THE IMPACT OF PRICING POLICY ON THE DEMAND FOR WATER IN
IRAN AGRICULTURAL SECTOR
By
AHMAD SADEGHI
Thesis Submitted to the School of Graduate Studies, University Putra Malaysia,
in Fulfilment of the Requirements for Degree of Doctor of Philosophy
March 2010
DEDICATION
This thesis is dedicated to:
The Late Imam Khomeyni, the enlightener and messiah of Iranian people, and my
dear family, especially my father, Late Barat Ali, and my mother, Fatemeh Soghra,
who have given their full support, encouragement and devotion to my completion of
the study; to my dear wife “Esmat” for her patience during the course of my study
and my beloved children Saeideh and Zahra who missed me always.
Finally, my beloved grandmother, “Ome Sallameh”, whom I lost in the first of my
study, but she is always with me
ii
ABSTRACT
Abstract of thesis presented to the Senate of Universiti Putra Malaysia in fulfilment
of the requirement for the degree of Doctor of Philosophy
THE IMPACT OF PRICING POLICY ON THE DEMAND FOR WATER IN
IRAN AGRICULTURAL SECTOR
By
AHMAD SADEGHI
March
2010
Chairman: Mohd. Ghazali Mohayidin, PhD
Faculty: Agriculture
Iran is located in an arid and semi-arid area with scarcity of water resource for its
agricultural activities. About 93 percent of the total annual Iranian water
consumption is for agricultural activities. Notwithstanding high irrigation water
demand in the agricultural sector, farmers pay very low price for water and the cost
borne by farmers for irrigation water is much lower than the actual value of water.
The low irrigation fee has caused not only inefficient allocation of water resources in
this sector, but it has also resulted in farmers producing crops which require
relatively large amount of water as well as non-essential crops.
The main objective of this study is to analyze alternative water pricing mechanisms
and to determine its impact on water demand. The specific objectives of this research
include: to estimate the demand for irrigation water, to analyze the effects of
iii
increasing water price to farmers and finally to recommend a suitable mechanism for
determining an efficient pricing system for irrigation water. Data for this study were
collected from 28 provinces in Iran, obtained from the relevant provincial statistical
reports of 2001 to 2006. Crops grown in the 28 provinces were wheat, barley, lentil,
pea, pinto bean, onion, tomato, potato, cucumber, water melon, cotton, sugar beet,
and these are based on the producer provinces statistical reports from 2001 to 2006.
Demand functions for irrigation water were estimated as functions of water price,
land rental, price of fertilizer, machinery rental, price of seed, wage rate, price of
animal fertilizer and pesticide, and irrigated production level for each of the crops
mentioned above. Based on the natural log functional forms and the estimated
regression coefficients, pricing systems for irrigation water for 28 selected provinces
were developed.
Parameters for the demand functions were estimated using Ordinary Least Squares
(OLS) or Generalized Least Squares (GLS), Estimated Generalized Least Squares
(EGLS), and or Weighted Least Squares (WLS). The parameters of models were
estimated using the econometric method on panel data.
A major conclusion that emerged from this research is that the pricing elasticity of
irrigation water demand for most crops in Iran agricultural sector is perfectly
inelastic. Furthermore, the estimated results of price elasticity of irrigation water
demand, in terms of water supply cost (MC and AVC), showed that they were in fact
very low, and perfectly inelastic for most crops.
iv
Also, this study showed that the price elasticity of water demand is relatively
inelastic based on value of marginal product, and it can approximately lead to an
efficient use of irrigation water. The study concluded that the Iranian authorities
could make use of suggested pricing mechanism as an important and effective policy
tool for water conservation.
v
ABSTRAK
Abstrak tesis yang dikemukakan kepada Senat Universiti Putra Malaysia sebagai
memenuhi keperluan untuk ijazah Doktor Falsafah
IMPAK POLICI HARGA TERHADAP PERMINTAAN AIR DI SEKTOR
PERTANIAN IRAN
Oleh
AHMAD SADEGHI
March
2010
Pengerusi: Mohd. Ghazali Mohayidin, PhD
Fakulti: Pertanian
Negara Iran terletak di kawasan kering dan separa-kering yang menghadapi masalah
kekurangan air bagi aktiviti pertanian. Sekitar 93 peratus daripada jumlah
penggunaan air tahunan adalah untuk aktiviti pertanian. Walaupun terdapat
permintaan air dalam kuantiti yang tinggi dalam sektor pertanian, harga yang dibayar
untuk air oleh petani adalah sangat rendah, dan harga yang dibayar untuk air
pengairan adalah kurang daripada nilai sebenar air. Harga air yang rendah bukan
hanya telah menyebabkan pengagihan air yang tidak cekap, tetapi ia juga telah
menyebabkan petani menggeluarkan tanaman yang memerlukan banyak air dan
tanaman yang tidak begitu penting.
Objektif utama kajian ini adalah untuk menganalisis alternatif kaedah penentuan
harga air dan kesannya terhadap permintaan air. Objektif khusus kajian ini termasuk:
untuk membuat anggaran terhadap permintaan air untuk pengairan, menganalisis
vi
kenaikan harga air yang disalurkan kepada para petani dan juga mencadangkan
mekanisma atau sistem yang cekap dan sesuai untuk menentukan harga untuk air
pengairan. Data yang digunakan untuk kajian ini adalah berdasarkan kepada laporan
statistik dari 28 wilayah di Iran bagi tahun 2001 sehingga 2006. Mengikut laporan
statistik itu juga, tanaman yang terdapat di 28 wilayah tersebut ialah gandum, barli,
lentil, kacang polong, kacang-kacangan luna, bawang merah, tomato, kentang, timun,
semangkar, kekabu, dan gula bit.
Fungsi permintaan air untuk pengairan dianggar sebagai fungsi kepada harga air,
sewa tanah, harga baja, sewa jentera, harga biji-benih, kadar upah, harga baja ternak,
harga racun perosak dan dan tahap pengeluaran untuk setiap produk yang tersebut
diatas. Berdasarkan bentuk logaritma asli dan anggaran pekali regresi, satu sistem
penilaian telah dibentuk bagi menentukan harga atau nilai air pengairan untuk 28
wilayah yang dinyatakan di atas.
Parameter fungsi permintaan dianggar dengan menggunakan “Ordinary Least
Squares” (OLS) atau “Generalized Least Squares” (GLS), “Estimated Generalized
Least Squares” (EGLS), dan/atau “Weighted Least Squares” (WLS). Penganggaran
model dibuat dengan menggunakan kaedah ekonometrik ke atas data panel.
Kesimpulan utama yang ditemui daripada kajian ini ialah kaedah semasa
menentukan harga air untuk pengairan adalah tidak cekap; iaitu, keanjalan harga
permintaan untuk air pengairan bagi kebanyakan tanaman dalam sektor pertanian di
Iran sangat tidak anjal. Seterunya, anggaran keanjalan harga air pengairan
vii
berdasarkan kos marginal dan kos berubah purata adalah sangat rendah, atau dengan
kata sangat tidak anjal untuk kebanyakan tanaman.
Selain daripada itu, kajian ini juga mendapati bahawa keanjalan harga permintaan
untuk air pengairan adalah agak anjal jika berdasarkan nilai keluaran marginal, oleh
itu ia boleh membawa kepada penggunaan air yang lebih cekap. Kajian ini
merumuskan bahawa pihak berkuasa Iran boleh menggunakan mekanisma perletakan
harga yang dicadangkan sebagai alat dasar yang penting dan cekap untuk
pemuliharaan sumber air.
viii
ACKNOWLEDGEMENTS
Thank God that whatever my heart ever desired
God gave me that, and more than I ever could seek.
(Hafiz - e Shirazi)
I thank God for all His blessings on me and thank Him for giving me courage and
strength to finish my study. It is understood that human beings cannot repay one
another enough. Hence, it is better to request Almighty Allah to reward the person
who did a favor and to give his best.
A person cannot go through life without the help and guidance from others. One is
invariably indebted, knowingly or unknowingly. These debts may be of physical,
mental, psychological or intellectual in nature but they cannot be denied. To enlist all
of them is not easy. To repay them even in words is beyond my capability. The
present work is an imprint of many persons who have made significant contribution
to its materialization.
The success of this thesis would not have been possible without various contributions
and support to this work directly or indirectly, and I would like to convey my special
appreciation to those who made it possible. I wish to express my deep sense of
appreciation and gratitude towards my committee chairman and supervisor, Prof. Dr.
Mohd Ghazali Mohayidin for his valuable patient, guidance and supervision of this
dissertation. Your morality, constant support and encouragement have helped me to
press on until the research written and completed. I learned and experienced a lot to
doing a good research.
ix
I am grateful to my advisory committee members, Prof. Dr. Md. Ariff Hussein, and
Dr. Jalal Attari for your recommendations and guidance that lead this thesis to
successful completion. Please accept my heartiest gratitude, you all have been
sources of help, encouragement, and valuable advice to me, I am also grateful for
your valuable suggestions and guidance during this study, without which the
completion of my research would not have been possible.
I am thankful to all staff of UPM, especially those of the Agriculture and Economics
Faculties who contributed to my learning process, especially Prof. Dr. Zainal Abidin
Mohamed Head of Department of Agribusiness and Information Systems. You
behaviour was so friendly; I enjoyed a lot, thanks a lot to all of you.
Words are not enough to express my gratitude to my family for their patience and
perseverance during my absence and for keeping me warm even when out of the
country. I owe a lot to my parents and for accepting inconveniences of my absence
during my study. They have been a constant source of encouragement. Finally I am
especially grateful to my dear wife and children for their patience during the course
of my study, and my dear brothers Mohammad Ali, Mohammad Hossein and my
sisters.
I am deeply indebted to many individuals who have assisted me to perform the
research and finalize this thesis by providing scientific, technical, administrative and
moral support. I would like to offer my sincere gratitude to the previous Chancellor
of Power and Water University of Technology Dr.Naghashan. Special thanks to Dr.
Khodabakhsh and Mr. Ahmadzadeh staff of Iran Water Resources Management
x
Company, Dr. Mohaddes, Dr. Sanaei, Dr Montazer Hojat, Dr. Vakilpour, and Dr.
Mortazavei, Babaei, Khomamei, Barangi, and my entire friends who helped me.
xi
APPROVAL
I certify that a Thesis Examination Committee has met on 5th March 2010 to conduct
the final examination of Ahmad Sadeghi on his thesis entitled “The Impact of Pricing
Policy on the Demand for Water in Iran Agricultural Sector” in accordance with the
Universities and University Colleges Act 1971 and the Constitution of the Universiti
Putra Malaysia [P.U. (A) 106] 15 March 1998. The committee recommends that the
student be awarded the Doctor of Philosophy.
Members of the Examination Committee are as follows:
Zainal Abidin Mohamed, PhD
Professor
Faculty of Agriculture
Universiti Putra Malaysia
(Chairperson)
Khalid Abdul Rahim, PhD
Professor
Faculty of Economics and Management
Universiti Putra Malaysia
(Internal Examiner)
Amin Mahir Abdullah, PhD
Lecturer
Faculty of Agriculture
Universiti Putra Malaysia
(Internal Examiner)
Mohd Fazui Bin Mohd Jani, PhD
Professor
Faculty of Economics and Business
University Kebangsaan Malaysia
(External Examiner)
BUJANG KIM HUAT, PhD
Professor and Deputy Dean
School of Graduate Studies
Universiti Putra Malaysia
Date: 5 March 2010
xii
This thesis was submitted to the Senate of Universiti Putra Malaysia and has been
accepted as fulfilment of the requirement for the degree of Doctor of Philosophy.
The members of the Supervisory Committee were as follows:
Mohd Ghazali Mohayidin, PhD
Professor
Faculty of Agriculture
Universiti Putra Malaysia
(Chairman)
Md Ariff Hussein, PhD
Professor
Faculty of Agriculture
Universiti Putra Malaysia
(Member)
Jalal Attari, PhD
Lecturer
Faculty of Water Engineering
Power & water University of Thechnology
(Member)
HASANAH MOHD GHAZALI, PHD
Professor and Dean
School of Graduate Studies
Universiti Putra Malaysia
Date:
xiii
DECLARATION
I declare that the thesis is based on my original work except for quotations and
citations which have been duly acknowledged. I also declare that it has not been
previously, and is not concurrently, submitted for any other degree at University
Putra Malaysia or at any other institution.
AHMAD SADEGHI
Date: 5 March 2010
xiv
TABLE OF CONTENT
iii vi ix xii xiv xvii xviii xix xx 1 ABSTRACT
ABSTRAK
ACKNOWLEDGEMENTS
APPROVAL
DECLARATION
LIST OF TABLES
LIST OF FIGURE
LIST OF APPENDICES
LIST OF ABBREVIATIONS
CHAPTER
1 1 INTRODUCTION
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 The Importance of Water
General Overview of Water
1.2.1 Water and Population
1.2.2 Water Uses
1.2.3 A Synopsis of Global Water Crisis
Study Scope
1.3.1 Background of Iran
1.3.2 Overview of Climate in Iran
1.3.3 Overview of population in Iran
1.3.4 Overview of Agriculture Status in Iran
Description of the problem
Research Questions
Research Hypotheses
Research Objectives
Significance of the study
Organization of the study
23 2 LITERATURE REVIEW
2.1 2.2 2.3 2.4 General Survey of Water Pricing
Pricing Methods of Irrigation Water
2.2.1 Volumetric Pricing Method
2.2.2 Non – Volumetric Pricing Method
2.2.3 The Market - based Pricing Method
2.2.4 The Quotas Method
Pricing Mechanism and Accomplished Studies in Iran
2.3.1 Urban Water
2.3.2 Agricultural Water
Discussion and Deduction
3.3 23 28 29 32 34 37 37 37 41 45 50 3 METHODOLOGY
3.1 3.2 1 1 1 3 4 5 6 7 10 10 11 18 19 19 20 21 Introduction
Theoretical Framework
3.2.1 Demand Function
3.2.2 Elasticity- A General Survey
3.2.3 Irrigation Water Demand Function
Econometric Methodology
xv
50 50 51 62 66 71 3.4 3.5 3.6 Econometric Model
Data Collection
Description of Variables
3.6.1 Water Demand
3.6.2 Water Price
3.6.3 Output Price
3.6.4 Wage
3.6.5 Cultivated Area
3.6.6 Other Explanatory Variables
3.6.7 Value of Marginal Product (VMP)
3.6.8 Average cost
3.6.9 Short-run Marginal Cost
4 RESULTS AND DISCUSSION
4.1 4.2 4.3 4.4 Introduction
Estimation of the Model with Current Price
4.2.1 Wheat
4.2.2 Barley
4.2.3 Lentil
4.2.4 Pea
4.2.5 Pinto Bean
4.2.6 Onion
4.2.7 Tomato
4.2.8 Potato
4.2.9 Cucumber
4.2.10 Watermelon
4.2.11 Cotton
4.2.12 Sugar Beet
Estimations of the Model with Alternative Prices
Discuassion
5 SUMMARY, CONCLUSION AND RECOMMENDATIONS
5.1 Summary
5.1.1 Purpose and Objectives
5.1.2 Research Procedures
5.1.3 Research Findings
5.1.4 Contributions of the Study
5.2 Conclusion
5.2.1 Policy Implications of the Empirical Findings
5.2.2 Limitations of the Current Research
5.2.3 Recommendations for Future Research
REFERENCES
APPENDICES
BIODATA OF STUDENT
LIST OF PUBLICATIONS
xvi
78 79 80 80 83 83 83 84 84 84 86 86 87 87 87 88 92 95 99 102 105 108 112 115 117 121 124 127 131 134 134 134 135 136 142 144 145 147 148 150 158 203 204 LIST OF TABLES
Table
Page
Shares of total water use
4 Main features of regions of Iran
9 Agricultural lands area on holdings with cropland
13 Time process of irrigation fee in Iran
15 Comparison of pricing methods
30 Ranges of irrigation efficiency in some provinces in Iran
83 Water demand function for wheat
89 Water demand function for barley
93 Water demand function for lentil
97 Water demand function for pea
100 Water demand function for pinto bean
104 Water demand function for onion
107 Water demand function for tomato
110 Water demand function for potato
114 Water demand function for cucumber
116 Water demand function for watermelon
119 Water demand function for cotton
123 Water demand function for sugar beet
126 Descriptive statistics of current price and water alternative prices
128 Estimated coefficients for the alternative prices of water
130 xvii
LIST OF FIGURE
Figure
Page
Project Water Scarcity in 2025
5 Map of Iran’s Borders
7 Iran Climate Map
8 Main Basins of Iran [IWRMC]
9 xviii
LIST OF APPENDICES
A.
Water in World and Iran
B.
Characteristics of Iran Provinces and their Agricultural Products
C.
A Review on Studies Carried Out
D.
Irrigation Demand Function
E.
Schematic of the Compute Stages of the Water Demand
F.
Estimation Results of the Water Demand Functions
G.
Studies with High R – Square
H.
Descriptive Statistics
xix
LIST OF ABBREVIATIONS
AFC
AP
ASCE
ATC
AVC
CWP
ECM
EGLS
FAO
FEM
GDP
GE
GLS
IIWM
IWRMC
MB
MC
MCP
MNB
MP
MR
MSB
MSC
O&M
OECD
OLS
PE
REM
SRMC
STATA
SUR
SURE
VMP
WLS
Average Fixed Cost
Average Product
American Society of Civil Engineers
Average Total Cost
Average Variable Cost
Crop Water Productivity
Error Components Model
Estimated Generalized Least Squares
Food and Agriculture Organization
Fixed Effects Model
Gross Domestic Product
General Equilibrium
Generalized Least Squares
International Institute of Water Management
Iran Water Resources Management Company
Marginal Benefit
Marginal Cost
Marginal Cost Pricing
Marginal Net Benefit
Marginal Product
Marginal Revenue
Marginal Social Benefit
Marginal Social cost
Operation and Maintenance
Organization for Economic Co-operation and Development
Ordinary Least Squares
Partial Equilibrium
Random Effects Model
Short Run Marginal Cost
Data Analysis Statistical Software
Seemingly Unrelated Regression
Seemingly Unrelated Regression Estimation
Value of Marginal Product
Weighted Least Squares
xx
CHAPTER I
1
1.1
INTRODUCTION
The Importance of Water
Water plays a very important role in the formation and continuation of civilizations,
and it is a necessary factor for economic development. Water is a fundamental
precondition for increasing food production, for energy development, and for
industrial activities and consequently, to provide higher employment opportunities,
security of potable water hygiene, and to protect biological diversity and ecosystem.
Water is available almost everywhere, but there is much difference between supply
and demand for water in quality and quantity, timing and place.
In the Post-World War II period, economic development became the main goal of
various communities. In this regard, water was known as one of the main
developmental factors, and great efforts were made in using new technical and
financial possibilities in water supply and demand throughout the world. As
economic activities increase, the importance of water increases, and so does the
importance of studies relating to supply and demand of water.
1.2
1.2.1
General Overview of Water
Water and Population
Water is the source of life on the planet earth. About 70 percent of the earth’s surface
and also human’s body is made from water. Most of the earth surface water (about
1338 billion km3) is in the form of oceans, and fresh water only forms three percent
of the total volume of water on earth. Almost 75 percent of total fresh water (about
24 million km3) is in the form of ice caps and glaciers located in polar areas. Finally,
between 12.5 and 14 billion cubic meters of water is available for human use which
amounted to about 6.8 billion liters per person in the year 2001. By the year 2025
global per capita availability of freshwater is projected to drop to 4.8 billion liters per
person as another two billion people join the world population.
Global per capita figures on water availability, however, give a false picture, as the
world’s available freshwater supply is not distributed evenly around the globe,
throughout the seasons, or from year to year. Thus, in many cases water is not
sufficient where or when it is necessary. In other cases we have too much water, in
the wrong place, at the wrong time (Hinrichsen, Robey, & Upadhyay, 1997). For
example, the amount of renewable freshwater available per capita on an annual basis
ranges from 600 million liters in Iceland to only 75,000 liters per person in Kuwait,
as estimated in 1995 (Gardner-Outlaw & Engleman, 1997).
The world’s population is growing by about eighty million people each year. This
number implies an increased demand for freshwater of about 64 billion cubic meters
a year (Clarke, 1991). While population growth rates have somewhat slowed down,
the absolute number of people added in each year have not done so. For example,
because nearly two billion people have been added to the planet since 1970, per
capita availability of water is one - third lower now than it was then (Postel, 1997).
The population growth rates have ominous implications for per capita water supply
in some regions such as Africa and the Near East. As their population grows, more
and more countries encounter water shortage.
2
Water stress and water scarcity are the two words used to describe countries with
water in short supply. When annual water supplies drop below 1,700 cubic meters
(between 1,700 and 1,000 cubic meters) per person, a country is said to be
encountering water stress, and as annual water supplies drop below 1,000 cubic
meters per person, it is said to be facing with water scarcity. The chronic shortage of
freshwater is a threat to food production, economic development, and ecosystems.
Falkenmark and Widstrand (1992) developed the concepts of water stress and water
scarcity based on an index of per capita freshwater needs. They estimated a
minimum need of 100 liters per day per person for household use and between 5 and
20 times as much for agricultural and industrial uses. Population, fertility rate and
growing water shortage for some selected countries are shown in the Table A.1 in the
Appendix.
1.2.2
Water
Water Uses
uses
include
agricultural,
industrial,
household,
recreational
and
environmental activities. Agriculture is the biggest user of water in the world. It is
estimated that approximately 69 percent of total water consumed in the world is for
irrigation. Approximately 23 percent of world water uses are industrial activities
such as power plants, ore and oil refineries, and manufacturing plants. About 8
percent of total world water is consumed for household related purposes.
The world is steadily being affected by climate changes. The level of economic
development, weather condition, and population are factors that affect the patterns of
water usage in different countries. For instance, in Africa, 88 percent of water is used
3
for agricultural purposes where as in developed countries, most of renewable water is
used in industrial activities and electricity production (Refer to Table 1-1).
Table 1-1 Shares of total water use
Countries
Worldwide
OECD*
Iran
Uses
(%)
(%)
(%)
Domestic
8
5
6
Industry
23
65
1
Agriculture
69
35
93
*OECD: The member countries of Organization for Economic Co-operation and
Development
In most countries around the world, a very small proportion of water is used for
drinking, cooking, washing, cleaning, irrigation of house gardens, and public
services. In addition, when standard of living improves, demand for household water
increases.
1.2.3
A Synopsis of Global Water Crisis
Water crisis is a term referring to a situation in which the world’s water resources are
disproportionate with human demand. The major aspects of the water crisis are
overall scarcity of usable water, and water pollution. Approximately one billion
people don't have access to clean potable water. Furthermore, more than two billion
lack access to adequate sanitation. Millions of people perish every year as a result of
preventable water related diseases, and this is certainly one of the biggest
development failures in the modern era. Based on estimates of the Pacific Institute,
over 34 million people might die in the next 20 years as a result of water-related
diseases (Gleick, 2002).
4
On the other hand, global water resources are endangered by climate changes, abuse,
and pollution. This, of course, could be prevented if authorities pay respect to
protection of the environment using innovative water efficiency and conservation
strategies, community-scale projects, smart economics, and new technologies.
Figure 1-1 shows that, in 2025, water shortages will be more prevalent among poorer
countries where population growth is rapid and resources are limited. This group of
countries include all African and Middle Eastern countries, and parts of Asia that
will have less than 650 m3 of water per person which is a severe water shortage by
any standard (Johansson, 2000).
Figure 1-1 Project Water Scarcity in 2025 (based on Seckler et al., 1998)
1.3
Study Scope
This research focuses on water usage in agriculture sector of Iran. The crops that will
be the focus of the valuation study are; wheat, barley, lentil, pea, pinto bean, onion,
tomato, potato, cucumber, water melon, cotton, sugar beet.
5
Iran is the eighteenth largest country in the world. Iran consists of 30 provinces, each
governed by an appointed governor. The provinces are divided into counties, and
subdivided into districts and sub-districts. Important information on Iran provinces
such as its capital, area, annual precipitation, and population is shown in Table B.1 in
the Appendix.
1.3.1
Background of Iran
Iran is located in the Middle East, the northern temperate zone to the South West of
Asia. Iran’s land area is about 1.648 million km2 (636,296 square miles), and is
bounded by Azerbaijan (759 km) and Armenia (48 km) in the northwest, the Caspian
Sea in the north, Turkmenistan (1205 km) in the northeast, Pakistan (978 km) and
Afghanistan (945km) in the east, Southern coastline (2045 km) in the south and,
finally, Turkey (511 km) and Iraq (1,609 km) in the west. The Persian Gulf located
in the south has an area of 232,850 square kilometers and stretches 930 km from the
Arvandrood River to the Oman Sea, with an average width of 288 km. Iran has more
than 3,450 rivers among which Karoun River, Atrak, Zayanderood, Sefidrood,
Hirmand and Uroomiyeh are the most important ones.
About 65 percent of Iran is covered by deserts, salt flats, and bare-rock Mountains.
11 percent of Iran’s total surface is covered by forests, and also, 7 percent is covered
by cities, towns, villages, industrial areas and roads (See Figure1-2).
6
Figure 1-2 Map of Iran’s Borders
1.3.2
Overview of Climate in Iran
Iran is an arid and semi-arid country which is characterized by long warm and dry
periods covering nearly ninety percent of the country. It has a variable climate. In the
south, the summers are very hot and winters are mild. In the northwest, the winters
are cold with heavy snowfall and subfreezing temperatures during December and
January. Spring (March, April and May) and fall (September, October and
November) are relatively mild, while summers (June, July and August) are dry and
hot (See Figure1-3).
7
Caspian Mild and We
Caspian Mild
Mediterranean with Spring Rains
Mediterranean
Cold Mountains
Very Cold Mountains
Cold Semi-Desert
Hot Semi-Desert
Dry Desert
Hot Dry Desert
Hot Coastal Dry
Coastal Dry
Figure 1-3 Iran Climate Map
The main source of water in Iran is precipitation (both Rainfall and Snow), and water
entering from Border Rivers. Total precipitation is estimated to be about 413 billion
cubic meters, of which almost 295 billion cubic meters evaporates. In Iran, the mean
annual rainfall is about 250 mm which is about 30 percent of the mean annual
precipitation in the world. Western part of Zagrous Mountains in the west of Iran has
one of the regions with heaviest precipitation. Altogether, the total potential of
renewable water resources in Iran has been estimated to be 130 billion cubic meters.
The country is divided into 6 main hydrological basins which are, in turn, subdivided
into 37 basins (See Figure 1-4).
8
Figure 1-4 Main Basins of Iran [IWRMC]
Table 1-2 shows areas, the volume per year of precipitation, and the renewable water
resources (precipitation minus evapotranspiration) for the six basins of Iran.
Table 1-2 Main features of regions of Iran
Basin
Region Name
Area
(km2)
Precipitation
Volume
(mm3/year)
1
Caspian Sea
173,730
2
Urumieh Lake
51,866
3
Persian Gulf
419,802
4
Central
851,126
5
Moshkil Hirmand
107,369
6
Kashaf-rood
44,107
Source: Jamab Consulting Engineers (1999)
9
84,190
22,300
153,820
27,510
13,480
11,860
Net
Precipitation
Volume
(mm3/year)
24,834
7,207
62,035
29,584
1,910
2,430
1.3.3
Overview of population in Iran
The rate of population growth in Iran is high. The highest recorded rate of 3.9
percent occurred in 1986, and the lowest recorded rate of 1.45 percent occured
during the years of 1986-1996 (Ghazi, 2002). According to the census in the year
2005, Iran’s population was 71.4 million people approximately 99 percent of which
are Muslim. About 65 percent of Iran’s people are of Aryan origin. Iran’s population
is relatively young; almost 34 percent of the population is under the age of 14 and 61
percent between 15 and 64 years of age. It is also expected that the population may
double by 2021 (Plan and Budget Organization, 1999).
1.3.4
Overview of Agriculture Status in Iran
Agriculture plays an important role in the Iranian economy. According to a report
from Iran Statistics Centre in the year 2005, agriculture sector forms 11.5 percent
($170 billion) of the Gross Domestic Product (GDP), one third of non-oil exports
(Around $55 billion).
Moreover, the sector employs about 23.4 percent of the labor force and provides
more than 80 and 90 percent of the national food requirements and raw materials for
domestic industries respectively.
Iran’s agricultural sector is one of the most important economic sectors of the
country. One-third of Iran’s total area is suitable for agriculture. However, due to
poor soil and lack of adequate water distribution most of the areas are not under
cultivation. In fact, only about 20 percent of the total land area is under cultivation in
the form of cultivatable land, gardens and etc. According to published statistics in the
10
year 2003, about 8 million hectares of the cultivated area were irrigated; and about 9
million hectares were rain fed (See Table B.2 in the Appendix). The western and
north western parts of the country have the most fertile soils.
According to reports from some agencies, the various climatic zones make it possible
to prepare the land for planting a great variety of crops, including fruits
(pomegranate, fig, melon, orange, grape, peach, and date), cereals (rice, maize,
barley, and wheat), vegetables, cotton, sugar beet and sugarcane, pistachio (38% of
the world’s output in 2005), nuts, olive, spices (Press TV, 2008), tea, tobacco and
medicinal herbs (Iran Daily, 2007). More than 2,000 plant species grow in Iran; only
100 out of which are being used in pharmaceutical industries. The land covered by
Iran’s nature flora is four times that of the Europe’s (Mehr News, 2007).
1.4
Description of the problem
Water resources are among major assets to every country. In the past decades,
accompanied by increasing population, urbanization, and industrial development,
there has been increased demand for water. The increasing water demand has caused
an alarming decrease in annual per capita renewable water resources. Based on the
studies conducted by the United Nations (UN) experts, the per capita water resources
of Iran are projected to be about 726-860 m3 in 2025 compared with 2,200 m3 in
1990. By the year 2025, Iran is expected to fall into the category of countries with
critical water scarcity (Mousavi, 2005).
Nowadays in the most arid and semi-arid areas of Iran, people are facing with
insufficient supply of water and it is recognized as one of the major constraints to
11
economic development. In such areas, the main problem in water management is
matching supply and demand for water. Demand for water increases due to
population growth and economic activities especially those in agriculture. However,
the supply of water has remained constant, resulting in shortage or scarcity of water
for future needs.
Currently, about 89.5 out of 130 cubic billion meters of renewable water (68.85%) is
yearly used in the country. Thus, in view of the current population of Iran (71.4
million), the current consumption of renewable water is 1900 cubic meters per
capita. Hence, based on Falken Mark Index, Iran is on the verge of water crisis.
Likewise, based on the indices of International Institute of Water Management
(IIWM), Iran is in severe water crisis.
According to IIWM report, Iran should add to its water resources around 112% to
maintain the current situation, an amount that seems impossible in light of the
capacity of existing water resources.
The world population growth and limited water resources have resulted in a shortage
of agricultural products in Middle Eastern countries, especially Iran. Iran is located
in an arid and semi-arid area. Based upon the accomplished studies in Iran Water
Comprehensive Plan, renewable water resources of Iran amount to 130 billion cubic
meters approximately. It has been show that of all current renewable water resources
of Iran, about 83 billion cubic meters (93%) is taken up by the agricultural sector. As
one of the major economic sectors in Iran, the agricultural sector is encountering
rareness of water resources. Of all 165 million hectares of the country’s area, about
12
20 million hectares (12.2%) are irrigated, and 17 million hectares (10.3%) are in the
form of dryland. According to (Keshavarz, Ashraft, Hydari, Pouran, & Farzaneh,
2005), 6.4 million out of the 37 million hectares of agricultural lands are under
annually irrigated crops, 2 million hectares are under horticultural crops, and about
6.2 million hectares are under annual dryland crops; the remaining 3.9 million
hectares are fallow. Of the country’s total area, 90 million hectares are pasture land
and 12.4 million hectares are forests. Based upon the available data, the areas
receiving full irrigation in Iran total to 5 million hectares (Refer to Table 1-3 ). At
least 1.6 and 1.8 million hectars of the irrigated areas are suffering from severe and
moderate water stress, respectively (Keshavarz et al., 2005).
Table 1-3 Agricultural lands area on holdings with cropland
Area
Irrigated
Rainfed
Uses
Cropland (ha)
Cropland (ha)
Land under annual crops
5141153
6505877
Fallow
1873689
2676453
Orchards & nuserie
1282188
185846
Total
8297031
9368176
Source: Statistical center of Iran - Iran statistical book 2006
Total
(ha)
11647026
4550142
1468034
17665198
In the past 80 years, rapid growth of population has been one of the most important
factors contributing to the decrease in per capita renewable water in Iran. Within a
period of 80 years, the population of Iran has increased about seven times.
Consequently, annual per capita of renewable water has decreased from 13000 cubic
meters in the year 1921 to 1750 cubic meters in the year 2006.
In view of the decreasing per capita renewable water, it is anticipated that food
security will be a serious challenge in the coming decades. On the other side, due to
low efficiency of agricultural irrigation systems, about 50 to 60 percent of the
13
renewable water devoted to agriculture is squandered. This has caused agricultural
water productivity to be very low (Keshavarz et al., 2005).
Efficient usage of irrigation water in Iran is one of the most important contributing
factors to producing as much food as required at present and in the future. Suitable
planning, price policies, management, and education for efficient application in this
sector is one of the most important policies of the Iran’s government (Keshavarz et
al., 2005).
Studies have revealed that the overall irrigation efficiency in Iran ranges from 33 to
37 percent which is lower than average worldwide irrigation efficiency. This rate of
irrigation efficiency implies that the average consumption of irrigation water in the
country is high compared to corresponding worldwide usage (Keshavarz et al.,
2005).
A research conducted by the Iranian Agricultural Engineering Research Institute in
1999 revealed that the consumption efficiency depended on farm management,
method of irrigation, growth stage, and type of crop. Furthermore, consumption
efficiency was found to vary between 24.7 and 55.7 percent. The overall irrigation
including transmission efficiency was found to vary between 15 and 36 percent
(Keshavarz et al., 2005).
The study also pointed out that increasing the economic value of water is one of the
major objectives in the Iran Economic Development Program. Increase in the
economic value of water is possible when the yield or return per specific volume of
14
water increases. For this reason it is preferable to use the available water supply for
producing commodities with higher economic efficiency, and/or to use it in regions
where its return is of greater economic value.
Not paying the real price of water, may convey the demander the illusion that the
value of water is really the low price they are paying. Hence, the cheapness of water
would create no incentive for the economization of water. Furthermore, it wouldn’t
encourage the firms to invest in water sources, since the profit is very low. Studies
has indicated that underpricing of irrigation water results in an inefficient use of
scarce water resources. In other words, water is squandered due to underpricing.
Based on current regulations, the price of regulated surface water is between 1 and 3
percent of the value of the cultivated crops. A chronicle of Iranian irrigation pricing
policies, since 1937 till now is shown in Table 1-4.
Table 1-4 Time process of irrigation fee in Iran
Date
Pricing System
1937
1942
1954
1967
1971
1981
1985
1986
1988
1999
Politic - Social
Politic - Social
Economic - Social
Economic - Social
Economic - Social
Economic - Social
Economic - Social
Politic - Social
Politic - Social
Economic - Social
Computing Method
of water rate
Stone Four Stream
Farmer Situation
Volumetric (m3)
Volumetric (m3)
Volumetric (m3)
Volumetric (m3)
Volumetric (m3)
Hectare
Hectare
Volumetric (m3)
Identification Base of water rate
Operating cost and maintenance
Local price
Average cost
Average cost
Average cost
Average cost
Average cost
Per four hectare, $ 0.015
Four percent of product
1-3 percent of value of the
cultivated crop
Source: Iran Water Resources Management Company (2008)
Social and political water pricing is used to cover maintenance and operation costs.
The purpose of the method is to supply cheap water to farmers, financial
15
independence to manager and protection to consumers. Government in this way
accepts investment costs while the manager does not participate in operational costs.
This way, the price of water are kept extremely low and used to increase agricultural
income from irrigated areas. This pricing method is not related to actual costs.
Managers in such systems receive their remuneration through operational costs. In
cases where operation and maintenance costs are not paid, government usually
reimburses the system through subsidy.
In the system, water rate is determined to cover all or part of the investment costs
(with interest or without interest) including maintenance and operational cost. The
purpose of the method is to encourage farmers to participate in investment.
Essentially, the investment done by the government will be reimbursed by farmers
through pricing. The reimbursement is done in two ways
A) Repay part of the total investment done
B) To repay the total used investment
Based on the above discussions, determination of crop water requirements and
cropping pattern for each region and specification of volumetric allocation of water
have been considered to be the major objectives in increasing the economic value of
water.
Price mechanism is applied to balance production and consumption of economical
goods and services. Being an economical good, irrigation water must be optimally
allocated based on economic theories and price mechanism. Thus, planning for
efficient use and conservation of water resources is of special interest, and in this
16
regard, prices can play an effective and an important role in achieving and
maintaining equilibrium between supply and demand of water. Determination of per
cubic meter price for water, and establishing a suitable allocation of it among
different activities such as agriculture, industry and urban use have always been one
of the most fundamental problems economists, policy makers, and planners in the
water and agricultural sector have been encountering.
According to Latinopoulos et al., (2004) in developed countries, the price of
agricultural water is far below its economic value and farmers often pay for water
little or nothing at all. Consequently, farmers have little incentive to conserve water
or avoid from cultivation of highly water-consuming crops. Apart from politics, the
other critical factor equally contributing to the inefficiency of water allocation is the
conspicuous lack of suitable pricing of irrigation water (Johansson, 2000;
Latinopoulos et al., 2004).
Based on the available agricultural statistics in the year 2001, 93% of gathered water
in Iran was consumed by the agricultural sector. Despite the huge demand of this
sector for water, the price for water that farmers pay is very low. A considerable
number of studies have been carried out to measure the value of water in the
agricultural sector. All the studies agree on the fact that the paid price for water is
lower than the real value to farmers. An initial estimation shows that the average
water price in modern irrigation networks is only $ 0.0003 per m3. At the equilibrium
condition, the water price must be equal to the value of an increase in yield on one
side and the increase in gathered water cost on the other side. Price can and has been
17
used to equilibrate supply and demand. Nonetheless, unfortunately, the pricing
system has not been exerted to solve the water problem in Iran’s agricultural sector.
1.5
Research Questions
Previous studies shows that the water usage in the agricultural sector of development
countries is inefficient. Hence, to improve efficiency, considerable improvement is
necessary through policy changes, including reforming the existing water pricing
structure. There are many factors that determine an effective pricing mechanism for
irrigation water. This research focuses on issues related the estimates of water
demand for the agricultural sector in Iran, the price elasticity of irrigation water and
effective pricing policies. Hence, this research will find answers to the following
questions:
1. Are the current prices in Iran’s agricultural sector effective?
2. Would an increase in irrigation fee have an effect on decreasing the
agricultural water usage?
3. Which pricing mechanism (pricing based on AVC, MC, and VMP) is
efficient for irrigation water in Iran’s agricultural sector?
4. Can water price play an effective role in determining the quantity of irrigation
water consumed?
5. Is the price elasticity of demand for irrigation water greater than one in Iran’s
agricultural sector?
6. Is there a positive relationship between crop quantity and demand for water?
18
1.6
Research Hypotheses
The hypotheses of this research include:
1. Price of Irrigation water is inefficient. In other words, water demand is
inelastic with respect to its current price.
2. Price elasticity of water in Iran’s agricultural sector is negative and less than
one.
3. In Iran’s agricultural sector, pricing structure, based on marginal cost and
average variable cost of water (from water supplier), has no effect on
decreasing water usage by farmers.
4. In Iran’s agricultural sector, pricing structure, based on value of marginal
product, has the effect of decreasing water usage.
5. In Iran’s agricultural sector, crop quantity has a significant effect on water
usage.
1.7
Research Objectives
The main objective of this research is to analyze alternative pricing mechanisms and
to determine its impact on water demand. The specific objectives are:
1. To investigate the existing water pricing policy
2.
To estimate water demand functions according to current price, Value of
Marginal Product (VMP), Average Variable Cost (AVC) and Short Run
Marginal Cost (SRMC).
3. To analyze the impact of alternative pricing mechanism on the demand for
irrigation water
19
4. To recommend a suitable mechanism for determining an efficient pricing
system for irrigation water
1.8
Significance of the study
As aforementioned, Iran is located in arid and semi-arid areas, and Iranian people are
encountering insufficient supply of water. The demand for water increases due to
population growth and economic activities especially in agriculture. However, the
supply of water has remained unchanged resulting in shortage or scarcity of water for
future needs. On one hand, the agricultural sector is one of the most important
economic sectors of Iran; on the other hand, water is the most limiting factor for
production. In fact, close to 93 percent of the renewable water in Iran is used in the
agricultural sector. Unfortunately, due to cheapness of water and losses through
leakage, evaporation, and etc., only 30 to 40 percent of the irrigation water (of the
93%) is effectively available for crop production.
Iranian authorities are well aware of this problem and have always been conscious of
the urgent need to improve the efficient application of water in agriculture by proper
policy changes, technological solutions, and other alternatives. According to (Molle
& Berkoff, 2007) water pricing mechanism can be used as primary means for
regulating irrigation water consumption, cost recovery, and conservation. Hence, a
study on the irrigation water pricing is necessary to achieve an efficient application
of irrigation water in Iran.
This study is able to fill some of the gaps present in researches conducted previously.
First, it provides consistent and comprehensive estimates of water prices (values)
20
across several regions of Iran. Second, this study reveals that kind and value of crops,
and geographical factors have significant effect on water demand and price. Finally,
this study shows that with current prices, farmers are not using water efficiently.
So far, no such study on the irrigation water pricing has been conducted in Iran. It
follows that reform in agricultural water pricing can significantly save a large
amount of water. Indeed, low water prices generally lead to waste while higher prices
promotes conservation. Results of the study would be important in highlighting that
water is wasted because it is underpriced.
1.9
Organization of the study
This study is organized into five chapters. The current chapter serves as the
introduction of the subject matter and the problem statement of the study. It also
highlighted the importance of the study. It also offers the outline of the study.
Chapter two is a comprehensive review of water pricing mechanism based on
relevant literatures. More so, the existing water pricing practice in Iran is described
in the chapter. The method of analysis adopted in this study will equally be
elaborated in the chapter also. Chapter three discusses the theoretical framework
represented in the study. It includes a conditional derived demand for irrigated water
based on cost, description, justification of the variables involved, experimental model
designs, and methods used to achieve the stated objectives.
Chapter four presents an econometric model of panel data used to estimate the
demand for irrigated water for wheat, barley, lentil, pea, pinto bean, onion, tomato,
21
potato, cucumber, water melon, cotton, and sugar beet. Next, the inputs price
elasticity and crop amount are extracted and key information highlighted in the form
of text and tables. Chapter four also presents the results, interpretations and
conclusions associated with them. Finally, the chapter also computes input price
elasticity and crop quantity based on average variable cost, marginal cost and value
of marginal production. The final chapter presents a summary of findings, policy
implication and conclusion of the study. Recommendations for future researches are
also presented.
22
CHAPTER II
2
2.1
LITERATURE REVIEW
General Survey of Water Pricing
Water pricing is a means of achieving one or more usage-policy purposes. According
to (Molle & Berkoff, 2007), water charge is capable of acting as a financial means
directed to recover all or part of the recurrent and capital costs. It’s noteworthy that
recovery of recurrent costs is critical, especially to keep the physical integrity of the
system when money that belongs to the government is to be used for public goods.
Moreover, water charge can be an economic tool designed to preserve water and
raise its productivity by promoting, (i) cautious management and water conservation;
(ii) cultivation of crops requiring less water, and investments in water-saving
technologies; and (iii) diverting water to high value agriculture and/or other sectors.
Finally, a charge may also be an environmental instrument to oppose water pollution
and improve water quality (Molle & Berkoff, 2007).
There has been recently a trend in the world to promote the efficiency of water use.
As an economic good, water can no longer be treated as a free commodity. It has
been proven to be quite necessary that in order to achieve a sustainable, efficient and
environmentally sound water resource development and management, all the user
sectors such as agriculture, industry and household must pay for their water usage.
Water pricing is an important means of improving water allocation and encouraging
users to conserve water resources (Sahibzada, 2002).
23
Several studies have been conducted on the impact of water price on demand for it.
Gottlieb (1963) formulated the effects of income and price on domestic demand for
water. The demand was an aggregation of industrial, commercial, and public uses
served by municipal water systems throughout the Kansas state. He drew the
conclusion that “price increase tends to depress per capita consumption of water
temporarily.”
Gardner and Schick (1964) examined factors affecting urban household uses of water
in the northern Utah. They found that plumbing price and lot size were significant.
North (1967) claimed that the best fitting equation for estimating residential use of
water within the city of Athens must include number of family members, number of
baths in dwellings, existence and area of gardens and lawn areas, and market value of
residences. The income and price elasticities were 0.83 and 0.67 respectively.
Linaweaver et al. (1967) found that water demands throughout a country vary over a
wide range from one season/area to another. Most of the differences between
summer and winter usage of water in residential areas is attributed to lawn irrigation.
Thus, authors separate domestic and sprinkling demands from each other by
subtracting the amount of seasonal variation from the total. Lesser water use in
apartment areas in comparison to single family dwelling residential areas reflects the
relatively smaller lawn areas adjacent to apartment buildings. In areas with metered
public water and sewer systems, higher water use in the west results from the lack of
natural precipitation during summers. In areas with septic tanks, the annual water use
was less than that in areas with public answers. Water use in flat rate areas is more
than twice as great as that in metered block rate areas.
24
Guilbe (1969) found that type of dwelling, and property value, public apartments
were not suitable as a basis to explain and predict water demand. Instead, the number
of bedrooms was found to be more appropriate for this purpose.
Saunders (1969) identified the factors closely related to water usage in urban areas,
both for inter- and intra-community basis. The population, value added, land area,
and number of autos turned out to be the primary variables affecting total water
usage in metropolises.
Burke (1970) developed an econometric model of municipal water requirements
which unified variables reflecting the various factors affecting water demand. He
used log-linear functions and concluded that in New York city, important variables
were estimated precipitation, number of families and people served; whereas, in
California, they were estimated number of retail establishments, value added by
manufacturers, and population being served.
Hanke (1970) investigated the effects of a change from flat rate to a metered one of
the price structure for the residential water demand in Boulder, Colorado. According
to his findings, (i) domestic demand was reduced by 36 percent but stabilized at the
lower level thereafter; (ii) with the introduction of meters, sprinkling demand not
only was substantially reduced but also continued to decline every year during the
study period.
Wong (1972) investigated the impact of income, price and average summer
temperature on consumer’s behavior of demand for water in Chicago. He found that,
25
(i) summer temperature was the most significant variable in both areas; (ii) the
income elasticity obtained from the time series analysis for Chicago was significant
at 2 percent, but not for outside communities; (iii) the price elasticity obtained from a
time series analysis was not significant for Chicago, but was significant at a level of
5 percent for outside communities; (iv) the price elasticity obtained for both areas
from a cross-sectional analysis was significant at a level of 5 percent; however, the
income elasticity was significant at a level of 2 percent only in the two more
populated areas; (v) cross-sectional analysis yielded larger coefficients of income
and price elasticity than those from time series analysis; however, they showed larger
values of standard error of the estimate and lower R2 values.
Morgan (1973) collected information and data about number of persons per dwelling
unit, annual water usage, and market value of a dwelling unit as a function of annual
quantity of demand for domestic water and estimated the function of water demand.
Eliminating water price since it was fixed. He found that the number of people per
dwelling unit to be the primary factor in the determination of domestic water
consumption.
Howe (1982) used two marginal price variables and price difference for estimating
linear function of water demand for both non domestic and domestic usages.
Williams and Suh (1986) estimated the model of water demand separately for
business, household, and industry consumption. They introduced market size and the
level of economic activities as variables in functions for business and industry
demands. In functions for household water demand, they incorporated variables such
26
as price, per capita income, temperature, rainfall, dimension of dwelling as the major
factors.
Aghthe and Billings (1987) estimated a simultaneous equation model of demand to
determine the price elasticity of demand for households within each income group.
They found that higher income households not only used more water but also had
lower elasticity of demand. That is, a uniform proportional rate increase causes a
larger percentage drop in water usage among low income households than among
those with lower income.
Rhodes and Sampath (1988) showed how optimal pricing system depends on the
relative capital intensities between large and small farmers. They also presented a
large volume of literature dealing with irrigation water management in general and
water pricing in particular. In the documents, also are compared six alternative
methods of distribution and pricing irrigation water in developing countries. The
methods were then ranked on the basis of equity in the distribution of income,
allocation efficiency in production, and cost recovery to the supplying authority.
Nieswiadomy and Molina (1991) indicated that households are responding to
marginal price when facing with increasing block rate structures, and to average
price when facing with decreasing block rates.
Tsur and Dinar (2002) highlighted the equity and efficiency performance of various
irrigation water pricing methods. A review of water pricing practices enabled them to
arrive at two conclusions, (i) water pricing can improve income reallocation only if
27
water quota rules are enforced; (ii) water pricing methods that affect the demand for
irrigation water ensure its efficient use.
These studies are mainly on the characteristics of different price systems. Being
based on a variety of case studies on the execution of water pricing reforms, the
studies indicate that there are many difficulties in executing more efficient pricing
rules. Theoretical water pricing models are, however, rare and mostly scattered in the
scientific literature. These studies are also important to water public service
managers and to water supply industry regulators who have to establish exact water
pricing systems for particular conditions in which their customers operate.
Ghazali et al. (2010) recently published a paper titled review of water pricing
theories and related models which developed a wide range of methods for pricing
water.
2.2
Pricing Methods of Irrigation Water
According to Johansson (2000), the basic role of prices is to help distribute rare
resources among competing users and uses. Pricing of water affects distribution
considerations by different user. He stated that a diversity of methods for water
pricing has arisen which are dependent on economic and natural conditions.
The usual pricing methods for irrigation water include volumetric pricing, nonvolumetric pricing methods (such as output pricing and input pricing, or area
pricing), market-based pricing methods, and quotas. These methods often result from
insufficient information regarding actual consumption amounts. Volumetric pricing
28
mechanisms charge for irrigation water based on use of actual volume of water
(Johansson, 2000). He also indicated that non-volumetric methods charge for
irrigation water is based on land values, or per area or per output/input basis.
However, the mechanisms of pricing based on market are needed to address waterpricing inefficiencies intrinsic to existing irrigation institutions. Market-based
mechanisms rely on market pressures and nicely specified water rights to determine
the irrigation water price (Johansson, 2000). The efficiency, equity and
implementation costs associated with these practices are summarized in Table 2-1.
2.2.1
Volumetric Pricing Method
According to Johansson (2000), volumetric pricing mechanisms charge for irrigation
water using a measurement of the quantity of water consumed. He added that such
mechanisms call for information on the volume of water used by each user. He also
pointed out that water meters make volumetric pricing straightforward, involving
routine maintenance and periodic meter readings. Volumetric pricing costs are fairly
high and entail water user association, or require a central water authority to set a
price, collect fees, and monitor usage. Volumetric approach consists of, (i) indirect
calculations based on measurement of minutes of known flow (as from a reservoir)
or minutes of uncertain flow (proportion of the flow of a river); and (ii) charges for a
given minimal volume even if it is not actually consumed. Volumetric charges are
fairly common in groundwater irrigation systems (Sahibzada, 2002).
29
Table 2-1 Comparison of pricing methods
Pricing
scheme
Potential
efficiency
Time
Equity
horizon of
efficiency
Short-run User-pays
Fairness
Implementati
on costs
Characteristics
Single-rate
volumetric
First-best
Complicated
Requires water
use monitoring
Multi-rate
Volumetric
(Tiered)
Two-part
First-best
Short-run
Can be
used to
target ...
Relatively
complicated
Requires water
use monitoring
First-best
Long-run
As above
As above
Output/input Secondbest
Per area
Secondbest
Short-run
As above
Short-run/
long-run
As above
Relatively
complicated
Less
complicated
Easy
Quotas
Short-run
As above
Easy
Requires cost
and benefit
information ...
Requires
developed
water ....
Water
markets
First-best
(when
tradable)
First-best
Short-run/
long-run
Depends
Difficult
on type of
market
Source: Johansson et al. 2002 (Adapted from Tsur and Dinar ,1995).
Requires
input/output...
Requires
cropping ...
He also pointed out that the optimal volumetric pricing rule requires that water price
is set equal to marginal cost of water supply. In the absence of a water market, a
water user organization or central water authority is needed to monitor use, set the
price, and collect fees. The implementation cost associated with volumetric pricing is
relatively high. In fact, an optimal volumetric price would equal, (i) in the absence of
implementation costs, the (variable) cost of supply equal the cost of delivery; (ii) in
the presence of implementation costs, the marginal delivery cost plus marginal
implementation cost.
In multi-rate volumetric method (tiered pricing), water rates change as the amount of
water consumed exceeds certain threshold values. It is usual when water supply or
30
demand have periodic daily/seasonal variations. During periods of excess demand,
the water price accounts for water scarcity, and is increased by the scarcity rent. On
the other hand, during periods of excess supply, setting the water price equal to the
marginal cost of supply would result in (short-run) efficiency.
On the other side is a two-part tariff pricing method, which includes two elements, (i)
an access charge (aims to recover, at a minimum, the fixed costs of the service
associated with supplying the user); (ii) a constant marginal price per unit of water
purchased (volumetric marginal cost pricing). This pricing method has been
supported, and practiced, in situations where a public utility produces with marginal
cost below average cost, and needs to recover variable and fixed costs (Johansson et
al., 2002).
According to Sahibzada (2002), the two - part tariff includes two main components, a
volumetric charge and an access charge. The main objective of the two-part tariff is
to achieve an efficient pricing which addresses both demand and supply sides.
Supply efficiency needs sufficient recuperation of costs to sustain the provision of
services required by customers. Demand efficiency requires that customers are
charged not less nor more than the cost of producing a unit of service for them.
Brazil uses the two - part tariff systems (a volumetric water charge to recover
operation and maintenance costs, and a per hectare water charge to recover the public
investment in off - farm irrigation infrastructure) in water pricing. In addition, the
government is allowed to recover from small farmers public investments to acquire
land and on - farm equipment (Sahibzada, 2002).
31
The two-part tariff system is an ideal pricing system for the water sector when the
objective is cost improvement and financial sustainability of the system.
Nevertheless, it suffers from a potential inefficiency effect if the access charge
exceeds the net benefit of the service for an individual user (which might happen for
small users). In such a case, the user will withdraw from the system since they are
better off without it even though the marginal cost is less than the marginal benefit of
the service (Sahibzade, 2002).
He also indicated that irrigation water charges is comprises a two - part tariff
containing a volumetric water charge to recover maintenance and operation costs,
and a per hectare water charge to recover the public investment on an off - farm
irrigation foundation.
2.2.2
Non – Volumetric Pricing Method
Non-volumetric pricing is utilized when volumetric pricing is impossible. As a
mentioned above, several pricing methods have been proposed for irrigation service
in practice: per area pricing, per output pricing (which calls for knowledge of user’s
outputs, but obviates the need for water use measurement), per input pricing (which
charges users for water consumption through a tax on inputs, for example per unit
charge for each kilogram of fertilizer purchased), and betterment levy pricing
(Johansson et al, 2002).
32
2.3.2.1 The Output Pricing Method
According to Tsur and Dinar (2002), output pricing procedures charge farmers a
water fee for each unit of output they produce. Thus, output pricing requires data
from the output level of each farmer. Its advantage is that it does away with the need
for measuring individual water consumption which is a more costly and unachievable
task in many regions (particularly in developing countries).
When implementation costs are zero, the allocation earned under output pricing is
second - best, and output pricing is inferior to volumetric pricing (that is, it produces
a lower social benefit). On the other hand, when implementation costs are not zero,
the allocation earned under output pricing is the first - best allocation.
2.3.2.2 The Input Pricing Method
Input pricing procedures charge water use via taxing inputs. Irrigators pay a water
fee per each unit of a certain input use (for example, fertilizer).
2.3.2.3 The Area Pricing Method
Area pricing charge for water is the most usual procedure of irrigation water pricing
found. Under this pricing mechanism, farmers are charged for water used per
irrigated area. The procedure often depends on the type and quantity of crops
supplied with water, the season, and some other factors. Rates are typically greater
for pumped water from storage than for gravity flow from stream diversion
(Johansson, 2000).
33
Area pricing is easy to perform and administer and does not require water
transmission facilities to be metered. This procedure requires only farm size data (if a
unified fee is used) or only land - by - crop data (if the per hectare water fees vary
across crops). Simplicity and low agency costs associated with this method are
reflected in its (Johansson et al., 2002).
In area pricing, for the right to receive irrigation water, users pay a fixed per
hectare/acre fee that, once paid, can no longer affect decisions regarding input and
output, but can affect the choice of crop or persuade some farmers to switch to unirrigated farming. For users who pay the water fee, the demand for irrigation water is
greater than it would be under marginal cost pricing, and the resulting water
allocation is inefficient. However, the execution costs connected with per area
pricing are lower than those associated with volumetric or output pricing. Thus, area
pricing may well generate a higher social benefit (Johansson et al., 2002).
2.2.3
The Market - based Pricing Method
Tsur and Dinar (2002) indicated that water markets are of different forms throughout
the world, in industrial and developing countries. They also pointed out that water
markets may be formal or informal, organized or natural. Their participants may
trade water on the spot or for future delivery, or they may trade water rights (for
example, the right to purchase some quantities of water at a particular price during
specific periods of time).
They also indicated that, in a stylized water market, yearly, each farmer is given a
water endowment and is free to sell or buy such shares of entitlements from other
34
farmers at the going rate. Such entitlement may be based on historical or legal rights,
or they may be set by a deputed or assigned committee or water agency.
In the absence of implementation costs, the basic premise of modern economics is
that under certain conditions (that include a competitive environment, fully informed
agents operating under certainty, no externalities, and no increasing returns to scale
in production), markets achieve a first - best efficiency (Johansson, 2000). In the case
of water, due to expensiveness of its transport, water markets tend to be localized,
consisting of a limited number of participants some of whom may be able to
influence outcomes.
Meanwhile, water markets induce the transfer of water from less productive to more
productive farmers, and eliminate corruption incentives to which centralized
allocation mechanisms are more sensitive.
Water supply is often uncertain; thus, water resources may be shared by many users
who inflict externalities on one another. Finally, water supply systems may exhibit
increasing returns to scale. Hence, water markets are unlikely to achieve a first - best
allocation in practice. Nowadays, special attention is paid to the use of markets in
water allocation due to emerging scarcity and inefficient allocation and use of water
resources. An increasing number of studies have focused on specific policy
implications of water market and trading.
Dinar et al. (1997) stated that market-based allocation is considered economically
efficient from both individual and social viewpoints. They argued that under certain
35
conditions, market mechanism is able to secure water supply for high-value uses in
different sectors without need for developing new, costly water resources. They
claimed that such conditions include, defining the original allocation of water rights,
creating the institutional and legal frameworks for trade, and investing in basic
necessary infrastructure to allow water transfers. They also indicated that by
allowing compensation for water sold by low-value uses, water markets provide a
stimulus for more efficient water use.
According to Sahibzada (2002), there are cases of both informal (when recharge is
adequate and there are a sufficient number of sellers in the market) and formal (when
the legalizing of water trading and recording of water-use right exist) water markets.
Most informal markets are located in the irrigated areas of South Asia while the
formal ones can be found in North and South America.
Easter et al., (1997) stated six essential arrangements for an efficient, sustainable and
equitable water market. When such arrangements are distorted, achieving first - best
allocations is unlikely. For example, when the development of water markets is
generally localized (for water is expensive to transport), the number of users and
suppliers is limited which, in turn, may lead to a non-competitive market, and may
preclude first-best allocations. By the way, when first-best conditions are distorted,
second-best market allocations may surpass volumetric pricing in efficiency.
The essential arrangements mentioned above requires, (i) a management
organization; (ii) institutional arrangements that set up water rights as detachable
from land rights (iii) an effective resolution mechanism; (iv) a flexible infrastructure;
36
and, (v) internalization of externalities. Besides, equity concerns, such as future and
social goals, need to be addressed.
2.2.4
The Quotas Method
In this method, water user (farmer) will have to pay some part of their productions to
the water authority (water sellers) as water fee. Johansson et al. (2002), stated that
the quotas method is efficient when using base prices on the marginal cost of
acquiring more water plus its rareness value. They also argued that incidental prices
based on marginal costs are often too high for low income farmers. They alleged that
quota allotments are, in many cases, used to mitigate fairness or resource
management issues that arise with a water market or marginal cost pricing.
2.3
Pricing Mechanism and Accomplished Studies in Iran
In every region of Iran, water rate is determined based on the aforementioned
theories, and economic, social and geographical conditions.
2.3.1
Urban Water
Methods usually employed to determine urban water price will be discussed below.
2.4.1.1 One-part tariff and multi-part tariff
The simplest form of water tariff is a fixed monthly (flat-rate), as well as another
form of water rate which is in order to consumption water unit that measure by
meter.
37
Multi-part tariff comprises two, three or more components, and also additional
components such as annual evaluation on the basis of minimum asset value, water
rate, fixed water rate and etc.
The most common type of tariff is a two – parted one with flat rate. Using this type
of tariff, without reference to consumption in the past, season or region, is cause
parsimony in subscriber water consumption.
2.4.1.2 Declining Block Rate or Promotional Pricing
In this type of rate, water consumption per bill is divided into some blocks, and
distinct prices for each block are determined. The prices decline with use increase
and form declining block rate and water rates decline consecutively. Using this kind
of rate does not create motive for parsimony.
2.4.1.3 Incremental Block Pricing
Under this type of rate, with an increase in consumption block price is increased in
the same direction. This tariff may be when user pays one price for total usage, but
with usage increase, price for upper block is increased. Implicit assumption of this
type of rate is recognition of remunerative consumer and justice in payment for
water.
38
2.4.1.4 Mixed Block Rate
This type of rate is a combination of promotional pricing and incremental block
pricing as its components. Normally, prices are firstly ascending and then
descending. In this sort of rate, there always is a minimum payment. Block rate and
prices are declining for users with under - consumption and increasing for those with
over - consumption.
2.4.1.5 Seasonal Rate
Demand for water and its production cost varies all year round (demand is increased
when weather is warm and dry) and water authorities offer various prices for
different seasons. For example, in summer water authorities use higher prices to
encourage consumers to decrease their consumption of water. Using various rates in
summer is the most effective method in comparison with using maximum prices in
this season. Whereas seasonal price differential reflects seasonal change of
parsimony costs, rates could be a strong motive for economical return, conservation
and justice.
2.4.1.6 Accomplished Studies
Kollahi (1991) estimated the potable water demand in Shiraz city. He used time
series data for estimation of demand functions such as total demand, domestic
demand, non-domestic demand and demand for various seasons. He used time series
- cross section statistics for variables such as number per dwelling unit of people in
house area and court yard area. Results showed that consumption of potable water
for domestic and non-domestic purposes in various seasons has been more than
39
quantity of minimal water necessary for living. On the other hand, water demand has
a positive relationship with house area and a negative one with number per dwelling
unit of people in court yard area.
Saeid nia (1996) estimated the demand function of urban water for Qom city. He
used in his estimation variables such as water consumption, water average price,
number of subscriber and house income average. He found that which price elasticity
and water income were negative.
Sadr et al. (1994) estimated the quantity of water demand for Tehran city with the
assumption that it was a function of water price, income, number of subscribers and
temperature. They used statistic and information of time series for a period of 14 year
and 10 seasons. They found that variables such as income and temperature were not
significant, and quantity of water demanded had a direct relationship with number of
subscribers and an indirect one with price.
Hassanli (2006) analyzed the cost and water values of components by considering
water as an economical good, and measured the volume of water used for citrus
production by drip system in Darab region in Iran. He found that the real value of
water is much more than the final water cost, and suggested a price which indicates
the high value of water in the study region. He obtained water value, final water cost,
the value of marginal production of water, and a price for water. His suggested prices
for water were 824, 319.5, 374.4 and 55 Rials/m3, respectively.
40
Jafari (2007) provided an overview of the theoretical issues and operational models
for estimating the value and cost of water in Alavian Dam, Iran. He concluded that
the full cost of water for dam and the irrigation network would be 92 and 182
Rials/m3 respectively with an average of 174 Rial/m3.
2.3.2
Agricultural Water
As mentioned in chapter one, Iran is facing with drought and agriculture sector uses
about 93 percent of irrigation water in Iran. Water is used for crop irrigation, and to
leach salt from soils, and to regulate crop temperature. The system of agriculture in
Iran uses surface and ground water resources prominently. Users have no control
over the availability and quantity of water, and there is no a market for trading
ground/surface water in Iran.
According to Keshavarz et al. (2005), half of the fully irrigated areas are equipped
with modern irrigation systems and are operated by governmental organizations; the
rest are operated by the private sector, with groundwater as resources.
Current prices for agricultural water depends on crops, source kind of water supply
(surface or groundwater), and area to be irrigated. In Iran’s agricultural sector, the
current price of irrigation water is derived from the value of the crops to which it is
applied. For instance, for traditional networks, semi - modern irrigation networks and
modern irrigation networks, water rates are one, two and three percents from the
value of the crops respectively. Hence, this method needs information on prices of
outputs in each province. Its advantage is that it does not require measurement of
41
individual’s water consumption which is very expensive or even impossible in many
regions.
Private sector supplies water from groundwater resources, and charges farmers a
water fee based on abstraction charge (almost one percent of the value of the crops)
which is determined by government. In determining the real price of water,
compensation costs (in groundwater) and cost of pumping water should be taken into
account.
Although water prices have gone up from time to time during recent decades, they
have never risen as fast as the prices received for agricultural commodities. This
pattern of charges encourages wasteful use of the country’s most limiting resources.
According to Sahibzada (2002) noting the growing shortage of water in Iran, a
reasonable water price policy could be the key to rare water resource development.
2.4.2.1 Accomplished Studies
Keramatzade et al. (2006) used linear programming technique for determination of
the economic value of agricultural water in Shirvan Barzo Dam, Iran. They have
been considered the shadow price of water resource as an economic value which was
equal value of marginal product after earned optimal cropping pattern for integrated
farm and horticulture.
Hossein zad & Salami (2005) evaluated the effect of the choice of production
function on the estimated structural parameters. They also revealed the importance of
correct functional form specification to prevent incorrect policy implications.
42
They estimated a number of flexible and inflexible functional forms which are
assumed to represent the wheat production process in Alavian region. Next, they
computed the economic values of water based on the estimated parameters of various
production functions. They compared the economic value of water input which was
derived from parameters of the correctly specified functional form; then, they
indicated that using the econometric criterion and specification of tests, with those of
the inappropriate functional forms, reveals a substantial difference between the
computed values for the water input.
Salami & Mohammad nejad (2002) used the flexible production functions, and
estimated the economic value of irrigation water in Saveh region. They found the
shadow price per cubic meter of irrigation water at farm gate to be 215, 386, 342 and
265 Rials for use in wheat, cotton, cucumbers, and pomegranate productions,
respectively. A comparison of estimated and the current price of water shows that the
economic value of irrigation water is much higher than what is currently received by
the local water authorities. Under these circumstances, an inefficient use of water and
the lack of incentives in investing in water saving technology are expected.
Hossein zad et al. (2007) stated that some inputs in agricultural products are quasifixed in nature (like water in Iran), and there is no defined market for them. These
input prices were not determined by market and there were no suitable and efficient
price for them. Next, they employed non-market methods for determining the
economic value (real price) of water for staple crops (wheat and onion) in the
Maragheh- Bonab Plain using flexible production functions approach. They found
43
that the economic value of water used in wheat and onion production were almost
248 and 291 Rials per cubic meter, respectively. They concluded that the estimated
prices are much higher than local prices of these inputs.
Chizari et al. (2006) used goal programming approach for determination of optimum
cropping pattern and economic value of irrigation water in three regions in Shirvan
Barzo Dam located in the north of Khorasan province. They computed the economic
value or shadow price of water with sensitivity analysis of constraints. According to
their study the estimated economic value of water ranges between 56 and 2227 Rials.
Using an engineering economic approach, Mansouri & Ghiasi (2002) estimated the
cost of irrigation water at the point of reservoir dams in Azarbaijan Gharbi province
for the two year period of 1998-99. They found the real price of water to be much
higher than that applied by water authorities. They pointed out that the mentality
behind such changes should be towards promoting water from a free input, as it is
considered by many at present, as a commodity of economic value.
Asadi et al. (2007) in their seminal study, estimated (1) the value of marginal
product of irrigation water, (2) the cost of irrigation water, (3) the cost of production
per hectare for different groups, (4) the price elasticity of water demand; and (5) the
water price by Gardner method in Ghazvin plain in 1995. The irrigation area under
Taleghan Dam was divided into five homogenous regions on the basis of the cost for
providing water and the length of the water canal. The required data were then
obtained from 127 farmers who were selected from 24 villages of Ghazvin plain
using a suitable sampling method. In this study, water demand function and the
44
values of marginal production will be estimated by linear programming, econometric
and engineering economic methods. Results show that the price elasticity is less than
one, and demand for water relative to its price tends to be inelastic. The marginal
production value of water was more than the price received by authorities. Average
price of agriculture water is estimated about 65 rials per (m3) by Gardner method.
The marginal production value of water for users groups (<10ha) in five regions
estimated 65,148,190,230 and 102 rials while for farmers groups (>10 ha) was 208,
113, 77, 69 and 120 rials, respectively.
Zare (2006) estimated the suitable production function for economic value
calculation per unit of groundwater in Kerman province. He also estimated the cost
and social welfare functions, and calculated the side effects of excessive pumping
using social welfare function was earned. Lastly, using production function, he
calculated the demand elasticity for input water. He concluded that the best way to
increase irrigation efficiency is to promote efficient irrigation methods.
2.4
Discussion and Deduction
As mentioned in chapter one, the main objective of this study is to analyse current
pricing mechanism, and to create an appropriate mechanism for determining an
efficient pricing system in Iran agricultural water. Therefore, the demand side of
agriculture water will be focused upon, and an irrigation water demand function via
minimization of cost function will be derived. The estimated coefficients of irrigation
water demand functions will be used for analysing current water pricing system and
alternative water pricing systems. The estimated elasticities of irrigation water will
be used to compute marginal value product of water.
45
On the supply side, water pricing system based on average and marginal cost will be
used. Secondary data on expenditures of a district on irrigation water supply delivery
will be used to calculate average and marginl cost per cubic meter of water. The
average variable cost of water supply delivery and has been calculated using
operating and maintaining expenditures on the irrigation system. Estimated
agricultural water use, calculated under each price alternative, can then be compared
with actual water use derived from secondary data.
As mentioned above, there have several water pricing system for irrigation water:
average cost pricing, marginal cost pricing (short run and long run or two part
tariffs), pricing based on value of marginal product and etc.
Average cost pricing is one accepted pattern for recovery of the partial or full cost of
the irrigation works. According to Sahibzada (2002), it is called the cost of service
approach which to public utility rates has both an economic and an equitable appeal.
In this approach farmers should be charged only a quantity sufficient to cover the
outlay incurred in providing service. There are two variants of this approach: 1)
Charging rates which cover only current maintenance and repair and used for as
partial cost recovery or the rock bottom variant. 2) Full cost recovery insists on
charges which not only cover maintenance but also yield a depreciation allowance
and some net return on the historical capital costs of the canal (Sahibzada, 2002).
He also illustrated that average cost pricing involve inefficiencies in water use. Lewis
(1969) cited by Sahibzada (2002) claimed that average cost pricing would mean that
a cultivator using an extra unit of water for crop production would be charged less
46
for it than it costs the community to provide. He also pointed out that this pricing
takes only the supply side into account and ignores the demand side, and its
application under both increasing (lead to profits) and decreasing (lead to
subsidization) average costs leads to inefficient outcomes.
According to Tsur et al. (2004), average cost relaxes the need to use public funds, but
entails an efficiency loss in the irrigation sector. In addition, the farmers carry most
of the burden of the welfare loss. Therefore, according to the problem mentioned in
chapter one, I think that this criterion is not appropriate to economic decision making
in Iran.
Marginal cost pricing is another criterion that is adopted for determining rates in the
irrigation water. Marginal cost pricing sets the price of irrigation water equal to the
marginal cost of providing it or incremental costs associated with incremental
production. According to Dinar et al. (1997) a marginal cost pricing mechanism,
targets a price for water to equal the marginal cost of supplying the last unit of that
water. One of the most important advantages of this pricing is that it is theoretically
efficient. But Dinar et al. (1997) and Sahibzada (2002) in their studies showed that
using of this pricing system confronts some practical problems such as: 1) Marginal
cost alters with the nature of the irrigation decision with which the irrigation methods
are concerned. 2) The marginal cost varies with the period over which it is measured
(like seasonal differences and short - run vs. long - run) and space (the tail end and
near to the source of water supply) which will require that different prices be charged
at different times. 3) This method is difficult to estimate and apply in real conditions.
Therefore, pricing based on marginal cost as a result from, would necessitate
47
charging varying prices within a single irrigation system and also overtime.
Consequently, I believe that this criterion is not appropriate to economic decision
making in Iran, too.
Another famously accepted criterion for determining rates in the agriculture water
sector is pricing system based on value of marginal product of irrigation water.
According to Sahibzada (2002), in this method, prices will be just low enough so that
all water available is used, but just high enough so that no farmer wants more
irrigation water at the price facing him. As mentioned later, on the value of marginal
product water which at equilibrium, this will be equal to the price farmers are willing
to pay for water.
On the other hand, Shiferaw et al. (2008) pointed out where no market price exists,
optimal allocation of irrigation water will require the shadow price to be equal to its
marginal value product. According to Dinar et al. (1997) an allocation which equates
water’s unit price (the water’s marginal value product) with the marginal cost is
considered an economically efficient, or socially optimal, allocation of water
resources. Therefore, I think that this criterion ought to be more appropriate to
economic decision making in Iran.
Five methods of estimating the marginal product value of water include;
(i)
The residual imputation which deducts from gross product value
the costs of inputs other than water, and then, attributes the whole
of the remainder to the water input.
48
(ii)
The linear programming method which is well suited to estimate
the marginal value of water.
(iii)
The production function method which is used to derive the
marginal product value of water. In this method, The first
estimates a crop-water production function from field trials and
then scales this physical production function by the price of the
product (Colby, 1989) ; (Penzhorn & Marais, 1998); (Conradie &
Hoag, 2004).
(iv)
The derived demand function method which is to estimate a
demand function directly from water price data. Griffin and Perry
(1985) presented an econometric model using panel data of
irrigation prices (volumetric and flat rate water charges) in Texas.
(v)
The fifth approach is to use Hedonic pricing methods to measure
the contribution of water value to farm prices.
The next chapter will start with the presentation of the theoretical framework,
which will specify the foundation of the study. The first section of the chapter will
also illustrate and compare various pricing models that have been mentioned above.
The procedure of computing the value of marginal product will also be shown in
Chapter 3.
49
CHAPTER III
3
3.1
METHODOLOGY
Introduction
Economics mainly deals with explaining and foreseeing events. To do so, theoretical
and empirical methods of study are utilized. Even though a blend of the two
approaches is mostly used in practice, they should be differentiated from each other.
On one hand, starting with some assumptions, theoretical approaches use some
abstract inferences, and come up with some conclusions eventually; on the other
hand, empirical approaches of study are, to some extent, inherently inductive. After
all, the two methods are complementary in that theories provide guidelines for
empirical investigations. On the other hand, empirical studies furnish some means for
testing the basic assumptions of theoretical methods. Likewise, they provide a way
for juxtaposing with reality of the deductions made from theoretical studies
(Henderson & Quandt, 1980). In this research the demand functions of irrigation
water will be estimated using Cobb-Douglas functional form and Panel Data
econometric methods.
3.2
Theoretical Framework
In the next sections, the theoretical framework of demand and cost functions will be
discussed first.
50
3.2.1
Demand Function
According to microeconomic theories, there are two kinds of demand functions. The
first one is the demand function of output or consumer’s ordinary demand function
which can be derived from the analysis of utility maximization. The second one is
the input or derived demand function which can be obtained from the analysis of
profit maximization (input unconditional demand) or cost minimization (input
conditional demand).
3.2.1.1 Output Demand Function
According to Henderson and Quant (1980), consumer’s ordinary demand function
expresses the quantity of commodity they buy as a function of commodity price and
their income. The function can be derived from the analysis of utility maximization.
Equation 3-1 shows the typical form of utility function.
Utility  u ( X 1 , X 2 ,.... X n )
Equation 3-1
Likewise, the budget constraint can be written as,
I  P1 X1  P2 X 2 ...  Pn X n
Equation 3-2
Where, I denote income.
The generalization of the Lagrange–multipliers method to n variables can be easily
done if we write the selected variables in subscript notations. The objective function
will then be equation (3-1) subject to constraints of equation (3-2). Thus, the
Lagrangian function will be as follows,
51
L  U  x1 , x 2 ,, x n     I  p1x1  p2 x 2  pn x n 
Equation 3-3
In the output demand function, the first-order conditions for maximization consist of
x1, x2, x3, x4, …, xn and  . The demand functions are obtained by solving this system
for the unknowns x1, x2, x3, …,xn and  . The solutions for x1, x2, x3, …,xn are in terms
of the parameters p1, p2, p3, .., pn and Ī. The quantity of xi as consumer purchases,
depends on their income, and in the general case, on the price of all commodities
(Henderson & Quandt, 1980). The first-order condition will consist of the following
(n+1) simultaneous equations,
L U

  p1  0
x1 x1
L U

  p2  0
x2 x2
.
.
Equation 3-4
.
L U

  pn  0
xn xn
L
 I  p1 x1  p2 x2  ...  pn xn  0

For any two goods we have,
U / xi
p
 i
U / x j p j
Equation 3-5
This implies that at the optimal allocation of income,
MRS (xi for x j ) 
pi
pj
Equation 3-6
52

U / xn
U / x1 U / x2

 ... 
p1
p2
pn

MU x1
p1

MU x2
p2
 ... 
Equation 3-7
MU xn
pn
Where,  denotes the marginal utility from an extra amount of consumption
expenditure (the marginal utility of income). At the margin, the price of a commodity
represents the consumer’s evaluation of the utility of the last unit consumed i.e., how
much the consumer is willing to pay for the last unit.
pi 
MU xi

Equation 3-8
Solving for xi , i = 1..n, …, xn gives the demand functions
x1 
I
2 P1
x2 
I
2 P2
.
Equation 3-9
.
.
xn 
I
2 Pn
The demand functions extracted in this fashion are contingent on continued
optimizing behavior by the consumer. The demand curve derived from this function
looks at the relationship between x1 and p1 while keeping p2, p3,..,pn, Ī and preference
unchanged. In other words, it shows the following relationship,
53
x *1  x1  p1 , p2 , , pn , I 
x *2  x1  p1 , p2 , , pn , I 
.
Equation 3-10
.
.
x *n  x1  p1 , p2 , , pn , I 
3.2.1.2 Input Demand Function
In the input demand function, we will discuss production function of a firm or farm.
Given that production is a function of the values of variables x1, x2, x3, x4,…,xn
corresponding to inputs, it can be written as follows,
q  f  x1 , x2 , x3 , x4 ,, xn 
Equation 3-11
Frequently, economists assume that the problem of optimum input combinations has
been solved, and conduct their analysis of a firm in terms of its revenues and costs
expressed as functions of output. The problem of the entrepreneur is then to select an
output maximizing their profit. Equation 3-12 expresses the total cost of production
(C) as a linear function,
C  Ø q  b
Ø  q   ri xi
Equation 3-12
C  ri xi  b
Where ri’s stand for the prices corresponding to of xi’s, and b is the fixed inputs cost
(the cost function vanishes in the long-run). A firm then maximizes output subject to
a cost constraint. It follows from the Lagrange function that,
54
V f
 x1 , x2 , x3 , x4 ,, xn 
   C0  r1 x1  r2 x2    rn xn  b 
Equation 3-13
Where, 0 is an undetermined Lagrange multiplier. Setting the partial derivatives
of V with respect to xi, i = 1,…,n, and  equal to zero will lead to the following set of
equations,
V f

  r1  0
x1 x1
V f

  r2  0
x2 x2
.
.
.
Equation 3-14
V
f

  rn  0
xn xn
V
 C0  r1 x1  r2 x2  ....  rn xn  b  0

Transferring the price terms to the right hand side of the first two equations, then
dividing the first equation by the second one gives the equation below for any two
goods,
f
x i
MPxi f i ri
f


 
x j MPxj f j r j
Equation 3-15
First order conditions state that the rate of technical substitution (RTS) or ratio of the
MP of xi to the MP of xj (MPs ratio of xi and xj) must be equal to their ratio of prices.
Thus, the contribution to output of the last dollar expended upon each input must
equate.
55
A number of special cost relations which are also functions of output level can be
derived from Equation (3-12). Average total cost, average variable cost, and average
fixed cost are defined as the respective total, variable, and fixed costs divided by the
level of output. This is shown in the following equation,
ATC  [Ø  q   b] / q
; AVC  Ø  q  ; AFC 
b / q
Equation 3-16
Where, ATC is the sum of AVC and AFC. Marginal cost is the derivative of total cost
with respect to output; in other words,
MC 
C
q
Equation 3-17
For an entrepreneur who sells their output at a fixed price, revenue is also a function
of the level of their output. It follows that, their profit is a function of the level of
their output too. This fact is shown by the equation below,
  TR  C
  p .q   (q )  b
Equation 3-18
Setting the first derivative with respect to “q” of
in Eq. (3-18) equal to zero yields
the maximized profit, i.e.,

 p   '(q )  0
q
 '(q )  MC
p   '(q )  MC
Equation 3-19
An entrepreneur must equate their MC with their constant selling price of output.
They can increase their profit by expanding their output if the addition to their
56
revenue of selling another unit exceeds the addition to their cost (Huffaker,
Whittlesey, Michelsen, Taylor, & McGuckin, 1998).The second order condition for
profit maximization requires that MC must be increasing at the profit-maximizing
output. Equation (3-20) expresses this fact.
 2 / q 2   2C / q 2  0
Equation 3-20
Multiplying by -1, and inverting yields the inequality  2C / q 2  0 . If MC were
decreasing, the equality of price and MC would give a point of minimum profit.
(Henderson and Quandt, 1980: 87)
The analysis of a firm is easily generalized to cover a production process with s
outputs and n inputs. The production function is stated in an implicit form like
f(q1,q2,…,qs ; x1,x2,…xn). Profit is the difference between total revenue from sale of all
outputs and the expenditure upon all inputs,
   pi qi – rj x j
Equation 3-21
An entrepreneur desires to maximize profit subject to the technical rules given by
production function. Letting,
J   pi qi – rj x j   f  q1 , q2 ,, qs ; x1 , x2 , xn 
and setting each of its s + n + 1 partial derivatives equal to zero yields,
57
Equation 3-22
J / q i  p i   Fi  0
J / x j   r j   Fs  j  0
J /  f  q1 , q 2 , , q s , x1 , x 2 ,  x n   0
Pi / p k  f i / f k  q k / q i
j , k  1, 2, , s
ri / p k   f s  j / f k  q k / x j
or r j  p k .q k / x i
Equation 3-23
i  1, 2,, s ; j  1, 2, ., n
An important duality for the firm exists between the production and cost functions.
Letting a firm’s isoquant q0 defined by q0 = f(x1,x2), and that the first - order
condition for cost minimization for this output be - dx2/dx1 = r1/r2. The input
functions can be derived in the following form,
X 1   1  r1 / r2 , q 0 
Equation 3-24
X 2   2  r1 / r2 , q0 
Now differentiate the cost equation,
C  r1 x1  r2 x2
Equation 3-25
Given of (3-25) and the first order conditions r1 = λ f1:
C / r1  xi   ( f1. 1 / ri  f 2 . 2 / ri )  xi  0
i  1, 2
Equation 3-26
Where,
X1 and X2 are cost - minimizing values expressed as functions of the ratio of the
input prices to the recommended output level.
λ is the Lagrange multiplier in the constrained cost minimization problem.
58
The bracketed term equals  q0 / ri = 0 along the isoquant.
Equation 3-26 is known as Shephared’s lemma. Technically, it is one result of the
envelope theorem (it concerns how the optimal value for a particular function
changes when a parameter of the function changes).
The partial derivates of the cost function with respect to input prices equal the cost minimizing values which for the inputs are as follows,
xi  C  r1 , r2 , q  / ri
Equation 3-27
We know that a firm maximizes profit by selling a quantity for which MC = P. In the
short run, the supply function of a perfectly competitive firm states the quantity that
it will produce as a function of market price. The function can be derived from the
first-order condition for profit maximization. (Henderson & Quandt, 1980:140)
The approach mentioned above is Primal Approach. We can extract supply and
demand functions for inputs directly from profit or cost functions. This approach is
called Duality Approach. In this method, the partial differentiation of the first order
condition of profit function with respect to output price yields output supply which is
a function of output and inputs prices.
  r1 , r2 , q  / P  q  P, r1 , r2 
Equation 3-28
59
While the partial differentiation of the first order condition of profit function with
respect to price of a particular input by using Hotelling’s lemma yields (the negative
of) the corresponding input demand. This demand function is called the
unconditional demand for input, and is a function of output and input prices. As the
profit function itself is homogeneous of degree one, both of the functions described
above are homogenous of degree zero. That is, doubling both output and input prices
will not change the input levels that a firm chooses, nor will this change the firm’s
profit - maximizing output level (Nicholson, 2004).
  r1 , r2 , q  / r1   x1  P, r1 , r2 
Equation 3-29
Finally, the partial differentiation of the first order condition of cost function with
respect to price of a particular factory by using Shephard’s lemma yields the
corresponding factor demand. This demand function is called the conditional
(contingent) demand for input. Conditional factor demand is a function that gives the
optimal demand for each of the several inputs as a function of the output expected,
and the prices of inputs.
  r1 , r2 , q  / r1  x1c  q, r1 , r2 
Equation 3-30
At the profit-maximizing choice for input x1 (for example labour input), these two
concepts agree about the amount of labour hired, that is,
x1c  q, r1 , r2   x1  P, r1 , r2 
Equation 3-31
60
Differentiation of this identity with respect to the market wage yields,
x 1  P , r1 , r2  x 1c q , r1 , r2  x 1c q , r1 , r2  q

[
. ]
r1
r1
q
r1
If
Equation 3-32
q q  P  MC  MC

.
r1
MC
r1
Equation 3-33
Then,
x 1c q x 1c q  P  MC  MC
=
.
.
Out put Effect 
q r1
q
MC
r1
Equation 3-34
This equation shows that the total effect of a change in the input price (wage for
labours employed) can be decomposed into two components: (1) the change in
contingent labour demand, holding q constant (the substitution effect); and (2) the
change in contingent labour demand from a change in the level of output (the output
effect).
Both of these effects mentioned above are negative. The first of these effects is
obviously negative because of the convexity of the firms’ isoquants. The second
effect is clearly negative since: (1)  q/ MC in equation above is negative, that is,
for a given market price a shift upward in the marginal cost curve causes reduction in
production. (2) For a normal good, both  x1c/ q and  MC / r1 are positive. So the
output effect will definitely be negative. But even in the pesky case of an inferior
input, both of these derivatives will be negative, so their product is positive.
Therefore, even for inferior goods the output effect is negative (Nicholson, 2004).
61
An increase in price of input affects both substitution and output since the amount of
input demanded decreases.
3.2.2
Elasticity- A General Survey
After the derivation of the demand function, now it’s possible to derive own
elasticity of demand and also income (output) elasticity of it. The own elasticity of
demand for an input X is defined as the ratio of proportionate rate of change of the
input to the proportionate rate of change of its own price where price and output Q
remains constant. In the other word, the own price elasticity of demand for an input
is defined as the percentage change in the quantity of the input taken from the market
divided by the percentage change in the price of that input. According to Debertin
(2002), the output - price elasticity is the percentage change in the quantity of the
input taken from the market divided by the percentage change in the price of the
output. Using calculus, the own price input demand elasticity is
ex , p x 
x / x
x px


px / px px x
Equation 3-35
“A numerically large value for elasticity means that a quantity is proportionately
very sensitive to price changes. When elasticity is less than negative one (ε11 < -1),
demand is elastic whereas input with numerically small elasticity more than negative
one (ε11 > -1) is demand is inelastic and it is called necessity. Besides, if elasticity is
between negative one and zero (-1 < ε11 < 0), demand is inelastic. However, when
elasticity vanishes (ε11 = 0), demand is infinitely inelastic. On the other hand, when
62
elasticity equals unlimited, demand is infinitely elastic in which case MC = MB and
MNB = 0.
If there were more inputs to the production process than one, both own price and
cross-price elasticities can be defined. A cross-price elasticity of demand for the
derived demand function relates the percentage change in the quantity of input xi
taken from the market divided by the percentage change in the price of input xj. In
other words, the cross-price elasticity of demand measures the rate of response of
quantity demanded of one input (for example hire) to a price change of another input
(for example water). The common formula for the cross–price elasticity of demand is
given by,
 21    lnx 2  /   lnp x  
1
p
x1
 
/ x 2 . x 2 /  p x1

Equation 3-36
According to economic theories, inputs can be technical complements and still
substitute for each other along a downward- sloping isoquant. A simple example of
technical complements in agriculture would be two different kinds of fertilizer
nutrients in corn production. For example, the presence of adequate quantities of
phosphate may make the productivity of nitrogen fertilizer greater.
In the Cobb Douglas type of production functions, an input (x2) is said to be
technically independent of another input if when the use of x2 is increased, the
marginal product of x1 (MPPx1) does not change, but an input (x2) is said to be
technically competitive with another input (x1) if when the use of x2 is increased, the
marginal product of x1 (MPPx1) decreases.
63
Cross–price elasticity may be either positive or negative. If two inputs are substitutes
(that is, sign of elasticity is positive), it should be expected that consumers use more
of one input when the price of its substitute increases. Similarly, if the two inputs are
complements (that is, sign of elasticity is negative), a price rise in one input is
observed to cause the demand for both inputs to fall. According to Nicholson
(2004), change in the price of an input will cause the firm to change its input mix.
That is, k/l changes in response to a change in w/v while holding q constant. In other
words, we wish to examine the derivative along an isoquant.
k 
 
l 
w 
 
v 
Equation 3-37
Putting this in proportional terms as
s
 (k / l ) w / v  ln(k / l )


 (w / v ) k / l  ln(w / v )
Equation 3-38
The equation also indicates an alternative definition of the elasticity of substitution:
(i) in the two-input cases, s must be nonnegative, that is, an increase in w/v will be
met by an increase in k/l; and ii) a numerically large value of s means that firms
change their input mix significantly if input prices change.
The partial elasticity of substitution between two inputs (xi and xj) with prices wi and
wj is given by,
64
sij 
 ( xi / x j ) w j / wi  ln( xi / x j )


 ( w j / wi ) xi / x j  ln( w j / wi )
Equation 3-39
sij is a more flexible concept than s since it allows the firm to adjust the usage of
inputs other than xi and xj when input prices change.
A quantitative elasticity of demand in a derived demand function is defined as
proportionate change in the quantity of an input relative to that in output while prices
remain constant.
i    ln xi  /   ln Q    Q / xi  . (xi / Q)
Equation 3-40
Where, ηi denotes the quantitative elasticity of demand for an input xi. According to
economic theories, quantitative elasticity should be positive meaning that when
demand for an output increases, the demand for input will increase too.
Finally, when a firm maximizes profit in a competitive market, MR = MC, and it can
be seen that,

1
MC  p 1 
 eq , p




Equation 3-41
p  MC
1

p
eq , p
According to Nicholson (2004), the gap between price and marginal cost is an
important measure of inefficient resource allocation as shown below:
65
1. The gap between price and marginal cost is zero, there is no gap (eq,p = - ∞),
the demand curve facing the firm becomes perfectly elastic and resource
allocation is intensively efficient.
2. The demand curve facing the firm is more elastic (eq,p < - 1) the resource
allocation is relatively efficient.
3. The demand curve facing the firm is relatively inelastic (eq,p > - 1), would
imply impossibility and the resource allocation is relatively inefficient.
4. The gap between price and marginal cost is very high (eq,p = 0) the demand
curve facing the firm becomes perfectly inelastic and resource allocation is
intensively inefficient.
The gap between price and marginal cost will fall as the demand curve facing the
firm becomes more elastic. If eq,p > -1, then MC < 0 meaning that firms will choose
to operate only at points on the demand curve where demand is elastic.
3.2.3
Irrigation Water Demand Function
3.2.3.1 Empirical Model
In the estimation of input demand and output supply, different approaches have been
suggested and adopted. Timmer (1974) as cited by Chembezi (1990), identified two
approaches, namely direct and indirect estimations. Indirect approaches involve
derivation of demand functions from agronomic response functions and research.
Direct methods, on the other hand, involve estimation of demand functions directly
from observed market data on input consumption and prices, and also the prices or
66
quantities of farm output. For the purpose of this study, the direct method approach
will be used to estimate the water demand functions. According to Jorgenson (2000):
“Under increasing returns and competitive markets for output and all
inputs, producer equilibrium is not defined by profit maximization, since
no maximum profit exists. However, in regulated industries the price of
output is set by regulatory authority. With output fixed from the point of
view of the producer, necessary conditions for equilibrium can be derived
from cost minimization. Where total cost is defined as the sum of
expenditures on all inputs, the minimum value of cost can be expressed
as a function of the level of output and the prices of all inputs.”
Thus as mentioned earlier, conditional factor demand is a function that gives the
optimal demand for each of several inputs as a function of the expected output, and
the prices of inputs. Conditional demand functions are obtained using the Shepard’s
Lemma where the cost minimization problem is the production of a specified level of
output with the least expenditure on inputs (Arrigada, 2004). In this study, it is
assumed that, under cost minimization, the water demand function is a function in
terms of crop quantity and the prices of the eight inputs namely, water price, land
rent, fertilizer price, machinery rent and cost, seed price, wage, animal fertilizer price
and pesticide price. The most widely used forms of production functions in the
analysis of agriculture are the Cobb-Douglas, Modified Cobb-Douglas is called
Transcendental and the Translog (Sahibzada, 2002). However, Translog is a flexible
functional form which places no prior restrictions on the production technology such
as constant returns to scale, homogeneity, separability and constant elasticity of
substitution (Sahibzada, 2002). It is a second order Taylor series approximation, and
thus requires a larger number of parameters to be estimated. This in turn, results in a
decrease in the degree of freedom, an increase in variance and may make it
impossible to reject the null hypothesis. By the way, multi - collinearity is often a
67
problem of estimating Translog production function in estimating single equation
(Sahibzada, 2002).
In this research, the Cobb-Douglas functional form is used due to;
1. To avoid of mentioned problems above
2. It is very common in agricultural production studies
3. Its stinginess in parameters
4. Ease of interpretation
5. Computational simplicity
6. The resulting coefficients make it possible to interpret the elasticity of
production with respect to inputs.
7. Indicate the relative importance of each input with respect to output.
Several studies have made use of this form primarily because the resulting
coefficients make it possible to interpret the elasticity of production with respect to
inputs, and also indicate the relative importance of each input with respect to output
(Sahibzada, 2002).
He used an initial Cobb-Douglas production function to estimate the relationship
among total aggregated farm output, fertilizer use, labor supply, tractor use, and
irrigation water input. The findings suggest that irrigation water demand is price
inelastic, and that predicted water usage exceeds actual use across the sample. The
Cobb-Douglas functional form was proposed by Wicksell (1851-1926), and tested
against statistical evidence by Charles Cobb and Paul Douglas in 1928. The general
form of the Cobb-Douglas production functions is as follows,
68
n
q  A X iBi
Equation 3-42
i 1
Where, q and Xi denote output and each bundle of inputs respectively. A, and Bi are
n
parameters. This function exhibits constant returns to scale if  Bi  1 . Under
i 1
constant returns to scale Cobb-Douglas function, Bi is the elasticity of q with respect
to an input Xi. Since 0  Bi  1 , each input exhibits diminishing marginal
productivity. Any degree of increasing returns to scale can be incorporated into this
n
function depending on    Bi (Nicholson, 2004). The reason for computational
i 1
attractiveness of this form is that it becomes linear in terms of the logarithms of the
variables, that is:
ln q  ln A   Bi ln xi
Assuming that our objective function is as follows,
min TC  Pw .W  Pl .L  Pf .F  Pp .P  Pr .R  Pm .M  Pa .Fa  PS .S
Equation 3-43
Equation 3-44
Subject to the constraint,
Y  AW a1 La2 F a3 P a4 R a5 M a6 Fa a7 S a8
Equation 3-45
where,
Y = Total aggregated output; F = Fertilizer; L = Labour; M = Tractor and machinery
services; Fa = Animal Fertiliser; R = irrigated area; S= Seed; P = Pesticide; and W
=Consumed (Demanded) water; Pi = input prices; cost minimization problem for a
firm can be written as a constrained optimisation equation, as below:
69
l  (Pw.W  Pl .L  Pf .F  Pp.P  Pr .R  Pm.M  Pa.Fa  pS .S)  (Y 0  AWa1 La2 Fa3 Pa4 Ra5 Ma6 Fa7 Sa8 )
Equation 3-46
where  is the lagrangian multiplier. After applying the first order conditions for cost
minimization and rearranging,
C(Pw, Pl , Pf , Pp , Pr , Pm, Pa , PS ,Y)  Pw.W  Pl . L  Pf . F  Pp. P  Pr . R  Pm. M  Pa. Fa  PS . S
Equation 3-47
Using Sheppard’s Lemma the firm’s system of cost minimizing input demand
functions (the conditional factor demands), differentiating the cost function, and
rearranging the terms we will obtain the following:
C ( Pw .Pl .Pf .Pp .Pr .Pm .Pa .PS )  B.Y ay .( Pwb1 .Pl b2 .Pf b3 .Pp b4 .Pr b5 .Pm b6 .Pa b7 .PS b8 ) 
( Pw1 2 b1  Pl1 2 b2  Pf1 2 b3  Pp1 2 b4  Pr1 2 b5  Pm1 2 b6  Pa1 2 b7  p1s  2 b8 )
Equation 3-48
C
W
 PW
c
Y
ay
. B . Pw b 1 Pl b 2 . P f b3 . P p b 4 . Pr b5 . Pm b6 . Pa b7 . p S b8
Equation 3-49
In logarithmic terms, it yields,
lnW  ln B  ay lnY b1 ln Pw b2 ln Pl b3 ln Pf  b4 ln Pp  b5 ln Pr b6 ln Pm b7 ln Pa b8 ln PS
Equation 3-50
A detailed derivation of the input demand function for irrigation water is given in
Appendix D.
70
3.3
Econometric Methodology
According to Johansson (2005) when data on water use exist, researchers generally
employ econometric approaches to determine the value of irrigation water to
producers (See Table C.3 in Appendix about Econometric Studies of Water Values).
When, however, the number of observations of irrigation water usage available is not
adequate, mathematical programming approaches are useful to estimate water
demand, and the value of irrigation water (See Table C.4 in Appendix on
Mathematical Programming Studies of Water Values). In the econometrics research,
the determination of the type of econometric methodology is very important.
Different methods have been employed to estimate the water demand function. These
methods are generally divided into two main groups, nonparametric and parametric
(econometric). Nonparametric methods are either based on mathematical techniques
such as linear programming and mathematical programming, or on accounting
calculations and system of farm budgeting. In nonparametric method, the economic
value of water input is extracted, and the farmer reaction to different prices of water
is estimated. This method is used where no market price and amount exists (e.g. for
water).
In parametric econometric methods, a researcher will estimate a demand function for
input either directly (derived demand function) or indirectly. In the former case,
production function or profit function, and or by crop cost function making clear, and
then the demand function of input will be extracted. On the other hand, the success of
any econometric analysis eventully depends on the availability of the appropriate
data. Three types of data may be available for empirical analysis: time series, crosssection, and pooled (i.e., combination of time series and crosssection) data. A time
71
series is a set of observations on the values that a variable takes at different times.
Most empirical work based on time series data assume that the underlying time series
are stationary. Cross-section data are data on one or more variables collected
simultaneously. Of course, the cross - sectional data have their own problems, in
particular the problem of heterogeneity. In pooled or combined data, the elements of
both time series and cross - section data are pooled together. Panel, Longitudinal, or
Micropanel Data is a special type of pooled data in which the same cross-sectional
unit (say, a family or a firm) is surveyed over time (Gujarati, 2002). This study
focuses on the irrigation water demand function via using a panel data set. A panel
data set is one that follows a given sample of individuals over time, and thus
provides multiple observations on each individual in the sample (Hsiao, 2003).
According to Baltagi (1983), the major reasons for carrying out the analysis using
panel data rather than time series or cross sectional data are as listed below,
(1) Checking for individual heterogeneity.
(2) Panel data give more informative data with more variability, less
collinearity among the variables, more degrees of freedom and more
efficiency.
(3) Panel data offers better ability to study the dynamics of adjustment.
(4) Panel data are better in identifying and measuring effects that are not
simply detectable in pure cross-section/time-series data.
(5) Panel data models allow us to construct and test more complicated
behavioral models than purely cross-sectional or time-series data.
(6) Micro panel data gathered on individuals, firms and households may
be more accurately measured than similar variables measured at a macro
72
level. Biases resulting from aggregation over firms or individuals may be
reduced or eliminated.
(7) Macro panel data on the other hand have a longer time series and are
free from the problem of nonstandard distributions typical of unit roots
tests in time-series analysis.
According to Hsiao (2003), longitudinal data allow a researcher to analyze a number
of important economic questions that cannot be answered using cross - sectional or
time series data sets. Anyway, by taking proper account of selectivity and
heterogeneity biases in panel data, one can have confidence in the results obtained.
The panel data used in this study involve information on outputs mentioned earlier
and various inputs applied for their production over a period of 6 years across 28
provinces in Iran. Schoengold et al.(2006) estimated a model of agricultural water
demand based on the role of water in the farm production function. Likewise, they
presented estimates of the parameters of the model using a unique panel data set
from San Joaquin Valley, California. They also found that agricultural water demand
is more elastic than shown in previous work on urban water demand, a result which
is of high interest in clarification of differences in the design of optimal policies
directed at agricultural users of water in comparison to urban users.
I. Pooling Data:
Assuming that the intercept and slope coefficients are constant across time and space,
and that the error term captures differences over time and between individuals, the
following relationship will be generated,
73
Yit  1   2 X 2it  3 X 3it  uit
i  1, 2, 3, ..., N
;
Equation 3-51
t  1, 2,  , T
II. Panel Data:
1. The Fixed Effects Model (FEM)
According to Gujarati (2002), estimation of equation (3-51) depends on the
assumptions we make about the slope coefficients, the intercept, and the error term,
uit. There are several possibilities,
i) The slope coefficients are constant but the intercept varies among individuals.
Yit  1i  2 X2it  3 X3it  .  uit
for the cross section effect
Yit  0t   2 X 2it  3 X 3it  .  uit
for the time effect.
Equation 3-52
Equation 3-53
ii) The slope coefficients are constant but the intercept varies among individuals and
over time.
Yit  1i  0t   2 X 2i  3 X 3i  uit
Equation 3-54
iii) All coefficients (the intercept as well as slope coefficients) vary among
individuals.
Yit  1i   2it X 2i  3it X 3i  uit
Equation 3-55
1.4 The intercept as well as slope coefficients vary over individuals and time.
74
Yit  1i  0t   2it X 2i  3it X 3i  uit
Equation 3-56
2. The Random Effects Model (REM)
Gujarati (2002) states that in a random effect model it is assumed that the intercept of
an individual unit is a random drawing from a much larger population with a
constant mean value. The individual intercept is then expressed as a deviation from
this constant mean value; that is, there is a common mean intercept, but actual
intercepts vary randomly (error components). In other words, instead of treating β1i
as fixed, we assume that it is a random variable with a mean value of β1 (no subscript
i here). The intercept value for an individual company can be expressed as,
 1i   1   i
Equation 3-57
Y it  1   2 X 2it  3X 3it   i  u it  1   2 X 2it  3 X 3it  ..  w it
Equation 3-58
Where,
wit   i  uit
Equation 3-59
The composite error term wit consists of two components, εi which is the crosssection or individual-specific error component, and uit which is a combination of
time series and cross-section error components.
Gujarati (2002) illustrates the difference between FEM and ECM. He states that in
FEM each cross-sectional unit has its own (fixed) intercept value in all N such values
for N cross-sectional units. In ECM, on the other hand, the intercept β1 represents the
mean value of all the (cross-sectional) intercepts, and the error component εi
represents the (random) deviation of individual intercept from this mean value.
75
However, keep in mind that εi is not directly observable; it is what is known as an
unobservable, or latent, variable. He showed that a formal test called restricted F test
(Chow Test) could help to choose between Pool and Panel approachs. In this test, the
pooled regression model should first be employed as the baseline for comparison.
Then, after estimating the model as panel the restricted F test is used.
F
F
2
2
( R panel
 R pool
) / df
2
(1  R panel
) / df
2
2
( R panel
 R pool
) / number of new regressors
Equation 3-60
2
(1  R panel
) / (= n - number of parameters in the new model)
When the F value is highly significant, the pool regression seems to be invalid. By
the way, another method helping choose between Pool and Panel approaches is
Breush and Pagan Lagrangian Multiplier Test for random effects, with which is
equipped the STATA software package. In this test, the model regresses underlying
random effects to obtain Chi-squraed value. If the calculated value exceeds tabulated
Chi-squraed (χ2) value, can be drawn the conclusion that the random effect model is
more appropriate than pooled model. In this study, Breush and Pagan Lagrangian
Multiplier Test was used to choose between the Pool and the Panel approaches.
Gujarati (2002) also illustrated that:
“A formal test will help us choose between FEM and ECM was
developed by Hausman in 1978. The null hypothesis underlying the
Hausman test is that the FEM and ECM estimators do not differ
substantially. The test statistics developed by Hausman has an asymptotic
 2 distribution. If the null hypothesis is rejected, the conclusion is that
ECM is not appropriate and that we may be better off using FEM, in
which case statistical inferences will be conditional depending on the εi
in the sample.”
76
III. The Seemingly Unrelated Regression
Zellner’s (1962) indicated that, originally at a given time the seemingly unrelated
regression is a technique for analyzing a system of multiple equations with cross equation parameter restrictions and correlated error terms. Once seemingly unrelated
regression model estimates are obtained, inferences are mainly about testing the
validity of cross-equation parameter restrictions.
According to (Gujarati, 2004), applications of the SUR procedure with time-series or
cross-section data are too numerous to cite.
He also continues that in several
instances in economics, one needs to estimate a set of equations. This could be a set
of demand equations, across different sectors, industries or regions or the estimation
of a translog cost function along with the corresponding cost share equations. In
these cases, seemingly unrelated regressions (SUR) approach is popular since it
captures the efficiency due to the correlation of the disturbances across equations.
Consequently, we are faced with three kinds of seemingly unrelated regressions
(SUR) approach:
1. A set of equations, across different sectors, industries or region with crosssection data.
2. A set of equations, across different sectors, industries or region with timeseries data.
3. A set of equations, across different region with time-series and cross-section
data (Panel Data). In panel data the same cross-sectional unit (say a family or
77
a company or a province) is surveyed over time. In short, panel data have
space as well as time dimensions.
For illustrative purposes we obtained data from 28 provinces. Data for each province
on were more than five variables spanning through 2001–2006. Thus, there are 28
cross-sectional units and six time periods. In all, therefore, we have 168
observations. According to (Gujarati 2004, P: 646) we could run 28 cross-sectional
regressions, one for each year. For a given time, it is possible that the error term for
2001 is correlated with the error term for 2002 or both 2002 and 2003 etc. This leads
to seemingly unrelated regression (SURE) modeling. By the way, EViews simply
treat panel data as a set of stacked observations.
3.4
Econometric Model
The functional form for conditional factor demand may be derived in consonance
with an assumed production function. Here in, however, the water demand will be
specified directly using a water demand function that includes output quantity and
input prices. The estimation of the water demand function using the methodology
presented in the previous chapter allows identification of the significant variables
that explain its consumption. An empirical specification of the water demand is given
by,
ln Dw
i ,t
  0   1ln Pw
i ,t
  2 ln Pf i ,t   3 ln R l i ,t   4 ln Ps i ,t
  5 ln W i ,t   6 lnQ i ,t   7 ln Pfa i ,t   8 ln Cl i ,t   i ,t
Equation 3-61
Where,
Dwi,t = amount of water demanded (consumed) in i th region in year t (Cubic Meter)
78
Pwi,t = the vector of water prices used in production in i th region in year t.
(Toman/m3)
Pfi,t = the vector of fertilizer prices used in production in i th region in year t
(Toman/kg)
Pfai,t = the vector of animal fertilizer prices used in production in i th region in year t
(Toman/kg)
Cli,t = the vector of prepare cost of land used in production in i th region in year t
(Toman/kg)
Psi,t = the vector of seed prices used in production in i th region in year t (Toman/kg)
W i,t = wages paid for production in i th region in year t (Toman/Man day)
Qi,t = Irrigated Production in i th region in year t (Kg)
Rli,t = Land rent in i th region in year t (Toman/ m2)
εi,t denotes the effects of the omitted variables that are peculiar to both the individual
units, and time periods. In this study i denotes the provinces of Iran and t indicates
year (i= 1,2, … , 28 ; t = 2001, 2002, …, 2006).
3.5
Data Collection
This study is based on secondary data. All necessary information was collected from
28 provincial statistical reports from published data, government reports and online
statistical database of the provinces, Ministry of Energy, Iran Water Resources
Management Company, Ministry of Keshavarzi Jehad, Iran Meteorological
Organization, Soil and Water Research Institute, Centre of Iran statistic and relevant
institutions in Iran. Many of the agricultural statistics are gathered from the United
Nations Food and Agriculture Organization while others were inferred from official
documentations.
79
This study used secondary cross sectional - time series data (Panel Data) over a
period of 6 years and from 28 provinces. Annual data for the period 2001 to 2006
were used for the study. The study used Panel data to improve the analysis of water
demand with respect to previous studies.
Finally, the data was analyzed using descriptive statistics presented in Tables H.1,
H.2 and H.3 (see Appendix H) includes the water demand, water price, land rent,
seed price, fertilizer price pesticide, wage, land preparation cost, machinery services
cost, animal fertilizer price, irrigated crop production and VMP for all crops.
3.6
Description of Variables
Of all factors influencing demand side for agricultural water those found to be of
considerable effect from the author’s view point include, product quantity, plant
water required (water demand or water consumed), amount and price of inputs (such
as labor, fertilizer, machinery, area, seed, pesticide, water and etc) used in the crop
production processes, and finally average and marginal costs on the supply side of
water.
3.6.1
Water Demand
Determination of crops water requirement is the basic measure in irrigation and
water resource planning. Several methods (Blaney-Criddle, Radiation, Modified
Penman, and Pan Evaporation methods) have been developed over the last 50 years
to estimate reference crop evapotranspiration from climatic variables are given in
FAO (1984).
80
Briefly mentioning, the different steps in the calculation of consumed (demanded)
water are,
1. To collect existing climatic data from the Meteorological Organization,
consisting of daily readings from multiple stations.
2. To select the best method to use for determining crop water requirements.
Abbas Keshavarz and et al. (2005) illustrated that:
“The Penman-Monteith method was selected as the best method for
determining crop water requirements during formulation of The National
Document of Crop Water Requirements in Iran (Ministry of Agriculture,
1998).”
3. To calculate Reference Crop Evapotranspiration Standard of Grass (ETO)
values and to determine their validity.
4. The annual crops and fruit trees under consideration in each plain obtained
and the crop coefficient (Kc) was determined by the method recommended by
FAO, regional condition and previous experiments.
5. To calculate the crop water requirements (ETcrop) for the selected crops.
6. To calculate the Efficient Rainfall based on the method presented by
American Society of Civil Engineers (ASCE).
7. To calculate the total irrigation requirement that is given by the following
equation,
Irrigation water
net requirement
(IRReq)
crop water
= requirements
(ETcrop)
-
Efficient
rainfall (Peff)
Equation 3-62
Where,
81
[Ptot  (125  (0.2  Ptot))]
125
Peff  125   0.1 Ptot 
Peff 
for rainfall less than 250 mm in month
for rainfall more than 250 mm in month
Equation 3-63
And Ptot is Total Rainfall
8. To calculate the consumed (demanded) water. This is done using the
following equation,
Consumed
Water (m3)
Total Irrigation
= Requirement
+
[Total Irrigation requirement
Irrigation Efficiency
(1 
)]
100
Equation 3-64
These stages are shown in Appendix E.
In this study, steps one through six were done based on the data extracted from
published reports of Soil and Water Research Institute (SWRI) for various crops in
different regions of Iran. Upon extraction of the crop water requirements from the
aforementioned report, the net requirement of irrigation water will be computed by
subtracting the amount of effective rainfall based on meteorological information
from Iran Meteorological Organization. Data on annual average rainfall (Ptot) were
collected at a sub-district level from Iran’s Meteorological Organization. Finally, the
total amount of consumed water will be computed based on equation (3-64). It is
claimed that the overall irrigation efficiency in Iran is between 24 to 57 percent as
shown in Table (3-1).
82
Table 3-1 Ranges of irrigation efficiency in some provinces in Iran
Range of Irrigation
Range of Irrigation
Provinces
Efficiency (%)
Provinces
Efficiency (%)
West Azerbaijan
28-41
Gazvin
27-38
Ardabil
28-39
Kordestan
25-40
Isfahan
28-42
Glolestan
28-40
Boshehr
24-30
Gillan
38-54
Chaor mahal & Ba
30-39
Mazanderan
37-57
Korasan
30-37
Markazi
29-39
Kozestan
27-37
Hamedan
27-38
Zanjan
25-38
Yazd
30-40
Semnan
30-40
Source: (Keshavarz, Ashraft, Hydari, Pouran, & Farzaneh, 2005)
3.6.2
Water Price
Water price is computed from the cost of water which includes the cost paid to the
regional organization of the irrigation networks, the electricity/fuel cost of pumping,
and the simulated price of water that is given by the following relationship,
Water Price  Toman / m3  
The Cost of Water  Toman / ha 
Irrigation water net requirement  ha / m3 
Equation 3-65
3.6.3
Output Price
The main reference for price-per-kilogram (Toman/kg) of each crop for the various
provinces was The Iranian Statistical Yearbook made available by the Iranian Center
of Statistics for the period 2001-2006 in order of provinces.
3.6.4
Wage
Wage rates (Toman/man-day) corresponding to each type of activities (including,
sowing, hoeing, pre-sowing, irrigation, harvesting and threshing) were obtained from
83
The Iranian Statistical Yearbook provided by the Iranian Center of Statistics for the
period 2001-2006 in order of provinces. One man-day is equivalent to the amount of
work done by one labor in eight hours during one day.
3.6.5
Cultivated Area
The main reference on area under cultivation (ha) for total, irrigated and rain-fed
areas was the annual report provided for each province by the Ministry of Jahad-eKeshavarzi for the period 2001-2006.
3.6.6
Other Explanatory Variables
There are some other explanatory variables such as production in irrigated areas;
cultivated area; yield, seed, pesticide and fertilizer prices; machinery and land rents;
and cost of land preparation. Their main source was annual reports made available
for various provinces by the Ministry of Jahad-e-Keshavarzi (Agriculture) for the
period 2001-2006.
3.6.7
Value of Marginal Product (VMP)
The marginal product of an input indicates the additional output that might be
expected from an additional unit of that input ceteris paribus. The value of marginal
product of a factor is the extra value of output generated by employing one more unit
of that factor (VMPw = MPw.Py).
As mentioned earlier, all parameters in the Cobb-Douglas production function are
elasticity of production. Consequently, the value of marginal product (VMP) for a
84
certain input can be calculated from the derived demand function. The water demand
function assumed is as follows,
lnW  ln B  ay lnY b1 ln Pw b2 ln Pl b3 ln Pf  b4 ln Pp  b5 ln Pr b6 ln Pm b7 ln Pa b8 ln PS
Equation 3-66
Hence, elasticity of water input with respect to output is,
 
 ln W
 ln Y
W

W
 Y
Y

W
 Y

Y
W
 ay
Equation 3-67
and its marginal product can be computed as follows,
MP 
Y
1
1
Y

 AP 

W

ay W
Equation 3-68
The value of marginal product (VMP) was computed by multiplying the marginal
physical products of each crop in each region by their prices,
V M P  Py 
Y
1
Y
 Py 

W
ay W
Equation 3-69
where, Y and W denote output and the level of input respectively; and MP and AP are
the marginal and average physical product for water respectively; Py is crop price in
each region;  = ay is the elasticity for water with respect to quantity produced, and
1
is the elasticity of production with respect to water input.
ay
85
3.6.8
Average cost
On the supply side determination of the average cost of water was based on,
secondary data of provincial expenditures on irrigation water supply delivery, which
includes the annual cost of operation and maintenance (O&M) or the average
variable cost of supplying water. The annual costs of operation and maintenance
were collected from Iran Water Resources Management Company.
3.6.9
Short-run Marginal Cost
The short-run marginal cost will be computed from the average cost obtained earlier
by dividing incremental expenditure by incremental water supply over the six year
period.
86
CHAPTER IV
4
4.1
RESULTS AND DISCUSSION
Introduction
This chapter presents the research findings on the basis of research questions and
hypotheses mentioned previously in Chapter one. The findings of this study are
presented in two main parts. The first part focuses on a brief introduction to various
crops, the estimated irrigation water demand functions, with respect to their current
prices, and elasticity estimates. In the second part, the estimated irrigation water
demand functions for crops mentioned above, with respect to its value of marginal
product water supply side average cost and marginal cost, are presented together with
the estimation of the relevant elasticity and efficient pricing for irrigation water in
Iran.
4.2
Estimation of the Model with Current Price
The differences in climatic, atmospheric, and geographical conditions in each of the
provinces have pronounced influences on crop varieties in Iran. In this section, after
the introduction of each crop, its water demand function is estimated. In the ensuing
sections, the water demand functions, with respect to the current prices for strategic
products (such as wheat, barley, lentil, pea, pinto bean, onion, tomato, cucumber,
watermelon, cotton, and sugar beet) in Iran’s agricultural sector together with the
estimation of the relevant elasticity, are presented and discussed.
87
4.2.1
Wheat
Wheat is the primary crop among all cereals produced in Iran. In 2006, Iran’s total
production, cultivated area, and irrigated and rainfed yield of wheat were about 14.66
million ton, 6878919 ha, and 3754.03 and 1084 kilogaram per hectare, respectively.
Wheat yields from irrigated areas in Iran’s provinces range from 1717 to 5359 (kg /
ha). The largest irrigated cultivation area and production level reported were 457695
ha and 2044409 ton respectively in Fars province of Iran. The total irrigated
production in 2006 was reported to be about 10137770, for which 795151952 man
days were hired.
As mentioned earlier, the Panel Data method comprising of 168 observations was
used to estimate the irrigation water demand in 28 provinces for the period between
2001 and 2006. First, Breush and Pagan Lagrangian Multiplier Test was used to
choose between Pool and Panel Data approaches. This test revealed that the Panel
Data approach was more appropriate than the Pool Data approach. Hausman’s
specification test was also conducted on STATA 10 (an econometric software
package) to choose one method out of the fixed effect and random effect. Based on
the test, the best model for the irrigation water demand function of wheat was found
to be fixed effect approach. Next, pretest and diagnostic checks were done upon
which the best model was estimated. Listed in Table 4-1 are the estimated
parameters.
88
Table 4-1 Water demand function for wheat
Dependent Variable:
LDWT
Independent variable
Method: Panel Least Squares
Coefficient Std. Error t-Statistic Prob.
C (Intercept)
14.53***
0.59
24.48
***
LPW(water price)
-0.036
0.01
-4.05
***
LQ (output quantity)
0.381
0.05
7.23
LPS (seed price)
0.032**
0.01
2.11
***
LPF (fertilizer price)
0.056
0.01
6.10
**
LPM (machinery rent)
-0.050
0.02
-2.17
***
LW (wage)
0.054
0.01
3.86
AR(1)
0.316***
0.09
3.55
Cross-section fixed (dummy variables)
0.00
0.00
0.00
0.04
0.00
0.03
0.00
0.00
R-squared
0.99
Adjusted R-squared
F-statistic
751.92
Prob.(F-statistic)
Durbin-Watson stat
2.06
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
0.99
0.00
Based on the results presented in Table 4.1, the estimated coefficient for water price
is negative and it is significant at 1% level. This indicates that even though the
estimated coefficient is very low, farmers tend to use less water when the price is
higher. In other words, a one percent increase in water price will cause water demand
to decrease by 0.036 percent. Similar results were obtained for machinery rent; that
is, water and machinery services were found to be complementary inputs. On the
other hand, coefficients on seed price, fertilizer price and wage rate are positive and
significant at 1% or 5% level. One interpretation is that water with fertilizer, seed
and labour force are substitute inputs. Thus, a one percent increase in the
aforementioned input prices will cause water demand to increase by 0.032, 0.06 and
0.05 percent, respectively.
89
The positive sign of the above coefficient may stem from the fact that a fullymechanized cultivation is not possible in certain regions of the country, and thus,
most of the activities associated with cultivation, maintenance and harvest of barley
are to be done by labour force. Indeed, after costs incurred by water use, labour
force has the third highest cost share in barley production.
The independent variables were not only found to be significant, but the values of the
coefficients were also generally almost zero; that is, water demand is infinitely
inelastic with respect to the above input prices. This indicates that farmers are
insensitive to changes in input prices since they deem the above mentioned inputs
essential for crop yields (Arriagada, 2004).
The estimated coefficient for the output quantity is significant at 1% level. As
mentioned in the previous chapter, the functional form used to estimate water
demand is linear - logarithm. Hence, the estimated parameter coefficient indicates the
elasticity of water use given the changes in the output quantity, i.e. that one percent
increase in the demand for wheat quantity will lead to a 0.38 percent increase in
water usage. This relationship could be used to determine the impacts of production
quotas or other wheat policies on water use (Arriagada, 2004).
Thus, the obtained coefficients do not contradict with the fifth hypothesis outlined
for the research on the agricultural sector in Iran, i.e. crop amount has significant
effect on the water usage.
90
The most popular test for detecting serial correlation is the one developed by
statisticians named Durbin and Watson. In the initial estimation autocorrelation was
found to exist and AR (1) was used to remedy it. AR (1) is known as Markov firstorder autoregressive scheme, or simply the first-order autoregressive scheme. It is
significant at 1% level. The sequential disturbances are positively correlated, with a
coefficient of autocorrelation of +0.32, a weak degree of dependence between
members of series of observations ordered in time.
The term “fixed effects” is due to the fact that, although the Intercept may differ
across individuals (here the 28 provinces), each individual’s intercept does not vary
overtime; that is, it is time invariant. In FEM the intercept in the regression model is
allowed to differ among individuals in recognition of the fact each province may
have some special characteristics (type of soil, climate, economic, social and
geographical conditions) of its own.
In this case, R2 is equal 0.99. According to Gujarati (2005), the quantity of R-square
is known as the sample determination coefficient and is the most commonly used
measure of the goodness of fit in regression. Its range is 0 ≤ R2 ≤ 1. An R 2 equals 1
means a perfect fit, while an R2 equals zero means that there is no relationship
between the regressor and the predictors whatsoever. Several studies in panel data
model have revealed a high R-Square (see Appendix G).
91
4.2.2
Barley
Among all cereals in Iran, barley is the second most important crop after wheat.
Barley production averaged 2,956,032 ton, with an estimated annual value of $ 473
million in 2006. Cultivated areas of barley amounted to 1567454 ha, and the average
application of water for barley cultivation was 4.8 billion cubic meters in 2006.The
total irrigated production in the same year was 1,972,399 tons, while 20,178,506 man
days were employed.
The yield for barley in irrigated areas of Iran’s provinces is between 1717 to 5359
kg/ha. As in the case of wheat, Fars province has the largest irrigated cultivation area
(457695 ha) and highest production quantity (2,044,409 ton) for this particular crop.
In 2006, the water productivity for barley in Iran ranged from 0.2 to 0.82.
Meanwhile, the water productivity, average application of water, and yield in the
aforementioned province for barley were reported to be about 0.51, 8659 and 4467
kg/ha, respectively. Other important information on barley is presented in Table B.4
(Appendix B).
The Cobb-Douglas production function was used to estimate of the water demand
function for barley. The same water demand function for barley was estimated using
equation (3-61). The water demand is a function of the current water price, fertilizer
price, seed price, wage, land rent, and the amount of output.
The panel data corresponding to a total of 156 observations were obtained from 26
provinces for the period between 2001 and 2006. Therefore, to achieve a suitable
92
function, the Breush and Pagan Lagrangian Multiplier Test was initially employed to
choose between the Pool and Panel data approaches. In this study, the Panel data
model was found to be better than the Pool data model, and for this reason, the
Hausman’s specification test was used to choose between the fixed effect and
random effect. Finally, the fixed effect approach was found to be the best model for
the irrigation water demand function of barley. These tests were conducted using
econometric software STATA 10.
After conducting the data stationary test, co-integration test (by Levin, Lin & Chu t*
statistic) and diagnostic checks, the best model was estimated. Table 4-2 shows these
estimated parameters.
Table 4-2 Water demand function for barley
Dependent Variable: LDWT
Method: Panel EGLS (Cross-section weights)
Independent variable
Coefficient Std. Error t-Statistic Prob.
4.596***
0.59
7.79
**
-0.017
0.01
-2.05
0.812***
0.03
23.52
-0.067***
0.02
-4.05
0.038
0.03
1.27
-0.118
0.07
-1.57
Cross-section fixed (dummy variables)
0.00
0.04
0.00
0.00
0.21
0.12
R-squared
0.99
Adjusted R-squared
F-statistic
518.5 Durbin-Watson stat
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
0.99
1.78
C
LPW
LQ
LRL
LW
LPF
93
The estimated water price coefficient was found to be negative. This variable was
found significant, but its value was almost zero, that is, water demand is infinitely
inelastic. This finding indicates that farmers are less sensitive to the price of water
since they consider this input as an essential input. However, based on these results,
farmers tend to reduce the use of water when its price increases, but this is done only
in a very small amount. Therefore, the obtained coefficients do not contradict with
the first and second hypotheses of the research, which are related to the existence of
a negative relationship between price of water and the amount of demand for it, and
price of water is not efficient.
The coefficient of price of fertilizer and land rental was also found to be negative.
One interpretation is that water with fertilizer and land are complementary inputs.
However, the coefficient on wage is positive. One interpretation is that water and
labour force are substitute inputs, whereby a one percent increase in the labour force
wage will cause water demand to increase by 0.038 percent. As discussed in the
earlier section, the reason for this substitution relationship is due to the farmers’
effort in enhancing efficiency of irrigation and in preventions wastage of available
water. In this way, as more labour force is employed in farms during irrigation, the
sooner the irrigation water will cover the irrigated area. This has finally led to a
substitution relationship between labour force and consumption of water.
The estimated coefficient for the quantity of output is significant at 1% level. As
elaborated in the previous chapter, the functional form used to estimate water
demand is linear-logarithm. Meanwhile, the estimated parameter coefficient shows
94
the elasticity of water use, provided that the changes in the quantity of output is
0.812, which indicates that a one percentage increase in the output (barley) quantity
leads to a 0.81 percent change in the use of water. Therefore, the estimated
coefficients do not contradict the fifth hypothesis of the research in Iran agricultural
sector, i.e. crop amount has significant effect on the usage of water. The R-square
value for the regression model was 0.99, indicating a nearly perfect fit. In any
empirical research, when the data are improved from time series to panel data
or from cross sectional to panel data, the number of observation increases. Hence, if
the power of the model goes up, an expected explanation for this is that R-square has
increased.
4.2.3
Lentil
Lentil is among the major crops produced in the agricultural sector of Iran. Table B.5
(see Appendix B) shows the irrigated area, irrigated production, yield, average
application of water, and crop water productivity of lentil in the year 2006. The total
amount of production for this particular crop in 2006 was reported to be about
209067 ton. In the same year, the cultivated area, and irrigated and rainfed yield
totaled to about 100784.1 (ha), 1129.52 and 293.75 kg/ha, respectively. The yield for
lentil was obtained from one hectare of irrigated areas in the different provinces in
Iran, averaging between 734 and 2792 kg. With 6337 (ha) cultivation area and a
production amount of 8145.14 (Ton), Fars province is obviously the primary
producer of lentil in Iran. In 2006, the total irrigated production was aggregated to
16663.13 metric tons, while about 588392 man-days were hired to produce such a
quantity of lentil.
95
The worldwide water productivity for cereal grains, as aforementioned, is between
0.2 and 1.5. Nonetheless, the water productivity range for lentil in 2006 in Iran was
between 0.11 and 0.41. Meanwhile, the water productivity, average application of
water, and yield in Fars province were around 0.21, 4617 and 2792 (kg/ha),
respectively. As mentioned in the previous section, the most prevalently used form of
derived demand functions are the Cobb-Douglas functions, since the resulting
coefficients have made it possible to interpret the elasticity of demand, with respect
to price inputs and amount of output.
Therefore, using the methodology presented earlier, the estimation of the water
demand function enables the identification of major variables that can explain its
consumption. The main equation to be estimated is the water demand equation
which has been presented for the crops mentioned previously, and demand for water
is a function of the current price of water, output amount, prices of fertilizer, seed, as
well as wage, land rent, etc.
Similarly, the panel data method was employed to estimate the irrigation water
demand. For this purpose, a total of 54 observations were made on 9 producer
provinces during 2001-2006. The Breush and Pagan Lagrangian Multiplier Test was
initially done to choose between the Pool and Panel data approaches so as to come up
with an appropriate function. Based on this test, the Panel data model was proven to
be better. Next, the Hausman’s specification test was used to select between the fixed
effect and random effect. Based on the test conducted on Eviwes 6, period weights
was found to be the best model for the irrigation water demand function of lentil.
96
EViews
estimates
a
feasible
GLS
specification
correcting
for
period
heteroskedasticity.
The software Eviews 6 was then used to estimate the best model, after the pretest and
diagnostic checks had been done. The estimated parameters are shown in Table 4-3.
Table 4-3 Water demand function for lentil
Dependent Variable: LDWT
Method: Panel EGLS
Linear estimation after one-step weighting matrix
Period weights (PCSE) standard errors & covariance (no d.f. correction)
Independent Variables
Coefficient
C
6.28***
LPW
-0.25***
LW
-0.90***
LPS
-0.86***
LQ
1.11***
LPP
0.31*
LPM
-0.28**
LPA
0.52***
LPF
1.06***
R-squared
Adjusted R-squared
Durbin-Watson stat
F-statistic
70.31
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
Std. Error t-Statistic Prob.
0.93
0.03
0.09
0.24
0.03
0.15
0.11
0.05
0.21
0.98
0.96
2.53
6.73
-9.07
-10.3
-3.56
31.72
2.03
-2.52
11.37
4.94
0.00
0.00
0.00
0.00
0.00
0.06
0.03
0.00
0.00
The estimated water price coefficient was found to be negative. The value of the
estimated coefficient is about -0.25, indicating that the demand for water is
approximately inelastic. In more specific, farmers are almost sensitive to the price of
water, and they tend to reduce the use of water when the price increases. In other
words, a one percent increase in the price of water will decrease the demand for it by
97
0.25 percent. Thus, the estimated coefficients do not contradict with the first and
second hypotheses of the research due to the existence of the negative relationship
between price of water and the quantity demanded for it, and thus, the price of water
is not efficient.
The coefficient on the price of seed, rental of machinery and wage is also negative.
One interpretation is that water with seed, machinery service and labour force are
complementary inputs, and an increase of one percent in the prices of seed,
machinary rental rate and wage, and land rent will decrease the demand for water by
0.86, 0.28 and 0.90 percent, respectively.
On the other hand, the coefficients of the prices of fertilizer (animal, chemical) and
pesticide are positive. This may be construed that water, fertilizer and pesticide are
substitute inputs, and there is a substitution relationship between the inputs and
agricultural water input. Therefore, to increase the efficiency of irrigation and save
the amount of water used for the production of crops, farmers employ higher amount
of the above inputs. Based on these value and price, the usage of the said inputs per
unit area is increased so as to improve irrigation but lower water consumption. This
means that a one percent increase in the prices of fertilizer and pesticide will cause
demand for water to increase by (0.52, 1.06) and 0.31, respectively.
The estimated coefficient for the output quantity is significant at 1% level. As
discussed in the previous section, the functional form used to estimate water demand
is linear-logarithm. Given the changes in the quantity of the output, the estimated
parameter coefficient shows the elasticity of water usage, i.e. a one percent increase
98
in the output quantity of lentil will lead to a 1.11 percent rise in water consumption.
Hence, the estimated coefficients are consistent with the fifth hypothesis of the
research on Iran agricultural sector, i.e. the amount of crops has marked impact on
the usage of water.
4.2.4
Pea
Pea is another important crop in Iran’s agricultural sector. Table B.6 (Appendix B)
lists some useful information on the production of pea, including in the irrigated area,
irrigated production, etc. The information is given according to province for the year
2006. The total production, cultivated area, and yield (irrigated and rainfed) of peas
in Iran in 2006 were aggregated to about 324786.1 ton, 602557 (ha), 1438 and 433
kilogaram per hectare, respectively. Meanwhile, the yield of peas from the irrigated
areas ranged from 764.43 to 1623.19 (kg / ha) by province. The largest irrigated
cultivation area (2912 ha) and production amount (3474 Ton) were reported in West
Azarbaijan and Fars provinces. Hiring 522340 man days, the irrigated production
during 2006 was estimated to be about 16159 metric tons.
In the same year, the water productivity of pea averaged between 0.13 and 0.42
(kg/m3). The water productivity, average application of water and yield were
reported to be around 0.24, 3788.8 (m3/ha) and 919 (kg/ha), respectively in West
Azarbaijan province. As for the Fars province, the same items were reported to be
0.24, 5896(m3/ha), and 1441 (kg/ha), respectively.
As in previous crops, and due to the same reasons stated in the earlier section, the
Cobb-Douglas form was used for the demand functions. This helps to recognize the
99
major variables explaining water consumption. The primary thing to be estimated,
once again, is the demand for water as a function of the current price of water, output
amount, fertilizer price, seed price, wage, and land rental.
The panel data method, with a total number of 66 observations from 11 producer
provinces during 2001-2006, was used to estimate irrigation water demand. For this
purpose, the Breush and Pagan Lagrangian Multiplier Test was once again used and
concluded that the Panel model was more appropriate than the Pool model. Likewise,
using Hausman’s specification test done on STATA 10, the fixed effect approach
was found to be the best model for the irrigation water demand function of pea.
Therefore, the best model was eventually estimated on Eviews 6 software, based on
the data stationary test and diagnostic checks. The estimated parameters are shown in
Table 4- 4.
Table 4-4 Water demand function for pea
Dependent Variable: LDWT
Method: Panel Least Squares
Independent Variables
Coefficient
Std. Error t-Statistic Prob.
6.02***
0.79
*
0.056
-0.11
0.05
0.96***
***
0.09
-0.31
***
0.06
0.39
-0.10
0.08
0.15
-0.48***
0.15
0.09
Cross-section fixed (dummy variables)
R-squared
0.98
Adjusted R-squared
0.97
Durbin-Watson stat
2.56
F-statistic
58.47
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
C
LPW
LQ
LW
LPM
LPA
LPF
LPS
100
4
-1.79
18.06
-3.51
6.18
-1.25
-3.21
1.64
0.00
0.09
0.00
0.00
0.00
0.23
0.01
0.12
The estimated coefficient for the price of water is negative. It was found significant
at 10% level with a coefficient value of about -0.11, implying that the demand for
water is infinitely inelastic. This means that farmers are insensitive to water price,
and thus, are interested in lowering their water usage when the price increases.
Hence, if the price of water rises by one percent, the demand for it will be lowered by
0.11 percent. The coefficients gained, thus, do not refute the first and second
hypotheses of the research, and this indicates the presence of a negative relationship
between water price and the amount demanded for it, and that the price of water is
not efficient.
The coefficients of the price of fertilizer (animal, chemical) and wage are also
negative. Hence, water, fertilizer, and labour force can be regarded as
complementary inputs. This means, a one percent increase in the price of fertilizer
(animal, chemical) and wage will lower the demand for water by 0.1, 0.48 and 0.31
percent, respectively.
On the contrary, the coefficients of the price of seeds and machinery rental are
positive. It could be deduced that water with seeds, and machinary services are
substitute inputs, in which there will be a rise in the demand for water by 0.15 and
0.39 percent respectively when there is a one percent appreciation in the price of
seeds and machinery rental.
The estimated coefficient for the quantity of output is significant at 1% level. The
estimated parameter coefficient indicates the elasticity of water used given the
variations in the quantity of output. A one percent increase in the quantity of output
101
will result in a 0.96 percent increase in the use of water. Therefore, the estimated
coefficients do not contradict the fifth hypothesis of the research, i.e. the amount of
crop has an important bearing on the water usage in the agricultural sector in Iran.
4.2.5
Pinto Bean
Among the crops produced in the agricultural sector in Iran is pinto bean. In 2006,
the total production, cultivated area, and irrigated yield for pinto beans in Iran were
reported to be around 33923 ton, 208285.7 (ha), 2117.6 kilogaram per hectare,
respectively.
Meanwhile, the yield of pinto beans from irrigated areas in Iran
provinces ranged from 1429 to 3362 (kg/ ha), with the provinces of Markazi and Fars
as the largest irrigated cultivation area and production amount in Iran, respectively.
The total irrigated production in the same year was about 202377 metric tons, while a
total of 6383835 man days were utilized. The water productivity of pinto beans in
Iran during 2006 was estimated between 0.21 and 0.78. In addition, the water
productivity, average application of water, and yield were reported to be about 0.27,
7803(m3/ha) and 2118 (kg/ha) in Markazi province. The corresponding figures for
Fars province were 0.38, 8024(m3/ha), and 3058 (kg/m3), respetively.
Table B.7 (see Appendix B) presents the yield, irrigated area and production, average
application of water, and crop water productivity of pinto bean holdings according to
province for the year 2006.
Like other products aforementioned, the Cobb-Douglas functions were used to
estimate the demand for water and identify major variables explaining water
consumption. The water demand equation presented in Chapter 3 was then estimated.
102
The equation reveals the demand for water in terms of the current price of water,
fertilizer and seed prices, wage, land rental rate and amo unt of the output.
A total of 60 observations corresponding to 10 producer provinces in period between
2001 and 2006 were used in the panel data method. The Breush and Pagan
Lagrangian Multiplier Test revealed that the Panel data approach was a better choice
than the Pool data. Meanwhile, Eviews was utilized to perform Hausman’s
specification test. The test showed that a period SUR was the most suitable method.
When selected Period SUR label, in the combo box labeled Weights, Eviews corrects
for both period heteroskedasticity and general correlation of observations within a
given cross-section (EViews 6 User’s Guide II. 2007: 483).
After the pretest and diagnostic checks, Eviews 6 was then used to estimate the best
model. The estimated parameters are shown in Table 4-5.
The estimated coefficient for the price of water was negative. The estimated
coefficient was -0.03, indicating that demand for water is infinitely inelastic. In other
words, farmers are insensitive to the price of water. Thus, the estimated coefficients
support the first and second hypotheses of this study that there is a negative
relationship between the price of water and the amount of demand for it, and that the
price of water is not efficient.
103
Table 4-5 Water demand function for pinto bean
Dependent Variable: LDWT
Method: Panel EGLS (Period SUR)
Linear estimation after one-step weighting matrix
Period weights (PCSE) standard errors & covariance (no d.f. correction)
Independent Variables
Coefficient Std. Error t-Statistic Prob.
0.75
2.73
0.00
C
2.05***
LPW
-0.03
0.02
-1.20
0.24
0.03
32.44
0.00
LQ
0.99***
***
0.06
-3.23
0.00
LW
-0.19
***
0.04
3.93
0.00
LPS
0.17
R-squared
0.96
Adjusted R-squared
0.95
Durbin-Watson stat
1.71
F-statistic
252.94
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
The coefficient for wage was also found to be negative and significant at 1% level.
This may be construed as an evidence of the fact that water and labour force are
complementary inputs, whereby an increase of one percent in the wage will decrease
the demand for water by 0.20 percent.
Unlike the coefficients for the price of water and wage, the ones derived for the price
of seeds are positive, indicating that water and seed could be regarded as substitute
inputs which, in turn, means that a one percent rise in the price of seed will cause the
demand for water to rise by 0.17 percent. Although the existence of a complementary
relationship between water and seed inputs produces a negative relationship between
the price and the quantity of demand for the other, while the accumulation of the
seed price of the respective crops could be deemed as the primary factor which
results in the sign of coefficient of this variable in the demand function. Likewise,
taking into consideration the fact that the prices of input were used as major
104
contributing factors in the price of water, the sign of the estimated coefficient seems
to be duely correct.
The estimated coefficient for the quantity of output is significant at 1% level. The
estimated parameter coefficient is an indication of the elasticity of water use given
the changes in the quantity of output. This implies that a one percent increase in the
quantity of output (pinto beans) will result in a 0.99 percent increase in the use of
water. Thus, the estimated coefficients do not contradict with the fifth hypothesis of
the research in the agricultural sector in Iran and the amount of crop strongly
influences the usage of water usage.
4.2.6
Onion
Onion is also another principal crop produced in Iran. In 2006, its production
amounted to about 14663745.3 (ton). The area allocated for its cultivation and
average application of water for this particular crop were 6878919 (ha) and about
12.9 billion cubic meter respectively in the same year. The key properties of onion
are presented in Table B.8 (Appendix B).
The yields for onion from the irrigated areas in various provinces in Iran vary from
15360 to 53090 (kg / ha). The biggest reported irrigated cultivation area of 9940 ha is
located in Hormozgan. Likewise, East Azarbaijan has the largest reported production
of onion in Iran, , with a production amount of 404497.22 ton. In 2006 alone, the
total amount of yield from the irrigated production was reported to be about 1670367
metric tons. In order to realize such production level 6043375 man days were
utilized.
105
Meanwhile, the water productivity for onions was found to average between 1.2 and
6.8 in Iran in 2006. The water productivity, average application of water, and yields
of onion in Hormozgan and East Azarbaijan provinces were about 0.51, 8659, and
4467 (kg/ha) and 0.51, 8659, and 4467 (kg/ha), respictively.
As in the other crops, the Cobb-Douglas functions were used and they yielded
coefficients which made it possible to interpret the elasticity of demand with respect
to the price of inputs and the quantity of output.
Employing the methodology elaborated in the previous chapter permitted the
researcher to identify the principal variables which could be used to explain water
consumption. The main equation estimated is the water demand equation which was
presented in the previous chapter. The equation expresses the demand for water in
terms of the current price of water, the prices of seed and fertilizer, wage rates, land
rental and the amount of output.
The demand for irrigation water of 16 producer provinces for the period 2001-2006
was estimated using a total number of 96 observations in the form of Panel Data. In
fact, the Breush and Pagan Lagrangian Multiplier Test showed that the Panel data
approach was superior to the Pool data. After considering the results derived from
the Hausman’s specification test (which was performed using the econometric
software STATA 10) among the fixed effect and random effect, the fixed effect
approach was found to be the most suitable one to model the irrigation water demand
function for onion. Table 4-6 presents the parameters estimated in the most suitable
106
model, which was achieved after the data were gathered from the pretest and
diagnostic checks.
Table 4-6 Water demand function for onion
Dependent Variable: LDWT
Method: Panel Least Squares
White cross-section standard errors & covariance (no d.f. correction)
Coefficient
C
LPW
LQ
LRL
LPF
LPM
1.92***
-0.02**
0.89***
-0.05*
-0.14*
0.04*
Std. Error
0.50
0.01
0.02
0.03
0.05
0.02
t-Statistic Prob.
3.86
-1.95
37.54
-1.76
-2.84
1.76
0.00
0.05
0.00
0.08
0.06
0.08
Cross-section fixed (dummy variables)
R-squared
0.99
Adjusted R-squared
0.99
Durbin-Watson stat
2.05
F-statistic
380.5
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
At 5% level, the estimated coefficient for water price is negative. The demand for
water was infinitely inelastic since the water price coefficient was about -0.02.
Hence, farmers are insensitive to the price of water. Therefore, the estimated
coefficients do not refute the first and second hypotheses of the research, i.e. there is
a negative relationship between the price of water and the quantity demanded for it,
and that the price of water is not efficient.
The coefficients for the price of fertilizer and land rental are also negative and
significant at 10% level. These findings imply that water, fertilizer, and land are
107
complementary inputs, and a one percent increase in the price of fertilizer and land
rental will reduce the demand for water by 0.14 and 0.05 percent, respectively.
However, the coefficient on machinery rental is positive. Water and machinary
services are thus substitute inputs, i.e. a one percent rise in the rental of machinery
will result in an increase of 0.04 percent in the demand for water. In particular,
farmers use tractor as both supplementary and substitution input for water input. The
application of machinery services is increased with rise in the area under onion
cultivation. In addition, more of this input is used for a quicker and easier irrigation.
To do so, some ridges (locally known as “Dookhisheh”) are created on the ground to
facilitate water flow after disking and trowelling. This causes the inputs-to-prices
ratio to vary.
The estimated coefficient for the quantity of output is significant at 1% level. The
elasticity of water use is 0.89, given the variations in the quantity of output,
indicating that a one percent growth in the output (onion) quantity will result in a
0.89 percent increase in the use of water. The estimated coefficients also confirm the
fifth hypothesis of the research, i.e. the amount of crops has a significant effect on
the usage of water in Iran’s agricultural sector.
4.2.7
Tomato
The crop production and agriculture-based industries form important part of Iran’s
economy. Another crop which is of interest among all vegetable crops cultivated in
Iran is tomato. About 5064571 tonnes of tomatoes were produced in Iran in 2006.
The cultivated areas for tomatoes were 147462 (ha), while the average water usage
108
per hectare for it was 9023 in 2006. Utilizing around 19365603 man days, the total
production in the irrigated areas in the year under consideration was nearly 5054830
tons. FAO reported that the exported volume of tomato paste crop to be about 51026
metric tons with a value of $ 26,626 million in 2004.
The tomato yields in the irrigated areas of various provinces in Iran range from 9304
to 50866 (kg/ha) on the average. The largest irrigated cultivation areas and
production quantity reported were 16478 (ha) and 798725 (metric ton) for the Jiroft
city and Fars province, respectively. The total irrigated area for the same year was
about 146837 ha, while 19365603 man days were hired to reach the above
production level.
The comprehensive data set available from the studies performed worldwide
revealed that the global values for Crop Water Productivity (CWP) are generally
very high for crops like tomato. In Iran, the crop water productivity for tomato in the
year 2006 ranged from 1.27 in Sistan-and-Balouchestan province to 8.44 in
Mazandaran province. Meanwhile, the water productivity, average application of
water, and yield in the Fars province for this particular crop were about 5.09 (kg/m3),
9992 (m3/ha), and 50866 (kg/ha) respictively. The major by-province features of
tomato holdings in Iran in 2006 are presented in Table B.9 (see Appendix B).
Derived demand functions were estimated utilizing Cobb-Douglas function forms
due to the possibility of interpretation of the resulting coefficients in connection with
elasticity of demand, with respect to input prices, and amount of output as mentioned
in the earlier section. One hundred and fifty observations carried out on 25 producer
109
provinces were used in the Panel Data method to estimate the irrigation water
demand within the period between 2001 and 2006. The choice between Pool and
Panel data approaches in reaching a suitable function was made based on the results
from the Breush and Pagan Lagrangian Multiplier Test. It revealed that the Panel
model showed advantages over the Pool data approach. Moreover, the Hausman’s
specification test indicated that the fixed-fixed effect (two - way) was the best to be
used as compared to the fixed effect and random effect approaches. The test was
conducted using the econometric software STATA 10. Having the pretest and
diagnostic checks done, the best model was also estimated. Table 4-7 presents
estimated parameters.
Table 4-7 Water demand function for tomato
Dependent Variable: LDWT
Method: Panel Least Squares
White cross-section standard errors & covariance (no d.f. correction)
Coefficient
C
LPW
LQ
LPS
LRL
Std. Error
t-Statistic Prob.
15.26***
0.60
***
0.03
-0.06
0.04
0.30***
0.04
-0.13***
0.04
-0.12***
Cross-section fixed (dummy variables)
Period fixed (dummy variables)
R-squared
0.95
Adjusted R-squared
0.93
Durbin-Watson stat
1.52
F-statistic
58.66
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
110
25.57
-2.28
6.60
-3.56
-3.26
0.00
0.02
0.00
0.00
0.00
Using the methodology presented in the earlier section, it is important to note that the
estimation of the water demand function has made it feasible to identify the
important variables explaining its consumption. The water demand, as a function of
water current price, seed price, land rent, output amount, is the main quantity to be
estimated.
The estimated coefficients for the price of water at 1% level was found to be 0.06,
showing that water demand was infinitely inelastic. The result also indicated that
farmers were insensitive to the price of water. Therefore, the estimated coefficients
do not negate the first and second hypotheses of the research that there exists a
negative relationship between the price of water and the amount demanded for it, and
that the price of water is not efficient.
The coefficients for the price of seed and land rental are also negative and significant
at 1% level. This indicates that water, seed, and land are complementary inputs, and
that is one percent increase in the price of seed and land rental will lower the water
demanded by 0.13 and 0.12 percent, respectively.
On the other hand, the estimated coefficient for the quantity of output is positive and
significant at 1% level. Thus, the estimated parameter coefficient shows the elasticity
of water usage, given the changes in the quantity of output to be such that an increase
of one percent in the output quantity raises the water usage by 0.30 percent. Hence,
the estimated coefficients do not reject the fifth hypothesis of the research, i.e. crop
amount profoundly affects usage of water in the agricultural sector of Iran.
111
The term “fixed - fixed effects” is due to the fact that the intercept of each province
or individual is time variant. It may be noted that the FEM assumes that the (slope)
coefficients of the regression do not vary across individual or overtime. In FEM the
intercept in the regression model is allowed to differ among individuals and overtime
in recognition of the fact each province and time may have some special
characteristics (type of soil, climate, economic, social and geographical conditions)
of its own.
4.2.8
Potato
Potato is another crop of interest in Iran, with an average production of 4218522
metric tons in 2006. In the same year, an area of 163843.5 (ha) was devoted to the
cultivation of potatoes, for which the application of water was 8561 (m3/ha) on the
average. The irrigated production of the crop was aggregated at 4188207 (metric
tons) in 2006 and 15883756 man days were employed. The irrigated potato yields in
the provinces in Iran were also averagely ranged from 12110 to 36219 (kg/ha). The
whole irrigated production in 2006 was reported to be about 4188207 metric ton.
This was achieved via 15883756 man days of work forced hired. With a production
of 875129 metric ton, Hamedan province had the largest proportion of the total
production mentioned above. In the same year, an area of 25738 (ha) was devoted to
the cultivation of potatoes in the province of Ardabil, which was the highest value
amongst all the country’s provinces. With a value of 4.13(kg/m3), this province also
had the highest water productivity for potatoes. On the other hand, Kerman province
was reported to have the lowest value for water productivity of 1.43 (kg/m3). The
average application of water, and the yield of potatoes in Aradabil province were
reported to be about 6414, and 26527 (kg/ha), respectively. The main specifications
112
of the potato holdings according to province for the year 2006 are listed in Table
B.10 (see Appendix B).
In estimating the derived demand function, the Cobb-Douglas functional form was
applied. Using this method, the identification of the important variables explaining
water consumption becomes easier. The equation for demand of water, as a function
of the current price of water, price of seed, land rental, output amount, is the main
equation to be estimated.
Meanwhile, the method used to estimate irrigation water demand was the Panel data,
in which 138 observations from 23 producer provinces in period 2001-2006 were
used. The Breush and Pagan Lagrangian Multiplier Test was also used to choose
between the Pool and Panel data approaches. In view of that test, the Panel data was
chosen as a more appropriate model to be employed. Additionally, the Hausman’s
specification test (performed on the econometric software package, STATA 10)
showed that the fixed effect approach was the best one for modeling the irrigation
water demand function of potato among all the other approaches. Next, the data
stationary test and diagnostic checks were also done, and the best model was
subsequently estimated. The estimated parameters are presented in Table 4-8.
The estimated coefficient for the price of water is negative at 5% level. With a value
of 0.06, the demand for water is infinitely inelasticity. This, in turn, attests that
farmers are insensitive to the changes that have taken place in the price of water.
Consequently, the estimated coefficients comply with the first and second hypotheses
113
of the research; these are, the price of water and the quantity demanded for it are
negatively related, and that the price of water is inefficient.
Table 4-8 Water demand function for potato
Dependent Variable: LDWT
Method: Panel Least Squares
White cross-section standard errors & covariance (no d.f. correction)
Coefficient
C
LPW
LQ
LW
LPP
Std. Error
t-Statistic Prob.
3.83***
0.45
0.026
-0.06**
0.028
0.79***
***
0.03
-0.15
0.022
0.07***
Cross-section fixed (dummy variables)
8.48
-2.1
27.95
-4.39
3.21
0.00
0.04
0.00
0.00
0.00
R-squared
0.99
Adjusted R-squared
0.99
Durbin-Watson stat
2.21
F-statistic
616.71
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
The coefficient on wage is (-0.15) also negative and significant at 1% level. Hence,
water and labour force could be considered as complementary inputs, i.e. a one
percent rise in the wage will reduce the damand for water by 0.15 percent.
On the other hand, the estimated coefficient for the quantity of output is significant
and positive at 1% level. The estimated parameter coefficient shows that an increase
of one percent in the amount of output induced a 0.79 percent growth in the usage of
water. Thus, the coefficients are compatible with the fifth hypothesis of the research
on the agriculture sector in Iran, i.e. the amount of crops has a great influence on the
usage of water.
114
4.2.9
Cucumber
With an average production of 1938491 (metric tons) in 2006, cucumber is yet
another crop of interest. Total land size devoted to its cultivation was aggregated to
82350 (ha), and the average application of water was 7190 (m3/ha) in the same year.
FAO reported the export level of cucumber paste crop as nearly 36948 metric tons,
with a value of $ 11.012 million in 2004.
The yields for cucumber from the irrigated farms in various provinces in Iran ranged
from 9044 to 37228 (kg / ha). In 2006, Jiroft city had the largest total irrigated
cultivation area for cucumber with 20373 (ha). Meanwhile, with 608939.6 (metric
ton), Kerman province had the highest production level reported. A total of 8751533
man days were employed to realize the total production level of 1933975 in the same
year.
The provinces of Tehran and Kohkiloyeh-va-Boyrahmad had the lowest and highest
water productivity ranges of 2.49 and 7.69, respectively. The water productivity,
average application of water, and yields for cucumber were reported to be about
3.29, 8812, and 29889 (kg/ha) respectively in Jiroft city located in Kerman province.
Table B.11 (see Appendix B) contains the main specifications of cucumber holdings
according to province for the year 2006.
Just like for the other crops, the Cobb-Douglas functions were used as the functional
form of the derived demand functions to facilitate the interpretation of the elasticity
of demand in relation to the price of inputs, and the amount of output. For this, the
115
demand for water, as a function of the current price of water, price of fertilizer and
seed, wage, land rental, and the amount of output, was estimated.
The estimation of irrigation water demand was carried out using the Panel Data,
based on 108 observations conducted within the period between 2001 and 2006, from
18 producer provinces. Once again, the Breush and Pagan Lagrangian Multiplier Test
was performed to select an appropriate function to be employed and the outcome of
the test indicated that the Panel model was more suitable to be used than the Pool
model. Based on the Hausman’s specification test done on STATA 10, the fixed
effect approach was selected as the most appropriate model for the irrigation water
demand function of cucumber. After that, the diagnostic checks and pretest were
performed, and the best model was estimated. These estimated parameters are
presented in Table 4-9.
Table 4-9 Water demand function for cucumber
Dependent Variable: LDWT
Method: Panel Least Squares
White cross-section standard errors & covariance (no d.f. correction)
Coefficient
C
LPW
LQ
LW
LPM
LPA
Std. Error
t-Statistic Prob.
5.57***
1.01
*
0.01
-0.02
***
0.06
0.68
0.06
-0.14***
0.04
0.08**
0.02
-0.03*
Cross-section fixed (dummy variables)
5.51
-1.8
12.01
-2.62
1.96
-1.76
0
0.07
0
0.01
0.05
0.08
R-squared
0.99
Adjusted R-squared
0.99
Durbin-Watson stat
1.87
F-statistic
414.21
Prob.(F-statistic)
0
*
Statistically significant at the 10% level ; ** Statistically significant at the 5% level
***
Statistically significant at the 1% level
116
The estimated coefficient for the price of water is negative at 10% level. The
coefficient was found to be about -0.02, and this indicated that the demand for water
was infinitely inelastic, and that the farmers were insensitive to the changes in the
price of water. Therefore, the estimated coefficients do not refute the first and second
hypotheses of the research; there is a negative relationship between the price of price
and the amount of demand for it, and that the price of water is inefficient.
The coefficients on wage and animal fertilizer price are also negative and significant
at 1% and 10% levels. This is a signal which indicates that water with labour force
and animal fertilizer are complementary inputs, in which a one percent growth in
wage and the price of animal fertilizer will lessen the demand for water by 0.14 and
0.03 percent, respectively.
The estimated coefficient for the output quantity is positive and significant at 1%
level. With an elasticity of 0.68, a one percentage rise in the output (cucumber) will
result in an increase of water usage by 0.68 percent. Accordingly, the estimated
coefficients support the fifth hypothesis of the research, i.e. the amount of crop
amount has a substantial impact on the water usage in the Iranian agricultural sector.
4.2.10 Watermelon
Watermelon is also another major crop whose production and export contribute to
the economy of Iran. Its production averaged at 2866324 (metric tons) in 2006. In the
same year, the cultivated areas for watermelons and the average application of water
for it were 119096 (ha) and 7542 (m3/ha), respectively.
117
The total irrigated production during the year under consideration was reported to be
about 2719320 tons, using 5761113 man days of labour. As reported by FAO, the
export level of watermelon crop was about 90775 metric tons, with value of $ 14516
million in 2004.
Meanwhile, the yields for watermelon in the Iranian provinces were indicated to
range from 4364 to 44386 (kg / ha). In particular, Khozestan province had the
largest irrigated cultivation area (21536 ha) and the highest production level of
680394 metric ton among all the provinces in Iran. Iran’s total irrigated area was
95718 ha in 2006. As mentioned in the section devoted for tomato, a comprehensive
dataset of studies conducted worldwide revealed that the global values for crop water
productivity (CWP) were generally very high for crops like tomatoes and
watermelons. With 6.38 and 2.01, Kerman and Fars provinces took the first and last
places respectively in terms of water productivity levels. The water productivity,
average application of water and yield for this crop in Khozestan province were 3.87,
8147, and 31593 (kg/ha), respectively. The main specifications of the watermelon
holdings for the different provinces in 2006 are presented in Table B.12.
The most prevalently used form of the derived demand functions are the CobbDouglas functions, in that the resulting coefficients have made it possible to interpret
the elasticity of demand, with respect to price of inputs, and amount of output.
The equation of water demand, as a function of the current price of water, fertilizer
and seed prices, wage, land rent, and the output amount, was then estimated using the
panel data method comprising of 126 observations from 21 producer provinces for
118
the period between 2001 and 2006. The Breush and Pagan Lagrangian Multiplier
Test was used to juxtapose the Pool and Panel Data approaches seeking for a suitable
function. Finally, the Panel model was proven to be more appropriate than the Pool
model. Meanwhile, the fixed effect and random effect were compared in the
Hausman’s specification test which was run on STATA 10. The results revealed that
the irrigation water demand function of watermelon could be best derived using the
fixed effect approach. The estimation for the best model was employed to carry out
the pretest and diagnostic checks. These estimated parameters are presented in Table
4-10.
Table 4-10 Water demand function for watermelon
Dependent Variable: LDWT
Method: Panel EGLS (Cross-section weights)
Linear estimation after one-step weighting matrix
White cross-section standard errors & covariance (no d.f. correction)
Independent Variable
C
LQ
LPW
LW
LPP
LRL
LPS
LPM
LCL
Coefficient Std. Error t-Statistic Prob.
1.10*
0.81**
-0.09***
-0.21***
0.047***
-0.05**
0.04***
-0.02***
0.26***
0.66
0.02
0.02
0.05
0.02
0.02
0.01
0.01
0.06
1.66
37.22
-5.39
-4.56
2.96
-2.07
2.84
-2.78
4.19
Cross-section fixed (dummy variables)
R-squared
0.99
Adjusted R-squared
0.99
Durbin-Watson stat
2.06
686.773
F-statistic
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
119
0.10
0.00
0.00
0.00
0.00
0.04
0.01
0.01
0.00
Based on the results, the estimated coefficient for the price of water is negative at 1%
level. The coefficient was found to be approximately -0.09. This indicates that the
demand for water is infinitely inelasticity, and that the farmers are insensitive to the
changes in the price of water. Thus, the estimated coefficients vindicate the first and
second hypotheses of the research; the price of water and the amount demanded for it
are negatively related, and that the price of water is not efficient.
Likewise, the coefficients for wage, machinery rental and land rental are negative,
and they are significant at 1%, 1% and 5% levels, respectively. These suggest that
water, labour force, land, and machinery services are complementary inputs, which
explains that a one percent increase in the wage, machinery rental, and land rental
will decrease the demand for water by 0.21, 0.02 and 0.05 percent, respectively.
The coefficients for the price of seeds, price of pesticide, and cost incurred in land
preparation are all positive and significant at 1%. Hence, it could be deduced that
water and all the inputs given above are substitute inputs, and this means that a one
percent increase in price of seed, pesticide, and cost in land preparation will eleviate
demand for water by 0.04, 0.05, and 0.26 percent, respectively. Among the reasons
for the existence of a substitution relationship between land preparation and water
usage is that quality of soil and its ability by absorption and retention of water was
considered as a whole.
The estimated coefficient for quantity of output is positive and significant at 1%
level. The estimated parameter coefficient suggests the elasticity of water use, with
respect to the quantity of output, is 0.81. This also means that a one percent increase
120
in the output (watermelon) will result in a 0.81 percent increase in the use of water.
The yielded coefficients also confirm the fifth hypothesis of the research, i.e. the
amount of crops has a strong effect on the usage of water.
When we select Cross section weights, EViews will estimate a feasible GLS
specification assuming the presence of cross-section heteroskedasticity.
4.2.11 Cotton
With a production of about 283673.34 (metric tons) in 2006, cotton is a crop of great
importance in Iran’s agricultural sector. In the same year, the areas cultivated for
cotton amounted to 116560 (ha), and an average application of water for it was about
100884 (m3/ha).
Cotton yields from various irrigated areas in Iran provinces ranged from 1663 to
3318.55 (kg/ ha). With a production of 152348 metric tons, Khorasan province had
highest production level. The province also contained a total cultivation area of
63722 ha, putting it in the first place among all other Iranian provinces for the
production of cotton. The total irrigated production of cotton in Iran for 2006 was
about 113345 metric tons, employing 8870633 man days of labour force.
The water productivity for cotton in Iran was between 0.013 and 0.12 in 2006. The
water productivity, average application of water and yield in the Khorasan province
for this crop were repored to be approximately 0.021, 110436 (m3/ha), and 2321.14
(kg/ha), respestively. The key specifications of cotton holdings by provinces in 2006
are presented in Table B.13.
121
The Cobb-Douglas form was used for the derived demand functions. The water
demand equation explained in the previous chapter (including the current price of
water, price of fertilizer and seed, wage, land rental, and amount of output) was
estimated.
The panel data method was used to estimate irrigation water demand based on 78
observations from 13 producer provinces for the period between 2001 and 2006. The
Breush and Pagan Lagrangian Multiplier Test made it possible to find the Panel Data
approach which had been proven as better than the Pool Data approach. Similarly,
the Hausman’s specification test was performed on STATA 10, and it suggested that
the fixed – fixed effect approach was the best choice among all other approaches.
Finally, the best model was estimated after the pretest and diagnostic checks had
been carried out. The estimated parameters are listed in Table 4-11.
The results show that the estimated coefficient for the price of water is negative at
1% level. Infinite inelasticity of water demand could be discerned from -0.41, as the
value of this variable coefficient. In other words, this also indicates that farmers are
relatively sensitive to the changes which have taken place in the price of water.
Hence, the estimated coefficients justify the first and second hypotheses of the
research; there is a negative relationship between the price of water and its demand
level, and the price of water is not efficient.
122
Table 4-11 Water demand function for cotton
Dependent Variable: LDWT
Method: Panel Least Squares
White cross-section standard errors & covariance (no d.f. correction)
Independent Variable
Coefficient Std. Error t-satistic Prob.
C
LPW
LQ
LW
LRL
LCL
LPS
2.49
2.76
-0.41
0.08
0.73
0.19
-0.27
0.16
0.12
0.036
0.56
0.156
-0.09
0.126
Cross-section fixed (dummy variables)
R-squared
0.97
Adjusted R-squared
0.96
Durbin-Watson stat
2.04
F-statistic
80
Prob(F-statistic)
0
*
Statistically significant at the 10% level;
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
0.90
-5.02
3.81
-1.70
3.66
3.75
-0.70
0.37
0
0.0
0.1
0.0
0.0
0.48
The coefficients for the cost in land preparation and land rental are positive and
significant at 1% level. From this, it can be concluded that water and all the above
inputs are mere substitutions. For this reason, a one percent rise in the cost for land
preparation and land rental will increase the demand for water by 0.56 and 0.12
percent, respectively.
Likewise, the coefficients for the price of seed and wage are negative. Hence, water
with seed and labour force comprise complementary inputs. This means, a one
percent increase in the price of seed and wage will lower the demand for water by
0.09 and 0.27 percent, respectively.
123
The estimated coefficient for the quantity of output is, however, positive and
significant at 1% level. The elasticity of water use, given the changes in the quantity
of output is indicated by the estimated parameter coefficient. This also means that a
one percent increase in the output (cotton) quantity will bring about a 0.73 percent
increase in the consumption of water. Accordingly, the obtained coefficients are in
compliance with the fifth hypothesis of the research, i.e. in Iran’s agricultural sector,
the amount of crops has a decisive effect on the usage of water.
4.2.12 Sugar Beet
Another major industrial crop produced in Iran is sugar beet, with a production level,
cultivated area, and average water application in 2006 to be around 6709112 (metric
tons), 185888 (ha), and 14664 (m3/ha), respectively. Sugar beet yields in irrigated
areas in the Iranian provinces ranged from 25000 to 54411 kg/ha for the same year.
Among all the provinces, Khorasan had the largest irrigated cultivation area totalling
63279 ha and the highest production level of 2006475 metric tons. For the total
irrigated production of about 6709112 metric tons in 2006, 15394605 man days of
labour were hired.
In Iran, the water productivity for sugar beet was between 1.4 and 3.42 in 2006. The
water productivity, average application of water, and yield in the Khorasan province
for sugar beet were reported to be about 1.92, 14779 (m3/ha) and 28348 (kg),
respectively. Table B.14 (Appendix B) shows the important specifications of sugar
beet holdings in the different provinces in 2006.
124
The main equation to be estimated is the water demand equation, as presented in the
previous chapter. In this equation, the demand for water serves as a function of the
current price of water, prices of fertilizer, seed, wage, land rental, and amount of
output.
The method used to estimate irrigation water demand was the Panel Data.
Meanwhile, a total of 84 observations taken from 14 producer provinces during the
period between 2001 and 2006 were used in the model. As the results of the Breush
and Pagan Lagrangian Multiplier Test conducted, the Panel Data approach was used
as a more appropriate one to be used to obtain a suitable function. Based on the
Hausman’s specification test performed on STATA 10, the fixed effect approach was
found to be the best among the fixed effect approach and random effect. Thus, the
fixed effect approach was used to model the irrigation water demand function for
sugar beet. Finally, the pretest and diagnostic checks were done and the most suitable
model was estimated. These estimated parameters are presented in Table 4-12.
The results show that the estimated coefficient for the price of water is -0.04 and it is
significant at 1% level. Hence, the demand for water is infinitely inelastic. It also
indicates that farmers are very sensitive to the change in the price of water.
Therefore, the estimated coefficients do not contradict with the first and second
hypotheses of the research; that there is a negative relationship between the price of
water and the amount demanded for it and that the price of water is not efficient.
125
Table 4-12 Water demand function for sugar beet
Dependent Variable: LDWT
Method: Panel EGLS (Cross-section weights)
Linear estimation after one-step weighting matrix
White cross-section standard errors & covariance (no d.f. correction)
Independent Variable
C
LPW
LW
LQ
LPS
LPM
LCL
Coefficient Std. Error t-Statistic Prob.
2.41***
-0.04***
-0.16***
0.88***
-0.03***
-0.04***
0.09
0.25
0.01
0.04
0.02
0.005
0.02
0.06
9.53
-3.68
-3.60
44.53
-7.94
-2.32
1.57
0.00
0.001
0.001
0.00
0.00
0.02
0.12
Cross-section fixed (dummy variables)
R-squared
0.99
Adjusted R-squared
0.99
Durbin-Watson stat
2.22
F-statistic
811.13
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
On the contrary, the coefficients on wage, machinery rental and price of seed are
negative and they are significant at 1%, 5% and 1% levels, respectively. Hence,
water with labour force, machinery services, and seed comprise complementary
inputs. In more specific, a one percent increase in wage, machinery rental, and price
of seed will reduce the water demanded by 0.16, 0.04 and 0.03, respectively.
On the other hand, the coefficient for the cost in land preparation is positive,
indicating that water and land preparation are substitute inputs, in which a one
percent rise in the cost for land preparation will increase the demand for water by
0.09 percent.
126
Similarly, the estimated coefficient for the quantity of output is positive and
significant at 1% level. The estimated parameter coefficient gives an estimated
elasticity of water use with respect to the quantity of output. Thus, a one percent
increase in the output (sugar beet) quantity will lead to a 0.88 percent rise in the
usage of water. This also shows that the estimated coefficient is consistent with the
fifth hypothesis outlined in the current research, i.e. in the agricultural sector of Iran;
the usage of water is very much determined or influenced by the amount of crops.
4.3
Estimations of the Model with Alternative Prices
As mentioned in Chapter one, the main goal of this study was to design a suitable
mechanism for determining an efficient pricing system for irrigation water. The
current prices are not efficient because the water demand elasticity with respect to
the current price is close to zero (infinitely inelasticity). Therefore, an irrigation
water demand function via minimization of cost function will be used. The estimated
coefficients of irrigation water demand will be used for analyzing water pricing
system.
On the supply side, water pricing system based on average and marginal cost will be
used. Secondary data on expenditures of a district on irrigated water supply delivery
will be used to calculate average and marginl cost per cubic meter of water. The
average variable cost of water supply delivery has been calculated using operating
and maintaining expenditures on the irrigation system.
127
The descriptive data presented in Table 4-13 includes the current price and water
alternative prices for all crops. While variables are defined for estimate as natural
logs are represent the original values.
Table 4-13 Descriptive statistics of current price and water alternative prices
Variable
MC
Crop
AVC
WCP
Tomato
VMP
WCP
Watermelon
VMP
WCP
Pinto bean
VMP
WCP
Sugar beet
VMP
WCP
Lentil
VMP
WCP
Onion
VMP
WCP
Potato
VMP
WCP
Pea
VMP
WCP
Barley
VMP
WCP
Cotton
VMP
WCP
Wheat
VMP
WCP
Cucumber
VMP
* WCP is Water Current Price
Mean
544.0456
31.09896
14.96
830.52
2999.25
308.04
10.44
142.98
8.84
161.01
6.36
65.55
13.38
341.26
12.92
529.12
10.66
96.70
5.93
57.91
3.49
66.39
6.71
138.29
21.54
613.09
Std. Dev.
5604.868
92.05422
11.24
597.67
2940.58
202.90
11.96
65.60
6.02
310.85
4.84
37.88
8.38
213.31
10.33
2044.01
12.17
71.94
4.22
22.29
2.69
57.08
4.28
55.56
18.31
410.21
Observations
105
105
144
144
123
126
58
58
82
82
44
44
100
100
138
138
63
63
151
151
78
78
167
167
108
108
The average MC is 544 and a standard deviation of 5605, while average AVC is 31
and a standard deviation of 92. The average water current price (WCP) for tomato,
watermelon, pinto bean, sugar beet, lentil, onion, potato, pea, barley, cotton, wheat
128
and cucumber are 14.96, 2999, 10.44, 8.84, 6.36, 13.38, 12.92, 10.66, 5,93, 3.49,
6.71 and 21.54 Toman/m3, and with a standard deviation of 11.24, 2940.6, 11.96,
6.02, 4.84, 5.38, 10.33, 12.17, 4.22, 2.7, 4.28 and 18.31, respectively. While average
VMP for tomato, watermelon, pinto bean, sugar beet, lentil, onion, potato, pea,
barley, cotton, wheat and cucumber are 830, 308, 143, 161, 655, 341.3, 529.1, 96.7,
57.9, 66.4, 138.3 and 613 Toman / m3and with a standard deviation of 597.7, 202.9,
65.6, 310.85,37.88, 213.31, 2044, 71.94, 22.3, 57.08, 55.56 and 410.21, respectively.
The shadow price of water (labelled as VMP of water) that could be calculated from
the equation (3-69), the average variable cost of water supply delivery (using O&M
expenditures on the irrigation system), and the short-run marginal cost would use an
alternative price instead of the current price.
As mentioned earlier, the Panel Data method was used to estimate the irrigation
water demand. First, Breush and Pagan Lagrangian Multiplier test was used to
choose between Pool and Panel Data approaches. This test revealed that the Panel
Data approach was more appropriate than the Pool Data approach. Hausman’s
specification test was also conducted on STATA 10 (an econometric software
package) to choose one method out of the fixed effect and random effect. Based on
the test, the best model for the irrigation water demand functions for all crops were
found to be fixed effect approach. Next, pretest and diagnostic checks were done
upon which the best model was estimated. The estimated results are presented in
Appendix F. Meanwhile, the estimated coefficients for the water alternative prices
(VMP, AVC and MC) are shown in Table 4-14
129
Table 4-14 Estimated coefficients for the alternative prices of water
Dependent Variable: LDWT
Independent AVC
MC
Crop
Coefficient Std.
Coefficient
***
Wheat
-0.01
0.006 -0.07***
Barley
-0.075***
0.02 -0.03*
0.04 -0.10***
Lentil
-0.09*
*
0.01 -0.06*
Pea
-0.03
0.015 -0.01
Pinto Bean -0.06***
***
0.01 -0.01***
Onion
-0.04
0.02 -0.01
Tomato
-0.04**
Potato
-0.004
0.01 -0.02
0.01 -0.01***
Cucumber
-0.01**
0.003 0.006***
Watermelon -0.005*
***
0.01 0.09*
Cotton
0.02
0.005 -0.08**
Sugar Beet -0.01**
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the1% level
VMP
Std.
Coefficient
0.007 -0.20***
0.02 -0.66***
0.03 -0.88***
0.03 -0.49***
0.05 -0.39***
0.02 -0.20***
0.01 -0.89***
0.01 -0.16***
0.003 -0.21***
0.009 -0.26***
0.05 -0.88***
0.03 -0.50***
Std.
0.05
0.02
0.04
0.04
0.05
0.05
0.03
0.02
0.02
0.04
0.08
0.10
The regression results show that the estimated coefficients for the water alternative
prices (VMP, AVC and MC) are negative and significant at 1% level. These
coefficients are also very low when the price of water is equal to AVC and MC. It is
because both average variable cost and marginal cost are very close to the current
price. Therefore, the estimated coefficients do not contradict with the third
hypothesis of the research for the Iranian agricultural sector, i.e. pricing based on the
marginal cost and average variable cost of water has no effect on decreasing water
usage.
Nonetheless, it is approximately efficient when the price of water is equal to the
value of marginal product (VMP). These coefficients indicate that when the price of
water is equal to the VMP, a one percent increase in the price of water will cause
water demanded for tomatoes to decrease by 0.89 percent. In other words, farmers
will use lesser amounts of water when the price is higher.
130
Therefore, the estimated coefficients do not contradict the fourth hypothesis of the
research conducted in the Iranian agricultural sector, i.e. pricing based on the value
of marginal product has an efficient effect on decreasing usage of water.
One of the most significant findings emerging from this study is that the range of the
estimated coefficients for the alternative prices of water, based on the VMP is
between 0.14 (for cotton) and 0.89 (for tomato).
These results may be explained by the facts that there is no close substitute for water,
and that farmers allocate such a tiny fraction of the costs they bear to water (Sloman,
2003). This means that the price of water is rather low, that is, each factor is totally
inelastic if the price is very low.
4.4
Discuassion
The estimated results indicate that the social and political pricing system (that the
water rate is determined so that only the maintenance and operation costs covered)
and current pricing system (social and economic pricing that is based on the average
cost and water rate is determined so that cover all or part of the investment costs in
addition to costs of the maintenance and operation.) is inefficient in Iran’s
agricultural sector.
According to ( Johansson, 2000), an economically efficient allocation of water is one
that results in the highest return for a given water resource. He also suggested that to
attain this effectiveness, the price of water should be identical to the marginal cost of
131
supplying an additional unit of water plus the shortage value of the resource. Garcia
and Reynaud (2004) noted that maximizing social welfare leads to public utility of
marginal-cost pricing (MCP). Maximizing aggregate net surplus leads to the famous
law of equality of price and social marginal-cost. They also argued that due to a
number of criticisms against marginal cost pricing (First - best water pricing), the
“revenue-recovery principle” has played the primary rule in the design of water
prices.
By the way, marginal cost changes since irrigation decisions are functions of
geographical conditions and seasonal differences. This fact requires that different
prices are charged at different times. Likewise, the marginal cost to society of
delivering one unit of water to a farmer at tail end may be higher than that of the
same unit of water to a farmer nearer to the source of water supply. Indeed, water
supply costs should include items corresponding to maintenance (Easter, 1987),
collection of water and relevant fees (Small and Carruthers, 1991), social cost
(benefit), scarcity, infrastructure, extraction cost externalities (Johansson, 2000). In
2001, Griffin showed that water price should also include opportunity costs such as,
user’s marginal cost of water (to take into account sacrifice of future uses of
unrenowned groundwater supplies); marginal value of raw water (surface water and
fully renewable ground water sources, in scarcity situations); marginal capacity cost
(when water supplied with capacity installed is less than water demand).
In 2002, Sahibzada noted that water rate determination principle often favored by
economists is to base charges on the value of service, i.e. on the marginal product
value of water which equals, at equilibrium, the price farmers are willing to pay for
132
water. He also stated that the opportunity cost of the various inputs is normally taken
to be their market prices, in the absence of their shadow prices. Where no market
price exists, the cost of a particular input in its best alternative use is substituted for
its price.”
Shiferaw et al. (2008) illustrated that in the absence of a water market, optimal
allocation of irrigation water will require the shadow price to equal its marginal
product value.
Therefore, the estimated results reveal that water price is approximately efficient
when the price of water is equal to the value of marginal product (VMP).
Consequently, Water pricing can be used as an important means of improving water
allocation and encouraging users to conserve water resources as mentioned in chapter
two, this criterion ought to be more appropriate to economic decision making in Iran.
133
CHAPTER V
5
5.1
SUMMARY, CONCLUSION AND RECOMMENDATIONS
Summary
Planning for an efficient use of water resources is one of the most important current
discussions about water resources management. This is largely due to the fact that
water is a fundamental prerequisite to achieve goals such as increasing level of food
production, energy development, and wide industrial activities. Moreover, water also
plays a key role in agricultural activities as it is an important component of the
climatic system.
In most arid and semi - arid areas like most parts of Iran, inadequate supply of water
is confronting many people, and this is one of the major constraints on economic
development in the country. In such areas, the core issue in water management is
equilibrating the supply and demand for water. On one hand, the supply of water is
often limited, while on the other, the quantity demanded is constantly rising, mainly
as a result of growth in the country’s population. Therefore, establishing a wise water
pricing system is crucial to attain an optimal allocation of water resources.
5.1.1
Purpose and Objectives
In this study, the main goals were to analyze the current pricing mechanism (policy)
for water and determine alternative pricing policy so as to optimize the utilization of
this scarce resource. This study focused to determine suitable and efficient pricing
134
system for irrigation water. Thus, water demand functions for major crops in the
region were estimated in order achieve these objectives.
Reflecting on the hypothesis posed at the beginning of this study, it is now possible
to state that an increase in the price of water delivered to the agricultural sector is
effective as a measure to be used in conserving this rare resource and in ensuring its
efficient consumption.
The hypotheses on which this research was based on are listed below:
1. Price of Irrigation water is inefficient. In other words, water demand is
inelastic with respect to its price.
2. Price elasticity of water in Iran’s agricultural sector is less than one.
3. In Iran’s agricultural sector, pricing based on marginal cost and average
variable cost of water supply has no effect on decreasing water usage by
farmers.
4. In Iran’s agricultural sector, pricing based on value of marginal product has
an effect on decreasing water usage.
5. In Iran’s agricultural sector, crop quantity has a significant effect on water
usage.
5.1.2
Research Procedures
In approaching the aforementioned objectives, this research used secondary data to
assess the efficiency of water price. Major crops considered herein include wheat,
barley, lentil, pea, pinto bean, onion, tomato, potato, cucumber, watermelon, cotton,
and sugar beet. Data and information related to these crops were collected from the
135
statistical reports for 28 producing provinces in Iran within the period between 2001
and 2006. The analysis was based on the econometric method of panel data.
Meanwhile, demands for irrigation water were estimated as functions of water price,
land rent, fertilizer price, as well as rental rate for machinery, seed price, wage,
animal fertilizer price, pesticide price, and irrigated production level. In this study,
regression coefficients were estimated and a suitable pricing system of irrigation
water in these selected provinces was also estimated upon performing the relevant
statistical tests.
The natural logarithm functional forms were used to estimate the functions of water
demand. The parameters of the demand functions were estimated using the ordinary
Least Squares (OLS) or Generalized Least Squares (GLS), Estimated Generalized
Least Squares (EGLS), and/or Weighted Least Squares (WLS).
5.1.3
Research Findings
The following conclusions were drawn from the present study:
1. The price elasticity of demand for water at the price based on VMP is
relatively inelastic (0 < Ed < 1) for most of crops.
2. Based on these findings, farmers will generally use lesser amount of water
when the price is higher. Hence, decision makers and authorities can use
price as a policy instrument for water conservation.
136
3. For most of the crops, the price elasticity of demand for water is perfectly
inelastic at the current price, and this is similar for the prices which are based
on AVC and MC. Indeed, each factor is totally inelastic when its price is very
low. Consequently, the water prices based on the current price and on AVC
and MC are inefficient. A possible explanation for this is that at these prices,
farmers are less responsive to changes in the quantity of water.
4. The price of the water for most of the crops is close to efficient when the
water price is given based on the value of marginal product.
5. As mentioned above, the price elasticity of water demand at the current price,
and also, at prices based on AVC and MC is very low for the crops. This is
because there is no substitute for water, and water cost comprises only a
negligible fraction of total costs borne by farmers, i.e. the price of water is
very low.
6. As for wheat, the price elasticity of demand for water at the current price, and
the prices based on VMP, AVC and MC were found to be -0.036, -0.20, -0.02
and -0.05, respectively; that is, a one percent increase in the price of water
will cause demand for water to decrease by 0.036, 0.20, 0.02 and 0.05
percent, respectively.
7. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC were -0.02, -0.66, -0.07 and -0.03 respectively
for barley. This indicates that a one percent increase in the price of water will
137
cause water demand to decrease respectively by 0.02, 0.66, 0.07 and 0.03
percent.
8. For lentil, the price elasticity of demand for water at the current price, and the
prices based on VMP, AVC and MC were -0.25, -0.88, -0.09 and -0.1,
respectively. In other words, a one percent increase in the price of water will
cause water demand to decrease by 0.25, 0.88, 0.09 and 0.1 percent,
respectively.
9. As for pea, the price elasticity of demand for water at the current price, and
the prices based on VMP, AVC and MC were -0.11, -0.49, - 0.03 and -0.06,
respectively, indicating that a one percent rise in the price of water will lead
to a decrease in water demand by 0.11, 0.49, 0.03 and 0.06 percent,
respectively.
10. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC for pinto bean were -0.03, -0.39, -0.06 and
-0.01, respectively. This means a one percent increase in the price of water
will lower demand for water by 0.03, 0.39, 0.06 and 0.01 percent,
respectively.
11. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC for onion were -0.02, -0.20, -0.04 and -0.01,
respectively; that is, a one percent rise in water price can reduce demand for
water by 0.02, 0.20, 0.04 and 0.01 percent, respectively.
138
12. For tomato crop, the price elasticity of demand for water at the current price,
and the prices based on VMP, AVC and MC were -0.06, -0.89, -0.04 and
-0.01, respectively. This indicates that a one percent increase in the price of
water will decrease demand for water by 0.06, 0.89, 0.04 and 0.01 percent,
respectively.
13. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC were -0.06, -0.16, - 0.004 and -0.02
respectively for potato crop. In other words, a one percent grow in the price
of water will lower demand for water by 0.06, 0.16, 0.004 and 0.02 percent,
respectively.
14. For cucumber crop, the price elasticity of demand for water at the current
price, and the prices based on VMP, AVC and MC were -0.02, -0.21, -0.01
and-0.01, respectively. This also means that a one percent increase in the
price of water will reduce demand for water by 0.02, 0.21, 0.01 and 0.01
percent, respectively.
15. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC for were -0.09, -0.26, -0.01 and -0.01
respectively for watermelon crop; that is, a one percent rise in the price of
water will cause demand for water to reduce by 0.09, 0.26, 0.01 and 0.01
percent, respectively.
139
16. The price elasticity of demand for water at the current price, and the prices
based on VMP, AVC and MC were found to be -0.41, -0.88, -0.02 and -0.09
respectively for cotton crop. This reveals that a one percent increase in the
price of water will lessen demand for water by 0.15, 0.14, 0.02 and 0.09
percent, respectively.
17. For sugar beet crop, the price elasticity of demand for water at the current
price, and the prices based on VMP, AVC and MC were indicated at -0.04,
-0.50, -0.01 and -0.08, respectively. This means that a one percent increase in
the price of water will lead to a decrease demand for water by 0.04, 0.50, 0.01
and 0.08 percent, respectively.
18. The estimated coefficients for output are positive and significant for all crops.
These coefficients, in logarithmic functions, indicate the elasticity of water
usage given a change in the quantity of output. This means farmers tend to
use more water when the demand for crops is higher. Decision makers and
authorities can use this elasticity (coefficients) as a policy instrument for
water conservation. The coefficients for wheat, barley, lentil, pea, pinto bean,
onion, tomato, potato, cucumber, water melon, cotton, sugar beet were 0.38,
0.81, 1.11, 0.96, 0.99, 0.89, 0.30, 0.79, 0.68, 0.81, 0.73, and 0.88,
respectively. For example, a one percent increase in the demand for wheat
leads to a 0.38 percent rise in the usage of water. This relationship could be
used to determine the impact(s) of production quotas on water supply.
140
19. The marginal production value of water for each crop was computed via these
estimated coefficients (output amount coefficients).
20. The estimated coefficients for wage for most of the crops are negative and
significant, implying that water and labour are complementary inputs. In fact,
a one percent increase in wage will decrease demand for water by Bi percent.
21. The estimated coefficients for land rental are negative and significant for
most of the crops, illustrating that water and land are also complementary
inputs. Therefore, a one percent rise in the rental of land will reduce demand
for water by Bi percent.
22. The estimated coefficients for the price of fertilizer for the majority of crops
are negative and significant, implying that water and fertilizer are
complementary inputs. Hence, a one percent increase in the price of fertilizer
will lower demand for water by Bi percent.
23. The estimated coefficients for the price of seed (or the cross price elasticity
with respect to the price of seed) for almost all the crops are both positive and
significant. These indicate that a one percent rise in the price of seed will
cause a decrease in water demanded by Bi percent. Although the existence of
a complementary relationship between water and seed inputs produces a
negative relationship between the price of one and the quantity of demand, as
well as the accumulation of the price of seed for respective crops could be
141
deemed as the primary factor in the formation of the variable in the demand
function.
24. Likewise, noting the fact that the prices of input were used as major factors
contributing to the price of water, the sign of the estimated coefficient seems
to be appropriate. Furthermore, in light of a large proportion of seed cost
(second after that of water’s) in the total production cost, this particular input
and water appear to be considered by farmers as two substitutes. Another reason for this could be the employment of price index of seed, and also, the existing variety being used.
25. The estimated coefficients for the price of pesticide for most crops are
positive and significant. Hence, water and pesticide are substitute inputs, in
which a one percent increase in the price of pesticide will cause demand for
water to increase by Bi percent.
26. The estimated coefficients for the rental of machinery services for most of the
crops are also positive and significant. This indicates that water and
machinery services are substitute inputs. In this relationship, a one percent
increase in the rental of machinery services will increase demand for water by
Bi percent.
5.1.4
Contributions of the Study
This study has filled in the gaps present in the research conducted previously. First, it
provides consistent and comprehensive estimations for the price water (values)
142
across several regions in Iran. These are based on the farmers’ optimizing behaviour
over some of the outputs (including wheat, barley, lentil, pea, pinto bean, onion,
tomato, potato, cucumber, water melon, cotton, and sugar beet) and inputs (price of
water, land rental, price of fertilizer, machinery rental, price of seed, wage, price of
animal fertilizer and pesticide, and irrigated production level of the crops). Although
there has been much estimation done or suggested for the value of water in Iran’s
agriculture, no estimation for the values across different regions in Iran is available.
Several studies have found that the value of water changes with or according to
region. Keramatzade et al. (2006) used a linear programming technique to determine
the economic value of agricultural water in Shirvan Barzo Dam. On the contrary,
Salami and Mohammad Nejad (2002) employed flexible production functions, and
estimated the economic value of irrigation water in Saveh region. Meanwhile,
Mansouri and Ghiasi (2002) estimated the cost at the point of reservoir dams of
irrigation water in the West Azarbaijan province in 1998-99 using the engineering
economic approach.
Second, this study has also revealed that the type and value of crops, as well as the
geographical factors have significant effects on demand for water and its price.
However, previous studies on Iran did not test the role of these factors. For instance,
Chizari et al. (2006) used a goal programming approach to determine the optimum
cropping pattern and economic value of irrigation water in three regions in Shirvan
Barzo Dam located in the north of Khorasan province.
143
Finally, this study has shown that at the current prices, farmers are not using water
efficiently. Thus, to fill these gaps, this research used a microeconomic theory by the
panel data econometric method to measure the economic inefficiency of water
resources in the agricultural sector of Iran. As discussed in the above mentioned
studies, this particular model is different from the prior models and methods which
have mostly been applied in the agricultural sector. In the panel data econometric
method, the intercept coefficients for every province are considered. These
coefficients can show regional variations such as climatic or soil characteristics, and
prices of output.
As mentioned previously, the water productivity calculated provides a wide range of
economic information and policy implications which can be used to choose proper
crops to be grown during droughts.
In the empirical field, this research has provided the necessary economic information
to the policy makers and authorities alike in Iran’s water sector to conserve water
resources and manage their usage in the optimal way. Finally, the outcomes of this
research can be used by policy makers and provincial governments to determine an
efficient price for water.
5.2
Conclusion
Water is a blessing from the Almighty God, and it is easily accessible at a minimal
cost. It has been known as a public good that needs finance to cover the cost of
servicing worldwide.
144
As aforementioned, the quantity of water supplied is usually constant while the
quantity demanded, mainly because of the growth in population, is continuously
rising. Thus, authorities should fulfil the basic needs of the people using innovative
water efficiency and conservation strategies, community - scale projects, welfare
economics, and new technology. Consequently, planning for an efficient use of water
resources is of special interest. Furthermore, prices have been proven to play an
important role in the equilibrium generation between supply and demand of water.
This study has been directed towards evaluation and modification of the current price
of water in order to save and optimally use this rare resource. For the purpose of the
analysis conducted in this study, the derived demand functions of irrigated water
were estimated, and the price elasticity were extracted from these irrigated water
functions. Expectedly, the results of this study suggest that the current prices of water
are not efficient since the estimated coefficients elasticity were found to be relatively
low. In fact, only a small proportion of the total costs borne by farmers is that of
water (Sloman, 2003), indicating that the price of water is very low.
5.2.1
Policy Implications of the Empirical Findings
The findings of this study have a number of important implications for both practice
and prospective polices in the future. These implications include:
1. A major conclusion may be drawn from this study, i.e. the reforms in the
price of irrigation water would have a significant effect on the usage of this
rare resource. In fact, this study could be instructive to policy makers when
145
they attempted to propose a fair suggestion for modifications in the pricing of
water, particularly in case of water shortage.
2. The results of the study also show that significant increases in the price of
water generally lead to lesser use of water resources, but higher productivity
of them.
3. Another implication of these findings is that both price of water and quantity
of crop should be taken into account by the authorities whenever there is
water shortage.
4. An important practical implication is that this study has shown yield, crop
water productivity, and average application of water for many crops in each
province. This useful information reveals the water requirements for each
crop in different provinces. It also indicates the efficiency of water usage in
terms of the utmost agricultural production per unit irrigation water. In other
words, this information provides the comparative advantage of production in
relation to the use of water for each crop in every province. At the same time,
this information can be used by the relevant authorities in decision making
when rainfall is limited or when there is water shortage.
5. The findings of this study can help policy makers in choosing the kind of
crops for production during drought. As prices of water differ from one crop
to another, farmers will normally opt to cultivate crops that incur lower cost
in usage of water or lower prices. For this, reason, it is important for water
146
authorities to control the allocation of water to the farmers and encourage
certain cropping patterns.
6. The findings of this study suggest that the price of water is fairly efficient
where its value of marginal productivity equals the price of water.
7. This study also revealed that the productions of wheat, barley, lentil, peas,
and tomatoes were prominent in Fars province in terms of agro-socio-climatic
conditions.
8. The findings of this study pointed out that the agriculture sector must manage
its shrinking water supplies more efficiently, and that authorities should
employ control in water pricing to achieve this goal.
9. Finally, the farmers’ situation and the welfare of irrigators must be
considered in planning for water conservation based on the reforms in
irrigation water pricing. In fact, a naive policy may raise serious economic
problems such as unemployment in villages and emmigration to cities,
hindering economic development, threatening food production based on
agricultural crops, etc.
5.2.2
Limitations of the Current Research
A number of important limitations need to be considered, herein. These limitations
are given below:
147
1. The first limitation of this study is that the secondary data employed do not
reflect the actual farmers’ characteristics, i.e. the analyses are at a macrolevel rather than at a farm - level.
2. The second limitation is that for some crops, the number of observations
differs from the other panel members. In fact, this study relied heavily on the
estimation of unbalanced panel.
3. Avoidance of model complexity, multi - colinearity and the loss of a certain
degree of freedom are among the other limitations of this study. Moreover,
this study had relied on the Cobb-Douglas functional form.
4. Another limitation of this study is that for some crops, the number of
producer provinces is limited. Consequently, the results derived from them
may offer a little bias.
5. Furthermore, the information about the market price of water and various
inputs such as labour, electricity, material, and capital used in production
processes is also not available. Likewise, the information on demand for
water is also limited to only 6 years (2001-2006).
5.2.3
Recommendations for Future Research
The limitations in relation to the methods used in this study could serve as bases for
future studies; whereby it would be better to estimate sub - regional demand
functions from representative farms. It is recommended that further research should
148
focus on provinces such as Fars, Khozestan, Khorasan, Esfahan, Kohkiloy &
Boyerahma, and Kermanshah.
Several methods have been suggested to be used for estimating water demand
functions for conservation of water resources, and these should be investigated to
provide the most complete information to authorities. The coefficients estimated
through the econometric methods which rely on the secondary data tend to be more
inelastic than the ones recommended by the mathematical programming models.
However, they are very elastic in some cases. Thus, a further study with more focus
on various methods is suggested. Furthermore, more information on water pricing
reforms would help policy makers to establish a greater degree of accuracy in this
regard.
149
REFERENCES
Aghthe, E. D., & Billings, R. B. (1987). Equity, Price Elasticity and Household
Income under Increasing Block Rates for Water. The American Journalof
Economics and Sociology, 46(3).
Arriagada, R. A. (2004). Estimating Profitability and Fertilizer Demand for Rice
Production Around the Palo Verde National Park, Costa Rica. North
Carolina State University.
Asadi, H., Soltani, G. R., & Torkamaani, J. (2007). Irrigation water pricing in Iran: A
case study on land downstream of Taleghan dam. Eqtesad - E Keshavarzi Va
Towse'e, 15(58), 61-90 [in Persion].
Baltagi, B. H., & Griffin, J. M. (1983). Gasoline demand in the OECD : An
application of pooling and testing procedures. European Economic Review,
22(2), 117-137.
Berck, P., Robinson, S., & Goldman, G. E. (1990). The Use of Computable General
Equilibrium Models to Assess Water Policies. In A. Dinar & D. Zilberman
(Eds.), The Economics and Management of Water and Drainage in
Agriculture. Amsterdam: Kluwer Publishing Company.
Burke, T. R. (1970). A Maunicipal Water Demand Model for the Counterminous
United States. Water Resource Bulletin, 6(July-August), 661-681.
Chembezi, D. M. (1990). Estimating Fertilizer Demand and Output Supply for
Malawi's Smailholder Agriculture. Agricultural Systems, 33, 293-314.
Chizari, A. H., Sharzehi, G. A., & Keramatzadeh, A. (2006). Determination of the
economic value of the irrigation water using goal programming approach:
(Case study of Shirvan Barzo Dam). Tahghighat -E-Eghtesadi, 71, 39-66. [in
Persion].
Clarke, R. (1991). Water: The International Crisis: Eart Scan.
Coelli, T., Lloyd-Smith, J., Morrison, D., & Thomas, J. (1991). Hedonic Pricing For
a Cost Benefit Analysis of a Public Water Supply Scheme. The Australian
Journal of Agricultural Economics, 35(1), 1-19.
Colby, G. B. (1989). Estimating the value of water in alternative uses. Nat.Resour. J,
29(2), 511-527.
Conradie, B. I., & Hoag, D. L. (2004). A review of mathematical programming
models of irrigation water values. Water S A, 30(3), 287-292.
150
Dandy, G., McBean, C., & Hutchinson, B. (1984). A model for constrained optimum
water pricing and capacity expansion. Water Resources Research, 20(5), 511520.
Debertin, D. L. (2002). Agricultural production economics (2 ed.). Kentucky: David
L. Debertin.
Dinar, A., Rosegrant, M. W., & Meinzen-Dick, R. (1997). Water Allocation
Mechanisms: Principles and Examples: World Bank and IFPRI.
Easter, K. W. (1987). Inadequate Management and Declining Infrastructure: The
Critical Recurring Cost Problem Facing Irrigationin Asia (No. Report ER872): Institute of Agriculture, Forestry and Home Economics, University of
Minnesota.
Easter, K. W., Becker, N., & Tsur, Y. (1997). Economic Mechanisms for Managing
Water Resources: Pricing, Permits, and Markets, in A. K. Biswas (ed.). In
water Resources: Environmental Planning, Management and Development.
New York: McGraw-Hill.
Falkenmark, M., & Widstrand, C. (1992). Population and water resources: A delicate
balance. Population Bulletin 47(3), 1-36.
Faux, J., & Perry, G. M. (1999). Estimating irrigation water value using hedonic
price analysis: A case study in Malheur county, Oregon. Land Economics,
75(3), 440-452.
Garcia, S., & Reynaud, A. (2004). Estimating the benefits of efficient water pricing
in France. Resource and Energy Economics, 26(1), 1-25.
Gardner-Outlaw, T., & Engleman, R. (1997). Sustaining Water, Easing Scarcity: A
Second Update. Washington, D.C.: Population Action International.
Gardner, B. D., & Schick, S. H. (1964). Factors Affecting Consumption. of. Urban
Household Water in Northern Utah. Bulletin 449, Nov.
Ghazali, M., Attari, J., Sadeghi, A., & Arrif, H. (2009). Review of water pricing
theories and related models. African Journal of Agricultural Research, 4(13), 15361544.
Ghazi, I. (2002). Water Resources Management and Planning in Iran: Report to the
University of Isfahan.
Gleick, P. H. (2002). Dirty Water: Estimated Deaths from Water-Related Diseases
2000-2020: Pacific Institute for Studies in Development, Environment, and
Security.
Gottlieb, M. (1963). Urban Domestic Demand for Water: A Kansas Case Study.
Land Economics, 34, 204-210.
151
Griffin, R. C., & Perry, G. M. (1985). Volumetric Pricing of Agricultural Water
Supplies: A case study. Water Resource Research, 21(7), 944-950.
Griffin, R. C. (2001). Effective water pricing. Journal of the American Water
Resources Association, 37(5), 1335-1347.
Guilbe, A. (1969). Quantitative Analysis of Residential Water-Use Patterns: Water
Resources Research Institue.University of Puerto Rico.
Gujarati, D. N. (2002). Basic Econometrics: Irwin Professional Pub.
Gysi, m., & Loucks, d. (1971). Some Long Run Effects of Water-Pricing Policies.
Water Resource Research, 7(6), 1371-1382.
Hanke, S. H. (1970). Demand for Water under Dynamic Conditions. Water Resource
Bulletin, 6(OCT), 1253-1261.
Henderson, J. M., & Quandt, R. E. (1980). Microeconomic theory: a mathematical
approach (Vol. c1980). New York: McGraw-Hill.
Hinrichsen, D., Robey, B., & Upadhyay, U. D. (1997). Solutions for a Water-Short
World. Series M, No.14.
Hossein zad, J., & Salami, H. A. (2005). Choosing an empirical production function
to estimate economic value of irrigation water: A case study of wheat
production. Eqtesad - E Keshavarzi Va Towse'e, 12(48), 53-74 [in Persion].
Hossein zad, J., Salami, H. A., & Sadr, S. K. (2007). Estimation of economic value
of water used in agricultural products using flexible production function:
(Case study: Margheh-Bonab Plain). Journal of Agricultural Science
(University of Tabriz), 17(2), 1-14.
Howe, C. W. (1982). The Impact of Price on Residential Water Demand: Some New
Insights. Water Resources Research, 18(4), 713-716.
Hsiao, C. (2003). Analysis of Panel Data Cambridge: Cambridge University Press,
Cambridge, the United Kingdom.
Huffaker, R. G., Whittlesey, N. K., Michelsen, A., Taylor, R. G., & McGuckin, T.
(1998). Evaluating the Effectiveness of Conservation Water-Pricing
Programs. Journal of Agriculture and Resource Economics, 23, 12-19.
Hussain, I., Nazir, A., Ahmad, A., & Jehangir, W. A. (2007). Impact of Irrigation
Infrastructure Development on Dynamics of Incomes and Poverty:
Econometric Evidence Using Panel Data from Pakistan: Japan Bank for
International Cooperation.
IranDaily. (2007). Fars Medicinal Herbs Face Extinction [Electronic Version] from
http://iran-daily.com/1385/2761/html/panorama.htm#s203427.
152
Johansson, R. C. (2000). Pricing Irrigation Water. A Iiterature Survey. World Bank
Policy Research Working Paper 2449.
Johansson, R. C. (2005). Micro and Macro - Level Approaches for Assessing the
Value of Irrigation Water. World Bank Policy Research Working Paper 3778.
Johansson, R. C., Tsur, Y., Roe, T. L., Doukkali, R., & Dinar, A. (2002). Pricing
irrigation water: a review of theory and practice. Water Policy, 4(2), 173-199.
Jorgenson, D. W. (2000). Econometrics: Econometric Modeling of Producer
Behavior (Vol. 1). London: Cambridge, Mass. : MIT Press.
Keramatzadeh, A., Chizari, A. H., & Mirzaei, A. (2006). Determining the economic
value of irrigation water through: Optimal cropping pattern for integrated
farm and horticulture Eqtesad - E Keshavarzi Va Towse'e, 14(2), 35-60 [in
Persion].
Keshavarz, A., Ashraft, S., Hydari, N., Pouran, M., & Farzaneh, E.-A. (2005). Water
Allocation and Pricing in Agriculture of Iran. Paper presented at the Water
Conservation, Reuse, and Recycling:Proceedings of an Iranian-American
Workshop, Iran.
Kollahi, R. (1991). Potable Water Demand Estimation of Shiraz Unpublished M.A,
Shiraz University, Shiraz [in Persion].
Krueger, A. O., Schiff, M., & Valdes, A. (1991). The Political Economy of
Agricultural Pricing Policy: Latin America (World Bank). Baltimore: The
Johns Hopkins University Press.
Laffont, J., & Tirole, J. (1993). A Theory of Incentives in Procurement and
Regulation. Cambridge, Massachussets: The MIT Press.
Latinopoulos, P., Tziakas, V., & Mallios, Z. (2004). Valuation of irrigation water by
the hedonic price method : A case study in Chalkidiki, Greece. Water, Air,
and Soil Pollution, 4(253-262).
Lewis, J. N. (1969). Criteria for Determination of Water Rates in West Pakistan.
. Lahore: Planning and Development Board.
Linaweaver, F. P., John, J. R., Geyer, J. C., & Wolff, J. B. (1967). A Study of
Residential Water use. Paper presented at the Baltimore, MD: Johns Hopkins
University for the Federal Housing Administration and the Department of
Housing and Urban Development.
Manning, R., & Gallagher, D. (1982). Optimal water pricing and storage: The effect
of discounting. Water Resources Research, 18(1), 65-70.
Mansouri, M., & Ghiasi, A. (2002). Estimation of irrigation water cost price at
reservoir dams, using engineering economic approach: A case study of
153
Bukan, Mahabad and Barun Reservoires. Eqtesad - E Keshavarzi Va
Towse'e, 10(37), 171-192 [in Persion].
Mas-Collel, A., M.D.Whinston, & Green., J. R. (1995). Microeconomic Theory. New
York: Oxford University Press.
MehrNews. (2007). Iran’s Share of Worldwide Medicinal Plant Trade Barely 2%
[Electronic Version] from
http://www.mehrnews.com/en/NewsDetail.aspx?NewsID=442498.
Mitra, A. K. (1997). Economic aspects of irrigation management in magor and
medium surface irrigation system in India. Artha Vijnana, xxxix(3).
Molle, F., & Berkoff, J. (2007). Irrigation Water Pricing:The Gap Between Theory
and Practice. London: Biddles Ltd, King’s Lynn.
Moncur, J. (1987). Urban Water Pricing and Drought Management. Water Resources
Research 23(3), 393-398.
Monteiro, H. (2005). Water Pricing Models: a survey. Lisbo, Portugal: Instituto
Superior de Ciencias do Trabalho e da Empresa.
Morgan, W. D. (1973). Residential Water Demand: The Case from Micro Data.
Water Resources Research, 9(4), 1065-1073.
Mousavi, S. F. (2005). Agricultural Drought Management in Iran. In Water
Conservation, Reuse, and Recycling: Proceedings of an Iranian-American
Workshop (pp. 106 -114): National Academies Press.
Nicholson, W. (2004). Microeconomc Theory :Basic Principles and Extensions
United State: South-Western College.
Nieswiadomy, L. M., & Molina, D. J. (1991). A note on Price Perception in Water
Demand Models. Land Economics, 67(3), 352-359.
North, R. M. (1967). Consumer Responses to Prices of Residential Water. Paper
presented at the The Third Annual American Water Resource Conferance,
American Water Resource ASSA; Urbana, ILL.
Penzhorn, N., & Marais, D. (1998). Deficit irrigation has financial benefits. Farmers
Weekly September 18, 26-28.
Postel, S. (1997). Last Oasis: Facing Water Scarcity. New York: Norton.
PressTV. (2008). Iran's Saffron Exports Exceed $14m [Electronic Version] from
http://www.presstv.ir/detail.aspx?id=38184&sectionid=351020102.
Rhodes, G. F., & Sampath, R. K. (1988). Efficiency, Equity and Cost Recovery
Implications of Water Pricing and Allocation Schemes in Developing
Countries. Canadian Journal of Agricultural Economics, 36, 103-117.
154
Riley, J., & Scherer, C. (1979). Optimal water pricing and storage with cyclical
supply and demand. Water Resources Research, 15(2), 233-239.
Riordan, C. (1971). General multistage marginal cost dynamic programming model
for the optimization of a class of investment-pricing decisions. Water
Resources Research, 7(2), 245-253.
Riordan, C. (1971). Multistage marginal cost model of investment-pricing decisions:
Application to urban water supply treatment facilities. Water Resources
Research, 7(3), 463-478.
Sadr, S. K., Khodarahmi, R., & Abdian, M. (1994). Water Demand Estimation of
Tehran City. Science and Technology Bulletin, 13 [in Persion].
Saeid nia, E. (1996). Estimation of Demand Function for Potable and its Pricing
Policy. Unpublished M. A, Tarbit Modares University, Tehran [in Persion].
Sahibzada, S. A. (2002). Efficient irrigation water development in Pakistan: Pricing
issues and options. Unpublished Ph.D., State University of New York at
Binghamton, United States - New York.
Salami, H. A., & Mohammad nejad, A. (2002). Economic value of irrigation water
using the flexible production functions (Saveh case study). Agricultural
Sciences and Technology, 16(2), 58-97 [in Persion].
Sampath, R. K. (1992). Issues in irrigation pricing in developing countries. World
Development, 20(7), 967-977.
Samuelson, P. A. (1976). Economics (10th ed.): McGraw-Hill Inc.,US.
Saunders, R. J. (1969). Forecasting Water Demand: An Inter- and Intra Community
Study. Business and Economics Studies, 11(2).
Schoengold, K., Sunding, D. L., & Moreno, G. (2006). Price elasticity reconsidered:
Panel estimation of an agricultural water demand function. Water Resources
Research, 42.
Schuck, E. C., & Green, G. P. (2002). Supply-based water pricing in a conjunctive
use system: implications for resource and energy use. Resource and Energy
Economics, 24(3), 175-192.
Seagraves, J. A., & Easter, K. W. (1983). Pricing irrigation water in developing
countries. Journal of the American Water Resources Association, 19(4), 663672.
Shiferaw, B., Reddy, V. R., & Wani, S. P. (2008). Watershed externalities, shifting
cropping patterns and groundwater depletion in Indian semi-arid villages: The
effect of alternative water pricing policies. Ecological Economics, 67(2), 327340.
155
Sloman, J. (2003). Economics (fifth ed.). London: Financial Times/Prentice Hall.
Small, L. E., & Carruthers, I. D. (1991). Farmer Financed Irrigation:The Economics
of Reform. Cambridge: Cambridge University Press.
Small, L. E., & Rimal, A. (1996). Effects of alternative water distribution rules on
irrigation system performance: A simulation analysis,. Irrigation & Drainage
Systems, 10, 25-45.
Smith, R. B. W., & Tsur, Y. (1997). Asymmetric Information and the Pricing of
Natural Resources: The Case of Unmetered Water. Paper presented at the
Fifth Joint Conference on Agriculture, Food, and the Environment, June 1718, 1996, Padova, Italy.
Spulber, N., & Sabbaghi, A. (1994). Economics of Water Resources. Norwell,
Massachusetts Kluwer Academic Publishers.
Sunding, D., Zilberman, D., Howitt, R., Dinar, A., & MacDougall, N. (1994).
Modeling the Impacts of Reducing Agricultural Water Supplies: Lessons from
California’s Bay/Delta Problem. Paper presented at the Conference Name|.
Retrieved Access Date from URL.
Sunding, D. L. (2005). The Economics of Agricultural Water Use and the Role of
Prices. Paper presented at the Water Conservation, Reuse, and
Recycling:Proceedings of an Iranian - American workshop, Washington,.
D.C.
Tijani, A. A., & Osotimehin, K. O. (2007). Economics of pesticides use among
maize farmers in Edo State, Nigeria. Research Journal of Agricultural
Biological Sciences, 3(3), 129-132.
Torell, L. A., Libben, J. D., & Miller, M. D. (1990). The market value of water in the
Ogallala aquifer. Land Economics, 66(2), 163-175.
Tsur, Y., Dinar, A., Doukkali, R. M., & Roe, T. L. (2004). Irrigation water pricing:
Policy implications based on international comparison. Environment and
Development Economics, 9(06), 735-755.
Williams, M., & Suh, B. (1986). The demand for urban water by customer class.
Applied Economics, 18(12), 1275-1289.
Wong, S. T. (1972). A Model on Municipal Water Demand: A Case Study of
Northeastern Illinois Land Economics, 48(1), 34-44.
Yaffee, R. (2003). A Primer for Panel Data Analysis: New York University.
Information Technology Services.
Zare, M. R. (2006). Valuation of groundwater for agricultural uses A Case study of
Kerman, Iran. Tarbiat Modares University, Tehran.
156
Zarnikau, J. (1994). Spot market pricing of water resources and efficient means of
rationing water during scarcity (water pricing). Resource and Energy
Economics, 16(3), 189-210.
157
APPENDICES
Appendix A: Water in World and Iran
Table A.1. Population size and growth and renewable freshwater availability in
water –short countries, (1995 and 2025)
Country
Population Water per Population Water per
2025
capita
1995
capita
(million)
2025
(million)
1995
(m3 per
(m3 per
year)
year)
28.1
527
47.3
313
0.6
161
0.9
104
0.3
192
0.3
169
6.1
594
12.3
292
0.4
777
0.7
442
Algeria
Bahrain
Barbados
Burundi
Cape
Verde
Comoros
0.6
1,667
1.3
Cyprus
0.7
1,208
1.0
Egypt
62.1
936
95.8
Ethiopia
56.4
1,950
136.3
Haiti
7.1
1,544
12.5
Iran
68.4
1,719
128.3
Jordan
5.4
318
11.9
Kenya
27.2
1,112
50.2
Kuwait
1.7
95
2.9
Libya
5.4
111
12.9
Malawi
9.7
1,933
20.4
Malta
0.4
82
0.4
Morocco
26.5
1.131
39.9
Oman
2.2
874
6.5
Qatar
0.5
91
0.8
Rwanda
5.2
1,215
13.0
Saudi
18.3
249
42.4
Arabia
Singapore
3.3
180
4.2
Somalia
9.5
1,422
23.7
South
41.5
1,206
71.6
Africa
Tunisia
9.0
434
13.5
United
2.2
902
3.3
Arab
Emirates
Yemen
15.0
349
39.6
Source: (Hinrichsen, Robey, & Upadhyay, 1997)
158
Total
fertility
rate
%
Growth
rate 1998
4.4
3.2
1.7
6.6
5.3
2.4
2.0
0.5
2.5
2.9
760
947
607
807
879
916
144
602
55
47
917
71
751
295
64
485
107
5.1
2.1
3.6
7.0
4.8
3.0
4.4
4.5
3.2
6.3
5.9
2.1
3.3
7.1
4.1
6.0
6.4
2.7
0.7
2.2
2.5
2.1
1.8
2.5
2.0
2.3
3.7
1.7
0.6
1.8
3.9
1.7
2.1
1.1
142
570
698
1.7
7.0
3.3
1.1
3.2
1.6
288
604
3.2
4.9
1.9
2.2
131
7.3
3.3
Table A.2. Baseline (year 2000) and Projected (year 2021) Characteristics of
Water Resources Management in Iran
Indicator
Unit
2000
Total volume of exploited water
Share of water resources, by source
Groundwater
Surface water
Recycled (domestic, industrial)
Share of consumption, by sector
Agriculture and aquaculture
Urban & Rural
Industry & Mine
Water loss, by sector
Agriculture
Urban
Volume of return flow
Effluents and Wastewater
Urban
Industria
Investment
Total gross investment
Contribution of private sector
Importance in national economy
(NE)
Contribution of water investment
from GDP
Contribution of water value and
related services in NE
Contribution of capital return of
expenses of governmental projects
Urban water
Agriculture water
Economic revenue of water in
different sectors (average)
Economic revenue of water in
farming subsector
Productivity of agricultural water
Role of water in the production of
cereals
Role of water in the production of
other yields
Source : Ardakanian (2005) page 29
bcm
97
%
%
%
52
48
-
42
55
3
81
114
-
-19
+15
-
%
%
%
94
6
1.2
86
7
3
92
117
250
-9
+17
+150
%
%
bcm/yr
bcm/yr
bcm/yr
bcm/yr
64
27
29
4.5
3.7
0.8
60
10
40
8
5.5
2.5
94
37
138
178
149
312
-6
-63
+38
+78
+49
+213
1012 Rls
1012 Rls
41
40
262
32
635
80
+539
-20
%
1.2
2.6
217
+117
%
7.5
9.8
131
+31
%
%
Rls/cm
22
6
1614
50
23
5018
227
383
311
+127
+287
+211
kg/ m3
%
0.6
69
1.1
73
183
106
+83
+6
%
90100
90-100
-
-
159
2021
Ratio Percent
(selected (%) change
scenario)
120
124
+24
Appendix B: Characteristics of Iran Provinces and their Agricultural Products
Table B.1. Characteristics of Iran Provinces
Province
Capital
Qom
Qom
Hamadan
Hamadan
Golestan
Gorgan
Azarbaijan, West
Urmia
Kermanshah
Kermanshah
Azarbaijan, East
Tabriz
Qazvin
Qazvin
Ardabil
Ardabil
Yazd
Yazd
Khorasan, South
Birjand
Mazandaran
Sari
Khuzestan
Ahvaz
Tehran
Tehran
Lorestan
Khorramabad
Semnan
Semnan
Kurdistan
Sanandaj
Chahar Mahaal &
Shahrekord
Bakhtiari
Markazi
Arak
Yasuj
Kohgiluyeh &
Boyer-Ahmad
Zanjan
Zanjan
Esfahan
Esfahan
Khorasan, Razavi
Mashhad
Bushehr
Bushehr
Fars
Shiraz
Khorasan, North
Bojnourd
Ilam
Ilam
Hormozgan
Bandar Abbas
Gilan
Rasht
Kerman
Kerman
Sistan &Baluchistan Zahedan
Source: Statistical Centre of Iran
64,055
18,814
28,294
97,491
29,137
16,332
Annual
Precipitation
(mm) /2006
111.1
283.1
522.3
372.2
430
128.9
325.1
237.4
43.8
134.8
683.9
184.3
226.5
510.1
176.8
449.4
413
1,046,737
1,703,267
1,617,087
2,873,459
1,879,385
3,603,456
1.143,200
1,228,155
990,818
636,420
2,922,432
4,274,979
13,422,366
1,716,527
589,742
1,440,156
857,910
29,130
15,504
283.4
776.7
1,351,257
634,299
21,773
107,029
144,681
22,743
122,608
28,434
20,133
70,669
14,042
180,836
181,785
307.8
219.7
223.3
223.9
134.8
240.8
554.6
276.6
1475.8
134.3
54.6
964,601
4,559,256
5,593,079
886,267
4,336,878
811,572
545,787
1,403,674
2,404,861
2,652,413
2,405,742
Area (km²)
11,526
19,368
20,195
37,437
24,998
45,650
15,549
17,800
129,285
69,555
160
Population
(person)
Table B.2. Agricultural lands area on holdings with irrigated and rain fed
cropland 2003
Province
East Azarbayejan
West Azarbayejan
Ardebil
Esfahan
Ilam
Bushehr
Tehran
Chaharmahal & Bak.
South Khorasan
Khorasan-e-Razavi
North Khorasan
Khuzestan
Zanjan
Semnan
Total (ha)
1,319,713
887,187
731,167
423,858
322,655
348,641
221,256
194,813
149,653
2,225,941
473,669
1,266,123
737,997
153,826
Sistan & Baluchi...
245,328
Fars
1,254,511
Qazvin
413,895
Qom
80,662
Kordestan
940,609
Kerman
667,633
Kermanshah
751,012
Kohgiluyeh & Boye..
157,251
Golestan
538,967
Gilan
253,403
Lorestan
768,924
Mazandaran
360,657
Markazi
670,852
Hormozgan
129,147
Hamedan
844,580
Yazd
131,265
Total country
17,665,198
Source: Statistical Centre of Iran
Irrigated (ha)
378,825
379,541
221,725
372,699
76,188
70,151
211,912
107,202
78,756
1,138,514
197,095
690,764
155,194
131,298
Rain fed (ha)
225,104
890,838
249,657
78,520
127,852
662,305
161,132
37,988
188,484
172,851
199,113
231,175
317,791
118,028
296,090
130,238
8,297,031
20225
363672
164239
2143
812758
5327
589879
119263
350483
80551
569811
129483
353061
11120
548490
1027
9,368,167
161
940888
507647
509442
51159
246467
278489
9345
87611
70898
1087427
276573
575359
582803
22528
Table B.3. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Wheat Holdings by Province in 2006 Year
Produced
Provinces
Irrigated
Area
(ha)*
Irrigated Irrigated
Average
Crop Water
Production
Application Productivity
Yield
( Metric
of Water
(Kg/ha)*
(Kg/ m3)**
tons )*
(m3/ha)**
Qom
11894
52158.33 4385.26
9344.49
0.47
Hamadan
104719
381317
3641.34
7863.73
0.46
Golestan
156335
491029.8 3140.88
6088.48
0.52
West Azarba
115912
372171.1 3210.81
6563.14
0.49
Kermanshah
96886.5 517566.9 5341.99
6547.97
0.82
East Azarba
101809
363356.6
3569
6398
0.56
Qazvin
74667
344936.2 4619.66
7341.02
0.63
Ardabil
76093
305414.6
4013.7
5925
0.68
Yazd
26138
89202.92 3412.77
10886.53
0.31
Mazandaran
3264
9336.03
2860.3
3473.66
0.82
Khuzestan
370299
1260262
3404.01
6921.26
0.49
Tehran
69217.5 370931.2 5358.92
7908.1
0.68
Lorestan
102322
303611.4 2967.22
8450.14
0.35
Semnan
33853
134622
3976.66
7924.87
0.50
Kurdistan
38554
161441.9 4187.42
6970.90
0.60
Chaharl &Ba
33156
129458.9 3904.54
7743.66
0.50
Markazi
73324
317229.1
4326.4
8571.31
0.50
Kohgil& Boy
30426
93987.78 3089.06
6291.76
0.49
Zanjan
24348
94406.95
3877.4
8430.22
0.46
Esfahan
113088
572628.3 5063.56
8805
0.57
Bushehr
20437
55303
2706.02
7538.74
0.36
Fars
457695
2044409
4466.75
8659.28
0.52
Khorasan,
360945
1033285
2862.72
8789.19
0.33
Ilam
40618
154869.1 3812.82
6217
0.61
Hormozgan
13560
56661.4
4178.57
7421
0.56
Gilan
94
161.38
1716.84
3475
0.49
Kerman
104443.6 328862.9 3148.71
9420
0.33
Sistan &Ba.
52968
99149.27 1871.87
9657.58
0.19
Country
2706996 10137770 3745.03
7241
0.52
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
162
Table B.4. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Barley Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)*
( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
21611.9 63147.59 2921.89 6398.46
0.46
West Azarbaijan
14764 41187.59 2789.73 6563.14
0.42
Ardabil
22232 58602.24 2635.94 5924.93
0.44
Esfahan
48636.2 226054.5 4647.86 8805.34
0.53
Tehran
36179.5 144711.9 3999.83 7908.08
0.51
CharMahal & Bakhtiari
6512 20682.02 3175.99 7743.66
0.41
khorasan
171364 480079.9 2801
8789.19
0.35
Khozestan
25675 44628.62 1738.21 6921.26
0.25
Zanjan
5015 16999.44 3389.72 8430.22
0.40
Semnan
13460 46898.02 3484.25 7924.86
0.44
Sistan & Baloshesta
17409.5 23872.99 1371.26 9657.58
0.14
Fars
34716 109853.8 3164.36 8659.28
0.36
Ghazvin
27445 86059.16 3135.7 7341.02
0.43
Ghom
27374 102761.2 3753.97 9344.49
0.40
Kordestan
4905 15631.37 3186.82 6790.90
0.47
Kerman
18829.6 46306.09 2459.22 9419.84
0.26
Kermanshah
14118 74065.59 5246.18 6547.97
0.80
Kohkiloyeh & Boyrahmad 4013.5 11598.27 2889.81 6291.76
0.46
Golestan
8353 25544.77 3058.15 6088.48
0.50
Lorestan
8661
17737 2047.92 8450.14
0.24
Mazandaran
2328
3751.34 1611.4 3474.66
0.46
Markazi
36883 133180.9 3610.9 8571.31
0.42
Hormozgan
1315
2634.66 2003.54 7421.04
0.27
Hamedan
34204 132394.2 3870.72 7863.73
0.49
jiroft & kahnoj
9213 19694.01 2137.63 8019.84
0.27
Yazd
6277 17639.21 2810.13 10886.53
0.26
country
624491.2 1972399 3158.41
7417
0.42
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
163
Table B.5. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Lentil Holdings by Province in 2006 Year
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
Esfahan
1428
2950.6 2066.24 5063.04
0.41
CharMahal & Bakhtiari
631.2
901.3 1427.91 4100.8
0.35
khorasan
1234
1211.1 2792.19 4617.06
0.21
Fars
6337 8145.14 1285.33
7272
0.18
Lorestan
932
834.89 895.8
7120
0.13
Mazandaran
139
102.06 734.24 3785.6
0.19
Yazd
94
99.27 1056.07 9489.56
0.11
Kerman
285.5
351.6 1231.53 8459.67
0.14
Kohkiloyeh & Boyrahmad 121
134.05 1107.82 3131.2
0.35
Country
13377.5 16663.13 1245.61 5893.21
0.23
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
Produced Provinces
164
Table B.6. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Pea Holdings by Province in 2006 Year
Produced
Provinces
Irrigated Irrigated Average
Irrigated
Crop Water
*
Area (ha) Production ( Yield Application Productivity
Metric tons )* (Kg/ha)* of Water
(Kg/ m3)**
3
**
(m /ha)
West Azarbaijan
2912
1913.99
919.3
3788.8
0.24
Esfahan
901.5
1628.87
1623.19
5463.04
0.30
jiroft & kahnoj
527
450.37
886.55
3467.67
0.26
Khorasan
2149.5
2218.46
1020.96
7353.06
0.14
Fars
2842
3474.4
1441.06
5896
0.24
Lorestan
1270
743.79
764.43
2160
0.35
Markazi
768
1389.12
1438.01
4875.2
0.29
Hamedan
264.5
675.21
1192.94
2811.2
0.42
Yazd
391.1
110
940.18
7121.56
0.136
Kerman
906
1022.48
1234.13
3467.67
0.36
Country
15459.6
16159.25 1175.82 102027.46
0.25
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
165
Table B.7. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Pinto Bean Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
4510
8335.7 1848.27 8778.01
0.21
Esfahan
4273 10870.71 2544.05 6919.04
0.37
Tehran
115
213.65 1857.82 8151.14
0.23
CharMahal & Bakhtiari
3972.3 10159.77 2557.65 6408.75
0.40
Fars
14994 45849.12 3057.83
8024
0.38
Geilan
944.5 1772.18 1876.31 6281.80
0.30
Kohkiloyeh & Boyrahmad 1505 5059.86 3362.04 4283.2
0.78
Lorestan
12080 22972.66 1901.71
7120
0.27
Mazandaran
486.5
695.14 1428.85 3785.6
0.38
Markazi
16019.5 33922.91 2117.6 7803.2
0.27
Country
92980.8 202377.26 2176.55
7488
0.39
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
166
Table B.8. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Onion Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)*
( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
9149.5 404497.22 44209.76 13850.01
3.19
West Azarbaijan
1396 59103.73 42337.92 10364.8
4.08
Esfahan
5618 298106.09 53062.67 12599.04
4.21
Sistan & Baloshesta
5152 131796.29 25581.58 12797.44
1.20
Fars
4299 202713.19 47153.57 12072
3.91
Kordestan
577
13690.81 23727.57 11761.6
2.02
Kerman
281.5 8988.69 31931.42 4635.67
6.89
Kohkiloyeh & Boyrahmad 258
7055.7 27347.69 9707.2
2.82
Markazi
659.8 29638.53 44920.47 12107.2
3.71
Hormozgan
9940 165442.15 16644.08 6724.8
2.47
Yazd
445
22483.13 50523.88 18657.56
2.71
Boushehr
825
12672.4 15360.48 6340.51
2.42
Tehran
449
23837.65 53090.53 11479.14
4.62
Khozestan
3631 122034.03 33608.93 6658.71
5.05
Zanjan
3350 105961.87 31630.41 10147.2
3.12
Country
48758.3 1670367.4134258.11 10660
3.55
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
167
Table B.9. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Tomato Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
7151
West Azarbaijan
Ardabil
Esfahan
Ilam
Boshehr
Tehran
jiroft & kahnoj
Khorasan
Khozestan
Zanjan
Semnan
5211.5
1690
2185.5
663
14810
5453
16478
16845
11185
3944
1798
315427.6 44109.6 10074.01
4.38
189140.5
63673.81
78937.39
11967.94
449676.7
226689
487313.6
572013
360901
93853.85
58442.14
4.41
4.1
3.32
1.63
4.21
3.62
2.89
6.33
3.57
1.64
3.00
36292.9
37676.8
36118.7
18051.2
30363
41571.4
29573.6
33956
32266.5
23796.6
32504
8220.8
9165.93
10887.04
11048
7204.51
11479.14
10217.06
5362.71
9027.2
14498.02
10829.44
Sistan & Baloshesta
1361.2 22410.37 16463.7 12968
1.27
Fars
15702.5 798725 50866.1
9992
5.09
Ghazvin
7469 304139.3 40720.2 11479.14
3.55
Golestan
9562 263550.3 27562.3 7612.8
3.62
Lorestan
1856 39493.62 21278.9 11136
1.91
Mazandaran
4221.5 207395.8 49128.5 5817.6
8.44
Markazi
947
31929.5 33716.5 9547.2
3.53
Hormozgan
9597 221392.5 23068.9 8916.8
2.59
Hamedan
3662 141349.4 38599
8955.2
4.31
Yazd
506.5 15339.72 30285.7 11361.56
2.67
Kordestan
1114 18520.02 16624.8 9569.6
1.74
Kerman
638.5 10148.01 15893.5 12363.67
1.28
Kermanshah
2435 60660.02 24911.7
9888
2.52
Kohkiloyeh & Boyrahmad 254
8796.59 34632.2 11211.2
3.09
Country
146837 5054830 30554.9
9023
3.18
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
168
Table B.10. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Potato Holdings by Province in 2006 Year
Produced Provinces Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
9513
West Azarbaijan
1359
Ardabil
25738
Esfahan
21019.5
Tehran
1125
jiroft & kahnoj
8900
CharMahal & Bakhtiari 3894.4
khorasan
7059.5
Khozestan
4612
Zanjan
7133
Semnan
5303
Fars
6427
Ghazvin
649
Golestan
6448
Gilan
77
292057.5 30700.9
26818.93
682763.2
513592.9
22697.61
184304.9
129998.1
179003.4
98902.78
171653.5
118268.8
128200.7
13411.06
109184.9
1279.26
8746.01
3.51
19734.3 7324.8
26527.4 6413.93
24434.1 9495.04
20175.7 9463.14
20708.4 5819.67
33380.8 10084.8
65112.7 11244.26
21444.7 5842.71
24064.7 9347.2
22302.3 12450.02
19947.2 10264
20664.2 10024
16905.2 6636.8
16613.7 6025.6
2.69
4.14
2.57
2.13
3.56
3.31
2.25
3.67
2.57
1.79
1.949
2.06
2.55
2.768
Lorestan
3201 67249.44 21008.9
7968
2.64
Mazandaran
680
8234.99 12110.3 6025.6
2.01
Markazi
5136 134358.4 26160.1 8811.2
2.97
Hamedan
24162 875128.6 36219.2 9003.2
4.024
Yazd
115
2426.85 21103 9873.56
2.14
Kordestan
9573 274583.3 28683.1 10257.6
2.8
Kerman
5409 118749.8 21954.1 5819.67
3.77
Kermanshah
1284 18391.38 14323.5
9952
1.44
Country
159875 4188207 26196.8
8561
3.06
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
169
Table B.11. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Cucumber Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
2081 39236.4 18854.6 7178.0096
2.63
West Azarbaijan
1068 15895.7 14883.6 6252.8
2.38
Esfahan
2039.5 59345.5 29098.1 6919.04
4.20
Ilam
6195
118327 19100.5
6472
2.95
Tehran
1495 30421.3 20348.7 8151.1424
2.50
khorasan
3617 58540.5 16219.3 7801.0624
2.07
Khozestan
8166
184690 22617 8098.7136
2.79
Fars
6315
147803 23405.1
8024
2.92
Qazvin
594
11136 18747.5
7448
2.52
Kordestan
1916 37204.6 19417.8 6062.4
3.20
Kohkiloyeh & Boyrahmad 1465 48237.6 32926.7 4283.2
7.69
Golestan
1897
42388 22344.8 5804.8
3.85
Lorestan
8662
184874 21343.1
7120
2.30
Markazi
95
2275.9 23956.9 7803.2
3.07
Hormozgan
7519
131922 17545.1 7572.8
2.32
Hamedan
3379
108084 31987.1 5067.2
6.31
Yazd
493
12549.1 25454.6 8257.5616
3.08
jiroft & kahnoj
20373 608940 29889.5 8811.6736
3.39
Country
81562.2 1933975 23712
7190
3.29
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
170
Table B.12. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Watermelon Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average Crop Water
Area Production Yield Application Productivity
(ha)* ( Metric (Kg/ha)* of Water (Kg/ m3)**
tons )*
(m3/ha)**
East Azarbaijan
609
14009.3 27523.2 7794.01
3.53
West Azarbaijan
807
42071.7 30398.6 5564.8
5.46
Esfahan
1303.6 28121.9 27489.7 7639.04
3.60
Ilam
3423
86988 30651.2
6360
4.82
Boshehr
2514 50467.5 21923.3 9956.5
2.20
Tehran
645
11984.1 44385.6 9463.1
4.69
jiroft & kahnoj
14407 502702 29603.8 5819. 7
5.09
Khorasan
15066.7 327326 22920.6 7593.1
3.02
Khozestan
22069 680394 31593.4 8146.7
3.88
Semnan
3150 43107.5 27283.2
6906
3.95
Sistan & Baloshesta
14931.5 138506 21233.5 10237.4
2.07
Fars
9195.5 99726.7 24144
11976
2.02
Gilan
441
40602.7 25517
6636.8
3.85
Golestan
660.5 24521.9 23784.6 6636.8
3.58
Markazi
559
16007.5 28231.8 8811.2
3.20
Hormozgan
6342
111206 24861.7 4532.8
5.48
Hamedan
3194 85803.6 36527.7
8392
4.35
Yazd
1133.9 29020.4 25640.9 9873. 6
2.60
Kordestan
913
19955.2 22074.3
8392
2.63
Kerman
8327.5 214101 37138.2 5819. 7
6.38
Kohkiloyeh & Boyrahmad 945
29010.5 34743.2 7883.2
4.41
Country
116338 2719320 28409.8
5887
3.77
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
171
Table B.13.Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Cotton Holdings by Province in 2006 Year
Produced
Provinces
East Azarbaijan
Ardabil
Esfahan
Tehran
khorasan
Irrigated
Irrigated
Area Production (
(ha)* Metric tons )*
2069
4957
4904
1969
63722
6866.08
15147.34
14898.1
6007.09
35470.6
Irrigated
Yield
(Kg/ha)*
3318.55
3055.75
3037.95
3050.83
2321.14
Average
Crop Water
Application of Productivity
Water
(Kg/ m3)**
3
**
(m /ha)
73887.58
0.05
73887.58
0.04
136958.40
0.02
107729.76
0.03
110436.21
0.02
Semnan
5261
12850.06
2442.51
123232.52
0.02
Fars
10692
29737.8
2781.31
115834.54
0.02
Ghom
3331
7713.11
2315.55
127817.06
0.02
Golestan
10515
21157.19
2012.1
17014.19
0.12
Mazandaran
151
286.29
1895.93
17014.19
0.11
Markazi
1530.5
2545.12
1662.93
92387.46
0.02
Yazd
738.5
1847.01
2501.02
187972.22
0.01
Kerman
2694
5732.8
2127.99
151930
0.01
country
113345 279337.56
2542.21
100884.04
0.02
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
172
Table B.14. Irrigated Area, Irrigated Production, Yield, Water Price Average,
Average Application of Water, and Crop Water Productivity of
Sugar Beet Holdings by Province in 2006 Year
Produced Provinces
Irrigated Irrigated Irrigated Average
Crop Water
Area Production Yield Application Productivity
(ha)*
( Metric (Kg/ha)* of Water
(Kg/ m3)**
*
3
**
tons )
(m /ha)
West Azarbaijan
340
11816
34753.1 10151.5
3.42
Esfahan
6976
238081 34128.5
13519
2.52
CharMahal & Bakhtiari
3490.7
108693 31137.8 11316.8
2.75
khorasan
63279.1 668825 28347.6 14778.7
1.92
Semnan
3387
111920 33043.9
12770
2.59
Fars
22310
701818 31457.6 16291.2
1.93
Ghazvin
3647.5
132876 36429.5 12647.1
2.88
Lorestan
8444
266935 31612.4
14224
2.22
Markazi
2666
81729.5 30656.2 13627.2
2.25
Hamedan
8773
318067 36255.2 12747.2
2.84
Yazd
265
7825
29528.3 21105.6
1.40
Kerman
3097.5 82848.7 26746.9 13995.7
1.91
Kermanshah
15535 1667956 49631.2
15632
3.17
Country
185888 6709112 36092.3
14664
2.46
Source: *Ministry of Keshavarzi Jehad, 2006. ** Based on research findings
173
Appendix C: A Review on Studies Carried Out
Table C.1. A review on partial equilibrium analyses
First-best
Hearne and
Easter (1998)
Thobani
(1998)
Water markets
Marginal cost
pricing
Easter et al.
(1998)
Water markets
Tsur and
Dinar (1997)
Marginal cost
pricing
It has long been recognized that markets provide a
means to efficiently allocate water
Marginal cost pricing has also been called
opportunity cost pricing, implying that the price of
water should be set equal to the opportunity cost of
providing it.
Such things as monitoring, return flows, third-party
effects, and instream uses have to be considered,
when deciding what to include in water
transactions.
When water supplied is of different quality the
marginal value of supply should be reflected in the
price.
Second-best
Easter (1999)
Externalities
Summarizes water conditions, irrigation systems,
and their potential externalities.
Willis et al.
(1998)
Externalities
Third-party effects of return-flow from large
irrigation dam projects recently have accounted for
environmental degradation in Colorado.
Smith and
Tsur (1997)
Asymmetric
information
Use mechanism design theory to propose a waterpricing scheme, which depends only on observable
outputs.
Easter et al.
(1997)
Public goods
It is useful to categorize irrigation service based on
their public good nature, depending upon the
evolution of technology or institutions.
Tsur and
Dinar (1997)
Transaction
costs
Effects of implementation costs on the performance
of different pricing methods are significant in the
sense that small changes in costs can change the
order of optimality of those methods.
Zilberman
(1997)
Scarcity
Develops an optimal water pricing, allocation, and
conveyance system over space to capture different
upstream and downstream incentives.
Shah et al.
(1995)
Scarcity
Find that it may be optimal to increase water prices
to encourage more quickly the adoption of water
conserving technologies used with groundwater.
174
MacDonnell
et al. (1994)
Externalities
Discuss the third-party effects of American West
dams and water banking
Seagraves
and Easter
(1983)
Equity
Equity concerns include such things as the recovery
of costs from users, subsidized food production,
and income redistribution.
Saliba and
Bush (1987)
Equity
Note that higher costs associated with the purchase
of water rights may force some users out of the
market.
Sampath
(1991)
Equity
Argue that consumers benefit from agricultural
investments through lower food prices and so
should be expected to share in covering the costs.
Sampath
(1992)
Equity
Notes equity concerns surrounding income
redistribution via irrigation distribution have
become one of the most important objectives across
disciplines.
Easter (1993)
Equity
Illustrates the effect of ‘‘fairness’’ on efficient
management of four irrigation systems.
Equity
Concerns
Tsur and
Equity
Equity effects of pricing are primarily dependent on
Dinar (1997)
land endowments.
Source: R.C. Johansson et al. (2002) 178
175
Table C.2. A review on general equilibrium analyses
First-best
Hurwicz (1998)
Binswanger,
Deininger, and Feder
(1993)
Berk, Robinson, and
Goldman (1991)
Second-best
Kohn (1998)
Roe and Diao (1997)
Smith and Roumasset
(1998)
Diao and Roe (1995)
Vaux and Howitt
(1984)
Elbasha and Roe
(1995)
Mohtadi (1996)
Rausser and Zusman
(1998)
Schaible (1997)
Equity Concerns
Diao and Roe (2000)
Derives the optimality conditions for GE treatments of
market failure and second-best policies
Discuss GE assumption in first-best and second-best
analysis
Compare the advantages and disadvantages GE and partial
equilibrium analyses
Externalities Illustrates a simple Nash-game scenario that
both countries will opt for environmental
taxes.
Externalities Describes a situation found where two
countries share water resources and thus the
water-use decisions of each country will
affect the water availability of the other
country.
Trade
Provide a model for water management with
multiple sources and transport technologies.
Trade
Focus on the environmental and health
effects of changing trading patterns.
Trade
Examine the interregional equilibrium supply
and demand relationship for California.
Endogenous Incorporate pollution and abatement efforts
growth
into three types of endogenous growth
models.
Endogenous Show how optimal growth depends upon the
growth
type and extent of environmental regulation
Scarcity
Explore the affects of water scarcity on the
political power balance in a GE format.
Scarcity
Examines groundwater demand responses to
conservation pricing policies.
Equity
Water pricing may have a role in policies
aimed at affecting income distribution
between farming and non-farming sectors.
Just, Netanyahu, and
Equity
Examine the equity considerations of water
Horowitz (1997)
pricing.
Carruthers, Rosegrant, Equity
Generate various scenarios regarding equity
and Seckler (1997)
concerns as a function of global food supply
and demand linked by trade in a GE
framework.
Rosegrant (1997)
Equity
The effects on food security of changing
investment levels can be evaluated for a
variety of regions and periods.
Source: R.C. Johansson et al. (2002) P179
176
Table C.3. Econometric Studies of Water Values
Author
Data
Place
Kim and
Schaible
2000
Nebraska
Moore,
Gollehon,
and Hellerstein
2000
Pacific
Northwest,
U.S.
Estimation
Method
Linear and
nonlinear
estimation
of crop
water functions
Tobit
regression of
censored
water price
experiments
Pazvakawambwa 2000
and van der Zaag
Nyanyadzi, OLS
Zimbabwe estimation of
crop-water
function
Droogers and
Allen
2002
World
Water and
Climate
Atlas
Functional
forms for
evapotranspiration
Quba’a, El
Fadel, and
Darwish
2002
TyreQasmieh
region of
South
Lebanon
Sahibzada
2002
Pakistan
Linear
programming
of multiple
crops and
municipal
water
consumption.
OLS estimated
Cobb-Douglas
177
Description of Results
Estimating derived
demand using applied
water rather than
consumptive water may
bias results upwards,
which may lead to under
investment in improved
irrigation technology
Producer surplus
response to price
changes is essentially
inelastic due to the
substitution opportunities
underlying the multioutput
production model.
Findings indicate that the
marginal value of water
(rainfall and irrigation)
given a maize
price of $0.10/kg is
$0.15/m3.
Results indicate that the
PM is superior when
predicting
evapotranspiration, but
the MG method is
preferred under uncertain
data conditions, such as
that one might encounter
in many countries
Currently, water is underpriced and overly
subsidized. Optimal
pricing may lead to
decreased agricultural
production and increased
tourism-based enterprise.
Elasticity found to be
approximately –0.50,
with a derived value of
water of Rs 415 – 445
per acre inch.
Table C.3. Econometric Studies of Water Values (Cont.)
Author
Data
Place
Gopalakrishnan
and Cox
2003
Schaible and
Aillery
2003
Oahu,
Hawaii
tourism
industry
Pacific
Northwest
and MidPlain
States
Schuck and
Green
2003
Ranganathan and 2004
Palanisami
Qiuqiong Huang
et al.
2008
Estimation
Method
OLS reduced
form equation
for water
demand.
Estimation of
irrigation
technology
transitions
Description of Results
Estimates provided for
derived demand as a
function of water price
Water price elasticities
of technology adoption
are inelastic for the
Pacific Northwest and
less so for the Mid-Plain
States.
Arvin
Logit model of Surface water prices
Edison
groundwater
range from $50.35/af to
Water
adoption
$87.73/af; groundwater
Storage
costs range from $62.85/
District,
af to $112.31/af. The
switching point between
the two is found to
occur when surface
water prices are 62
percent of groundwater
costs.
Srivilliputh Quadratic crop Water VMPs range from
ur Big
water response Rs. 146.60 (maize) to
Tank:
functions and
Rs. 385.64 (cotton).
Tamilnadu, exponential
Demand estimated to
India
water demand. be:
(- 0.00001742w)
y = 389 e
Rural
classical
Results indicate that
China
econometric
there is a large gap
methods and
between the value of
GME estimator water and the current
water cost in many
places.
178
Table C.4. Mathematical Programming Studies of Water Values
Author
Date
Place
Estimation
Method
Multiproduct,
restricted
equilibrium
model in a
mathematical
programming
approach
Three-stage
procedure
combining
mathematical
programming,
cropgrowth
model, and
econometric
estimation of
generated data.
Stochastic
dynamic
mathematical
programming
Schaible
2000
Pacific
Northwest,
U.S.
Bontemps,
Couture,
and Favard
2002
Southwest
France
Carey and
Zilberman
2002
Westlands
Water
Distict, CA
Ray
2002
Maharashtra, Linear
India
programming
Draper et
al.
2003
California
Network flow
optimization
179
Description of Results
Producer willingness-toaccept is lowest for
regulatory policy
($4/af - $18/af) and highest
for conservation-incentive
policies ($67/af - $208/af).
Elasticities ranging between
-0.31 for a feedback model
and -0.34 for an open-loop
model were obtained.
Demand is elastic when
price of water > 0.30 F/m3
in a wet year and > 1.60
F/m3 in a dry year.
Examines dynamic
adoption of modern
technologies when a water
market exists with
uncertain prices. Water
purchases range from
$44/af in 1988 to $115/af in
1995.
Because irrigation water
prices were significantly
below the scarcity value of
water, the potential for a
system of tradable water
rights seemed high. Hurdles
include: raising prices to
their opportunity value,
allocation system
inefficiencies, and crop
prices are set inefficiently
The largest shadow value
for water in 2020 was for
urban users in the Castaic
Lake region -- $8/m3,
which was reduced to
$0.50/m3. Marginal
willingness to pay measures
approached $200/ m3.
Table C.4. Mathematical Programming Studies of Water Values(Cont.)
Author
Date Place
Estimation
Description of Results
Method
Schaible
2000 Pacific
Multiproduct,
Producer willingness-toNorthwest,
restricted
accept is lowest for
U.S.
equilibrium
regulatory policy
model in a
($4/af - $18/af) and highest
mathematical
for conservation-incentive
programming
policies ($67/af - $208/af).
approach
Alverez
2004 Castilla – La Combines
Depth for maximum crop
et al.
Mancha,
irrigation
yield is lower than the
Spain
scheduling, crop irrigation depth for
growth,
maximum gross margin,
economic, and
which is lower than the
crop rotation
depth for maximum
modules using
economic efficiency.
nonlinear
programming.
The difference between a
Ghahraman 2004 Khorasan
Linear and
simple nonlinear model and
nonlinear
and
province,
programming of an integrated linear model
Iran
Sepaskhah
become more pronounced
crop-water
the greater the water
functions
constraint.
Rodríguez 2004 Duero
Linear
Producers are assumed to
and
Valley,
programming
maximize profit, minimize
Martínez
Spain
risk, and minimize labor
input. A simulated spot
market for water shows
prices ranging from 0.005
€/m3 to 0.29 €/m3
depending on scarcity
assumptions.
Water value of 0.035
Tsur et al.
2004 Case studies Linear
programming for Yuan/m3 for China;
for
R0.07/m3 for South Africa;
Morocco,
Morocco,
0.46 Dh/m3 – 3.0 Dh/m3
PMP for China,
China,
for Morocco; and TL12
Mexico, South
Mexico,
mil./ha – TL16mil./ha for
Africa,
South
Turkey were estimated
and Turkey
Africa,
and Turkey
Results show the usefulness
Multi-Attribute
2004 Area in the
Gómezof differential analysis in
Utility Theory
Duero
Limón and
evaluating the impact of a
(MAUT)
Valley in
Laura
water pricing policy.
mathematical
Spain
Riesg
programming
models
180
Appendix D: Irrigation Demand Function
The demand function of irrigation water is a function of output amount and inputs
price under cost minimization. Assuming that our objectives function as follows:
minimize
TC  Pw .W  Pl .L  Pf .F  Pp .P  Pr .R  Pm .M  Pa .Fa  PS .S
subject to:
Y  AW a1 La2 F a3 P a4 R a5 M a6 Fa a7 S a8
Where,
F = Fertiliser; L = Labour; M = Tractor and machinery services; Fa = Animal
Fertiliser; R = irrigated area; S= Seed; P = Pesticide; and W =Consumed (Demanded)
water. Pi are respective input prices.
Cost minimisation problem for a firm can be written as a constraint optimisation
equation, as:
l  (Pw.W  Pl .L  Pf .F  Pp .P  Pr .R  Pm.M  Pa.Fa  pS .S)  (Y 0  AWa1 La2 Fa3 Pa4 Ra5 M a6 Fa7 S a8 )
Where λ is the lagrangian multiplier.
The first-order conditions for cost minimisation are:
dl
 Pw   Aa2W a1 1 La2 F a3 P a4 R a5 M a6 F a7 S a8  0
dW
181
dl
 Pl   Aa2W a1 La2 1F a3 P a4 R a5 M a6 F a7 S a8  0
dL
dl
 Pf   Aa3W a1 La2 F a3 1 P a4 R a5 M a6 F a7 S a8  0
dF
dl
 Pp   Aa4W a1 La2 F a3 P a4 1 R a5 M a6 F a7 S a8  0
dP
dl
 Pr   Aa5W a1 La2 F a3 P a4 R a5 1M a6 F a7 S a8  0
dR
dl
 Pm   Aa6W a1 La2 F a3 P a4 R a5 M a6 1 F a7 S a8  0
dM
dl
 Pa   Aa7W a1 La2 F a3 P a4 R a5 M a6 F a7 1S a8  0
dFa
dl
 Ps   Aa8W a1 La2 F a3 P a4 R a5 M a6 F a7 S a8 1  0
dS
dl
 Y 0  AW a1 La2 F a3 P a4 R a5 M a6 F a7 S a8  0
d
Dividing equations from 2 to 8 by equation 1 and after rearranging the terms are,
Pl a2 W
a P
   RTSl , w  L  W  2  w
Pw a1 L
a1 Pl
Pf
Pw
Pp
Pw

a3 W
a P
  RTS f , w  F  W  3  w
a1 F
a1 Pf

a4 W
a P
  RTS p , w  P  W  4  w
a1 P
a1 Pp
a P
Pr a5 W
   RTS r , w  R  W  5  w
Pw a1 R
a1 Pr
Pm a6 W
a P
 
 RTS m , w  M  W  6  w
Pw a1 M
a1 pm
182
Pa a7 W
a P
 
 RTS a , w  Fa  W  7  w
Pw a1 Fa
a1 Pa
Ps a8 W
a P
   RTS s , w  S  W  8  w
Pw a1 S
a1 Ps
Ps a8 W
a P
a P
   RTS s , w  W  S  1  s  M  1  m  ....
Pw a1 S
a8 Pw
a6 Pw
Then substituting above equation in production function and will have,
a
2

a P
a P 
. a1 . W  2  w  . W  3  w 
Y  AW
a1 Pl  
a1 Pf 

a3
a4
a
a
a
a
6
7
8
5

a P 
a P  
a P 
a P
a P 
. W  4  w  . W  5  w  . W  6  w  . W  7  w  . W  8  w 
a1 Pp  
a1 Pr  
a1 pm  
a1 Pa  
a1 Ps 

a3
a
2
 a2 Pw   a3 Pw   a4 Pw 

Y  AW
.
.   .   .  
 a1 Pl   a1 Pf   a1 Pp 
a4
ai
a P 
. 5  w 
 a1 Pr 
a5
a6
a P 
. 6  w 
 a1 pm 
a7
a P 
. 7  w 
 a1 Pa 
a8
a P 
. 8  w 
 a1 Ps 
Solving for W,
W 
Y
ai
a P 
A.  2  w 
 a1 Pl 
a2
a P 
. 3  w 
 a1 Pf 
a3
a P 
. 4  w 
 a1 Pp 
a4
a P 
. 5  w 
 a1 Pr 
a5
a P 
. 6  w 
 a1 pm 
a6
a P 
. 7  w 
 a1 Pa 
a7
a P 
. 8  w 
 a1 Ps 
Y .a 1 ( a 2  a 3  a 4  a 5  a 6  a 7  a 8 ) . Pl a 2 . P f a 3 . P p a 4 . Pr a 5 . Pm a 6 . Pa a 7 . PS a 8
ai


W
A .a 2 a 2 .a 3 a 3 .a 4 a 4 .a 5 a 5 .a 6 a 6 .a 7 a 7 .a 8 a 8 . PW( a 2  a 3  a 4  a 5  a 6  a 7  a 8 )
W

Y
ay
. B . Pl b 2 . P
f
b3
. P p b4 . Pr b5 . Pm
P wb 1
183
b6
.Pa b7 .PS
b8
a8
Then solve for other factors and will have,
L 
F
. B . P wb 1 . P
b3
f
. P p b4 . Pr b5 . Pm
b6
.Pa b7 .PS
b8
b6
.Pa b7 .PS
b8
b6
.Pa b7 .PS
b8
Pl b 2
. B . P l b 2 . P wb 1 . P p b 4 . P r b 5 . P m
ay
Y

P
.B .Pl b2 .P
ay
Y
P 
b3
f
ay
Y
. B . Pl b 2 . P f
b3
b 3
f
. P wb 1 . P r b 5 . P m
b 4
p
P
R 
M
ay
Y
. P p b 4 . P wb 1 . P m b 6 . P a b 7 . P S b 8
P rb 5
a y
Y

b
f
3
.P
b
p
4
P mb
Y
F a 
ay
. B . Pl b2 . P
f
b3
.Pr
b
5
. P wb 1 . P a
b
7
.PS
b8
6
. P p b4 . Pr b5 . Pm
b6
. P wb 1 . P S
b8
P ab 7
a y
Y
S 
.B .Plb2 .P
.B .Pl b2 .P
f
b3
.P
b
p
p
4
.P r b5 .Pm
b6
.Pa
b7
.p
b1
w
b 8
s
Where,
a1
B 
1
a2
( a 2  a3  a 4  a5  a 6  a 7  a8 )
ai
a3

a5
a4
a6
a7
a8
ai
ai
ai
ai
ai
ai
ai
ai
A  .a 2  .a 3  .a 4  .a 5  .a 6  .a 7  .a 8 
ay 
b5 
; b1 
a1
; b2 
 ai
a2
; b3 
 ai
a3
; b4 
 ai
a5
; b6 
 ai
a6
; b7 
 ai
a7
; b8 
 ai
a8
 ai

1
ai
Now, equations put into the cost function and will have,
184
a4
;
 ai
C(Pw, Pl , Pf , Pp , Pr , Pm, Pa , PS ,Y)  Pw.W  Pl . L  Pf . F  Pp. P  Pr . R  Pm. M  Pa. Fa  PS . S
By Shepard’s Lemma the firm’s system of cost minimizing input demand functions
(the conditional factor demands) will be obtained by differentiating the cost function
with respect to input prices:
Y ay .B.Pl b2 .Pf b3 .Ppb4 .Prb5 .Pmb6 .Pab7 . pS b8 
C(Pw, Pl , Pf , Pp , Pr , Pm , Pa , PS ,Y )  Pw. 

b1
P
w


Y ay .B.Pl b2 .Pwb1.Ppb4 .Prb5 .Pmb6 .Pab7 .PS b8 
Y ay .B.Pwb1.Pf b3 .Ppb4 .Prb5 .Pmb6 .Pab7 .PS b8 
Pl . 

  Pf . 
b2
b3
P
P
l
f




Y ay .B.Pl b2 .Pf b3 .Pwb1.Prb5 .Pmb6 .Pab7 .PS b8 
Y ay .B.Pl b2 .Pf b3 .Ppb4 .Pwb1.Pmb6 .Pab7 .PS b8 
Pp . 

  Pr . 
b4
b5
P
P
p
r




Y ay .B.Pl b2 .Pf b3 .Ppb4 .Prb5 .Pmb6 .Pwb1.PS b8 
Y ay .B.Pl b2 .Pf b3 .Ppb4 .Prb5 .Pwb1.Pab7 .PS b8 
Pm. 
  Pa . 

b7
b6
P
P
a
m




PS .
Y ay .B.Pl b2 .Pf b3 .Ppb4 .Prb5 .Pmb6 .Pab7 .Pwb1
Psb8
After rearranging the terms will have,
C ( Pw .Pl .Pf .Pp .Pr .Pm .Pa .PS )  B.Y ay .( Pwb1 .Pl b2 .Pf b3 .Pp b4 .Pr b5 .Pm b6 .Pa b7 .PS b8 ) 
( Pw1 2 b1  Pl1 2 b2  Pf1 2 b3  Pp1 2 b4  Pr1 2 b5  Pm1 2 b6  Pa1 2 b7  p1s  2 b8 )
C
W
 PW
c
Y
ay
. B . Pw b 1 Pl b 2 . P f b3 . P p b 4 . Pr b5 . Pm b6 . Pa b7 . p S b8
In logarithms, it becomes:
lnW  ln B  ay lnY b1 ln Pw  b2 ln Pl b3 ln Pf b4 ln Pp b5 ln Pr b6 ln Pm  b7 ln Pa b8 ln PS
185
Appendix E: Schematic of the Compute Stages of the Water Demand
I. The steps were extracte of published report of Soil and Water Research Institute
(SWRI)
Gathering of climate existing data from the Meteorological Organization, consisting
of daily readings from multiple stations.
To select the best method to use for determining crop water requirements.
( The Penman-Monteith method was selected as the best method)
To calculate Reference Crop Evapotranspiration Standard of Grass (ETo) values and
validity them.
To select the annual crops and fruit trees under consideration of each plain and to
determine of crop coefficient (Kc) them by recommended method via FAO, regional
condition and previous experimental.
To calculate the Efficient Rainfall based on the presented method via American
Society of Civil Engineers (ASCE).
II. The steps which were performed by researcher
To calculate the total irrigation requirement that is given by the following equation:
Irrigation water
net requirement
(IRReq)
crop water
= requirements
(ETcrop)
-
Efficient
rainfall (Peff)
To calculate the consumed (demanded) water that is given by the following equation:
Consumed
Water (m3)
Total Irrigation
= Requirement
186
+
[Total Irrigation requirement
Irrigation Efficiency
(1 
)]
100
Appendix F: Estimation Results of the Water Demand Functions
Table F.1. Estimation results of the water demand function for wheat
production with VMP, AVC and MC (2001-2006)
Method: Panel
EGLS
Dependent Variable: LDWT
Method: Panel Least Squares
VMP
Coef.
Independent variable
AVC
Std.
Error
C
14.120***
0.92
LAVC
LMC
LVMP
-0.20***
0.05
***
LQ
0.47
0.09
**
LW
0.06
0.03
***
LRL
0.12
0.03
***
LPP
0.03
0.01
LCL
LPS
R2
0.99
Adjusted R2
0.99
Durbin-Watson stat
1.72
F-statistic
846.36
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
187
MC
Coef.
Std.
Error
Coef.
Std.
Error
3. 65***
-0.02**
0.43
0.01
1.50***
0.42
-0.07***
0.007
0.98***
-0.16***
0.02
0.08
0.25*
0.97
0.97
2.41
248.05
0
0.15
0.96***
-0.18***
0.05
0.03
0.14***
0.04
0.99
0.99
2.00
433.26
0
Table F.2. Estimation results of the water demand function for barley
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
VMP
AVC
MC
Coef. Std.
Independent variable
Error
***
0.64
C
2.53
LAVC
LMC
LVMP
-0.66***
0.02
***
LQ
0.89
0.05
***
LW
0.15
0.03
***
LPS
0.38
0.06
D(LPfa)
0.02*
0.01
LPP
LPM
LRL
-0.03
0.008
2
R
0.99
2
Adjusted R
0.99
Durbin-Watson stat
1.81
F-statistic
2240.20
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
Coef.
0.56
-0.075***
Std.
Error
1.19
0.02
0.896***
0.16
0.03
0.11
0.02***
0.07**
0.14***
0.004
0.03
0.04
0.99
0.99
2.46
499.7
0
Coef.
2.18***
Std.
Error
0.36
-0.03*
0.02
0.91***
-0.29***
0.78***
0.02
0.08
0.16
-0.22***
0.06
0.97
0.95
1.80
53.45
0
The sign D shows it has a unit root, meaning the time series under consideration
(logarithm of animal fertilizer price) was nonstationary and we make it stationary.
188
Table F.3. Estimation results of the water demand function for lentil
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method:
Independent
variable
Panel EGLS
(Period SUR)
Panel EGLS
(Period SUR)
VMP
Coefficient
Std.
Error
Panel Least
Squares
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
1.57**
0.78
LAVC
LMC
LVMP
-0.88***
0.04
***
LQ
1.01
0.03
***
LW
0.34
0.07
***
LPS
0.31
0.08
***
LRL
-0.1
0.04
LCL
LPM
0.07
0.05
LPP
LPF
0.98
R2
2
Adjusted R
0.98
Durbin-Watson stat
2.04
F-statistic
440.13
Prob(F-statistic)
0
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
189
8.36***
-0.09*
2.08
0.04
0.71***
0.07
0.03
0.1
-0.53***
0.245***
0.14
0.05
0.99
0.97
2.4
49.33
0
3.69*
1.8
-0.1***
0.03
0.95***
-0.26***
0.11
0.08
-0.45***
1.18**
0.96
0.78
1.66
10.7
0.001
0.115
0.41
Table F.4. Estimation results of the water demand function for pea
production with VMP, AVC and MC ( 2001-2006 )
Dependent Variable: LDWT
Method:
Independent
variable
Panel Least Squares
VMP
Coefficient
Std.
Error
Panel EGLS
Panel Least Squares
(Cross-section
weights)
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
1.12*
0.59
LAVC
LMC
LVMP
-0.49***
0.04
***
LQ
0.94
0.02
***
LW
-0.08
0.01
***
LPS
0.68
0.04
***
LRL
-0.05
0.0
LCL
LPM
LPP
LPF
0.10*
0.06
LPA
R2
0.99
2
Adjusted R
0.99
Durbin-Watson stat
1.83
F-statistic
294.45
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
190
2.45
-0.03*
1.83
0.01
1.83*
0.95
-0.06*
0.03
0.98**
0.05
0.78***
-0.56***
0.05
1.69***
0.49***
0.46
0.16
0.24*
0.22*
0.18**
-0.52*
0.22***
0.99
0.99
2.36
180.81
0.00
0.10
0.06
0.04
0.09
0.01
-0.36**
-2.45***
0.13
0.59
0.96
0.91
1.97
17.79
0.00
Table F.5. Estimation results of the water demand function for pinto bean
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
5.08***
0.53
LAVC
LMC
LVMP
-0.39***
0.05
***
LQ
0.89
0.02
LW
-0.04
0.03
*
LPS
0.16
0.09
**
LCL
-0.07
0.03
LPM
LPP
LPF
R2
0.99
2
Adjusted R
0.99
Durbin-Watson stat
2.02
F-statistic
711.81
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
191
-1.99*
-0.06***
1.01***
0.36***
0.06
-0.09*
0.98
0.97
1.65
94.25
0.00
1.14
0.015
0.04
0.05
0.12
0.05
0.28***
0.30
-0.01
0.05
0.92***
0.15*
0.64***
0.01
0.06
0.06
0.21
-0.49
0.05
0.92
0.74
2.23
4.92
0.03
0.45
0.63
0.01
Table F.6. Estimation results of the water demand function for onion
production with VMP, AVC and MC ( 2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
0.68*
0.40
LAVC
LMC
LVMP
-0.20***
0.05
***
LQ
0.88
0.03
LW
LPS
LPP
LPM
LPfa
LRL
R2
0.99
2
Adjusted R
0.99
Durbin-Watson stat
1.99
F-statistic
405.32
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
192
0.52
0.60
2.40***
0.18
-0.04***
0.01
-0.01***
0.02
0.89***
0.03
0.84***
-0.12***
-0.11***
0.14***
0.09***
0.01
0.02
0.01
0.00
0.02
0.04**
0.03
0.99
0.99
1.81
299.53
0.00
0.02
0.02
-0.02
0.01
0.99
0.99
2.20
1740.93
0.00
Table F.7. Estimation results of the water demand function for tomato
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
8.57***
0.34224
LAVC
LMC
LVMP
-0.89***
0.02917
***
LQ
0.88
0.025
***
LW
-0.43
0.05427
**
LPM
-0.05
0.02437
LPS
-0.01
0.01408
***
LPF
-0.27
0.05195
LPfa
LPP
LRL
0.99
R2
2
Adjusted R
0.99
Durbin-Watson stat
2.00
F-statistic
663.549
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
193
-1.10
-0.04**
0.98
0.02
0.95***
0.13**
0.07
0.06
-0.08***
-0.06***
0.04**
0.02
0.02
0.02
0.99
0.99
2.46
269.72
0.00
0.04
1.14
-0.01
0.01
0.92***
0.07
-0.06***
0.02
0.03*
0.99
0.99
2.56
253.04
0.00
0.02
Table F.8. Estimation results of the water demand function for potato
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
3.37***
0.50
LAVC
LMC
LVMP
-0.16***
0.02
***
LQ
0.82
0.02
LW
-0.04
0.03
LPM
LPS
0.05**
0.02
LPF
LPfa
LPP
LRL
R2
0.99
2
Adjusted R
0.99
Durbin-Watson stat
2.02
F-statistic
928.134
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
194
2.72***
-0.004
0.71
0.01
0.80***
0.04
-0.01*
0.01
0.99
0.99
2.23
606.397
0.00
3.88**
1.7
-0.02
0.01
0.69***
0.13***
-0.07**
0.07
0.06
0.02
0.99
0.99
2.23
315.6
0.00
Table F.9. Estimation results of the water demand function for cucumber
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
4.41***
0.49
LAVC
LMC
LVMP
-0.21***
0.02
***
LQ
0.75
0.03
**
LW
0.09
0.04
LPM
0.09***
0.04
**
LCL
-0.08
0.04
LPF
LPfa
LPP
LPS
0.99
R2
2
Adjusted R
0.99
Durbin-Watson stat
1.99
F-statistic
520.282
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
195
4.34***
-0.01**
0.71***
-0.03***
-0.03
***
0.99
0.99
2.06
400
0.00
1.25
0.01
0.07
0.01
6.84***
0.29
-0.01***
0.003
0.65***
-0.09**
-0.13***
0.02
0.04
0.01
0.22***
0.06
-0.08***
-0.07***
0.99
0.99
2.09
855.37
0.00
0.02
0.01
0.01
Table F.10. Estimation results of the water demand function for watermelon
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
1.40***
0.26
LAVC
LMC
LVMP
-0.26***
0.04
***
LQ
0.84
0.0
***
LW
-0.08
0.03
*
LPM
-0.04
0.03
***
LCL
0.13
0.04
LPF
LRL
LPP
0.07***
0.03
**
LPS
0.05
0.02
R-squared
0.99
Adjusted R-square
0.99
Durbin-Watson stat
2.02
F-statistic
557.215
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
196
-0.73***
-0.005*
0.012***
0.051***
-0.016*
-0.03*
0.004
0.65
0.56
1.91
7.30
0.00
0.165
0.003
0.0005
0.012
0.01
0.017
0.004
0.209***
0.016
0.006***
0.009
0.0003
0.0006
***
0.080
0.020
-0.017**
0.008
-0.014***
0.003
-0.005**
0.002
0.55
0.47
1.97
6.33
0.00
Table F.11. Estimation results of the water demand function for cotton
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
VMP
Coefficient
Std.
Error
1.29*
1.51
Independent
variable
C
LAVC
LMC
LVMP
-0.88***
0.08
***
LQ
0.99
0.04
***
LW
0.24
0.11
***
LPF
0.35
0.17
LRL
0.13***
0.04
*
LPP
0.08
0.05
**
LPS
0.16
0.08
LPA
LPM
R2
0.98
2
Adjusted R
0.98
Durbin-Watson stat
2.41
F-statistic
557.215
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
***
***
-10.27
1.27
-15.94
1.72
***
0.02
0.01
0.09*
0.05
197
0.98***
0.82
0.27
0.02
0.41***
0.38***
0.23***
-0.15
0.98
0.97
2.35
80
0.00
0.04
0.01
0.07
0.07
0.24
1.22***
1.29***
-0.72***
0.05
0.09
0.24
0.33***
1.01**
0.09
0.40
-0.59*
0.96
0.92
2.26
20.8
0.00
0.29
Table F.12. Estimation results of the water demand function for sugar beet
production with VMP, AVC and MC (2001-2006)
Dependent Variable: LDWT
Method: Panel Least Squares
Independent
variable
VMP
Coefficient
Std.
Error
AVC
MC
Coefficient Std. Coefficient Std.
Error
Error
C
5.11***
0.71
LAVC
LMC
LVMP
-0.50***
0.10
***
LQ
0.89
0.014
***
LW
-0.31
0.07
LPF
-0.15***
0.05
***
LPS
-0.04
0.01
***
LPM
-0.15
0.04
***
LCL
0.23
0.06
LPP
LPA
0.98
R2
2
Adjusted R
0.98
Durbin-Watson stat
2.02
F-statistic
557.215
Prob(F-statistic)
0.00
*
Statistically significant at the 10% level
**
Statistically significant at the 5% level
***
Statistically significant at the 1% level
198
0.88
-0.01**
0.77
0.005
0.84***
0.34***
-0.33***
-0.14***
0.05
0.11
0.10
0.03
0.10***
0.04
0.98
0.97
2.44
222.6
0.00
1.20***
1.25
-0.08**
0.03
0.93
-0.45***
0.06
0.14
0.1***
0.04
0.15***
0.96
0.93
2.09
29.98
0.00
0.06
Appendix G: Studies with high R2
Author(s)
Year
Title of Research
Rana Hasan
2001 The impact of trade and labor market
regulations on employment and wages
Aurella.B.M.,
2001 Economic Growth and CO2 Emissions
Francisco.H.T.,and
in the European Union
Inmaculada. M.Z
Hossein zad, J., 2005 Choosing an empirical production
Salami, H. A
function to estimate economic value …..
Ruth-Aïda Nahum 2005 Income inequality and growth: A panel
study of swedish counties 1960-2000
Matz Dahlberg
2005 Inequality and crime: separating the
and Magnus
effects of permanent and transitory
Gustavsson
income
Mohammad
2006 A panel data analysis of banglaesh’s
Mafizur Rahman
trade: The gravity model approach
Moolman. C.E,
2006 Modelling the marginal revenue of water
Blignaut .J.N & R
in selected agricultural commodities: A
van Eyden
panel data approach
Andrew Leigh
2007 How closely do top income shares track
other measures of inequality
Victor Brescia
2007 Supply elasticities for selected
Daniel Lema
commodities in Mercosur and Bolivia
Antonio Estache 2007 Regulatory agencies: Impact on firm
performance and consumer welfare
Marius Brulhart
2008 Sectoral agglomeration economies in a
And Mathys. N.A
panel of European regions
Batool Asiri
2008 Testing weak-form efficiency in the
Bahrain stock market
Emin Koksal
2008 An analysis of public expenditures using
the median voter theorem for Turkey
Vialou Alexandre, 2008 Impact of GMO crop adoption on
et al.
quality-adjusted pesticide use in corn
and soybeans: A full picture
Hassen et al.
2008 The effect of heritability estimates on
high-density SNP analyses with related
animals
Richard Frensch
2009 Trade liberalisation and import margins
Jalaie and Naghavi 2009 The analysis of Iran and European
Union regionalism in agricultural sector
Anna Aizer
2009 The gender wage gap and domestic
violence
Gabriele Ruoff
2009 Grow rich and clean up later? Joint
effects of IGO membership and
democracy on environmental
performance in developing countries
199
Amount of
R2
0.99
0.37 in
(Pool) and
0.99 in Panel
0.91, 0.94
and 0.96
0.90, 0.98
and 0.99
0.97 and
0.99
0.92 and
0.86
0.995
0.91, 0.98
and 0.99
0.78, 0.85
and 0.98
0.98 and
0.99
0.88 and
0.99
0.0001 and
0.9995
0.98 and
0.99
0.98 and
0.99
0.99
0.99
0.989
0.96, 0.98
and 0.99
0.99 and
1.00
Appendix H: Descriptive Statistics
Table H.1. The descriptive data of barley, cotton, wheat and cucumber
Barley
Cotton
Std.
Variable
Mean
Dev.
Observations Mean
Std. Dev. Observations
Demand Water
1.95E+08 2.93E+08
151
3.58E+08 6.88E+08
78
Water Price
5.93
4.22
151
3.49
2.69
78
Land Rent
8.99
5.26
151
165321.20 119595.50
78
Seed Price
135.34
36.64
151
294.19
95.00
78
Fertilizer Price
52.42
11.19
151
54.29
12.21
78
Pesticide Price
1754.14
1298.26
151
2987.63
1583.18
78
Wage
4727.44
1940.94
151
4048.76
1677.67
78
Land Prepare Cost
3.56
1.67
151
48907.44 21875.51
78
Machinery Rent
Cost
13.84
7.13
151
9.77
4.10
78
Animal Fertilizer
Price
3.12
2.97
151
5.14
3.98
78
Irrigated Crop
Production
73611395 1.07E+08
151
27215377 39565536
78
VMP
57.91
22.29
151
66.39
57.08
78
Crop
Wheat
Cucumber
Std.
Variable
Mean
Dev.
Observations Mean
Std. Dev. Observations
Demand Water
6.66E+08 8.03E+08
167
31328535 38851345
108
Water Price
6.71
4.28
167
21.54
18.31
108
Land Rent
11.20
6.90
167
168604.90 95938.16
108
Seed Price
169.90
45.56
167
48612.32 19546.59
108
Fertilizer Price
52.55
10.61
167
63.78
23.06
108
Pesticide Price
2248.81
1206.84
167
3684.12
1551.29
108
Wage
4758.18
1940.24
167
4467.75
1939.58
108
Land Prepare Cost
3.70
1.55
167
74255.38 34614.42
108
Machinery Rent
Cost
16.95
8.59
167
20.17
11.13
108
Animal Fertilizer
Price
3.12
3.18
167
7.28
5.94
108
Irrigated Crop
Production
3.14E+08 3.85E+08
167
84787937 1.01E+08
108
VMP
138.29
55.56
167
613.09
410.21
108
Crop
200
Crop
Table H.2. The descriptive data of lentil, onion, potato and pea
Lentil
Onion
Variable
Demand Water
Water Price
Land Rent
Seed Price
Fertilizer Price
Pesticide Price
Wage
Land Prepare
Cost
Machinery Rent
Cost
Animal Fertilizer
Price
Irrigated Crop
Production
VMP
Crop
Variable
Demand Water
Water Price
Land Rent
Seed Price
Fertilizer Price
Pesticide Price
Wage
Land Prepare
Cost
Machinery Rent
Cost
Animal Fertilizer
Price
Irrigated Crop
Production
VMP
Mean
8315997
6.36
79062.68
381.32
47.86
1857.72
4054.87
Std. Dev. Observations Mean
Std. Dev. Observations
12706948
44
28089994 31102646
100
4.84
44
13.38
8.38
100
51065.85
44
196064.50 137970.50
100
169.68
44
17170.59 20028.28
100
19.39
44
58.79
16.82
100
1744.47
44
4021.00
1764.34
100
1832.63
44
4425.47
1819.48
100
34681.45
17455.58
44
82724.69
55612.74
100
7.16
4.41
44
7968.58
38272.94
100
2.70
3.82
44
7.74
7.70
100
1521337
65.55
2176976
37.88
Potato
44
44
90032257
341.26
99025506
213.31
Pea
100
100
Mean
Std. Dev. Observations
63756810 58803710
138
12.92
10.33
138
182828.00 99512.18
138
151.40
56.55
138
56.45
13.75
138
2821.59
1191.90
138
4699.71
1890.24
138
Mean
6918225
10.66
72522.90
408.65
52.43
2230.69
5127.19
Std. Dev.
7931064
12.17
51149.92
151.23
17.48
1579.51
3046.06
Observations
63
63
63
63
63
63
63
59456.63
35096.87
138
35206.93
19908.76
63
22.80
14.05
138
6.56
3.94
63
6.04
6.00
138
2.59
3.86
63
1.78E+08
529.12
1.91E+08
2044.01
138
138
1515501
96.70
1288191
71.94
63
63
201
Crop
Table H.3. The descriptive data of pinto been, sugar beet, tomato and
watermelon
Pinto bean
Sugar beet
Variable
Demand Water
Water Price
Land Rent
Seed Price
Fertilizer Price
Pesticide Price
Wage
Land Prepare
Cost
Machinery Rent
Cost
Animal Fertilizer
Price
Irrigated Crop
Production
VMP
Crop
Variable
Demand Water
Water Price
Land Rent
Seed Price
Fertilizer Price
Pesticide Price
Wage
Land Prepare
Cost
Machinery Rent
Cost
Animal Fertilizer
Price
Irrigated Crop
Production
VMP
Mean
52884859
10.44
55219.01
571.04
104.19
2302.41
5004.43
Std. Dev.
54077290
11.96
36741.30
185.02
120.77
1349.03
2201.26
Observations
58
58
58
58
58
58
58
Mean
1.31E+08
8.84
161092.60
4328.92
56.10
2777.55
4505.52
Std. Dev.
2.03E+08
6.02
86925.56
6424.30
16.04
1300.33
1835.24
Observations
82
82
82
82
82
82
82
154457.50
96979.07
58
43308.14
19607.71
82
10.71
5.24
58
369.62
2858.80
82
2.91
2.84
58
2.47
2.00
82
14872056
142.98
15575644
65.60
Tomato
58
58
3.21E+08
161.01
4.82E+08
310.85
Watermelon
82
82
Mean
31722172
2999.25
111229.00
23659.45
50.05
2999.25
3693.46
Std. Dev. Observations
40303550
126
2940.58
123
103095.30
125
25540.23
126
37.77
126
2940.58
123
2577.01
125
Mean
Std. Dev. Observations
51687870 52447859
144
14.96
11.24
144
201880.60 129428.90
144
54680.76 31047.86
144
62.56
20.49
144
3671.63
1803.86
144
4482.34
1740.89
144
71633.11
31883.39
144
49306.86
35536.62
124
22.43
12.16
144
15.13
11.39
125
5.85
5.99
144
7.57
8.00
125
1.68E+08
830.52
1.87E+08
597.67
144
144
1.09E+08
308.04
1.47E+08
202.90
126
126
202
BIODATA OF STUDENT
Ahmad Sadeghi son of Baratali was born on 18 March 1965 in Tehran (Capital of
I.R.Iran). He completed elementary guidance and high school studies in Tehran city.
He got his high school diploma in the field of experimental sciences in 1983. After
two years compulsory soldiering, he passed entrance exams and started his education
in the field of Agricultural Economics in Tehran University where he backed his
B.Sc. in 19 Feb 1992.
He worked in the Ministry of Education as a teacher in the same year. Then he
obtained master’s degree at Sistan and Balouchestan University in 2000. He passed
all requisite courses for M.Sc. under Dr. Sepehrdoust, under whom he carried out the
thesis titled “To study on the effects of economic policies - social government on
rural industries of Sistan and Balouchestan”.
He has been working in Tarbiat Modares University as Head of Research Office,
College of Agriculture, from 1995 to 2002 year. He worked in Power and Water
University of Technology (PWUT), Tehran, Iran as lecturer from 2002 to until now.
He began his PhD program under the supervision of Prof. Dr. Mohd Gazali
Mohayidin in the field of Agricultural Economics in 2006 in University Putra
Malaysia (UPM). He is married and has two daughters.
203
LIST OF PUBLICATIONS
Ghazali, M., Attari, J., Sadeghi, A., & Arrif, H. (2009). Review of water pricing
theories and related models. African Journal of Agricultural Research, 4(13), 15361544.
Sadeghi, A., Mohayidin, G., Hussein, A., & Baheiraie, A. (2009). Determining the
Economic Value of the Irrigation Water in Production of Wheat in Iran. Australian
Journal of Basic and Applied Sciences,.
Sadeghi, A., Mohayidin, G., Hussein, A., & Attari, J. (2010). Estimation of Irrigation
Water Demand for Barley in Iran: The panel Data Evidence. Journal of Agricultural
Science,2-2,.
204