Review of Literature on Productivity

2.1. Introduction

The review in this study is restricted mainly to literature on productivity in

agriculture. There are studies at international level comparing the level and growth
trend of countries, developed as well as developing, studies at national and regional
level. However, with respect to the time frame, the methodology employed, the
nature of data used, the number of variables examined and the estimation procedure
adopted, the conclusion of the studies vary widely. The studies reviewed are
arranged in chronological order so that it enables us to trace the historical evolution
of the methodology used, the improvement in data coverage and estimation
procedure and the contribution of each piece of research to the stock of knowledge.

2.2. International Studies

A significant impetus in understanding the factors influencing international

agricultural productivity was provided by Hayami (1969, 1970)1,2 and Hayami and
Ruttan (1970)3, with updating by Nguyen (1979)4, Kawagoe and Hayami (1983)5,
and Kawagoe, Hayami and Ruttan (1985) 6. They stressed the influence of education
and human capital on productivity growth. By conducting a separate analysis for
developed and developing countries, Hayami and Ruttan (1970) implicitly assumed
that technology might change with the level of development. This raises the question
of how agricultural productivity varies across countries.

In an attempt to explain cross-country differences in agricultural

productivity, Evenson and Kislev (1975)7 emphasized the role of research. using a
Cobb-Douglas production function. Antle. J.M (1983)8 focused on the effect of
transportation and communication infrastructure on aggregate agricultural
productivity for the period 1965-66 for 66 countries comprising 47 less developed
and 19 developed nations for all crops using secondary data and Cobb-Douglas
function. The variables used in the study are land, labour and livestock variables that
10
represent the country’s resource endowments of these inputs. The fertilizer variable
is used as a proxy for ‘modern’ inputs. The education variable as a measure of the
country’s human capital stock in 1965. The research variable was included to
capture the effect of investments in agricultural research on the Hicks-neutral
productivity level of each country’s agricultural sector. The study tested Hayami-
Ruttan hypothesis that production function coefficients are equal across the
developed and less developed country groups. The hypothesis was not rejected for a
model that accounts for variation in the Hicks neutral productivity levels due to
difference across countries in human capital, research and infrastructure.

Using an econometric approach, Nehru and Dhareshwar (1994)9 examined

sources of TFP growth in 83 industrial and developing countries for the period 1960-
1990. They found that human capital formation was three to four times more
important than raw labour in explaining output growth. Using human capital as a
separate variable, they found that the countries with the fastest growing economies
have based their growth on factor accumulation (human capital, labour and physical
capital), not growth in efficiency or technology.

Developing a non-parametric approach to production analysis Chavas

(2001)10 investigates international agricultural productivity. The analysis develops
productivity indexes based on Shephard's (1970)11 input distance function and non-
parametric representation of technology. The approach is applied to FAO annual data
of inputs and outputs for twelve countries for the period 1960-1994. It estimates
separate production frontier for each country and therefore it does not require the
assumption that the production technology is similar across countries. It was found
that technical efficiency was decreasing in most of the countries during the late 1960s
probably due to environmental degradation.

Data Envelopment Analysis to derive Malmquist productivity indices was

utilised by Rao, and Coelli (2004)12 to examine level and trends in agricultural output
and productivity for 1980 – 1995 for 97 countries treating crops and livestock as
output variable. Time series data provided by the FAO and the ILO have been used
to estimate Total Factor Productivity. The study found that TFP was growing at rate
of 2.7 percent annually. China witnessed spectacular performance recording 6.8
percent TFP growth. India posted a Total Factor Productivity growth of 1.6 percent.

11
Asia is the major performer with TFP growth rate of 3.1percent and Africa registered
the weakest performance with 0 .9 percent growth.

2.3. National Level Studies – Rest of the World

Numerous studies were conducted on the productivity theme in the United

States and other Western European countries. The present review does not touch up
on these studies, as they will be used in the context of methodology. Only a few
studies are taken up for review.

On the sources of productivity gains in Japanese agriculture as compared with

the U.S experience for the period 1952-54 and 1959-61 Kaneda (1967) 13 used the
output elasticity estimates derived from Cobb-Douglas production function as the
weights for land, labour and capital in the computation of total productivity index.
The annual growth rate of residuals (rate of technical change) is found to be
2.9percent. Land productivity and mechanization are the two factors that contributed
95percent of productivity gain in labour.

Wen G. J (1993)14 examines the performance of the successive rural

institutions in China in terms of changes in Total Factor Productivity for 1952-89
using Secondary time series data for crop and non-crop output. The study estimates
PFP and Total Factor Productivity using TT index. The study concluded that
Household Responsibility System (HRS) was more efficient in resource allocation
than the commune system in increasing Total Factor Productivity. In more than 20
years the commune system succeeded in improving land productivity but worsened
labour productivity. In the context of HRS both rose.

Using farm level cross-sectional survey data Wadud and Ben (2002) 15
measured farm-specific technical inefficiency of rice farmers in Bangladesh.
Technical inefficiency effects are modelled as a function of environmental factors,
irrigation infrastructure and farm-specific socio-economic factors.

The study estimated a Translog stochastic frontier production model with the
specification of technical inefficiency effects model which included farm-specific
factors like age, education of farmers, land fragmentation, irrigation infrastructure
and land degradation, using the single stage formulation applying maximize
likelihood techniques. The study revealed that technical inefficiency effects tend to

12
be significantly influenced by the factors measuring environmental degradation and
irrigation infrastructure. Younger farmers, who adopt new technology, are
educated and those who have easier access to credit are most likely to operate
farming activities efficiently. Moreover, larger the land plot size the greater is the
technical efficiency. Land management and land tenure policies in reducing the land
fragmentation remain critical to improving technical efficiency and hence household
welfare.

2.4. National Level Studies – India

Relatively large volume of literature on productivity is available in Indian

Agricultural economics. Majority of studies conducted at all India level compute
partial productivity measures, particularly land productivity. Mukherjee and
Vaidyanathan (1980)16 used econometric technique and fitted two models to study
the effect of rainfall on yield variation and alternatively the effect of rainfall, inputs,
the interaction between their effects and sustained changes in output due to
productivity changes for the period 1950-51, 1975-76 using time series data. As
regards the use of inputs Kerala’s input index showed 13percent fall over the period
considered.

Bhalla & Alagh (1983)17 made an attempt to examine the levels and changes
in labour productivity in agriculture in 281 districts of India during 1962-65 to 1970-
73 and to statistically explain the inter-district variation in labour productivity in
these regions by factors like intensity of use of capital and other important inputs.
The variation in male worker productivity is explained using double log regression
analysis. The study found that tractor alone was able to explain fairly large variation
in labour productivity in both periods. Since high growth districts were the
predominant users of available tractors and tube wells, capital intensity per worker
was the main determinant of labour productivity in these districts. Fixed capital alone
was important, but additional variables like gross cropped area and fertilizers
resulted in significant increase in the explanation of variation in labour productivity.
As investment in modern equipment like tractors and tube wells was limited and
even the working capital requirements were quite low in traditional agriculture,
labour productivity depended more on the vagaries of monsoons than on quantum of
fixed and variable capital.

13
 Bhatia (1992)18 tried to understand the changes in input efficiency in the
production of wheat in the major growing states of India during post-green period,
1970-71 to 1990-91. Using time series data on wheat for Punjab, Haryana, U.P, and
M.P the partial Factor Productivity of land, labour, fertilizer and total factor
productivity are computed using multiple linear regressions. The study found
improved input use efficiency over the years; labour productivity and land
productivity have improved resulting in the reduction in unit cost due to the use of
land saving technologies of high yielding variety seeds, fertilisers, insecticides, etc.
and the marginal productivity of fertiliser came down particularly in Punjab due to
its over use.

Roségrant and Evenson (1992)19 tried to assess Total Factor Productivity

growth, traced the sources of productivity growth and estimated the rate of return to
public investment in research and extension in India for 1956 –1987 using cross
section and time series data at district level. Törnqvist -Theil index was computed for
271 districts in 13 states in India for the said period. To assess the determinants,
Total Factor Productivity index was regressed as a function of variables such as
investment in research, extension, human capital and infrastructure. Altogether, 15
independent variables were used in regression. Dummy variables were used for
agro-climatic zones. The study found that main sources of productivity growth were
public research and extension and private research. The rate of return to public
research and extension were high indicating continued profitability of public
investment in agricultural research and extension.

Dholakia and Dholakia (1993)20 aimed to estimate the sources of growth of

Indian agriculture for three sub-periods during 1950-51 to 1988-89 for all crops. It
also estimated the contribution of adverse weather conditions and intensity of
resource use to total factor productivity growth. The analysis was developed in
growth accounting framework, worked out the growth rates of output and inputs,
land, labour and capital using time series data. Principal component analysis was
used to identify the determinants of TFP growth so as to avoid the problem of multi
co-linearity, which is unavoidable in time series data. The study confirmed the
hypothesis that modern agricultural inputs like fertilizer, HYV seeds and irrigation
determine the TFP corrected for weather.

14
 In order to address the issues like the contribution of growth to total output
growth, the sources of productivity growth and the returns to agricultural research,
Kumar and Roségrant (1994)21 attempt to assesses Total Factor Productivity growth
for rice in different regions in India and examine the sources of productivity growth.
Index number approach to growth accounting has been used. Divisia -Törnqvist
index for TFP was calculated. To estimate the impact of probable variables on TFP,
the TFP index was treated as the dependent variable and regressed by nine relevant
variables. Time series data from different regions were pooled and dummy variables
are included for regions, keeping the eastern region as the reference region.
Estimation was undertaken using a fixed effects approach for the pooled cross
section time series regional level data set for 1970-71 to 1988-89. The study
revealed that increase in area and production of crop was highly associated with their
relative profitability Rice area increased slowly since 1980s mainly through
substitution from coarse cereals. Market infrastructure, research, canal irrigation,
balanced use of fertilizers were found to the important source of TFP. Future
productivity gains in rice production would have to be achieved from the eastern and
the southern regions of India.

Panigrahi, Ramakrishna (1995)22 analysed the growth pattern and identified

factors affecting productivity levels across states and examined the role of population
in land productivity growth for the period 1967 – 68 to 1991-92 in India during the
post green revolution period for principal crops divided into food crops and non-food
crops. Secondary time series data was used and multiple regression method was
applied.

Bhalla and Singh (1997)23 examined the contribution of Indian agriculture in

the productivity frame for the period 1980-83 to 1992-95 for the states in India. The
contribution of area to the output growth has drastically diminished; the expansion of
gross cropped area through double cropping has increased. Cropping pattern changes
were positive since low yield and low value coarse cereals were replaced by high
value oilseeds as well as rice and wheat, without adversely affecting food grain
output. A rise in labour productivity was another striking feature during 1980-83 to
1992-95.

15
 On studying the relationship between farm prices and aggregate agricultural
production, as well as agricultural productivity in 20 Indian districts belonging to
seven states taking into account seventeen harvests, Schafer (1997)24 examined
whether changes in farm prices were an important, or even the sufficient factor in
effecting changes in cultivated area or in aggregate agricultural production.
Regression method is employed. The expected price at time t was estimated by
Nerlovian Model. The study could find only a weak positive correlation between
farm prices and agricultural area. No statistically significant relation could be
established between farm prices and agricultural productivity. No significant
relationship between agricultural production and farm prices was found. However,
there is relationship between infrastructure, proxied by irrigation or length of roads,
and agricultural production. This supports the view that prices do not have a proven
short-term effect on aggregate agricultural production but infrastructure plays a good
role. Policy implication is that non-market factor, including provision of better
infrastructure and socio-economic environment influences productivity rather than
the accepted wisdom of market factors.

Mittal and Kumar (2000)25 used Total Factor Productivity as a measure of

quality of inputs and technology. Using cross section and time series data for 1973-
95 for wheat and rice Törnqvist-Theil index is worked out. The period was
subdivided into 1973-90: pre-trade liberalization; 1990-95: short term liberalized
economic environment; 1973-95: long term perspective. In order to examine the
impact of literacy on TFP simultaneous equation model was utilised. The study
found that literacy was positively related to crop productivity and strongly linked to
farm modernisation.

Assessing the TFP growth, measuring the changes in real cost of production,
identifying the sources of productivity and estimating marginal rate of return to
public investment on wheat production in different states (Punjab, Haryana, UP, MP
and Rajastan) in India were the aims of Mittal, Surabhi & Lal, R.C (2001) 26.
Secondary data on inputs, their prices were collected for the period 1971-72 and
1995-95. State level time series data on area, yield, production, irrigation, HYVs,
rural literacy rate, village electrification, sources of irrigation etc. were taken from

16
published reports of GOI. Using the accounting framework the Divisia-Törnqvist
index is used to compute the total output, input and input price indices.

Cross section time series data were used in the estimation of TFP
decomposition model using three stage least squares (3SLS) estimation framework.
The study found that the TFP index for wheat rose at the rate of 0.9percent per
annum and contributed about 24percent to the output growth. Research investment,
quality of inputs and rural infrastructure are the most important determinants of TFP
growth. There is a need to target public investments to agricultural research, ground
water irrigation, electrification to areas where the yield levels are still low.

Long run relationship between land productivity of food grain and

government investment in agriculture in India was the thrust of Shyjan (2003) 27 using
secondary data of food grains for the period 1974-75 to 200-01. The study found that
growth rate in productivity was not uniform through out. Growth rates of
productivity were higher for cereals than for pulses. Kerala and Punjab had levels of
investment higher than the national average during all the plan periods. They stood
above for food productivity too. Lag in food grains productivity and investment is
more than 12 years in Kerala and Punjab. The lag is more in those states where share
of cereals is more in food grains.

For analysing the temporal and spatial variations in the sustainability status of
the crop sector in Indo Gangetic Plane (IGP) and identifying sources of Total Factor
Productivity of food crop sector in the region Kumar, et. al (2004)28 used district wise
secondary data for the period 1980 – 81 to 1996 – 97 for the states, Punjab, Haryana,
U.P, Bihar, and W. Bengal. Growth accounting approach was adopted for computing
TFP growth. Divisia-Törnqvist index was used. To find the determinants of TFP, its
index was regressed on a number of variables. In order to find the impact of
infrastructure on total factor productivity an Infrastructure Index was computed using
six major infrastructural facilities – transport, energy, irrigation, banking, education
and health. Energy, transport, irrigation, and finance are economic infrastructural
facilities while education institutions and health facilities are considered as social
infrastructure. The major findings of the study are: The TFP index of the crop sector
in IGP rose by 1.2percent annually during 1981 to 1987; productivity alone
contributed 1/3 of the increase in output; productivity growth attained in 1980s was

17
not sustained in 1990s and The public policies such as investment in research,
extension, education and infrastructure (road, electrification, educational institutions,
health facilities, banking etc.) have been the major sources of TFP growth.

2.5. Regional Studies –other than Kerala

In this section, we deviate from the chronological approach generally

followed hitherto in reviewing earlier studies. The studies here are clubbed state wise
so that region wise comparison of the studies is possbile.

Majority of regional studies during the 1960s and 70s focussed on testing the
Shultzian hypothesis that traditional farmer utilises resources efficiently. Reddy
(1967)29 tested the hypothesis and approved it for Andhra Pradesh. Agricultural
production in India may not be increased simply by increasing all inputs in the
traditional state of art but should introduce a modern technology package. The
package should consist of new inputs, agricultural education, special skills, and
techniques, and guidance in farm planning. Primary data for 1957-58 in West
Godavari district for rice and tobacco in 3 villages were collected and used CD
function to estimate the marginal factor cost and the marginal value product.
Efficiency is defined in terms of marginal factor cost and marginal value product,
MFC MVP ≤1 .

Iqbal and Azeemuddin (1993)30 assessed and evaluated the impact of tube
well irrigation on input utilization in the paddy crop and to find out its relationship
with output and assessed the productivity and profit of paddy growers in Tribal and
Non-tribal area of AP. Using Cross-section data for 1992 the study could find that
variables like season, seeds, fertilizers, pesticides and hired labour components were
significant in influencing the value per acre of paddy crop. Manures and acres were
found to be nonsignificant.

The objective of Kumar et. al (1984)31 was to estimate the yield per hectare of
wheat in the three agro-climatic zones (low hills, mid hills and high hills) so as to
determine the levels of land productivity in different soil and climatic conditions of
wheat in Himachal Pradesh. Average yield for three years was computed as a
combined average of yield of individual years. Sample data for three years from
1979-80 to 1981-82 were analysed. The yield of irrigated wheat is highest in the low
hills. The yield of irrigated wheat has been found to fall as the altitude increases.
18
 Juyal. R.K (1987)32 was to enquire into the effect of health on productivity,
using district level aggregate data in respect of 54 districts of Uttar Pradesh. Scores
of variables effect productivity such as labour related, like literacy, type of
employment, health. Others relate to non-labour or physical variables, like
urbanization, roads, consumption of electricity, fertilizers. A set of 17 variables was
included. And, stepwise regression was run. It was shown that among the variables
concerned with attributes of labour, health variable has positive and significant
contribution toward productivity.

Sau (1990)33 analysed the productivity differentials in rice across the major
rice producing states of India and in rice and food grains across the districts of West
Bengal. It uses econometric method. Productivity in agriculture as measured by the
output per hectare depends mainly on two factors: the level of inputs used and
resource base and infrastructure development of the region. The indicators of inputs
used are: Percentage of gross area irrigated to gross cropped area; Fertilizer used per
hectare of cropped area; Percentage of HYV area to gross cropped area and credit
availability per hectare or per capita. Infrastructure, in its broad sense covers the
items that provide external economies to the farmers as a whole. Its indicators are:
Road mileage per square kilometer; percentage of electrified villages and Literacy
rate.

Pillai (2001)34 focuses on the role of input utilization in paddy cultivation

during the 80’s and early 90’s. The exercise is carried out for West Bengal and
Orissa by examining a) changes in aggregate, state level factor productivity growth
b) the extent of technical inefficiency, and difference if any, across seasons, crop
varieties, status and overtime and c) whether there is perceptible shift in production
technology overtime in the case of west Bengal. The upsurge in agricultural
productivity in general is attributed to two major factors: 1) Role of institutional
changes and 2) Wider adoption of new technology, better utilization of fertilizers,
credit and so on. Using time series data for 1971 to 92 for West Bengal, for 1971 to
93 for Orissa and farm level study (cross section data) is carried out for 1986-87 for
West Bengal and 1990-91 of Orissa separated by season and crop variety. Growth
accounting method is used and Divisia-Törnqvist index is computed. The study

19
found that input productivity played an important role in the growth performance in
80’s and 90’s. The performance of West Bengal was better than Orissa.

Ahmed and Bhowmick (1991)35 attempted to examine: the extent of

variability in area, yield and price; the nature of price, productivity and acreage
movement over the period in relation to technology; and the factors responsible for
determining the area under the principal crops. Price, productivity (kg/hect),
productivity of competing crops are the factors influencing the variation of acreage
of a crop in Assam. The study is based on five principal crops, rice, rapeseed &
mustard, jute, wheat, and potato in Assam. Time series data were used. Time period
variable was used as a proxy for technology. Simple linear model was used for the
purpose with P, Y, A = f(t). Supply response function was fitted in line with the
Nerlovian type. Both linear and double log form of the function were fitted and linear
form was found to be more representative. The study concluded that acreage under
the principal crops in Assam is influenced by the yield of the crop, lag acreage and
area under irrigation. Farmers in Assam were not at all price responsive for the crops
considered here with respect to allocation of acreage. It means that farmers in this
area are yet to take farming as a business.

Sarma (1999)36 analysed the impact of mechanisation on land productivity in

the Jorhat district of Assam for 1991-92 covering the crops: Rice, Mustard, and
Potato. Field level data was used and multiple regression was run. The study found
that mechanisation improved productivity. Borbora and Mahanta (2001)37 examined
the factor intensity in agriculture of Assam. To examine the impact of fertilizer and
rainfall on agricultural productivity for the period 1990 – 1999 in all the 23 districts
of Assam Time series data was utilised to estimate marginal productivity of land
specifying Cobb-Douglas production function. The parameters were estimated by
using least squares method. Borbora and Mahanta (2002) 38 tried to identify inter-
district disparities in the agricultural sector in 23 districts of Assam. The period of
study is:1998-99. Crops are not specified. Secondary data, cross-section was used
for estimating land productivity. The method of analysis is: Principal component
method used to determine factors influencing production. To measure inter-district
disparity in agricultural development in the state 13 indicators were developed. The

20
study found that inter-district disparity existed in proportion to the area under
cultivation.

Patil and Kalyankar (1992)39 studied the impact of research on productivity of

major crops grown in Marathawada region of Maharastra. Time series data on area,
production and productivity per hectare was collected from 1970-1990 and trend
analysis was done using regression. Production and productivity of crops included in
the study increased during the period covered.

Singh and Kaur (1992)40 attempted to study the growth in productivity in

different districts of the Punjab during the 70s and 80s; to examine the inter-district
gap and variation in agricultural productivity in Punjab; to study the relationship
between incremental productivity and incremental capital investments and type of
capital investments. Growth rates were worked out. Rank method was adopted to
compare the productivity differential among districts. Agricultural productivity in
Punjab was increasing at the rate of 4percent per annum during the 70s and 80s.

Sidhu and Byerkee (2001)41 tried to find out the major sources of growth in
wheat productivity in the post green revolution period after HYVs were widely
adopted; the prospects for continuing to exploit these sources of growth; slower
growth in productivity reflect changing input-output ratio that provides a disincentive
for investing in improved technology; increased yields and changing practices
reflected in changes in costs of production and TFP in wheat; extent of increase of
productivity in wheat been captured by producers or passed on to consumers in the
form of lower prices. Data collected from the Comprehensive Scheme for the Study
of Costs of Cultivation of Principal Crops in Punjab has made use of. Total
productivity index has been worked out. This index is superior to the usual approach
in India of equating productivity with yield per unit of land area, which does not
account for increasing importance of purchased inputs in substituting for land in
wheat production. More over, the TFP index has been proposed as way of measuring
sustainability. The study concluded that important source of productivity has been
labour saving technology. The use of Bio-chemical technologies has only modestly
increased TFP. Further productivity gains in wheat production could be achieved by
more efficient use of inputs such as fertiliser and water.

21
 Adams and Bumb (1993)42 tried to know the sources of district wise
agricultural productivity in Rajastan for 1971 covering all crops. Seventeen variables
were included. Land Productivity was estimated using Factor analysis. The result of
the study showed that land productivity depended directly upon three things: supplies
of conventional inputs, the cropping pattern and cropping intensity, and the use of
modern mechanical and chemical technologies. Infrastructure and institutions are
facilitative or intermediary variables.

Singh and Singh (1993)43 examined the growth rates of area, production and
productivity of gram in different districts of Bihar; estimate the factors affecting the
area, production and productivity of gram. Based on the district wise secondary time
series data on area, production, productivity, prices, irrigated area, and annual rainfall
from 1960-61 to 1989-90. Compound growth rates have been estimated. In order to
study the impact of various factors on yield a Cobb Douglas type function was fitted
using the output per hectare as the dependent variable and annual rain fall, area under
irrigation, current year prices (Rs./qtl.) and one year lagged prices (Rs./qtl.) as
independent variables. Conclusion was that though the area under pulses were
mostly pre-determined and as the area under irrigation increased it was gradually
substituted by cereal or cash crops; the production could be increased either by
motivating the farmers of non-traditional pulses growing area or by introducing
pulses crops in no-traditional crop seasons.

Nanavati and Patel (1994)44 examined the relative position of districts after
the green revolution for 1970-71 to 1984-85 taking food and non-food crops in
Gujarat. Land productivity was estimated using secondary cross-section data. The
method of analysis was simple in that coefficient of variation of output among
districts was worked out. It concluded that the share of food and non-food crops in
SDP remained 50:50; govt. policy towards agriculture should be on the basis of agro-
climatic zones. Shiyani and Maurvi (2002)45 estimated total factor productivity
growth for 4 major oilseed crops viz., groundnut, castor, mustard and sesame in
Gujarat. TFP is measured as an index of total output to index of total factor inputs
and encompasses the impact of technical change and change in the level of all inputs.
Time series data on area, production and yield per hectare of four oil seeds have been
collected. Data on crop inputs included human labour (man days), bullock labour

22
(days/hectare), seed (kg/hect.), insecticides/pesticides (Rs/hect.), miscellaneous cost
which includes the cost of transportation and other paid out expenditure (Rs/hect.),
depreciation cost of farm buildings and implements (Rs./hect.), and rental value of
owned land (Rs./ha.). The Tornquist Theil index was used for computing the total
output index, total factor productivity index. The growth rate of total output indices,
total input indices and total factor productivity indices were worked out. The acreage
and yield per hectare of all the crops improved substantially during the 40 year
period. Positive rate of growth was found in the case of castor, while others showed
negative growth rate.

Singh and Nandal (1995)46 examined the effect of land degradation on

agricultural productivity and its magnitude for the period 1989-90 covering all crops.
Primary data were used to compute land productivity. Cobb-Douglas production
function was estimated using regression technique for Haryana. It found that
appropriate choice of technology like surface and sub surface drainage, improved
crop varieties etc. could reverse the process of degradation and improve productivity.

Jain and Idnani (1996)47 was to envisage the cost return structure according to
size of holding; determine the resource productivity of rainfed paddy in different size
of holding and; to suggest measures to improve the productivity of rainfed paddy.
Secondary data on 80 cultivators in Chatisgarh region of Madhyapradesh was
collected. A Cobb-Douglas production function was fitted for human labour, bullock
labour, capital inputs and fixed inputs on paddy production. Estimating land
productivity the study concluded that rainfed cultivation was profitable only for
medium and large scale farmers. Kawadia (2000)48 was to decompose labour
productivity to identify the factors contributing it. As cultivators and agricultural
labourers cannot be distinguished average product of workers is termed as labour
productivity for all crops. The method used by Bhalla and Alagh (1983) was
employed. Labour productivity growth is decomposed into growth rate of land
productivity, cropping intensity and land labour ratio.

Mythili and Shanmugam (2000)49 measured farm level technical efficiency

for rice in TamilNadu using farm level data for 1990-91, 91-92, 92-93. Frontier
Production Function Technique was applied using the Cobb-Douglas production
function type applying Maximum Likelihood method of estimation. Total output in

23
quintals is regressed on human labour in man hours, area in hectares, fertilizer in kg.,
capital expenditure incurred on bullock labour, machinery and pesticides in money
terms. It concluded that technical efficiency ranged between 46.5 and 96.7 mean TE
being 82percent. Extension services and training programmes do contribute to TE.
Govindarajan et. al (2004)50 analysed the productivity variation over time and space
effects using error component model and to suggest suitable policy measure for
improving the rice productivity. It covered the period: 1980-2002. The area of study
was Cauvery Delta region consisting Kumbakonam of Tanjore district, Mannagudi
taluk of Nagapattinam district of Tamil Nadu. Both primary and secondary data were
used to estimate land productivity. Productivity is hypothesised to depending on
inputs, time, region and other random factors. Time effects are changes in
technology, management efficiency, soil quality, climate, status of management,
input quality such as water, labour skill and the level of service provided in the
region. The productivity variation can be decomposed into input effect, regional
effect, time effect and stochastic effect. The variation in output due to changes in the
level of input used is the input effect on output. The regional difference in output
due to variation in availability of say water, quality water, management services is
the spatial effect. The variation in output due to change in time period was termed as
the regional effect. Productivity decomposition is carried out by the CD form of
Time Varying Stochastic Frontier Production Function (TVSFPF). The study found
that except for seeds, all other inputs contributed positively to productivity. There is
very little space for improving productivity as majority of sample farmers have
reached more than 80 percent efficiency levels. Farmers should be taught to adopt
modern practices of farming by intensifying the extension activities of the State
Department of Agriculture.

2.6. Regional Studies – Kerala

There are studies focusing on one or more crops or on the agricultural sector
as such. Invariably all studies attempt to estimate land productivity defined as the
output per hectare. One earlier study on Kerala economy on productivity theme was
that of Oommen (1962)51. Defining productivity as yield per acre the author worked
out land productivity of important crops except rubber in Kerala during 1950-51 to
1960-61. A simple linear regression was fitted for time to find the growth rate. The

24
study found that though productivity of cereals and pulses generally showed a rising
trend, many other crops especially cash crops, did not show a steady rise in
productivity. Factors behind productivity are identified as irrigation, improved seeds,
fertilizer consumption, and disease controls.

Pillai (1969)52 was analysing the impact of irrigation on rice production in

Kerala using aggregate data for the period 1952- 53 to 1965-66 for rice using
secondary data. Using OLS method of regression and treating irrigating and un-
irrigated area as exogenous variables the author estimated land productivity. The
study concluded that irrigated farms yield more output than un-irrigated farms.

George (1979)53 worked out land productivity for food and non-food crops.
Simple linear growth rate was used to estimate the trend in output and area of crop
cultivation. The period covered is 1952-53 and 1974-75. Time series data was
utilised. Growth rate in agriculture was contributed more by increase in area under
cultivation than by increase in productivity during the period. The price non-price
factors responsible for increase in productivity were examined and found that price
relations have adversely affected the cultivators in respect of most of the crops. Large
increase in the growth rate of area under non-food crops, accompanied by negative
growth rate in productivity should be a matter of serious concern. The stagnant
growth rate of food production especially rice, despite large increase in the supply of
irrigation, calls for closer look at our irrigation projects. In view of favourable
response of individual crop areas to price, there is need for evolving a long-term
input-output price policy consistent with the land use capability and the requirements
of overall economic growth. Long-term policy agricultural research is called for it
will enhance technological progress and thereby agricultural productivity.

Jeemol Unni (1981) 54 analysed the reason for the shift in cropping pattern
away from rice to coconut for the period1960-61 to 1978-79 using secondary data.
Land productivity was computed by simple statistical values like ratios, percentages
etc. The result of the study was that coconut was substituting for rice in wetlands
because of relative profitability of coconut single cropped paddy land than double
cropped converted faster. Higher labour fertiliser cost in paddy while paddy prices
fluctuated.

25
 A first comprehensive study on Kerala agriculture economy is that of Pillai
(1982)55. The aim of the study was to analyse in summary measures the growth of
agricultural output in Kerala during the period from 1952-53 to 1978-79 as against
the overall performance of the state economy. It also looked at (i) whether the green
revolution in Indian agriculture brought any noticeable impact on Kerala’s
agriculture, (ii) whether this sector was yet poised for sustained growth with stability
and (iii) whether agricultural growth started decelerating in the recent past. Time
series data has been used to construct index of area, production and partial
productivity of all crops, food crops and non-food crops separately. Annual
compound growth rates as a summary measure of growth and range and standard
deviations as measures of variations to examine the stability or otherwise in the
growth were used. The study concludes that there has been a gradual decline in the
share of agriculture in the total SDP, particularly from the mid-sixties. During the
mid sixties per worker contribution to SDP of the primary sector has been on the
decline and that of the secondary sector on the increase. There has been a shift in the
cropping pattern in favour of plantation crops. Among the food crops, increase in the
production of tapioca has been substantial. Pulses, pepper, cashew nuts and coconuts
declined. Sugarcane and banana experienced moderate gains in yield. Rice, tapioca,
rubber and tea recorded impressive increase in yield.

In Indian agriculture the general performance of agricultural production and

yield was better during the 70’s than in the 60’s. However, it was the other way
round in Kerala. In Kerala’s agriculture the 60’s witnessed a period of accelerated
growth and after the 70’s a period of decelerated growth. Kerala’s agriculture
witnessed during the sixties much higher growth rates in area as well as output, but
not in productivity. The growth of productivity in Kerala’s agriculture always lagged
behind that in Indian agriculture. As arable land is limited in Kerala and, the only
way to increase production is to concentrate on productivity. Sivanandan (1985) 56
attempted to know the nature of variation among crops, seasons and regions and to
identify factors influencing such variations for the period 1960-61 to 1982-83. The
method of curve fitting was effectively used; growth curve of second degree was
fitted. Since 1970-71 was taken as the base year for calculation, 1977-78 gives the
year in which growth rate is zero and turns to be negative. It was in 1975-76, the
new system of land utilization survey was introduced. Area classification was
26
according to agro-climatic zones. The study found that determinants of growth or
decay of production could be identified as variations in land capability, effectiveness
of irrigation, use of fertilizers, spread of HYVs and availability of credit.

George and Mukherjee (1986)57 analysed the changes in the growth pattern of
rice in Kerala over and across time (between 1960-61 to 1974-75, I period and 1975-
76 to 1983-84, period II), across seasons (autumn, winter and summer), and across
space (major rice growing districts), irrigation and relative prices in explaining the
changes in area, yield and production of rice. Regression is run to estimate compound
growth rate of area, yield and production. Additive decomposition model was used
to identify the contribution of area, yield in explaining the changes in production
over the period following Minhas and Vaidyanthan (1965), Dharam Narain (1977)
and Vidya Sagar (1980). The study concludes that it is unlikely that the area under
paddy in the state could be increased. While maintaining the parity between paddy
prices and wage rates might prevent farmers from keeping land fallow, price
incentives are unlikely to induce a shift in the cropping pattern in favour of paddy.
There is scope, however for increased production through changes in technology,
particularly HYV and fertiliser application. However, this can be effective only if
irrigation facilities, both surface and ground water are utilised efficiently. Thus, the
strategy for increased rice production in Kerala should be based on improved
utilisation of irrigation facilities, use of HYV and efforts to maintain farm level
income either through remunerative output prices or through stable cost of
production. It is important to explore the possibilities of increasing yield through
institutional mechanism, particularly consolidation of holdings as proposed by Raj
(1985).

Kannan and Pushpangadhan (1988, 1999)58 explained the agricultural

stagnation that set in Kerala since the mid 70s. The study covers the period between
1962-63 and 1985-86, which seems to show that there has been two distinct phases in
terms of agricultural growth. During the sixties and up to the mid 70s (1962-63 to
1974-75) there has been an over all increase in the rate of growth of area, production
and yield for all the crops while in the following period 1975-76 to 1985-86 there has
been a near stagnation in the growth rate of aggregate area, production and land
productivity.

27
 Index numbers were constructed for each crop with 1962-63 as the base in
order to overcome the problems of different units of physical output. A measure of
aggregate output was obtained on weighting index number of output of each crop by
its share in the total gross cropped area for the year 1962-63. These aggregate
measures are then used for the analysis of the performance of Kerala’s agriculture.

The overall performance of the agricultural sector can be measured by the

growth rate of output and its components, namely, area and yield. Two methods are
used: the decomposition method pioneered by Minhas and Vaidyanathan (1965) and
the second is a statistical estimation method using various functional forms. A
second-degree exponential function is used for measuring growth rate. The period
wise growth rate is based on kinked exponential function.

The study concludes that the decline in agriculture during the 70s has wiped
out the gain in growth during the 60s. The stagnation in Kerala’s agriculture is
attributed to the inadequate as well as ill conceived development of critical factors
such as water management and land development and increasing environmental
degradation taking place in Kerala since mid 70s.

To test the hypothesis statistically that though in general Kerala agriculture

was stagnant, it was severe for individual crops, Pushpangadan (1988) 59 studied land
productivity of tapioca using secondary data for the period 1963-64 to 1974-75 and
1975-76 to 1985-86. District wise analysis of the growth rate of tapioca using the 2 nd
degree function in terms of area, yield and output was conducted. The result of the
study reveals that state level finding is not valid even for single district. Causes of
stagnation of tapioca are identified as: demand constraints, low profitability
compared to competing crops; and relative price of and income from competing
crops, rubber.

Thomas et. al (1991)60 analysed the trend in area, production and land
productivity and estimated the output response behaviour of tapioca in the state.
Secondary data supplied by GOK, Statistics for Planning for the period 1960-61 to
1986-87 is used for the analysis. Trends are estimated by a semi logarithmic function.
Area and yield response are studied by alternative linear forms of Nerlovian type
function. The lagged adjustment model proposed by Nerlove postulates that actual
acreage under a crop in each period is adjusted in proportion to the influence between
28
the desired area and actual area in the previous period. The result of the study is
summarised in the table below.

The acreage showed a declining trend. The positive growth rate of

productivity offset the decreasing trend in area. 1960s witnessed increasing rate of
growth of area, output and productivity. 1970s witnessed negative trend in all the
three. 1980s showed a slight increase in productivity coupled with a fall in area and
output. The area under tapioca declined and that of natural rubber has increased. It is
at the expense of tapioca. Natural rubber was considered as the competing crop with
tapioca for area. The acreage under tapioca is determined mainly by its harvest price.
The factor that determines yield variation is found to be rainfall. Through
advancement of technology and diversification market higher level of output of
tapioca could be attained.

Lelithabhai (1993)61 studied sources of production of paddy and

coconut in Kalliassery panchayat using primary data. The sources of production
were identified as: yield, area, cropping pattern, and interaction of area, and cropping
pattern. The result of the study showed that yield is positively correlated with size of
land. Larger size enjoys greater advantage. Density of plants in smallholdings is
higher that results in low yield. Three major constraints identified are:1) Drainage
problems in coastal plains caused by change in cropping pattern, infrastructural
development and spread of human habitation; 2) Increasing problem of salinity as a
result of unrestricted ground water utilization in the coastal plane and siltation of
valley rises causing floods in monsoon and water scarcity in summer. Extension
work in cultivation practices and community intervention are suggested for
improving productivity. The social milieu is such that agriculture is considered to be
a subsidiary occupation. Social institutions have to be developed to create better
utilisation of land and water resources.

On evaluating the economic impact due to technical innovations in paddy

cultivation in Kerala using district wise secondary data Shoba Varghese (1995) 62
identified technical change as TFP. Divisia index has been worked out. TFP and
TFPG index from 1961-1992 has been computed. Taking HYV seeds as embodied
technical change the study estimated 55-63percent growth in TFP from 1961 to 1992
with an average annual growth of 1.79percent. The study found declining trend in

29
agricultural production since 1970s. Technical change in agriculture is due to
fertilizer (embodied/chemical technology). There is continuous neglect of extension
activities. Spread of HYV technology is shrinking – spread effect is low and lab to
land linkage is missing.

Sathees Babu et. al (1996)63 attempted to analyse the trends in area,

production and productivity of pepper and to identify research needs and constraints
in the cultivation of pepper. Time series data for 1956-57 to 1989-90 has been made
use of. Exponential growth function is fitted for getting growth rate. The factors
influencing land productivity were identified by step-wise multiple linear regression.
Independent variables are climatological, technological and economic factors. The
growth rate of area, production and average productivity are tabled below: The
monthly rainfall and monthly variation in average farm harvest prices of pepper on
productivity were examined by a linear regression model. May rain in summer
contributed to more pepper productivity while January and July rains had negative
impact. It was found that average farm harvest prices prevailing during January had
influence on pepper productivity.

Thomas (1996)64 examined trends in area, productivity of paddy in Kerala and

economic causes of decline in paddy cultivation and tried to identify current
problems of paddy for the period 1960-61 to 1991-92 using both secondary time
series data and primary data to study current problems. - 10 villages were collected,
5 in each Kuttanad regions in Alapuzha and Kottayam districts in June 1996. Land
productivity was computed. Growth rate calculations were made to examine
performance of area and productivity. Additive decomposition scheme was used to
measure the change in output.

The study found that none of the sources of productivity had affected
productivity of paddy. HYV seeds, chemical fertilizers, rainfall, plant protection
measures are identified as the determinants that influence productivity. Productivity
in Kerala since mid 70’s is attributed to the sustained decline in area under paddy, a
process in which marginal paddy lands with comparatively lesser productivity had
been going out of cultivation. Growing pressure on land, land price differentials and
rise in paddy fields for non-agricultural purposes have resulted in the area under
paddy.

30
 Job and George (2002)65 assess technical efficiency in rice production of the
State using primary data collected from Kuttanad for 1996. Maximum Likelihood
Estimation is resorted. Farm specific Technical efficiency is worked out as the ratio
of the production function per hectare of the ith farm to the frontier production per
hectare of the same farm. Mean efficiency is found to be 85.01 for puncha, 84 for
Virippu. Extension service and training programmes are recommended to increase
farm level Technical efficiency.

Job and Nandamohan (2004)66 attempts to analyse the changes in the growth
pattern of rice in Kerala across time and across seasons; between the period 1975-76
and 1998-99., and for autumn, winter and summer. Secondary time series data on
area, production and productivity were compiled. Methodology is based on the
computation of compound growth rate, decomposition of growth, and measures of
instability. Compound growth rates of area, production and productivity or rice in the
three crop seasons for the state was estimated using exponential growth model. Rice
production was decomposed into area and yield effect. Area effect and yield effect
ware computed using a multiplicative model. Result of the study reveals that area
under rice and production showed significant negative trend and positive trend in
productivity. The factors responsible for the decline in area are weather, which is a
short run phenomenon and price of rice. Before mid 70s rice was one of the
profitable crops that got reversed on the combined effect of falling price and rising
cost of production. As rice production is labour intensive, labour cost is the most
important item of cost and more than 90 percent of labour used in rice cultivation is
hired labour. Apart from the above, rising land value for housing and brick making
industry and increasing urbanisation have contributed for the contraction of rice
producing area. Rising labour cost coupled with the unmanageability and
unavailability of labour had also prompted the relatively resourceful farmers to
convert rice lands to be used for other relatively labour non-intensive crops.

Subash et. al (2004)67 measured farm specific technical efficiency (TE)

among rice growers in Kumarkam village of Kuttanad area in Kottayam for the
period 2002. Using primary data technical efficiency was computed. Econometric
method was used– Translog stochastic frontier production function estimation using
the computer package Frontier 4.0. Result of the study showed that TE varied

31
between 60 to 98 percent. Mean efficiency being 92percent. The factors that
determine TE are: age, which means experience counts, education. The variables that
determine TE regressed are: raw materials, family and hired labour, fertilizer, usage
per acre tractor use, irrigation, electricity charges, extension services, sources of
credit, farm size in acres, quantities of farm output, infrastructure and data relating to
schooling, age, size of house hold.

2.7. Concluding Observations

We have reviewed important studies relating to agricultural productivity

conducted in international, national and regional perspective. Probably because of
the easiness of computations, majority of the studies seem to have measured land
productivity, which is only a partial measure that does not reveal the relationship
among variables in agricultural production. Of course, some of the recent macro
level studies conducted by using aggregate data at all India level focus on total factor
productivity. Even micro level studies in different regions of India attempt to
measure total factor productivity. In Kerala the scene is quite different. The review
of studies spanning for over 64 years from 1962 to 2004 in Kerala convincingly
show the fact that not a single study attempted to estimate the Total Factor
Productivity Growth of the agricultural sector in Kerala but relied on conventional
measure of partial productivity, particularly land productivity. Many studies have
shown that the only way out for Kerala agriculture from the stagnated state to growth
path is through technological improvement [Pillai, 1982; Kannan & Pushpangadan,
1988]. Since TFP is a measure of technical change, the primary concern in this
context is to identify factors contributing to TFP changes.

Only a few studies have taken into account the variety of crops covered in
this study. Similarly analysis based on a longer period, extending 45 years from
1960-61 to 2004-05 as we proposed to cover in this study is not seen in other studies.
Further, the methodology of analysis, particularly the cointegration approach, which
is the contemporary technique used for time series data analysis and the cost function
approach undertaken for estimating TFP and its decomposition into technical change
and returns to scale were not undertaken by any studies so far. The present study is
an endeavour in this direction

32
Notes and References

33
1 , 2
Hayami, Y. (1969): Sources of agricultural productivity gap among selected countries. American
Journal of Agricultural Economics, 51: 564-575.
Hayami, Y. (1970): On the use of the Cobb-Douglas production function on the cross-section analysis of
agricultural production. American Journal of Agricultural Economics, 52: 327-329.
3
Hayami, Y. & and Ruttan, V.W. (1970): Agricultural productivity differences among countries.
American Economic Review, 60: 895-911.
4
Nguyen, D. (1979): On agricultural productivity differences among countries. American Journal of
Agricultural Economics, 61: 565-70.
5
Kawagoe, T. & Hayami, Y. 1983. The production structure of world agriculture: an inter-country cross-
section analysis. The Developing Economies, 21: 189-206.
6
Kawagoe, T., Hayami, Y. & Ruttan, V.W. 1985. The inter-country agricultural production function and
productivity differences among countries. Journal of Development Economics, 19: 113-132.
7
Evenson, R.E. & Kislev, Y. 1975. Agricultural research and productivity. New Haven: Yale University
Press.
8
Antle. J.M (1983): “Infrastructure and Aggregate Agricultural Productivity”, Economic Development
and Cultural Change, Vol. 31, No. 3, April, pp. 609 – 619.
9
Nehru, V. & Dhareshwar, A. (1994): New estimates of total factor productivity growth for developing
and industrial countries. World Bank Policy Research Working Paper #1313, June. Washington, DC,
World Bank.
0
1
Chavas, J.P.(2001), “An International Analysis of Agricultural Productivity”, in L. Zepeda, (ed.),
Agricultural Investment and Productivity in Developing Countries, FAO, Rome.
1
1
Shephard, R.W. (1970): Theory of cost and production functions. Princeton: Princeton University
Press.
2
1
Rao, D. S. Prasada and Coelli. T.J. (2004): “Catch up and Convergence in Global Agricultural
Productivity”, Indian Economic Review, Vol. XXXIX, No.1, pp. 123 – 148.
1 3
Kaneda, Hiromitsu (1967): “Sources of Productivity gains in Japanese as compared with the U.S
Experience”, Journal of Farm Economics, Vol. 49, No. 5, December, pp. 1443 - 1454.
1 4
Wen.G.J (1993): “Total Factor Productivity change in China's farming sector: 1952 – 1989”, Economic
Development and Cultural Change, Vol. 42, No. 1, October, pp. 1 –41.
1 5
Wadud. Abdul & White Ben (2002)15: “The determinants of Technical efficiency of Farms in
Bangladesh” Indian Economic Review: Vol. XXXVII No.2, pp.183 – 97.
6
1
Mukherjee and Vaidyanathan (1980): “Growth and Fluctuations in Food grain yields per hectare”,
Indian Journal of Agricultural Economics, Vol. XXXV, No. 2, April – June, pp. 60 – 70.
7
1
Bhalla, G.S & Alagh Y.K. (1983): “Labour Productivity in Indian Agriculture”, Economic and
Political Weekly, Annual Number, 1983, pp. 825 – 834.
8
1
Bhatia. M. S (1992):“Input use efficiency in Wheat Production in India”, Agricultural Situation in
India, Vol. XLVII, No. 5, pp. 339 – 344.
9
1
Roségrant, Mark.W and Evenson R.E (1992): “Agricultural Productivity and Sources of Growth in
South Asia”, American Journal of Agricultural Economics, Vol. 74, No.3, August, pp. 757 – 761.
0
2
Dholakia H.R and Dholakia H.B (1993)20:”Growth of Total Factor Productivity in Indian Agriculture”,
Indian Economic Review, Vol. XXVIII, No.1, pp. 25 – 40.
1
2
Praduman Kumar & Mark W. Roségrant (1994): “Productivity growth for Rice in India” Economic and
Political Weekly, Dec, 31. pp. A183 – A188.
2
2
Panigrahi, Ramakrishna (1995): Patterns of Agricultural Growth in India: A State wise analysis”,
M.Phil Dissertation submitted to Jawaharlal Nehru University, New Delhi.
3
2
Bhalla. G.S and Singh G. (1997): “Recent development in Indian agriculture: A State level Analysis”,
Economic and Political Weekly, Vol. XXXII, NO.13, 29 March, pp.A2– A18.
4
2
Schafer, Hans-Bernd (1997): “Farm Price and Agricultural Production – an empirical study 120 Indian
districts”, Economic Development and Agricultural Productivity, Amit Bhaduri and Rune Skarstein
(eds.) Edward Elgar, U.K. pp.98 – 117.
5
2
Mittal, Surabhi and Kumar, Praduman (2000): “Literacy, Technology adoption, Factor demand and
Productivity: An Econometric Approach”, Indian Journal of Agricultural Economics, Vol. 55, No. 3,
 July – September, pp. 490 – 499.
6
2
Mittal, Surabhi & Lal, R.C (2001): “Productivity and Sources of Growth for wheat in India”,
Agricultural Economic Research, Vol. 14, No. 2, July – December, pp. 109 – 120.
7
2
Shyjan. D (2003)27: “Public Investment and Agriculture Productivity: The case of food grains in India”,
M.Phil. dissertation submitted to Jawaharlal Nehru University, New Delhi.
8
2
Kumar, Praduman, Anjani and Mittal, Surabi (2004): “TFP of crop sector in the Indo-Gangetic Plain of
India: Sustainability issues revisited”, Indian Economic Review, Vol. XXXIX, No.1, pp. 16 – 201.
9
2
Chenna Reddy, Venkareddy (1967): “Production efficiency in South Indian Agriculture”, Journal of
Farm Economics, Vol. 49, No. 4, November, pp. 816 – 820.
0
3
Iqbal Ali, Mohammad and Azeemuddin Khan (1993): “A Multivariate Analysis of Productivity and
Profit in Paddy Crop – A study in Tube Well Areas”, Indian Journal of Regional Science, Vol. XXV,
No.2, pp. 13 – 19.
1
3
Kumar, Surinder, Kapur.R.C, Mirchandani,R (1984): “Levels of Productivity of Wheat Crop in
different Agro-climatic zones in Himachal Pradesh”, Agricultural Situation in India, Vol. XXXVIII, No.
12, March, pp. 817 – 819.
2
3
Juyal. R.K (1987)32: “Contribution of health to Productivity in a Developing Area: An Economic
Analysis” Margin April-June, pp. 80 – 85.
3
3
Sachindanda Sau (1990): “Productivity Differentials in Rice and Food grains: A Regional Study”,
Indian Journal of Regional Science, Vol. XXII, No.2, pp.55 – 69.
4
3
Renuka Pillai (2001): “An Analysis of productivity growth in West Bengal and Orissa”, Indian Journal
of Agricultural Economics, Vol. 56, No.4, Oct – Dec. pp. 613 – 630.
5
3
Ahmed.A.U, Bhowmick.B.C (1991)35: “Behaviour of Prices, Productivity, and Acreage Response of
Some Important Crops in Assam”, Agricultural Situation in India, Vol.XLVI, No.4, July, pp. 225 – 229.
6
3
Sarma. A.K (1999):”Impact of Farm Mechanisation through Power Tiller on Productivity and
Employment in Jorhat District of Assam”, Agricultural Situation in India, Vol. LV, No. 12, pp. 727 –
734.
7
3
Borbora, Soundarya and Mahanta, Ratul (2001)37: “District wise study of agricultural production in
Assam using Cobb-Douglas Production Function”, Indian Journal of Regional Science, Vol. XXXIII,
No. 1, pp. 33 – 40.
8
3
Borbora, Soundarya and Mahanta, Ratul (2002): “An Analysis of inter-district disparity in agricultural
development in Assam”, Indian Journal of Regional Science, Vol. XXXIV, No. 2, pp. 28 – 36.
9
3
Patil.V.K. and Kalyankar. S.P (1992): “Impact of research on the Productivity of Agriculture in
Maharastra”, Agricultural Situation in India, Vol. XLVII, No.5, August, pp. 321 – 328.
0
4
Singh, Karam and Kaur, Kulwider (1992)40: “Growth in Agricultural Productivity and Nature of
Technological change in Punjab Agriculture”, Agricultural Situation in India, Vol. XLVII, No.5,
August, pp.355 – 358.
1
4
Sidhu.D.S and Byerkee, Derek (2001): “Technical Change and Wheat Productivity in Post-Green
Revolution in Punjab”, Dynamics of Agricultural Development, Dhinds.K.S. and Anju Sharma (eds.),
Vol. 2. Concept Publishing Company.
2
4
Adams, John and Bumb, balu (1993)42 : “Determinants of Agricultural Productivity in Rajastan, India:
The impact of Inputs, Technology, and Context on Land Productivity”, Economic Development and
Cultural Change, Vol. 42, No. 1, October, pp. 705 – 722.
3
4
Singh. K. M, Chaudhari. J. N and Singh R.K.P (1993)43: “An analysis of compound Growth Rates and
Factors Affecting Area, Production and Productivity of Gram in Bihar” , Agricultural Situation in India,
Vol. XLVII, No. 11, February, pp. 841 – 846.
4
4
Nanavati, Arti and Patel, B.C (1994): “Inter-district variation in Agricultural productivity in Gujarat”,
Productivity, Vol. 35, No.2, July-Sept, pp.
5
4
Shiyani, R.L and Pandya, Maurvi N (2002): “Total Factor Productivity Changes of Oilseed Crops in
Gujarat” Agricultural Situation in India, No.12, March, pp. 791 – 798.
6
4
Singh, Jai and Nandal, D.S (1995)46: “Land degradation and Agricultural Productivity in Haryana”,
Productivity, Vol. 35, No.4, pp. 716 – 720.
 7
4
Jain. S. K. and Idnani.NL (1996) :”Resource Productivity and Sustainability of Rainfed Paddy in
Madhya Pradesh”, Agricultural Situation in India, Vol. LIII, No.8, November, pp.571 – 574.
8
4
Kawadia, Ganesh (2000): “Productivity growth in agriculture economy of Madhya Pradesh”, The
Asian Economic Review, Vol. 42, No. 1, April, pp. 137 – 142.
9
4
Mythili, G and Shanmugam, K.R (2000): “Technical Efficiency of rice growers in Tamil Nadu: A study
based on panel data”, Indian Journal of Agricultural Economics, Vol. 55, No. 1, January – March, pp. 15
– 25.
0
5
Govindarajan.K, Karunakaran.K.R, Shanmugam.T.R, Swaminathan.L.P. (2004) 50: “Rice Productivity
Variation In Cauvery Delta Region of Tamil Nadu”, Asian-African Journal of Economics and
Econometrics, Vol.4, N0.2, pp.85 – 96.
1
5
Oommen. M. A. (1962): “Agricultural Productivity Trends in Kerala”, Agricultural Situations in India,
Vol. XVII, and No. 4, July 1968, pp. 333 – 336.
2
5
Pillai (1969): “Towards a Planning Yardstick for Irrigation of rice in Kerala”, Agricultural Situation in
India, June, pp. 207 –208.
3
5
George. M.V. (1979): “Recent trends in Production and Productivity in Kerala Agriculture”, in Kerala
Economy since Independence, Oommen. M.A (ed.) Oxford & IBH publishing company, New Delhi. pp.
36 – 47.
4
5
Jeemol Unni (1981) : “An Analysis of Change in Cropping pattern in Kerala with Special Reference to
the Substitution of Coconut for Rice: 1960-61 to 1978-79”. M. Phil Dissertation submitted to Jawaharlal
Nehru University, New Delhi.
5
5
Pillai. P. P. (1982): “Growth of Agricultural Output in Kerala during 1952-53 to 1978-79” (pp. 18 -
50) in Agricultural Development in Kerala Pillai P.P. (ed.) Agricole Publishing Academy, New Delhi
6
5
Sivanandan. P. K (1985): “Kerala's Agricultural Performance: Differential Trends and Determinants
of Growth” M. Phil dissertation submitted to Jawaharlal Nehru University, New Delhi.
7
5
George P.S. and Mukherjee, Chandan (1986)57, “A Disaggregate Analysis of the Growth Performance
of Rice in Kerala”, Indian Journal of Agricultural Economics, Vol.41, No.1 Jan – March. Pp. 1 – 16.
8
5
Kannan. K.P and Pushpangadhan. K (1988): “Agricultural Stagnation in Kerala: An Exploratory
Analysis”, Economic and Political Weekly, Vol. XXIII, No. 39, September 24, pp. A-120 – A-128.
Kannan. K.P and Pushpangadhan. K (1999): “Agricultural Development in a Regional Perspective: A
Study of Kerala”, Centre for Development Studies, Thiruvananthapuram.
9
5
Pushpangadhan. K (1988): “Agricultural Stagnation in Kerala: An econometric study of tapioca”,
Working Paper No.226, Centre for Development Studies, Thiruvananthapuram.
0
6
Thomas. E.K, Jessy. T. K, and Devi, Indira (1991): “Growth and output response of tapioca in Kerala”,
Agricultural Situation in India, Vol. XLVI, No.4, July, pp. 215 – 218.
1
6
Lelithabhai. K. N (1993): “Constraints to Productivity Growth in Kerala Agriculture: A Micro level
study”, M. Phil, dissertation submitted to Jawaharlal Nehru University, New Delhi.
2
6
Shoba Varghese (1995)62: “Agricultural Technology: A study of Innovation and Diffusion in Kerala”,
M. Phil dissertation submitted to Jawaharlal Nehru University, New Delhi.
3
6
Sathees Babu.K, Balakrishnan Asan R, Mohanakumaran N, Bhaskaran C and Mohammad Kunju U
(1996): “Trends in Area, Production and Productivity of Pepper in Kerala: An Analysis”, Agricultural
Situation is India, Vol. LIII, No.8, pp. 555 – 558.
4
6
Thomas, P.M (1996): “Decline of Paddy Cultivation in Kerala: A Study of Economic Causes”
Unpublished Ph. D thesis submitted to the Calicut University.
5
6
Job, Elsamma & George, M.V (2002): “Technical Efficiency in Rice Production – A Frontier
Production Function Approach”, Agricultural Economic Research Review, Vol. 15, No.1, pp. 50 –55.
6
6
Job, Elsamma and V. Nandamohan (2004): “Rice production in Kerala – Trend and Instability
Analysis”, Agricultural Situation in India, No.3, June, pp. 135 – 139.
7
6
Subash S, Bino Paul and Ramanathan. A. (2004): “Technical Efficiency in rice Production: An
Application of Stochastic Production Frontier Analysis”, Indian Journal of Regional Science, Vol.
XXXVI, No.1, pp. 118 – 126.