.S. agriculture., eds R. Mendelsohn
and J. Newmann: Cambridge University Press, London: 18-54.
Attavanich, W., B.A. McCarl, Z. Ahmedov, S. Fuller, and D.V. Vedenov. 2013. Effects of
climate change on US grain transport. Nature Climate Change 3: 638–643.
Buddhaboon, C., S. Kongton, and A. Jintrawet. 2005. Climate scenario verification and impact
on rain-fed rice production. In The study of future climate change impact on water resource
and rain-fed agricultural production., eds S. Chinvanna and A. Snidvongs. Proceedings of
the APN CAPABLE CB-01 Synthesis Workshop, Vientaine, Lao PDR, 29-30 July 2004.
SEA START RC Technical Report No. 13.
Deschenes, O., and M. Greenstone. 2007. The economic impacts of climate change: Evidence
from agricultural output and random fluctuations in weather. American Economic Review
97 (1): 354–85.
Dinar, A., R. Mendelsohn, R. Evenson, J. Parikh, A. Sanghi, K. Kumar, J. McKinsey, and S.
Lonergan (eds) 1998. Measuring the impact of climate change on Indian agriculture.
World Bank Technical Paper No. 402, Washington, DC.
IPCC, ed. 2007a. Climate change 2007: The physical science basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change., eds. S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M.
Tignor and H. L. Miller. Cambridge, UK: Cambridge University Press.
———. 2007b. Climate change 2007: Impacts, adaptation and vulnerability. Contribution of
Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change., eds. M. L. Parry, O. F. Canziani, P. J. Palutikof, P. J. van der Linden
and C. E. Hanson. Cambridge, UK: Cambridge University Press.
———. 2001a. Climate change 2001: The scientific basis. Contribution of Working Group I to
the Third Assessment Report of the Intergovernmental Panel on Climate Change., eds. J.
T. Houghton, Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Maskell
and C. A. Johnson. Cambridge, UK: Cambridge University Press.
17
———. 2001b. Climate change 2001: Impacts, adaptation, and vulnerability. Contribution of
Working Group II to the Third Assessment Report of the Intergovernmental Panel on
Climate Change., eds. J. J. McCarthy, O. F. Canziani, N. A. Leary, D. J. Dokken and K.
S. White. Cambridge, UK: Cambridge University Press.
Isvilanonda, S., S. Praneetvatakul, and C. Khamwong. 2009. Economic impact assessment of
climate change on rice production in Thailand. Project Code: RDG5130015. The
Thailand Research Fund, Thailand.
Khamwong, C., and S. Praneetvatakul. 2011. Impact of climate change on agricultural revenue
and adaptation strategies of farmers in Northeasterm Thailand. Kasetsart Journal (Social
Science) 32: 214-228.
Kumar, K. and J. Parikh. 2001. Indian agriculture and climate sensitivity. Global Environmental
Change 11: 147–54.
McCarl, B. A., X. Villavicencio, and X. Wu. 2008. Climate change and future analysis: Is
stationarity dying? American Journal of Agricultural Economics 90 (5): 1241–47.
Mendelsohn, R (eds.). 2001. Global Warming and the American Economy: A Regional
Assessment of Climate Change. Cheltenham: Edward Elgar Publishing.
Mendelsohn, R., J. Arellano-Gonzalez, and P. Christensen. 2010. A Ricardian analysis of
Mexican farms. Environment and Development Economics 15 (2): 153–71.
Mendelsohn, R., W. D. Nordhaus, and D. Shaw. 1994. The impact of global warming on
agriculture: a Ricardian analysis. American Economic Review 84: 753–71.
Mendelsohn, R., and M. Reinsborough. 2007. A Ricardian analysis of U.S. and Canadian
farmland. Climatic Change 81 (1): 9–17.
Office of Environmental Policy and Planning. 2000. Thailand’s Initial National Communication
under the United Nations Framework Convention on Climate Change. Ministry of
Science, Technology, and Environment. Bangkok, Thailand. 100 p.
Pannangpetch K., V. Sarawat, S. Boonpradub, S. Ratanasriwong, S. Kongton, S. Nilpunt, C.
Buddhaboon, K. Kunket, I. Buddhasimma, P. Kapetch, K. Ek-un, W. Damrhikhemtrakul.
2009. Impacts of global warming on rice, sugarcane, cassava, and maize production in
Thailand. Project Code: RDG5130007. The Thailand Research Fund, Thailand.
18
Reilly, J., F. Tubiello, B. A. McCarl, D. Abler, R. Darwin, K. Fuglie, S. Hollinger, C. Izaurralde,
S. Jagtap, and J. Jones. 2003. U.S. agriculture and climate change: new results. Climatic
Change 57: 43–67.
Reinsborough, M. J. 2003. A Ricardian model of climate change in Canada. Canadian Journal of
Economics 36: 21-40.
Schlenker, W., W. M. Hanemann, and A. C. Fisher. 2005. Will U.S. agriculture really benefit
from global warming? Accounting for irrigation in the hedonic approach. American
Economic Review 95 (1): 395–406.
Schlenker, W., and M. J. Roberts. 2009. Nonlinear temperature effects indicate severe damages
to U.S. crop yields under climate change. Proceedings of the National Academy of
Sciences 106 (37): 15594.
Seo, S. N., and R. Mendelsohn. 2008. An analysis of crop choice: Adapting to climate change in
Latin American farms. Ecological Economics 67: 109–16.
Seo, S., R. Mendelsohn, and M. Munasinghe. 2005. Climate change and agriculture in Sri Lanka:
A Ricardian valuation. Environmental and Development Economics 10: 581–96.
Smith, J. B., and D. A. Tirpak (eds.). 1989. The Potential effects of global climate change on the
United States: Report to Congress, United States, Environmental Protection Agency.
Office of Policy, Planning, and Evaluation.
Chinvanno, S., V. Laung-Aram, C. Sangmanee, and J. Thanakitmetavut. 2009. Simulation of
future climate scenario for Thailand and surrounding countries. Contract
No.RDG5030032. Thailand Research Fund.
Wang, J., R. Mendelsohn, A. Dinar, J. Huang, S. Rozelle, and L. Zhang. 2009. The impact of
climate change on China's agriculture. Agricultural Economics 40 (3): 323–37.
