an economic analysis of institutional reforms in...
TRANSCRIPT
i
AN ECONOMIC ANALYSIS OF INSTITUTIONAL
REFORMS IN IRRIGATION SECTOR IN PUNJAB
PAKISTAN
By
MUHAMMAD ARIF RAZA
M.Sc (Hons) AGRICULTURAL ECONOMICS
A THESIS SUBMITTED IN PARTIAL FULFILLMENT FOR THE DEGREE
OF
DOCTOR OF PHILOSOPHY
IN
AGRICULTURAL ECONOMICS
Faculty of Agricultural Economics and Rural Sociology
University of Agriculture Faisalabad
2008
i
TABLE OF CONTENTS
Chapter
No. Title Page No.
List of Tables vii
List of Figures ix
Acknowledgements x
Dedication xii
Abstract xiii
1 INTRODUCTION 1
1.1 A Brief History of Irrigation Reforms in the World 3
1.2 Agriculture and Economy of Pakistan 5
1.3 Irrigation Sector in Punjab 9
1.4 Canal Irrigation System of Punjab 10
1.4.1 Main Canal Level 10
1.4.2 Distributary Level 10
1.4.3 Watercourse Level 10
1.5 Objectives of Irrigation System in Punjab 12
1.5.1 Design Stage Objectives 12
1.5.2 Operational Objectives. 12
1.6 Issues of Irrigation Management 13
1.6.1 Low Irrigation and Water Use Efficiencies 13
1.6.2 Poor Aabiana Collection (Water Charges) 14
1.6.3 Inequitable Distribution of Irrigating Water 14
1.6.4 Inadequate Maintenance of the System 15
1.6.5 Poor Irrigation Infrastructure 15
1.6.6 Illegal Diversion from Distributaries 16
1.6.7 Inadequate Farmer‘s Participation in Irrigation
Management 16
1.6.8 Corruption and Rent Seeking Behavior of the
Officials 16
1.7 Need for Institutional Reforms in Irrigation
Sector 16
ii
1.7.1 Irrigation Management Transfer 17
1.7.2 Participatory Irrigation Management 17
1.8 PIM Model for Punjab 17
1.9 Description and Functions of PIDA, AWB and FOs 19
1.9.1 Punjab Irrigation and Drainage Authority
(PIDA) 19
1.9.2 Area Water Board (AWB) 20
1.9.3 Farmers Organizations (FOs) 20
1.10 Schedule for Formation of FOs 20
1.11 Legal Framework and Composition of PIDA 21
1.12 Need for the Study 24
1.13 Objectives of the Study 25
1.14 Summary 25
2 REVIEW OF LITERATURE 26
2.1 Selected Studies from Outside Pakistan 26
2.2 Selected Studies Related to Pakistan 41
2.3 Summary 49
3 METHODOLOGY 50
3.1 Selection of the Study Area 50
3.2 Characteristics of the Study Area 50
3.2.1 A Brief Description of the Districts in the Study
Area 51
3.2.2 Specific Characteristics of the LCC (East) 53
3.3 Sampling Framework 56
3.3.1 Sampling Size 59
3.3.2 Questionnaire Development 59
3.3.3 Pre-Testing of the Questionnaire 60
3.4 Data Sources and Collection 60
3.4.1 Primary Data Collection 60
iii
3.4.2 Secondary Data Collection 61
3.5 Data Entry 62
3.6 Cleaning and Organization of Data 62
3.7 Indicators for Performance Measurement 62
3.8 Summary 66
4 CONCEPTUAL FRAMEWORK 67
4.1 Single Equation Estimation: OLS Approach 67
4.1.1 Assumptions of OLS 67
4.1.2 General Form of the Model 69
4.2 Concept of Production Function 69
4.2.1 Cobb-Douglas Production Function 70
4.3 Efficiency 71
4.3.1 Input Oriented Efficiency Measures 72
4.3.2 Concept of Production Frontier 73
4.3.3 Measurement of Technical Efficiency 74
4.3.4 Parametric Frontier Production Function 75
4.3.5 Stochastic Frontier Production Function 76
4.3.6 Estimation of Stochastic Frontier
Production Function 78
4.3.7 Estimation of Mean and Firm-Level Technical
Efficiencies 79
4.3.8 Technical Inefficiency Effects Model 80
4.4 Summary 82
5 METHODS OF ANALYSIS 83
5.1 Approaches for Analysis 83
5.2 Quantitative Analysis 83
5.2.1 Comparison of Means, Averages, Percentages
and Frequencies 84
5.2.2 Estimation of Econometric Models for Different
Crops 85
iv
1 Estimation of Model for Gross Value Product
of Wheat 85
2 Estimation of Model for Wheat Yield 87
3 Estimation of Model for Gross Value Product
of Rice Crop 89
4 Estimation of Model For Rice Yield 91
5 Estimation of Model for Gross Value Product
of Sugarcane Crop 92
6 Estimation of model for Sugarcane Yield 94
5.3 Economic Inefficiency Model 96
5.3.1 Empirical Model and Efficiency Analysis 96
5.4 Summary 100
6 COMPARISON OF PERFORMANCE INDICATORS 101
6.1 Socio-Economic Profile of the Respondents Across the
Study Area 101
6.1.1 Social Indicators 102
6.1.2 Agricultural Indicators 104
6.2 Comparison of Indicators Developed from Primary Data
Sources 105
6.2.1 Opinion of the Farmers Regarding O&M and
Quantity of Irrigation Water 106
6.2.2 Farmer‘s Perception on Reduction in Water
Theft Cases 109
6.2.3 Average Yield of Major Crops Across the
System 110
6.2.4 Average Gross Value Product (GVP) of Major
Crops Across the System 112
6.2.5 Average Cost of Production of Major Crops
Across the System 114
6.2.6 Comparison of Gross Margin of Major Crops 124
v
Across the System in Pre and Post Reform
Period
6.2.7 Ratio of Gross Margin to Cost of Production
Across the System 127
6.2.8 Cropping Intensity in the Study Area in Pre and
Post- reform Period 128
6.3 Comparison of Indicators Developed from Secondary
Data Sources 129
6.3.1 Comparison of Aabiana Assessment and
Collection in Pre and Post-reform Period 129
6.3.2 Comparison of Per Hectare Operation and
Maintenance Expenditures in Pre and Post-
reform Period
131
6.3.3 Per Hectare Salary Expenditures in Pre and Post-
reform Period 133
6.3.4 Per Hectare Contingency Expenditures in Pre
and Post-reform Period 134
6.3.5 Delivery Performance Ratio of Selected
Distributaries in Pre and Post-reform Period 135
6.3.6 Comparison of Head-Tail Equity in Water
Distribution on the Selected Distributaries 136
6.4 Summary 139
7 ESTIMATION OF REGRESSION MODELS 140
7.1 Results of Wheat Regression Models 140
7.1.1 Estimation of Regression Model for Average
Gross Value Product of Wheat 142
7.1.2 Estimation of Regression Model for Average
Yield of Wheat 144
7.2 Results of Sugarcane Regression Models 145
7.2.1 Estimation of Regression Model for Average 146
vi
Gross Value Product of Sugarcane
7.2.2 Estimation of Regression Model for Average
Yield of Sugarcane 148
7.3 Results of Rice Regression Models 150
7.3.1 Estimation of Regression Model for Average
Gross Value Product of Rice 151
7.3.2 Estimation of Regression Model for Average
Yield of Rice 153
7.4 Summary 155
8 STOCHASTIC FRONTIER PRODUCTION FUNCTION
AND TECHNICAL INEFFICIENCY EFFECTS MODEL 156
8.1 Maximum Likelihood Estimates for Parameters of
Stochastic Frontier Production Function 156
8.2 Technical Inefficiency Effects Model 159
8.3 Technical Efficiencies in Crop Production 161
8.4 Summary 163
9 SUMMARY AND CONCLUSIONS 164
9.1 Summary 164
9.2 Conclusions 170
9.3 Limitations of the Study 172
9.4 Problems Faced During Study 172
9.5 Policy Recommendations 173
9.6 Future Areas of Research 175
LITERATURE CITED 177
ANNEXURES 193
vii
LIST OF TABLES
Table
No. Title Page No.
1.1 Countries or States Those Have Adopted IMT During the Past
30 Years
4
1.2 Actual Surface Water Availability in Pakistan 7
1.3 Agricultural Water Demand for Major Crops in Pakistan 8
1.4 Irrigation Network of Punjab 9
1.5 Schedule for Formation of FOs in LCC East Irrigation System of
the Punjab
21
3.1 Salient Features of the Selected Irrigation System 54
3.2 Sampled Distributaries and their Characteristics 55
3.3 Broader Category of Indicators Used at Various Levels in the
Study
63
6.1 Structure and Family Size of the Respondents in the Study Area 102
6.2 Source of Income of the Respondents in the Study Area 103
6.3 Educational Qualification of the Respondents in the Study Area. 104
6.4 Land Holding of the Respondents in the Study Area 104
6.5 Farming Experience of the Respondents in the Study Area 105
6.6 Opinion of the Farmers Regarding O&M and Quantity of
Irrigation Water
108
6.7 Opinion of the Farmers Regarding Cases of Water Theft 109
6.8 Average Yield of Major Crops Across the System 112
6.9 Average Gross Value Product of Major Crops Across the System 114
6.10 Variable Cost of Production of Wheat Crop Across the System
in Pre and Post Reform Period
117
6.11 Variable Cost of Production of Rice Crop Across the System in
Pre and Post Reform Period
119
6.12 Variable Cost of Production of Sugarcane Crop Across the
System in Pre and Post Reform Period
121
viii
6.13 Average Cost of Production of Major Crops Across the System 122
6.14 Average Gross Margin of Major Crops Across the System in Pre
and Post Reform Period
125
6.15 Ratio of Gross Margin to Cost of Production Across the System
in Pre and Post Reform Period
127
6.16 Comparison of Cropping Intensity of the Study Area in Pre and
Post Reform Period
128
6.17 Overall Comparison of Aabiana Assessment and Collection in
Pre and Post Reform Period
130
6.18 Overall Comparison of Per Hectare Operation and Maintenance
(O&M) Expenditures in Pre and Post Reform Period
132
6.19 Per Hectare Salary Expenditures in Pre and Post Reform Period 133
6.20 Per Hectare Contingency Expenditures in Pre and Post Reform
Period
134
6.21 Comparison of Delivery Performance Ratio at Head and Tail of
the Selected Distributaries in Pre and Post Reform Period
137
6.22 Comparison of Equity in Water Distribution on the Selected
Distributaries in Pre and Post Reform Period
138
7.1 Descriptive Statistics of Important Variables for Wheat Crop 141
7.2 Estimated Parameters of the Income Model for Wheat Crop 143
7.3 Estimated Parameters of the Yield Model for Wheat Crop 144
7.4 Descriptive Statistics of Important Variables for Sugarcane Crop 146
7.5 Estimated Parameters of the Income Model for Sugarcane Crop 148
7.6 Estimated Parameters of the Yield Model for Sugarcane Crop 150
7.7 Descriptive Statistics of Important Variables for Rice Crop 151
7.8 Estimated Parameters of the Income Model for Rice Crop 153
7.9 Estimated Parameters of the Yield Model for Rice Crop 154
8.1 Maximum Likelihood Estimates for Parameters of Stochastic
Frontier Production Function and Inefficiency Effects Model
157
8.2 Distribution of Technical Efficiency Estimates 162
ix
LIST OF FIGURES
Figure
No. Title Page No.
1.1 Schematic Diagram of Canal System in Punjab 11
1.2 Flow Diagram Showing the Composition of Punjab Irrigation and
Development Authority
23
3.1 Sampling Framework and Design 58
4.1 Input-Oriented Measures for Technical and Allocative Efficiency 72
4.2 Stochastic Frontier Outputs 78
6.1 Average Yield of Wheat Crop Across the System in Pre and Post
Reform Period
113
6.2 Average Yield of Sugarcane Crop Across the System in Pre and Post
Reform Period
113
6.3 Average Yield of Rice Crop Across the System in Pre and Post
Reform Period
113
6.4 Average Cost of Production Wheat Crop Across the System in Pre and
Post Reform Period
123
6.5 Average Cost of Production of Sugarcane Crop Across the System in
Pre and Post Reform Period
123
6.6 Average Cost of Production of Rice Crop Across the System in Pre
and Post Reform Period.
123
6.7 Comparison of Gross Margin of Wheat Crop Across the System in Pre
and Post Reform Period
126
6.8 Comparison of Gross Margin of Sugarcane Crop Across the System in
Pre and Post Reform Period.
126
6.9 Comparison of Gross Margin of Rice Crop Across the System in Pre
and Post Reform Period
126
6.10 Comparison of Cropping Intensity of the Study Area in Pre and Post
Reform Period.
128
x
All praises and thanks are for ALMIGHTY ALLAH The compassionate, The
merciful, The only creator of the universe, and the source of all knowledge and
wisdom Who blessed me with health, thoughts, talented teachers, co-operative
friends and opportunity to make some contribution to the already existing ocean
of knowledge. I offer my humblest thanks to the greatest social reformer and
Madina-tul-Ilm, The Holy Prophet Hazrat Muhammad (PBUH), for His humanity.
I deem it my utmost pleasure in expressing my heartiest gratitude with the
profound benedictions to Dr. Muhammad Ashfaq, Associate Professor,
Department of Agricultural Economics, University of Agriculture, Faisalabad
for providing me with strategic command at every step. I extend deep emotions
of appreciation, gratitude and indebtedness for his valuable guidance.
I also wish to express my feelings of sincerest gratitude to my Committee
member Dr. Sarfraz Hassan, Associate Professor, Department of Environment and
Resource Economics for his sincere cooperation and encouragements. I deem it
to my utmost pleasure to avail this opportunity in recording my deep feelings of
regards and sense of gratitude to Dr. Intizar Hussain, Senior Water Resources
Management Specialist, Islamic Development Bank, Jeddah, who in spite of his busiest
schedule provided substantial guidance and invaluable assistance in bringing
this dissertation to its present form.
I also feel proud to acknowledge the sincere help of Dr. M. Iqbal Zafar, Dean
Faculty of Agri Economics and Rural Sociology, for inspiring guidance during
my study period.
I don’t have words at command in acknowledging that all the credit goes to my
loving Father, Mother, Brothers, Sister and Sweet wife for their amicable
attitude and love, immense orison, mellifluous affections, inspiration, well
wishing and keen interest which hearten me to achieve success in every sphere
of life. I can never forget the innocent prayers of my sons, Ahmed Hassan,
Muhammad Taha Arif and daughter, Aleena Arif. Their prayers are the roots of my
success.
xi
My thanks are also due to the kindness and love of my Friends specially Irfan
Ahmad Baig, who always encouraged me and played a pivotal role in achieving
higher goals of life.
Cordial love and thanks to my beloved Nieces Dr. Ambreen Pervaiz, Sania Pervaiz
and Rabia Pervaiz, Aysha and Harrm whose hearts beat with golden sentiments,
whose hands always raised for my success.
I have no words of thanks for my Bhabi Farhit-un-Nisa (Late) who always wished
to see me glittering high on the sky of success during her life. May Allah rest her
soul in peace.
(MUHAMMAD ARIF RAZA)
xii
DEDICATED
TO
My Sweet and Beloved parents
And
My Loving and Caring
Family Members
xiii
ABSTRACT
Agriculture is crucial to Pakistan‘s economy and irrigation is the lifeblood of agriculture. The
irrigation system of Pakistan is the largest integrated irrigation network in the world. The state
managed surface irrigation in Punjab had not been performing well and was deteriorating day
by day due to financial, managerial and socio-political factors. Keeping in view the above
discussed problems, the World Bank proposed commercialization and privatizations of the
irrigation system as the only choice for rehabilitation. However after a series of negotiations,
the government of Pakistan agreed upon institutional reforms in water sector of the Punjab.
Consequently, in 1997, Pakistan‘s provincial assemblies passed bills to implement
institutional reforms in the country‘s irrigation sector. In the province of Punjab, institutional
reforms have been introduced in the Lower Chenab Canal (LCC) East irrigation system of the
Punjab as a pilot project through PIDA Act of 1997.
Under these reforms, management at secondary canal level (distributaries) has been handed
over to the Farmers Organizations FOs). The present study was designed to assess the
effectiveness of ongoing irrigation reforms in terms of improving water delivery, operation
and maintenance (O&M) of irrigation system, equity in water distribution and overall
management of irrigation system. It also envisaged the early effects/ impacts of irrigation
reforms on overall agricultural productivity and farm income.
A well represented sample size of 30 distributaries and 360 farm households was selected for
data collection. A multistage sampling technique was used for sample selection. The study
employed two level analysis. At first level, assessment of reforms in LCC East (Reform Area)
was made on the basis of information from the secondary sources taking into account ―Before
and After‖ reform scenario. At second level, assessment of reforms was carried out on the
basis of primary data collected at farm household level. Quantitative analysis was conducted
by making comparison of set of well established indicators developed by secondary
information at distributary level to determine the impact of irrigation reforms on water
charges (Aabiana) collection, operation and maintenance of the system, delivery performance
ratio (DPR) at head and tail of the distributaries etc. A single equation model was used to
capture the impact of irrigation reforms on farm income and productivity. Economic
xiv
Inefficiency model was also estimated to determine the negative impact of irrigation reforms
on inefficiency of the respondents.
The results of the study based on comparison of indicators from primary data showed that
there was an increase in the crop yields. On an overall basis, all the major crops (wheat,
sugarcane and rice) showed an increasing trend in yields. Wheat yield increased by 10
percent, sugarcane by 5 percent and rice by 13 percent respectively. Average gross margin of
wheat, sugarcane and rice increased by 6 percent, 38 percent and 43 percent respectively in
post reform period. The results of the study showed that cost of production of major crops
reduced after reform process. While estimating regression model, Average gross value
Product (GVP/acre) of crops (in real prices) was taken as dependent variable to capture the
effect of reform process, location of the farm along the distributary and important components
of variable cost of production. Similarly, average yield per acre of crops was taken as
dependent variable to determine the impact of reform process. The results of the regression
model for wheat, sugarcane and rice yield showed that F-Value was 7.08, 6.6 and 5.5
respectively, showing that over models were significant at less than 5 percent significance
level. For the estimation of stochastic frontier production function and inefficiency effect
model Cobb-Douglas form of production function and translog were used. The key finding of
the Inefficiency Effects Model was that the dummy for reforms had negative impact on
inefficiency effect for all the crops.
The results of the study showed that Aabiana collection increased from 42 percent to 62
percent in post reform period. The study also showed that delivery performance ratio at the
tail of the distributary increased after introduction of reform process in the province of
Punjab. It was concluded that that the institutional reforms in the irrigation sector have
positive impacts on the yield and productivity of the farmers for all the major crops. It was
also evident that the reforms also have significant impact on the farms located at the tail
clusters of the distributaries.
1
CHAPTER 1
INTRODUCTION
Water is important for human and plant life on the earth. It plays a decisive role in the
sustainable livelihoods of rural people. Approximately 40 percent of the world‘s food supply
is produced on the irrigated land (Johnson III 1995). Improvement in access to water serves as
a powerful tool to diversify livelihoods and reduce exposure for small producers. Irrigation
water creates options for extended production across the year, increases yields and outputs,
and creates employment opportunities. Increased household income may be spent locally thus
helping to stimulate the rural economy. For the last two decades, an ever-increasing number
of countries around the world have been turning over the management authority for irrigation
systems from government agencies to farmers or other non-governmental organizations. This
phenomenon is generally referred to as management transfer or devolution.
Water is precious resource. Only 2.5 percent of the world water is not salty, and of two- third
is locked up in the form of ice caps and glaciers. Due to the continuous hydrological cycle,
about two-third of remaining water is lost to evaporation while, some 20 percent of the
remaining potentially useable water is in areas too remote for human access. After deducting
all the quantities of water which can not be utilized by the human beings (for example excess
water during the monsoon or the flood water), only 0.08 percent of the total water on the
planet is actually utilized by the mankind (Lashari et al. 2003). Agriculture is the largest
consumer of water. Ever increasing population of the earth is putting more pressure on the
agriculture sector to meet the demands of the increased population, especially food
requirements. Irrigation water, the single most important input for the agriculture sector, is
even under more stress as compared to other inputs due to limited supplies. Need for
improvement in efficiency and productivity of irrigation water has become one of the key
issues for the irrigation as well of the agriculture sector. It has been observed that the state
owned irrigation systems have not been performing well and are deteriorating day by day,
especially in developing countries due to financial, managerial and socio-political factors
(Haq 1998). Literature and world experiences on irrigated agriculture have clearly indicated
2
that without integrated approach of water resources that includes irrigation, drainage and
environment, the agricultural productivity and sustainability would not be possible in the
developing countries. The linkages and coordination among all stakeholders of irrigated
agriculture is the most important institutional intervention. The irrigation and drainage sector
plays a vital role in the food supply as well as in the economy of Pakistan.
The Indus Basin Irrigation system of Pakistan is the largest contiguous irrigation system in the
world, serving in excess of 14 million hectares (Johnson III 2004). The system is fed by the
waters of the Indus River and its tributaries. It consists of three major storage reservoirs,
namely, Tarbela and Chashma on River Indus, and Mangla on River Jhelum, with a present
live-storage of about 15.4 BM3 (12.5 MAF), 19 barrages; 12 inter-river link canals and 43
independent irrigation canal commands. The total length of main canals alone is 58,500 Km.
Water courses comprise another 1,621,000 Kms (Shaikh 2004).
The Indus Basin Irrigation System of Pakistan is now facing multiple problems like
deterioration of infrastructure, high conveyance losses and inequitable water distribution both
under normal supply and shortage conditions. There has been chronic inequity with the
upstream water users receiving more water than their due share, while those in the Tail1
reaches of the canal command receiving less. The system is steadily deteriorating and
performing far below user‘s expectations; and there is a great mistrust between the irrigation
department and the users. Some of the causes of ever declining system performance are
inequity in water distribution, poor operation and maintenance (O&M) of the system, poor
water charges recovery (Aabiana), rent-seeking, political interference and weak capacities of
government institutions.
Keeping in view the above mentioned problems, the World Bank proposed that involving the
stakeholders in decision making and operation and maintenance process of the irrigation
system is the only solution for rehabilitation of the existing irrigation system. Consequently,
the government of Pakistan agreed to introduce institutional reforms in the irrigation sector of
provinces. In 1997, Pakistan‘s provincial Assemblies passed bills to implement institutional
1 Last 20 percent portion of the total length of canal, distributary or minor.
3
reforms in the country‘s irrigation sector (Nakashima 1998). In the province of Punjab, the
institutional reforms were introduced through the PIDA Act of 1997.
1.1 A Brief History of Irrigation Reforms in the World
―Until the late 1800s, the bulk of irrigation in the world was developed by users and operated
through a participatory process at the village level. These irrigation systems were developed,
operated, and maintained using local resources largely provided by the water users. Working
together, users made decisions about water allocation, established priorities for repairs and
system expansion and jointly established contributions in cash and kind to be provided by all
who received irrigation and drainage services from the system (Martin et al. 1986).
In contrast, the vast amount of irrigation and drainage development that has occurred since the
early 1900s was quite different. Most of these large-scale irrigation and drainage schemes
were developed by public agencies. Although some early development, such as that in British
India and the Western United States, that were carried out by large agri-business firms, yet
these were gradually shifted to public sources. Various development organizations and
international financial institutions became an increasingly important part of this mix.
Irrigation thus changed from being an activity under the control of local community to a
responsibility of some public organizations. Rather than becoming the responsible parties,
users became passive recipients of the services. Through the 1950s and 1960s, with massive
help from the donor agencies public irrigation development followed a model that excluded
users from active involvement in management.
By the early 1970s, however, it was becoming obvious that this model had created irrigation
systems that were difficult to operate and maintain and were open to extensive rent-seeking,
and becoming less and less sustainable (Repetto 1986). After a period of rapid expansion of
irrigated area from the 1950s to the early 1980s, many governments found it difficult to
finance the recurring costs of irrigation or to collect water charges from the farmers. Centrally
financed bureaucracies tended to lack the capacity to be effective providers of water services
to large number of small farmers. These factors led to rapid deterioration of infrastructure,
shrinkage of area irrigated, maldistribution and wastage of water (Vermillion and Sagardoy
1999). At this stage, a number of irrigation specialists articulated the need for all new
4
paradigm for irrigation development as they recognized that sustainable irrigation systems
require active participation of the users in order to be properly operated and maintained
(Coward and Jr 1987).
Since the mid 1980s, there has been an upsurge in efforts by governments around the world to
transfer management for irrigation systems from government agencies to farmer organizations
or other non-governmental entities. This has occurred more or less in developed and
developing countries. Generally, governments hope that Irrigation Management Transfer
(IMT) will reduce the cost burden of irrigation on the government and will increase
productivity and profitability of irrigated agriculture enough to compensate for any increase in
the cost of irrigation to the farmers.‖
Among the pioneers having experience in irrigation management transfer (IMT) USA,
Mexico, Australia and Turkey are few good examples. Countries, such as Chile, Mexico and
China, are well along in this process. Other countries, such as Indonesia and the Philippines,
and some States in India have embarked on transfer programmes but appear to be bogged
down in problems of implementation. Some countries have transferred small scale systems
and now are considering transferring large scale systems (Vermillion and Sagardoy 1999).
Table 1.1 shows list of countries that have adopted irrigation management policies over the
last 30 years.
Table 1.1: Countries or States Those Have Adopted IMT During the Past 30 Years
Latin America South, South East and
East Asia
Africa& Near East Europe &Central Asia
Brazil, Chile,
Colombia,Dominican
Republic, Ecuador, El
Salvador, Guatemala,
Peru, Mexico
Bangladesh, China,
India, Indonesia, Laos,
Nepal, Pakistan,
Philippines, Sri Lanka
Vietnam
Ethiopia, Ghana,
Jordan, Madagascar,
Mali, Mauritania,
Morocco, Niger,
Nigeria, Senegal,
Somalia, South
Africa, Sudan,
Turkey, Zimbabwe
Albania, Armenia,
Bulgaria, Cyprus,
Georgia,
Kazakhstan,
Macedonia,
Moldova, Romania
Source: Vermillion and Sagardoy 1999
Overtime, irrigated agriculture has increased its importance in the world as a source of food
security, higher farm incomes, and increase in welfare of both rural and urban populations.
5
The development of irrigation in 20th
century played an important role in generating food
surplus that have led to economic development in Asia. Over 60 percent of the world‘s
irrigation is in Asia and since 1965 the irrigated area has almost doubled (Barker 2002). The
world has moved beyond the period when food security was the major goal and construction
of large dams and surface irrigation systems was seen as the major investment needed to
achieve that goal. It is now a challenge to produce more food with less water, to enhance
livelihood and alleviate poverty in the rural areas, and to manage water to protect the
environment and human health. This calls for a new approach to water management.
1.2 Agriculture and Economy of Pakistan
Pakistan is, basically an agricultural country. Agriculture is the backbone of Pakistan‘s
economy. Agriculture is not only the main stay of the populace, directly or indirectly, for
seeking food, clothing employment and perhaps everything, but also viewed as a dominant
way of life. Our culture, habits and attitudes derive their roots, inter alias, from the agriculture
being practiced in this part of world. Therefore, the development of agriculture is synonymous
to the development of the people as well as the country. The importance of water for Pakistan
can not be under-estimated, particularly for irrigated agriculture in the country. In Pakistan,
irrigated agriculture covers 16.2 million hectare (74 percent) out of the total cultivated area of
22 million hectare. Irrigated agriculture uses 97 percent of the available water and provides
over 90 percent of agricultural produce (Shaikh 2004). Agriculture accounts for 20.9 percent
of GDP, and employs 43.4 percent of labor force. It supplies most of the country‘s needed
food grains and also is a source of raw materials for major domestic industries. As such,
agriculture is at the center of the national economic policies and has been designated by the
Government as the engine of national economic growth and poverty reduction. The major
crops grown are cotton, rice and sugarcane in the rainy season and wheat in the dry season.
Cotton alone accounts for 8.6 percent of the value added in agriculture and about 1.8 percent
to GDP (GOP 2007). The former crops are produced in the irrigated areas and the latter is
produced in both irrigated and non-irrigated areas. In the dry season fodder crops are
produced in the irrigated areas because of high importance of livestock sector in Pakistan
(Nakashima 1998). The health of the agricultural sector also has important implications for
poverty reduction and private sector development.
6
There are two principal crop seasons in Pakistan, namely the Kharief, the sowing season of
which begins in April-June and harvesting during October-December; and the Rabi, which
begins in October-December and ends in April-May. Rice, sugarcane, cotton and maize are
major Kharief crops while wheat, gram, lentil tobacco, rapeseed, barley and mustard are Rabi
crops. Major crops, such as, wheat, rice, cotton and sugarcane account for 88.7 percent of the
value added in the major crops. The value added in major crops accounts for 36.3 percent of
the value added in overall agriculture. Thus, the four major crops (wheat, rice, cotton, and
sugarcane), on average, contribute 32.2 percent to the value added in overall agriculture. The
minor crops account for 11.7 percent of the value added in overall agriculture (GOP 2007).
In Pakistan, average rainfall is less than 240 mm a year. In the cultivable plains, rainfall
ranges from about 500 mm a year along the Punjab border with India (which receives some
rainfall from the summer monsoon) to only 100 mm a year in the western parts of Pakistan.
The low precipitation level mean that rain-fed, or barani, agriculture is not possible on a large
scale in Pakistan. Throughout history people have adapted to the low and poorly distributed
rainfall by either living along river banks or by careful management of local water resources
(World Bank 2005). The balance between population and available water already makes
Pakistan one of the most water-stressed countries of the world and with rapid population
growth it will soon enter a condition of absolute water scarcity (World Bank 1994). The
irrigated agriculture of Pakistan mainly depends on Indus River System and its tributaries.
The annual flow of Indus River is 143 MAF out of which 103 MAF is diverted into different
canal commands. Being semi-arid climate of the country, having an annual rainfall of 240
mm, the 90 percent of the irrigated agriculture is being carried out in Indus Plains. The 80
percent flow of the Indus River is generated during monsoon i.e. from June to August, which
necessitates effective water management for sustainability of irrigated agriculture (Qureshi
and Haq 2006). Pakistan‘s agricultural output is closely linked to the supply of irrigation
water. Actual surface water availability in Pakistan against the normal surface water
availability at canal Heads is shown in Table 1.2.
7
Table 1.2: Actual Surface Water Availability in Pakistan
Period Kharief
(MAF)
Percentage
increase/decrease
over the average
Rabi
(MAF)
Percentage
increase/decrease
over the average
Total
(MAF)
2001-02 54.7 -18.4 18.4 -49.5 73.1
2002-03 62.8 -6.4 25 -31.3 87.8
2003-04 65.9 -1.8 31.5 -13.5 97.4
2004-05 59.1 -11.9 23.1 -36.5 82.2
2005-06 70.8 5.5 30.1 -17.3 100.9
2006-07 63.1 -6 31.2 -14.3 94.3
Average system usage
67.1 _ 36.4 _ 103.5
Source: GOP 2007
Above Table 1.2 shows that during the fiscal year of 2006-07, the availability of water for
Kharief 2006 for major crops such as rice, sugarcane and cotton has been 6.0 percent less than
that for the normal supplies and 10.8 percent less than last year‘s Kharief. The water
availability during Rabi season (for major crops such as wheat) was estimated at 31.2 MAF,
which was 14.3 percent less than the normal availability. Thus it is clear that from the above
Table that water availability during the last many years have gone down.
Large part of Pakistan has good soils, abundant sunshine and hardworking farmers. And yet
crop yields, both per hectare and per cubic meter of water, are much lower than international
benchmarks and much lower than in neighboring areas of India. The quality of water service
plays an important role in enhancing productivity through equity and reliability of surface
water.
Pakistan is required to double its annual food production every 15 years in order to maintain
its status quo in meeting requirements of food. This target, on the surface, may not look so
demanding, as the country is bestowed with enough fertile and productive lands and sufficient
freshwater-resources. Despite the availability of these basic resources, unfortunately the
country has to import large quantities of food commodities every year. With the current
population of about 151.55 million people which are growing at the rate of almost 2.6 percent
per annum (GOP 2007), the country would have to feed 120 million additional mouths by the
year 2025 (Kahlown 2005). In such a situation, it is inevitable to keep a balance between
8
production of crops and crop water requirements. Table 1.3 shows the production and water-
requirements of some major crops needed to maintain self-sufficiency in food grains.
Table 1.3: Agricultural Water Demand for Major Crops in Pakistan
Crops Agricultural Water Demand (MAF)
1990 2000 2025
Wheat 26.27 28.8 56.91
Rice 18.78 22.24 56.91
Cotton 13.68 15.71 16.68
Sugarcane 11.35 13.41 19.35
Other Crops 28.93 30.59 46.74
Total with losses @70
percent
168.32 188.28 261.14
Source: GOP 2000
1.3 Irrigation Sector in Punjab
Punjab is the largest province of Pakistan, located in the middle of the country with fertile
irrigated lands. All of its total irrigated area lies in the Indus basin, which is one of the largest
irrigation systems in the world. Irrigation system of Punjab consists of major part of Indus
Basin Irrigation System (PIDA 2005). This system was built up almost a century ago on run
off river basis. This system is considered as a life line for sustainable agriculture in Punjab
which feeds almost whole the country. The irrigation net work of Punjab is a part of Indus
Basin Irrigation System of Pakistan designed for initial cropping intensity ranging from 60 to
80 percent (PIDA 2005). The Punjab irrigation net work comprises of irrigation canals,
drains, tube-wells, small dams and flood protection infrastructures. There are 14 main
barrages on five rivers supplying 120,000 cusecs of water to the fields with 110,000 cusecs
capacity of inter link canals. The colossal net work of over 23000 miles of irrigation channels
provides irrigation facilities to about 8.41 million hectares (PIDA 2005).
Irrigation being a provincial subject, Irrigation Department has historically been responsible
for all water sector activities at provincial level, including administration, planning,
development, operation and maintenance of irrigation and drainage, flood control and
reclamation work. The Irrigation and Power Department is a government body, Headed by the
Secretary to the government of the Punjab, Irrigation and Power Department with Chief
9
Engineers, Superintending Engineers and Executive Engineers to work in the field offices
along with other staff to provide irrigation facilities to the farmers. Present status of irrigation
net work in Punjab is shown in Table 1.4.
Table 1.4: Irrigation Network of Punjab
Head Works/ Barrages
14 No.
Main Canal System
21No.
Length of Main Canals & Branches
6,389 Km.
Distributary & Minors
2,794 Km.
Total off-taking Capacity
120,000 Cfs.
Outlets
50,000 Nos.
GCA/ CCA
23.35/ 20.78 Mac.
Annual Permissible Irrigation
13.96 Mac.
Design Intensity
67 percent
Actual Irrigation
25.50 Mac.
Actual Irrigation Intensity
122 percent
Source: GOP 2005
The irrigation network of the Punjab was initially designed for 67 percent cropping intensity.
But presently actual irrigation intensity has increased to 122 percent, showing that the system
is already overburdened.
1.4 Canal Irrigation System of Punjab
The main objective of any effective canal system is to provide ensured water supply to the
farmers throughout the year without any interruption. As water is scarce resource in Pakistan,
hence the objective was to allocate this resource over a large geographic area on an equitable
basis. The desired pattern of water allocation was to be achieved through design of systems‘
10
structure. The canal irrigation system of Punjab is classified into three distribution level.
These are: main canal level, distributary level and watercourse level (PIDA, 2005).
1.4.1 Main Canal Level
Main canals are the most crucial constituent of the irrigation system for controlled diversion
of irrigation supplies to sustain irrigated agriculture in Punjab. Main canals originate from the
river and are the main supply line for branch canals. In Punjab, there are 21 main canals with
total length of 6389 km, which support all the irrigated area of the province. The effective and
efficient operation and maintenance of these canals is the responsibility of Punjab irrigation
department.
1.4.2 Distributary Level
The distributaries off take from the main canal system. There is a complex network of
distributaries which supply water directly to outlets. The outlets are designed to take water
automatically from the distributary. Irrigation and Power Department of the Punjab has been
responsible for maintaining irrigation water flow in the distributaries.
1.4.3 Watercourse Level
This is tertiary level which directly supplies water to the farmers‘ fields. It is based on the
rotational operation plan, which is controlled on hourly basis to irrigate each farmland. From
the outlets water is directly supplied to watercourses. Farmers are responsible for operation
and maintenance of watercourses at this level. The structure of irrigation network of Punjab is
depicted from the Figure 1.1.
11
Figure.1.1: Schematic Diagram of Canal System in Punjab
Source: GOP 2005
Figure 1.1 shows the structure of irrigation network from Head works to the farmer‘s field.
The major responsibility for irrigation system management rests with the provincial irrigation
department (PID), and some of its elements are with Provincial Agriculture departments. PID
undertakes some construction works, but primarily attend to the operation and maintenance
(O&M) of irrigation facilities, extending from barrages and main canals to outlets, upkeep and
maintenance of drainage and flood works, assessment of water charges and resolution of
conflicts among water users (Haq 1998).
_________________________
Some text in section 1.5 has been derived and reproduced from ―Managing Irrigation for Environmentally
Sustainable Agriculture in Pakistan‖ by Asrar-Ul-Haq (1998).
12
1.5 Objectives of Irrigation System in Punjab
The Irrigation system of the Punjab has the following two kinds of objectives.
1.5.1 Design Stage Objectives
To improve control and command for the acquisition and distribution of
irrigation water.
Optimal allocation and utilization of scarce water resources.
Bringing maximum area under cultivation to benefit the rural population with
available irrigation water.
Partial irrigation with restricted cropping intensities.
Operation of the system with minimum human intervention.
Equitable and proportional distribution of available irrigation supplies.
1.5.2 Operational Objectives
Effective and efficient management of irrigation and drainage infrastructure.
Equitable distribution of available canal supplies at Head, Middle and Tail end
reaches.
Control of illegal water abstractions.
Water resources development.
Control of water logging and salinity.
Revenue generation through efficient assessment of water fees.
Control of environmental degradation of land and water resources.
Resolution of conflicts related to the mutual rights of the share holders.
13
The irrigation performance needs to be reviewed in the context of system design objectives,
operational constraints, institutional systems and the broader socio-economic framework. The
various studies (World Bank 1994, Haq and Shahid 1997, Nakashima 1998, Latif and Pomee
2003) indicated that Pakistan‘s extensive irrigation system has progressively deteriorated
because of inadequate maintenance funding, overstressing of channels to meet an escalating
water demand, and a phenomenal increase in the uses of canal banks by the human, animal
and vehicular traffic. The increased trespassing has been triggered by rapid population
growth, farm mechanization, changing social order, and weakening controls. The system
deterioration is characterized by weak canal banks, eroded brim, channel cuts and breaches,
frequent sedimentation of distributaries and minors and dilapidated condition of canal
structures (Haq 1998). It is, therefore, obvious that the Punjab Irrigation Department (PID)
remained unable to meet the operational objectives.
1.6 Issues of Irrigation Management
Pakistan has significant natural water resources, but they are inadequate for crop production
on the available land. River flows are highly seasonal. Roughly 85 percent of annual flows are
in the Kharief (summer) season and only 15 percent in the Rabi (winter) season. Moreover,
due to inadequate water availability in winter and at the beginning and end of the summer,
cropping intensity is exceptionally low. The stagnant crop yields and increase in the country‘s
population demand enhancement of agricultural production in the irrigated areas. The
agriculture in Pakistan is dependent on irrigation. The system is performing poorly (World
Bank 1994a). The deterioration of the irrigation system is considered the main cause of
stagnant agriculture (Vermillion et al. 1997).The irrigation system of the Punjab is beset with
large number of problems. These were:
1.6.1 Low Irrigation and Water Use Efficiencies
The overall efficiency is the product of conveyance losses, distribution losses and the
application losses. The overall efficiency of Pakistan‘s irrigation system is estimated to range
between 35-40 percent of water from canal Head to the root zone (World Bank 1994). It
implies that for every 100 units of water diverted at the canal Head, only 40 units are
14
available to the root zone. In practical terms, this means that much surface water is currently
lost enroot-water that could be profitably used by farmers. The Indus irrigation system, which
is based on gravity flow, has low use-efficiency. Moreover, it is supply based, and so unable
to accommodate changing water demands during the crop season. Inefficient water delivery
and water use also mean that, in reality, water does not reach users toward the Tail end of the
system.
1.6.2 Poor Aabiana Collection (Water Charges)
Irrigation fees had been sufficient to cover operation and maintenance (O&M) costs until the
irrigation sector started to deteriorate in the 1970s. Due to weakened discipline, the collection
of irrigation fees from the farmers declined, and revenues fell short of government O&M
expenditures. The gap between recoveries and expenditures through water fees was 44 percent
in 1992, which was high and increasing (World Bank, 1994). The current studies also showed
that cost recovery in Punjab before the irrigation management transfer was 44 percent only
(Latif and Pomee 2003).
1.6.3 Inequitable Distribution of Irrigating Water
Water distribution, contrary to the system‘s objective, is not equitable (World Bank 1994).
Equity in water allocation and distribution has many dimensions and levels; inter and intra
canal equity, inter and intra distributary equity, and inter and intra watercourse equity.
Inequity in irrigation water distribution is the most serious problem for the farmers. It has
been observed that inequity in water distribution increases with the increasing length of
distributary. There are three categories of farmers receiving canal water. The categories are;
1) farmers getting more water than their due rights, 2) farmers getting the right share of water,
and 3) farmers suffering from scarcity of water. The categories 2 and 3 are un-satisfied
farmers since the basic system is designed for an inadequate supply of water and even the
right share is not sufficient. According to an official source, 20 percent of the farmers fall in
category 2 and 3 are from the Punjab province. Out of this group, 10 percent are severely hit
farmers, receiving negligible or no water (Nakashima 1998).
15
1.6.4 Inadequate Maintenance of the System
The operation and maintenance (O&M) of the entire irrigation network is one of the major
management issues of the water sector, which starts from the rim station to the farmer‘s fields.
Maintenance of distributaries include desilting, restoration of distributary banks and secure
hoes and rebuilding of regulators and bridges. Such works are carried out by contract unless
they are very small. The extent of canal maintenance is thus determined by the amount of
non-development budget and what remains after paying for the establishment.
In 1996, collapses of two major structures, the out-fall structure of Balloki-Suleimanki-I Link
and regulator-cum- bridge on Marala-Ravi Link, have raised serious concerns about the state
of health of the irrigation network in the province (Haq 1998). The Government has not even
adequately met the requirements of an administered system. It has failed to make budgetary
provisions for operations and maintenance. Moreover, the public body responsible for
irrigation maintenance is separate from the agency responsible for revenue collection. In the
past, administrative discipline was adequate but it has now broken down and the cost of
irrigation maintenance has vastly increased. The issue has become increasingly severe over
the last many years, and is the outcome of a host of factors. Increasing water demand,
deferred maintenance, siltation of distributary prisms, excessive withdrawal by outlets and
illegal water extractions all contribute towards the increasing inequity in the system. This has
eroded the confidence of the Tail users in the integrity of the system.
1.6.5 Poor Irrigation Infrastructure
Most of the water infrastructure is in poor condition. Pakistan is extremely dependent on its
water infrastructure, and it has invested in it massively. Due to a combination of factors such
as age, time neglecting attitude of the department towards repair and maintenance of existing
infrastructure, much of the infrastructure is crumbling. This is true even for some of the major
barrages, which serve millions of hectares and where failure would be catastrophic. There is
no modern Asset Management Plan for any of the major infrastructure (World Bank 2005).
16
1.6.6 Illegal Diversion from Distributaries
There have been increasing incidences of tampering of outlets and other regulation structures,
combined with cuts and breaches at the Head and the Middle of the distributary, which caused
serious distortions in the established operational patterns, thereby impinging adversely on the
system performance. It was found that the farmers at the Tail reaches were badly effecting
from such kind of breach of irrigation water.
1.6.7 Inadequate Farmer’s Participation in Irrigation Management
The performance of the state managed irrigation systems, including the huge integrated Indus
Basin Irrigation System of Pakistan, has been on decline, and is an equal cause of concern for
policy makers, managers and the users. One main reason for this decline is lack of
involvement of users in the management of the system (Mirza et al. 2000). The desired level
and frequency of contact between the irrigation agency staff and the farmers has, however,
diminished overtime. Resultantly, the water theft and water use of canal roads is on the
increase. There has been a growing recognition that any worthwhile improvement may not be
possible without implementing effective programmes for farmer participation in the
prevailing socio-economic environment (Haq 1998).
1.6.8 Corruption and Rent Seeking Behavior of the Officials
Irrigation water is scarce in Pakistan and is distributed to the farmers through administrative
rationing. In response to growing water scarcity, farmers increasingly engage in informal
negotiations and extralegal transactions with irrigation agency officials in order to obtain
more water than their legal quota (Rinaudo 2002). Hence such kind of negotiations may lead
to appropriation of extra canal water by the farmers favorably located at Head end reaches of
canal, at the cost of farmers located at the Tail end reaches of canal.
1.7 Need for Institutional Reforms in Irrigation Sector
The poor functioning of the irrigation system in Pakistan has been a source of concern since
the 1960s and since then it has been the subject of considerable external assistance and
internal policy reforms (Latif and Pomee 2003). However, the need for improving irrigation
17
management has become a major priority on the agenda of most national and international
agencies in the recent past. This was triggered by the declining irrigation performance despite
sizeable investments in the rehabilitation of irrigation infrastructure. Thus learning from the
experience of other nations of the world, it was recognized that the existing irrigation
management organization be changed by shifting the responsibility of management from
government managed agency (PID) to the farmers. Considering the situation, the Government
has introduced institutional reforms in the irrigation sector of the Punjab. One important
aspect of the reforms is irrigation management transfer (IMT) by user‘s participation in the
management of the system called ―Participatory Irrigation Management‖ (PIM).
1.7.1 Irrigation Management Transfer (IMT)
The term ―Irrigation Management Transfer‖ means the relocation of responsibility and
authority for irrigation management from government agencies to non-government
organizations, such as water users associations. It may include all or partial transfer of
management functions. It may be implemented at sub-system levels, such as distributary canal
commands, or for entire irrigation systems or tube-well commands (Vermillion and Sagardoy
1999).
1.7.2 Participatory Irrigation Management (PIM)
The term ―Participatory Irrigation Management‖ normally refers to involvement of water
users in irrigation management along with the government (Vermillion 1999). It is assumed
that the transfer of authority from the government to farmers would result in better
management of irrigation system in terms of dispute resolution, equity in water distribution
and overall management of the system as farmers understand the irrigation problems at farm
level much better than the government staff.
1.8 PIM Model for Punjab
The irrigated agriculture is facing organizational changes worldwide. There is growing
recognition that irrigation water management is a service provided to customers with better
results when operated by decentralized organizations (Lashari 2006). On cognizant of the
problems in irrigated agriculture and water management in the Province, the Government of
18
the Punjab decided to adopt the institutional reforms in the irrigation sector. Hence, in 1997,
the Punjab Provincial Assembly passed the ―Punjab Irrigation and Drainage Act‖. The Punjab
Irrigation and Drainage Authority (PIDA) have been set up at provincial level with the
representation of the farmers and the government representatives.
The main objectives of reforms initiatives as brought out in PIDA Act are as follows.
Implement the strategy of the government of the Punjab for streamlining the Irrigation
and Drainage system.
To replace the existing administrative organization and procedures with more
responsive, efficient and transparent manners.
To achieve economical and effective O&M of the irrigation, drainage and flood
control system.
To make the irrigation and drainage network sustainable on a long term basis; and
Introduce the participation of beneficiaries in the O&M of the irrigation and drainage
systems.
However, the principal objective of reforms in irrigation sector of the Punjab Province is to
reverse the deteriorating performance of its irrigation system and the consequent stagnating
productivity of irrigated agriculture in the Indus Basin. Under the reform model, new
institutions have been designed and implemented on pilot testing basis in Punjab. This is to be
accomplished primarily through two institutional policy initiatives.
One is the restructuring and decentralization of Punjab irrigation department (PID)
into autonomous Punjab Irrigation and Drainage Authorities (PIDA) and Area Water
Boards (AWB).
________________________
Some text in section 1.9 has been derived and reproduced from ―Reforms in Irrigation Sector: Vision,
Implementation, Achievements.‖ Report published by PIDA (2006).
19
The other is to include local irrigation communities in the management of the
irrigation system by developing operational and maintenance as well as fiscal
responsibilities for secondary canal channels through independent, self-sustaining
Farmer Organizations (FO). Thus the institutional frame consists of three entities:
1. Punjab Irrigation and Drainage Authority (PIDA)
2. Area Water Board (AWB)
3. Farmers Organizations (FOs)
1.9 Description and Functions of PIDA, AWB and FOs
The formation and functions of Punjab Irrigation Development Authority (PIDA), Area Water
Board (AWB) and former organizations is given below.
1.9.1 Punjab Irrigation and Drainage Authority (PIDA)
PIDA is an autonomous body which performs the following main functions:
a. To carry out all the functions of Irrigation Wing of the Punjab Irrigation
Department (PID) with independent revenue collection and purchasing authority.
b. Responsible for policy formulation, legal enactment and supervision of the overall
management of Irrigation and Drainage System in the province.
c. To plan, designs construct, operate and maintain the Irrigation and Drainage and
flood control infrastructure located within the territorial jurisdiction.
d. To take measures for reducing O&M expenditures and improving maintenance
planning.
e. To make the Authority financially self-sustaining with regard to O&M cost of
canal irrigation and drainage within a period of 7-10 years.
20
1.9.2 Area Water Board (AWB)
Under PIDA, Area Water Board (AWB) is financially self-sufficient entity at the canal
command level which performs the following functions.
a) Responsible for the irrigation and drainage management of the main canal system,
including bulk water supplies to the Head of the distributaries.
b) Review and monitor the operation and maintenance work plan of the canal.
c) Participate in the operation, implementation and regulation of the rotational
programmed of water distribution in the Area Water Board.
d) Monitor the operation of irrigation systems in the area.
e) Assist the Authority and the government in the formation, promotion and
development of Farmers Organizations
f) Monitor the working of Farmers Organizations
g) Responsible for the management of the irrigation system from barrages to the
distributary Heads, drainage and flood control structures within its territorial
jurisdiction.
1.9.3 Farmers Organizations (FOs)
Under the Area Water Boards, Farmers Organizations have been formed at distributary level.
The overall management of the distributaries has been transferred to these FOs. The FO is the
basic management unit responsible to operate and manage the irrigation and drainage
infrastructure within its jurisdiction.
1.10 Schedule for Formation of FOs
In the Lower Chenab Canal East (LCC East) command area, 85 FOs have been formed having
a total command area of 1564.5 acres. Management transfer to these 85 FOs in pilot Area
Water Board (AWB) was done in three phases of transfer to 20, 49 and 16 FOs respectively
21
during the year 2005. As a first step towards irrigation management transfer in LCC (East), 20
FOs were formed in March 2005 and IMT was transferred to them. The second and third
group of FOs were completed in June and December 2005 respectively. The detail of
formation of FOs is given in Table 1.5 below.
Table1.5: Schedule for Formation of FOs in LCC (East) Irrigation System of the
Punjab
Group of FOs No. of FOs CCA(Acres) Dated for IMT Status
First Group 20 545552 08-03-2005 Completed
Second Group 49 612924 30-06-2005 Completed
Third Group 16 405996 08-12-2005 Completed
Total 85 1564472
Source: PIDA 2005
FOs in general, are performing the following functions.
a) To obtain irrigation water supplies from the main or branch canal at its Head regulator.
b) To manage, operate and maintain the irrigation infrastructure of the channels including
any hydraulic structure according to the approved design.
c) To supply the irrigation water equitably and efficiently to the farmers.
d) To access the water charges and other irrigation related charges to be collected from
the canal water users.
To collect the water charges, fees and other dues from the farmers.
e) To deposit the collected amount of water charges with the Authority after retaining its
share in the collected amount, as agreed with the Authority.
f) To settle water disputes relating to the farmers or other water users of the area.
1.11 Legal Framework and Composition of PIDA
For implementation the reform initiatives, a legal framework has been evolved by the PIDA
which is based on the participation of the farmers at all levels of irrigation management that is
at provincial level, canal command level and at distributary level. This frame work legitimizes
the reforms process and stipulates the conditions for establishing Farmers Organizations and
22
Area Water Boards through Rules and Regulations under PIDA Act. The detail of the legal
frame work is as under:
The Punjab Irrigation and Drainage Authority Act 1997.
The Area Water Board (Rules) 2005.
Farmers Organizations (Rules) 1999, replaced with the new Rules, 2005.
FOs (Election) Regulation, 1999.
FOs (Registration) Regulation, 1999.
FOs (Financial) Regulation 2000.
FOs (Conduct of Business) Regulations 2000.
Irrigation Management Transfer Agreement between FO and AWB/PIDA.
Under this legal framework the existing management of irrigation system has been converted
into multi-tier system for management of irrigation infrastructure.
23
Punjab Irrigation Development Authority (PIDA)
(At Provincial Level)
Chairman
(Minister for Irrigation, Punjab)
Farmer Members: 6
Non-Farmer Members: 5
Chairman P&D Board
Sectary, Irrigation & Power Deptt.
Secretary, Agri. Deptt.
Sectary, Finance Deptt.
Managing Director PIDA
Chairman
(Elected Out of Farmer Members)
Farmer Members: 10
(Elected out of Farmer Organizations):
Non Farmer Members: 9
(Representative of Allied Govt. Deptt.
and Technical Experts
Farmer Organizations (FOs)
(At Distributary Level)
Khal Punchayat
(At Water Course Level)
Chairman: 1
Members: 4
General Body
Consists of Chairmen of
respective Khal
Punchayats
Management Committee
(Elected by General Body)
President: 1
Vice President: 1
Secretary: 1
Treasure: 1
Executive Members:5
(Three from Tail reaches)
Figure 1.2: Flow Diagram Showing the Composition of Punjab Irrigation and
Development Authority (PIDA)
Source: Adapted from PIDA 2005
Area Water Board (AWB)
At Canal Command Level
24
The process of Institutional Reforms commences with community development at village
level to make aware of the farming community about Participatory Irrigation Management
(PIM) initiatives and to organize them for establishment of Farmers Organization. Such
organizations have also been formed at tertiary level i.e. watercourses and are called water
user‘s associations. Thus our farmer community is already familiar with the concept of
community development.
1.12 Need for the Study
As mentioned earlier, the institutional reforms have been introduced in the Punjab Irrigation
Sector through the PIDA act of 1997 and this process of management transformation is
scheduled to be completed by the year 2011 in all Punjab. LCC (East) system of the Punjab
Irrigation is selected as a pilot project for the implementation of the reform process.
Government is investing huge amount for the amelioration of irrigation system. While in-
depth assessment of impacts of these management changes is yet to come, there are
indications from claims made so far that such transformation has brought many benefits
including improved management, lower management costs, improved O&M of the system,
improved water delivery, lower water thefts, reduced conflicts on water and the empowerment
of farmers/beneficiaries that has multiplier effects in community building and quality of life.
However, on the other hand, there have also been concerns on the potential negative effects of
new management institutions especially for small poor farmers. The results of this study will
serve as a role model for implementation and expansion of reforms to other systems in Punjab
and Pakistan.
This study will also provide the researchers/research institutions a very good footing ground
for their future research as this would be the very first quantitative analysis of highly
important relationships between water charges, O&M of the system with water use efficiency
and Head-Tail equity in the scenario of newly designed and implemented water sector
reforms.
25
1.13 Objectives of the Study
The specific objectives of the study were:
1- To assess the effectiveness of ongoing irrigation reforms in terms of improving water
delivery, operation and maintenance (O&M) of irrigation system, equity in water
distribution and overall management of irrigation system.
2- To analyze the early effects/ impacts of PIM reforms on overall agricultural
productivity and farm income.
3- To examine the institutional and financial sustainability of newly created institution
under PIM reforms.
4- To suggest the policy recommendations and guidelines for improvement in PIM
reforms for enhancing their effectiveness, sustainability and overall benefit to society.
1.14 Summary
Agriculture continues to play an important role in the economy of Pakistan. Water is crucial
for our agrarian economy of because overwhelming majority of population is directly or
indirectly dependent on agriculture as a source of livelihood. In addition to it, agricultural
produce is the economic foundation for foreign exchange earnings, making water all the more
important. The existence of gigantic irrigation system and vast networks of irrigation channels
alone, do not necessarily mean that a high level of agricultural productivity is ensured to help
poverty alleviation. The higher level of agricultural productivity, which also is
environmentally sustainable, has many other important ingredients that need to be considered.
Among these, the most important one is reliable and equitable supply of irrigation water to the
farmers in genera, and Tail-enders in particular. Past research in Pakistan shows that a low
level of agricultural productivity is associated with low performance and poor management of
the irrigation system (Mirza et al. 2000). The poor performance of the irrigation system
results in skewed and unreliable water supplies associated with many other problems. This
problem invited the attention of the government to introduce reforms in irrigation sector of
Punjab province consequently, management was transferred to the farmers under the PIDA
Act, 1997.
26
CHAPTER 2
REVIEW OF LITERATURE
The present study is an attempt to measure the performance of the newly created institution
(Punjab Irrigation and Drainage Authority) in the irrigation sector of the Punjab, Pakistan.
The study in hand is the first effort of its kind based on the primary as well as the secondary
data. So far, in Pakistan, very few researchers have made an attempt to directly probe in to the
problem that is why there is a little evidence of comparable results and no empirical literature
directly related to impact assessment of irrigation reforms is available. However, in
international scenario, many studies were conducted to evaluate the performance of Irrigation
Management Transfer (IMT). This chapter is divided in to two sections. Section 2.1 contains
review of irrigation reforms from outside Pakistan and section 2.2 studies from Pakistan.
2.1 Selected Studies from Outside Pakistan
Danial et al. (1980) indicated that the water requirement for enhancing agricultural
productivity could be better achieved at least through five measures. These measures
included: 1) Rehabilitation of existing irrigation projects in terms of modifying the
distribution net work. 2) More intensive operation and maintenance of irrigation
infrastructure. 3) More careful planning of cropping patterns. 4) Greater care in allocation
and scheduling of water both among and within systems. 5) Greater enforcement of irrigation
reforms, rules and regulations in the system.
World Bank (1994) pointed out many problems the irrigation system of Pakistan is
facing. Such problems included, water logging and salinity, over-exploitation of fresh
groundwater, low efficiency in delivery and use, inequitable water distribution, unreliable
water delivery, and insufficient cost recovery. The study also pointed out that due to age,
overuse, and poor maintenance, the efficiency of delivery of the canal system was low,
ranging from 35-40 percent from canal Head to the root zone. It was also observed that,
unless the Pakistan government changes its approach, no future strategy for irrigation and
drainage would succeed. The long-term option for the government would be to define
27
individual water property rights, which were necessary to ensure equity in water distribution.
This would address the problems of Tail-enders, on the one hand, and relieving pressure on
ground water resources, on the other.
Johnson III (1995) studied the process of Irrigation Management Transfer (IMT) in
Indonesia, Colombia, New Zealand and Nigeria. He found that in countries where
governments have had the political will to increase water fees to close to the real O&M cost,
the process of irrigation transfer was smoother. This reflected the fact that the water users
were encouraged to take over management responsibilities in order to reduce water costs. An
equally important inducement for water users to accept additional management responsibility
was better and more dependable delivery services from irrigation agencies.
Turral (1995) discussed responses to the under-performance of irrigated agriculture,
highlighting the changing relationships between the state and water users in the operation and
maintenance of publicly-funded irrigation schemes. The study begun with a summary of the
reasons for and responses to under-performance in the irrigation sector and then outline the
institutional and economic background that favoured local cooperation and coordination in
irrigation management. The focus of the study was on management transfer and privatization
of irrigation, with reference to the broader problems of inter-sectoral water allocation and the
accompanying transformation of the role of state agencies from implementation to service
provision.
Biswas (1996) suggested that water was likely to be one of the most critical resource
issues of the developing world in the early part of the 21st century. He also suggested that a
balanced and sustainable approach to water development was mandatory if the adverse
impacts of the impending water crisis were to be avoided. He was of the view that the issue of
water had been basically ignored in the international agenda in the recent past years. The
perspectives of the North and the South differ on water in certain fundamental ways. It was
examined that the global water requirements were increased almost ten-fold during the 20th
century. The first and the foremost reason was over population, which was expected to be
double to about 10.64 billion by the year 2050. The second reason was increasing water
requirement, which was basically ignored by the water planners in all developing countries.
28
He suggested that it was essential that both the developing and the developed nations should
make a concerted attempt to put water higher up in the international political agenda.
Dick (1997) examined trends in the irrigation policies towards farmer‘s participation
in irrigation management over the past 20 years. Special attention was paid to experiences of
induced participation and management transfer programs in Philippines, Sri Lanka, Pakistan,
Senegal, Columbia Basin USA, and Mexico. This paper provided key lessons related to the
value of social organizers as catalysts, the role of the irrigation agency as partner and the
enabling conditions for farmer‘s participation. As level of income and infrastructure rose,
more formal organizations that enabled farmers to deal with bank accounts, service contracts,
water rights, water markets, and advanced technology in irrigation systems were being
expected. It was also found that the impact of participation on irrigation performance needs to
be evaluated not just in terms of reductions in government costs, but also in terms of
improvement in physical infrastructures and farmers‘ control over water.
Kloezen et al. (1997) undertook the impact assessment of irrigation management
transfer in the Alto Rio Lerma irrigation district in Mexico. The study tested the hypothesis
that in general, Irrigation Management Transfer (IMT) had positive impacts on operational
performance, managerial accountability, O&M budgeting, overall O&M expenditures, cost of
water to farmers and agricultural and economic productivity. The study found that irrigation
management transfer has had very little impact, if any, on surface water allocation and
distribution and the use of groundwater. Changes, if any, in agricultural and economic
productivity and costs to farmers were related to the wider set of neoliberal agricultural and
economic reforms. On the other hand, there was strong evidence indicating that transfer
resulted in improvements in system maintenance and O&M cost recovery.
Vermillion (1997) conducted a study entitled ―Impact of Irrigation Management
Transfer: A Review of the Evidence‖. In this study he synthesized the evidence about the
impacts of irrigation management transfer programmes on the financial viability of irrigation
systems, the quality of irrigation operation and maintenance, the physical sustainability of
irrigation infrastructure, agricultural and economic activity and the environment. He
concluded that the irrigation reforms reduced the costs of the government and eventually the
29
cost of irrigation of the farmers increased. In this way the farmers made the more judicious
use of surface water. In this study, it was reported that the relationship between management
transfer and agricultural and economic productivity was less direct than the relationship
between transfer and O&M performance or financial viability. It was also found that there
was a substantial increase in water fee collection rates in the post-reform period. This increase
in water fee collection ranged from 30 percent to 80 percent. The study also concluded that, in
the year 1992/93 season, following the transfer, 70 percent of distributary canals and 60
percent of field-channel lengths were cleaned by farmer groups. As a result, 10 percent more
wheat and 8 percent more maize were grown in the dry season compared with previous years.
Oza (1998) observed that India had placed a great emphasis on development of the
irrigation sector right from independence. However, the problems of under-utilization, lack of
access by Tail-end farmers, poor maintenance and non-viability of the irrigation systems
persisted in the government owned surface irrigation schemes. It was also investigated that
small, privately owned irrigation systems (dug wells and tube wells) were more efficient and
provided more than 50 percent of irrigation water in India. In fact, the average water rate was
only 3 percent of the estimated net benefit from irrigation. Because of the low water rates and
poor recovery rates, revenue from the irrigation sector covered only 20 percent of the cost of
operation and maintenance, making the sector highly subsidized and non-viable. These
problems in the irrigation sector were more or less found in all states of India. Since water is a
major priority, farmers were constantly looking for alternatives.
Edward et al. (1999) analyzed that dilapidated and poorly maintained water
distribution infrastructure below the canal turnout; the system of farmer managed
watercourses was responsible for a significant proportion of overall system water losses. This
was specially the case in canal command areas where SCARP tube wells had been sited at the
watercourse Heads, but for which no provision for watercourse remodeling had been in
SCARP projects. The resulting water losses, reported are 40 percent or more, contributed to
the growing problems of water logging and were a primary cause for shortage in irrigation
supplies for Tail end farmers. They predicted that the new irrigation system and drainage
would both end the present crisis in the Indus Basin canal systems and brought about
sustainable benefits to farmers, including greater farm output and income. It would guarantee
30
a reliable supply of the minimum amount of water determined by the basic right, provide
opportunities for farmer organizations to plan ahead for purchases of additional water, and
reduce water losses as well as water charges. It was also believed that O&M would be carried
out more efficiently by the farmers.
Samad and Vermillion (1999) examined the impact of the partial management
reforms on the performance of irrigation system. The study was based on the methodology
developed by IWMI to examine the modalities of IMT in different country settings. The
performance of transferred system was compared with that of non transferred system. The
regression model was used to analyze trends in government investment with the annual O&M
costs/ha during the period 1985-95. The results showed that there were decline in
government‘s recurrent costs for irrigation during the period 1985-95 across all the categories
of schemes. The results indicated that there was a statistically significant declining trend in
government expenditure for O&M during the pre-IMT period. In the post-IMT period, there
was a slight reversal in the trend. In this study the regression model was used to analyze
trends in cropping intensities in different groups of schemes. The analysis indicated that there
were no significant differences in trends in cropping intensities in any of the four groups of
schemes in the period before and after transfer. Economic returns per unit of land and water
were measured in terms of gross value output (GVO) per hectare of cultivated land and per
cubic meter of water respectively.
Sarah et al. (1999) addressed the different inefficiencies in the irrigation system of
Turkmenistan. They found that irrigation efficiency was between 25 percent and 30 percent in
the system. Rehabilitation in the existing irrigation system could make substantial water
savings but the cost would be high. Estimates for rehabilitation for Turkmenistan put the
figure at $2000 per hectare which, for the republic as a whole, would be $3.5 billion. It
required a long term investment in the irrigation system of the country.
Wichelns (1999) noted that irrigation water policies could be enhanced by
considering the economic dimensions of farm-level decisions and public goals regarding
limited land and water resources. Policies that modified economic parameters could motivate
farmers to choose crops and irrigation methods that were consistent with public goals. Such
31
policies included water prices or allotments, subsidies for improving irrigation methods and
the removal of output price distortions that favoured crops with large water requirements in
water-short regions. Economic issues regarding water policy in the Nile Valley and Delta
were also included in the paper.
The author also argued that policy reforms could encourage farmers to use irrigation
and drainage resources efficiently by motivating improvements in the water management
practices, while generated revenue for operation, maintenance and capital replacement. The
author suggested that the design of appropriate policies could be enhanced by considering the
economic dimensions of farm-level irrigation decisions and public goals regarding limited
land and water resources. It was also concluded that farm-level costs of water could be
modified to reflect the delivery costs, opportunity costs and off farm effects of irrigation
decisions. Economic analysis of farm- level decisions regarding cropping patterns and
irrigation in Egypt demonstrated the potential role of economic parameters in the selection of
policies to improve resource use.
Facon (2000) argued that the notion of water delivery service and generalized service
orientation of institutions in the irrigation sector has become central in new concepts and
definitions of participatory irrigation management and irrigation management transfer. The
sustainability of the water users associations was depended on their capacity to provide an
adequate water delivery service that allowed the agricultural productivity to increase. In the
context of Asia, diversification of rice crops was a major issue for increased income by
farmers and improved agricultural and water productivity. The author reviewed the
implications of improved agricultural productivity and improved water productivity in terms
of required quality of water service, particularly in the case of rice and diversification of rice-
based farming systems. The author argued that concepts of irrigation management transfer /
participatory irrigation management transfer and modernization of irrigation systems
operation were therefore converging. However, there were still some substantial differences:
the infrastructural physical improvements which must be supported should be designed with a
view to improve equity and reliability of water delivery service and evolve towards increasing
levels of flexibility.
32
Amoah and Gowing (2001) examined a case study of irrigation management transfer
of a rice irrigation scheme in Ghana. The criteria used relate to agricultural, financial and
economic performance and environmental sustainability. The study showed that while both
cropping intensity and cultivated area decreased after transfer, the relatively high yields were
sustained. The average production cost after transfer decreased by about 7 percent from $827
per hectare to $774 per hectare after transfer. Most importantly, average net income increased
by more than 100 percent from $260 per hectare to $549 per hectare after transfer. Very high
financial self-sufficiency ratios coupled with low running cost achieved by the farmers‘ co-
operative led to the conclusion that transfer has resulted in better performance so far.
Dinar (2001) elaborated the implementation of existing concepts and strategies for an
integrated water management in Germany. In his paper, the author highlighted the politics
associated with various water sector management reforms at domestic and international
levels. At domestic level, reforms may include water pricing, water rights, and privatization.
At international level, arrangements may include, for instance, water allocation agreements
and handling of externalities (quantity and quality). Since politics of water reforms did not get
appropriate attention in the design and implementation of the reform process, this paper called
upon inclusion of political consideration in water sector reforms. Moreover, the paper
recognized the strong interactions between domestic and international politics in the water
sector, and suggested comprehensive rather than specific- localized approaches.
Hussain and Biltonen (2001) highlighted that agriculture sector in the Asia and
Pacific region faced the dual challenges of increased food demand and looming water
scarcity. The study found that the overall benefits from irrigation development were largely
skewed and unequal. The study examined that the pro-poor impact of irrigation differed
significantly from one setting to another. The extent of benefits depend on factors such as
land and water distribution, the quality of irrigation and infrastructural management, the
availability of inputs and support services, and water and agricultural policies. The study
suggested that there was no trade-off between equity/poverty and productivity. It was also
found that in settings with greater inequities in land and water distribution, as in India,
Pakistan and Bangladesh, low level of irrigation charges did not necessarily benefit the poor,
and it could be disadvantageous to the poor where charges lead to under-spending on O&M
33
and the system performance suffered. In this study, it was suggested that irrigation
interventions could be designed to re-distribute benefits in favor of the poor.
Barker (2002) presented a framework for examining the evolution of modern
irrigation development in South and Southeast Asia. Dissatisfaction with the performance of
these systems and pressures to reduce government budgets led to a period of irrigation
management reform. The growing scarcity and competition for water was leading rapidly to
the need that water should be treated as a resource and an economic good with a wide range
of uses that could benefit various members and sectors of society. Water must be allocated
equitably across sectors i.e. irrigation, domestic, industrial, and environment sector. There
must be coordination between farm and basin level management of water and between surface
and groundwater management. The impact of irrigation development on environment and
human health must be carefully analyzed.
Koppen et al. (2002) examined the ways to measure relative income poverty within
large scale canal irrigation schemes in India. The author assessed the different impacts of IMT
programmes on poor and non-poor farmers. Two different IMT programmes-the State-Wide
programme under The Andra Pradesh farmers‘ management of irrigation systems Act of 1999
(APFMIS) and the pilot programme under the participatory irrigation management (PIM) in
Gujrat were selected for the study. A sample of seven hundred land owning and tenant
farmers from seven water users associations were selected. The impacts observed were five
years of programme implementation in Gujrat and two years in Andra Pradesh. The results of
the study showed that there was an improved access to canal water after IMT for both farmers
at Middle and Tail end reaches of the command area. On an overall basis the average increase
was 15 percent to 25 percent. It was also found that there was an extension in irrigated area to
about 2 percent (average 0.66 ha).
Yercan et al. (2003) proposed performance criteria for irrigation according to the
situation of before and after irrigation management transfer in Turkey. The selected irrigation
schemes in Gediz river basin were examined and assessed for their physical and economic
performance criteria according to the situation of before and after management transfer
process. The analysis was based on time series; covering a period of 10 years (5 years before
34
transfer and 5 years after transfer). The results of the study showed that the transfer process
affected positively the rate of irrigation performance on overall basis. There was an increasing
tendency from 51-57 percent on an average rate of irrigation in the schemes studied. It was
also found that the rate of water fee collection was between 15-30 percent before transfer
process, and it was higher than 75 percent after the transfer and thus, cost recovery was
improved dramatically. The study concluded that public costs of operation and maintenance
begun to fall and would very likely continue to do so over the next few years. So, the positive
impact of transfer programme was reduction in the cost of irrigation to the government.
However, private costs had increased and would most likely to increase as more and more
responsibility was transferred to local agencies.
Abdelhadi et al. (2004) envisaged that the participatory management approach could
be successfully applied to a huge scheme such as the Gezira with large number of
smallholders under one administration. IMT was implemented as a pilot project and the
results were analyzed taking into consideration the interplay of the main factors in process
and the historical background and the attitude of the tenants. The average yield of four crops
before the project started (1999-2000) and in the first year of the project (2000-01) compared
with the scheme‘s average scheme. It was found that the crop yields, with the exception of
cotton, during 1999 were below the scheme‘s average and reduction was more pronounced for
wheat. The average increase in cotton ranged between 30-45 percent. The average yield of
sorghum increased between 36-310 percent. The cropping intensity was raised from 39-67
percent. The farmers were able to grow some new crops such as sweet potato, potato and
chicken peas in addition to expanding into vegetables. At the same time, there were some
fears about the consequences of immature transfer of responsibilities to farmers committees
without proper and sufficient training.
Penov (2004) determined the institutional alternatives and evaluated their impacts on
the irrigation system of Bulgaria. It was found that the current institutional settings could not
provide for sustainable water usage. The appropriation transactions regarding water were
regulated by a mixture of market (local monopoly) and hierarchy (state price intervention).
Three types of institutional options regarding irrigation in Bulgaria were discussed in the
paper. The first type aimed at improving coordination at the local level. Non-state
35
organization of irrigation water supply was recommended in villages with sufficient social
capital. In this respect, stimulating the development of small water user groups was seen as an
intermediate step toward establishment of water user associations. The second type of option
aimed at limiting the market imperfections (local monopoly). Inclusion of farmers'
representatives in the irrigation company management was recommended as a way of
increasing their bargaining position. However, this option was only feasible in areas with
well-established organizations of farmers. Finally, the third type of options aimed at
strengthening the external conflict resolution and sanctioning mechanisms.
Awosola et al. (2005) made an assessment of the economic, social and financial
performance of the Ogun-Oshun River Basin and Rural Development Authority‘s
(OORBRDA) irrigation projects in Nigeria. A structured questionnaire was used to gather
primary data from 73 participating farmers. Documented primary data on projects‘ activities
from the 1995/96 to 2001/02 seasons were summarized into social, economic and financial
performance indicators. In the Sepeteri project, service payment was enforced at a cost
recovery level of about 96 percent However, the project was not financially viable, because
only 29 percent of the expenditure was covered. Furthermore, the farmers did not have much
stake in determining the project's success with 67 percent social capacity level. The relative
profit level of irrigated cropping was 1:1.13, and did not present sufficient evidence that
farmers demanded (or preferred) irrigated cropping to rain-fed cropping. The project covered
about 50 percent of its total expenditure. The farmers did not have much stake in determining
the project's success, with a 33 percent social capacity level. The relative cropping profit
levels of 1:0.79 showed some evidence of higher demand or preference for irrigated land
compared to rain-fed cropping.
Doukkali (2005) reviewed and evaluated the institutional reforms in Morocco. The
study suggested that, considering their overall thrust and direction, the institutional reforms
undertaken in Morocco were truly remarkable. While these reforms have paved a solid
institutional foundation for promoting an economically responsive water sector, there were
still serious reform gaps, especially in areas such as groundwater regulation and supportive
institutions for irrigated agriculture. The reform experience of Morocco indicated that
although undertaking initial reform could be difficult, subsequent reforms were relatively
36
easier when the political opportunities for reforms provided by both endogenous and
exogenous factors were well exploited.
Hearne and Donoso (2005) offered a review of the institutional reforms in the water
sector in Chile. They analyzed that the factors that motivated institutional changes in Chile‘s
water management including ideology, transaction costs, interest-group behavior and path
dependency. The already observed institutional changes, such as transferable water rights,
water markets and urban water reforms were all significant. Furthermore, reforms were
delayed by the deliberate legislative process required for changes. However, institutional
changes during the period of military rule were rapid, ambitious and favored by private sector
and agrarian interest groups that supported their implementation.
Heyns (2005) highlighted the importance of the realization of political reforms within
a modern democratic framework called for wide ranging reforms in all sectors of the economy
to which the water sector was not an exception. Institutional reforms in the water sector were
undertaken with an overall aim of introducing integrated water resource management as a
durable solution to the water challenges of the arid environment prevailing in Namibia. The
reforms included the development of a new national water policy to manage and regulate
activities in water sector through institutional changes. It was also found that although
institutional reforms in the water sector were necessary to meet the demands of a new nation,
they could not succeed without the required level of skills and capacity both within and
outside water administration. While it was relatively easier to formulate new policies,
promulgated legislation and create new organizations, it was very difficult for an emerging
country to develop quickly the human capacity necessary to handle the reforms, especially
when inadequate funding constraints created a conflict between resource development and
capacity building. As a result, very little progress had been made to address the real needs of
the water sector in Namibia.
Livingston (2005) presented a framework to understand the potential and the need for
change in water institutions. The pressure for institutional change could be analyzed at the
micro and meso levels. At the micro level, the model pointed to how evolving subjective
interests and changing objective realities can be combined to shape the forces for institutional
37
change. At the meso level, the model focused on the probability that pressure at the micro
level would result in actual change. It was observed that the role of political agents and
structure of institutions in the status quo was critical. The concepts of nesting, path
dependency and institutional transaction costs had been used at this level of change.
McKay (2005) reviewed the water institutional reforms in Australia. It was found that
the reforms initiated in 1995 were notable for their comprehensiveness, fiscal incentives and
clear and time bounded targets to be achieved. Although water institutions in Australia had
undergone remarkable changes, however, there were still issues and challenges inherent in
reform process in Australia. It was found that Australia needed further reforms; its recent
reform experience provided considerable insights into the understanding of both the theory
and practice of water institutional reforms.
Saleth and Dinar (2005) examined the conceptual, analytical and theoretical aspects
of water institutional reforms and a synthesis of the main findings from the reform
experiences of six countries: Australia, Chile, Morocco, Namibia, South Africa and Sri Lanka.
Based on the latest developments in the literature on the subject, this paper presented the
analytics of unbundling water institutions to show their endogenous and exogenous linkages,
the transaction cost approach as a diagnostic framework for understanding the role of factors
affecting water institutions, and a stage-based perspective to provide insights into the internal
mechanics and dynamics evident in the process of water institutional change. Using this
analytical framework and theoretical approach, the paper also identified a few practically
relevant principles for reform design and implementation. Based on a review of individual
country reform experiences, the paper also synthesized reform theories with actual practices
by providing anecdotal evidence for various theoretical postulates and practical reform
principles.
Samad (2005) examined the institutional reforms in water sector in Sri Lanka by
comparing those observed during the 1980s with those proposed during the 1990s. It was
found that the earlier reforms focused on the irrigation sector that yielded quicker benefits and
low political risks. The later reforms covered macro institutions and the whole water sector
where the benefits were gradual and less visible but with the high political risks. As the earlier
38
reforms were packaged as part of larger investments, they had built-in incentives and strong
proponents. The recent reforms not only lacked such conditions but also faced an
ideologically charged hostile environment. The public at large remained passive with the real
intent of the proposed reforms especially the issue of creating a public support to the ideology
and politically based opposition.
Smajgl et al. (2005) developed a conceptual framework for water reforms in
Australia, and used an Applied General Equilibrium (AGE) model to investigate the impacts
of potential water reform. The first part of this paper was regarding the application of soft
systems methodology to develop a conceptual model of water reforms for irrigators. The
second part of the paper developed a quantitative model of water reform and presented the
results of scenario analysis performed by the AGE model. Specially, the outcomes of an
important water reform option were shown for economics, social and environmental
objectives. It was assumed that a linear reduction in the use of ground water was necessary to
achieve an optimal ground water level. The results of the study showed that higher levels of
restriction lead to significantly higher efficiency gains but also large welfare decline. A
doubling of groundwater table would lead to an increase of water use efficiency by up to 46
percent while, the highest welfare loss was 25 percent. An increase of the groundwater table
by 33 percent would lead to an efficiency effect of 1 percent and the highest welfare loss was
less than 3 percent. Connected with an increase of the groundwater table by 25 percent was no
effect at all.
Zhovtonog et al. (2005) analyzed that in the beginning of 1990s the period of
transition from a central planning economy to a market economy started in all Central and
Eastern European countries (CEEC). Many common issues in the agriculture and water
management were recognized. In irrigation sector, the majority of CEEC faced similar
problems like decrease in agricultural production, deterioration of irrigation and drainage
infrastructure. The process of reforms in agriculture and water sector in different countries
was initiated with respect to local natural conditions, traditional farm practices as well as
institutional and legislative developments. He studied the reform process and drew some
lessons regarding the IMT. It was found that intensity and shape of the reform process were
different from country to country according to the prevailing economic and political structure.
39
In some countries like the Czech Republic, Germany or Poland, radical restructuring of the
irrigation sector took place, while in others like Solvenia and Macedonia, some governmental
organizations still played an important role in the management of irrigation system. In various
other countries, like Bulgaria, Romania, Russia and Ukraine, the process of reorganization
was just started. He suggested that in order to enhance the irrigation reform process it was
vital to clearly specify the roles, functions and responsibilities of the various actors involved
in the reform process. Depending on the stage of the reform process and on the economic
situation at farm level, governments would have to retreat step by step from involvement in
management and operation and maintenance functions.
Bandaragoda (2006) synthesized studies in five selected Asian countries on their
water policy reform initiatives. Of the five countries, China stands out as the country that has
derived the most from on-going global efforts in promoting water sector institutional reforms
and the concept of integrated water resources management (IWRM). China emerged as the
leader in adapting these concepts to suit the context of the country. Advanced stages of water
development in many parts of the country and increased water shortages due to rapid
economic development have prompted China to forge ahead in the search for institutional
solutions to make the water sector more productive, and the management of water resources
more sustainable. In the other selected countries, efforts to replicate the models of developed
countries without much adaptation and due reference to their stages of development have
generally failed.
Molden et al. (2007) concluded that effective irrigation reforms could provide the
environment for productive and sustainable agriculture that is vital for economic growth and
uplifting lives of the people and keeping them out of poverty. Poorly managed irrigation could
have the similar effect. Irrigation performance assessment was an important management tool
to aid in providing sound service. Performance assessment in irrigation and drainage was the
systematic observation, documentation and interpretation of activities related to irrigate
agriculture with the objective of continuous improvement.
Bandyopadhyay et al. (2007) analyzed the impacts of irrigation management transfer
in Philippine. It was found that the motivation behind IMT was that it would reduce
40
government responsibilities for operation and maintenance and simultaneously increase
farmer‘s supervision over water use. By lowering government expenditures and strengthening
local governance, IMT was expected to have a long-term impact on the country‘s agricultural
and natural resource sectors. With an IMT contract, the water user associations could make
better decisions regarding water delivery and timeliness and could organize themselves to
resolve conflicts and maintain infrastructure. Without local control, associations have to wait
for the national agency to come in and undertake repairs – with IMT they perform repairs as
and when needed. The results of the study also showed that the implementation of IMT was
associated with an increase in maintenance activities undertaken by irrigation associations.
While Irrigation Associations with and without IMT contracts both undertake canal
maintenance, the frequency of maintenance in IMT was higher. IMT areas also had higher
rice yields to the extent of 2 to 6percent relative to non-IMT areas. The analysis also showed
that IMT was associated with a reduction in technical inefficiencies in production. Thus,
increasing local control over water delivery did appear to help with farm productivity.
Madhav (2007) identified the factors contributing to the inefficiency in the canal
irrigation system in India. He found that the persistent under- funding of O&M works had
resulted in rapid deterioration of the network and large conveyance losses. The resources were
thinly spread over a large number of projects, leading to substantial time and cost-overrun. He
also analyzed that poor water management hindered the delivery of adequate, reliable and
equitable irrigation. Farmers in the Head reaches of major and medium irrigation schemes
drew water far in excess of their allocation, and as a consequence, water did not flow into
areas downstream. The solution as recommended to the Andhra Pradesh Government by the
World Bank was to focus on cost recovery and decentralizing irrigation management by
vesting greater powers and responsibilities in water users. It was believed that this would lead
to improvement in quality and cost efficiency of irrigation management.
41
2.2 Selected Studies Related to Pakistan
Radosevich (1975) worked out the role of water users organizations for improving
irrigated agriculture in Pakistan. He concluded that Pakistan has no institutionalized system of
water users associations by which her farmers could jointly pursue optimizing mutual tasks.
Many of the problems faced by the farmers, particularly the small farmers, can be partially
solved by providing them an opportunity to formally organize with their neighbors to increase
agricultural production by improving water management activities. The water user
associations could become the nucleus of mutual on-farm activities pursuing objectives of
equitable distribution of water, resolving disputes, watercourse rehabilitation and collection of
irrigation fees and assessments.
Haq and Shahid (1997) presented a brief overview of Pakistan‘s irrigation system
and the rural society as wells as the major issues facing irrigation management in Punjab.
They summarized the strategies and models proposed by various agencies for participatory
irrigation management and evaluated their main strengths and weaknesses. The authors
mentioned that irrigation water demand in Pakistan increased tremendously over the past
three decades due to agricultural development far beyond the designed parameters. The
system and the supply constraints limited the capacity of the Punjab Irrigation Department
(PID) to respond effectively to the increased water demand. The changing socio-political
situation and general decline in discipline of the society added to the problems of irrigation
management. The paper also discussed major issues and options for improving irrigation
performance. It was recognized that greater farmers‘ participation in irrigation management
was of considerable value and needs to be pursued. The extent and arrangements for farmers‘
participation, however, depend on local environment. The patterns of other countries could
serve as useful models but did not necessarily had universal application. Past experience
suggested that in Pakistan, farmer‘s participation could be done the best through gradual and
phased process.
Latif and Zaman (1998) identified the constraints in the performance of irrigation
systems of Indus Basin with special reference to the performance of the system and equity of
water. The main issues discussed were, reduced capacities of canals and distributaries, poor
42
performance of the system, deteriorating institutional structure and inequitable distribution of
irrigation water. It was found that the annual water supply varied from 248 mm to 398 mm
per unit area against designed value of 437 mm. Moreover, less water was supplied in winter
season than summer. It was also concluded that at canal command level the informal
operational practices and role of operators seemed to be the main reason for inequitable
distribution. At distributary level, illegal abstraction of water and location characteristics was
important factors affecting equitable water distribution.
Nakashima (1998) conducted a study on Pakistan‘s institutional reforms in the
irrigation sector in Punjab province. Based on the past and present experiences in water users
organizations, a few necessary conditions for user governance of irrigation water were
discussed. He found that irrigation fees had been sufficient to cover operation and
maintenance (O&M) costs until the irrigation sector started to deteriorate in the 1970s. Due to
weakened discipline, the collection of irrigation fees from the farmers declined, and revenues
fell short of government O&M expenditures. The gap between recoveries and expenditures
through water fees was 44 percent in 1992, which was high and increasing. Consequential
deterioration of irrigation infrastructure, together with eroding institutions, had brought
irregular water distribution to the canals, resulting in unequal irrigation water distribution
among farmers. Inequity in irrigation water distribution was a major problem for farmers. The
irrigation sector was suffering from these and other similar problems.
Jehangir et al. (1999) found that the government of Pakistan was spending heavily on
the operation and maintenance of the irrigation system yet shortage of funds was a major
reason for deferred maintenance, which threatened the operational integrity of the irrigation
system. The shortfall in O&M funding was estimated to be more than 24 percent in 1993. The
poor O&M had direct effect on the productivity of agriculture; indirectly it affected the whole
economy. The allocation of funds for the increasing O&M costs was becoming a problem for
the Government of Pakistan every successive year. This paper aimed at estimating the level of
operation and maintenance expenditures of the H-4-R Distributary (Bahawal Nagar) and the
situation of the Aabiana collection and the extent of its leakage through different means. The
results of the analysis of Aabiana collected and O&M expenses incurred showed that the
Aabiana collection was almost always in excess to the operational and maintenance expenses
43
incurred on H-4-R Distributary. From the analysis, it was clear that the real cause of low
Aabiana collection was Aabiana under-assessment. So Aabiana assessment procedure should
be improved. It was anticipated that if the activities of Aabiana assessment, collection, and
incurring of O&M expenses were with farmer organization, it could improve since both
would become interdependent. High Aabiana recovery would also lead raise the ability of the
farmer organization to spend more to improve system O&M.
Mirza et al. (2000) concluded that both national and international experience
suggested that the small irrigation schemes, fully managed by the water users could improve
the irrigation system O&M and bring equity in water distribution for all water users. This in
turn could help in increasing crop productivity and environmentally sustainable irrigated
agriculture. However, successful implementation of this process needs political will, legal
framework, independent collective decision making by Farmer Organizations (FOs) and
effective and continuous institutional support from relevant agencies. They also found that the
gradual deterioration of the irrigation systems throughout the world exposed serious
institutional deficiencies and government failure to deliver the services in most water
resources systems. It was also examined that in most of the developing countries financial
crises and inability of government to deliver services was visible. This included lack of
motivation and accountability of agency staff, high levels of political interference and rent
seeking and inadequate concern for needs of water users. On the other hand, without reliable
and equitable supply of water, users were not ready to share the ever increasing costs of
operation and maintenance (O&M). The options left behind were to involve the users of the
system in sharing some of the responsibilities of O&M for sustained and improved
agricultural production.
Rinaudo et al. (2000) investigated the nature of the existing economic rent in public
irrigation systems. The author proposed a conceptual framework to analyze the distribution
between private and public actors involved in irrigation management. A case study was
conducted in Pakistan's Indus basin to quantify the overall level of rent granted to the
irrigation sector and to show how farmers compete for the appropriation of the greatest
possible share of this rent. The study found that rent-seeking activities were not confined to
wealthy and politically influential farmers only but small and medium-size farmers also
44
interfere in water distribution, especially when they were located in the upper reach of the
hydraulic system. It was also highlighted that rents were shared between three types of actors:
politicians, officials of the irrigation agency and water users through a system of
administrative and political corruption. As a consequence, a significant share of the rent
intended to benefit farmers was transferred out of the agricultural sector, which was
detrimental to the level of investment in the sector and to its productivity.
Tahir and Habib (2001) examined the spatial variation in production across canal
commands using gross production indicators i.e. Gross Value of Production (GVP) per unit of
land and GVP per unit of water. Given the data constraints, Punjab province was selected for
the analysis, which consisted of major network of 12 inter-linked and a total of 23 canals out
of 45 canals of Indus Basin Irrigation System (IBIS). The analysis was performed at the canal
command level. In the study, GIS application was developed and validated to convert district
level information into canal command (hydraulic unit) level. Punjab province has 35 districts
and 23 main canals. A big variation in cropping intensities across Punjab canals was shown
by the secondary data, ranging from less than 60 percent to 160 percent, annually. In Kharief
season, GVP per unit of CCA varies with a ratio of 1:10 (Rs. 1,451 per hectare to Rs. 13,836
per hectare), and GVP per unit of CA varied with a ratio of about 1:4 (Rs. 4,368 per hectare to
Rs. 15,649 per hectare). In Rabi season, GVP per unit of CCA varies with a ratio of 1:4 (Rs.
1,566 per hectare to Rs. 6,258 per hectare), and GVP per unit of CCA varied with a ratio of
about 1:2 (Rs. 6,251 per hectare to Rs. 9503 per hectare). In Kharief season, GVP per unit of
water varied with a ratio of 1:6 (Rs. 0.21 per cubic meter to Rs. 1.47 per cubic meter). In Rabi
season, GVP per unit of water varies with a ratio 1:6 of (Rs. 0.39 per cubic meter to Rs. 2.41
per cubic meter). Annual GVP per unit of CCA among 23 canal of Punjab varied with a ratio
of about 1:5 (Rs. 3844 per hectare to Rs. 18326 per hectare). Annual GVP per unit of water
available varied with a ratio of 1:5 (Rs. 0.35 per cubic meter to Rs. 1.57 per cubic meter).
Rinaudo (2002) examined that corruption could determine the allocation of water in a
large public canal irrigation system. The socio-economic characteristics of farmers who
participated in illegal exchanges were analyzed using hydraulic and socio-economic field data
collected from 420 canal outlets of Southern Punjab irrigation system in Pakistan. The
analysis showed that corruption did not only involve economically and politically powerful
45
farmers but it also concerned the lower social segments of rural society. He concluded that
corruption in Pakistan‘s irrigation system arose from the scarcity of a resource, which was
priced far below its marginal value. The under- pricing of water generated a demand for this
resource that exceeded the available quantity in system, thus creating economic incentives for
farmers to resort to illegal means for obtaining more water than their official quota. It was
also concluded that the irrigation agency officials illegally supply more water to farmers who
offered bribe to them. Water corruption without involvement of department officials was quite
difficult.
Latif and Pomee (2003) designed a study to evaluate a farmer-managed distributary
in southern Punjab under the IMT period. As a result of these improvements imparted by the
FO in system management, the extent of irrigated area was increased on an average by 6 to
percent, even under severe drought-like conditions prevailing in the country during those
years. It was also found that cost recovery increased by an amount of 14 percent for summer
and 23 percent for winter growing seasons, respectively, from the irrigated area of the
distributary. Based on the results of the study, it was concluded that there was proportionate
and equitable distribution of irrigation supplies particularly to Tail end water users. The extent
of irrigated area increased, on an average, by 6 percent to percent and the cost recovery by 14
percent to 23 percent during the post transfer summer and winter seasons, respectively.
Lashari et al. (2003) suggested that the participation of beneficiaries (water users)
was one of the best tools to ensure as well as gauge the reliability of water delivery and equity
in water distribution. Based on the results, they concluded that the institutional reforms as
envisaged by ensuring the participation of farmers in irrigation sector was the only option
towards an integrated water resource management where irrigation and drainage function
simultaneously under the umbrella of one institution and active involvement of beneficiaries.
Dinar et al. (2004) developed an approach to assess the political risk associated with
implementation of institutional reforms in the water sector. The approach consisted of a two-
tier process to assess the institutional feasibility of reform implementation. The first tier was a
structured analysis of power distribution among the power groups interested in the outcome of
the reform. The second tier was a Delphi process, reflecting the opinions of experts. The
46
approach was applied to the case of the National Drainage Program Project (NDP) in
Pakistan. Several hypotheses regarding likely progress were tested, using the feedback
provided by a panel of experts in the Delphi process. It was found that reform process was
implemented at slower pace and low level of education among the farmers was a hurdle in
implementing the reform process.
Johnson III et al. (2004) examined the range of institutional options available to
manage the changing irrigation sector of the 21st century. The paper described the various
institutional approaches that were being used and explored the dimensions of the national
contexts within which they were being implemented. Lessons and experiences from Australia,
France, Mali, Mexico and New Zealand were shared by the author. Drawing from experience
in different countries, the paper examined performance as related to the Pakistan‘s‘ context
and organizational framework. The study also provided a guideline for management transfer.
It was perceived that in Pakistan, replacing the Punjab irrigation department (PID) with
Provincial irrigation authority (PIDA), creating Area water board (AWB), formation of
former organizations (FOs) depend upon water charging system and sustainable cost
recovery. It was necessary for the stakeholders to pay water charges to cover the O&M costs.
The author concluded that in Pakistan, the process of irrigation management transfer (IMT)
would be completed in next 20 years and irrigation in Pakistan would be decentralized. It was
suggested that the institutional reform must be the result of a specific policy decision made by
the senior administrators and policymakers in the country. It could not simply be a program
agreed to under pressure from a donor but requires high-level involvement and commitment
from the public also.
World Bank (2005) made an analysis that the water economy of Pakistan depend
fundamentally on a gigantic and complex hydraulic infrastructure system. There were a set of
related challenges which need to be addressed – how to maintain what has been built. It was
also found that many elements of the vast hydraulic system now reached the end of their
designed lives, and have to be re-built. There is an enormous backlog of deferred
maintenance. Most recent irrigation and water supply ―investments‖ from donors, including
the World Bank, had been for the rehabilitation of poorly maintained systems. There was no
systematic Asset Management Plan at either the Federal or Provincial level which described
47
the condition of the assets, the requirements for replacement, rehabilitation (or retirement) and
operations and maintenance of the whole system.
Hussain et al. (2006) reviewed a synthesis of key lessons, messages and examples of
better practices in participatory irrigation management (PIM) in the world. The paper covered
a fairly wide range of countries and regions where PIM reforms were being implemented. The
study covered a wide variety of hydrological and agricultural environments and irrigation
systems including small, medium and large systems, and systems with short and long history
of PIM reform experiences. The study explored the links between irrigation and poverty
alleviation in six Asian countries (India, Pakistan, Bangladesh, China, Vietnam and
Indonesia) with the aim to determine realistic options for increasing returns to poor farmers in
the low productivity irrigated systems. The study was based on primary data collected from
over 5400 rural households covering 26 irrigation systems, supplemented with reliable
secondary data and review of global literature on the subject. One of the main conclusions of
the study was that irrigation did indeed significantly reduced poverty as measured by the
household income. Poverty outside of irrigation systems in nearby non-irrigated settings was
much higher than that within irrigation systems. The author concluded that pro-poor impact of
irrigation differed significantly from one setting to another depending on factors such as land
and water distribution, the quality of irrigation and infrastructural management, the
availability of inputs and support services, and water and irrigation policies. The study also
concluded that inequity and insecurity in access and rights to land and water were bad for
both productivity and poverty. It was also found that in South Asia institutional reforms in
irrigation sector were moving at snail‘s pace and only on limited scale. The study concluded
that, in South Asia, unless irrigation reforms were sharpened with a pro-poor focus the poor
might be bypassed. Irrigation reforms were likely to generate significant benefits for the poor
where land and water were less inequitably distributed.
Lashari (2006) observed the two elements of equity on the basis of results collected in
the pre-transfer period. External equity issues were regarding water allocation and delivery
between different distributaries. Internal equity issues were regarding how water was shared
among watercourses along a canal. He found that in the two canals with favorable water
deliveries at the Head there was no noticeable Head-Tail difference, and all the farmers got at
48
least designed discharge during the peak of the summer season. The third canal which got
close to designed discharge showed a marked disparity between Head and Tail, with Tail-
enders more or less deprived of reliable water. He suggested that greater internal equity of
water distribution be achieved when the internal mechanism was made stronger and farmer
organizations be given a full responsibility for operation and maintenance.
Sarwar (2006) studied the on going reform process in the Province of the Punjab. He
analyzed the role of donors, the influence of existing institutions, the Canal and Drainage Act
and other organizations like PID as well as role of elites with PIDA. He found that all the
agencies had major influence on institutional and organizational change. The conclusions
were based on the historical information about the reforms, the influencing factors before and
after their introduction, the implementation process, and the current functioning of water
management organization and the current situation of reforms. He found that the pressure
from the donors for reforms, particularly from the World Bank, was based on increasing
financial crises and big gap between revenue and expenditure. He mentioned that top-down
strategy was adopted in the introduction of donor driven reforms, regarding the supply
induced economic changes under the pressure from international financing agencies. The
anticipated changes did not yet occur, due to the opposition of the existing organization i.e. the
PID. Different stakeholders especially the staff of old irrigation departments and the elites
were opposing the reforms before their introduction and later were causing the reforms
process to slow down. The study showed that the supply-induced change had been slow and
not successful due to lack of support from the powerful stakeholders. Bureaucratic elites of
PID have been able to significantly affect the reform implementation process to the extent of
change in ‗model‘ with change in personalities. He further added that the newly formed PIDA
(Punjab Irrigation Development Authority) and the reforms were facing many difficulties due
to weak legal framework and bottlenecks in the PIDA Act 1997. If the reforms were
implemented sincerely according to the provisions outlined in the legislation, the process
would go in the right direction. The inclusion of Canal and Drainage Act in the new act added
further complexities for PIDA in the new setup. It was still questionable that the old Act
would be completely drawn out for better functioning of the reformed institutions.
49
2.3 Summary
The main focus of this chapter was to review past studies conducted to determine the impact
of institutional reforms in the irrigation sector of the world as well as of Pakistan. The review
of literature showed that most of the studies were directly or indirectly related with the study
in hand. Each study had specific features on the basis of objectives and methodology adopted;
however, the findings were more or less same. The national and international experience
reflected that impacts of irrigation reforms in the world had both positive and negative
implications for small, medium and large farmers. It was clearly reflected that equity in water
distribution was achieved by the involvement of farmers in the decision process. However, it
differed significantly from one setting to another depending on factors such as land and water
distribution, the quality of irrigation and infrastructural management, the availability of inputs
and support services, and water and irrigation policies. In some countries like Mexico, USA,
Turkey and Egypt, the reform process remained successful.
As for Pakistan, many studies indicated that more than a century old irrigation network
supervised by the Punjab Irrigation Department had not been working according to the
satisfaction of the stakeholders. It remained unable to reduce the Head-Tail equity in the
distribution of water. The gap between O&M expenditures and cost recovery was increasing
with the passage of time. Consequently, the government of Pakistan introduced institutional
restructuring in the irrigation sector of the Punjab. These studies found that the IMT, in
general have both positive as well as negative implications on operational performance,
managerial accountability, operation and maintenance of the system, cost of irrigation water
to the farmers, and agricultural productivity.
50
CHAPTER 3 METHODOLOGY
Analytical study comprises of systematic and appropriate techniques for analysis. The
selection of sample, data source and methodology is important to analyze, verify and describe
the relationships. The finding and analysis of data comprising qualitative and quantitative
variables need in-depth interpretation. The data presentation and dissemination leads to
successful completion of the study. Keeping in view the objectives of the study, following
research methodology was adopted.
3.1 Selection of Study Area
The area of Lower Chenab Canal LCC (East) was selected as study area. The process of
irrigation reforms was initiated as a pilot project and completed in the area of LCC (East). As
the main objective of study was to assess the impact of the irrigation reforms in terms of
improving water delivery, equity in water distribution and overall agricultural productivity.
Therefore, based on the above consideration and to meet the objectives of the study, the LCC
(East) was selected as a case study on account of the reason that first phase of reform process
was initiated and completed on this canal division. It has four canal divisions. These are:
1. Khanki Division
2. Upper Gogera Division
3. Lower Gogera Division, and
4. Burala Division
The LCC (East) irrigates the fertile land of Gujranwala, Hafizabad, Sheikhupura, Faisalabad,
and Jhang districts.
3.2 Characteristics of the Study Area
The characteristics of the study area are divided in to two parts. In the first part a brief
description of the districts irrigated by LCC (East) is given, and in the second part specific
characteristics of the LCC (East) are given.
51
3.2.1 A Brief Description of the Districts in the Study Area
A brief description of the districts which are being irrigated by LCC (East) and where
irrigation management has been transferred to the stakeholders is given below. The districts
include Hafizabad, Gujranwala, , Sheikhupura, Faisalabad, and Jhang.
Hafizabad District
Hafizabad was given the status of district in July 1993. It is bounded on the north by Sialkot
district, on the east by Gujranwala district, on the south by Jhang and Faisalabad districts and
on the west by Sargodha district. The total area of the Hafizabad district is 2,367 square
kilometers, of which 80 percent is under cultivation. It is generally plain and there is no hilly
area in the district. The river Chenab passes through the district of Hafizabad. There are some
marshy areas along the river side. Nature of land is sandy and clay-loam. The entire area is
irrigated through canals and tube wells. The sub-soil water is almost sweet (fit for drinking
and irrigation). The LCC (East) irrigates the whole district through branch canals.
The climate of the district is hot for most of the year. The average annual rainfall in the
district is 790 millimeters (GOP 1998). During summer, temperature goes up to 45 degree
centigrade with minimum 20 degree centigrade. The average rainfall per month ranges from
50 to 75 millimeters. The major crops of the district are rice, wheat and sugarcane. The best
variety of Basmati rice (Karnal) is cultivated in the fertile lands of Hafizabad district.
Gujranwala District
The district of Gujranwala is located at 700 feet above the sea-level. It has an area of about
5,988 square kilometers (GOP 1998). The district is bounded on the north-west by Gujarat
district, on the east by Sialkot district, on the south by Sheikhupura and Faisalabad districts
and on the west by Jhang and Sargodha districts. The district has extremes of climate. The
summer season begins in April and continues still September. May, June and July are the
hottest months. The mean maximum and minimum temperatures during this period are 39°C
and 26°C respectively. The winter season begins in November and continues till March.
December, January and February are the coldest months. The maximum and minimum
temperatures during this period are 20°C and 7°C (GOP 1998). The monsoon usually begins
52
in July and continues with usual muggy spell until the middle of September. It is one of the
most prominent districts of rice growing areas in Pakistan. Major crops grown in this district
are rice, wheat and sugarcane. In some areas vegetables and fruits are also grown.
Faisalabad District
Population wise, Faisalabad is the third largest district of Pakistan and second largest of
Punjab. It is bounded in the North by Gujranwala and Sheikhupura Districts, in the East by
Sheikhupura and Sahiwal districts, in the south by Sahiwal and Toba Tek Singh districts and
in the West by Jhang district. It has an area of 5,856 sq. km. and population of approximately
3.54 million people, (GOP 1998). This district consists of six sub-divisions and Faisalabad
city is the districts Headquarter. Out of total area of 1.44 million acres of this district, 1.15
million acres of land are irrigated through canals (GOP 1998). Rest of the area is either Barani
(Rainfed) or cultivated through tube wells. The district has flat alluvial plain formed by
Chenab and Ravi rivers. The river Ravi flows along the south eastern boundary of the district.
The land close to the river is relatively lower than that away from the river towards the west.
The climate of the district touches two extremes. The maximum temperature in summer
reaches up to 50°C or 122°F. In winter, it may, at times fall below the freezing point. The
mean maximum and minimum temperature in summer are 39 and 27°C respectively and in
winter, 21 and 6°C respectively (GOP 1998). The summer season starts from April and
continues until October. May, June and July are the hottest months. The winter season, on the
other hand, starts from November and continues until March. December, January and
February are the coldest months. Faisalabad district is un-parallel for its agricultural
productivity. Major crops grown are wheat, rice and sugarcane and in some areas cotton is
also cultivated.
Sheikhupura District
Sheikhupura city is located 32 km west of Lahore, and is linked to the other regional centers
of Punjab. Sheikhupura district is bounded on the north by Gujranwala and Sialkot districts,
on the east by Lahore and Kasur districts, on the south by Sahiwal and on the west by
Faisalabad. The Sheikhupura district is spread over an area of 5,960 square kilometers (GOP
53
1998) with a population of more than a million people. District has extreme climate; the
summer season starts from April and continues until October. During the summer season,
temperature ranges from 30 to 45 degrees Celsius. The winter season starts from November
and continues until March. December and January are the coldest months with a mean
minimum temperature of 5 degrees. The rainfall is 500 mm per year. The mean minimum and
maximum humidity during winter is 37 percent and 84 percent, respectively. There is a
network of canals in the district for irrigation. The Upper Chenab Canal and Lower Chenab
Canals are two major perennial canals, which supply water for irrigation. Major crops are rice,
wheat and sugarcane. Sheikhupura is famous for Basmati rice production in the world.
Jhang District
The district of Jhang is bounded on the north by Sargodha and Gujranwala districts, on the
west by Mianwali, on the south by Multan and on the east by Faisalabad. The surface of the
district presents three levels: on the extreme west are the high dunes of the desert Thal; in the
center are the two low-lying valleys and on the extreme east is a portion of the old Sandalbar.
The rivers Chenab and Jhelum pass through the district. The climate of the district is hot in
summer and cold in winter. The maximum and minimum mean temperature in summer is 45.5
degree centigrades and 29 degree centigrades and dust storms are common during the summer
season. The main crops are wheat, cotton, sugarcane and grams. Land is irrigated by canals
and tube wells. The total area irrigated by all sources is 126 thousand acres (GOP 1998).
3.2.2 Specific Characteristics of the LCC (East)
Lower Chenab Canal (East) off takes from the left flank of Khanki Headwork which is
constructed at the river Chenab. It was constructed during the last decade of 18th century and is
about 114 old years. The designed cropping intensity of the system is 40 percent to 75 percent
with water allowance of 1.52 Cs/1000 to 2.85 Cs/1000 acres. The cropping intensity has increased
to 135 percent. The system irrigates about 3.04 million acres (123 million hectares) of cultureable
command area located in the system (GOP 2006). Table 3.1 shows the salient features of the LCC
(East) irrigation system of the Punjab.
54
Table 3.1: Salient Features of the Selected Irrigation System
Canal
Division
Total
Area
(Acres)
Total
GCA
(Acres)
Total
CCA
(Acres)
Discharge
Major Crops
Distributary
No of
Outlets Sanctioned
(Cusec)
Proposed
(Cusec) No
Length
(canal
mile)
Khanki 296,181 274,016 239,658 2076 2584 Wheat, Rice 25 144.07 598
Upper
Gogera
701,133 655,998 574,095 1965 3192 Wheat, rice
sugarcane 26 307.54 1182
Lower
Gogera
679,140 630,631 529,598 1870 2708
Wheat,sugarcane,
cotton, sesamum,
maize
39 283.71 1087
Burala 586,298 561,007 504,539 2338 2745
Wheat,
sugarcane,
cotton, sesamum,
maize
34 286.22 1034
Total
LCC(E)
262,752 2,121,652 1,857,890 8249 11,229 124 1,025.4 3892
Source: GOP 2005
In the study area, Khanki canal division is the smallest and Upper Gogera canal division is the
largest with respect to CCA. Total GCA of the system is 262,752 acres and total number of
distributaries is 124 with total length of 1,025.4 miles. This system irrigates the major portion
of central Punjab.
The cropping patterns in all the four canal divisions differ from each other. In Khanki division
the cropping pattern followed was wheat-rice-wheat. In Upper Gogera cropping pattern
followed was wheat- rice- sugarcane. Whereas, in Lower Gogera division mixed cropping
pattern was found. Here the cropping pattern adopted was wheat-sugarcane-cotton-sasamum-
maize. This division touches the cotton belt therefore; the farmers adjoining to this belt grow
more cotton as compared to farmers far from this belt. In Burala division the cropping pattern
was similar to Lower Gogera with slight difference in area under cotton crop. Table 3.2 shows
the list of distributaries which were visited by the respondent for collection of data.
Table 3.2 represents the sampled distributaries from all the four canal divisions along with
their CCA. This sample represented more or less small, medium and large distributaries at
Head, Middle and Tail of the main canals. These distributaries were physically visited by the
55
Table 3.2: Sampled Distributaries and their Characteristics
Distributary CCA
(hectares)
Length
(miles)
No. of water
course
Designed capacity
(cusecs)
Channi 1517 7 17 18
Hafizabad 5383 33 26 52
Jurrian 6548 13 49 31
Gajjiana 13205 26.23 93 64
Laggar 6656 15 38 33
Shahkot 19394 31 164 102
Ghordor 8276 14 48 45
Sharqpur 37963 101 240 209
High Level 8310 35 94 58
Kabbarwala 3266 3.6 20 10
Annuana 3729 3.53 21 17
Buchiana 2254 6.5 17 12
Khanuana 12694 25 105 64
Pauliani 12449 22 102 64
Moungi 19202 43 162 96
Awagat 6973 14 59 40
Khewwala 2280 5 12 9
Buttiwala 2872 3 22 13
Terkhani 31340 46 261 162
Jaranwala 1813 12 14 11
Jassuana 5471 14 44 54
Kaluaana 2335 3.26 11 19.82
Rassiana 6519 13 54 34
Korukhatwan 2099 14 12 11.62
Talyara 1057 3.3 12 20
Satyana 1545 3.60 8 29.71
Tandianwala 27444 42 385 136
Killianwala 29855 53 212 147
Munianwala 2215 7 19 18
Bhalak 22552 49 154 123
Source: GOP 2006
56
researcher and examined their existing condition after the implementation of the irrigation
management transfer. The physical condition of the sampled distributaries was determined by
interviewing the farmers in the area.
3.3 Sampling Framework
Statistically well designed sampling procedure is of prime importance to drive the needed
inferences. A multistage sampling procedure was adopted in selecting the sample. Study area
was initially selected purposively on the ground that irrigation management was transferred in
the area of LCC (East). At the second stage, study area was stratified based on the following
criteria;
First batch of the distributaries where irrigation management was transferred in
March 2005 was included in the sample.
Second batch of the distributaries where irrigation management was transferred in
June 2005 was included in the sample.
Third batch of the distributaries where irrigation management was transferred in
December 2005 was not included in the sample because the rights of the IMT were
transferred to third batch just before the time of data collection process.
By using simple random sampling technique, 15 distributaries were selected from each first
and second batch of the distributaries making a sample of 30 distributaries in the reform area.
It covered first level analysis. Out of these 30 distributaries, 10 distributaries were selected by
using purposive random sampling technique on the basis of homogeneous characteristics. At
this stage, purposive sampling technique was adopted to second cover the entire system from
Head of the main canal to the Tail. The sampled distributaries represented Head2, Middle
3 and
Tail of the command area. Out of the 10 selected distributaries, 6 watercourses per
distributary (2 each from Head, Middle and Tail) were selected through stratified random
sampling giving a sum of 60 watercourses. From these 60 watercourses, 6 farm household
2 First 40 percent portion of total length of canal or distributary.
3 Middle portion of canal or distributary comprising of 40 percent of total length.
57
were randomly selected. A sample size of 360 farmers was collected for level analysis. The
schematic diagram of sampling procedure is shown in Figure 3.1.
58
Punjab Irrigation System
Purposive Sampling
Simple Random Sampling
Purposive Sampling to cover the entire
system from Head, Middle & Tail
Two each from Head, Middle & Tail
through Stratified Random Sampling
6 HH from each water course by SRS
Figure 3.1: Sampling Framework and Design
Reform Area
LCC (East)
Non-Reform Area
LCC (West)
1st Batch
(20 Disty
2nd
Batch
(49 Disty 3
rd Batch
(16 Disty
First Level Analysis 30 Distributaries
15 Distributaries 15 Distributaries
10 Distributaries
60 Water Courses
360 Farm Households
Second Level Analysis
59
3.3.1 Sample Size
A representative sample of 30 distributaries and 360 farm households were selected for
collecting primary information from the field and secondary information from the FOs and
irrigation department offices. The distributaries and farmers were taken from all the districts
which fall in the area of Lower Chenab Canal (East) i.e. reform area.
3.3.2 Questionnaire Development
A comprehensive questionnaire was constructed and pre-tested in order to make necessary
changes regarding variation in situation across different selected distributaries. After
necessary amendments, the final version of the questionnaire was developed. The
questionnaire consisted of four modules arranged as follows.
Basic information module:
This module was designed to gather information about the household, such as
household members, their ages, schooling, sources of income, employment, non-
farm income, area owned, and area rented-in, area rented-out and land rent in the
area.
Irrigation infrastructure module:
This module gathered information on sources of irrigation water, number of
irrigations, type of outlets (mogah), cultivated area, operation and maintenance of
irrigation infrastructure and overall condition of the distributary.
Agricultural production module:
This module obtained information on the farming situation before and after the
implementation of the irrigation management, cost and value of agricultural
production
60
Retrospective questions module
This module was designed to obtain historical information over the last four years (before and
after IMT) on availability of surface water, water delivery problems at their turn, percentage
of ground water used, quality of ground water, crop yields, and production of major crops and
irrigation related problems.
The questionnaire was carefully edited to frame the questions to suit the local context, in so
far as units of measurement, local connotations, or other common usage of phrases or words
were concerned. This made the questionnaire easier to understand by the respondents.
3.3.3 Pre-Testing of the Questionnaire
Keeping in view the objectives of the study, pre-testing was under taken. Information such as
the clarity of the questions, length of time required to complete a questionnaire, quality of the
answers, relevancy of the questions and logistical requirements were gathered during the pre-
testing period. A general review of the questionnaire was conducted and necessary changes
based on pre-testing were incorporated.
3.4 Data Sources and Collection
Two types of data were collected.
Primary data
Secondary data
3.4.1 Primary Data Collection
For collection of primary data, the researcher interviewed the respondents personally at their
farms. Although questionnaire was constructed in English, yet the questions were asked in
their local language (Punjabi) for the convenience of interviewees to get the required
information with maximum accuracy. While interviewing, researcher tried his best to
maintain informal and friendly atmosphere in order to obtain the data from the respondents.
Records of irrigation activities and transaction records for the years 2004-05 and 2006-07
61
growing seasons were also collected. Fieldwork in the community was the ultimate focus of
study as farmers were the most important stakeholders in the process.
3.4.2 Secondary Data Collection
Various secondary data sets were also collected in order to support the results on primary data
set. Pre-IMT data for two years (2003 and 2004) was collected from the Punjab Irrigation
Department (PID). Data included assessment of water charges, collection of water charges,
O&M expenditures, non-development expenditures, Head-Tail equity in water distribution
and disputes reported from Head and Tail end reaches. Information regarding design
discharge of distributary at the Head was also collected. Post-IMT data for the last two years
(2005 and 2006) was collected from the FOs working in the field
In this research, selected distributaries in the LCC (East) were examined and assessed for their
physical and economic performance before and after management transfer. The research work
was based on:
a) Literature Review,
b) Interviews with different stake holders including
(i) FO representatives particularly President and members of management
committee
(ii) Officials of Punjab Irrigation Department (PID) and Punjab Irrigation and
Drainage Authority (PIDA)
(iii) Farmers of the study area
(iv) Irrigation management experts from different walks of life
Literature review contributed towards the understanding of the existing irrigation system and
the proposed institutional reforms. Interviews with the FO representatives and relevant people
from the government and relevant institutions enhanced the understandings of the irrigation
system management and broadened the vision regarding the irrigation reforms.
62
3.5 Data Entry
The questionnaires were edited every day before moving on to the next day. In this way a
quality data of 30 distributaries and about 360 respondents was collected. For data entry, a
format was prepared on the Microsoft Excel work sheet. It was also required to convert data
recorded in different units in the questionnaire to standard units prior to entering in the
database. All the data was carefully entered.
3.6 Cleaning and Organization of Data
Cleaning is the integral part of the data management process before using it in the final
analysis. The entered data was examined for errors or bad entry, missing values and zeros.
Errors identified were immediately corrected. The data were also examined cell by cell to
detect any error. The data base in Excel was converted into Statistical Package for Social
Scientists (SPSS) format and further cleaning was undertaken. Tables were generated for all
variables and were examined for outliers, errors in coding, as well as other errors. Variables
with such errors were sorted and the case number identified and doubtful cases were verified
by checking back with the questionnaire. Subsequently each data file was examined by
individual row or column to detect any error across variables or within a variable.
3.7 Indicators for Performance Measurement
Following is a list of indicators that were used to assess the outcome and impacts of Irrigation
Management Transfer (IMT) in the study area. The rationale for adopting IMT is that it would
improve quality and efficiency of irrigation system, to improve Head-Tail equity, better
financial and physical sustainability of irrigation systems, and increased productivity and
profitability of irrigated agriculture. A well defined set of indicators was carefully recorded.
63
Table 3.3: Broader Category of Indicators Used at Various Levels in the Study
Broad category of
indicators
Level of
study Indicators
Physical indicators At both levels Condition of the system, water delivery, ground
water contribution
Institutional Indicators At level first
Presence of new institutions, working of new
institutions, user participation, impact of new
institutions, dispute settlement
Productivity indicators At level two cropping intensity and pattern, cropped area,
GVP, gross Margin, net income of farm
Sustainability
Indicators At both levels
O&M expenditure, water charges collection,
operational expenditures
Socio-economic
Indicators At both level
Age, education, income level
Equity At level first Delivery performance ratio, Head-Tail equity
Institutional Indicators At Second
level
Capacity building and training of FO
representatives. dispute settlement
Source: Hussain and Biltonen 2001
Few of the above indicators are explained below while others (average yield, gross value
product, average cost of production, variable cost of production, gross margins, Aabiana
assessment, Aabiana collection, per hectare O&M expenditures, per hectare salary
expenditures) are self explanatory.
64
1. Equity in Water Delivery
a. Delivery Performance Ratio (DPR):
Observed discharge at Head
DPR (H) = -----------------------------------
Sanctioned discharge at Head
The ratio should be closer to 1.
Observed discharge at Tail
DPR (T) = -----------------------------------
Sanctioned discharge at Tail
Ideally the ratio should be closer to 1.
DPR at Head
b. Head-Tail Equity: = ---------------------
DPR at Tail
c. Tail End Supply Ratio (TSR):
No. of days sufficient water supply reached at the end of the distributary
TSR = ----------------------------------------------------------------------------------------
Total number of days.
Ideally the ratio should be closer to 1.
d. Head-Tail Equity in output
2 Financial Sustainability
a. Aabiana collection performance:
Total amount of Aabiana collected in a year
= ----------------------------------------------------------
Total amount of Aabiana assessed in a year
b. Gap between Aabiana assessment and Actual collection
3. Operation and Maintenance (O&M)
a. O&M expenditures per acre of CCA
65
4. Institutional/Capacity Building
a. Trainings obtained by the FO Executive Body
b. Conducted regular General Body meetings
c. Conducted regular Executive Body meetings
d. Participation rate in General Body meetings
e. Participation rate in Executive Body meetings
f. Preparation of Annual Development and Maintenance Plan
5. Dispute Settlement
Total Number of disputes settled
Ratio of disputes: = --------------------------------------------
Total Number of disputes registered
6. Productivity
a. Out put per unit area at Head, Middle and Tail (Rs. Per acre):
Total annual value of agricultural production
= ----------------------------------------------------------
Total command area
b. Out put per unit irrigated area (Rs. Per acre):
Total value of agricultural production
= -----------------------------------------------------------
Total annual irrigated cropped area
c. Cropping Intensity:
Gross cultivated area in a year
= ----------------------------------------------------------
Gross cultureable area
66
3.8 Summary
This chapter comprises of different methods and techniques used for data collection from the
study area. The area of Lower Chenab Canal East LCC (East) was selected for study purpose.
As the reform process in Punjab was initiated in the area of LCC (East) therefore the same
area was selected for study purpose. The study area was consisted of four canal divisions,
namely, Khanki, Lower Gogera, Upper Gogera and Burala. The whole study area fell in the
districts of Gujranwala, Hafizabad, Sheikhupura, Faisalabad and Jhang. A multi-stage
sampling technique was use for data collection from the field. A well represented sample of
30 distributaries and 360 farm households were selected for collecting primary information
from the field and secondary information from the FOs and Irrigation department offices.
67
CHAPTER 4
CEPTUAL FRAMEWORK
In this chapter, conceptual frameworks underlying the analytical techniques used in the study
have been discussed. This chapter is divided in to three parts:
A. Single Equation Estimation: OLS Approach
B. Concept Regarding Production Function
C. Efficiency Concepts and Measurement
4.1. Single Equation Estimation: OLS Approach
In this study different single equation models have been estimated to capture the impact of
irrigation reforms on farm income and farm productivity. Models were estimated by using
Multiple Regression models through Ordinary Least Square (OLS) estimation procedures.
Multiple Regression was used to account for (predict) the variance in an interval dependent,
based on linear combinations of interval, dichotomous, or dummy independent variables.
4.1.1 Assumptions of OLS
Ordinary Least Square estimates require certain assumptions to be met. These
assumptions have been described by Gujarati (2003) and Green (2005) and are discussed
below.
(i) The first assumption is about Measurement: Two types of variables are included in
OLS i.e. dependent and independent variables. All Independent Variables may be in
the form of interval, ratio, or dichotomous, whereas the Dependent Variable can be
continuous, unbounded, and estimated on an interval or ratio scale. All variables are
estimated without error.
(ii) Second assumption is regarding Specification: All predictors having impact on
dependent variable are included in the analysis whereas irrelevant predictors of the
68
dependent variable are excluded from the analysis. The form of the relationship is
linear (allowing for transformations of dependent variables or independent variables).
Expected value of error: E(ε) = 0.
(iii) Next assumption is about the normality of errors: It is assumed that the errors are
normally distributed for each set of values of the independent variables.
(iv) We assume that there is no correlation between the error terms and the
independent variables: The error terms are uncorrelated with the independent
variables. i.e. E (εi, xi) = 0
(v) An important assumption of OLS is that, there does not exist perfect multicollinearity.
None of the independent variables is a perfect linear combination of the other
independent variables in multiple regressions. If there is only one predictor,
multicollinearity is not an issue. This condition is also known as Full rank (Green
2002).
(vi) Liner relationship exists between dependent variables and independent variables
yi = xilβi + xi2β2 + .....+ xiKβK + εi
The given model specifies a linear relationship between y and X\, . . . , XK
(vii) It is assumed that independent variables are exogenous:
E [εi | Xjl, Xj2, . . . , XJK] = 0
This says that the expected value of the disturbance in the sample is not a
function of the independent variables observed at any observation.
(viii) Homoscedasticity and non-autocorrelation are other important assumptions of OLS.
Each disturbance, εi has the same finite variance,σ2 which is uncorrelated with every
other disturbance, εj.
(ix) Another assumption is about exogenously generated data. The data in (XJ1 Xj2, . . . XJK)
may contain a mixture of constants and random variables. The process generating the
69
data operates outside the assumptions of the model—that is, independently of the
process that generates εi.
(x) Last assumption is that the disturbances are normally distributed.
4.1.2 General Form of the Model
The multiple regression equation can be written in the following form:
y = c + b1x1 + b2X2 + ... + bnxn
The ‗b's‘ (Betas) are the regression coefficients, representing the amount the dependent
variable ‗y‘ changes when the corresponding independent variables changes by one unit. The
‗c‘ is the constant, where the regression line intercepts the ‗y‘ axis, representing the amount
the dependent ‗y‘ will have when all the independent variables are zero.
4.2 Concept of Production Function
Generally, production function shows how the factors of production such as land, labour,
capital and entrepreneur are combined to produce output. Factors of production have derived
demand because only factor of production does not provide utility to human being e.g.
fertilizer provides no utility to human beings but it increases out put when used in production
process. According to Beattie and Taylor (1985), production function is the highest output
that a farmer can get from a given set of inputs with the given technology. A production
function in mathematical form is expressed as:
Y= f (X) (1)
where ‗Y‘ denotes output of a farmer, ‗X‘ shows a vector of inputs used in the production
process and (f) represents suitable functional form.
There are two types of inputs which are used in the production process. They include variable
inputs and fixed inputs. Variable inputs can be defined as the inputs the amounts of which can
be changed during a production process whereas fixed inputs are those inputs whose amount
does not change in a production process for certain of time. So we can say there are two time
70
periods namely short run period and long run period. All inputs used in the production
process are assumed as variable inputs in the long run time period. Whereas in the case of
short run time period one input is assumed as variable input and all others remain fixed.
Doll and Orazem (1984) and Beattie and Taylor (1985) have discussed different assumptions
for a production function which are as follows.
1. The first assumption says that the production activity of a farm in one time period is
absolutely incompatible or independent of production in preceding and following time
periods.
2. All inputs and outputs in the production process of a farm are homogeneous and it is
called homogeneity.
3. The production function can be differentiated twice.
4. The production function and output and input prices are given.
5. The accessibility and availability of input is unlimited indicating that there is no
budget constraint.
4.2.1 Cobb-Douglas Production Function
In economic literature, the Cobb-Douglas functional form of production functions is
extensively employed to estimate the relationship of an output to inputs particularly in
agriculture sector. It was proposed by Knut Wicksell and tested against statistical evidence by
Paul Douglas and Charles Cobb in 1928. General form of the production function is:
Y= ALαK
β
Where,
Y = total output (the monetary value of all goods produced in a year)
L = labor input
K = capital input
A, α and β are constants determined by technology.
If
α+ β=l
71
the production function has constant returns to scale. That is, if L and K are each increased by
20 %, Y increases by 20 %.
If
α+ β < 1,
the production function has decreasing returns to scale,
and if,
α+ β >l
the production function has increasing returns to scale.
Assuming perfect competition, α and β are labour and capital's share of output respectively.
4.3 Efficiency
Economic performance of a firm, farm or organization is estimated by efficiency. It is defined
as the economic or productive efficiency of a firm, farm or organization, meaning that it is
thriving in producing as much output as feasible from a known set of inputs (Farrell 1957).
According to Koopmans (1951) efficient producer is one who has to sacrifice at least one unit
of any output to obtain at least one extra unit of other output or who could save at least one
unit of any input at the cost of reduction in the quantity of at least one output. There are two
components of efficiency such as technical and allocative efficiency. Technical efficiency
means the capability of a farm to produce as much output as achievable with given sets of
inputs (input oriented efficiency measure) or the capacity of a farm to use as least inputs as
possible for a given level of output (output oriented measure). Farrell (1957) estimated the
difference between technical and allocative efficiency. Technical inefficiency may be there
when maximum output is not achieved with given factors of production and the causes of
technical inefficiency include mismanagement of timing and method of application of
production inputs.
Allocative or price inefficiency arises when the ratio of marginal products of inputs is unequal
to the ratio of market prices. According to Lovell (1993), allocatively efficient firms combine
optimal combination of inputs and output while keeping in view the established prices.
Economic efficiency is the product of technical and allocative efficiency.
72
4.3.1 Input Oriented Efficiency Measures
Farrell (1957) used input oriented efficiency measures to explain the constant return to scale
relation between multi-inputs and a single output. Using this approach, he illustrated the
concepts of technical, allocative and economic efficiency. In Figure 4.1 input oriented
efficiency measure has been explained in detail. Two input factors namely, X1 and X2 are
used to produce a single output (Y). A fixed level of output produced by using different
combination of two inputs is shown by the unit isoquant input IQIQ. Point B shows the
technically output level on the efficient unit isoquant, IQIQ.A firm operating at point A
produces the same level of output Y as produced on unit isoquant, IQIQ. A line drawn from
the origin O, to the point A explains the technical efficiency of the given firm. This OA line
passing through the point B indicates that the same level of output, Y, is produced with X1
and X2 inputs at the point B (Coelli et al. 1998) implying that the observed firm id technically
inefficient. Thus, the technical efficiency of the observed firm is defined as the ratio of the
distance from the point B to the origin over the distance of the point A from the origin.
TE = OB / OA
We can estimate allocative efficiency in the presence of given input prices. In Figure 4.1
isocost line is shown by ww which is tangent to the unit isoquant at the point D. The
allocative efficiency is defined as
AE = OC / OB
O X1
Figure 4.1: Input-Oriented Measures for Technical and Allocative Efficiency (Reproduced from Coelli et al. (1998).
X2 IQ
IQ
W
73
Both technically and allocatively efficient output can be produced at the point D, but point B
shows technically efficient production only (Coelli et al 1998).
Economic efficiency is defined as the product of technical and allocative efficiency.
EE = TE* AE
EE = [OB/OA) OC/OB] = OC/OA
4.3.2 Concept of Production Frontier
A standard against which the technical efficiency of production of a firm, farm or
organization is measured is known as production frontier and Farrell (1957) referred it as the
best practice frontier. Two methods are commonly used to analyze the internal and external
factors affecting the efficiency and input, output relationship. They include average
production function and production frontier function. The important distinction between the
production frontier and the average production function is that average production function
represents the mean output for a given level of input while, the production frontier shows the
best practice output.
In agriculture, Cobb-Douglas production function has often been used. Number of studies
(e.g. Hoper 1965 and Welsch 1965) has used average production function in these types of
studies. However, Upton (1976) criticized the average production function as it is an
inadequate representation of complex and dynamic farming system. Simultaneous equation
bias and the problem of multicollinearity are other shortcomings of this function pointed out
by Yotopoulous and Nugents (1976) and Lau and Yotopoulous (1971). Ghatak and Ingersent
(1984) argue that average production approach cannot distinguish between technical and
allocative efficiency. On the basis of these drawbacks of average production function,
production frontier has been chosen for the present study to determine technical efficiency of
the farmers in the study area to analyze the impacts of reforms and other factors on the
efficiency/inefficiency.
74
4.3.3 Measurement of Technical Efficiency
Two approaches namely frontier approach and non frontier approach are used to estimate
technical efficiency. There are two types of frontier approach i.e. non-statistical and statistical
methods. Non-statistical methods comprise non-parametric and parametric approaches. The
non-parametric approach is called deterministic approach as it has no fixed functional form
for frontier including all observations in the model. It is also known as data envelopment
analysis (DEA) method. The parametric approach is called probabilistic approach based on
Cobb Douglas or other forms. Statistical methods consist of non-stochastic and stochastic
methods. Non-stochastic frontier approach estimates technical efficiency and it implies that
all variation from frontier is due to presence of inefficiency. Stochastic frontier function (SF)
shows that deviation from frontier is owing to two factors, random effect and inefficiency.
Maximum Likelihood (ML) and corrected ordinary least squares (COLS) methods are used in
statistical methods (Ali and Byerlee 1991). However, in literature, both parametric and non-
parametric approaches have been used to find out technical efficiency of various enterprises.
Parametric approach makes use of econometric modeling and mathematical modeling is done
in non-parametric (DEA) approach. There are certain advantages and disadvantages of these
two approaches over each other which are discussed by Battese (1992), Bravo-Ureta and
Pinheiro (1993), Forsund et al. (1980), Fried et al. (1993), Coelli (1996a) and Coelli and
Perelman (1999).
Schmidt (1986) argued that the results obtained by non-parametric approach are less precise
because it makes less use of information compared to the parametric approach. Further the
DEA approach is sensitive to extreme observations and measurement error (Forsund et al.
1980). Another limitation of this approach is that it is difficult conceptually to separate the
effects of uncontrollable environmental variables and measurement error from the effect of
differences in farm management (Jaforrullah and Whiteman 1999). Moreover, tests of
hypothesis in relation to differences in technical efficiency cannot be performed statistically
(Schmidt 1986, Jaforrullah and Whiteman 1999).
75
4.3.4 Parametric Frontier Production Function
DEA approach has not been used because of disadvantages of this approach as discussed
above. Thus the parametric approach is used for the present study. Parametric approach has
two types, deterministic and stochastic frontier production functions. Development in
econometric frontier production function has been reviewed by Battese (1992) and Coelli, et
al. (1998). The methodology developed by Coelli et al. (1998) has been followed for this
study.
Following the work of Farrell (1957), we assumed that the production function of fully
efficient firms is known. However, in practice the production function is not known. Farrell
(1957) gave the solution to this problem. According to Farell, the sample data could be used
to estimate the production function by implying a non-parametric piece-wise linear
technology or a parametric function, such as the Cobb Douglas production function. Aigner
and Chu (1968) used the parametric functional approach for a sample of N firms
Yi = f (xi, β)e-µi
i = 1,2…….,N (2)
We can write the above production function in log form as
ln(Yi, ) = xi β--µi
(3)
Yi denotes the output for the i-th firm, xi represents a vector of k-input used by the i-th firm, β
is a vector of unknown parameters to be estimated, µi is associated with technical inefficiency
(TEi) in production and In represents natural logarithm. The above model is called
deterministic frontier production function because the observed output Yi is bounded above
by the deterministic quantity exp (xi β). Technical efficiency (TEi) of the i-th firm is defined
as:
TEi = Yi / exp (xi β) = exp (xi -µi) / exp (xi β) = exp (-µi) (4)
One of the most important distinctions between non-parametric and parametric deterministic
frontier model is functional form specification. Schmidt (1976) argued that the MLE of the
parameters for the statistical model can be obtained by linear and quadratic programming if
the µi‘s shave exponential or half-normal distribution, respectively.
76
Russell and Young (1983) criticized the deterministic frontier production model, since it
assumes that all deviations from the frontier are the outcome of technical inefficiency.
However, two types of factors can affect the performance of a firm. They include exclusively
outside the control of the firm and under the control of the firm. Factors fully outside the
control of the firm are weather, climate, and failure of market and measurement errors;
whereas, factors under the firm's control include socioeconomic characteristics and
management practices. Thus, a parametric frontier production function was developed to
incorporate these effects while estimating technical efficiency of the firms (Aigner et al. 1977,
Mceusen and Van den Broeck 1977).
The stochastic frontier production function takes account of firm's specific random shocks and
technical efficiency separately into the analysis. Aigner et al. (1977) and Mceusen and Van
den Broeck (1977) pointed out that deviations from the production frontiers are because of
two types of factors, such as factors entirely outside the control of the firm or farmer and
factors under the control of the firm or farmer. This signifies that deviations are not
completely under the control of the firm or farmer, but some factors such as bad weather,
measurement errors, etc. are totally outside the control of the firm or farmer.
4.3.5 The Stochastic Frontier Production Function
In order to overcome the deficiencies of traditional stochastic frontier production model,
Aigner et al. (1977) and Mceusen and Van den Broeck (1977) gave independently the
stochastic frontier production function, including both types of factors into the model. In such
type of model, error term is decomposed into two components, factors outside the control of
the firm or farmer and factors under the control of the firm or farmer. Therefore, this model is
also called composed error model. This model shows that the firm's output can be affected by
technical inefficiency along with measurement errors and other factors, such as effects of
weather, luck, etc., combined effects of unspecified/omitted variables in the model (Coelli et
al 1998).
77
For the cross-sectional data, the stochastic frontier production function model is as follows:
Yi = f (xi β)eε
i (5)
And εi = vi—µi where i =1,2,....., N
Here vi represents a random error and it takes in to account of measurement errors and other
random factors outside the control of a firm or a farmer. Aigner et al. (1977) assumed that vi‘s
independent and identically distributed (i.i.d) normal variables with mean zero and constant
variance, σ2
v independent of the µi‘s. The µi‘s shows the technical inefficiency effects and
they are connected with technical inefficiency of the firm or farmer. The µi‘s are assumed to
be identically and independently distributed exponential or half normal random variables
(Coelli et al. 1998).
The stochastic frontier production function is detailed in Figure 4.2. The horizontal axis
shows the inputs units used in the production process and outputs are represented on the
vertical axis. The deterministic frontier production function in the figure assumes declining
return to scale. Two firms, i and j are considered. Suppose firm i produces output Yi, and firm
j produces output Yj. Firm i makes use of xi units of inputs to produce output Y*i, = exp
(xiβ+vi) and this output level lies above the deterministic output level, Yi, = exp (xiβ+vi)
because the vi‘s are positive. Now consider the firm j using j-th units of inputs and generating
output Y*j =exp (xiβ+vi).This output level lies below the deterministic output level, since vi‘s
are negative. Thus we can say that the observed output may be higher than the deterministic
frontier production function if the random errors are greater than the inefficiency effects Y*i
>exp (xiβ)if vi>µi). However, the observed output would be smaller than that of the
deterministic frontier production function if the random errors are less than the inefficiency
effects (Y*J >exp (xiβ)if vj>µi) (Coelli et al. 1998).
78
Figure 4.2: Stochastic Frontier Outputs (Reproduced from Coelli et al. 1998).
4.3.6 Estimation of Stochastic Frontier Production
The maximum likelihood (ML) method or corrected ordinary least square (COLS) method
can be used to estimate the parameters of the stochastic frontier production function. Coelli
(1995) concludes that the maximum likelihood estimator is asymptotically more efficient than
the COLS estimator. Computer software such as LIMDEP econometrics packages (Greene
1992) and the FRONTIER 4.1 program (Coelli 1996) are used to determine technical
efficiency.
Frontier output
Yj,if Vj< 0
Frontier output
Yi, ifVi>0
Deterministic production
frontier
X Xi Xj
Y
79
Aigner et al. (1977) developed the log likelihood function for the model in which half-normal
distribution for the technical inefficiency effects is assumed. Aigner et al. (1977) explained
the likelihood function in terms of the two variance parameters as σ2
s = σ2 + σ
2v and λ = σ / σv.
Nevertheless, Battese and Corra (1977) suggested that the parameter, y= σ2 / σ
2s used because
it has a value between 1 and 0 and the λ parameter could be any non-negative value. In
general, technical efficiency lies between 0 and 1. When Technical efficiency is equal to one,
it implies that the firm or farmer is producing on the production frontier with available
resources and technology and it is the indication that the firm or farmer is technically
efficient. When value of technical efficiency is less than zero, it implies that the firm or
farmer is producing below the production frontier for given technology and resources and it is
said that firm or farmer is technically inefficient. Using the y parameterization, the log
likelihood function by Battese and Corra (1977) is as under:
In(L)= - N/2 In [(II) / 2] - N/2log (σ2
s) + ΣN
i = 1 In[1 - Ǿ(zi)] - 1/2 σ2 Σ
Ni = 1(In Yi - xiβ)..... (6)
where,
Zi= (In Y-xiβ) / σs
and θ(.) is the distribution of the standard random variable (Coelli et al. 1998).
4.3.7 Estimation of Mean and Firm-Level Technical Efficiencies
The mean technical efficiency, TEi = exp(-µi) for the firms or farmers can be
estimated as under:
E[exp(-µi)] = 2[1-Ǿ(σs y)]exp(-yσ2s / 2) (7)
Jondrow et al. (1982) recommended that the technical efficiency of the i-th firm can be
80
estimated by using 1 – E[µi / εi] Battese and Coelli (1988) point out that the best
predictor of exp(-µi) is obtained by using
E[exp(-µi) / εi] = 1-Ǿ(σA+ yεi / σA) exp(yεi+ σA 2
/ 2) (8)
4.3.8 Technical Inefficiency Effects Model
A variety of factors, such as distinctiveness of firms, management, physical, institutional
and environmental aspects can affect technical inefficiencies in the production process of
the firms or farmers. Kalirajan (1981) and Pit and Lee (1981) regressed the predicted
technical inefficiency effects on various explanatory variables, such as firm size, age,
education of the manager, etc. In the above studies two staged approach has been
employed however, this approach has been criticized due to serious problems pertaining
to assumptions made for the µi. In the first stage, the technical inefficiency effects are
assumed to be independently and identically distributed using the approach of Jondrow et
al. (1982) to estimate firm or farm level technical inefficiency. However, the predicted
technical inefficiency effects are assumed to be a function of a number of firm-specific
factors in the second stage, which implies that they are not identically distributed, unless
all the coefficients of the factors are simultaneously equal to zero (Coelli et al. 1998).
Kumbhakar et al. (1991), Reifschneider and Stevenson (1991) and Battese and Coelli
(1995) criticized the second stage used to estimate the determinants of technical
efficiency. They specified stochastic frontier models in which the inefficiency effects are
defined to be explicit functions of some firm-specific factors, and all parameters are
estimated in a single-stage maximum likelihood procedure. Wang and Schmidt (2002)
critically discussed biasness in two-step estimation of the effects of exogenous variables
in technical efficiency levels by proposing a class of one-step models based on the
scaling property that u equals a function of z times a one-sided error u whose distribution
did not depend on z.
Huang and Liu (1994) suggested a model for a stochastic frontier production function, in
which the technical inefficiency effects are specified to be a function of some firm-
specific factors, in conjunction with their interactions with the input variables of the
81
frontier function. Bravo-Ureta and Pinheiro (1993) also studied the association between
technical efficiencies and various socio-economic variables, such as age and level of
education, firm size, access to credit and utilization of extension services. Battese and
Coelli (1993, 1995) proposed the technical inefficiency effects model for panel data. This
model estimates stochastic frontier production function and inefficiency effects in a
single step to avoid problems of two-step models. The single stage model is used for the
present study and the model for cross-sectional data is detailed below.
To begin with the stochastic frontier production function
Yi=exp (xiβ+vi-µi) (9)
From this model the technical inefficiency effects are function of a set of explanatory
variables, namely firm size, age, education of the manager and other socio-economic
factors and vectors of unknown parameters to be estimated and a random error (Battese
and Coelli 1995). It is assumed that the technical inefficiency effects are independently
distributed non-negative random variables, however, these effects are not identically
distributed random variables. Therefore, the technical inefficiency effect, µi. in the
stochastic frontier model is specified as:
µi. =zi. + wi (10)
Where zi shows a (1 x P) vector of firm specific variables influencing the efficiency of ith
firm, and Б is a (Pxl) vector of parameters to be estimated and wi are unobservable
random variables which are assumed to be independently distributed and obtained by
truncation of the normal distribution with zero mean and constant variance (σ2 ). Factors
affecting technical efficiency, such as land size, land tenure, credit availability,
extension and education have been studied by Kalirajan and Flinn (1983), Lingard et al.
(1983) and Shapiro and Muller (1977). Ali and Flinn (1989) included farm size, method
of cultivation and share tenancy as explanatory variables for rice growers in Pakistan
besides above determinants. Parikh et al. (1995) included family size, off-farm work, and
82
fragmentation and non-farm assets in addition to above mentioned explanatory variables
to estimate the effects on efficiency in agriculture sector of Pakistan.
4.4 Summary
This chapter was divided into three parts. In the first part, single equation models were
estimated to capture the impact of irrigation reforms on farm income and productivity.
Multiple Regression model was estimated by using Ordinary Least Square (OLS)
estimation procedure. Cobb-Douglas production function along with its general form was
also discussed in detail. The previous studies showed that Cobb-Douglas production
function was widely used to determine the input-output relationship in agriculture sector.
It was also found that the Cobb-Douglas production function was not developed on the
basis of any knowledge of engineering, technology, or management of the production
process. It was instead developed because it had attractive mathematical characteristics,
such as diminishing marginal returns to either factor of production.
The concepts of efficiency, input oriented efficiency measures and measures of technical
efficiency were also discussed in this chapter. Frontier approach was used to measure the
technical inefficiency. It was divided in to two types i.e. non-statistical and statistical
methods. Statistical methods consisted of non-stochastic and stochastic methods.
83
CHAPTER 5
METHODS OF ANALYSIS
This chapter consists of econometric techniques to analyze the data. Different approaches
were used for quantitative analysis of data which have been discussed in this chapter.
5.1 Approaches for the Analysis
In order to have comprehensive evaluation of performance, achievements and early
impacts of reforms in LCC (East) irrigation System, the study employed two approaches
in evaluating performance and impact of newly created institutions.
A) First Level Approach
This level included the Review and Assessment of reforms in LCC East (Reform Area)
on the basis of information from the secondary data. At this level ―before and after‖ IMT
situation was compared.
B) Second Level Approach
The performance of distributaries after the reforms was evaluated on the basis of primary
data collected through well structured questionnaire at farm household level.
5.2 Quantitative Analysis
Quantitative analysis of the data was made by using the following econometric
techniques.
1. Comparison of set of indicators for secondary data collected from PID and PIDA.
Means, averages, percentages and frequencies were estimated to determine the
impact of irrigation reforms in pre and post-reform periods.
84
2. Estimations of Single Equation Models were done to capture the impact of
irrigation reforms on farm income and productivity.
3. Estimation of Economic Inefficiency Model was carried out to determine the
negative impact of irrigation reforms on inefficiency of the respondents.
5.2.1 Comparison of Means, Averages, Percentages and Frequencies
The situation of the selected distributaries before irrigation management transfer was
compared with that after management transfer. For comparison purpose following
variables analyzed.
1. Operation and Maintenance (O&M) of the system which included all the
expenditures incurred on the selected distributaries for smooth operations of
the distributaries. Such expenditures were desilting the canal, brim cutting,
repair and maintenance of the defective outlets and gauges. These expenditures
were for financial year starting from July 1, 2004 to June 30, 2006.
2. Contingency expenditures included all expenditures other than used for
operation and maintenance of the distributary. Such expenditures were
stationary expenditures, telephone and electricity bills, office maintenance and
purchase of office furniture and equipment. All these expenditures were
compared in the before and after irrigation management context. The
contingency expenditures were collected on annual basis.
3. Salary expenditures included the salary expenditures of all the staff working
on the distributary. These expenditures were taken on yearly basis.
4. Aabiana or water charges are collected on the particularly distributary in one
year before irrigation management transfer was compared with that collected
in the same distributary after the irrigation management transfer. Aabiana
(Water charges) assessment is made twice in a year i.e. in Rabi and Kharief
seasons. A flat rate system of Aabiana is adopted in Punjab after Aabiana
collection of 2004. According to this system, a fixed rate of Rs. 85 per acre in
85
Kharief and Rs. 50 per acre in Rabi are charged as water charges from the
farmers. However, this rate is charged half on all the lands which are situated
at the Tail(the command area which is situated below the last 20 percent of the
entire length of the distrbutary). The flat rate is liable on the cultivatable
command area regardless of the fact whether it was actually cultivated / sown
or not. This system envisaged assessing Aabiana on the basis of entire area
under CCA i.e. the area to which irrigation water is sanctioned irrespective of
the cropping pattern, intensity of irrigation and area actually sown for a current
crop. The Chairman Khal Punchayat (KP) is responsible for Aabiana
collection and sending the collected amount to his respective FO. KP
Chairman is paid 5 percent of the 100 percent Aabiana collection as an
incentive. Each FO, then keep 40 percent of this collected amount for O&M of
the system and for office management. The remaining 60 percent is sent to
PIDA as its share.
5.2.2 Estimation of Econometric Models for Different Crops
Econometric analysis of gross value product (GVP) and average yield for wheat, rice and
sugarcane was carried out by developing desired relationships i.e. impact of PIM reforms
on agriculture productivity and farm income by implying Cobb-Douglas Production.
1. Estimation of Model for Gross Value Product of Wheat
For estimation of income and productivity of wheat crop for different farm households
before and after irrigation management transfer, relationships were developed through
Cobb-Douglas Production. The log form of the production function for estimating gross
value product for wheat crop is written as:
Lnwagvp = β0 + β1 lnwareaij + β2 lnwscostij + β3 lnwfcostij + β4 lnwsicostij
+ β5 lnwticostij + β6 lnwmcostij + β7 lnwlcostij + β8 lneduij + β9 D1 + β10 D2
+ β11 D1 D2 + µij
Where:
β0 = Constant
86
lnwagvp = Natural Log of average real GVP of wheat crop of the i-th farm in the
sample area expressed in Rs. per acre calculated (by using GDP deflator
for the year 2001-02 as base).
lnwareaij = Natural Log of area under wheat crop in the sample area measured in
acres.
lnwscostij = Natural Log of seed cost of the i-th farm for wheat crop in the sample area
measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnwfcostij = Natural Log of fertilizer cost of the i-th farm for wheat crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnwsicostij = Natural Log of surface irrigation cost of the i-th farm for wheat crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnwticostij = Natural Log of tube-well irrigation cost of the i-th farm for wheat crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnwmcostij = Natural Log of cost of mechanized operations of the i-th farm for wheat
crop in the sample area measured in real price (by using GDP deflator for
the year 2001-02 as base) and expressed in Rs. per acre.
lnwlcostij = Natural Log of cost of labour operations of the i-th farm for wheat crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer for wheat crop in the
sample area.
87
D1 = Dummy variable for location of outlet of specific farm. If D1= 1 then it
represents location at Tail of the distributary otherwise Head or Middle of
the distributary.
D2 = Dummy variable for taking into account the implementation of reform
process. If D2 = 1 then it represents post-reform era and if value is 0 then it
represents pre-reform period.
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
µij = Error Term
2. Estimation of Model for Wheat Yield
The log form of the production function for wheat crop was taken into the account the
major relationships between different inputs, qualitative variables and average yield of
wheat crop (maunds per acre) is written as:
Lnay = β0 + β1 lnwareaij + β2 lnwscostij + β3 lnwfcostij + β4 lnwsicostij + β5
lnwticostij + β6 lnwmcostij + β7 lnwlcostij + β8 lneduij + β9 D1 + β10 D2 + β11
D1D2 + µij
Where:
β0 = Constant
lnay = Natural Log of average yield of wheat crop for i-th farmer.
lnwareaij = Natural Log of area under wheat crop in the sample area measured in
acres.
lnwscostij= Natural Log of seed cost of the i-th farm for wheat crop in the sample area
measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
88
lnwfcostij = Natural Log of fertilizer cost of the i-th farm for wheat crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnwsicostij = Natural Log of surface irrigation cost of the i-th farm for wheat crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnwticostij = Natural Log of tube-well irrigation cost of the i-th farm for wheat crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnwmcostij = Natural Log of cost of mechanized operations of the i-th farm for wheat
crop in the sample area measured in real price (by using GDP deflator for
the year 2001-02 as base) and expressed in Rs.per acre.
lnwlcostij = Natural Log of cost of labour operations of the i-th farm for wheat crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer for wheat crop in the
sample area.
D1 = Dummy variable for location of outlet of specific farm. If D1= 1 then it
represents location at Tail of the distributary otherwise Head or Middle of
the distributary.
D2 = It is a dummy variable for taking into account the implementation of
reform process. If D2= 1 then it represents post-reform era and if value is 0
then it represents pre-reform era.
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
89
µij = Error Term
3. Estimation of Model for Gross Value Product of Rice Crop
A single equation model for estimation of income from rice in pre and post-reform period
was used. The Cobb-Douglas production function was found to be an adequate
representation of the data. Cobb-Douglas production function for rice is given as:
Lnragvp = β0 + β1 lnrareaij + β2 lnrscostij + β3 lnrfcostij + β4 lnrsicostij + β5 lnrticostij
+ β6 lnrmcostij + β7 lnrlcostij + β8 lneduij + β9 D1 + β10 D2 + β11 D1 D2 + µij
Where:
β0 = Constant
lnragvp = Natural Log of average real GVP of rice crop of the i-th farm in the
sample area expressed in Rs. per acre calculated (by using GDP deflator
for the year 2001-02 as base).
lnrareaij = Natural Log of area under rice crop in the sample area measured in acres.
lnrscostij = Natural Log of seed cost of the i-th farm for rice crop in the sample area
measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnrfcostij = Natural Log of fertilizer cost of the i-th farm for rice crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnrsicostij = Natural Log of surface irrigation cost of the i-th farm for rice crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
90
lnrticostij = Natural Log of tube-well irrigation cost of the i-th farm for rice crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnrmcostij = Natural Log of cost of mechanized operations of the i-th farm for rice crop
in the sample area measured in real price (by using GDP deflator for the
year 2001-02 as base) and expressed in Rs. per acre.
lnrlcostij = Natural Log of cost of labour operations of the i-th farm for rice crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer rice for crop in the
sample area.
D1 = Dummy variable for location of outlet of specific farm. If D1= 1 then it
represents location at Tail of the distributary otherwise Head or Middle of
the distributary.
D2 = Dummy variable to capture the effectiveness of reform process the
implementation of reform process. If D2= 1 then it represents post-reform
era and if value is 0 then it represents pre-reform period.
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
µij = Error Term
91
4. Estimation of Model for Rice Yield
lnay = β0 + β1 lnrareaij + β2 lnrscostij + β3 lnrfcostij + β4 lnrsicostij + β5 lnrticostij
+ β6 lnrmcostij + β7 lnrlcostij + β8 lneduij + β9 D1 + β10 D2 + β11 D1D2 + µij
Where:
β0 = Constant
lnay = Natural Log of average yield of rice of the i-th farm in the sample area
measured in maunds per acre.
lnrareaij = Natural Log of area under the rice crop in the sample area measured in
acres.
lnrscostij = Natural Log of seed cost of the i-th farm for rice crop in the sample area
measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnrfcostij = Natural Log of fertilizer cost of the i-th farm for rice crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnrsicostij = Natural Log of surface irrigation cost of the i-th farm for rice crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnrticostij = Natural Log of tube-well irrigation cost of the i-th farm for rice crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnrmcostij = Natural Log of cost of mechanized operations of the i-th farm for rice crop
in the sample area measured in real price (by using GDP deflator for the
year 2001-02 as base) and expressed in Rs. per acre.
92
lnrlcostij = Natural Log of cost of labour operations of the i-th farm for rice crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer rice for crop in the
sample area.
D1 = It is a dummy variable for location of outlet of specific farm. If D1= 1 then
it represents location at Tail of the distributary otherwise Head or Middle
of the distributary.
D2 = It is a dummy variable to capture the effectiveness of reform process the
implementation of reform process. If D2= 1 then it represents post-reform
era and if value is 0 then it represents pre-reform era
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
µij = Error Term
5. Estimation of Model for Gross Value Product of Sugarcane Crop
Cobb-Douglas production function for estimation of Grass Value Product (GVP) of
sugarcane before and after irrigation reforms is written as:
lnsagvp = β0 + β1 lnsareaij + β2 lnsscostij + β3 lnsfcostij + β4 lnssicostij + β5 lnsticostij
+ β6 lnsmcostij + β7 lnslcostij + β8 lneduij + β9 D1 + β10 D2 + β11 D1D2 + µij
Where:
β0 = Constant
93
lnsagvp = Natural Log of average real GVP of sugarcane crop of the i-th farm in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnsareaij = Natural Log of area under the sugarcane crop in the sample area measured
in acres.
lnsscostij = Natural Log of seed cost of the i-th farm for sugarcane crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnsfcostij = Natural Log of fertilizer cost of the i-th farm for sugarcane crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnssicostij = Natural Log of surface irrigation cost of the i-th farm for sugarcane crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnsticostij = Natural Log of tube-well irrigation cost of the i-th farm for sugarcane crop
in the sample area measured in real price (by using GDP deflator for the
year 2001-02 as base) and expressed in Rs. per acre.
lnsmcostij = Natural Log of cost of mechanized operations of the i-th farm for
sugarcane crop in the sample area measured in real price (by using GDP
deflator for the year 2001-02 as base) and expressed in Rs. per acre.
lnslcostij = Natural Log of cost of labour operations of the i-th farm for sugarcane
crop in the sample area measured in real price (by using GDP deflator for
the year 2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer for sugarcane crop in
the sample area.
94
D1 = Dummy variable for location of outlet of specific farm. If D1= 1 then it
represents location at Tail of the distributary otherwise Head or Middle of
the distributary.
D2 = Dummy variable for taking into account the implementation of reform
process. If D2= 1 then it represents post-reform era and if value is 0 then it
represents pre-reform period.
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
µij = Error Term
6. Estimation of Model for Sugarcane Yield
Sugarcane is important annual major cash crop of Punjab. It requires a wide range of
inputs like fertilizer, irrigation water and intensive labour at different stages of growth.
The nature of output of sugarcane crop also differs from other cash crops. For estimation
of income and productivity of sugarcane for different farm households, relationships were
developed through Cobb-Douglas Production. The log form of the production function
for sugarcane is written as:
lnay = β0 + β1 lnsareaij + β2 lnsscostij + β3 lnsfcostij + β4 lnssicostij + β5 lnsticostij
+ β6 lnsmcostij + β7 lnslcostij + β8 lneduij + β9 D1 + β10 D2 + β11 D1D2 + µij
Where:
β0 = Constant
lnay = Natural Log of average yield of sugarcane of the i-th farm in the sample
area measured in maunds per acre
95
lnsareaij = Natural Log of area under the sugarcane crop in the sample area measured
in acres.
lnsscostij = Natural Log of seed cost of the i-th farm for sugarcane crop in the sample
area measured in real price (by using GDP deflator for the year 2001-02 as
base) and expressed in Rs. per acre.
lnsfcostij = Natural Log of fertilizer cost of the i-th farm for sugarcane crop in the
sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnssicostij = Natural Log of surface irrigation cost of the i-th farm for sugarcane crop in
the sample area measured in real price (by using GDP deflator for the year
2001-02 as base) and expressed in Rs. per acre.
lnsticostij = Natural Log of tube-well irrigation cost of the i-th farm for sugarcane crop
in the sample area measured in real price (by using GDP deflator for the
year 2001-02 as base) and expressed in Rs. per acre.
lnsmcosti = Natural Log of cost of mechanized operations of the i-th farm for
sugarcane crop in the sample area measured in real price (by using GDP
deflator for the year 2001-02 as base) and expressed in Rs. per acre.
lnslcostij = Natural Log of cost of labour operations of the i-th farm for sugarcane
crop in the sample area measured in real price (by using GDP deflator for
the year 2001-02 as base) and expressed in Rs. per acre.
lneduij = Natural Log of years of schooling of the i-th farmer for sugarcane crop in
the sample area.
D1 = Dummy variable for location of outlet of specific farm. If D1= 1 the it
represents location at Tail of the distributary otherwise Head or Middle of
the distributary.
96
D2 = Dummy variable for taking into account the implementation of reform
process. If D2= 1 then it represents post-reform era and if value is 0 then it
represents pre-reform period.
D1D2 = Interaction variable of two dummies i.e. D1D2 was used to capture the
impact of reform process on the farms located at the Tail of the
distributary.
µij = Error Term
5.3 Economic Inefficiency Model
This section describes the likelihood ratio test followed by methods to estimate
efficiency/inefficiency effects of different factors on crops grown in the study area.
5.3.1 Empirical Model and Efficiency Analysis
Wheat, sugarcane and rice are major crops cultivated in the study area i.e. LCC (East)
irrigation system. Each crop requires wide-ranging levels of inputs during different
growth stages such as different types of fertilizers being utilized by the farmers, different
kinds of plant protection measures (including chemicals) used, different kinds of labour
such as part time labour, full time labour or family labour and variety of other inputs.
Keeping all these facts in view it is more logical determine technical efficiency of all
crops separately. Therefore, this study was designed to determine technical efficiency of
selected crops separately.
Stochastic frontier production function was adopted to estimate technical efficiency for
the selected crops. Several studies used the stochastic production function approach to
determine technical efficiency (Parikh and Shah 1994, and Hassan 2004). Several
production function forms have been adopted in the different empirical studies,
depending on the nature of data on hand. However, Cobb-Douglas functional form has
been widely used to examine farm efficiency in spite of its well-known limitations
(Battese 1992, Bravo-Ureta and Pinheiro 1993 and Hassan 2004). Kopp and Smith
(1980) indicated that functional form has a distinct but rather small impact on estimated
97
efficiency. Ahmad and Bravo-Ureta (1996) rejected the Cobb-Douglas functional form
and favoured translog form, but concluded that technical efficiency measures do not
come out to be affected by the selection of the functional form (cited in Thiam et al.
2001). However, the Cobb-Douglas functional form is used even with its well-known
limitations because it is easy to estimate and mathematically manipulate.
In Yij = β0 + Σ8
i = 1 βij In xij + vij - µij and i = 1, 2……..n j = 1, 2…... n
Where, Yij is the dependent variable in the production function showing GVP (Rs. per
acre) for the i-th farm growing j-th crop. In represents natural logarithm. Crop GVP and
input variables are expressed in logarithms. Eight input categories were defined as
explanatory variables in the production function. xij is a vector of k inputs used in the
production of j-th crop and xij are defined as under:
x1ij shows number of acres under the j-th crop in the specific years
x2ij represents the cost of seed (Rs. per acre) used on the i-th farm growing the j-th
crop.
x3ij represents the cost incurred on different fertilizers (Rs. per acre) on the j- th crop
x4ij represents cost of surface irrigation water used to irrigate one acre of land under
the j-th crop.
x5ij indicates cost of tube-well irrigation (Rs. per acre) on the i-th farm for the j-th
crop.
x6ij shows the cost of plant protection measures including pesticides, herbicides and
fungicides (Rs. per acre).
x7ij j shows the cost of mechanized operation (Rs. per acre) for the i-th farmer on j-th
crop.
x8ij represents the cost of labour incurred (Rs. per acre) for the i-th farmer on j-th
crop.
98
β0, βij are unknown parameters to be estimated and vij and uij are defined earlier.
ui’s are non-negative random variables, associated with technical inefficiency of
production of the farmers, assumed to be independently distributed, such that the
technical inefficiency effect for the i-th farmer growing the j-th crop is obtained by
truncation (at zero) of the normal distribution with mean ui and variance σ2, such that
uij = Б0+ Б1Z1ij + Б2Z2ij + Б3Z3ij+ Б4Z4ij+ Б5Z5ij+ Б6Z6ij+ωij
Where:
Z1ij represents the age of farmers in years.
Z2ij indicates years of schooling of the farmers.
Z3ij represents farming experience of the farmers in years.
Z4ij dummy variable indicating the location of the farmers at the distributary.
(if the farmer is locating at tail of the distributary, then it has the value of one,
otherwise zero).
Z5ij dummy variable showing the implementation of irrigation reforms or
otherwise (Value of the dummy is one for the Posr-reform period and zero
otherwise).
Z6ij represents the total farm area operated by the farmers in acres.
It is assumed that some farmers produce on the frontier and others do not produce on the
frontier. Therefore, the need arises to find out factors causing technical inefficiency. The
technical inefficiency model has been developed for this study to concentrate on this
issue. The technical inefficiency effects model incorporates farm and fanner's specific
characteristics, institutional and environmental factors. The above-mentioned variables
included in the technical inefficiency effects model are detailed below with their
expected effects on technical inefficiency.
Age of the primary decision maker was included in the above model to estimate the
impact of age on the level of technical inefficiency. Age is an important characteristic
which can affect the ability of the farmer to learn and adopt new practices and in
99
characterization of its traditional behaviour. It is anticipated that more the age of the
farmer, more he will be risk averser and conservative in nature.
The variable education, the years of schooling of the primary decision makers in crop
production is used a proxy variable for managerial input. Increased farming experience
together with higher level of education could guide to better management of farming
practices. Thus, this variable is expected to reduce technical inefficacity in farming.
The variable farming experience may have a positive effect on technical inefficiency if it
is not linked with age of the farmer.
Location of the farmers (or farm) at Head, Middle or Tail of the distributary can also
have an impact on the technical efficiency of the farmers since it has been established in
few studies especially with respect to study area that majority of the farmers at tail
clusters are small and poor (Hussain and Biltonen 2001).
Another important binary variable included in the model is to take into the account
implementation of reform process in the area. It was expected that implementation of
reform process may have a positive impact on the technical efficiency of farmers.
Farm area operated by the farmers under the specific crop is another variable included in
the model to determine its impact on technical inefficiency. Several studies unearth a
significant relationship between technical efficiency and farm size (Wang et al. 1999).
Yet some studies found no such association (Byrnes et al. 1987). Thus, no a priori
expectation for this variable is made here.
100
5.4 Summary
In the chapter of ―Methods of Analysis‖ different analytical approaches used for
quantitative analysis were discussed. Quantitative analysis was carried out for pre and
post-reform periods. Analysis of data was made by using three different econometric
techniques. These were:
1. Comparison of set of indicators for secondary data collected from PID and PIDA.
Means, averages, percentages and frequencies were estimated to determine the
impact of irrigation reforms in pre and post-reform period.
2. Estimation of Single Equation Models were done to capture the impact of
irrigation reforms on farm income and productivity.
3. Estimation of Economic Inefficiency Model to determine the negative impact of
irrigation reforms on inefficiency of the respondents particularly the farmers at
the Tail reaches.
For comparison purpose different variables such as operation and maintenance of the
system (O&M), Aabiana (water charges) collection, salary and contingency expenditures,
delivery performance ratio (DPR) and Head-Tail equity in pre and post-reform period
were analyzed. Econometric analysis of gross value product (GVP) and average yield for
major crops like wheat, sugarcane and rice was carried out by developing desired
relationships i.e. impact of institutional reforms on agricultural productivity and farm
income by implying Cobb-Douglas production function.
101
CHAPTER 6
COMPARISON OF PERFORMANCE INDICATORS
To achieve the objectives of the study and to assess the performance of the system in
terms of improving water delivery, O&M of the system, equity in water distribution and
overall management of the system and agricultural and economic productivity, data were
collected separately from primary sources (farming households in the study area) and
secondary sources (office record of FOs, PID, PIDA and other published material).The
present chapter is divided in to three parts. The first part covers the socio-economic
profile of the respondents in the study area. The second part is based on the comparison
of indicators developed from primary data and the third part is based on the comparison
of indicators developed from secondary data. These indicators have been used in different
studies such as Hussain and Biltonen (2001), Hussain et al. (2003) and Molden et al.
(2007)
6.1 Socio-Economic Profile of the Respondents Across the Study Area
Apart from the direct measurement of the variables in the study, other basic
characteristics like, family size, sources of income of the respondents, education, tenancy
status and land distribution also provides information on the impact of irrigation reforms
in the study area. As it was observed that a large family size with low education level has
more tendencies to work on the farmland as compared to an educated family of the same
size. Various indicators were analyzed under the following categories:
Socio-economic indicators
Agricultural indicators
102
6.1.1 Social Indicators
Family Size in the Study Area
Family size represents the number of total family members (male and female) of the
respondent. A large number of farm household in the study area were averaged 6 family
members. About 11 percent of the respondents had family members totaling 7 or more.
About 23 percent of the respondents fall in the category of those having family members
from 2 to 4. About 28 percent of the respondents in the study area had 6 family members.
Thus Table 6.1 shows a heterogeneous family structure as observed in the study area.
Table 6.1: Structure and Family Size of the Respondents in the Study Area
Family Size No. of Respondents Percentage (%)
2 to 4 family members 82 22.7
5 family members 59 16.5
6 family members 102 28.4
7 family members 78 21.6
7 and above members 39 10.8
Total 360 100
Sources of Income
Farming is profitable business if it is managed scientifically/technically using balanced
inputs. In the study area, it was found that most of the farmers were growing major crops
(wheat, rice and sugarcane). Due to financial constraints, they were unable to use the
recommended inputs and hence they were not getting expected yields. It was also evident
that for large proportion of farmers in the study area agriculture was the only source of
income. However, there were few farmers who were jointly managing farmland along
with the other economic activities. More than 80 percent of the farmers were engaged in
agriculture for their livelihood. Only 7.50 percent of the respondents were either doing
government or private sector service. Some 2.66 percent of farmers were poultry farmers
and 4.55 percent were shopkeepers. A small percentage of farmers (about 0.45 percent)
were also getting remittances from abroad. It was also found that there were some
103
farmers who were also doing labour along with agriculture. This group of farmers
represented about three percent of the total respondents.
Table 6.2: Source of Income of the Respondents in the Study Area
Source of Income No. of Respondents Percentage (%)
Agriculture only 293 81.35
Agriculture +Job 27 7.50
Agriculture + Poultry Farming 10 2.66
Agriculture +Shop keeping 16 4.55
Agriculture +Remittances 2 0.45
Agriculture +Labour 9 2.45
Agriculture +Others 3 1.04
Total 360 100
Educational status of the respondents
Table 6.3 shows that 58 percent of the farmers in the study area were either illiterate or
under middle school level. When the respondents were asked about their low level of
education, they revealed that they got the household farming in inheritance from their
forefathers. For such respondents, agriculture was the only source of income, and due to
their limited resources they remained unable to get desired education. On the other hand,
only 1.6 percent of the respondents were those whom have completed 14 years or above
schooling. Out of the total sampled respondents 7.45 percent completed 10 years of
schooling. It was observed that majority of the educated farmers were also engaged in
other economic activities at local level. Thus it is clear from the above discussion that
most of the farmers in the study area were not equipped with education. Similar pattern of
educational qualification was found in FOs. The results of the study also showed that a
majority of the farmers were illiterate. Such farmers represented 58 percent of the total
respondents in the study area. While interviewing such farmers, it was observed that they
were laggard in adopting modern agricultural techniques. It was also found that only 6
percent of the farmers had higher school education.
104
Table 6.3: Educational Qualification of the Respondents in the Study Area
Qualification of the Respondents No. of Respondents Percentage (%)
B.A and above 6 1.6
Intermediate (Higher Secondary) 20 5.53
Matriculation (Secondary level) 27 7.45
Middle (8 years of schooling) 100 27.65
Illiterate 207 57.77
Total 360 100
6.1.2 Agricultural Indicators
Land Distribution
Land is unevenly distributed among the farmers in the study area. The number of farms in
different farm household classes varied during different time intervals. Land is being
fragmented in to smaller units due to our prevailing inherited land distribution system.
Over population has been one of the more apparent causes of this variation especially as
the farm sizes were decreasing. Table 6.4 shows that a significant proportion of the
farmers in the study area were operating less than 5 acres of land. Land holding in the
study area was mostly in the range of 3 to 5 acres. Similarly, about 80 percent of the
farmers owned land in the range of 3 to 5 acres in LCC (East) irrigation system of
Punjab. About 0.6 percent of the farmers were operating farm area more than 100 acres,
showing a meager proportion. On the other hand, thus heterogeneity was found regarding
the landholdings of the farmer.
Table 6.4: Landholding of the Respondents in the Study Area
Farmer Classes No. of Respondents Percentage (%)
Less than 5 acres 157 43.6
5 to 12.5 acres 135 37.5
12.51 to 25 acres 59 16.4
25.01 to 100 acres 7 1.9
More than 100 acres 2 0.6
Total 360 100.0
105
Farming Experience
Experience of farm household in agriculture has also implications on agricultural
productivity. About 4 percent of the farmers had less than 5 years of experience in
agriculture. Such farmers were engaged in other economic activities in addition to
agriculture. It was found that about 13 percent of the farmers had more than 36 years of
experience in farming. They were able to get more productivity of crops by timely
sowing of crops, avoid flood irrigation hence saving water and balanced use of fertilizers
on account of their experience. A large majority of respondents were doing agriculture
for the last 16 to 35 years. Table 6.5 given below shows the farming experience of the
respondents in the study area.
Table 6.5: Farming Experience of the Respondents in the Study Area
Farming Experience No. of Respondents Percentage (%)
Less than 5 years 13 3.5
6 to 15 years 83 23.0
16 to 25 years 100 27.7
26 to 35 years 120 33.55
36 years and above 44 12.25
Total 360 100
6.2 Comparison of Indicators Developed from Primary Data Sources
Primary data were obtained from farming households in the study area as per sampling
scheme discussed in Chapter 3. Primary data were collected for two years. Year 2004
represents pre-reform situation covering two crops i.e. Rabi 2003-04 and Kharief 2004
and year 2006 present post-reform situation covering two crops i.e. Rabi 2005-06 and
Kharief 2006. Year 2005 was transition period so it was not included in pre and post-
reform period.
The primary information comprised of opinions of the farmers regarding their perception
about reduction in water delivery problems at their turns, provision of allocated quantity
106
of water, improvement in quantity of water supplied after reforms and particularly at Tail
end reaches and improvement in operation and maintenance of the distributary after
irrigation management transfer were collected and analyzed. Data about average yield of
major crops, gross margin and cost of production of major crops like wheat, sugarcane
and rice were collected from the farmers in the field and then empirically examined.
6.2.1 Opinion of the Farmers Regarding O&M and Quantity of Irrigation Water
Irrigation water availability by design was less than its demand through out the study
area. Therefore, most of the farmers were concerned and look for ways and means to
increase their water availability. Farmers‘ perception about increased water availability
and FOs management are presented in the Table 6.6. It also shows the perception of the
farmers regarding reduction in water delivery problems, provision of allocated quantity of
water at their turns, improvement in quantity of water supplied after irrigation reforms
and particularly at the Tail reaches and improvement in O&M of the distributary
When respondents were asked about irrigation reforms and their impact in reduction of
water delivery problems after the irrigation reforms, 53 percent of them in Khanki
division were of the view that their problems regarding water delivery had been reduced
and FOs were performing better under PIDA Act, 1997. However farmers in Burala
divisions were more satisfied with the performance of the new system. About 90 percent
of the respondents perceived that water delivery problems have been considerably
reduced after irrigation management transfer. They justified their opinion with the
comments that after reforms, farmers were involved in the decision making process and
they were in a better position to solve their own problems. Similar trend was observed in
Upper Gogera and Lower Gogera canal divisions.
When the respondents were asked about their opinion regarding the provision of allocated
quantity of water at their turns after the reforms, 70 percent of the respondents agreed that
they were getting the allocated quantity of water at their turns. However, they reported
that time required to irrigate one acre by surface irrigation water was insufficient. They
reported that government allocated 24 to 30 minutes for irrigating one acre by canal water
107
whereas, practically it took 120 to 180 minutes to irrigate one acre. 30 percent of the
respondents in Khanki division were not agreed with the above statement. Similarly,
majority of the respondents in Upper Gogera, Lower Gogera and Burala canal divisions
agreed that they found improvement in water delivery at the Tail end reaches. However,
10 percent of respondents in Upper Gogera, 10 percent in Lower Gogera and 6 percent
respondents in Burala canal division reported no change in water supply at the Tail ends
after irrigation reforms. The reason they mentioned was that the new institution (PIDA)
would not be able to provide water according to demand because of shortage of supply at
the rim station.
Similarly majority of the respondents in Khanki (60%), Upper Gogera (70%), Lower
Gogera (54%) and Burala (77%) canal division agreed that they found improvement in
operation and maintenance of the distributaries after irrigation management transfer.
Another important opinion recorded was about the improvement in quantity of water
supplied to the farmers. It was observed that majority of the farmers were of opinion that
there was an improvement in quantity of water supplied. The percentages given in the
‗No‘ column shows the respondents who described the present water quantity supplied to
them as worse than the pre-reform period. As it is indicated from Table 6.6 that majority
of the farmers termed the present water supplies better than the previous system.
108
Table 6.6: Opinion of the Farmers Regarding O&M and Quantity of Irrigation Water
Variables under study
Khanki Canal
Division
Upper Gogera Canal
Division
Lower Gogera Canal
Division
Burala Canal
Division
Yes
(%)
No
(%)
No
Change
(%)
Yes
(%)
No
(%)
No
Change
(%)
Yes
(%)
No
(%)
No
Change
(%)
Yes
(%)
No
(%)
No
Change
(%)
Reduction in water delivery
problems 47 53 - 56 44 - 66 34 - 90 10 -
Provision of allocated
quantity of water 70 20 10 68 22 10 63 17 20 32 30 38
Improvement in quantity of
water supplied after reforms 66 34 0 60 13 27 63 22 15 79 10 11
After reforms improvement
in quantity of water supplied
to Tail ends
70 30 0 73 17 10 70 20 10 94 6 6
After reforms improvement
in O&M of the distributary 60 20 20 70 15 15 54 15 31 77 17 6
After reforms improvement
in quality of construction
works
66 33 1 46 8 46 51 14 35 44 16 40
109
6.2.2 Farmers’ Perception on Reduction in Water Theft Cases
Irrigation water theft has been a chronic problem in the LCC (East) irrigation system and
was root cause of many problems associated with conflicts and inefficient water
utilization. Farmer Organizations (FOs) were able to institutionalize a number of
measures that could lead to substantial reduction in water stealing. Table 6.7 shows that
there were 43 percent of the farmers who reported that water theft had almost been
controlled (theft cases reduced but comparatively at lower pace) after handing over the
irrigation system to the farmers. At the Middle reach of the distributary 48 percent of the
farmers gave their opinion that water theft decreased after IMT. Most of the farmers at
the Tail end were of the opinion that either water theft increased (10 percent) or was same
(10 percent). According to them there was no significant change in the extent of water
theft after change in management. Only the form of water theft changed. Earlier there
was open theft but now there was disguised theft of canal water. Only 7 percent of the
farmers reported that water theft increased but the percentage of those farmers who
reported that water theft decreased after irrigation management transfer, totaled 31
percent.
Table 6.7: Opinion of the Farmers Regarding Cases of Water Theft
Cases of water theft Head (%) Middle (%) Tail (%) Total (%)
Increased 5 6 10 7
Decreased 15 48 35 33
Controlled 55 34 40 43
Same 20 6 10 12
Do not know 5 6 5 5
110
6.2.3 Average Yield of Major Crops Across the System in pre and Post-reform
Period (maunds per acre)
Wheat, rice and sugarcane were the major crops grown in the study area. Table 6.8 shows
the average yield of major crops across the LCC (East) system in pre and post-reform
period. The results of the study showed a wide variation in yield across farms and
distributaries. It is evident from Table 6.8 given below that the irrigation reforms have
positive impacts on crop yields across the study area. In Khanki canal division wheat and
sugarcane were the two major crops. Average yield was 31.1 and 35 maunds per acre
respectively before and after irrigation management transfer. It is revealed that after
reforms, the increase in wheat yield was about 13 percent which was highest increase in
the LCC (East) irrigation system. The other canal divisions also showed increasing trend
in yield of wheat but at slower rate as compared to Khanki division. This substantial
increase in wheat yield in Khanki division was attributed to suitable soil conditions and
fertility of the soil. Agronomic factors like land preparation, timing of crop sowing,
variety and quality of seed, timing of application of inputs and quality of land and
adequacy of irrigation of water contributed significantly to increase in production of
wheat in the Khanki canal division. It was also found that farmers found sufficient time
for sowing of wheat after harvesting of rice, which resulted in their higher yield (Early
sowing of wheat produced high yield). It was also observed that some times farmers keep
their land fallow which conserves the fertility of the soil and resultantly more crop yield.
The percent increase in average yield of wheat was recorded as 12, 8 and 6 percent in
Upper Gogera, Lower Gogera and Burala divisions, respectively. Lowest percent increase
was found in Burala division. This lower yield was due to reason that farmers usually
habitual of late sowing of wheat after harvesting of rice in this area (Basmati rice is sown
in this area which is a late variety). The irrigation management transfer also affected
positively the average increase in sugarcane yield after IMT. In the post-reform period
average yield rose in all canal divisions but percent increase in yield of sugarcane was
more in Khanki division (7%) and Upper Gogera division (7%) as compared to other
canal divisions. Whereas, change in sugarcane yield was about 4 and 3 percent in Lower
Gogera and Burala canal divisions respectively. This increase in sugarcane in Khanki and
Upper Gogera was due to tendency of farmers and their experience in sugarcane
111
production. In Burala canal division, increase in sugarcane yield was lower i.e. 3 percent
but the farmers of this division were already getting maximum yield of sugarcane crop
before IMT and after IMT, therefore showing nominal increase in both the periods. Rice
was grown more or less in all canal divisions in LCC (East) irrigation system of the
Punjab. The average rice yield in the study area before IMT was 31.4 maunds per acre
which rose to 35.5 maunds per acre after IMT, thus showing 13 percent increase after
irrigation management transfer. Burala canal division showed 22 percent increase in rice
yield. Rice is water loving crop that requires plenty water from sowing up to maturity of
crop. An increase in yield of rice in Burala division showed that there was more
availability of surface water after IMT. Burala division is located at the Tail of main
canal where there was shortage of water before the management transfer. Thus increase
in rice yield on an average basis showed that there was improvement in water supply
situation at the Tail of the canal.
While collecting data from the field, it was observed that some farmers were getting more
yields of crops and others operate at low performance level. The difference in crop
productivity across the systems and across the various categories of farmers (small,
medium and large farmers) was attributed to number of factors, which included: land and
irrigation water related factors such as location of distributary with respect to main canal,
location of watercourse with respect to distributary and location of the farm with respect
to watercourse. Overall condition of the watercourse i.e. lined or unlined, also
significantly contributed towards surface water availability to the farmers.
A comparison of average yield of wheat, sugarcane and rice is shown in Figures 6.1, 6.2
and 6.3 respectively. These Figures given below show the yield difference of the said
crops in pre and post-reform period of management change.
112
Table 6.8: Average Yield of Major Crops Across the System (maunds per acre)
Canal
Division
Wheat Sugarcane Rice
Pre-
reform
Period
Post-
reform
period
%
change Pre-
reform
Period
Post-
reform
period
%
change Pre
reform
period
Post-
reform
Period
%
change
Khanki 31.1 35 13 560 597 7 32 35 9
Upper
Gogera 30.4 34.1 12 569 606 7 31.1 35 13 Lower
Gogera 33.6 36.4 8 612 636 4 32.1 35.2 10
Burala 34.8 36.7 6 618 637 3 30.3 37 22
Overall 32.4 35.5 10 590 619 5 31.4 35.5 13
6.2.4 Average Gross Value Product (GVP) of Major Crops Across the System in Pre
and Post-reform Period (Rs. per acre)
Average Gross Value Product of the major crops was calculated on per acre basis using
the real prices (calculated by using GDP deflator for the year 2001-02). GVP depicts the
income of the farmer for the specific crop. Table 6.9 shows that GVP of the farmers
calculated for three major crops (wheat, rice and sugarcane) of the system increased by 4
percent for wheat crop in the year 2006 (post-reform period) as compared to the year
2004 (pre-reform period). While average percentage in GVP for sugarcane increased
about 15 percent in post-reform periods. Most significant increase was in rice crop that
showed an increase of 43 percent across the system. This increase in GVP was attributed
to high increase in prices of rice output. Overall increase in wheat crop was 4 percent
which was lowest among other crops studied in the area. It shows that there was an actual
decline in real price of wheat output received by the majority of the farmers in year 2006
as compared to year 2004. Rice crop showed the maximum increase in real GVP i.e. 43
percent while the increase in average rice yield was about 13 percent depicting that there
was an increase in real price of rice in the year 2006 as compared to year 2004.
113
0
5
10
15
20
25
30
35
40
Khanki Upper Gogera Lower Gogera Burala
Canal Division
md
s/a
cre
Pre-reform Period Post reform period
Figure 6.1: Average Yield of Wheat Crop Across the System
in Pre and Post-reform Period
520
540
560
580
600
620
640
660
Khanki Upper Gogera Lower Gogera Burala
Canal Division
md
s/a
cre
Pre-reform Period Post reform period
Figure 6.2: Average Yield of Sugarcane Crop Across the
System in Pre and Post-reform Period
0
5
10
15
20
25
30
35
40
Khanki Upper Gogera Lower Gogera Burala
Canal Division
md
s/a
cre
Pre-reform Period Post-reform period
Figure 6.3: Average Yield of Rice Crop Across the System
in Pre and Post-reform Period
114
One interesting point is that percent increase in GVP of rice in Burala division was about
61percent which was highest among all the crops and canal divisions. While discussion
with the farmers in the field, it was found that this high increase in GVP of rice was on
account of two factors: one was the availability of sufficient quantity of water after
irrigation reforms and other was higher prices of output paid to the farmers.
Table 6.9: Average Gross Value Product (GVP) of Major Crops Across the System (Rs.
per acre)
Real prices of output are taken by using GDP deflator for the year 2001-02.
6.2.5 Average Cost of Production (COP) of Major Crops Across the System (Rs. per
acre)
Estimation of cost of production of a farm enterprise is a complex phenomenon. The cost
of one enterprise can not be determined precisely unless the cost of all other enterprises is
determined simultaneously. This is due to the reason that the farm enterprises are
interdependent and interrelated and cost of any one of them can not be determined in
isolation. The major cost items included in the study were seeds, fertilizer, chemicals,
tube well irrigation, surface irrigation, labour (permanent and hired) and mechanical. The
cost of production of wheat, sugarcane and rice is given below.
Canal
Division
Wheat Sugarcane Rice
Pre
Reform
Period
Post-
reform
Period
%age
Change
Pre
Reform
Period
Post-
reform
Period
%age
Change
Pre
Reform
Period
Post-
reform
Period
%age
Change
Khanki 11612 12430 7 26179 30561 17 5914 8260 40
Upper
Gogera 11348 12034 6 26600 31074 17 5828 8250 42
Lower
Gogera 12583 12896 2 28796 32618 13 6821 9118 34
Burala 13015 13102 0.7 29101 32645 12 5611 9049 61
Overall 12148 12616 4 27669 317245 15 6044 8669 43
115
i. Cost of Production of Wheat Crop
It is evident from Table 6.10 that the real cost of production of wheat crop for all the cost
items included in the analysis have increased in the post-reform period except the real
cost of surface irrigation and real cost of tube well irrigation. Increase in the real
fertilizer cost was mainly due to the increase in real prices of different types of fertilizers.
Nominal cost of urea (an important nitrogenous fertilizer) had increased from Rs. 430 to
Rs. 530. In real terms price of urea fertilizer has increased from Rs. 402 to Rs. 418.
Nominal price of the DAP fertilizer (an important phosphatic fertilizer) had increased
from Rs. 950 to Rs. 1150. While real prices of DAP had increased from Rs. 888 to Rs.
906. Same was the case with the prices of diesel that was one important reason for the
increase in the real cost of mechanized operations.
Cost of surface irrigation and cost of tube well irrigation had decreased in the post-reform
period. It should be kept in mind here that nominal cost of surface irrigation (water rates
or Aabiana) was fixed in the year 2004 on per acre basis (flat rate charges). Cost of tube
well irrigation had also decreased in the post-reform period (year 2006) in spite of the
fact that diesel prices had increased in this year as compared to the year 2004.
Comparing the real average variable cost per acre across the different canal divisions
showed that Khanki canal division had increased more than the other canal divisions.
Cost of production had increased about 14 percent in Khanki division while it increased
12, 12 and 11 percent respectively in Upper Gogera canal division, Lower Gogera canal
division and Burala canal division respectively. Khanki Canal division had the highest
average real cost of production as for as the wheat crop was concerned. In both periods
(pre and post-reform period), it had highest cost comprising Rs. 5642 and Rs. 6468
respectively.
Fertilizer cost was the main component of cost of production of wheat crop. It comprised
about 25 to 26 percent in the pre-reform period and 28 percent to 30 percent in post-
reform period. Cost of mechanized operation was the second main component of the cost
of production of wheat crop. It covered about 22 percent to 24 percent of the cost of
116
production in the pre-reform period while it covered about 22 percent to 25 percent of the
cost of production in the post-reform period.
Chemical cost was the smallest component of the cost of production as it covered only 3
percent to 5 percent of the cost of production in the pre-reform period and 4 percent to 6
percent in the post-reform period in different canal divisions across the study area.
Variable cost of production of wheat crop in pre and post-reform period is given in the
Table 6.10.
ii. Cost of Production of Rice Crop
Table 6.11 given below shows the detailed cost of production with major components of
variable cost in respect of rice crop. Rice is grown in the Kharief season starting from
June until the middle of November.
Cost of each component included in the analysis showed an increase in the post-reform
period except the real cost of surface irrigation and real cost of tube well irrigation which
showed same pattern as the case of wheat crop. Real fertilizer cost increased from Rs.
1510, Rs. 1475, Rs. 1516 and Rs. 1454 in Khanki, Upper Gogera, Lower Gogera and
Burala canal divisions respectively to Rs. 2032, Rs. 1945, Rs. 1835 and Rs. 1688
respectively. The increase in cost of fertilizer was mainly due to increase in real prices of
both nitrogenous and phosphatic fertilizers as have been discussed in previous section.
Fertilizer cost increased about 35 percent in Khanki canal division as compared to 16
percent increase in the Burala canal division.
Increase in the mechanization cost was also high. Cost of mechanized operations
increased about 53 percent in the Khanki canal division and Burala canal division. One
reason for increase in the cost of mechanization was the higher diesel fuel prices in the
year 2006 as compared to the year 2004.
117
Table 6.10: Variable Cost of Production of Wheat Crop Across the System in Pre and Post-reform Period (Rs. per acre)
Canal Division
Seed Cost Fertilizer
Cost
Surface
Irrigation
Cost
Tube Well
Irrigation Cost
Chemical
Cost
Mechanization
Cost Labour Cost Overall
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Khanki 365 428 1436 2023 40 34 1173 953 233 271 1375 1616 1020 1143 5642 6468
Upper
Gogera 380 402 1308 1570 40 34 814 805 270 312 1111 1234 1029 1169 4952 5526
Lower
Gogera 394 420 1363 1629 40 34 923 885 273 324 1234 1389 1134 1312 5361 5993
Burala
395 410 1363 1572 40 34 808 750 296 331 1143 1304 1107 1283 5152 5684
1. Cost of production does not include opportunity costs.
2. Cost of family labour is not included in the cost items.
3. Real prices of inputs are taken by using GDP deflator for the year 2001-02.
118
Cost of surface irrigation and cost of tube well irrigation had decreased in the post-reform
period. It should be kept in mind here that nominal cost of surface irrigation (water rates
or Aabiana) was fixed in the year 2004 on per acre basis. Cost of tube well irrigation had
declined in the post-reform period (year 2006). Cost of tube well irrigation had a decrease
of about 14 percent in the Burala canal division while reduction was negligible in the
Lower Gogera canal division. This reduction in tube well irrigation cost was in spite of
the fact that diesel prices had increased in the year 2006 as compared to the year 2004.
Comparing the real average variable cost per acre across the different canal divisions
showed that real average cost of production for rice crop across the canal divisions had
recorded an increase ranging from 9 percent in Lower Gogera division to 14 percent in
Khanki canal division in the year 2006 (post-reform period) as compared to the year 2004
(Pre-reform period). In Upper Gogera and Burala canal divisions, the increase was about
11 percent.
Tube well irrigation cost was the main component of the total variable cost for the rice
crop. It comprised of about 36 percent to 49 percent in the pre-reform period (36 percent
in the Lower Gogera canal division and 49 percent in Khanki canal division) and 32
percent to 42 percent in post-reform period (in Lower Gogera and Khanki canal divisions
respectively). Fertilizer cost the rice crop accounted for only 18 percent to 22 percent in
the pre-reform period and 20 percent to 24 percent in the post-reform period. Cost of
mechanized operation was another important component in the production of the rice
crop. It covered about 14 percent to 18 percent of the cost of production in the pre-reform
period while it covered about 17 percent to 21 percent of the cost of production in the
post-reform period.
Seed cost was the smallest component of the cost of production for the rice crop. It
covered only about 2 percent of the cost of production in the pre-reform period as well in
the post-reform period. Surface irrigation cost decreased from 0.84 percent to 1 percent in
for the rice crop in pre-reform period and 0.6 percent to 0.8 percent in the post-reform
period. Cost of production of the rice crop in the area of LCC (East) in pre and post-
reform period is shown in Table 6.11.
119
Table 6.11: Variable Cost of Production of Rice Crop Across the System in Pre and Post-reform Period (Rs. per acre)
1. Cost of production does not include opportunity costs.
2. Cost of family labour is not included in the cost items.
3. Real prices of inputs are taken by using GDP deflator for the year 2001-02.
Canal Division
Seed Cost Fertilizer
Cost
Surface Irri
Cost
Tube Well
Irrigation Cost
Chemical
Cost
Mechanization
Cost Labor Cost
Total Variable
Cost
Pre-
Refo
rm
Post-
reform
Pre-
Refo
rm
Post-
reform
Pre -
Refo
rm
Post-
reform
Pre -
Refo
rm
Post-
reform
Pre -
Refo
rm
Post-
reform
Pre-
Refo
rm
Post-
reform
Pre-
Refo
rm
Post-
reform
Pre-
Refo
rm
Post-
reform
Khanki 168 198 1510 2032 70 59 4085 3907 470 612 1134 1732 839 869 8276 9404
Upper
Gogera 135 159 1475 1945 70 59 3150 2950 415 408 1118 1370 975 1218 7338 8104
Lower
Gogera 113 135 1516 1835 70 59 2504 2500 426 348 1269 1368 1045 1286 6932 7517
Burala
108 125 1454 1688 70 59 3786 3262 282 427 1200 1843 958 1303 7858 8702
120
iii. Cost of Production of Sugarcane Crop
Sugarcane was the third important crop of the area under study. Sugarcane crop is an
annual crop as it takes about a year or so to vacate the land.
Table 6.12 shows the components of real average cost of production for producing 1 acre
of the sugarcane crop. The sugarcane crop required large quantity of seed to be used so
the seed cost calculated using the real prices was the largest component of average
variable cost (in real terms) to grow one acre of sugarcane. Seed cost average around 26
percent in the pre-reform period while it accounted for 28 percent in the post-reform
period. Cost of fertilizer also showed an increase in the post-reform period. Fertilizer cost
had an increase of 4 percent to 9 percent across the canal divisions in the study area. Real
fertilizer cost increased by 4 percent in Upper Gogera canal division while it recorded an
increased 9 percent in the Burala canal division.
Tube well irrigation cost was another important component of the cost of production of
the sugarcane crop. It accounted for 20 percent (in Burala canal division) to 28 percent
(Lower Gogera canal division) in the pre-reform period. While in post-reform period this
had declined to 18 percent to 24 percent in the same canal divisions respectively. Tube
well irrigation cost decreased up to 8 percent in the post-reform scenario for the sugar
cane crop. Labor cost accounted for 17 percent to 19 percent in the pre-reform period
while in post-reform period, it accounted for 18 percent to 20 percent of the average total
cost of production to raise one acre of sugarcane crop.
121
Table 6.12: Variable Cost of Production of Sugarcane Crop Across the System in Pre and Post-reform Period (Rs. Per acre)
Canal Division
Seed Cost Fertilizer
Cost
Surface Irri
Cost
Tube Well
Irrigation Cost
Chemical
Cost
Mechanization
Cost Labor Cost Overall
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Pre-
reform
Post-
reform
Khanki 4165 4305 2265 2460 112 94 3875 3585 388 397 1790 1965 2580 2885 15175 15691
Upper
Gogera 4342 4242 2303 2401 112 94 3895 3661 311 336 1890 1994 2659 2955 15512 15683
Lower
Gogera 4388 4231 2400 2659 112 94 4751 4379 389 435 2243 2713 2703 3123 16986 17634
Burala
4502 4313 2539 2760 112 94 3295 3111 440 545 2311 2694 3129 3406 16328 16923
1. Cost of production does not include opportunity costs.
2. Cost of family labour is not included in the cost items.
3. Real prices of inputs are taken by using GDP deflator for the year 2001-02.
122
iv. Comparison of Cost of Production of Major Crops Across the System
Table 6.13 given below shows the summary of the cost of production discussed in the
previous sections. Table 6.13 depicted that in case of wheat and sugarcane crop average
increase in the total variable cost of production estimated in real prices had been up to 11
percent to 12 percent while for rice crop increase in cost was about 3 percent. Maximum
increase in cost was in Khanki canal division ranging from 4 to 14 percent. One reason
might be larger land holdings of the farmers in Khanki canal division as compared to
other canal divisions in the study area.
Table 6.13: Average Cost of Production (COP) of Major Crops Across the System
(Rs. per acre)
1. Cost of production does not include opportunity costs.
2. Cost of family labour is not included in the cost items.
3. Real prices of inputs are taken by using GDP deflator for the year 2001-02.
Canal
Division
Wheat Sugarcane Rice
Pre -
Reform
Period
Posr-
reform
period
%
change
Pre-
Reform
period
Post-
Reform
Period
%
change
Pre-
Reform
Period
Post-
Reform
period
%
change
Khanki 5642 6468 14 15175 15691 4 8276 9404 14
Upper
Gogera 4952 5526 12 15512 15683 2 7338 8104 11
Lower
Gogera 5361 5993 12 16986 17634 4 6932 7517 9
Burala 5152 5684 11 16328 16923 4 7858 8702 11
Average 5276 5917 12 16044 16482 3 7601 8431 11
123
0
1000
2000
3000
4000
5000
6000
7000
Khanki Upper Gogera Low er Gogera Burala
Canal Division
(Rs.)
Pre Reform Period Post Reform period
Figure 6.4: Average Cost of Production of Wheat Crop
Across the System in Pre and Post-reform Period
13500
14000
14500
15000
15500
16000
16500
17000
17500
18000
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Rs
.)
Pre-reform Period Post-reform Period
Figure 6.5: Average Cost of Production of Sugarcane
Across the System in Pre and Post-reform Period
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Rs.)
Reform Period Post Reform period
Figure 6.6: Average Cost of Production of Rice Crop
Across the System in Pre and Post-reform Period
124
6.2.6 Comparison of Gross Margin of Major Crops Across the System in Pre and
Post-reform Period
To assess the impact of IMT programme on agricultural and economic productivity, gross
margin of major crops was compared in pre and post-reform periods. Gross margin was
calculated by deducting variable cost from the gross income of the farm on per acre basis.
Table 6.11 reveals that there was an increasing trend in gross margin of wheat across the
system but with minor variation from one canal division to the other. Per acre gross
margin of wheat was highest in Khanki division and lowest in Burala division. It was
found that per acre increase in gross margin of wheat in Khanki canal division was about
Rs. 195 showing increase of about 4 percent. In Burala canal division per acre increase in
gross margin was Rs. 553 with 6 percent increase. This high gross margin in the Burala
canal division was attributed to higher per acre yield and higher price of output as
compared to situation before IMT. It also clearly depicted that the farmers were able to
get more quantity of surface irrigation water after irrigation management transfer in the
study area. Per acre percent increase in gross margin of wheat in Khanki, Upper Gogera
and Lower Gogera division was about 4, 4 and 7 percent respectively showing an upward
trend in gross margin in post-reform period. On an over all basis gross margin increased
by 6 percent in case of wheat crop after IMT.
Similar increasing trend was observed in sugarcane across the whole length of the LCC
(East) irrigation system. On an overall basis, increase in gross margin of sugarcane was
found to be about 38 percent, with highest increase in Lower Gogera canal division. The
percent increase in gross margin in Khanki, Upper Gogera, and Lower Gogera and Burala
canal division was 40, 41, 43 and 31 percent respectively. This increase in gross margin
was attributed to higher yield in sugarcane, reasonable prices of output and sufficient
surface water supply at the Tail end of the distributaries.
It was also observed that gross margin increased in case of rice crop in the study area. It
was found that per acre gross margin was highest in Burala canal division which was
about 61 percent and lowest in Lower Gogera canal division. When the farmers were
asked about this highest per acre return in Burala division they showed their complete
satisfaction on the working of FOs. It was also observed that farmers shifted from wheat
125
cultivation to rice on account of sufficient surface water availability at the Head, Middle
and Tail of the distributary because support price of rice was higher than wheat crop in
Punjab. The farmers made the judicious use of all the inputs alongwith the major
production input i.e. surface water. It was also observed that surface water increased the
fertility of the soil as compared to tube well water that was previously used for irrigation,
resultantly more per acre yield. The above Table also revealed that on overall farm basis
increase in gross margin was about 31 percent across the LCC (East) irrigation system
showing that reforms have significant impact on crop productivity.
Table 6.14: Average Gross margin (GM) of Major Crops Across the System in Pre
and Post-reform Period (Rs. per acre)
Canal
Division
Wheat Sugarcane Rice
Pre-
reform
Period
(Rs.)
Post-
reform
period
(Rs.)
%
change
Pre-
reform
period
(Rs.)
Post-
reform
Period
(Rs.)
%
change
Pre-
reform
Period
(Rs.)
Post-
reform
period
(Rs.)
%
change
Khanki 5461 5656 4 10257 14380 40 5914 8260 40
Upper
Gogera 6240 6475 4 10920 15436 41 5828 8250 42
Lower
Gogera 6918 7391 7 11264 16093 43 6821 9118 34
Burala 7335 7888 8 12657 16523 31 5611 9049 61
Average 6489 6853 6 11275 15608 38 6044 8669 43
126
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Rs.)
Pre-reform Period Post reform period
Figure 6.7: Comparison of Gross Margin of Wheat
Crop Across the System in Pre and Post-reform Period
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Rs.)
Pre-reform period Post-reform Period
Figure 6.8: Comparison of Gross Margin of Sugarcane
Crop Across the System in Pre and Post-reform Period
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Rs.)
Pre-reform Period Post-reform period (RS)
Figure 6.9: Comparison of Gross Margin of Rice Crop
Across the System in Pre and Post-reform Period
127
6.2.7 Ratio of Gross Margin (GM) to Cost of Production (COP) Across the System
in Pre and Post-reform Period
Gross margin (GM) to cost of production (COP) ratio depicts the cost-benefit ratio of the
specific farmer. It depicts returns per unit of money spent by the farmer for incurring
inputs / variable costs. It is well evident from the Table 6.15 that each rupee spent for
incurring variable costs in post-reform scenario in the study area earned a higher return
than that of the return per rupee in the pre-reform period. Highest rate of return had been
observed in case of the rice crop in Lower Gogera and Burala canal divisions as
compared to other canal divisions and other crops. Comparing the figures for sugarcane
in pre as well post-reform period, GM to COP ratio increased from 0.68 to 0.97 showing
an increase of about 43 percent. Comparing the overall ratio of rice in the LCC (East)
irrigation system GM to COP ratio has increased from 0.85 to 1.34 showing an increase
of about 58 percent. Thus farmers as a whole in LCC (East) system has earned more for
each rupee spent on incurring inputs/variable costs after irrigation reforms. This better
rate of return was very vital for the farmers who were generally poor and scarce in
resources. It is also worth mentioning here that GM to COP ratios calculated in this study
are in line with those mentioned by Hussain et al. (2003).
Table 6.15: Ratio of Gross Margin to Cost of Production Across the System in Pre
and Post-reform Period
Canal
Divisions
Wheat Sugarcane Rice
Pre-
reform
Period
Post-
reform
period
Pre-
reform
Period
Post-
reform
period
Pre-
reform
Period
Post-
reform
period
Khanki 0.87 0.88 0.61 0.90 0.79 1.15
Upper
Gogera 1.18 1.21 0.70 0.99 0.78 1.20
Lower
Gogera 1.22 1.34 0.64 0.96 1.06 1.54
Burala 1.41 1.38 0.77 1.02 0.79 1.54
Overall 1.16 1.19 0.68 0.97 0.85 1.34
128
6.2.8 Cropping Intensity in the Study Area in Pre and Post-reform Period
Cropping intensity is defined as the ratio of gross cultivated area to designed command
area (Hussain et al. 2003) Table 6.16 summarizes the change in cropping intensity in the
study area before and after irrigation management transfer. It depicts that reforms in
irrigation sector have positive impact on cropping intensity. The highest cropping
intensity estimated at Khanki (13%) and Upper Gogera (13%) with lowest at Lower
Gogera and Burala (12%). There was an increasing trend found in all the four canal
divisions after IMT. On an overall basis, in the whole system of LCC (East), two years
after transfer of management cropping intensity changed from 163 percent to 182 percent
leading 12 percent increase after IMT. This suggests that there is discernible difference
between pre and post transfer period.
Table 6.16: Comparison of cropping Intensity of the Study Area in Pre and Post-
reform Period
Canal Division Pre-reform Period
(%)
Post-reform
period (%) % change
Khanki 156 176 13
Upper Gogera 165 186 13
Lower Gogera 168 188 12
Burala 164 179 12
Overall 163 182 12
0
20
40
60
80
100
120
140
160
180
200
Khanki Upper Gogera Lower Gogera Burala
Canal Division
(Per
cen
tag
e)
Pre-reform Period Post reform period (%)
Figure 6.10: Comparison of Cropping Intensity of the Study Area in
Pre and Post-reform Period
129
6.3 Comparison of Indicators Developed From Secondary Data Sources
The third part of this chapter consists of analysis of data based on secondary information
collected from different Farmer Organizations (FOs) working in the field. The
information collected was related to post irrigation management transfer period. Pre-
reform data were collected from the Punjab Irrigation Department (PID). Secondary data
was collected for four years, starting from 2003 to 2006. Pre-reform period comprised of
two years i.e. 2003 and 2004, and post-reform period consisted of two years i.e. 2005 and
2006. Pre-reform and post-reform situation was compared by using different set of
indicators as discussed in earlier chapter.
6.3.1 Comparison of Aabiana Assessment and Collection in Pre and Post-reform
Period
Aabiana or water charges is an important component of irrigation system. Institutional
reforms introduced in the LCC (East) have addressed this component. The World Bank
(1994) reported that in Pakistan Aabiana collection was only 44.4 percent and the system
was not financially sustained. The government realized that financial sustainability could
be achieved only through increased Aabiana collection. Thus after handing over the
system to the stakeholders, emphasis was made on improving the Aabiana collection.
Table 6.17 shows the performance of sampled distributaries for Aabiana assessment and
collection in pre (2003 and 2004) and post-reform period (2005 and 2006). It is evident
from the results of the study that Aabiana recovery percentage increased from 43 percent
before IMT to 62 percent after IMT, thus showing an increase of about 44 percent in the
post-reform period. This result is in accordance with previous study conducted by Haq
(1998). Table 6.17 also shows that before introducing reforms in irrigation sector of the
Punjab, Aabiana collection in Burala division was only 30.8 percent in the year 2003-
2004 which was lowest in the LCC (East) irrigation system. However after reforms, due
to better management of FOs, Aabiana collection rose to 65 percent in the same canal
division. Table 6.17 also shows that in pre-reform period Aabiana collection rate in
Lower Gogera division was about 53 percent which was highest among the canal
divisions.
130
Table 6.17: Overall Comparison of Aabiana Assessment and Collection in Pre and
Post-reform Period
Canal
Divisions
CCA
(ha)
Pre-reform period Post-reform period
Aabiana
assessed
(Rs. 000)
Aabiana
collected
(Rs. 000)
%age
Aabiana
collected
Aabiana
assessed
(Rs. 000)
Aabiana
collected
(Rs. 000)
%age
Aabiana
collected
Khanki 6900 3380.8 1364.7 40.3 3292.3 194 58.8
Upper
Gogera 100800 60792 25750.7 42.3 63350.4 3313 53
Lower
Gogera 113933 60302.8 32158.1 53.3 55874.9 3925 70.2
Burala 82068 46159.5 14244.1 30.8 45965.9 2995 65.1
Total 303701 170635.1 73517.6 41.67 168483.5 10427 61.80
While collecting data in the field it was observed that this high Aabiana collection rate in
Lower Gogera canal division was due to better performance of the PID staff responsible
for Aabiana collection. Another reason behind this reasonable Aabiana collection was
sufficient and reliable quantity of water availability at this division as indicated in the
Table in the next sections.
In the post-reform period, percentage of Aabiana collection increased in all the four canal
divisions under study as shown in Table 6.17. In Khanki division Aabiana collection
raised from 40 percent to about 59 percent. Whereas, in Upper Gogera, Lower Gogera
and Burala division it increased from 40 to 53 percent, 53 to 70 percent and 39 to 65
percent respectively. Thus, Table 6.17 clearly indicates that after irrigation reforms there
was significant increase in Aabiana recovery showing that the performance of FOs was
comparatively better than the system working under PID. The results of the study are
inline with the studies conducted by Yercan (2003) and Awosola (2005) in which they
concluded that water charges collection rate increased after management transfer to the
stakeholders. While collecting data from the field, it was found that some FOs/KPs were
poorly managed on account of the fact that FO Presidents were not qualified. They were
unable to collect Aabiana from the KP Chairmen and proper management of office
record. The record of FOs showed that during the first two crops after IMT (year of
131
IMT), Aabiana collection was even more than 90 percent on some distributaries but with
the passage of time the collection percentage dropped to 65 to 70 percent due to loose
administration of FOs and poor legal support from the government. It was also observed
that in most of the cases KP Chairman collected the Aabiana but did not deposited in the
government treasury and keep hold with them. It was also found that one of the reasons
for decreasing trends in Aabiana collection after reforms was that KP chairmen were
given the responsibilities of Aabiana collection but they were paid only a meager amount
of 6 percent incentive for 100 percent Aabiana collection. No one was paid even 6
percent incentives as no body was able to collect 100 percent Aabiana. For financial
sustainability of the system it is therefore necessary to increase the present Aabiana
collection in Punjab. However, the LCC (East) system as a whole showed significant
achievement in Aabiana collection in post-reform period.
6.3.2 Comparison of Per Hectare Operation and Maintenance (O&M) Expenditures
in Pre and Post-reform Period
Operation and maintenance of the system is an important aspect for the sustainability of
irrigation system. Irrigation system in Punjab has been ignored in the past and genuine
needs of the O&M have not been fulfilled (Haq 1998). Table 6.18 given below shows
performance of two systems under study in respect to O&M expenditures. O&M
expenditures include general maintenance and operation expenditures incurred on the
distributaries excluding salaries of the staff and expenditures on main canal. In the pre-
reform period there were a wide variation in the O&M expenditure/ha across the system.
Khanki division in LCC (East) system received major portion of the O&M expenditures
as compared to other canal divisions. In the post-reform period this variation across the
system has been minimized. The additional amounts spent in the Burala division were
due to the fact that Govt. of Punjab has sanctioned special funds for the maintenance of
Killianwali distributary due to its deteriorated situation. Another interesting fact is that in
the new system of reforms, each FO has been authorized to spend only 30 percent of the
total Aabiana collected for the purpose of O&M but actual expenditures incurred were
more than the allocated amount which came through from the savings of staff salaries and
132
contingency expenditures. The farmer survey which was carried out in the study area,
about 65 percent of the respondents in the reform area responded the existing condition of
O&M of the distributaries as better than the situation already existed under the provincial
irrigation department. While only 15 percent were of the opinion that previous system of
O&M was better. Table 6.18 shows the skewed pattern of operation and maintenance
(O&M) expenditures before the reforms. The study showed that average per hectare
O&M expenditures in Khanki. Upper Gogera, Lower Gogera and Burala division were
Rs. 147, Rs. 66, Rs. 57, and Rs. 41 respectively. It is clear from Table that O&M
expenditures incurred by the irrigation department were not uniformly spent in all the
canal divisions. In Khanki division per hectare expenditures were Rs. 147 and in Burala
division per hectare expenditures were only Rs. 41 which shows irrational decision of
irrigation department. However after the management transfer O&M expenditures were
need based. It is also clear from Table that most of the Aabiana collected in the pre-
reform period was utilized for O&M of the distributaries and there was no deposit in the
government exchequer. However after the irrigation management transfer significant
amount of Aabiana collection was deposited in the government exchequer. The above
results of the study are in conformity with the study conducted by Smad (2005).
Table 6.18: Comparison of per Hectare Operation and Maintenance (O&M)
Expenditures in Pre and Post-reform Period
Canal
Division
Actual O&M Expenditures/hectare
Pre-reform period Post-reform period
Actual O&M
expenditures
(Rs. 000)
Av.
O&M
expendi-
tures /
hectare
(Rs.)
Annual
O&M exp
as %age
of annual
Aabiana
collected
Actual O&M
expenditures
(Rs. 000)
Av. O&M
expenditures
/ hectare
(Rs.)
Annual
O&M exp
as %age
of annual
Aabiana
collected
Khanki 1014.5 147.0 74.3 326.3 47.3 16.9
Upper
Gogera 704.4 66.0 27.4 4044.8 40.1 12.2
Lower
Gogera 4998.9 57.0 15.5 8083.6 79.7 20.5
Burala 3389.4 41.0 23.8 5560.6 50.6 18.6
133
6.3.3 Per Hectare Salary Expenditures in Pre and Post-reform Period
A direct consequence of transfer process is the reduction in the staff position responsible
for managing distributaries. Salaries expenditures of the operational staff per unit of
irrigated area is another major indicator to assess the performance of the irrigation
systems. In Table 6.19, salary expenditures for the two years of pre-reform period and
two years of post-reform period were compared. Salary expenditures included the
monthly pays of the regular employees and those working on contractual basis. It is
beyond any doubt that irrigation department recruited a large number of employees
without taking in account the workload. Average salary expenditures per hectare before
reforms in Khanki canal division were the highest among all canal divisions. Whereas, in
Burala canal division per hectare salary expenditures were Rs.21. After irrigation
management transfer, there was significant decrease in such expenditures as is evident
from Table 6.19. In Khanki canal division average salary expenditures decreased to Rs.
28 per hectare. Thus showing 60 percent decrease in Khanki division. In the Upper
Gogera canal division average salary expenditures per hectare before reforms were Rs. 46
that decreased to Rs. 30 per hectare after irrigation reforms thus showing a decrease of 35
percent. Similar decreasing trend in salary expenditures was observed in the Lower
Gogera and Burala canal division. The above results show that reforms resulted in
lowering government expenditures. The results of this study are more or less similar to
those presented by Samad (2005).
Table 6.19: Per Hectare Salary Expenditures in Pre and Post-reform Period
Canal Division Staff salary expenditures / hectare
Pre-reform period Post-reform period
Actual salary
expenditures
(Rs.)
Av. salary
expenditures/
hectare
Actual salary
expenditures
(Rs.)
Av. salary
expenditures/
hectare
Khanki 483594 70 194345 28
Upper Gogera 4944364 46 3242127 30
Lower Gogera 5311456 47 4212720 37
Burala 1710322 21 1142983 14
134
6.3.4 Per Hectare Contingency Expenditures in the Pre and Post-reforms Period
Contingency expenditure is an important indicator for measuring the financial
sustainability of new institution. It includes the expenditures for office maintenance,
purchase and repair of equipments, office stationary, electricity and telephone bills and
all other expenditures required for smooth running of the office in a particular financial
year starting from July to June. These expenditures were collected from both Punjab
irrigation department (PID) and Farmer Organizations (FOs) for comparison purpose.
The analysis shows that average contingency expenditures on per hectare basis ranged
from Rs. 21 to Rs. 44 before the irrigation management transfer. After irrigation
management transfer, Farmer Organizations considerably reduced contingency
expenditures ranging from Rs. 6 per hectare to Rs. 13 per hectare. The results of the study
as summarized in Table 6.20 given below showed that after irrigation management
transfer there was 77 percent reduction in contingency expenditures in Khanki division as
compared to pre transfer period. Similarly, in Upper Gogera, Lower Gogera and Burala
division, percent decrease was 82 percent, 60 percent and 57 percent respectively. It is
evident from above discussion that a significant reduction in contingency expenditures
mean saving in government revenue.
6.20: Per Hectare Contingency Expenditures in the Pre and Post-reforms Period
Canal
Division
Contingency expenditures per hectare
Pre-reform period
Post-reform period
Annual
Contingency
exp. as %age
of annual
Aabiana
collected
Actual
contingency
expenditures
(Rs.)
Av.
Contingency
expenditures/
hectare
Actual
contingency
expenditures
(Rs.)
Av.
Contingency
expenditures/
hectare
Khanki 300602 44 70103 10 5.1
Upper
Gogera 3580358 34 593441 6 1.8
Lower
Gogera 3731225 33 1511774 13 3.8
Burala 1749806 21 706561 9 2.3
135
6.3.5 Delivery Performance Ratio of Selected Distributaries in Pre and Post-reform
Period
Delivery performance ratio (DPR) is defined as the ratio of actual to target (sanctioned)
volume of water delivered. Table 6.21 shows the delivery performance ratio at Head and
Tail of the selected distributaries before and after irrigation management transfer. It is
clear that there was large variation in DPR at Head and Tail of the selected distributaries
before the reforms. However, after reforms this variation in DPR was significantly
reduced. The study showed that in the year 2003 and 2004, average DPR at the Head of
Khanki, Upper Gogera, Lower Gogera and Burala canal division was 0.68, 0.59, 0.62 and
0.66 respectively. On the other hand, DPR at the Tail of the above mentioned canal
divisions were amazingly very low. The DPR at the selected distributaries of the Khanki
division was 0.1, which shows that most of the Tails were almost dry. After reforms, the
share of water at the Head of Khanki was reduced, whereas Tail were receiving sufficient
amount of water. It shows that reform process played a significant role for the Tail
enders. Similar pattern of water distribution at Head and Tail of the other distributary was
observed before the initiation of the institutional reforms in irrigation sector. The analysis
of the study also showed that after introducing reforms in year 2005, the Tails of the
distributaries which were deficient in water, those distributaries started getting water. It is
evident from Table 6.21 that the average DPR at the Tails of the selected distributaries of
LCC (East) irrigation system was only 0.19 which rose to 0.36 showing that Tails were
comparatively in a better position after reforms. The results of this study are in line with
the study conducted by Latif and Pomee (2003) in which they mentioned that most of the
outlets were drawing 50 to 65 percent more water than their due share and some other
outlet would certainly be drawing less than their due share. Thus we can conclude that
reforms have positive impact in terms of improving water delivery at Tails of the
distributaries. The results of the study are also in accordance with the study conducted by
Nakashima (1998) in which he mentioned that there were fewer fluctuations observed at
the Tail end of the distributaries and outlets of the watercourses after reforms compared
to corresponding pre-transfer period. Dick (1996) endorsed the above mentioned
statement that the equity in water distribution improved after the reforms in many
developed and developing countries of the world.
136
6.3.6 Comparison of Head-Tail Equity in Water Distribution on the Selected
Distributaries in Pre and Post-reform Period
Head-Tail equity with respect to discharge is defined as the ratio of water delivery
performance of Head reach to water delivery performance of Tail reach of the
distributaries. The discharges of selected distributaries at the Head and Tail reaches were
compared to calculate the above mentioned indicator. Equity does not mean equal water
supplies to the stakeholders but proportionate and fair share of irrigation water to all the
stakeholders regardless of their location along the distributary (Latif and Pomee 2003).
Table 6.22 shows the Head-Tail equity of the selected distributaries before and after
irrigation management transfer. It is clear from the Table 6.22 that there was significant
inequity in water distribution at Head and Tail of the distributaries before the irrigation
management transfer. However, the situation improved after the irrigation management
transfer.
137
Table 6.21: Comparison of Delivery Performance Ratio (DPR) at Head and Tail of the Selected Distributaries in Pre and
Post-reform Period
Canal
Division
Delivery Performance Ratio (Pre-reform)
Delivery Performance Ratio (Post-reform)
DPR 2003 DPR 2004 AV.DPR
(2003-04)
DPR 2005
DPR 2006 AV. DPR
(2005-06)
Head Tail Head Tail Head Tail Head Tail Head Tail Head Tail
Khanki 0.74 0.09 0.71 0.19 0.68 0.1 0.54 0.61 0.61 0.31 0.58 0.46
Upper
Gogera 0.66 0.20 0.56 0.25 0.59 0.19 0.62 0.30 0.66 0.27 0.64 0.29
Lower
Gogera 0.70 0.38 0.61 0.36 0.62 0.34 0.66 0.51 0.69 0.41 0.68 0.46
Burala 0.75 0.15 0.57 0.11
0.66
0.14 0.62 0.16 0.66 0.25 0.64 0.21
138
Table 6.22 shows that before irrigation reforms, on an average of 2003 and 2004, highest
value estimated was found for Khanki canal division, indicating higher inequity in water
distribution as compared to an estimate of 1.71 for Lower Gogera, showing equitable
distribution of water at various reaches of distributaries. However, after reforms, Khanki
division showed an excellent example of Head-Tail equity ratio, as the estimated
computed was around 1.92. For the distributaries of lower Gogera, the Head-Tail equity
ratio was 1.71 before reforms and 1.51 after reforms, indicating almost equitable
distribution of water before and after reforms. However, the Head-Tail estimate for the
whole system before introducing reform was 3.94, and this figure was 2.26 after reforms.
The higher Head-Tail equity estimate than the perfect equality estimate of 1 was due to
unauthorized or illegal modification in the size of outlets at the Head and Middle of the
distributaries.
Table 6.22: Comparison of Equity in Water Distribution on the Selected
Distributaries in Pre and Post-reform Period
Canal Division Year
2003
Year
2004
Average
(2003-04)
Year
2005
Year
2006
Average
(2005-06)
Khanki 8.21 3.72 5.96 1.86 1.97 1.92
Upper Gogera 3.28 2.24 2.76 2.07 2.49 2.28
Lower Gogera 1.87 1.71 1.79 1.30 1.71 1.51
Burala 5.05 5.46 5.26 3.99 2.64 3.32
Overall Average 4.60 3.28 3.94 2.31 2.20 2.26
139
6.4 Summary
This chapter was divided in to two parts. In the first part, comparison was made by
developing indicators from primary data sources. Data collected from the farmers in the
study area was regarding average yield of major crops across the study area, average
gross value product, cost of production, comparison of gross margin of major crops and
cropping intensity of major crops in the study area in pre and post-reform period. It was
found that irrigation reforms had positive impact on the above mentioned variables.
Average yield increased and cost of production of major crops decreased in the study
area. All the prices were taken in real terms by using GDP deflator for the year 2001-02.
In the second part of this chapter, analysis of data was based on secondary information
collected from the FOs and Punjab Irrigation department. The data collected were about
comparison of Aabiana assessment and collection (water charges), operation and
maintenance of the system, salary and contingency expenditures, delivery performance
ratio and Head-Tail equity in pre and post-reform period. It was concluded that on overall
basis, Aabiana collection increased from 44 percent to 62 percent in pre-reform period.
Similar trend was also found in other variables.
140
CHAPTER 7
ESTIMATION OF REGRESSION MODELS
In order to estimate the impacts of irrigation reforms on farm productivity and farm
income various single equation models were estimated and analyzed as discussed earlier.
A Cobb-Douglas production function most widely used in agriculture was found an
appropriate representation of the data.
Primary data, used in analysis, was collected from 360 farmers in the study area. The
collected data were arranged and cleaned before actually estimating the OLS equations.
The analysis was carried out for three major crops in the area i.e. wheat, sugarcane and
rice. Results are presented below for each of the crops.
7.1 Results of Wheat Regression Models
Data of 360 farmers were obtained in the study area for the two years i.e. 2003-04 and
2005-2006. Collected data was then cleaned and actual analysis was carried out for 320
farmers. The remaining farmers were dropped out due to the reason that they have not
sown wheat in any of these two years. Rigorous analysis of data was carried out to
estimate the effect of irrigation reforms on farm income from wheat crop and wheat
productivity. Initially problem of multicollinearity was detected among different
variables when VIF and tolerance of each variable was analyzed. To remove this problem
and the problem of high R2 and high adjusted R
2, analysis of only non-negative log
values of different variables was carried out.
Table 7.1 given below shows the mean and standard deviation of each variable included
in the analysis for the wheat crop. Although real average cost per acre was not included
in the model, it is shown in Table 7.1. Average yield per acre of wheat has been included
in the regression model (that will be discussed in section 7.2 of this chapter) as the
dependent variable.
141
Table 7.1: Descriptive Statistics of Important Variables for Wheat Crop
Variables Mean Value Std. Deviation
Real average gross value product/acre
(Rs.) 12091 1727
Wheat Area (acres) 6.97 10.7
Real seed cost (Rs.) 397 44.6
Real fertilizer cost (Rs.) 1493 377
Real Surface Irrigation cost(Rs.) 37 10.1
Real Tube well Irrigation cost (Rs.) 862 434.6
Real mechanization cost (Rs.) 1254 348.2
Real labour cost (Rs.) 1142 430.9
Real average cost/acre (Rs.) 5442 1014.1
Average Yield (maunds/acre) 33 4.9
7.1.1 Estimation of Regression Model for Average Gross Value Product of Wheat
In this model average gross value product (GVP/acre) of the wheat crop (in real prices)
has been taken as dependent variable to estimate the effect of reform process. Important
components of variable cost i.e. seed cost, fertilizer cost, surface irrigation cost, tube well
irrigation cost, mechanization and labour cost have been included in the regression
model. Variables have been entered in the model as natural log form.
Table 7.2 summarizes the results of the regression model. The Cobb-Douglas model for
wheat was estimated using lnwagvp (Natural log of per acre GVP, calculated by using
real prices) as dependent variable. Value of R2 was 0.19 indicating that the independent
variables included in the production function explained about 19 percent of the variations
in the dependent variable i.e. per acre GVP of the wheat crop. Reasonable F-value also
depicted that the overall model was significant. The estimated coefficients (βs) of the
explanatory variables showed the percentage change in dependent variable with one
142
percent change in explanatory variable. Table 7.2 shows that three βs (lnwfcost, lnwlcost
and lnedu) were significant at about less than 5 percent level of significance and one β
(D1) was significant at about 10 percent significance level, while rest of the coefficients
were non-significant at 10 percent level or below. Cost of chemical application was not
included in the model as it was observed that in the study area use of chemicals
application for wheat was very limited and only few farmers were applying the chemicals
(herbicides spray) to the crop.
Coefficients of wheat area (lnwarea), surface irrigation (lnwsicost) and dummy variable
(D1) were negative. It was observed that each percent increase in the area under wheat
crop for each farmer could reduce the respective average GVP by 0.002 percent, although
the result was statistically non significant using t-statistics. Increase in the cost of surface
irrigation water could also decrease the average GVP. There was an interesting finding
that showed that without making any considerable improvements in the surface irrigation
water supply, increase in water rates (Aabiana) could reduce the GVP of the farmers in
the study area, although the coefficient was also statistically significant at 10 percent
significance level.
The coefficients of dummy variable D1 and D2 which were introduced to measure the
effects of location of the farm and impact of reforms on the real GVP of the farmers can
be interpreted by using the Halvorsen and Palmquist device (Gujarati 2003). This device
uses the eβi
value to estimate the percentage change in the mean value of intercept (i.e.
base category). In case of D1, coefficient showed that farmers at the Tail reaches of the
distributary had about 5 percent less GVP as far as wheat crop was concerned. The result
was significant at about 10 percent significance level. Similar conclusions had been
drawn by Hussain at al. (2003) suggesting that Tail ends of the distrubutaries were
characterized by majority of small, poor farmers in Pakistan.
In case of D2, the coefficients depicted that although it was statistically non-significant yet
it has a positive impact on the GVP of the farmers and has positive sign which was
according to a priori expectations. In case of wheat crop, it seemed quite logical as it has
been discussed in previous chapter that real price of the wheat was decreased from Rs.
373 to Rs. 352 in the year 2006 as compared to year 2004. Percentage change in D2 was
about 7 percent from that of the mean value of intercept.
143
Interaction term D1D2, was introduced in the model between two dummy variables. The
coefficient was statistically non-significant but sit has a positive sign which showed that
reforms have tilted the water equity towards the Tail end farmers.
Table 7.2: Estimated Parameter of the Income Model for Wheat Crop
Variables Parameter T Value Significance Level
Constant 8.34 19.05 .000*
lnwarea -0.002 -0.195 0.84
lnwscost 0.07 1.20 0.22
lnwfcost 0.073 2.52 0.01*
lnwsicost -0.018 -0.40 0.68
lnwticost 0.006 1.06 0.28
lnwmcost 0.017 0.68 0.49
lnwlcost 0.017 4.25 0.00*
lnedu 0.005 2.06 0.03*
D1 -0.053 -1.76 0.08**
D2 0.068 1.36 0.17
D1D2 0.026 1.01 0.31
R2 0.19
Adjusted R2 0.16
F- Value 3.7 .000*
* Significant at 5 percent level. ** Significant at 10 percent level.
144
7.1.2 Estimation of Regression Model for Average Yield of Wheat
In this model, natural log of average yield of the wheat crop (maunds/acre) has been
taken as dependent variable to capture the effect of reform process. Table 7.3 shows the
results of the yield/productivity impact model. Estimated coefficients were more or less
similar to that of previous model. R2 and adjusted R
2 have improved further showing that
22 percent change in the wheat yield was being explained by the explanatory variables.
These low values of R2 and adjusted R
2 were in line with many previous studies like
Abedullah and Pandey (2004) and Abedullah and Ali (2006) who calculated similar
values of R2 and adjusted R
2 in respective studies conducted in agriculture sector of
Pakistan.
Table 7.3: Estimated Parameter of the Yield Model for Wheat Crop
Variables Parameter T Value Significance Level
Constant 2.42 5.53 .000*
lwarea -0.002 -0.195 0.84
lrwscost 0.07 1.20 0.22
lrwfcost 0.073 2.52 0.01*
lrwsicost -0.018 -0.40 0.68
lrwticost 0.006 1.06 0.28
lrwmcost 0.017 0.68 0.49
lrwlcost 0.011 4.25 0.00*
lnedu 0.006 2.06 0.03*
D1 -0.053 -1.76 0.08**
D2 0.084 3.036 0.00*
D1D2 0.026 1.01 0.31
R2 0.22
Adjusted R2 0.19
F- Value 7.08 .000*
* Significant at 5 percent level. ** Significant at 10 percent level
145
One very important change in coefficient was of D2, dummy for reform process that
showed that average yield of the farmers increased by 9 percent as compared to the base
category and result was statistically significant at the 5 percent level.
The above two models have revealed that introduction of the reform has positive impact
on the GVP and yield of wheat crop of the selected farmers in the study area but there
was significant scope for irrigation services to improve further to have their more
significant and positive impact on income and productivity.
7.2 Results of Sugarcane Regression Models
Sugarcane was one of the important crops of the study area. Analysis was carried out for
190 valid cases out of the 360 cases. Remaining cases were not included in the regression
model because the farmers had not sown sugarcane in any of the two years of the study
period. Descriptive statistics of the important variables provided in Table 7.4.
It should be kept in mind that sugarcane is grown as an annual crop in Pakistan and
surface irrigation rates (Aabiana) have been fixed separately for the sugarcane crop by
the Government of Punjab. Table 7.4 also shows that variation in the sugarcane yield was
quite high in the study area. Chemical cost was again not included in the analysis due to
the reason that farmers as a whole had no inclination for using chemicals for sugarcane
crop.
146
Table 7.4: Descriptive Statistics of Important Variables for Sugarcane Crop
Variables Mean Value Std. Deviation
Real average gross value product (Rs.) 29727 4462.3
Sugarcane area (acres) 3.9 3.2
Seed cost (Rs.) 4317 348
Fertilizer cost (Rs.) 2464 538
Surface Irrigation cost (Rs.) 99 28.5
Tube well irrigation cost (Rs.) 3952 1836.7
Mechanization cost (Rs.) 2222 863.9
Labour cost (Rs.) 2921 920.1
Average cost/acre (Rs.) 16240 3176.1
Average Yield (maunds/acre) 606 81.8
Prices are used in real term
7.2.1 Estimation of Regression Model for Average Gross Value Product of
Sugarcane
In this model AGVP (GVP/acre) of the sugarcane crop (in real prices) is the dependent
variable. Different combinations of the explanatory variables were used and the following
model was found appropriate to measure the effect.
Table 7.5 shows the results of the regression model for sugarcane crop in which natural
log of real GVP is the dependent variable. The model was estimated to find out the
impact of major variable inputs on the crop income. Impact of reforms, location of the
farm especially the farms at the Tail reaches of the selected distributaries in connection
with sugarcane crop was also analyzed by introducing dummy variables and the
interaction terms in the regression model. Impact of formal education on the crop income
was also analyzed by introducing the variable lnedu (natural log of schooling years of the
farmers included in the sample).
147
R2 and adjusted R
2 values indicated that 36 percent variations in the explained variables
have been explained by the explanatory variables. F test also indicated that overall model
was significant at 99 percent confidence level. Model was also checked for the presence
of heteroscedasticity by using the White‘s test (WGHT) and by using White‘s
heteroscedasticity corrected standard errors. The results obtained were not of
considerable difference from those reported in Table 7.5. White‘s Test has also accepted
the null hypothesis of no heteroscedasticity.
Five coefficients of parameters reported above, were significant at the 5 percent level of
significance. Coefficient of lnsicost (log of real surface irrigation cost) was negative and
significant. It reemphasized the results of the model of wheat crop and depicted that with
present quality of irrigation services; one percent increase in the surface irrigation cost
for sugarcane crop could reduce the GVP of sugarcane by 0.014 percent. Coefficients of
fertilizer cost (lnsfcost) and cost of mechanized operations were positive and significant
at 5 percent level of significance. One percent increase in real fertilizer cost could
increase the income of the crop by 0.15 percent while one percent increase in the real cost
of mechanized operations could increase the crop income by 0.08 percent.
Dummy variable for location of the farm at the distributary was negative which was also
according to the a priori expectations. Previous studies also revealed that farmers at the
Tail reaches of the distributary were comparatively poor, scarce in resources and
comparatively inefficient due to their limited access to the surface irrigation water.
Second dummy variable was introduced to capture the impact of irrigation reforms on the
crop income. Coefficient was positive and significant showing that reform process has
positively contributed towards the sugarcane income. Farmers in the post-reform period
have about 11 percent more income as compared to the farmers in the base category
(Intercept with D2 = 0 for pre-reform period).
Interaction term was introduced to capture the impact of reforms on the Tail reach
farmers. The coefficient was positive as a priori expectations but insignificant. Years of
schooling of the farmers also had positive and significant impact on the crop income.
148
Table 7.5: Estimated Parameter of the Income Model for Sugarcane Crop
Variables Parameter T Value Significance Level
Constant 8.40 10.95 .000*
lnsarea 0.007 0.695 0.48
lnsscost 0.039 0.43 0.66
lnsfcost 0.157 4.55 0.00*
lnssicost -0.141 -4.23 0.00*
lnsticost 0.004 0.98 0.32
lnsmcost 0.08 4.50 0.00*
lnslcost 0.004 1.30 0.19
lnedu 0.005 2.06 0.03*
D1 -0.034 -1.26 0.20
D2 0.11 6.21 0.00*
D1D2 0.04 1.52 0.12
R2 0.39
Adjusted R2 0.36
F- Value 14.5 .000*
* Significant at 5 percent level. ** Significant at 10 percent level
7.2.2 Estimation of Regression Model for Average Yield of Sugarcane
To capture the impact of other factors along with the irrigation reform process carried out
in the study area on the productivity of rice crop, OLS model using average yield of rice
crop (lray) measured in maunds/acre as explained variable in the model. Different
combinations of the inputs and socio economic variables were tested and following form
of the Cobb-Douglas production function was found appropriate. Results are presented in
Table 7.6.
149
Results of the productivity model were also more or less similar to those of income
model in case of sugarcane crop. Model was free from the problems of heteroscedasticity,
auto-correlation, and multi-collinearity as respective tests like White‘s test, Durbon-
Watson statistics and VIF values suggested. R2 and adjusted R
2 were also in satisfactory
range. The F-value indicated that the overall results of the model were also significant.
Results suggested that five coefficients were significant at less than 5 percent level of
significance. Out of five coefficients, four were positively related with the dependent
variable i.e. natural log of the average yield per acre (maunds/acre). One variable i.e.
natural log of the surface irrigation cost, was having negative sign. Thus implying that
any increase in surface irrigation cost would reduce the yield of the sugar cane crop
(results suggested that one percent increase in surface irrigation cost could decrease the
yield by about 0.2 percent). Reform dummy was significant and having positive sign that
was according to the a priori expectations. Coefficient of the dummy revealed that yield
was about 60 percent than that of base category. Coefficient of fertilizer cost (lnsfcost)
was also significant showing that one percent increase in the fertilizer cost could increase
the sugarcane yield by 0.15 percent. Coefficient of the location dummy was negative but
non significant.
Interaction term of the two dummies was also having positive sign and was significant at
about 12 percent. That showed that reforms were more in the favour of the farmers
located at the Tail clusters of the selected distributaries than the farmers located at Head
or Middle of the distributaries.
150
Table 7.6: Estimated Parameter of the Yield Model for Sugarcane Crop
Variables Parameter T Value Significance Level
Constant 8.40 10.95 .000*
lnrsarea 0.004 0.356 0.84
lnrsscost 0.051 0.559 0.57
lnrsfcost 0.15 4.30 0.00*
lnrssicost -0.118 -4. 04 0.00*
lnrsticost 0.004 0.98 0.32
lnrsmcost 0.10 5.25 0.00*
lnrslcost 0.003 0.917 0.36
D1 -0.034 -1.26 0.20
D2 0.47 2.59 0.01*
D1D2 0.04 1.54 0.12
lnedu 0.04 3.56 0.00*
R2 0.20
Adjusted R2 0.17
F- Value 6.6 .000*
* Significant at 5 percent level. ** Significant at 10 percent level
7.3 Results of Rice Regression Models
Rice was another important crop of the Kharief season in the study area especially in the
Khanki canal division and Upper Gogera canal division. Out of 360 farmers, data of 210
farmers were analyzed using separate model for income and productivity effects.
Following parameters were analyzed during the course of analysis. Table 7.7 shows the
important characteristics of the selected variables. The important descriptive statistics of
151
the variables that were included in the analysis along with information regarding the
average cost per acre in real terms. Apart from above variables, log of farmer‘s age and
education (no. of schooling years) were also included in the analysis and have been
discussed earlier parts of this chapter.
Table 7.7: Descriptive Statistics of Important Variables for Rice Crop
Variables Mean Value Std. Deviation
Real average gross value product (Rs.) 4846.61 4274.15
Cultivated area (acres) 7.10 8.25
Seed cost (Rs.) 149.53 62.89
Fertilizer cost (Rs.) 1689.55 483.21
Surface irrigation cost (Rs.) 64.71 15.95
Tube well irrigation cost (Rs.) 3165.06 1302.47
Mechanization cost (Rs.) 1287.63 544.44
Labour cost (Rs.) 1032.77 476.30
Average cost/acre (Rs.) 7856.96 1628.22
Av. Yield (maunds/acre) 32.99 5.63
Prices are used in real term
7.3.1 Estimation of Regression Model for Average Gross Value Product of Rice
In this model, average GVP of the rice crop measured in real prices on per acre basis was
taken as the response/explained variable to measure the effect of reforms on rice crop
income. Results of the first model were obtained after analyzing several combinations of
the variables and results of the most suited model is described in Table 7.8.
Table 7.8 shows that the coefficients of the variables entered in the model have behaved
according to expectations, theory, and previous experiences. Model was tested for the
presence of heteroscedasticity, auto-correlation and multi-collinearity and was found
152
statistically free from these potential problems through the series of statistical tests
carried out to check the presence of above problems in the post-reform analysis. R2 and
adjusted R2 were also satisfactory with significant F-value suggesting that model was
overall fit to the existing data.
Four coefficients of the variables were statistically significant at the 10 percent level.
Coefficient of the rice seed cost (lnrscost) was negative. This was due to the reason that
in rice crop, actually nursery has been sown in the fields and very small quantity of seed
is required to sow the nursery which require low cost. Another reason might be the un-
availability of any new good seed variety of rice crop. Surface irrigation cost was also
having negative sign due to the reasons discussed earlier.
Coefficient of the dummy variables were also showing a priory behaviors as dummy
variable for location of the farm was having negative sign (although statistically
insignificant at 10 percent significance level) thus showing that location of the farms
negatively effected the level of the farm income in the study area. Tail end farmers on the
selected distributaries would have about 5 percent less income as compared to the Head
or Middle reach farmers.
Reforms in irrigation sector have positive effect on the farm income (GVP of the rice
crop) earned through the rice crop as suggested by the results of the model estimated
using the data collected from the farmers in the study area from selected distributaries.
Interaction term was also showing positive sign thus reforms were better suited to the
farmers at the Tail ends of the distributaries as they were the target audiences of the
reform program carried out in the irrigation sector.
153
Table 7.8: Estimated Parameters of the Income Model for Rice Crop
Variables Parameter T Value Significance Level
Constant 9.44 21.179 .000*
lnrarea 0.005 0.404 .687
lnrscost -0.009 -.344 .731
lnrfcost .053 1.475 .141
lnrsicost -.080 -1.344 .180
lnrticost 0.034 1.554 .121
lnrmcost 0.039 1.59 .10**
lnrlcost 0.009 -1.723 .086**
lnedu 0.016 1.681 0.094**
D1 -0.042 -1.021 .308
D2 0.099 3.621 .000*
D1D2 0.022 0.556 .578
R2 0.19
Adjusted R2 0.16
F- Value 5.98 0.000*
* Significant at the 5 percent level. ** Significant at the 10 percent level
7.3.2 Estimation of Regression Model for Average Yield of Rice
Regression model for average yield of rice was estimated using the average yield
(maunds/acre) as the explanatory variable to measure the effect of the different important
cost components and reform process on the farm productivity. Table 7.9 shows the results
of the appropriate model. It is indicated from the Table 7.9 given below that 7
coefficients out of 11 were significant at 10 percent level of significance. Coefficient of
fertilizer cost and dummy for reform were significant at less than 5 percent level of
154
significance. While the coefficients of tube well irrigation costs, mechanization cost,
dummy for reform and interaction term were significant at less than 10 percent level of
significance. The coefficients of seed cost, surface irrigation cost and dummy for location
(D1) were having negative signs. Reasons have been spelled earlier and their negative
signs were also according to priori expectations.
Coefficient of the dummy for reform (D2) when explained using Halvorsen and Palmquist
device depicted that the average yield of the farmers have increased by 6 percent in the
post-reform scenario when compared with base category.
Table 7.9: Estimated Parameters of the Yield Model for Rice Crop
Variables Parameter T Value Significance Level
Constant 3.040 7.179 .000*
lnrarea .008 .713 0.477
lnrscost -.021 -.843 0.400
lnrfcost .089 2.539 0.012*
lnrsicost -.034 -.579 0.563
lnrticost 0.040 -1.883 0.061**
lnrmcost 0.43 1.788 0.075**
lnrlcost 0011 -2.099 0.037*
lnedu 0.016 1.681 0.094**
D1 -.021 -.523 0.601
D2 .056 2.111 0.036*
D1D2 .067 1.699 0.090**
R2 0.18
Adjusted R2 0.14
F- Value 5.531 0.000
* Significant at the 5 percent level. ** Significant at the 10 percent level
155
7.4 Summary
The impact of reform on farm income and productivity was measured by using single
equation models. A Cobb-Douglas production function was used for the estimation of
agricultural productivity and farm income. Econometric analysis was carried out for
wheat, sugarcane and rice crops. The Cobb-Douglas model for wheat was estimated by
using natural log of per acre GVP as dependent variable. The prices were taken in real
terms by using GDP deflator for the year 2001-02.The value of R2
was 0.19 percent
suggesting that independent variables included in the production function explained about
19 percent variation in the dependent variable. Dummy for reforms was also introduced
to determine the overall impact of reforms. While estimating regression model for
average yield of wheat, value of R2 was 0.22 percent showing that 22 percent in wheat
yield was being explained by explanatory variables. Sugarcane and rice crops also
showed more or less similar trend in GVP and yield. However, in case of sugarcane,
dummy variable for location of the farm at the distributary was negative which was
according to the a priori expectations.
156
CHAPTER 8
STOCHASTIC FRONTIER PRODUCTION FUNCTION
AND TECHNICAL INEFFICIENCY EFFECTS MODEL
In order to have maximum likelihood estimates of the parameters of the Cobb-Douglas
production function, a computer program Frontier, version 4.1 was used. Results can be
considerably affected by the choice of functional form in an empirical study. Most
commonly used forms in empirical studies are Cobb Douglas and Translog production
functions. However, Translog or flexible functional form causes serious problem of
multicolinearity whereas Cobb-Douglas function is easy to estimate and interpret and this
model is commonly used to determine technical inefficiency studies. Therefore, for the
present study, Cobb Douglas form has been used to estimate stochastic frontier
production function and inefficiency effect model. Cobb-Douglas form of production
function has been used in various studies like Battese and Broca (1997), Hassan (2004),
Kibbara (2005) and Baksh et al. (2006). Battese and Broca (1997) had preferred Cobb
Douglas type of production function over Translog type production function in
agriculture sector while estimating technical efficiency models.
8.1 Maximum Likelihood Estimates for Parameters of Stochastic
Frontier Production Function
The maximum likelihood estimates of the parameters of the model are given in Table 8.1.
The ratios of the estimated coefficients to their corresponding standard errors (t-ratios)
are used to test the statistical significance of the parameters. It is evident from Table 8.1
that five of the estimates of the coefficients associated with the production inputs for the
data sets of the wheat growers and rice growers were statistically different from zero at
the one percent level of significance. However in case of sugarcane, no coefficient was
statistically different from zero at 10 percent probability level.
157
Table 8.1: Maximum Likelihood Estimates for Parameters of Stochastic Frontier
Production Function and Inefficiency Effects Model
Variable/Crop Wheat Sugarcane Rice
Constant 4.058*
(0.356)
3.92*
(.079)
0.44*
(0.152)
lnArea -0.012 ns
(0.010)
.02ns
(.013)
-0.93*
(.033)
lnSeedcost 0.55*
(0.061)
0.55*
(0.04)
.052ns
(0.042)
lnFertilizercost 0.128*
(.036)
0.17*
(0.038)
0.427**
(0.048)
lnSurfaceirrigationcost 0.118*
(0.026)
-0.05**
(0.032)
0.865*
(0.072)
lnTubewellcost 0.02*
(0.005)
0.003ns
(0.20)
0.146*
(0.033)
lnChemicalcost -0.008ns
(0.005)
0.000ns
(0.020)
.012ns
(0.011)
lnMechanizationcost 0.10*
(0.034)
0.10*
(0.020)
0.27*
(0.037)
lnLaborcost 0.001ns
(0.004)
0.000ns
(0.16)
0.004ns
(0.010)
INEFFICIENCY MODEL
Constant -8.01*
(0.99)
0.22**
(0.07)
1.40*
(0.101)
Age 0.00ns
(0.01)
0.000ns
(0.00)
-0.01ns
(0.02)
Education -0.07ns
(0.038)
-0.000ns
(0.002)
-0.002ns
(0.004)
Experience -0.005ns
(0.014)
-0.000ns
(0.01)
-0.02*
(0.01)
Location dummy -1.72*
(0.26)
0.31*
(0.03)
0.50*
(0.04)
Reform dummy -1.74*
(0.21)
-0.18*
(0.04)
-0.25*
(0.03)
Area owned 0.005ns
(0.004)
0.001ns
(0.001)
0.008*
(0.003)
σ2 2.19
*
(0.207)
0.02*
(0.003)
0.07*
(0.00)
Log likelihood function 109.8 214.75 45.70
Figures in parenthesis are standard errors.
* and ** indicate that coefficients are statistically significant at five and ten percent level of
significance, respectively and ns stands for non-significant
158
The production elasticities have the expected positive signs in all selected crops with few
exceptions. The production elasticity for area under the specific crop (except the
sugarcane crop) was negative. This negative elasticity implied that in case of wheat crop,
one percent increase in area decreases the yield of the rice crop by 0.012 percent, and for
rice crop one percent increase in the area under rice crop could decrease the crop yield by
0.93 percent. However production elasticities of the crop area were statistically
insignificant for sugarcane and wheat crop while for rice crop it was significant and
negative. Negative elasticity was due to existence of diseconomies of scale.
The production elasticities for seed had the expected positive signs for all the selected
crops. Production elasticity of seed was higher in rice and wheat crop (0.55) for
sugarcane, and were statistically significant according to an asymptotic t-test. For rice
crop it was statistically insignificant which was also evident from as rice crop has very
low seed requirement and seed cost is low. An elasticity of 0.55 for wheat and sugarcane
implied that one percent increase in seed cost increased the yield by 0.55 percent. This
was due to the fact that yield directly depends on the number of plants per acre and
population of plants is directly related to the quantity of seed used. Moreover, positive
and significant elasticity for wheat seed and sugarcane seed indicated that there existed
potential to increase the yield by applying the more quantity of seed. This result is in full
agreement with those of Ahmad et al. (1999) and Battese and Broca (1997).
The production elasticities for the fertilizer variable were 0.128, 0.170 and 0.427 for
wheat, sugarcane and rice respectively. The variable had the expected signs and
significantly different from zero at 1 percent significance level except for rice crop where
it was significant below 10 percent significant level. The positive signs of fertilizer
elasticities indicated that one percent increase in the use of this variable would result in
an increased yield of wheat, sugarcane and rice by 0.128, 0.170 and 0.427 percent,
respectively. These elasticities were consistent with those calculated by Battese et al.
(1993), Ahmad et al. (1999) and Hassan (2004).
159
The sign of production elasticity for chemical cost was according to the expectations. For
rice crop it was not significantly different from zero at the 5 percent probability level
according to t-test. The signs of chemical cost for wheat and sugarcane crops were
negative but statistically insignificant from zero at 5 percent probability level. It showed
that chemicals were not being judiciously used in these crops in the study area.
The estimated elasticities for mechanical cost were positive, as expected. The positive
elasticities indicated that one percent increase in this variable would increase the yield of
wheat, sugarcane and rice by 0.105, 0.104, and 0.278 percent, respectively. These
elasticities were consistent with those of Battese et al. (1993) and Hassan (2004). All
these elasticity were significantly different from zero at the 5 percent level using t-test.
The production elasticities for surface irrigation water cost were positive for wheat and
rice crop but negative for the sugarcane crop. These coefficients were significant at 5
percent wheat and sugarcane crop and 10 percent for rice crop.
Tube-well cost variables were positive and statistically significant at 1 percent level for
wheat and rice crops but non significant for sugarcane crop. Coefficients of the tube well
irrigation cost were 0.02, 0.003 and 0.146 for wheat, sugarcane and rice crop
respectively. These showed that one percent increase in tube well irrigation cost could
change the yield of wheat, rice and sugarcane crop by 0.02 percent, 0.003 percent and
0.146 percent respectively. These results are inline with those of Ahmad et al (1999) and
Battese et al. (1993).
8.2 Technical Inefficiency Effects Model
Socioeconomic, demographic, environmental, institutional and non-physical factors are
expected to affect the efficiency (Ali and Chaudhry 1990). The study makes an attempt to
investigate determinants of technical inefficiency. And the coefficients of the explanatory
variables in the technical inefficiency model were of particular interest in terms of
making policy options. Dummy variables were introduced in this inefficiency
160
determination model to capture the impact of reforms and location on the inefficiency of
the farmers in the study area. Dummy for location was introduced with dummy (D1) = 0
for the farmers who were located at the Head or Middle portion of the distributary while
dummy (D1) = 1 for the farmers located at the Tail clusters of the distributary. While
second dummy variable which was introduced to capture the effect of reforms on
incidence of inefficiency in the farmers with dummy (D2) = 0 for pre-reform period and
dummy (D2) = 1 for reform period.
With regard to the sources of efficiency differentials among sample farmers of study area,
the estimates of technical inefficiency effects model provided some important insights.
The parameter estimates in Table 8.2 have the relevant signs, indicating the impact of
explanatory variables on technical (in) efficiency. Explanatory variables with a large
impact should be the main focus of efforts to improve efficiency in crop production, since
these can be influenced relatively easily.
Out of 6 variables explaining technical inefficiency, two variables each in rice and wheat
were significantly different from zero, according to an asymptotic t-test. Two important
variables i.e. location of the farm and implementation of irrigation reforms in the area
were the most important variables effecting technical efficiency in crop production. The
coefficients of location dummy variable did have the positive sign, as expected for
sugarcane and rice crops, significantly different from zero using t-test. This result
indicated that the technical inefficiency at the Tail of the distributaries was more as
compared to the farmers at the Head and Middle reaches of the distributary for sugarcane
and rice. These results are in line with those of Hussain and Biltonen (2001) stating that
majority of the farmers at Tail reaches were poor and scarce in resources. Negative and
significant coefficient has been estimated for wheat crop. The negative coefficient was
due to the fact that wheat crop required less delta of water as compared to rice and
sugarcane. And water availability at Tail reaches was far below than the requirement of
the said crops. So farmers growing wheat at Tail reaches were technically more efficient
than the growers of rice and sugarcane.
161
Another very interesting finding was that reform dummy had the negative sign and it
was statistically different from zero showing that introduction of reforms in the study
area had negative impact on the inefficiency of the farmers. After reform period is more
likely to be efficient as compared to before reform period. It is due to more availability of
water that improved technical efficiency in crops production. Farmers use more of
fertilizer, pesticide and labour because use of these inputs was related to water
availability. Our previous results are confirmed by negative sign of the coefficients of
reform period.
8.3 Technical Efficiencies in Crop Production
Table 8.2 given below shows the range of technical efficiency of the individual farmers
included in the study across the study area. Technical efficiency was calculated for three
major crops i.e. Wheat, Rice and Sugarcane individually. Overall efficiency was
calculated for the entire study period (year 2004 and year 2006). While mean technical
efficiency for the two periods i.e. pre-reform period and post-reform period has also been
estimated separately to capture the effects of irrigation reforms on the efficiency of the
farmers of the study area.
Efficiency estimates for the wheat crop ranged between 0.54 to 0.97 in the study area
during the study period. Overall mean efficiency of the wheat growers during this period
was 0.84 so there existed a potential for the increase in technical efficiency of the
farmers. Technical efficiency estimated of wheat crop for the pre-reform period was 0.82
which rose to 0.86 that showed that farmers were more efficient in the post-reform period
as compared to the pre-reform period. This is in conformity with the results of
inefficiency model estimated for the wheat crop.
162
Table 8.2: Distribution of Technical Efficiency Estimates
Technical Efficiency levels Wheat Sugarcane Rice
Overall mean 0.84 0.97 0.55
Overall minimum 0.54 0.82 0.15
Overall maximum 0.97 0.99 0.90
Before reform period mean 0.82 0.95 0.39
After reform period mean 0.86 0.98 0.65
Efficiency estimated for rice crop showed that minimum estimated efficiency for the
individual farmer was 0.15 while the maximum estimated efficiency was 0.90. Overall
mean efficiency was 0.55. It also depicted that rice farmers had greater variation as for as
the technical efficiency was concerned among them selves as compared to the other two
crops. Before reform technical efficiency was 0.39 which climbed to 0.65 in the post-
reform period. This also confirmed the results of the inefficiency model which was
estimated already.
As for as the sugarcane crop was concerned, predicted technical efficiency for sugarcane
crop ranged from 0.82 to 0.99 during the entire study period with a mean efficiency of
0.971. It also suggested that there existed a potential to increase per acre yield of
sugarcane. The important finding of Table 8.2 is that technical efficiency of sugarcane
growers improved from 0.95 to 0.98 and this finding is inline with the results of previous
findings.
Maximum increase in the technical efficiency between the two periods was observed
among the rice farmers. Efficiency has improved about 66 percent between the two
periods. It also depicted that rice farmers were quite inefficient in the pre-reform period
as compared to the rice and sugarcane growers. Minimum increase in efficiency was
observed in case of sugarcane crop which was estimated about 3.25 percent.
163
The results discussed above reveal that, in general, growers of selected crops were found
improving efficient use of available resources with given level of technology and
achieving the maximum possible output from new and existing technologies.
8.4 Summary
This chapter consisted of estimation of Maximum Likelihood for parameters of stochastic
frontier production function, technical inefficiency effects model and technical
efficiencies in crop production. The results of the Maximum Likelihood estimates of the
parameters showed that 5 of the estimates of the coefficients associated with the
production inputs for the data sets of the wheat growers and rice growers were
statistically different from zero at one percent level of significance. However in case of
sugarcane, no coefficient was statistically different from zero at 10 percent probability
level. These results suggested that the models for wheat and rice were fairly fit for the
data sets. The production elasticities of the crop area were statistically insignificant for
sugarcane and wheat crop while for rice crop it was significant and negative. Tube-well
cost variables were positive and statistically significant at 1 percent level for wheat and
rice crops but non significant for sugarcane crop. Dummy variables were introduced in
inefficiency effects model to capture the impact of reforms and location on the
inefficiency of the farmers in the study area. The results indicated that the technical
inefficiency at the Tail of the distributaries were technically more inefficient as compared
to the farmers at the Head and Middle reaches of the distributary for sugarcane and rice
crops.
Technical efficiency was also calculated for wheat, sugarcane and rice in the study area.
Efficiency estimates for wheat crop ranged from 0.54 to 0.97 in the study area. As far as
sugarcane was concerned, predicted technical efficiency ranged from 0.82 to 0.99 during
the entire study period, suggesting that there existed a potential to increase per acre yield
in sugarcane.
164
CHAPTER 9
SUMMARY AND CONCLUSIONS
This chapter is divided into six sections. The first section elaborates summary of the
study. The second section describes conclusions of the study. Limitations of the study are
discussed in the third section. Whereas problems faced during study, policy
recommendations and future research areas are given in the section 9.4, 9.5 and 9.6
respectively of this chapter.
9.1 Summary
Agriculture is the backbone of the economy of Pakistan. It accounts for 20.9 percent of
GDP and employs 43.4 percent of total labor force. It supplies most of the country‘s
needed food grains and also is a source of raw materials for major domestic industries. As
such, agriculture is at the center of the economic policies and has been designated by the
Government as the engine of national economic growth and poverty reduction. The
importance of water for Pakistan can not be under-estimated, particularly for irrigated
agriculture in the country. The major crops are wheat, cotton, rice and sugarcane. Cotton
alone accounts for 8.6 percent of the value added in agriculture and about 1.8 percent in
GDP. The health of the agricultural sector also has important implications for poverty
relief and private sector development. Our irrigated agriculture is fed by the waters of the
Indus River and its tributaries.
The Indus Basin Irrigation system of Pakistan is the largest contiguous irrigation system
in the world, serving in excess of 14 million hectares. It consists of three major storage
reservoirs, namely, Tarbela and Chashma on River Indus, and Mangla on River Jhelum,
with a present live-storage of about 12.5 MAF, 19 barrages; 12 inter-river link canals and
43 independent irrigation canal commands.
The Indus Basin Irrigation System of Pakistan was facing multiple problems like
deterioration of infrastructure, high conveyance losses and inequitable water distribution
165
both under normal supply and shortage conditions. There had been chronic inequity with
the upstream water users receiving more water than their due share, while those in the
Tail reaches of the canal command received less. The system was steadily deteriorating
and performing far below user‘s expectations; and there was a great mistrust between the
irrigation department and the users. Some of the causes of ever declining system
performance were inequity in water distribution, poor operation and maintenance (O&M)
of the system, poor cost recovery (Aabiana), political interference and weak capacities of
government institutions.
Keeping in view the above discussed problems, the World Bank proposed
commercialization and privatizations of the irrigation system as the only choice for
rehabilitation. However after a series of negotiations, the government of Pakistan agreed
upon institutional reforms in water sector of the Punjab. Consequently, in 1997,
Pakistan‘s provincial Assemblies passed bills to implement institutional reforms in the
country‘s irrigation sector. In the province of Punjab, the institutional reforms were
introduced through the PIDA Act of 1997.
The principal objective of reforms in irrigation sector of the Punjab Province was to
reverse the deteriorating performance of its irrigation system and the consequent
stagnating productivity of irrigated agriculture in the Indus Basin. Under the reform
model, new institutions were designed and implemented on pilot testing basis in LCC
(East) irrigation system in Punjab province. This had been accomplished primarily
through two institutional policy initiatives. One was the restructuring and decentralization
of Punjab irrigation department (PID) into autonomous Punjab Irrigation and Drainage
Authorities (PIDA) and Area Water Boards (AWB). The second was to include local
irrigation communities in the management of the irrigation system by developing
operational and maintenance as well as fiscal responsibilities for secondary canal
channels through independent, self-sustaining Farmer Organizations (FO). The irrigation
management was handed over to the farmers in year the 2005 and so far the system is
working with the participation of the stakeholders.
166
The present study was designed to assess the effectiveness of ongoing irrigation reforms
in terms of improving water delivery, operation and maintenance (O&M) of irrigation
system, equity in water distribution and overall management of irrigation system. The
impact of these reforms on overall agricultural productivity and farm income are also
examined.
The area of Lower Chenab Canal East LCC (East) was selected as study area because the
process of irrigation reforms was initiated as a pilot project and completed here in this
area in December 2005. A well represented sample of 30 distributaries and 360 farm
households were selected for collecting primary information from the field and secondary
information from the FOs and Irrigation department offices. A multistage sampling
procedure was adopted in selecting the sample. In order to have comprehensive
evaluation of performance, achievements and early impacts of reforms in LCC System,
the study employed two approaches in evaluating performance and impact of newly
created institutions. At first level, assessment of reforms in LCC East (Reform Area) was
made on the basis of information from the secondary data. At this level ―before and after‖
IMT situation was compared. At second level performance of distributaries after the
reforms was evaluated on the basis of primary data collected through well structured
questionnaire at farm household level. Quantitative analysis was carried out by making
comparison of set of indicators for secondary information at distributary level. Means,
averages, percentages and frequencies were estimated to determine the impact of
irrigation reforms. An econometric model was used to capture the impact of irrigation
reforms on farm income and productivity. Economic Inefficiency model, to determine the
negative impact of irrigation reforms on inefficiency of the respondents, was also
estimated.
The study showed that the average family size in the study area was 6 members and
average landholding was1- 5 acres. The analysis showed that there were 43 percent of the
farmers who reported that water theft had almost been controlled. At the Middle reach of
the distributary 48 percent of the farmers gave their opinion that water theft decreased.
Most of the farmers at the Tail end were of the opinion that either water theft increased
(10 percent) or was same (10 percent). According to them there was no significant
167
change in the extent of water theft. Only the form of water theft changed. Earlier there
was open theft but now there was disguised theft. Only 7 percent of the farmers reported
that water theft increased but the percentage of those farmers who reported that water
theft decreased after irrigation management transfer were 31 percent.
The results of the study based on comparison of indicators from primary data showed that
there was an increase in the crop yields which could be attributed to the outcome of
reform process in Punjab. On an overall basis, all the major crops (wheat, sugarcane and
rice) showed an increasing trend in yields. Wheat yield increased by 10 percent,
sugarcane by 5 percent and rice by 13 percent. The substantial increase in rice yield after
IMT indicated that surface water availability to the farmers increased due to better
management of the system by the FOs.
Average gross value product increased by 4 percent for wheat crop and 15 percent for
sugarcane in the year 2006 (Post-reform period). Most significant increase was in rice
crop that showed an increase of 43 percent across the system. Increase in GVP of wheat
crop was. 4 percent while the increase in average yields of wheat crop across the LCC
(East) system was 10 percent. The results of the study also showed that there was decline
in real price of wheat received by the majority of the farmers in year 2006 as compared to
year 2004. Rice crop showed the maximum increase in real GVP i.e. 43 percent, while
the increase in average rice yield was about 13 percent depicting that there was an
increase in real price of rice in the year 2006 as compared to year 2004.
It was found that cost of tube well irrigation and fertilizer were the main components of
the total cost. Cost of canal irrigation water was substantially less than the cost of ground
water in both pre and post-reform period. Thus with increase in surface water supply after
reforms, cost of production of the farmers reduced as they were using less ground water.
The results of the study showed that on an overall basis, cost of production of wheat,
sugarcane and rice reduced by 12 percent, 3 percent and 11 percent respectively in the
post-reform period.
168
Usually farmers take their decisions on the basis of gross margin and not on net income
basis. On an overall basis, gross margin increased by 6 percent in case of wheat crop after
IMT. Similar increasing trend was observed in sugarcane and rice across the whole length
of the LCC (East) irrigation system.
As far as cropping intensity in the study area was concerned, an increasing trend was
found in all the four canal divisions after IMT. On an overall basis, two years after
transfer of management, cropping intensity increased from 163 percent to 182 percent,
leading 13 percent increase after IMT.
In order to determine the early impacts of irrigation reforms on farm productivity and
farm income, Cobb-Douglas production function was estimated. The analysis was carried
out for three major crops in the area i.e. wheat, sugarcane and rice. Two aspects were
econometrically analyzed. In first analysis, Average gross value product (GVP/acre) of
crops (in real prices) was taken as dependent variable to capture the effect of reform
process, location of the farm along the distributary and important components of variable
cost of production. In the second analysis, average yield per acre (maunds per acre) was
taken as dependent variable to determine the impact of reform process.
For Average gross value product (AGVP) of wheat, variables estimated were area, seed
cost, fertilizer cost, surface irrigation cost, tube well irrigation cost mechanical cost and
years of education. The Cobb-Douglas model for wheat was estimated by using natural
log of per acre GVP as dependent variable. The F-Value for income model for wheat was
3.7 and it was significant at 99 percent confidence level. Estimated parameters for wheat
yield showed that F-Value for the overall model was 7.08 percent and it was significant at
99 percent confidence level. The results of the study showed the similar trend for
sugarcane and rice crop in the area.
For estimating Average gross value product (AGVP) of sugarcane, the regression model
was estimated to find out the impact of major variable inputs on the crop income. Impact
of reforms, location of the farm with respect to for wheat, rice and sugarcane was also
169
analyzed by introducing dummy variables and the interaction terms in the regression
model.
In order to capture the impact of the irrigation reform on the productivity of rice crop,
OLS model was estimated. Model was tested for the presence of heteroscedasticity, auto-
correlation and multicolinearity and was found statistically free from the potential
problems through the series of statistical tests. R2 and adjusted R
2 were also satisfactory
with significant F-value. Four coefficients of the variables were statistically significant at
90 percent or above confidence level. Coefficient of the dummy variables were also
showing a priory behaviors as dummy variable for location of the farm with respect to
distributary was having negative sign (although statistically insignificant) thus showing
that location of the farms did effect the level of the farm income in the study area. Tail
end farmers on the selected distributaries would have about 5 percent less income as
compared to the base category.
For the estimation of stochastic frontier production function and inefficiency effect model
Cobb-Douglas form of production function and translog production function were used.
The production elasticities were having the expected positive signs in all selected crops
with few exceptions. The production elasticity for area was negative implying that one
percent increase in area decreased the yield of the rice crop by 0.935 percent, however,
the sign of this variable was negative as for as wheat and rice crops were concerned but
the sign was positive for sugarcane crop. However production elasticities were
statistically non-significant except the rice crop where it was significant and negative.
Negative elasticity was due to existence of diseconomies of scale.
Aabiana or water charges was an important component of irrigation system. Institutional
reforms introduced in the LCC (East) have addressed this component. It was evident from
the results of the study that Aabiana recovery percentage increased from 43 percent
before IMT to 62 percent after IMT thus showed an increase of about 19 percent in the
post-reform period.
170
Operation and Maintenance of the system was an important aspect for the sustainability
of irrigation system, which was ignored in the past and genuine needs of the O&M had
not been fulfilled. O&M expenditures incurred by the irrigation department were not
uniformly spent in all the canal divisions. In Khanki division per hectare expenditures
were Rs. 147 and in Burala division per hectare expenditures were only Rs. 41 which
showed the irrational decision of irrigation department. However after the management
transfer O&M expenditures were need based.
Salaries expenditures of the operational staff per unit of irrigated area was a major
indicator to assess the performance of the irrigation systems. The results of the study
clearly showed that there was significant reduction in salary expenditures after IMT.
Thus a downward trend in salary expenditures was observed after reforms.
Delivery performance ratio (DPR) is the ratio of actual to target (sanctioned) volume of
water delivered. There was great variation in DPR at Head and Tail of the selected
distributaries before the reforms. However, after reforms this variation in DPR was
significantly reduced. The average DPR at the Tails of the selected distributaries of LCC
(East) irrigation system was only 0.19 which rose to 0.36 showed that Tails were
comparatively in a better position after reforms.
9.2 Conclusions
Reforms had positive impact on average yield of wheat, sugarcane and rice in the study
area. It was found that average yield of wheat on an overall basis increased by 10 percent,
sugarcane by 5 percent and rice by 13 percent. Average gross margins of these crops have
significantly increased after reforms. It was concluded that average increase in gross
margins of sugarcane and rice increased by 38 percent and 43 percent, respectively. It
was also concluded that cost of production of the farmers and particularly located at the
Tail reaches reduced as there was more supply of surface irrigation water that replaced
the tube well irrigation which was relatively more costly.
There was a relationship between the location of farms along the distributary and their
respective yield and income. It was found that, on an average the lowest yield was
171
achieved by the farmers at the Tail end reaches which was lower than the yields obtained
by the farmers at the Head and Middle reach areas. This low yield at the Tail reaches was
attributed to low surface water availability.
Aabiana recovery percentage on an overall basis, increased from 43 percent to 62 percent
after irrigation management transfer. However, it was observed that the Aabiana recovery
percentage considerably decreased in the second year and onward after IMT on account
of loose administration and loopholes in the new system, like no action against the
defaulter of Aabiana.
Per hectare contingency expenditures considerably reduced in all the canal divisions after
irrigation management transfer. It was due to the reason that these expenditures were
directly under the control of FO president who was accountable to other members.
Similarly, per hectare salary expenditures also reduced after reforms on account of the
reason that they did not inducted excessive staff in FOs.
The reform initiatives have enabled the farmers to participate in the irrigation
management in a constructive mode right from the initial stage of planning and
development, operation and maintenance, dispute resolution, equitable distribution of
water and Aabiana collection. The results of the study showed that IMT has created
confidence among the farmers through empowerment.
The process of IMT in Punjab is at its early stages. The new system is facing certain
administrative problems. It was also found that the officials of old irrigation department
are not happy with the creation of the new authority. They are endeavoring for the failure
of the new system. They are creating hindrance in smooth running of the new system.
These problems can be solved only through strong legal infrastructure and administrative
support. However, it is concluded with confidence that farmers‘ participation in irrigation
management in Punjab could bring many benefits to the farmers but it would require
sufficient time period.
172
9.3 Limitations of the Study
The findings of the study must be viewed in the light of some potential limitations listed
below.
1) The present study used ―Before and After ―approach for the analysis of data. An
important missing link was ―With and Without‖ approach. which could not be
incorporated in the present study due to financial constraint and time.
2) This study tends to evaluate short run impact analysis of reform process because
period under study was only two years. There is need to correlate this study with
the political reform process and socio-economic aspects of the farming system.
3) The present study could not be based on previous empirical studies as author
could not find much relevant econometric analysis on the issue of impact
assessment of institutional reforms in irrigation sector.
9.4 Problems Faced During Study
The researcher faced certain unavoidable difficulties and problems during the process of
primary and secondary data collection. Some of the problems faced during the survey
included the followings:
1) There were delays in contacting some respondents in the fields as they were
involved in wheat harvesting activities. They were unable to spare sufficient
time for interview.
2) Few respondents were found very accommodating and facilitated data
collection process but the others were reluctant to give required information.
At time it became unmanageable to make the respondent understand, because
they readily began to count their own problems.
3) Data collection became fairly difficult when respondent were illiterate.
Despite the fact, researcher tried his level best to let the respondent know
173
about the purpose of the study, some farmers were still suspicious about the
nature of the study and questioning. They thought that the information
collected may be used for taxation, especially when questions such as size of
land holding, area under crops and land rent were put to them.
4) Some respondents hesitated to give accurate information due to the fear that
the government might increase water charges or impose some new taxes on
them. However, the suspicious farmers were taken into confidence that the
information taken from them would be used only for research purpose.
5) The secondary data collection from the Punjab irrigation department was also
a tedious job. The clerical staff of the department was not willing to give any
kind of past information from the heap of files.
Despite all these problems and difficulties, the researcher tried his level best to manage
his efforts, time and money in judicious way.
9.5 Policy Recommendations
The following practical recommendations are made for technical, managerial and
institutional improvements of the new system.
1) Results of the study revealed that reforms have positive impacts on the yield and
productivity of the farmers for all the major crops. It is also evident that reforms
also have positive impacts on the farmers locating at the Tail clusters of the
distributaries. But location of the farm still has positive impact on the inefficiency
of the farmers as evident by inefficiency effect model. So there is need on
persistent basis that poor farmers as a whole and especially those locating at the
Tail clusters should be supported by providing subsidy on tube well installation
for improving their crop productivity and crop income. It is also mentioned that
small, poor farmers often situated at distributary Tails so by reducing
inefficiencies associated with the location of the farmers, productivity and income
status of the farmers can be up lifted.
174
2) Analysis of the quantitative data revealed that cost of surface irrigation was
negatively associated with crop income and crop yield. On the other hand it has
also been observed that irrigation charges make only 0.5 to 1.5 percent of the total
cost of production. It can be concluded that presently irrigation services are not up
to the standard and low irrigation charges do not instigate the formers for efficient
use of surface irrigation water. It is need of the time that irrigation services should
be improved and after improving their quality, rates of surface irrigation charges
(Aabiana) can be revised. In present scenario, any increase in irrigation charges
could have negative impact on crop income and productivity.
3) It was revealed during the analysis that real price of wheat has decreased and this
decrease in the real price of wheat has offset the effects of reform process to a
larger extent. So to capture the true impact of all the efforts made for
implementation of reform process, farmers should have real incentives for their
produces. It is therefore recommended that prices should be set taken into account
the inflationary process and rise in the cost of inputs.
4) The representatives /Office bearers of FO should be paid a reasonable alary on
regular basis like the other government servants. They loose their interest if they
are not paid, no body works without any interest for a longer time. The efficiency
and working of FO can be enhanced if they are paid properly and regularly.
5) Analysis of data showed that farmers have mostly paid the Aabiana but it was not
deposited to FOs in time. In most of the cases KP Chairman collected the
Aabiana but did not deposited in the government treasury and keep hold with
them. PIDA should devise such a system that no KP Chairman could keep the
collected Aabiana with him for more than a limited time period.
6) It was observed that with the passage of time the Aabiana recovery has declined.
In the first two crops after reforms, it touched to its climax, and afterward it
started decline. It was just because people do not accept moral pressure only. FO
Presidents are lacking strong legal support from the government and they have
been so far unable to lodge FIR against the Aabiana defaulters and those who
illegally pump water from distributaries. It is suggested that there should be
Recovery Magistrates who should stand responsible to all such kind of problems.
175
7) At present, out of total collected Aabiana, 60 percent share goes to PIDA and 40
percent to FO. Out of this 40 percent share received by FO, 70 percent is utilized
for Office management and 30 percent for O&M of the distributary. The share of
FO is not justifiable. It should be increased to 50 percent and out of this share; at
least 50 percent should be utilized for Operation and Maintenance of the
distributary.
8) An incentive of 6 percent is paid to KP chairman for 100 percent collection of
Aabiana. This is a meager amount as compared to the efforts involved in the
collection process. It is suggested that at least 10 percent share of Aabiana
collected and not conditioned with 100 percent collection should be given to KP
Chairman. It will increase the efficiency of the Aabiana Collectors.
9) There is need to inform the farmers about their entitlement of water, and its
expected quantities and timings. This information can enable them to plan the
cropping patterns, farming operations and use of other non-water inputs for
optimization. Farmer‘s participation can provide useful guidelines and feed back
in evolving effective policy frame work, which in turn can be helpful in
improving irrigation efficiencies.
9.6 Future Areas of Research
A lot of work is required to evaluate the performance of Punjab Irrigation and Drainage
Authority (PIDA), Area Water Board (AWB) and Farmer Organizations (FOs).
Performance measurement should concentrate on economic, institutional and technical
efficiency issues. There is need to evaluate empirically the extent of maintenance and
improvement activities, the distribution and productivity of water and changes in the size
and distribution of farm incomes before and after irrigation management transfer.
Substantive investigation is required to establish linkages of IMT with other sectors of
the economy. International, national and local research is needed on case studies to
illuminate detailed impact of irrigation management transfer on equity, water charges and
cost recovery, irrigation services and above all, agricultural productivity and farm
income. The present study was first empirical study of its kind on the issue of impact
176
assessment of irrigation reforms in Pakistan. In spite of the fact that empirical evidence
and literature review was not available, efforts have been made to capture all possible
dimensions in this study. It is suggested that further empirical research should be
conducted taking into account ―with and without‖ approach.
177
LITERATURE CITED
Abdelhadi, A.W., H.S. Adam, M.A. Hassan and T. Hata. 2004. ―Participatory
Management: Would it be turning point in the history of the Gezira
Scheme?‖ Journal of the International Commission on Irrigation
Drainage 53 (4): 429-436.
Abedullah and S. Pandey. 2004. ―Risk and fertilizer use in the rainfed rice ecosystem of
Tarlac, Philippines. Journal of Agricultural and Applied Economics, 36
(1): 241-250.
Abedullah and M. Ali. 2006. ―Quantifying the extent and nature of risk in alternative
cropping patterns in Claveria, Philippines. The Pakistan Development
Review, 45 (2): 261-280.
Ahmad, M. and B.E. Bravo-Ureta. 1996. ―Technical efficiency measures for dairy farms
using panel data- a comparison of alternative model specifications‖.
Journal of Productivity Analysis, 7:399-415.
Ahmad, M., M. Rafiq and A. Ali. 1999. ―An analysis of technical efficiency of rice
farmers in Pakistani Punjab‖. The Bangladesh Journal of Agricultural
Economics, 22:79-86.
Aigner, D.J. and S.F. Chu 1968. ―On estimating the industry production
function‖American Economic Review, 58: 826-839.
Aigner, D.J., C.A.K. Lovell and P. Schmidt. 1977. ―Formulation and Estimation of
Stochastic Frontier Function Models‖. Journal of Econometrics, 6: 21-37.
Ali, M. and M.A. Chaudhry. 1990. ―Inter-regional farm efficiency in Pakistan‘s Punjab:
A frontier production function study‖. Journal of Agricultural Economics,
41: 62-74.
178
Ali, M. and J.C. Flinn. 1989. ―Profit efficiency among Basmati rice producers in Pakistan
Punjab‖. American Journal of Agricultural Economics, 71: 303-310.
Ali, M. and D. Byerlee. 1991. ―Economic efficiency of small farmers in a changing
world: a survey of recent evidence‖. IRRI Social Science Division Paper
No. 91-04,Social Science Division, International Rice Research Institute,
Manila, Philippines.
Amoah, L.K.S and J.W. Gowing. 2001. ―The experience of irrigation management
transfer in Ghana: a case study of Dawhenya irrigation scheme‖.
Irrigation and Drainage Systems 15: 21:38. Netherlands.
Awosola, O.O., E.O. Idowu and V. Schalkwyk. 2005. ―Irrigation projects reform case
studies: An assessment of projects' performance‖. A paper presented at the
8th International River symposium, Australia.
Bandaragoda, D. J. 2006. ―Status of institutional reforms for integrated water resources
management in Asia: Indications from policy review in five countries‖.
Working Paper 108, International Water Management Institute (IWMI),
Colombo, Sri Lanka.
Bandyopadhyay, S.P. Shyamsundar and M. Xie. 2007. ―Yield impact of irrigation
management transfer: Story from the Philippines‖ Policy research working
paper 4298.The World Bank.
Barker, R. 2002. ―Recent developments in irrigation management in Asia and the
Pacific‖. Report of the Asian Productivity Organization (APO) Seminar on
Organizational Change for Participatory Irrigation Management,
Philippines.
Battese, G.E. and G.S. Corra. 1977. ―Estimation of Production Frontier Model: With
application to pastoral zone of eastern Australia‖. Australian Journal of
Agricultural Economics, 21: 169-179.
179
Beattie, B.R. and C.R. Taylor. 1985. ―The economics of production‖. John Wiley and
Sons Inc., New York.
Battese, G.E. and T.J. Coelli. 1988. ―Prediction of firm-level technical efficiencies with a
Generalized Frontier Production Function and panel data‖. Journal of
Econometrics, 38: 387-399.
Battese, G.E. 1992. ―Frontier production functions and technical efficienct: A survey of
empirical applications in Agricultural Economics‖. Agricultural
Economics, 7:185-208.
Battese, G.E., S.J. Malik and S. Broca. 1993. ―Production function for wheat farmers in
selected districts of Pakistan: An application of a stochastic frontier
production function with time-varying inefficiency effects‖. The Pakistan
Development Review, 32:233-68.
Battese, G.E. and T.J. Coelli. 1993. ―A stochastic frontier production function
incorporating a model for technical inefficiency effects‖. Working paper
on Econometrics and applied Statistics, No. 69, department of
Econometrics, University of New England, Armidale, Australia.
Battese, G.E. and T.J. Coelli. 1995. ―A model for technical inefficiency effects in a
Stochastic frontier production function for panel data‖. Empirical
Economics, 20: 325-332.
Battese, G.E. and S.S. Broca. 1997. ―Functional forms of stochastic frontier production
functions and models for technical efficiency effects‖. A comparative
study for wheat farmers in Pakistan. Journal of Productivity Analysis, 8:
395-414.
Biswas, A.K. 1996. ―Water for the Developing World in the 21st century: Issues and
implications‖. Journal of the International Commission on Irrigation
Drainage (ICID), 45 (2)
180
Bravo-Ureta, B.E. and A.E. Pinheiro. 1993. ―Efficiency analysis of developing country
agriculture: A review of the frontier function literature‖. Agriculture and
resource economics review, 22: 88-101.
Baksh, K., B. Ahmad and S. Hassan. 2006. ―Food security through increasing technical
efficiency‖. Asian journal of plant sciences, 5 (6): 970-976.
Byrnes, P., R. Färe, S. Grosskopf and S. Kraft. 1987. ―Technical efficiency and size: The
case of Illinois grain farms‖. European Review of Agricultural Economics,
14:367-81
Coelli, T.J. 1995. ―Recent developments in Frontier estimation and efficiency
measurement‖. Australian Journal of Agricultural Economics, 39: 219-
245.
Coelli, T.J. 1996. ―A guide to Frontier, version 4.1: A computer programme for
Stochastic Frontier production and cost function estimation‖. CEPA
working paper No 7/96, Department of Economics, University of New
England, Armidale, Australia.
Coelli, T.J., D.S.P. Rao and G.E. Battcse. 1998. ―An introduction to efficiency and
productivity analysis‖. Kluwer-Nijhoff, Boston.
Coelli, T.J. and S. Perclman. 1999. ―Efficiency measurement, multiple output
technologies and distance functions: with application to European
railways‖. Working paper in CREPP 96/05, University de Liege, Belgium.
Coward, E., and W. Jr. 1980. ―Irrigation development: Institutional and organizational
issues‖. Irrigation and Agricultural Development in Asia. Cornell
University press, USA.
181
Danial, W., C. Donald and D. E. Parker. 1980. ―Water reform and economic
development. Institutional aspects of water management in the developing
countries‖. Economic Development and Cultural Change 28 (2): 365-387,
The University of Chicago Press, USA.
Dick, R. M. 1997 ―Farmer participation in irrigation: 20 years of experience and lessons
for the future‖. International Food Policy Research Institute, Washington
D.C, USA.
Dinar, A. 2001. ―Reforming ourselves rather than our water resources: Politics of water
scarcity at local, national and international levels. Paper presented in the
international conference on fresh water, Bonn, Germany.
Dinar, A., T. Balakrishnan and J. Wambia. 2004. ―Politics of institutional reforms in the
water and drainage sector of Pakistan‖. Environment and Development
Economics, 9 : 409-445. Cambridge University, United Kingdom.
Doll, J.P. and F. Orazem. 1985. ―Production Economics‖. Theory with application, 2nd
edition, Wiley publishers, New York, USA.
Doukkali, M. R. 2005. ―Water institutional reforms in Morocco‖. Water Policy 7 (1): 71-
88.
Edward J., V.D. Velde and J. Tirmizi. 1999. ―Irrigation policy reforms in Pakistan:
Who‘s getting the process right?‖ International researchers‘ conference on
participatory irrigation management. The long road to commitment: A
socio-political perspective on the process of irrigation reform. Hyderabad,
India.
Facon, T. 2000. “Improving the irrigation service to farmers: Key issue in participatory
irrigation management‖. Food and Agriculture Organization (FAO),
Rome, Italy.
182
Farrell, M.J. 1957. ―The measurement of productivity efficiency‖. Journal of the Royal
Statistical Society Series A, 120: 253-290.
Forsund, F.R., C.A.K. Lovell and P. Schmidt. 1980. A survey of Frontier production
functions and their relationship to efficiency measurement‖. Journal of
Econometrics, 13, 5-25.
Fried, H.O., C.A.K. Lovell and S.S. Schmidt. 1993. ―The measurement of productive
efficiency: techniques and applications‖. Oxford University press, New
York, 68-119.
Ghatak, S. and K. Ingersent. 1984. ―Agricultural and Economic Development‖. Brighton
Wheatsheat books.
GOP. 1998. ―District census report of Faisalabad‖. Population Census Organization
Statistics Division, Islamabad, Pakistan.
GOP. 1998. ―District census report of Gujranwala‖. Population Census Organization
Statistics Division, Islamabad, Pakistan.
GOP. 1998. ―District census report of Hafizabad‖. Population Census Organization
Statistics Division, Islamabad, Pakistan.
GOP. 1998. ―District census report of Jhang‖. Population Census Organization Statistics
Division, Islamabad, Pakistan.
GOP. 1999. ―Irrigation reforms and water management project for the Hakra Branch
Canal‖. Report published by the government of the Punjab, Pakistan.
GOP. 2005. ―Agricultural Statistics of Pakistan 2005-06‖. Ministry of Food, Agriculture
and Livestock, Islamabad, Pakistan.
GOP, 2006. ―A brief of the Faisalabad irrigation zone‖. Irrigation department,
government of the Punjag, Lahore, Pakistan.
183
GOP. 2007. ―Economic Survey of Pakistan 2006-2007‖. Economic Advisor‘s Wing,
Finance Division, Islamabad, Pakistan.
Greene, W.H. 1992. ―LIMDEP Version 6.0‖: User‘s manual and reference guide,
Econometric Software Inc., New York.
Greene, W.H. 1995. 2005. ―Econometric Analysis‖. Fifth edition, published by Pearson
Education, Inc. New York, USA.
Gujrati, D.N. 2003. ―Basic Econometrics‖. Fourth edition, published by McGraw-Hill
companies, Inc. 1221 Avenue of the America, NY, 10020, USA.
Haq, A.U., and B.A. Shahid.1997. ―Public, Private or Participatory? Reforming Irrigation
Management in Pakistan‖ Journal of the International Commission on
Irrigation Drainage (ICID) 46 (1): 37-48.
Haq, A. U. 1998. ―Managing irrigation for environmentally sustainable agriculture in
Pakistan‖. A Case study of the Punjab Irrigation Department. Report No.
C-12, Pakistan National Programme, International Irrigation Management
Institute, Lahore, Pakistan
Hassan, S. 2004. ―An analysis of technical efficiency of wheat farmers in the mixed
farming system of the Punjab, Pakistan‖. A Ph.D. Dissertation,
Department of Farm Management, University of Agriculture, Faisalabad,
Pakistan.
Hearne, R. R. and G. Donoso. 2005. ―Water institutional reforms in Chile‖ Water Policy
7 (1): 53-69.
Heyns, P. 2005. ―Water institutional reforms in Chile‖ Water Policy 7 (1): 89-106.
Hopper, W.D. 1965. ―Allocative efficiency in a traditional Indian agriculture‖. Journal of
Farm Economics, 47: 611-624.
184
Hung, C. and J.T. Liu. 1994. ―Estimation of non-neutral stochastic frontier production
function‖. Journal of Productivity Analysis, 4: 171-180.
Hussain, I; and E. Biltonen. 2001. ―Pro-poor intervention strategies in irrigated
agriculture in Asia: Issues and Options‖. International Water Management
Institute, Colombo, Sri Lanka.
Hussain, I., W.A. Jehangir, M. Ashfaq and M. Mudasser. 2003. ―Pro-poor intervention
strategies in irrigated agriculture, Punjab, Pakistan. International Water
Management Institute, Lahore, Pakistan.
Hussain, I. 2003. ―Pro-poor intervention strategies in irrigated agriculture in Asia.
Poverty in irrigated agriculture: Issues and options, Pakistan‖.
International Water Management Institute, Colombo, Sri Lanka.
Hussain, I; Z.A. Gill; Z. Naseer and S. Sarwar. 2006. ―Institutional and technological
interventions for better irrigation management in the new millennium‖.
Proceedings of INPIM‘s 9th
international seminar, Pakistan.
Jaforrullah, M. and J. Whiteman. 1999. Scale efficiency in the New Zealand dairy
industry: a nonparametric approach. The Australian Journal of
Agricultural and Resource Economics, 43: 523-541.
Jehangir, W. A., M. Mudasser and Z. Hussain 1999. ―Participatory Irrigation
Management and its Financial Viability: A Case Study‖. The Pakistan
Development Review 38 (4): 629-640.
Johnson III, S. H. 1995. ―Selected experiences with Irrigation management transfer:
Economic implications‖. Water Resources Development 11: (1).
International Irrigation Management Institute, Colombo, Sri Lanka.
Johnson III, S.H., M. Sevendsen and F. Gonzalez. 2004. ―Institutional reforms, options in
the irrigation sector‖. Agriculture and Rural Development discussion
paper No. 5. World Bank, Washington, D.C.
185
Johnson III, S.H.; S. Mark, and J. Trava. 1997. ―Participatory Irrigation Management:
Benefits and second generation problems‖. Economic Development
Institute of the World Bank, Washington, D.C.
Jondrow, J., CA.K. Lovell, I.S. Matcrov and P. Schmidt. 1982. ―An estimation of
technical inefficiency in Stochastic Frontier production function model‖.
Journal of Econometrics, 19: 233-238
Kahlown, M.A.and A. Majeed. 2002. ―Water-resources situation in Pakistan: Challenges
and future strategies‖. Pakistan Council of Research in Water Research
(PCRWR), Islamabad.
Kalirajan, K.P. 1981. ―An econometric analysis of yield variability in paddy production‖.
Canadian Journal of agricultural Economics, 29: 283-294.
Kalirajan, K.P. and J.C. Flinn. 1983. ―The measurement of farm-specific technical
efficiency‖. Pakistan Journal of Applied Economics, 2: 167-180.
Kloezen W.H., C. G. Restrepo and S. Johnson. 1997. ―Impact assessment of Irrigation
Management Transfer in the Alto Rio Lerma irrigation district, Mexico‖.
Kibbara, B.W. 2005. ―Technical Efficiency in Kenyans‘ maize production: An
application of the stochastic frontier approach‖. M.Sc. dissertation of
Agriculture and Resource Economics, Colorado, USA.
Kopp, R.J. and V.K. Smith. 1980. ―Frontier production function estimates for steam
electric generation: a comparative analysis‖. Southern Economic Journal,
47:1049-1059.
Koppen,B.V., R. Parthasarathy and C. Saffiliou. 2002. ―Poverty dimensions of irrigation
management transfer in large scale canal irrigation in Andra Pradesh and
Gujrat, India‖. Report 61 International Water Management Institute
(IWMI), Colombo, Sri Lanka.
186
Koopmans, T.C. 1951. An analysis of production as an efficient combination of
activities‖. Activity analysis of production and allocation. Koopmans, T.C.
(ed.). Cowles Commission for research in economics, No. 13. Wiley, New
York, USA.
Kumbhakar, S.C. 1991. ―Efficiency measurement with multiple outputs and multiple
inputs‖. Journal of Productivity Analysis, 7: 225-255.
Lasharai, B.K. 2006. ―Improving equity of water distribution: The challenge for Farmer
Organizations in Sindh under the National Drainage Program‖. Mehran
University Research Journal of Engineering and Technology 25: (2)
Lashari, B., Y. Memon and N.A. Memon. 2003 ―Iinstitutional reforms in irrigation sector
of Pakistan: An approach towards integrated water resource management‖.
Paper presented at the 9th
International Drainage Workshop, The
Netherlands.
Latif, M., and W. U. Zaman. 1998 ―Constraints and main issues of Irrigation system in
the Indus Basin of Pakistan‖. Journal of Drainage and Water
Management 2: (2).
Latif, M., and M.S. Pomee. (2003) ―Impacts of institutional reforms on irrigated
agriculture in Pakistan‖. Research report, Centre of Excellence in Water
Resource Engineering, University of Engineering and Technology,
Lahore, Pakistan.
Lau, L.J. and P.A. Yotopoulous. 1971. ―A Test for relative efficiency and applications to
Indian Agriculture‖. American Economic Review 61:94-109
Lingard, J., L. Castillo and S. Jayasuriya. 1983. ―Comparative efficiency of rice farms in
Central Luzon, the Philippines‖. Journal of Agricultural Economics,
34:163-173.
187
Livingston, M. L. 2005. ―Evaluating changes in water institutions: Methodological issues
at the micro and meso level‖. Water Policy 7 (1): 21-34.
Lovell, C.A.K. 1993. ―Production Frontiers and productive efficiency‖. The
measurement of productive efficiency: techniques and applications. (cds)
H.O. Fried, C.A.K. Lovell and S.S. Schmidt, Oxford University press,
New York, USA, 3-67.
Madhav, R. 2007. ―Irrigation reforms in Andhra Pradesh: Whither the trajectory of legal
changes?‖ IELRC working paper, International Environmental law
Research Centre, Geneva, Switzerland.
Martin, E., R. Yoder and D. Groenfeldt. 1986. ―Farmer-managed irrigation: research
issues‖. Irrigation Management Network paper 86/3c. Overseas
Development Institute, London, England.
Mceusen, W. and J. Van den Broeck. 1977. ―Efficiency estimation from Cobb-Douglas
production functions with composed error‖. International Economic
Review, 18: 435-444.
McKay, J. 2005. ―Water institutional reforms in Australia‖ Water Policy 7 (1): 35-52.
Mirza, Z.I., M.U. Hassan and D.J. Bandaragoda. 2000. ―From state to jointly managed
irrigation systems‖. Social mobilization and management transfer
guidelines for large canal systems. Research Report No. R-108.
International Water Management Institute, Lahore, Pakistan.
Molden. D; M. Burton and M.G. Bos.2007. ―Performance assessment, irrigation service
delivery and poverty reduction: Benefits of improved system
management‖. Journal of Irrigation and Drainage 56: 307-320.
Nakashima, M., 1998. ―Water Users organization for institutiona reform in Pakistan‘s
irrigation sector‖. A report published by Faculty of international studies,
Hiroshima City University, Hiroshima, Japan.
188
Oza, A. 1998. ―NGOs and institutional reforms: A case study of irrigation sector reforms
in Gujarat, India‖. International workshop on Community Based Natural
Resource Management (CBNRM) Washington DC, United State.
Parikh, A., F. Ali and K. Shah. 1995. ―Measurement of economic efficiency in Pakistani
Agriculture‖. American Journal of Agricultural Economics, 77: 675-685.
PIDA, 2005. ―Reforms in irrigation sector: Vision, Strategy, Implementation,
Achievements‖. Report published by the Punjab Irrigation and Drainage
Authority, Punjab, Pakistan.
Penov, I. 2004. ―Institutional options for sustainable irrigation: An evidence from
Bulgaria‖.CEESA Discussion Paper No 2.
Pitt, M.M. and L.F. Lee. 1981. ―Measurement and sources of technical inefficiency in the
Indonesian weaving industry‖. Journal of Development Economics, 9: 34-
64.
Qureshi, M.A and A.U. Haq. 2006. ―Irrigation reforms in Punjab: The implementation,
experience and performance evaluation‖. Discussion paper presented in
the International Seminar, Lahore, Pakistan.
Rinaudo, J.D. 2002. ―Corruption and allocation of water: The case of public irrigation in
Pakistan‖. Water Policy 4: 405-422.
Rinaudo, J.D; P. Strosser and S. Thoyer. 2000. ―Distributing water or rent? Examples
from a public irrigation system in Pakistan‖.The Canadian Journal of
Development Studies XXI (1): 113-139.
Rinaudo, J.D. 2002. ―Corruption and allocation of water: The case of public irrigation in
Pakistan‖. Water Policy 4 (5): 405-422.
189
Radosevich, G.E. 1975. ―Water Users Organizations for improving irrigated agriculture:
Applicability to Pakistan‖. Water management technical report No. 44.
Water Management Research Project, Engineering Research Center,
Colorado state University Fort Collins, Colorado, U.S.A.
Reifschneider, D. and R. Stevenson. 1991. ―Systematic departures from the frontier: A
framework for the analysis of firm inefficiency‖. International Economics
Review, 32: 715-723.
Repetto, R. 1986. ―Skimming the water: Rent-seeking and the performance of public
irrigation systems‖. Report of the World Resource Institute, Washington,
DC.
Russel, N.P. and T. Young. 1983. ―Frontier production functions and measurement of
technical efficiency‖. Journal of Agricultural Economics, 34: 139-150.
Saleth, R.M. and A. Dinar. 2005. ―Water institutional reforms: Theory and Practice‖.
Water Policy 7 (1): 1-19
Sarah, L., O. Hara and T. Hannan. 1999. ―Irrigation and water management in
Turkmenistan: Past systems, present problems and future scenarios‖.
Europe- Asia studies, University of Glasgow 51 (1): 21-41.
Sarwar, S. 2006. ―On the road to irrigation reforms: From PID to PIDA. A review of the
irrigation management organizations in Punjab, Pakistan‖. A Master
Degree Thesis, University of Waghenien, Netherlands.
Schmidt, P. 1976. ―On the statistical estimation of parametric frontier production
functions‖. Review of Economics and Statistics, 58: 238-239.
Schmidt, P. 1986. ―Frontier Production Functions‖. Econometric Reviews, 4, 289-328.
Shaikh I.B. 2004. ―Efficient and sustainable irrigation management in Pakistan‖. Report
of the National Committee of ICID, Islamabad. Pakistan.
190
Shapiro, K.H. and J. Muller. 1977. ―Sources of technical efficiency: the role of
modernization and information‖. Economic Development and Cultural
Change, 25:293-310.
Smad, M. and D. Vermillion. 1999. ―Assessment of participatory management of
irrigation schemes in Sri Lanka: Partial reforms, partial benefits‖.
Research report No 34, IWMI, Sri Lanka.
Smajgl, A., R. Greiner and C. Mayocchi. 2005. ―Estimating the implications of water
reform for irrigators in a sugar growing region‖. Environmental modeling
and software 21 (2006): 1360-1367. available at www.sciencedirect.com
Samad, M. 2005. ―Water institutional reforms in Sri Lanka‖. Water Policy 7:125-140.
Svendsen, M. 1999. ―Integrated management of water in river basins‖. 20:20 Brief.
Washington DC: IFPRI.
Thiam, A., B.E. Bravo-Ureta and T.E. Rivas. 2001. ―Technical efficiency in developing
country agriculture: a meta-analysis‖. Agricultural Economics, 25:235-
243.
Tahir, Z. and Z. Habib. 2001. ―Land and water productivity: Trends across Punjab canal
commands‖. International Water Management Institute, Working Paper
No. 14, Pakistan Country Series Number 3.IIMI, Lahore.
Turral, H. 1995. ―Recent trends in irrigation- management changing directions for the
public sector‖. Alison Saxby Series, Overseas Development Institute,
Portland, UK.
Upton, M. 1976. ―The unproductive production function‖. Journal of Agricultural
Economics, 30: 179-194.
191
Vermillion, D.L.1997. ―Impact of Irrigation management transfer: A review of the
Evidence‖. Research Report No 11. International Irrigation Management
Institute, Colombo, Sri Lanka.
Vermillion, D.L. and J.A.Sagardoy. 1999. ―Transfer of irrigation management services
guidelines‖. Irrigation and Drainage Paper 58. Food and Agriculture
Organization, Rome, Italy.
Wang, H. and P. Schmidt. 2002. ―One-step and two-step estimation of the effects of
exogenous variables on technical efficiency levels‖. Journal of
Productivity Analysis, 18:129-144.
Welsch, D.E. 1965. ―Response of economic incentive by Abakaliki rice farmers in
Eastern Nigeria‖. Journal of Farm Economics, 47: 900-914.
Wichelns, D. 1999. ―Economic efficiency and irrigation water policy with an example
from Egypt‖. Water Resources Development 15 (4): 543-560.
World Bank.1994. ―A review of World Bank experience in Irrigation‖. Report No 13676.
Operations Evaluation Department, World Bank, Washington, DC.
World Bank. 1994a. ―Pakistan Irrigation and Drainage: Issues and Options‖. Agriculture
Operations Division. Afghanistan Maldives, Pakistan and Sri Lanka
Department, South Asia Region. Report No 11884.
World Bank. 2005. ―Pakistan, Country Water Resources Assistance Strategy, Water
economy running dry‖. Agriculture and Rural Development Unit, South
Asian Region. Report No 34081-PK.
Yercan, M., F. Dorsan and M.A. Ul. (2003) ―Comparative analysis of performance
criteria in irrigation schemes: A Case study of Gediz River Basin in
Turkey‖. Agricultural water Management 66 (2004): 259-266.
192
Yotopoulos, P.A. and J.B. Nugent. 1976. ―Economics of development‖. Harper and Row,
New York, USA.
Zhovtonog, O., W. Dirksen and K. Roest. 2005. ―Comparative assessment of irrigation
sector reforms in Central and Eastern European countries of Transition‖.
Journal of the International Commission on Irrigation Drainage (ICID)
54 (5): 487-500.
193
Annexure- I
List of Abbreviations and Acronyms
Aabiana Irrigation Water Charges
AWB Area Water Board
CCA Cultureable Command Area
DPR Delivery Performance Ratio
FAO Food and Agriculture Organization
FO Farmer Organization
GCA Gross Command Area
GDP Gross Domestic Product
GOP Government of Pakistan
GM Gross Margin
GVP Gross Value Product
HA Hectare
Hectare 2.47 acres
IMT Irrigation Management Transfer
KG Kilogram
Kharief Summertime, Warm-wet season, officially mid
April to mid October
LCC Lower Chenab Canal
LCC (E) Lower Chenab Canal (East)
MAF Million Acre Feet
Maund 40 Kilograms
Mds Maunds
PID Punjab Irrigation Department
PIDA Punjab Irrigation and Drainage Authority
Rabi Wintertime, Cool-dry season, officially mid
October to mid April
Rs. 62.65 Equivalent to 1 US $ (As on March 8, 2008)
194
Annexure- II
Snapshots of Field Visit
Physical work at Jaranwala distributary and minor
Community participation in O&M activities
195
Visit to the FO office and diversion Head Shahkot
Illegal water diversion through Pipe Damaged outlet structure
196
Informal discussion with farmers
Meeting with FO office bearers