Author Name
S.K. Srivastava, Prof. Ramesh Chand, Jaspal Singh
Choose Report Type
Publication Date
Report Upload
Download
(617.69 KB)
Publisher Name
Agricultural Economics Research Review
PDF Text
Agricultural Economics Research Review
Vol. 30 (Conference Number) 2017 pp 171-182
DOI: 10.5958/0974-0279.2017.00032.5
Changing Crop Production Cost in India: Input Prices,
Substitution and Technological Effects
S.K. Srivastava*, Ramesh Chand and Jaspal Singh
NITI Aayog, New Delhi-110 001
Abstract
The study has examined economics of crop cultivation at the aggregate level over the past 25 years,
identified sources of cost escalation and evaluated the effects of factor prices, substitution and technological
effects on the production cost. The results reveal that a disproportionate change in gross return vis-à-vis
cost resulted in varying rate of return from crop enterprise during the past 25 years. During 2007-08 to
2014-15, the average cost inflation reached the highest level of 13 per cent, more than half of which was
contributed by the rising labour cost alone. Further, at the aggregate level, use of physical inputs increased
only marginally and a large share of the increase in the cost of cultivation was attributed to the rising
prices of inputs. The estimated negative and inelastic demand of inputs revealed a great scope to reduce
the cost by keeping a check on input prices, particularly labour wages. The estimated elasticity of
substitution indicated imperfect substitution between labour and machine and the present level of farm
mechanization is inadequate to offset the wage-push cost inflation in Indian agriculture. It is therefore
necessary to accelerate appropriate farm mechanization through the development of farm machinery
suitable and economical at small farms and improvement in its access through the custom hiring. The
study has also revealed a slow rate of yield improvement to offset the rising cost.
Key words: Production cost, cost inflation, input price effects, factor substitution, technological effects
JEL Classification: Q12, Q14, Q16
Introduction
The agriculture sector, which engages 64 per cent
of the rural workforce, assumes a predominant role in
improving the overall welfare of rural society.
According to the latest available data in Situation
Assessment Survey of Agricultural Households
conducted by the National Sample Survey Office (NSS-
SAS), nearly half of the farmers’ income comes from
crop cultivation. The economic viability of crop
production sector, therefore, becomes an essential
condition to sustain interests of the farming community.
In this context, accurate information on the cost of
cultivation (COC) is indispensable. It not only helps
the farmers to decide on the allocation of limited
resources among alternate crop choices but also enables
an assessment of farm profitability, which in turn
influences their decision to invest in agriculture.
During the past five decades, Indian agriculture
has witnessed a significant change in input-use away
from traditional inputs like human labour, bullock
labour, farm-grown seeds, manure and traditional
methods of irrigation towards modern inputs like
improved seeds, chemical fertilizers, farm machine and
large-scale use of tubewells for irrigation. It is pertinent
to evaluate the effect of such transitions on crop
production cost and profitability of crop enterprise. It
is also important to ascertain whether the change in
COC, if any, is due to the changes in level of input-use
or its prices. The changing relative price of the factors
of production prompts farmers to partially substitute
*Author for correspondence
Email: shivendraiari@gmail.com 172 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
the related factors (e.g., farm labour with machinery)
in order to maximize their profits. The evaluation of
effect of factor substitution on crop cultivation cost is
useful in devising suitable strategies for controlling the
cost inflation in the country.
Most of the studies in the present literature have
used cost concepts as a supplementary tool to estimate
farm profitability or to assess the economic viability
of a technology or to evaluate the impact of policy
reforms (e.g., subsidy, MSP) on production cost. But
barring few (Sen and Bhatia, 2004; Raghavan, 2008),
no study in recent years has focussed exclusively on
the changing structure of COC in a comprehensive
manner. A properly-designed study on economics of
crop production assumes a significant importance,
particularly in the recent years when Indian agriculture
has witnessed a positive turnaround in its performance
since the year 2004-05 (Chand, 2014). With this
background, the present study examines the changes
in average real COC and relative profitability at the
aggregate level during the past 25 years and identifies
the sources of cost inflation and contribution of
different factors in rising COC. The paper also
evaluates the effects of factor prices, factor substitution
and technological improvements on production cost
by estimating price elasticity of input use, elasticity of
factor substitution and yield elasticity of cost in selected
crops, respectively.
Data and Methodology
The study is based on the state-level aggregate and
unit-level data on cost of cultivation collected under
the Comprehensive Scheme on Cost of Cultivation of
Principal Crops of Directorate of Economics and
Statistics, Ministry of Agriculture & Farmers Welfare,
Government of India. Presently, the COC data are
collected for 21 principal crops across major producing
states in the country. However, consistent time series
data over a long time period is available only for a few
crops. The present study uses time series data on COC
for ten crops across 19 major producing states for the
period 1990-91 to 2014-15. The selected crops are
paddy, wheat, maize and jowar from cereals group,
gram and arhar from pulses group, rapeseed & mustard
and groundnut from oilseeds group and sugarcane and
cotton among the other cash crops. These crops covered
66.11 per cent of gross cropped area (GCA) in the
country in the year 2014-15. The state-wise area
covered under the selected crops is given in Appendix
1. For evaluating the effects of factor prices and factor
substitution, plot- level data were used for the period
2000-01 to 2012-13 for the selected crops. The
technological effects on production cost were evaluated
using state-level panel data for the period 2000-01 to
2012-13 in the selected crops.
The trends in average cost and return from crop
cultivation were examined by constructing all-India
level aggregated time series of selected crops across
major producing states using crop area in respective
states as weight. The concept of Cost A
1+ imputed value
of family labour (Cost A
1+FL)
1
was used to represent
the cost. The cost and return were expressed in real
terms using Consumer Price Index for Agricultural
Labour (CPI_AL). The relative profitability of a crop
enterprise was examined from the ratio of CostA
1+FL
and value of gross output during the period 1990-91 to
2014-15. Based on the structural change in cost-output
ratio, the crop performance was examined during three
distinct sub-periods, viz. 1990-91 to 2002-03, 2002-
03 to 2007-08, and 2007-08 to 2014-15.
To estimate annual cost inflation and identify
sources of change in COC over time, cost index (with
base 2004-05=100) was constructed. Relative
contribution of different factors in cost inflation was
estimated using the following formula;
… (1)
where,
Z
it= Contribution of i
th
factor in cost inflation in the t
th
year
w
it= Share of i
th
factor in cost (A
1+FL) in the t
th
year
I
i= Inflation rate of i
th
factor cost in the t
th
year over
previous year, and
i = 1, 2,…, n inputs.
1
Cost A
1 comprises of all paid out cost components such as value of hired human labour, hired bullock labour, maintenance and
upkeep charges on owned bullock labour, upkeep charges of owned machines, hired machine charges, seed cost, pesticides cost,
manure cost, fertilizer cost, canal irrigation charges, depreciation of implements and farm buildings, land revenue cess and
other taxes, interest on working capital and miscellaneous expenses on other inputs. Imputed value of family labour was
estimated by multiplying working hours of family labour with prevailing wage rate. Srivastava et al. : Changing Crop Production Cost in India173
The effect of factor prices and factor substitution
on COC was evaluated by estimating the price elasticity
of factor demand and elasticity of technical substitution
between factors (labour and machine) in the selected
crops. The price elasticity of factor demand simulates
the response of input used to the changes in its prices,
while elasticity of technical substitution explains how
changes in relative prices of factors affect the share of
factors and income distribution. These elasticities were
estimated by fitting the transcendental logarithmic
(translog) cost function in selected crops for the period
2000-01 to 2012-2013. The translog functional form
captures many of the attributes of a cost function that
are implied by the economic theory. Because of this
flexibility, it has been widely used for studying
production relationships. Before logarithms are taken,
the function is:
.
… (2)
where, w is a vector of prices for the inputs to
production and y is a single output. N is the total
number of inputs and a’s are the parameters of the
function.
One disadvantage of the previous form is that it is
not linear in parameters. A standard technique when
dealing with power functions like the translog cost
function, is to take logarithms. The resulting function
is linear in parameters and standard statistical
techniques can be used for estimation. After taking
logarithms, the function is:
…(3)
While it is possible to include terms to account fortechnological progress, the specification used hereassumes that cost is independent of time. UsingShepherd’s lemma, the derived demand equations are
… (4)
where, is the cost share of the i
th
input. The
cost function is assumed to be continuous, so Young’s
Theorem concerning symmetry of the second
derivatives restricts
a
ij = a
ji for all i ≠ j .
The result of this derivation is a system of N C 1
equations consisting of N derived demand equations
and one cost function.
Homogeneity of the first degree implies
for all i and j .It is also possible to impose constant returns to
scale – equivalent to imposing homogeneity in y – and
details of this procedure can be found in Diewert and
Wales (1987). The global concavity can also be
imposed on this specification by forcing the matrix [a
ij]
to be negative semi definite. A technique for
accomplishing this can be found in Jorgenson (1986).
The elasticities of substitution are given by
… (5)
The price elasticities (own and cross) are given by
… (6)
In the empirical analysis, five production factors
— labour, machine, seed, fertilizer and irrigation were
taken into consideration. The model consisted of four
share equations each for the factors, namely labour,
machine, seed and fertilizer. The coefficient of
‘irrigation’ was estimated using homogeneity constraint
in the model.
While analysing impact of technological
improvement on production cost, it was assumed that
technological effects get manifested in crop yield.
Therefore, the impact of technological improvement
on production cost can be evaluated by estimating the 174 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
association between crop yield and production cost.
In the present study, yield elasticity of production cost
was estimated for the selected crops by fitting log-linear
state-level panel cost functions for the period 2000-01
to 2012-13. The general form of the cost function is
given by Equation (7).
Production cost = f(crop yield, seed prices, fertilizer
prices, labour wages, machine use
prices, irrigation prices, animal use
prices, trend)
… (7)
The appropriate models among fixed effects,
random effects and pooled data regression were
selected by following standard panel data modelling
process (Gujarati, 2005).
Results and Discussion
Trends in Cost and Returns
The trends in average real COC and return from
the selected crops during the past 25 years are depicted
in Figure 1. The average real COC witnessed a steady
rise with annual growth rate of 2.14 per cent over the
past 25 years. The rising COC is expected as it implies
growth in input use through higher investments in crop
cultivation. What matters from producers’ point of view
is whether increase in cost is accompanied by at least
a similar increase in the returns?
The ratio of cost to gross return revealed a
disproportionate change in the gross return as compared
to the cost during 1990-91 to 2014-15. Based on the
trend in the ratio, three distinct phases were delineated.
An increase in cost per 100 rupee of output during
1990-91 to 2002-03; a phase of sharp decline in the
production cost after 2002-03 till 2007-08, followed
by a phase of steep increase in the production cost
during 2007-08 to 2014-15.
During 1990-91 to 2002-03, the real COC
representing all the selected crops increased by 2.06
per cent per annum, whereas the real gross returns
remained stagnant. As a result, cost incurred to produce
100 rupees of crop output increased from ` 51 in 1990-
91 to ` 66 in 2002-03 and the net return declined at
the rate of 2.77 per cent per year (Figure 1). The
subsequent period till the year 2007-08 witnessed
revival in the real output, which witnessed a
substantially higher growth rate of 6.56 per cent against
a modest increase in real COC. This reduced the cost
of producing100 rupees of output to historically lowest
level of ¹ 48 by the year 2007-08. The crop profitability
witnessed a substantial improvement during this period.
However, the impressive growth in the real crop
output could not sustain after 2007-08. The value of
crop output deflated by CPI_AL during the year 2014-
15 dropped to the 2006-07 level. On the other hand,
the real COC increased rapidly by 3.22 per cent a year.
These changes led to the reversal in the declining cost
of production from ` 48 /100 rupee output in year 2007-
08 to ` 64 by the year 2014-15. Based on these results,
it can be concluded that during recent years, the growth
Figure 1. Trends in average cost and return from the crop cultivation in India Srivastava et al. : Changing Crop Production Cost in India175
Table 1. Cost of production in selected crops across the major producing states in 2014-15
(`/quintal)
State Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Punjab 515 562 934 - - - - - 2473 -
Uttarakhand 644 934 - - - - - - - 79
Haryana 911 842 - -1962 - - 1686 4156 -
Jharkhand 878 1505 987 - 1299 - - - - -
Rajasthan - 1029 1567 2283 2636- 3033 1923 2948 -
Kerala 1223 - - - - - - - - -
Madhya Pradesh 1151 801 1083 29171943 2968 - 1276 4624 -
Bihar 875 1036952 - - - - 1356 - -
Gujarat - 993 - - - 3578 3195 1423 2827 -
Karnataka 915 2085 1040 1933 1947 - 3639 - 3059 91
Chhattisgarh 938 - - - 2176 - - - - -
Andhra Pradesh 892 - 745 1431- - 3424 - 3311 145
Uttar Pradesh 1089 1220 1609 -4166 2772 - 2512 - 100
Tamil Nadu 1123 - - 2338 - - 2917 - 2974 134
Himachal Pradesh - 1594 1713 - - - - - - -
Maharashtra 1527 1811 2376 -4189 5014 - 3585 146
Odisha 1175 - 1061 - - 4336 - - 5228 -
West Bengal 1234 1311 - - - - - - - -
Assam 1139 - - - - - - 3339 - -
Overall (`/quintal) 1016 1011 1296 2279 2283 3703 3379 1933 3356 114
Output-cost ratio 1.40 1.74 1.23 1.28 1.70 1.51 1.32 1.82 1.22 2.29
in output of the major field crops has remained
inadequate to offset the rising COC leading to a
downward trend in the average net returns from the
crop cultivation. In real terms, the net returns received
by the farmers in 2014-15 were even less than the
returns which they received ten years back in 2005-
06. The effects of declining returns from the investment
in crop enterprises are reflected in the rising resentment
among the farmers across the country during the recent
years (Narayanamoorthy, 2013). As rising COC is not
translating into the improvement in crop output,
strategy to raise farmers’ income should include both
output acceleration and cost reduction measures.
The results presented in Table 1 show that
production cost varies substantially across the crops
and the producing states. For instance, the cost of
producing a quintal of paddy varied from ` 515 in
Punjab to ` 1234 in west Bengal in 2014-15. Similarly,
the cost of producing wheat in Karnataka was 3.7-times
the production cost in Punjab. The large variation in
the production cost of a crop across the states arises
due to difference in production technology (resulting
in differential COC), access to irrigation, and the level
of productivity. Therefore, in the states with low level
of productivity, the production cost can be reduced
substantially by improving crop yield.
Sources of Changes in Cost of Cultivation
The sources of changes in COC have been
identified by estimating the contribution of different
inputs in the average cost inflation during the three
sub-periods of the past 25 years. This in turn depends
on respective share of inputs in COC (weight) and
extent of rise in the COC during the period under
consideration. The composition of the average COC
during the three sub-periods is presented in Table 2.
The evidences showed that during the past 25 years,
Indian agriculture witnessed a steady shift from animal
labour towards machine-use. The share of human
labour in CostA
1+FL witnessed a fluctuating trend 176 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
during the successive periods and attained the highest
level of 47 per cent by TE 2014-15. The labour was
followed by machine, fertilizer, seed, animal labour
and insecticide with their respective shares of 14 per
cent, 11 per cent, 8 per cent, 5 per cent and 2 per cent.
During the past 25 years, the average annual
inflation in CostA
1+FL (2004-05=100) was about 10
per cent per annum (Table 3).The rise in COC was not
uniform during the period under consideration. The
average annual cost inflation declined from 10 per cent
during 1990-91 to 2002-03 to 6 per cent during 2002-
03 to 2007-08. But, the post 2007-08 period witnessed
a sharp increase in COC at the annual rate of 13 per
cent. The decomposition of cost inflation among
various factors revealed that labour alone contributed
53 per cent to the increase in COC during 2007-08 to
2014-15. The labour cost was followed by cost on
machine, fertilizer, seed, insecticides, and animal labour
with their respective contribution of 16 per cent, 9 per
cent, 7 per cent, 2 per cent and 2 per cent. Thus, the
evidences revealed that labour cost is the predominant
source of cost inflation, particularly in the recent years
and managing this factor of production alone can
substantially reduce the COC.
Effect of Input Prices on Cost of Cultivation
The effects of input prices and input-use on
increase in COC were seen from the trend in cost
expressed at current and at 2004-05 prices
2
. The trend
in CostA
1+FL at the base year prices represents changes
in the physical use of inputs. Figure 2 shows that at the
aggregate level, physical use of inputs has changed
only marginally
3
, whereas COC at current prices
witnessed a sharp increase which turned exponential
Table 2. Changing structure of cost of cultivation: TE 1990-91 to TE 2014-15
(per cent)
Year Share in cost of cultivation Cost
Seed Fertilizer Labour Animal Machine Insecticides Others
#
A
1+FL*
TE 1990-91 10 12 39 14 7 2 16 3737
TE 2002-03 8 11 42 12 10 2 15 9768
TE 2007-08 9 11 41 9 13 2 15 14856
TE 2014-15 8 11 47 5 14 2 13 34232
*`/ha (at current prices);
#
Others include manure, depreciation of implements and farm buildings, land revenue cess and
other taxes, interest on working capital and miscellaneous expenses on other inputs
Table 3. Contribution of factors in average cost inflation in India
(per cent)
Period Contribution in cost inflation Cost
Seed Fertilizer Labour Animal Machine Insecticides Others* inflation
1990-02 7 11 46 10 11 3 11 10
2002-07 12 8 34 8 21 3 20 6
2007-14 7 9 53 2 16 2 5 13
Overall 9 10 46 5 15 3 12 10
*Others include manure, depreciation of implements and farm buildings, land revenue cess and other taxes, interest on
working capital and miscellaneous expenses on other inputs
2
The CostA
1+FL at 2004-05 prices was arrived at by deflating cost of individual inputs with its implicit price deflator and
summing over all input costs.
3
The trend in CostA
1+FL at 2004-05 prices does not capture change in the use of individual inputs which might have witnessed
differential (increase/decrease) trend in its use Srivastava et al. : Changing Crop Production Cost in India177
after mid-2000. These changes imply that a large share
of the increase in cost is attributed to the rising prices
of the inputs.
Further, the input prices also exert an indirect effect
on the cost by influencing the actual use of the input.
Such effects can be predicted from the price elasticity
of inputs used in the crop cultivation (Chand, 1986).
In the present study, the price elasticities of inputs used
in the cultivation of selected crops were estimated by
fitting translog cost functions using the SURE
estimation technique. The estimated price elasticities
for labour, irrigation, seed, fertilizer and machine are
given in Table 4. The average estimated price
elasticities varied across the inputs used and the crops
taken into consideration. The elasticity values were
negative and less than one, which imply that the
increase in prices of the inputs would lead to less than
proportionate decline in their use. Therefore, in a
situation of rising input prices, COC will increase.
Thus, inelastic demand of inputs explains a rapid
increase in COC, especially during the recent years.
At the same time, inelastic demand for inputs reveals a
great scope to reduce the COC by controlling input
prices because reduction in input prices would lead to
a less than proportionate increase in input use which
in turn will result in the cost saving for the farmers.
Interestingly, labour, a predominant source of cost
inflation, exhibited the lowest price elasticity among
all the inputs in most of the crops. This implies that in
a situation of falling wage rates, COC will reduce
substantially. This fact also strengthens the argument
to manage the labour use in agriculture and devise a
strong strategy to offset the cost-push effects of wages,
which have risen sharply during the recent years
(Chand and Srivastava, 2014). On the other hand, cost-
reducing effect for the machine, which exhibited the
highest price elasticity among all the inputs used in all
crops, would be the lowest. Nevertheless, these results
Figure 2. Trend in costA
1+FL at current and 2004-05 prices
Table 4. Prices elasticity of inputs used in crop production
Inputs Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Labour -0.21 -0.27 -0.21 -0.17 -0.14 -0.21 -0.16 -0.25 -0.20 -0.23
Irrigation -0.25 -0.47 -0.10 -0.40 -0.30 -0.29 -0.04 -0.49 -0.38 -0.46
Seed -0.29 -0.26 -0.34 -0.36 -0.21 -0.28 -0.28 -0.18 -0.60 -0.44
Fertilizer -0.46-0.38 -0.38 -0.39 -0.38 -0.34 -0.46 -0.44 -0.42 -0.29
Machine -0.62 -0.53 -0.58 -0.55 -0.55-0.59 -0.64 -0.54 -0.65 -0.69
Note: The estimated parameters of the models are not given due to paucity of space and can be obtained from the authors 178 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
reveal an ample scope to reduce COC by controlling
the input prices.
Effect of Factor Substitution on Cost of Cultivation
Apart from controlling input prices, crop budget
can also be managed to some extent by substituting
the dearer inputs with technically feasible relatively
cheaper inputs. For instance, farmers can substitute
human labour with machine for several farm operations
if relative labour wages (to machine-use prices) rises.
It was observed that average labour-use in crop
cultivation witnessed a 13 per cent reduction during
2000-01 to 2014-15. Farm mechanization has played
a major role in reducing labour use in agriculture
(Reddy et al., 2014). However, inspite of declining
labour-use, its share in CostA
1+FL has increased during
the recent years (Table 2). Therefore, it is pertinent to
evaluate the effect of substitution between labour and
machine use on crop budget. This was examined by
estimating elasticity of technical substitution between
labour and machine (EoS) in cultivation of selected
crops.
The EoS between human labour and machine hour
varied from 0.47 for groundnut to 0.78 for arhar (Table
5). The positive but less than one value of elasticity
implies that labour and machine are inelastic substitutes
to each other. A 10 per cent increase in the ratio of
wages-machine use price would result in 6.4 per cent
(smaller) increase in ratio of machine use –labour use
in crop production. Therefore, in the situation of
increase in labour wages relative to machine-use prices,
labour is not completely substituted by the machine
and the share of labour in cost increases. The empirical
evidences showed increase in labour wages-machine
use prices ratio from 0.82 in 2001-02 to 1.03 in 2012-
13 and this was accompanied by the increase in ratio
of the share of labour and machine in crop outlay from
2.6 to 3.5 during the same period. Thus, inelastic
substitution between labour and machine along with
inelastic demand for labour appropriately explains why
the share of labour in COC is increasing in the recent
years despite reduction in the use of human labour in
farm operations.
The imperfect substitution between human labour
and machine use in agriculture also signifies the fact
that it is technically not feasible to replace all manual
farm operations with machine. Other reasons for the
smaller (inelastic) value of EoS might be the slow
progress in the development of efficient labour-saving
farm machinery as well as its suitability and
accessibility to the predominantly small and marginal
farmers in the country. Based on these results it can be
concluded that at present level of farm mechanization,
substitution between labour and machine is not
sufficient to offset the rising labour cost in Indian
agriculture. Greater efforts, therefore, are warranted
to accelerate the development of suitable farm
machinery and to improve its economic access to the
farmers through the institutional innovations like
custom hiring centres.
Effect of Technological Improvement on
Production Cost
While evaluating impact of technological
improvement on production cost, it was assumed that
technological improvement is manifested in the yield
of the crops. In a log-linear cost function, estimated
coefficient of crop yield represents cost elasticity of
yield which explains per cent change in production cost
due to one per cent change in crop yield. The estimates
of state –level panel cost functions for different crops
are given in Table 6. It is to be noted that cross-section
(state) effects were fixed to account for state-specific
differences in production environment and climatic
conditions. Further, inclusion of ‘time’ variable in the
regression captured the temporal changes in production
cost due to the factors other than those included in the
model.
The estimated coefficient of yield was negative for
all the selected crops indicating an inverse relationship
between yield and cost of production. These results
Table 5. Elasticity of substitution between labour and machine use in different crops
EoS Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Value 0.73 0.60 0.72 0.68 0.54 0.78 0.47 0.61 0.68 0.62 Srivastava et al. : Changing Crop Production Cost in India179
Table 6. Estimated coefficients of log-linear cost function for different cropsVariable Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Dependent variable: Cost of production ( `/qtl) in logarithmic form
Explanatory variables :
Intercept 7.402*** 6.593*** 7.440*** 5.327*** 4.578*** 7.552*** 4.832*** 6.099*** 6.722*** 4.971***
(0.323) (0.395) (0.455) (0.810) (0.495) (0.891) (0.476) (0.528) (0.510) (0.958)
ln(Yield) -0.758*** -0.933*** -0.812*** -0.251*** -0.251*** -0.638*** -0.651*** -0.962*** -0.621*** -0.428***
(0.061) (0.049) (0.054) (0.065) (0.053) (0.131) (0.060) (0.088) (0.053) (0.122)
ln(Seed prices) - 0.146* 0.094** 0.0050.287** -0.040 0.413*** -0.064 0.042* 0.059
- (0.083) (0.041) (0.075) (0.134) (0.137) (0.150) (0.073) (0.024) (0.095)
ln(Fertiliser prices) 0.065 0.310***-0.179 0.159 0.025 -0.321 0.037 0.304* 0.304** -0.202
(0.086) (0.102) (0.165) (0.303) (0.146) (0.216) (0.069) (0.164) (0.143) (0.221)
ln(Labour wages) 0.238*** 0.332***0.453*** 0.326** 0.392*** 0.585** 0.386*** 0.529*** 0.412*** -0.247
(0.055) (0.074) (0.080) (0.139) (0.073) (0.238) (0.080) (0.123) (0.089) (0.249)
ln(Machin-use prices) 0.082** 0.067*0.023 -0.052 0.133** 0.005 0.179*** 0.125* -0.055*** 0.521***
(0.033) (0.037) (0.035) (0.094) (0.055) (0.100) (0.063) (0.073) (0.028) (0.129)
ln(Irrigation machine- use prices) 0.052*** 0.121*** -0.014***0.012 0.021 0.035 -0.004 0.078 0.036 -0.124
(0.017) (0.024) (0.021) (0.046) (0.037) (0.067) (0.028) (0.057) (0.029) (0.192)
ln(Animal-use prices) -0.001 0.0040.120 0.255** 0.066 0.046 0.209*** 0.122** 0.125** -0.002
(0.032) (0.026) (0.039) (0.099) (0.052) (0.114) (0.058) (0.054) (0.055) (0.066)
Trend 0.002 -0.011*** -0.006 -0.014* -0.012* 0.030** 0.004 -0.012 -0.019** 0.027***
(0.003) (0.003) (0.005) (0.008) (0.007) (0.012) (0.006) (0.008) (0.007) (0.006)
Model (Fixed/Random/Pooled) F F F F P F P F R P
Cross-section effect Y Y Y Y N Y N Y Y N
Period effect N N N N N N N N N N
R
2
0.87 0.92 0.93 0.71 0.69 0.73 0.72 0.91 0.727 0.681
Hausman test 14.53** 15.550** - - - - - - 11.390 -
(7) (8)(8)
LR test: Cross section F - 16.880*** 10.506*** 2.912** - 8.556*** - 17.069*** - -
(10,124) (6,76) (4,52) (4, 52) (5,64)
LR test: Cross section: Chai square - 122.870***(10)54.966***(6)13.140**(4) - 32.871***(4) - 66.097***(5) -
Observations (13*11)143 (13*11)143 (13*7)91 (13*5)65 (13*6)78 (13*5)65 (13*5)65 (13*6)78 (13*9)117 (13*6) 78
***, **, * denote significance at 1, 5 and 10 per cent levels, respectively 180 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
show that yield improvement through technological
interventions offers an opportunity to absorb the rising
cost of production of crops. It is worth mentioning that
yield elasticities of production cost were less than one
for all the crops which implies that increase in yield
resulted in less than the proportionate increase in cost
of production during the past decade in the country.
This draws attention to the nature of technological
change in Indian agriculture. A one per cent increase
in yield resulted in 0.25 to 0.96 per cent reduction in
COC of different crops. The yield effect on reducing
cost was quite strong in the case of wheat and rapeseed
& mustard but quite weak in the case of gram and jowar.
Conclusions and Policy Implications
The aggregate cost of production and output of
ten major crops grown in India showed three distinct
patterns during 1990-91 to 2014-15. The period 1990-
91 to 2002-03 witnessed a steady rise in the real cost
of cultivation (COC) accompanied by a relatively
slower increase in the crop output. This mismatch
resulted into a decline in profitability and net returns
in real terms from crop production during this sub-
period. The subsequent period till the year 2007-08
witnessed a significant acceleration in growth of output
of the selected crops and the real cost of production
reached a historically low level. The crop profitability
registered a high growth during this period. However,
this could not sustain and growth in the crop output
remained inadequate to absorb the rising COC after
2007-08 till 2014-15. Over the 25 year period since
1990-91, the aggregate cost of cultivation of the
selected crops increased at a faster rate than the increase
in output during 1990-91 to 2014-15.
The average annual inflation in COC reached the
highest level of 13 per cent during 2007-08 to 2014-
15. More than half of the cost inflation during this
period was contributed by the rising labour cost.
Therefore, managing human labour alone would bring
a substantial reduction in the crop budget of the farmers.
Further, the results revealed that at the aggregate level
physical use of inputs increased at a smaller rate and a
large share of the increase in the COC was attributed
to the rising prices of inputs. The negative and inelastic
demand for farm inputs explains the sharp increase in
the COC due to rising prices of inputs in the recent
years. At the same time, keeping a check on input prices
offers a great scope to reduce cost as it would lead to a
less than proportionate increase in its use and the net
effect will be lower COC.
Apart from input price effect, elasticity of
substitution (EoS) between labour and machine is quite
important in influencing COC. The EoS between labour
and machine use was positive and less than one in all
the crops under study indicating imperfect substitution
between labour and machine. One consequence of this
has been increase in the share of labour in COC in the
recent years, despite declining labour-use for farm
operations. The evidences revealed that the present
level of farm mechanization is inadequate to offset the
wage-push cost inflation in Indian agriculture. It is
therefore necessary to promote efficient and appropriate
farm mechanization suitable for small farm situation.
Institutional innovations like custom hiring centres and
online taxi type model for machinery services to the
farmers seems to hold scope to reduce COC.
Nevertheless, the possibility of a perfect substitution
between labour and machine in Indian agriculture is
remote and therefore efforts should be made to improve
crop productivity to absorb rising COC.
The cost elasticity of yield has indicated an inverse
association between yield and cost of production in all
the crops. However, the elasticity coefficient in most
crops barring wheat and rapeseed & mustard was low
which revealed that the pace of improvement in the
crop yield was less than the increase in COC during
the past decade in the country. Such evidences indicate
the slow pace of technological improvement in Indian
agriculture and underline the need to accelerate efforts
to raise yield at a faster rate to offset the effects of
rising COC and maintain a fair profit margin in crop
cultivation.
Acknowledgements
The paper is partly drawn from the institute project
on Changing Structure of Crop Production Cost and
Technological Effects in India undertaken at ICAR-
National Institute of Agricultural Economics and Policy
Research.
References
Chand, R. (1986) Estimating effects of input - output prices
on input demand in Punjab agriculture-A profit function
approach. Artha Vijnana, 28(2):181-192.
Chand, R. (2014) From Slow Down to Fast Track: Indian
Agriculture since 1995. Working Paper 01/2014. ICAR- Srivastava et al. : Changing Crop Production Cost in India181
National Centre for Agricultural Economics and Policy
Research, New Delhi.
Chand, R. and Srivastava, S.K. (2014) Changes in rural
labour market and its implications for Indian
Agriculture. Economic and Political Weekly, 49(10):
47-54.
Diewert, W.E. and Wales, T.J. (1987) Flexible functional
forms and global curvature conditions. Econometrica,
55:43–68.
Gujarati, D.N. (2005) Basic Econometric (4
th
Edition). Tata
McGraw-Hill Publishing Company Limited, New
Delhi.
Jorgenson, D. (1986) Econometric methods for modeling
producer behavior. In: Handbook of Econometrics, Eds:
Z. Griliches and M.D. Intriligator. Elsevier. Amsterdam.
Narayanamoorthy, A. (2013) Profitability in crops
cultivation in India: Some evidence from cost of
cultivation survey data. Indian Journal of Agricultural
Economics, 68(1): 104-121.
Raghavan, M. (2008) Changing pattern of input use and
cost of cultivation. Economic and Political Weekly,
43(26-27): 123-129.
Reddy, A., Rani, C.R. and Reddy, G.P. (2014) Labour scarcity
and farm mechanization: A cross state comparison.
Indian Journal of Agricultural Economics, 69(3): 347-
358.
Sen, A. and Bhatia, M. S. (2004) Cost of Cultivation and
Farm Income. State of the Indian Farmer: A Millennium
Study, Vol. No. 14, Government of India, Academic
Foundation, New Delhi.
Vol. 30 (Conference Number) 2017 pp 171-182
DOI: 10.5958/0974-0279.2017.00032.5
Changing Crop Production Cost in India: Input Prices,
Substitution and Technological Effects
S.K. Srivastava*, Ramesh Chand and Jaspal Singh
NITI Aayog, New Delhi-110 001
Abstract
The study has examined economics of crop cultivation at the aggregate level over the past 25 years,
identified sources of cost escalation and evaluated the effects of factor prices, substitution and technological
effects on the production cost. The results reveal that a disproportionate change in gross return vis-à-vis
cost resulted in varying rate of return from crop enterprise during the past 25 years. During 2007-08 to
2014-15, the average cost inflation reached the highest level of 13 per cent, more than half of which was
contributed by the rising labour cost alone. Further, at the aggregate level, use of physical inputs increased
only marginally and a large share of the increase in the cost of cultivation was attributed to the rising
prices of inputs. The estimated negative and inelastic demand of inputs revealed a great scope to reduce
the cost by keeping a check on input prices, particularly labour wages. The estimated elasticity of
substitution indicated imperfect substitution between labour and machine and the present level of farm
mechanization is inadequate to offset the wage-push cost inflation in Indian agriculture. It is therefore
necessary to accelerate appropriate farm mechanization through the development of farm machinery
suitable and economical at small farms and improvement in its access through the custom hiring. The
study has also revealed a slow rate of yield improvement to offset the rising cost.
Key words: Production cost, cost inflation, input price effects, factor substitution, technological effects
JEL Classification: Q12, Q14, Q16
Introduction
The agriculture sector, which engages 64 per cent
of the rural workforce, assumes a predominant role in
improving the overall welfare of rural society.
According to the latest available data in Situation
Assessment Survey of Agricultural Households
conducted by the National Sample Survey Office (NSS-
SAS), nearly half of the farmers’ income comes from
crop cultivation. The economic viability of crop
production sector, therefore, becomes an essential
condition to sustain interests of the farming community.
In this context, accurate information on the cost of
cultivation (COC) is indispensable. It not only helps
the farmers to decide on the allocation of limited
resources among alternate crop choices but also enables
an assessment of farm profitability, which in turn
influences their decision to invest in agriculture.
During the past five decades, Indian agriculture
has witnessed a significant change in input-use away
from traditional inputs like human labour, bullock
labour, farm-grown seeds, manure and traditional
methods of irrigation towards modern inputs like
improved seeds, chemical fertilizers, farm machine and
large-scale use of tubewells for irrigation. It is pertinent
to evaluate the effect of such transitions on crop
production cost and profitability of crop enterprise. It
is also important to ascertain whether the change in
COC, if any, is due to the changes in level of input-use
or its prices. The changing relative price of the factors
of production prompts farmers to partially substitute
*Author for correspondence
Email: shivendraiari@gmail.com 172 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
the related factors (e.g., farm labour with machinery)
in order to maximize their profits. The evaluation of
effect of factor substitution on crop cultivation cost is
useful in devising suitable strategies for controlling the
cost inflation in the country.
Most of the studies in the present literature have
used cost concepts as a supplementary tool to estimate
farm profitability or to assess the economic viability
of a technology or to evaluate the impact of policy
reforms (e.g., subsidy, MSP) on production cost. But
barring few (Sen and Bhatia, 2004; Raghavan, 2008),
no study in recent years has focussed exclusively on
the changing structure of COC in a comprehensive
manner. A properly-designed study on economics of
crop production assumes a significant importance,
particularly in the recent years when Indian agriculture
has witnessed a positive turnaround in its performance
since the year 2004-05 (Chand, 2014). With this
background, the present study examines the changes
in average real COC and relative profitability at the
aggregate level during the past 25 years and identifies
the sources of cost inflation and contribution of
different factors in rising COC. The paper also
evaluates the effects of factor prices, factor substitution
and technological improvements on production cost
by estimating price elasticity of input use, elasticity of
factor substitution and yield elasticity of cost in selected
crops, respectively.
Data and Methodology
The study is based on the state-level aggregate and
unit-level data on cost of cultivation collected under
the Comprehensive Scheme on Cost of Cultivation of
Principal Crops of Directorate of Economics and
Statistics, Ministry of Agriculture & Farmers Welfare,
Government of India. Presently, the COC data are
collected for 21 principal crops across major producing
states in the country. However, consistent time series
data over a long time period is available only for a few
crops. The present study uses time series data on COC
for ten crops across 19 major producing states for the
period 1990-91 to 2014-15. The selected crops are
paddy, wheat, maize and jowar from cereals group,
gram and arhar from pulses group, rapeseed & mustard
and groundnut from oilseeds group and sugarcane and
cotton among the other cash crops. These crops covered
66.11 per cent of gross cropped area (GCA) in the
country in the year 2014-15. The state-wise area
covered under the selected crops is given in Appendix
1. For evaluating the effects of factor prices and factor
substitution, plot- level data were used for the period
2000-01 to 2012-13 for the selected crops. The
technological effects on production cost were evaluated
using state-level panel data for the period 2000-01 to
2012-13 in the selected crops.
The trends in average cost and return from crop
cultivation were examined by constructing all-India
level aggregated time series of selected crops across
major producing states using crop area in respective
states as weight. The concept of Cost A
1+ imputed value
of family labour (Cost A
1+FL)
1
was used to represent
the cost. The cost and return were expressed in real
terms using Consumer Price Index for Agricultural
Labour (CPI_AL). The relative profitability of a crop
enterprise was examined from the ratio of CostA
1+FL
and value of gross output during the period 1990-91 to
2014-15. Based on the structural change in cost-output
ratio, the crop performance was examined during three
distinct sub-periods, viz. 1990-91 to 2002-03, 2002-
03 to 2007-08, and 2007-08 to 2014-15.
To estimate annual cost inflation and identify
sources of change in COC over time, cost index (with
base 2004-05=100) was constructed. Relative
contribution of different factors in cost inflation was
estimated using the following formula;
… (1)
where,
Z
it= Contribution of i
th
factor in cost inflation in the t
th
year
w
it= Share of i
th
factor in cost (A
1+FL) in the t
th
year
I
i= Inflation rate of i
th
factor cost in the t
th
year over
previous year, and
i = 1, 2,…, n inputs.
1
Cost A
1 comprises of all paid out cost components such as value of hired human labour, hired bullock labour, maintenance and
upkeep charges on owned bullock labour, upkeep charges of owned machines, hired machine charges, seed cost, pesticides cost,
manure cost, fertilizer cost, canal irrigation charges, depreciation of implements and farm buildings, land revenue cess and
other taxes, interest on working capital and miscellaneous expenses on other inputs. Imputed value of family labour was
estimated by multiplying working hours of family labour with prevailing wage rate. Srivastava et al. : Changing Crop Production Cost in India173
The effect of factor prices and factor substitution
on COC was evaluated by estimating the price elasticity
of factor demand and elasticity of technical substitution
between factors (labour and machine) in the selected
crops. The price elasticity of factor demand simulates
the response of input used to the changes in its prices,
while elasticity of technical substitution explains how
changes in relative prices of factors affect the share of
factors and income distribution. These elasticities were
estimated by fitting the transcendental logarithmic
(translog) cost function in selected crops for the period
2000-01 to 2012-2013. The translog functional form
captures many of the attributes of a cost function that
are implied by the economic theory. Because of this
flexibility, it has been widely used for studying
production relationships. Before logarithms are taken,
the function is:
.
… (2)
where, w is a vector of prices for the inputs to
production and y is a single output. N is the total
number of inputs and a’s are the parameters of the
function.
One disadvantage of the previous form is that it is
not linear in parameters. A standard technique when
dealing with power functions like the translog cost
function, is to take logarithms. The resulting function
is linear in parameters and standard statistical
techniques can be used for estimation. After taking
logarithms, the function is:
…(3)
While it is possible to include terms to account fortechnological progress, the specification used hereassumes that cost is independent of time. UsingShepherd’s lemma, the derived demand equations are
… (4)
where, is the cost share of the i
th
input. The
cost function is assumed to be continuous, so Young’s
Theorem concerning symmetry of the second
derivatives restricts
a
ij = a
ji for all i ≠ j .
The result of this derivation is a system of N C 1
equations consisting of N derived demand equations
and one cost function.
Homogeneity of the first degree implies
for all i and j .It is also possible to impose constant returns to
scale – equivalent to imposing homogeneity in y – and
details of this procedure can be found in Diewert and
Wales (1987). The global concavity can also be
imposed on this specification by forcing the matrix [a
ij]
to be negative semi definite. A technique for
accomplishing this can be found in Jorgenson (1986).
The elasticities of substitution are given by
… (5)
The price elasticities (own and cross) are given by
… (6)
In the empirical analysis, five production factors
— labour, machine, seed, fertilizer and irrigation were
taken into consideration. The model consisted of four
share equations each for the factors, namely labour,
machine, seed and fertilizer. The coefficient of
‘irrigation’ was estimated using homogeneity constraint
in the model.
While analysing impact of technological
improvement on production cost, it was assumed that
technological effects get manifested in crop yield.
Therefore, the impact of technological improvement
on production cost can be evaluated by estimating the 174 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
association between crop yield and production cost.
In the present study, yield elasticity of production cost
was estimated for the selected crops by fitting log-linear
state-level panel cost functions for the period 2000-01
to 2012-13. The general form of the cost function is
given by Equation (7).
Production cost = f(crop yield, seed prices, fertilizer
prices, labour wages, machine use
prices, irrigation prices, animal use
prices, trend)
… (7)
The appropriate models among fixed effects,
random effects and pooled data regression were
selected by following standard panel data modelling
process (Gujarati, 2005).
Results and Discussion
Trends in Cost and Returns
The trends in average real COC and return from
the selected crops during the past 25 years are depicted
in Figure 1. The average real COC witnessed a steady
rise with annual growth rate of 2.14 per cent over the
past 25 years. The rising COC is expected as it implies
growth in input use through higher investments in crop
cultivation. What matters from producers’ point of view
is whether increase in cost is accompanied by at least
a similar increase in the returns?
The ratio of cost to gross return revealed a
disproportionate change in the gross return as compared
to the cost during 1990-91 to 2014-15. Based on the
trend in the ratio, three distinct phases were delineated.
An increase in cost per 100 rupee of output during
1990-91 to 2002-03; a phase of sharp decline in the
production cost after 2002-03 till 2007-08, followed
by a phase of steep increase in the production cost
during 2007-08 to 2014-15.
During 1990-91 to 2002-03, the real COC
representing all the selected crops increased by 2.06
per cent per annum, whereas the real gross returns
remained stagnant. As a result, cost incurred to produce
100 rupees of crop output increased from ` 51 in 1990-
91 to ` 66 in 2002-03 and the net return declined at
the rate of 2.77 per cent per year (Figure 1). The
subsequent period till the year 2007-08 witnessed
revival in the real output, which witnessed a
substantially higher growth rate of 6.56 per cent against
a modest increase in real COC. This reduced the cost
of producing100 rupees of output to historically lowest
level of ¹ 48 by the year 2007-08. The crop profitability
witnessed a substantial improvement during this period.
However, the impressive growth in the real crop
output could not sustain after 2007-08. The value of
crop output deflated by CPI_AL during the year 2014-
15 dropped to the 2006-07 level. On the other hand,
the real COC increased rapidly by 3.22 per cent a year.
These changes led to the reversal in the declining cost
of production from ` 48 /100 rupee output in year 2007-
08 to ` 64 by the year 2014-15. Based on these results,
it can be concluded that during recent years, the growth
Figure 1. Trends in average cost and return from the crop cultivation in India Srivastava et al. : Changing Crop Production Cost in India175
Table 1. Cost of production in selected crops across the major producing states in 2014-15
(`/quintal)
State Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Punjab 515 562 934 - - - - - 2473 -
Uttarakhand 644 934 - - - - - - - 79
Haryana 911 842 - -1962 - - 1686 4156 -
Jharkhand 878 1505 987 - 1299 - - - - -
Rajasthan - 1029 1567 2283 2636- 3033 1923 2948 -
Kerala 1223 - - - - - - - - -
Madhya Pradesh 1151 801 1083 29171943 2968 - 1276 4624 -
Bihar 875 1036952 - - - - 1356 - -
Gujarat - 993 - - - 3578 3195 1423 2827 -
Karnataka 915 2085 1040 1933 1947 - 3639 - 3059 91
Chhattisgarh 938 - - - 2176 - - - - -
Andhra Pradesh 892 - 745 1431- - 3424 - 3311 145
Uttar Pradesh 1089 1220 1609 -4166 2772 - 2512 - 100
Tamil Nadu 1123 - - 2338 - - 2917 - 2974 134
Himachal Pradesh - 1594 1713 - - - - - - -
Maharashtra 1527 1811 2376 -4189 5014 - 3585 146
Odisha 1175 - 1061 - - 4336 - - 5228 -
West Bengal 1234 1311 - - - - - - - -
Assam 1139 - - - - - - 3339 - -
Overall (`/quintal) 1016 1011 1296 2279 2283 3703 3379 1933 3356 114
Output-cost ratio 1.40 1.74 1.23 1.28 1.70 1.51 1.32 1.82 1.22 2.29
in output of the major field crops has remained
inadequate to offset the rising COC leading to a
downward trend in the average net returns from the
crop cultivation. In real terms, the net returns received
by the farmers in 2014-15 were even less than the
returns which they received ten years back in 2005-
06. The effects of declining returns from the investment
in crop enterprises are reflected in the rising resentment
among the farmers across the country during the recent
years (Narayanamoorthy, 2013). As rising COC is not
translating into the improvement in crop output,
strategy to raise farmers’ income should include both
output acceleration and cost reduction measures.
The results presented in Table 1 show that
production cost varies substantially across the crops
and the producing states. For instance, the cost of
producing a quintal of paddy varied from ` 515 in
Punjab to ` 1234 in west Bengal in 2014-15. Similarly,
the cost of producing wheat in Karnataka was 3.7-times
the production cost in Punjab. The large variation in
the production cost of a crop across the states arises
due to difference in production technology (resulting
in differential COC), access to irrigation, and the level
of productivity. Therefore, in the states with low level
of productivity, the production cost can be reduced
substantially by improving crop yield.
Sources of Changes in Cost of Cultivation
The sources of changes in COC have been
identified by estimating the contribution of different
inputs in the average cost inflation during the three
sub-periods of the past 25 years. This in turn depends
on respective share of inputs in COC (weight) and
extent of rise in the COC during the period under
consideration. The composition of the average COC
during the three sub-periods is presented in Table 2.
The evidences showed that during the past 25 years,
Indian agriculture witnessed a steady shift from animal
labour towards machine-use. The share of human
labour in CostA
1+FL witnessed a fluctuating trend 176 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
during the successive periods and attained the highest
level of 47 per cent by TE 2014-15. The labour was
followed by machine, fertilizer, seed, animal labour
and insecticide with their respective shares of 14 per
cent, 11 per cent, 8 per cent, 5 per cent and 2 per cent.
During the past 25 years, the average annual
inflation in CostA
1+FL (2004-05=100) was about 10
per cent per annum (Table 3).The rise in COC was not
uniform during the period under consideration. The
average annual cost inflation declined from 10 per cent
during 1990-91 to 2002-03 to 6 per cent during 2002-
03 to 2007-08. But, the post 2007-08 period witnessed
a sharp increase in COC at the annual rate of 13 per
cent. The decomposition of cost inflation among
various factors revealed that labour alone contributed
53 per cent to the increase in COC during 2007-08 to
2014-15. The labour cost was followed by cost on
machine, fertilizer, seed, insecticides, and animal labour
with their respective contribution of 16 per cent, 9 per
cent, 7 per cent, 2 per cent and 2 per cent. Thus, the
evidences revealed that labour cost is the predominant
source of cost inflation, particularly in the recent years
and managing this factor of production alone can
substantially reduce the COC.
Effect of Input Prices on Cost of Cultivation
The effects of input prices and input-use on
increase in COC were seen from the trend in cost
expressed at current and at 2004-05 prices
2
. The trend
in CostA
1+FL at the base year prices represents changes
in the physical use of inputs. Figure 2 shows that at the
aggregate level, physical use of inputs has changed
only marginally
3
, whereas COC at current prices
witnessed a sharp increase which turned exponential
Table 2. Changing structure of cost of cultivation: TE 1990-91 to TE 2014-15
(per cent)
Year Share in cost of cultivation Cost
Seed Fertilizer Labour Animal Machine Insecticides Others
#
A
1+FL*
TE 1990-91 10 12 39 14 7 2 16 3737
TE 2002-03 8 11 42 12 10 2 15 9768
TE 2007-08 9 11 41 9 13 2 15 14856
TE 2014-15 8 11 47 5 14 2 13 34232
*`/ha (at current prices);
#
Others include manure, depreciation of implements and farm buildings, land revenue cess and
other taxes, interest on working capital and miscellaneous expenses on other inputs
Table 3. Contribution of factors in average cost inflation in India
(per cent)
Period Contribution in cost inflation Cost
Seed Fertilizer Labour Animal Machine Insecticides Others* inflation
1990-02 7 11 46 10 11 3 11 10
2002-07 12 8 34 8 21 3 20 6
2007-14 7 9 53 2 16 2 5 13
Overall 9 10 46 5 15 3 12 10
*Others include manure, depreciation of implements and farm buildings, land revenue cess and other taxes, interest on
working capital and miscellaneous expenses on other inputs
2
The CostA
1+FL at 2004-05 prices was arrived at by deflating cost of individual inputs with its implicit price deflator and
summing over all input costs.
3
The trend in CostA
1+FL at 2004-05 prices does not capture change in the use of individual inputs which might have witnessed
differential (increase/decrease) trend in its use Srivastava et al. : Changing Crop Production Cost in India177
after mid-2000. These changes imply that a large share
of the increase in cost is attributed to the rising prices
of the inputs.
Further, the input prices also exert an indirect effect
on the cost by influencing the actual use of the input.
Such effects can be predicted from the price elasticity
of inputs used in the crop cultivation (Chand, 1986).
In the present study, the price elasticities of inputs used
in the cultivation of selected crops were estimated by
fitting translog cost functions using the SURE
estimation technique. The estimated price elasticities
for labour, irrigation, seed, fertilizer and machine are
given in Table 4. The average estimated price
elasticities varied across the inputs used and the crops
taken into consideration. The elasticity values were
negative and less than one, which imply that the
increase in prices of the inputs would lead to less than
proportionate decline in their use. Therefore, in a
situation of rising input prices, COC will increase.
Thus, inelastic demand of inputs explains a rapid
increase in COC, especially during the recent years.
At the same time, inelastic demand for inputs reveals a
great scope to reduce the COC by controlling input
prices because reduction in input prices would lead to
a less than proportionate increase in input use which
in turn will result in the cost saving for the farmers.
Interestingly, labour, a predominant source of cost
inflation, exhibited the lowest price elasticity among
all the inputs in most of the crops. This implies that in
a situation of falling wage rates, COC will reduce
substantially. This fact also strengthens the argument
to manage the labour use in agriculture and devise a
strong strategy to offset the cost-push effects of wages,
which have risen sharply during the recent years
(Chand and Srivastava, 2014). On the other hand, cost-
reducing effect for the machine, which exhibited the
highest price elasticity among all the inputs used in all
crops, would be the lowest. Nevertheless, these results
Figure 2. Trend in costA
1+FL at current and 2004-05 prices
Table 4. Prices elasticity of inputs used in crop production
Inputs Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Labour -0.21 -0.27 -0.21 -0.17 -0.14 -0.21 -0.16 -0.25 -0.20 -0.23
Irrigation -0.25 -0.47 -0.10 -0.40 -0.30 -0.29 -0.04 -0.49 -0.38 -0.46
Seed -0.29 -0.26 -0.34 -0.36 -0.21 -0.28 -0.28 -0.18 -0.60 -0.44
Fertilizer -0.46-0.38 -0.38 -0.39 -0.38 -0.34 -0.46 -0.44 -0.42 -0.29
Machine -0.62 -0.53 -0.58 -0.55 -0.55-0.59 -0.64 -0.54 -0.65 -0.69
Note: The estimated parameters of the models are not given due to paucity of space and can be obtained from the authors 178 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
reveal an ample scope to reduce COC by controlling
the input prices.
Effect of Factor Substitution on Cost of Cultivation
Apart from controlling input prices, crop budget
can also be managed to some extent by substituting
the dearer inputs with technically feasible relatively
cheaper inputs. For instance, farmers can substitute
human labour with machine for several farm operations
if relative labour wages (to machine-use prices) rises.
It was observed that average labour-use in crop
cultivation witnessed a 13 per cent reduction during
2000-01 to 2014-15. Farm mechanization has played
a major role in reducing labour use in agriculture
(Reddy et al., 2014). However, inspite of declining
labour-use, its share in CostA
1+FL has increased during
the recent years (Table 2). Therefore, it is pertinent to
evaluate the effect of substitution between labour and
machine use on crop budget. This was examined by
estimating elasticity of technical substitution between
labour and machine (EoS) in cultivation of selected
crops.
The EoS between human labour and machine hour
varied from 0.47 for groundnut to 0.78 for arhar (Table
5). The positive but less than one value of elasticity
implies that labour and machine are inelastic substitutes
to each other. A 10 per cent increase in the ratio of
wages-machine use price would result in 6.4 per cent
(smaller) increase in ratio of machine use –labour use
in crop production. Therefore, in the situation of
increase in labour wages relative to machine-use prices,
labour is not completely substituted by the machine
and the share of labour in cost increases. The empirical
evidences showed increase in labour wages-machine
use prices ratio from 0.82 in 2001-02 to 1.03 in 2012-
13 and this was accompanied by the increase in ratio
of the share of labour and machine in crop outlay from
2.6 to 3.5 during the same period. Thus, inelastic
substitution between labour and machine along with
inelastic demand for labour appropriately explains why
the share of labour in COC is increasing in the recent
years despite reduction in the use of human labour in
farm operations.
The imperfect substitution between human labour
and machine use in agriculture also signifies the fact
that it is technically not feasible to replace all manual
farm operations with machine. Other reasons for the
smaller (inelastic) value of EoS might be the slow
progress in the development of efficient labour-saving
farm machinery as well as its suitability and
accessibility to the predominantly small and marginal
farmers in the country. Based on these results it can be
concluded that at present level of farm mechanization,
substitution between labour and machine is not
sufficient to offset the rising labour cost in Indian
agriculture. Greater efforts, therefore, are warranted
to accelerate the development of suitable farm
machinery and to improve its economic access to the
farmers through the institutional innovations like
custom hiring centres.
Effect of Technological Improvement on
Production Cost
While evaluating impact of technological
improvement on production cost, it was assumed that
technological improvement is manifested in the yield
of the crops. In a log-linear cost function, estimated
coefficient of crop yield represents cost elasticity of
yield which explains per cent change in production cost
due to one per cent change in crop yield. The estimates
of state –level panel cost functions for different crops
are given in Table 6. It is to be noted that cross-section
(state) effects were fixed to account for state-specific
differences in production environment and climatic
conditions. Further, inclusion of ‘time’ variable in the
regression captured the temporal changes in production
cost due to the factors other than those included in the
model.
The estimated coefficient of yield was negative for
all the selected crops indicating an inverse relationship
between yield and cost of production. These results
Table 5. Elasticity of substitution between labour and machine use in different crops
EoS Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Value 0.73 0.60 0.72 0.68 0.54 0.78 0.47 0.61 0.68 0.62 Srivastava et al. : Changing Crop Production Cost in India179
Table 6. Estimated coefficients of log-linear cost function for different cropsVariable Paddy Wheat Maize Jowar Gram Arhar Groundnut Rapeseed Cotton Sugarcane
& mustard
Dependent variable: Cost of production ( `/qtl) in logarithmic form
Explanatory variables :
Intercept 7.402*** 6.593*** 7.440*** 5.327*** 4.578*** 7.552*** 4.832*** 6.099*** 6.722*** 4.971***
(0.323) (0.395) (0.455) (0.810) (0.495) (0.891) (0.476) (0.528) (0.510) (0.958)
ln(Yield) -0.758*** -0.933*** -0.812*** -0.251*** -0.251*** -0.638*** -0.651*** -0.962*** -0.621*** -0.428***
(0.061) (0.049) (0.054) (0.065) (0.053) (0.131) (0.060) (0.088) (0.053) (0.122)
ln(Seed prices) - 0.146* 0.094** 0.0050.287** -0.040 0.413*** -0.064 0.042* 0.059
- (0.083) (0.041) (0.075) (0.134) (0.137) (0.150) (0.073) (0.024) (0.095)
ln(Fertiliser prices) 0.065 0.310***-0.179 0.159 0.025 -0.321 0.037 0.304* 0.304** -0.202
(0.086) (0.102) (0.165) (0.303) (0.146) (0.216) (0.069) (0.164) (0.143) (0.221)
ln(Labour wages) 0.238*** 0.332***0.453*** 0.326** 0.392*** 0.585** 0.386*** 0.529*** 0.412*** -0.247
(0.055) (0.074) (0.080) (0.139) (0.073) (0.238) (0.080) (0.123) (0.089) (0.249)
ln(Machin-use prices) 0.082** 0.067*0.023 -0.052 0.133** 0.005 0.179*** 0.125* -0.055*** 0.521***
(0.033) (0.037) (0.035) (0.094) (0.055) (0.100) (0.063) (0.073) (0.028) (0.129)
ln(Irrigation machine- use prices) 0.052*** 0.121*** -0.014***0.012 0.021 0.035 -0.004 0.078 0.036 -0.124
(0.017) (0.024) (0.021) (0.046) (0.037) (0.067) (0.028) (0.057) (0.029) (0.192)
ln(Animal-use prices) -0.001 0.0040.120 0.255** 0.066 0.046 0.209*** 0.122** 0.125** -0.002
(0.032) (0.026) (0.039) (0.099) (0.052) (0.114) (0.058) (0.054) (0.055) (0.066)
Trend 0.002 -0.011*** -0.006 -0.014* -0.012* 0.030** 0.004 -0.012 -0.019** 0.027***
(0.003) (0.003) (0.005) (0.008) (0.007) (0.012) (0.006) (0.008) (0.007) (0.006)
Model (Fixed/Random/Pooled) F F F F P F P F R P
Cross-section effect Y Y Y Y N Y N Y Y N
Period effect N N N N N N N N N N
R
2
0.87 0.92 0.93 0.71 0.69 0.73 0.72 0.91 0.727 0.681
Hausman test 14.53** 15.550** - - - - - - 11.390 -
(7) (8)(8)
LR test: Cross section F - 16.880*** 10.506*** 2.912** - 8.556*** - 17.069*** - -
(10,124) (6,76) (4,52) (4, 52) (5,64)
LR test: Cross section: Chai square - 122.870***(10)54.966***(6)13.140**(4) - 32.871***(4) - 66.097***(5) -
Observations (13*11)143 (13*11)143 (13*7)91 (13*5)65 (13*6)78 (13*5)65 (13*5)65 (13*6)78 (13*9)117 (13*6) 78
***, **, * denote significance at 1, 5 and 10 per cent levels, respectively 180 Agricultural Economics Research Review Vol. 30 (Conference Number) 2017
show that yield improvement through technological
interventions offers an opportunity to absorb the rising
cost of production of crops. It is worth mentioning that
yield elasticities of production cost were less than one
for all the crops which implies that increase in yield
resulted in less than the proportionate increase in cost
of production during the past decade in the country.
This draws attention to the nature of technological
change in Indian agriculture. A one per cent increase
in yield resulted in 0.25 to 0.96 per cent reduction in
COC of different crops. The yield effect on reducing
cost was quite strong in the case of wheat and rapeseed
& mustard but quite weak in the case of gram and jowar.
Conclusions and Policy Implications
The aggregate cost of production and output of
ten major crops grown in India showed three distinct
patterns during 1990-91 to 2014-15. The period 1990-
91 to 2002-03 witnessed a steady rise in the real cost
of cultivation (COC) accompanied by a relatively
slower increase in the crop output. This mismatch
resulted into a decline in profitability and net returns
in real terms from crop production during this sub-
period. The subsequent period till the year 2007-08
witnessed a significant acceleration in growth of output
of the selected crops and the real cost of production
reached a historically low level. The crop profitability
registered a high growth during this period. However,
this could not sustain and growth in the crop output
remained inadequate to absorb the rising COC after
2007-08 till 2014-15. Over the 25 year period since
1990-91, the aggregate cost of cultivation of the
selected crops increased at a faster rate than the increase
in output during 1990-91 to 2014-15.
The average annual inflation in COC reached the
highest level of 13 per cent during 2007-08 to 2014-
15. More than half of the cost inflation during this
period was contributed by the rising labour cost.
Therefore, managing human labour alone would bring
a substantial reduction in the crop budget of the farmers.
Further, the results revealed that at the aggregate level
physical use of inputs increased at a smaller rate and a
large share of the increase in the COC was attributed
to the rising prices of inputs. The negative and inelastic
demand for farm inputs explains the sharp increase in
the COC due to rising prices of inputs in the recent
years. At the same time, keeping a check on input prices
offers a great scope to reduce cost as it would lead to a
less than proportionate increase in its use and the net
effect will be lower COC.
Apart from input price effect, elasticity of
substitution (EoS) between labour and machine is quite
important in influencing COC. The EoS between labour
and machine use was positive and less than one in all
the crops under study indicating imperfect substitution
between labour and machine. One consequence of this
has been increase in the share of labour in COC in the
recent years, despite declining labour-use for farm
operations. The evidences revealed that the present
level of farm mechanization is inadequate to offset the
wage-push cost inflation in Indian agriculture. It is
therefore necessary to promote efficient and appropriate
farm mechanization suitable for small farm situation.
Institutional innovations like custom hiring centres and
online taxi type model for machinery services to the
farmers seems to hold scope to reduce COC.
Nevertheless, the possibility of a perfect substitution
between labour and machine in Indian agriculture is
remote and therefore efforts should be made to improve
crop productivity to absorb rising COC.
The cost elasticity of yield has indicated an inverse
association between yield and cost of production in all
the crops. However, the elasticity coefficient in most
crops barring wheat and rapeseed & mustard was low
which revealed that the pace of improvement in the
crop yield was less than the increase in COC during
the past decade in the country. Such evidences indicate
the slow pace of technological improvement in Indian
agriculture and underline the need to accelerate efforts
to raise yield at a faster rate to offset the effects of
rising COC and maintain a fair profit margin in crop
cultivation.
Acknowledgements
The paper is partly drawn from the institute project
on Changing Structure of Crop Production Cost and
Technological Effects in India undertaken at ICAR-
National Institute of Agricultural Economics and Policy
Research.
References
Chand, R. (1986) Estimating effects of input - output prices
on input demand in Punjab agriculture-A profit function
approach. Artha Vijnana, 28(2):181-192.
Chand, R. (2014) From Slow Down to Fast Track: Indian
Agriculture since 1995. Working Paper 01/2014. ICAR- Srivastava et al. : Changing Crop Production Cost in India181
National Centre for Agricultural Economics and Policy
Research, New Delhi.
Chand, R. and Srivastava, S.K. (2014) Changes in rural
labour market and its implications for Indian
Agriculture. Economic and Political Weekly, 49(10):
47-54.
Diewert, W.E. and Wales, T.J. (1987) Flexible functional
forms and global curvature conditions. Econometrica,
55:43–68.
Gujarati, D.N. (2005) Basic Econometric (4
th
Edition). Tata
McGraw-Hill Publishing Company Limited, New
Delhi.
Jorgenson, D. (1986) Econometric methods for modeling
producer behavior. In: Handbook of Econometrics, Eds:
Z. Griliches and M.D. Intriligator. Elsevier. Amsterdam.
Narayanamoorthy, A. (2013) Profitability in crops
cultivation in India: Some evidence from cost of
cultivation survey data. Indian Journal of Agricultural
Economics, 68(1): 104-121.
Raghavan, M. (2008) Changing pattern of input use and
cost of cultivation. Economic and Political Weekly,
43(26-27): 123-129.
Reddy, A., Rani, C.R. and Reddy, G.P. (2014) Labour scarcity
and farm mechanization: A cross state comparison.
Indian Journal of Agricultural Economics, 69(3): 347-
358.
Sen, A. and Bhatia, M. S. (2004) Cost of Cultivation and
Farm Income. State of the Indian Farmer: A Millennium
Study, Vol. No. 14, Government of India, Academic
Foundation, New Delhi.