Dilemmas and
Challenges of Korean Rice Economy
after the Rice
Negotiation
Doo Bong Han1)
The special
treatment of rice import in
JEL Classification: Q170
Key Words: Rice Negotiation, MMA,
Korean Rice Economy
------------------------------------
1) Professor, Department of Food and Resource Economics, Korea
University, Anam Dong, Sungbuk Gu, Seoul, 136-701, Tel: +82-2-3290-3035,
e-mail: han@korea.ac.kr
Introduction
Rice as a main staple has contributed more
than 30 percent of daily per capita calories supplied in the respect of
national nutrition and is also of importance in farm economy since the income
from rice farming is about 50 percent of farm income. Rice policies such as
production encouragement and price support have been the core of agricultural
policy in
Dilemmas of Korean rice economy are the
excess supply of rice caused by declining consumption and the accumulation of public
storage. Although
The objectives of this study are: (1) to describe the importance of Korean rice economy, (2) to review the results of rice negotiation, (3) to analyze the impact of the rice negotiation on rice sector, (4) to discuss dilemmas and challenges of Korean rice economy after the rice negotiation. A Korean rice model for projections and policy simulations was developed to analyze the impact of the rice negotiation and also to discuss the implication of rice import expansion for both farmers and policy makers. The information from the model simulation would be helpful for rice farmers to make future production decisions, and for the Korean government to introduce a new policy programs such as new storage program and the program for promoting rice consumption.
Importance of Korean
Rice Economy
Rice in Korean Agriculture
Rice
is the most important staple crop in
The
acreage under rice cultivation decreased from 1,203 thousand ha in 1970 to
1,016 thousand ha in 2003, which accounts for 55% of total arable land in Table
1. Even though government has spent a sizable budget to expand the acreage of
paddy field per rice farm for increasing international competitiveness in
1990s, the acreage of paddy field per rice farm increased slightly from 0.60 ha
in 1970 to 1.08 ha in 2003, which is relatively very small compared to that of
western countries.
While
the proportion of farm income earned from rice to farm household income
decreased from 34.4% in 1970 to 20.0% in 2003 due to the relatively rapid
growth of off-farm income, the proportion of rice to farm income has been increased
from 42.2 % in 1980 to 50.8% in 2003 because rice market has been protected by
the MMA and government purchase with price support for rice has been maintained
last 10 years.
Table 1 Household, Acreage, and Income for Rice
Farming, 1970~2003
|
1970 |
1980 |
1990 |
2000 |
2003 |
Total Farm Household (1000 households, A) |
2,483 |
2,156 |
1,767 |
1,383 |
1,264 |
Farm Household Cultivating Rice (1000
households, B) |
2,011 |
1,837 |
1,506 |
1,078 |
945 |
B/A (%) |
81.0 |
85.2 |
85.2 |
77.9 |
74.8 |
(1000ha, C) |
2,298 |
2,196 |
2,091 |
1,889 |
1,846 |
Land for Cultivating
Rice (1000ha, D) |
1,203 |
1,233 |
1,244 |
1,072 |
1,016 |
D/C (%) |
52.3 |
56.1 |
59.5 |
56.7 |
55.0 |
Paddy Field per Rice Farm Household (ha, D/B) |
0.60 |
0.67 |
0.83 |
0.99 |
1.08 |
Annual Farm Household Income (1000Won, E) |
256 |
2,693 |
11,026 |
23,072 |
26,878 |
Farm Income (1000Won, F) |
194 |
1,755 |
6,264 |
10,897 |
10,572 |
Income Earned from Rice (1000 Won, G) |
88 |
741 |
3,097 |
5671 |
5369 |
F/E (%) |
75.8 |
65.2 |
56.8 |
47.2 |
39.3 |
G/E (%) |
34.4 |
27.5 |
28.1 |
24.6 |
20.0 |
G/F (%) |
45.4 |
42.2 |
49.4 |
52.0 |
50.8 |
Source:
Ministry of Agriculture & Forestry(MAF),
Korean National
Statistical Office
Changes
of Rice Production
Rice
yields almost doubled from the average level of 2.59 M/T per ha (milled weight)
in 1955~59 to 4.95 M/T in 1995~99 and 4.81 M/T per ha in 2000~03 as shown in Table
2. Annual rice yields were very fluctuating since paddy fields were not well
irrigated until the late 1960s and the degree of damage by flood and/or drought
almost determined the yields. In the 1970s, government invested in large-scale
irrigation projects for paddy fields, developed new high-yield varieties, and
guaranteed the rice price with government procurement. Consequently, the yield
jumped in the middle of the 1970s and has increased slightly afterward. In the
last ten years, the yields ranged from
Planted
acreage of rice decreased slightly from 1.1 million ha to 1.06 million ha for
the past 45 years. As the results of yield increase and a little decrease in
planted acreage, rice production has increased about 50% from 3.04 million M/T
in 1955~59 to 5.17 million M/T in 1995~99 and 4.95 million M/T in 2000~03. Even
though
Changes of Rice Consumption
Per capita
consumption of rice in
The rate of
rice consumption reduction is –0.3 percent during 1970s; -1.1 percent during
1980s; -2.2% during 1990s and –3.7 percent during 2000s respectively. Such a
acceleration of rice consumption reduction is known as a big problem too.
Table
2 Yield, Planted Acreage and Production for Rice, 1955~2003
|
Yields (Average, M/T/ha) |
Planted Acreage (Total, 1,000ha) |
Production (1,000M/T) |
1955~59 |
2.591 |
1,1023 |
3,0373 |
1960~64 |
3.032 |
1,1483 |
3,4233 |
1965~69 |
3.032 |
1,1933 |
3,6743 |
1970~74 |
3.432 |
1,1943 |
4,0563 |
1975~79 |
4.532 |
1,2253 |
5,5023 |
1980~84 |
4.322 |
1,2213 |
5,1933 |
1985~89 |
4.602 |
1,2503 |
5,7103 |
1990~94 |
4.522 |
1,2033 |
5,4403 |
1995~99 |
4.952 |
10573 |
51663 |
2000~03 |
4.814 |
10564 |
49464 |
1.
Yield of paddy rice.
2.
5-year average excluding the lowest and the
highest.
3.
5-year simple average.
4.
4-year simple average.
Source : MAF
The
reason of rapid reduction of rice consumption in households sector is because
consumption of meat and dining out increased due to westernization and
convenience trends of daily food system. Table 3 shows the change in food
consumption during the last three decades. The proportion of urban household
expenditure for rice to total expenditure for food and beverages dropped from
38.6% in 1970 to 6.7% in 2003. On the other hand, consumer expenditure for ¡°away
from home¡± food increased rapidly from 2.1% in 1970 to 45.8% in 2003. As the
Korean economy has grown, the Engel¡¯s index for rice, that is the proportion of
expenditure for rice to total expenditure, decreased from 15.4% to 1.5% during
the period. The share of rice among energy supply per capita per day due to
reduction of rice consumption has declined from 49.2 percent in 1970 to 30
percent in 2003.
Figure 1. Annual Rice Consumption Per Capita
(milled rice, Kg)
Source : MAF, Major Agricultural
Statistics
Table
3 Annual Expenditure of Food in Urban Household
and
Nutrition Intake from Rice in 1970~20003
|
1970 |
1980 |
1990 |
1995 |
2000 |
2003 |
Annual
Total Expenditure per Urban Household (1000 won, A) |
364 |
2,252 |
9,065 |
17110 |
22607 |
26688 |
Expenditure
on Food and Beverages (1000 won, B) |
145 |
930 |
2,634 |
4376 |
5364 |
6113 |
Expenditure
on Rice (1000 won, C) |
56 |
296 |
446 |
397 |
480 |
410 |
Expenditure
on ¡°Away from Home¡±(1000 won, D) |
3 |
35 |
538 |
1388 |
2112 |
2798 |
C/B
(%) |
38.6 |
31.8 |
16.9 |
9.1 |
8.9 |
6.7 |
D/B
(%) |
2.1 |
3.8 |
20.4 |
31.7 |
39.4 |
45.8 |
Engel`s
Index for Food and Beverage (B/A, %) |
39.8 |
41.3 |
29.1 |
25.6 |
23.7 |
22.9 |
Engel`s
Index for Rice (C/A, %) |
15.4 |
13.1 |
4.9 |
2.3 |
2.1 |
1.5 |
Daily Per
Capita Calories Supplied (kcal, E) |
2,533 |
2,485 |
2,853 |
2,959 |
3,010 |
2,985 |
Calories
Supplied from Rice (kcal, F) |
1,246 |
1,234 |
1,175 |
1,054 |
997 |
893 |
E/F(%) |
49.2 |
49.7 |
41.2 |
35.6 |
33.1 |
30.0 |
Review of Rice Negotiation Results
All agricultural
products have been liberalized with tariffication as the result of the UR
Agreement of Agriculture. However,
Table 4 Volume of the MMA by the
Unit :1,000 M/T
Year |
1995 |
1996 |
1997 |
1998 |
1999 |
2000 |
2001 |
2002 |
2003 |
2004 |
Volume (% of food consumption) |
51 (1) |
64 (1.3) |
77 (1.5) |
90 (1.8) |
103 (2) |
103 (2) |
128 (2.5) |
154 (3.0) |
180 (3.5) |
205 (4.0) |
After the rice
negotiation, the revised Country Schedule (CS) was submitted to WTO in
Special treatment for rice by the Minimum
Market Access (MMA) shall be extended for an additional 10 years from 2005 to
2014 and the MMA shall increase in equal annual installments. In the 5th
year, there shall be a multilateral review of the implementation for special
treatment. The import of the MMA will be allocated by country-specific quota on
existing MMA and global quota on additional MMA. Existing MMA volume of 205,228
metric tons, milled basis, shall be allocated to the past importing 4 member
countries as
It is also allowed for
Table 5. Volume of the MMA for Table and
Non-Table Uses
Unit :1,000 M/T
Year |
2005 |
2006 |
2007 |
2008 |
2009 |
2010 |
2011 |
2012 |
2013 |
2014 |
Total volume of MMA |
225.6 |
246.0 |
266.3 |
286.6 |
307.0 |
327.3 |
347.7 |
368.0 |
388.4 |
408.7 |
Table Use (%) |
22.6 |
37.7 |
52.8 |
67.9 |
83.1 |
98.2 |
104.3 |
110.4 |
116.5 |
122.6 |
(10) |
(15) |
(20) |
(24) |
(27) |
(30) |
(30) |
(30) |
(30) |
(30) |
|
Non-Table Use (%) |
203.0 |
208.2 |
213.5 |
218.7 |
223.9 |
229.1 |
243.4 |
257.6 |
271.8 |
286.1 |
(90) |
(85) |
(80) |
(76) |
(73) |
(70) |
(70) |
(70) |
(70) |
(70) |
Structure of Rice Model and Estimation
A rice model is developed to analyze the effects of import expansion expected in the Korean rice negotiation as shown in Table 6. The model is composed of demand and supply components. The components of supply are acreage, yield, import and inventory. Production (Prod) is derived from the multiplication of two estimated equations, acreage and yield. To analyze the factors of production changes, acreage and yield are estimated separately because acreage is in a decreasing trend but yield is increasing continuously. Because it is hard to transfer paddy land to upland or other purposes in the short-run, planted acreage, as shown in (1) equation, is specified to a function of previous real farm price as a proxy of expected price, previous yield and trend considering technology change. Real farm price is calculated by nominal farm price (FP) divided by the GDP deflator (GDF). Previous yield is included in acreage equation since current planting decisions are affected by previous harvest, which is mostly determined by yield.
General factors affecting rice yield are the speed of spreading out new varieties, technology and weather. As shown in equation (2), yield is specified to a function of trend and dummy variables to consider technology and weather. The variable, DM1, is 1 in 1980, 1993 and 2003, which were years of cold weather damage, and it is 0 in other years. The variable, DM2, is 1 in 1997 and 2001, which were exceptionally good harvest years, and it is 0 in other years. The supply in identity (4) is defined as the sum of previous production, previous ending stock (Stock) and import (IM). Inventory (Stock) is composed of private and government inventory. It is assumed that government inventory should be maintained to at least 0.7 million metric tons for food security from 2004 to 2014. On the other hand, private inventory is endogenously determined by an excess supply of rice market after satisfying government inventory.
To estimate
rice demand, the per capita food consumption (PerCon) in equation (5) is
specified to the function of own real consumer price (CP/GDF), real income
(PGDP/GDF) and previous per capita consumption since no substitute of rice was
found because rice has been a unique staple food in
To analyze the effects of rice negotiation on rice supply and demand, assumptions on exogenous variables are needed during the period of 2004-2014 year. Exogenous variables are macroeconomic variables as well as variables related to import. Based on various macroeconomic projections, the real GDP growth rate and inflation rate are assumed to be 5% and 3% per annum. It is assumed that the annual growth rate of the processing volume is 4% per annum. Loss volume is assumed to be 7 % of production, which is actual loss rate. Seed volume is assumed to be 0.075 % of previous production.
Table 6 A Structural Model for the Korean Rice Sector
(1) Acreage= f(FP(-1)/GDF(-1), Yield(-1),
Trend) |
(2) Yield = f(Trend, DM1, DM2) |
(3) Prod
= Acreage*Yield |
(4) Supply = Prod(-1) +Stock(-1) + IM |
(5) PerCon = f(CP/GDF, PGDP/GDF, PerCon(-1)) |
(6) CP/GDF = f(Supply/Pop, (Demand –Stock)/Pop, CP(-1)/GDF(-1)) |
(7) Demand=PerCon*Population + Processing + Seed +
Loss + EX + Stock |
(8) Stock=Government Stock + Private Stock |
(9) FP=f(CP) |
Five behavioral equations for per capita consumption,
planted acreage, yield, consumer price, and farm price are estimated by OLS
estimation as shown in Table 7. The rice model is composed of behavioral
equations and identity conditions and is dynamically solved to get simulated values
after the rice negotiation. To evaluate statistical adequacy on estimated
equations, several statistical tests are conducted. The statistical tests are
the Durbin-Watson test and Godfrey LM test for serial correlation, and the ARCH
test for time-varying heteroscedasticity. The overall stability of the model is
tested by ex-post simulation since the statistical
Table 7 Estimation Results of Behavioral Equations
LOG(Acreage) =32.43+0.46*LOG(FP(-1)/GDF(-1))-0.10*LOG(Yield(-1))-
0.01*TREND (18.4)
(8.8)
(-1.4)
(-14.2) Adj-R2 :
0.96 D-W stat :
1.82 Sample :
1990 - 2003 Yield =
-4119.21 + 2.30*Trend - 49.60*DM1 + 35.60*DM2 (-2.7)
(3.0) (-3.8)
( 2.7) Adj-R2 :
0.65 D-W stat :
2.09 Sample :
1985 - 2003 DM1 : 1 in 1980,
1993, 2003 and others are 0, DM2 : 1 in 1997, 2001 and others are 0 LOG(CP/GDF) =
0.66 -0.40*LOG(Supply/Pop) +0.32*LOG((Demand-stock)/Pop) + 0.97*LOG(CP(-1)/GDF(-1))
(0.8)
(-3.5) (2.1)
(9.6) Adj-R2 :
0.82 D-W stat :
1.77 Sample :
1982 - 2003 LOG(PerCon) =
2.54- 0.12*LOG(CP/GDF) - 0.076*LOG(PGDP/GDF) + 0.82*LOG(PerCon(-1))
(6.0) (-4.1)
(-6.2) (17.4) Adj-R2 :
0.99 D-W stat :
1.75 Sample
: 1977 - 2003 FP = - 541.12
+ 0.80*CP
(4.5) (56.9) Adj-R2 :
0.99 D-W stat : 1.33 Sample
: 1985 – 2003 |
significance of a single equation may not guarantee the
stability of the whole model. Annual data from 1975 to 2003 are used to
estimate the model.
Statistical
significance of the individual equation may not guarantee the overall model
stability and performance. Ex-post dynamic and static simulations[2]
are conducted to evaluate overall model stability from 1990 to 2003.The RMSPE
(Root Mean Square Percentage Error) calculated after the ex-post simulations is
as shown in Table 8. The rice model turns out reasonably stable because most
endogenous variables show below 5 or 10% of the RMSPE in both static and
dynamic ex-post simulations.
Variables |
Dynamic Ex-Post Simulation |
Static Ex-Post
Simulation |
Acreage |
3.95 |
2.10 |
Yield |
3.10 |
3.10 |
Production |
4.13 |
3.07 |
Per Capita Consumption |
6.83 |
2.04 |
Consumer Price |
8.50 |
3.21 |
Farm Price |
7.49 |
3.61 |
Impact of the
Rice Negotiation and Dilemmas
This
study investigates the effects of the rice negotiation by the MMA increase from
205 thousand M/T in 2004 to 408.7 thousand M/T in 2014. It is generally
expected that an adjustment by increasing the MMA would be slower than
tariffication because its effect transfers to rice markets and farmers through
additional supply, which increases indirectly.
As shown in Table 9, planted acreage for rice would decreased about 20% for the next 10 years from 1016 thousand ha to 878 thousand ha in 2009 and 796 thousand ha in 2014. However, production will decrease about 8% for the next 10 years from 4451 thousand M/T in 2003 to 4108 thousand M/T in 2014 because yield increases too. Per capita consumption of rice for food would decrease about 23% from 83.2 kg in 2003 to 65.8 kg in 2014 and per capita consumption for food and processing would also decrease about 14% from 90 kg in 2003 to 77.8 kg in 2014.
Table 9 Projections of Major Endogenous Variables after Rice Negotiation
Acreage (1000ha) |
Production (1000M/T) |
Import (1000M/T) |
Per Capita Consumption (kg) |
Inventory |
Self-Sufficiency ratio (%) |
|||
Food |
Food & Processing |
Amount (1,000M/T) |
Ratio (%) |
|||||
2003 |
1016 |
4451 |
180 |
83.2 |
90.0 |
1098 |
21.7 |
97.4 |
2009 |
878 |
4428 |
307.0 |
70.7 |
79.7 |
1305 |
30.5 |
106.1 |
2014 |
796 |
4108 |
408.7 |
65.8 |
77.8 |
1051 |
24.8 |
97.9 |
If the current situation of Korean rice
economy is maintained, the self-sufficiency ratio of rice would be at the level
of 106% in 2009 and 97.9 % in 2014 because the import would be fully controlled
by the government. The discrepancy between production and consumption will
increase inventory to the maximum 1.34 million M/T in 2008 and 1.05 million M/T
in 2014. Therefore, inventory ratio calculated as inventory over total
consumption will increase from 21.7% to the maximum 31.2% in 2008 and 24.8% in
2014. Inventory will increase until 2008 and then decrease steadily. Therefore,
it is the task assigned for
To reduce inventory ratio, the declining
trend of rice consumption should be lessened and the consumption pattern of
rice should be diversified to increase the price elasticity of minimum public
storage. Korean government is planned to store about 0.84 million M/T as 17% of
current consumption for food security. It is needed to diversify processing
products of rice to create new demand for rice focusing on young generation in
the medium and long run. If
In the short-run, supply control programs such as acreage reduction program and environmentally friendly farming or organic farming are also needed to balance supply and demand of rice in the short-run. Another way to reduce the inventory in the short run is to increase the portion of table use for imported rice more than 10% to 30 % in the mutual agreement of rice negotiation. In this case, rice price will drop and farm income problem will happen unless government increases the direct payment to compensate the decrease of rice income.
Concluding Remarks
The
special treatment called to the Minimum Market Access (MMA) for rice import in
It
may be very difficult for
In
the implementation period of the rice negotiation, Korean government tries to
reduce excess supply and huge public storage by diversifying and promoting rice
consumption in the medium and long run. In order to balance the supply and
demand of rice, supply control programs such as acreage reduction program and environmentally
friendly farming or organic farming are also needed in the short-run.
Han, Doo Bong, J. Shin, H. Son, (1999), ¡°Economic Impacts of
Rice Market Opening: A Comparative Study between Quota and Tariff,¡± Korean
Journal of Agricultural Economics 40
Han, Doo Bong, Jeong-Bin Im, (2003), ¡°Impacts of Greater
Market Access in WTO/DDA Agricultural Negotiations on Korean Rice Market,¡±
Conference Proceedings, 2003 Taiwan-Korea International Seminar on Agricultural
and Resource Economics, August 17-21,
Jeong, Tae-Ho, (2004), ¡°Characteristics and Debates of Rice
Negotiation,¡± NACF Research Institute.
Kako, Toshiyuki, Masahiko Gemma, Shoichi Ito. (1997), ¡°Implications
of the Minimum Access Rice Import on Supply and Demand Balance of Rice in
KREI (2005), 2005
Agricultural Outlook.
Lee, Dae-Seob, Eric J. Wailes, Jim M. Hansen. (1988), ¡°From
Minimum Access to Tariffication of Rice Imports in
Life and Resource Institute (2004), Proceedings for 2004 Korea-Japan Joint Symposium on Rice Consumption
Promotion Strategies,
Song, Yoo-Chul, Song-Soo Lim, Jin-Kyo Shu, etc. (2003), WTO/DDA
Negotiation on Agriculture: Evaluation for Modalities and Its Implication for
Agricultural Policies, KIEP.
WTO. (2003), Negotiations on Agriculture: First Draft of
Modalities for the Further Commitment.
WTO. (2003), The