The Impact of Fertilizer Subsidy On Paddy Cultivation in Sri Lanka

The Impact of Fertilizer Subsidy on Paddy Cultivation
in Sri Lanka
H. K. J. Ekanayake1
Abstract
The Government of Sri Lanka has been subsidizing fertilizer

for more than four decades. This study attempts to analyse the
impact of fertilizer subsidies with reference to paddy cultivation
in Sri Lanka. The issue is analysed by employing three separate
demand functions for major fertilizers by using simple regression
model. Regression results indicate that changes in the prices of
fertilizer and paddy do not have a significant effect on fertilizer
usage, which points to the fact that the fertilizer subsidy is not a
key determinant of the use of fertilizer in paddy cultivation. The
study also found that there is a relatively higher correlation
between fertilizer usage and paddy price than between fertilizer
usage and fertilizer price. These findings suggest that the fertilizer
subsidy could be withdrawn gradually over time. In its place,
appropriate infrastructure and institutional facilities that are
required to increase productivity in paddy cultivation and an
effective mechanism for marketing the output that would result in
favourable prices for paddy may be introduced for a more effective
outcome. (JEL N551, Q19)
I. Introduction
Paddy production in Sri Lanka has increased considerably during the last three
decades as a result of cultivation of high yielding varieties, increase in the
area of cultivation under irrigation and greater use of plant nutrients. Successive
governments have also provided support to stimulate paddy production by
way of introducing guaranteed price schemes, major irrigation schemes and
fertilizer subsidy schemes. Introduction of high yielding varieties was the major
factor, which contributed to increase in paddy production in the country. Given
that high yielding varieties are highly fertilizer responsive, proper fertilizer
1/ The views expressed in this paper are the author’s own and do not necessarily reflect
those of the Central Bank of Sri Lanka.
S TAFF STUDIES N O. 36
application is needed to obtain respective yield levels. With respect to paddy,

the supply of three major nutrients namely, Nitrogen (N) Phosporous (P) and
Potassium (K) are essential for cultivation. Among these nutrients Nitrogen is
a major nutrient demanded by paddy plants to increase yield. Since, over use
and misuse of these fertilizer can reduce yields and also make soil less
productive, fertilizer recommendations are done by respective crop research
institutions of the country. Fertilizer recommendation for paddy is given in
Appendix I.
Figure 1 : Annual Paddy Production and Fertilizer usage in Paddy Cultivation
Paddy Production & Fertilizer Usage in Paddy

Cultivation
3,500 4 00
3,0 00 3 50
2,500 3 00
2 50
’000 Mt
’000 Mt
2,0 00
2 00
1,5 00
1 50
1,00 0 1 00
500 50
- 0
197 0 1 974 1 978 19 82 19 86 199 0 199 4 199 8 20 02
Paddy Pro. Fertilizer. Consumption
Because of the importance of fertilizers in increasing agricultural output,

government intervention in the fertilizer market started in 1962 with the
introduction of fertilizer subsidy scheme. The main objective of the subsidy
scheme was to make fertilizer available as cheaply as possible in order to
encourage its wider use thereby increasing agriculture productivity. The
fertilizer subsidy has been available for more than four decades even though
with several changes, farmers expect it to be continued in the future.
Meanwhile, the government faces great difficulty with the increasing
expenditure on the fertilizer subsidy mainly in recent times due to its fiscal
implications. Expenditure on fertilizer subsidy in Sri Lanka has gone up from
Rs.600 million in 1987 to Rs. 3,572 million in 2004. This was further increased
to Rs. 6,846 mn in 2005. Public expenditure on fertilizer subsidy scheme for
last 10 years is given in Appendix II. Therefore, the purpose of this study is
(i) to analyse the fertilizer subsidy scheme on paddy cultivation on the basis
of the estimated fertilizer demand functions and (ii) to elaborate other measures
that can be used as alternatives to the fertilizer subsidy policy.
74
THE IMPACT OF FERTILIZER SUBSIDY ON P ADDY CULTIVATION IN SRI L ANKA
In order to analyse these issues this paper is organized as follows. The

rest of this section explains the history and evolution of fertilizer subsidy
policy in Sri Lanka. Section II reviews the existing literature on the subject.
Section III discusses the data and econometric tools used in this study while
in section IV presents the results of the econometric regressions. Section V
concludes the discussion on impact of fertilizer subsidy policy on paddy
cultivation.
A. Evolution of Fertilizer subsidy policy in Sri Lanka
As fertilizer is an essential input in agriculture strong fertilizer related policies

are crucial for any national effort aiming at improving agricultural productivity.
As a major step towards achieving this objective, the government of Sri Lanka
introduced a price subsidy for fertilizer in 1962. The main objective of the
subsidy scheme was to make fertilizer available as cheaply as possible in order
to encourage its wider use. A lower fertilizer price reduces cost of production
and increases demand for fertilizer and for other inputs depending on the
elasticity of substitution, income effect and elasticity of supply of other inputs.
If sufficient quantities of fertilizers are available to meet the increased demand
and if the additional fertilizer is properly applied, food production will increase.
Initially four main fertilizers namely, Urea, Sulphate of Amonia, Muriete
of Potash (MOP) and Triple Super Phosphate (TSP), which provide
Nitrogen (N), Phosperous (P) and Potassium (K), respectively were subsidized
at different rates. Until 1975, the subsidy level varied according to the type of
crop. This scheme was found to be unsuccessful as it allowed unauthorized
leakages of fertilizer between agricultural sub sectors. Hence, in 1975 the
government introduced a uniform subsidy scheme for all crop sectors. Subsidy
rates were varied according to the type of fertilizer and the subsidy rates were
subjected to revision overtime. Later, in 1990 the government completely
removed the fertilizer subsidy for all types of fertilizer, as it was heavy burden
on the government budget. However, due to various other reasons fertilizer
subsidy scheme was reintroduced in 1994. During the period 1994 to 1997,
the fertilizer subsidy scheme underwent several changes and in 1997 the
government decided to restrict the fertilizer subsidy only to Urea. The objective
of the new scheme was to provide a higher benefit to paddy farmers while
reducing the burden on the government budget. This scheme was also subjected
to revision on a seasonal basis. During this period subsidy had been given in
either of the following two ways.
1. Selling price of fertilizer is fixed allowing the subsidy component to
vary depending on the import price
2. Subsidy component is fixed allowing the selling price of fertilizer to
vary.
75
During the late 90’s the selling price was fixed with the intension of
reducing the burden on the farming community due to price fluctuations. Since
the price was fixed with a variable subsidy component there was no incentive
to the importers to import fertilizer when the world market prices were low.
To address this issue the government decided to fix the subsidy component
and allow the selling price to vary depending on the world market prices.
When the international prices were very high it had an adverse impact on the
farmers as the cost of production increased with the increase in fertilizer prices.
This situation created financial difficulties particularly for small farmers in
the dry zone where paddy cultivation largely exist. Therefore, again in 2004
the government decided to fix retail price of fertilizer. This system continued
until December 2005. In December 2005, the government decided to
reintroduce subsidy scheme for all types of fertilizer by fixing their selling
price. However, this scheme is restricted only for paddy farmers.
II. Literature Review
There are many studies that have estimated the demand for fertilizer or for the
specific nutrients (Nitrogen, Phosphorous and Potassium). In developed
countries, it is generally agreed that fertilizer demand is price inelastic. This
may be due to lack of an economic substitute to chemical fertilizer. Generally,
in less developed countries the demand for fertilizer is thought to be more
elastic under the assumption of readily available substitutes such as manure
and other organic materials. However, the demand for fertilizer may differ
from country to country due to the factors such as cultural practices, climate,
soil type, crops grown and farm structure. In this section, an attempt is made
to review several of impotent research work on the subject.
The demand for fertilizer as a main input in agriculture has been the
focus of many studies over the years. The early studies include work by
Griliches (1958, 1959), Heady and Yeh, Carman, Gunjal, Roberts and Heady,
and others. In general, emphasis has been on national or regional demand
estimates for total fertilizer or nutrient application on all crops. Griliches (1958)
estimated the aggregate demand functions for fertilizer use on all crops in the
United States. He showed that for the 1911 to 1956 period, most of the increase
in fertilizer use could be explained by changes in both fertilizer and crop prices,
and by the previous period’s use. Using the same model, Griliches (1959)
estimated regional demand functions for total fertilizer consumption over the
1931 to 1956 period. While the model explained a large portion of the variation
in the regional fertilizer use, he found that estimated price elasticities of demand
varied across regions, ranging from elastic to inelastic.
76
Heady and Yeh (1959) estimated fertilizer demand functions for total
fertilizer and for individual nutrient used on all crops in the United States.
The variables affecting the use included fertilizer prices, crop prices, total
cash receipts from crops, total crop acreage, acres of specific crops, wage
rates, the wholesale price index, and time. In addition, they estimated
relationships for total fertilizer use in ten different geographical regions of
the United States. Their study allowed a comparison between aggregate
fertilizer and individual demand elasticities. Heady and Yeh estimated a short-
run elasticity ranging from about -0.49 to -1.71 for total fertilizers, and
elasticities of -0.45, -0.45 and -0.40 for nitrogen, phosphate and potash. Those
results confirmed that the demand for N, P and K in the U.S. is inelastic.
There have been many studies conducted on fertilizer demand in
countries other than the United States. Boyle (1982) used a cost function
approach (translog) to estimate fertilizer demand and to obtain estimates of
the demand elasticities of nitrogen, phosphorous and potassium in the Republic
of Ireland. The major objective of his paper was to test the hypothesis of
nutrient substitution over time. His findings suggested inelastic own-elasticities
for phosphorous and potassium, and an elastic own-elasticity for nitrogen,
and significant cross-price elasticities between nitrogen, phosphorous,
potassium. He found that the adverse effects of own-price increases on the
consumption of nitrogen would be offset by substitution of nitrogen for
phosphorous and potassium.
Burrell (1982) had estimated the demand for fertilizer in the United
Kingdom. He used three different modeling approaches with two different
data sets (nitrogen, and all fertilizers). He found that the elasticity with respect
to crop price is about -0.4 to -0.5 for nitrogen and between -0.1 and -0.3 for
all fertilizer in most versions of the system model. This means that demand
for nitrogen fertilizer is more sensitive to its own price than is the case for
phosphate and potash fertilizer.
Gunawardena P. and Flinn C.J. (1987) in their analysis of supply response
and fertilizer demand in rice sector in Sri Lanka, estimated short run production
elasticities by using micro level data. They concluded that, both own price
elasticity (0.15) and cross price elasticities with respect to hired labour
(-0.12) and fertilizer (-0.01) indicated an inelastic production response
implying a relatively small incentive to increase paddy production in response
to increase in the price of paddy, or reduction in the price of fertilizer. Hence,
the results suggested that the guaranteed price scheme and the fertilizer subsidy
provide small incentive to increase paddy production.
Chandrapala H.A. and De Silva S. examined the economics of fertilizer
use in Major crop sectors of Sri Lanka in 1988. They analysed some economic
aspects of fertilizer use in the tea, rubber, coconut and paddy sectors of Sri
77
Lanka. Regression models estimated for fertilizer use indicated that the relative
price ratio of fertilizer and the output has a significant effect on fertilizer use,
which points out the fact that the removal of the subsidy for fertilizer will
further worsen the situation with respect to fertilizer use in these sectors.
III. Data and Model used
A. Description of Data
As mentioned earlier, the objective of this paper is to analyse factors affecting

fertilizer demand and to investigate the impact of fertilizer subsidy in paddy
cultivation. Generally, three main fertilizers are being used in paddy cultivation
namely, Urea, Muriate of Potash (MOP), and Triple Super Phosphate (TSP).
Therefore, factors affecting the usage (demand) of these fertilizers analysed
separately. The Study consists of two main parts.
General analysis of Fertilizer Subsidy Scheme
The Fertilizer Subsidy scheme divided into three periods according to nature
of the subsidy.
1. Period I – Subsidy provided for three main fertilizers
(1962–1989, 1995–1996)
2. Period II – Period of Subsidy removal (1990–1994)
3. Period III – Subsidy provided only for Urea (1997–2005)
Since, data is not sufficient to run statistical programme for each of

these periods, all the variables that relate to fertilizer usage are presented
graphically and analysed to illustrate the changes in those variables in different
periods.
Estimation of Fertilizer Demand Function
Factors affecting the demand of three main fertilizers namely, Urea, Muriate
of Potash (MOP), and Triple Super Phosphate (TSP) analysed separately.
Factors affecting the fertilizer application can be listed as follows:
1. Fertilizer Prices (Rs./mt)
78
2. Paddy Prices (Rs./mt)

3. Extent under irrigation schemes (Ha)
4. Policy changes
Separate fertilizer demand functions were specified for the three main
fertilizers, since the fertilizer subsidy scheme has been changed according to
the fertilizer type. Regression models were specified for annual issues of
fertilizer at the national level as dependent variables and current fertilizer
price and lagged nominal farm gate price and extent under irrigation schemes
as independent variables assuming all other factors that influence fertilizer
use were remained unchanged. Dummy variables are used to interpret policy
changes. Further, it was also assumed that total amount of different fertilizers
issued for paddy cultivation during the given year were utilized fully for paddy
during the year and no carry over stocks. This assumption was made due to
lack of information regarding quantities of paddy fertilizers that were used
for the cultivation of other food crops. Own price and cross price elasticities
of demand can be obtained from estimated demand function and those
elasticities explain the sensitivity (elasiticity) of fertilizer demand to change
in fertilizer prices and paddy prices which implies the impact of fertilizer
subsidy policy on paddy cultivation.
B. Description of Econometric tools used
Several econometric tests were carried out to estimate the fertilizer demand
function and to analyse the impact of fertilizer subsidy scheme in paddy
cultivation. These methods are explained bellow.
Unit root test
Before estimating the models, it is essential to verify whether the series

analysed are stationary or not, and whether there are one or several long run
equilibrium relationships between them. If a series is stationary it means, its
variance and autocovariences are independent of time. In order to test for unit
root to see whether these variables are level stationary (I) or first difference
stationary (I (I)), the Augmented Dickey- Fuller (ADF) test is carried out. If a
series is found to be (I (I)) then first difference may have to be used in analyzing
the relationship.
79
Regression Test
The natural logarithm values of all the variables were used to estimate fertilizer
demand function by using simple linear regression model. The model is as
follows:
Qt = f { (P(f)(t)), (P (p)(t-1)), (E(p)(t)), (D) }
Qt – Quantity of fertilizer (mt) used in year t

P(f)(t) – Price of Fertilizer (Rs./mt) in year t
P(p)(t-1) – Farm Gate price paddy (Rs./mt) in year t-1
E(p)(t) – Paddy Extent under Irrigation in year t
D – Dummy for policy changes
IV. Results and Discussion
In this section, the data described in section III are illustrated graphically and
those data are used to carry out the analysis of the fertilizer demand functions
for paddy cultivation in Sri Lanka by using the statistical and econometric
methods explained earlier. Fertilizer demand functions were estimated
separately for the different fertilizers, and the results discussed separately
under each fertilizer.
A. General analysis of Fertilizer Subsidy Scheme
Period I – subsidy provided for three main Fertilizers
During the period from 1962 to 1989 the subsidy scheme was continued for
all main fertilizers, which provide N, P, K to plants. Since, data are not available
from the beginning, data from 1980 to 1989 are used to represent the period.
During the period under review, usage of all three fertilizers increased
continuously. However, the ratio between three fertilizers was almost
maintained. A remarkable up rise in Urea consumption in 1980 was mainly
due to the low level of retail price (Rs.980/mt) prevailed during the year.
Fertilizer prices reduced sharply since the government increased its subsidy
rate from uniform subsidy of 50 per cent of Cost and Freight (C&F) price of
fertilizers to 85 per cent, 75 per cent, 75 per cent of the C&F price of Urea,
80
Figure 2 : Changes in Fertilizer Usage in Period I
Mt./Ha Changes in Fertilizer Usage

0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
2000
2001
2002
2003
2004
1997
1998
1999
Urea TSP MOP
MOP, TSP respectively. However, in 1981 the world market fertilizer prices
increased and its impact was severe with the deterioration of Rs./US$ exchange
rate. Hence, the government allowed fertilizer prices to increase by about
30 per cent by rearranging the subsidy components to 65 per cent, 65 per cent
and 40 per cent of C&F prices of Urea, MOP and TSP ,respectively.
This resulted in shortfall in fertilizer consumption in paddy cultivation
in 1981. Since then until 1984, fertilizer consumption remained almost stable
even though with several changes in subsidy components and retail prices.
In 1984 while the prices of fertilizer remained stable, farmers enjoyed an
increase in the guaranteed price of paddy. As a result fertilizer consumption
triggered upward. This trend continued until termination of fertilizer subsidy
scheme by the government in 1989.
Figure 3 : Changes in Fertilizer Price and Farm gate Price of Paddy in Period I
Changes in Fertilizer Prices Changes in Farm gate Price of Paddy
Rs./mt Rs./mt
4,000 7,000
6,000
3,500
5,000
3,000
4,000
2,500
3,000
2,000 2,000
1981
1989
1983
1984
1985
1987
1981
1986
1982
1988
1983
1989
1984
1985
1987
1986
1982
1988
81
The selling prices of fertilizer during this period is determined by the

National Fertilizer Secretariat in consultation with fertilizer whole sellers
taking in to consideration the cost of fertilizer and the subsidies made available
by the government. Therefore, subsequent to the reduction in the fertilizer
subsidies, an increase in the prices of subsidized fertilizers could occur.
Accordingly an increase in retail price of fertilizer observed in 1983 due to
the reduction of subsidy by around 15 per cent of C&F cost of fertilizer. These
fertilizer prices were not revised until 1988. In 1988 there was a sharp
escalation in the cost of fertilizer in the international market. Hence it became
necessary to increase the local price of fertilizers by reducing subsidy
component. Accordingly prices of three major fertilizers increased by around
30 percent in 1988. Increasing trend in paddy prices continued during the
period. However, in 1988/89 period, there was a sharp increase in the farm
gate price of paddy as the government increased guaranteed price of paddy
mainly to cushion farmers from the adverse impact of higher fertilizer prices.
Period II (1990-1994) – Period of Subsidy Removal
Increasing fertilizer prices in the international market with rising oil prices
together with depreciation of exchange rate caused considerable difficulties
in maintaining the fertilizer prices unchanged during 1989. Hence, the fertilizer
subsidy was completely withdrawn by the government with effect from
1 January 1990. With the removal of fertilizer subsidy there was sharp
escalation of fertilizer prices from the beginning of 1990. The total use of
Figure 4 : Changes in Fertilizer Usage in Period II
Changes in Fertilizer Usage

Mt/Ha
0.2
0.15
0.1
0.05
0
1990 1991 1992 1993 1994
Urea TSP MOP
82
fertilizer declined in 1990, in the wake of increase in fertilizer prices due to

the removal of the subsidy. However, the decrease in consumption was not as
high as was anticipated. It may well be that in anticipation of a price increase
in 1990 after removal of the subsidy farmers stocked their requirement for
their crops. Further, with the removal of the subsidy, in order to cushion the
adverse impact of sudden price increases of fertilizer, government revised
upward the guaranteed price of paddy.
In the years of 1991,1992, and in 1993, the overall consumption of
fertilizer increased by 3.5 per cent, 16.0 per cent and 17.0 per cent respectively
indicating that most of the farmers were gradually adjusting themselves to the
new prices. Furthermore, other scheme such as granting of credit in kind for
purchasing of seed and fertilizer under the agricultural trust fund and the
favourable farm gate price for paddy were some of the important factors
contributed to this improvement.
The sharp increase in fertilizer consumption in 1993 over that of 1992
was mainly due to the increase in area cultivated under paddy during that year
which indicates the positive relationship between fertilizer consumption with
extent under paddy. Continuing the increasing trend in 1994, the fertilizer
usage in paddy sector reached a record level representing around 14 per cent
growth over the previous year. Increase in area under paddy around 11 per
cent over the previous year and re introduction of fertilizer subsidy in October
1994 were the major factors that contributed to this increase.
Consequent to the removal of fertilizer subsidy, there was about
two-fold increase in the prices of all subsidized fertilizers. However, due to
the action taken by the government to increase farm gate price of paddy helped
farmers to minimize the adverse impact of sudden increase in fertilizer prices.
The drastic increase in selling prices of fertilizers were seen in 1991 as several
local price revisions were done according to the world market price changes.
Figure 5 : Changes in Fertilizer Prices and Farm gate Price of paddy in Period II
Changes in Fertilizer Prices Changes in Farm gate Price of

Rs/Mt Rs/mt Paddy
11000 9000
10000 8500
9000 8000
8000 7500
Urea MOP
TSP
7000 7000
1990 1991 1992 1993 1994 1990 1991 1992 1993 1994
83
However, during the period 1991 to October 1994, fertilizer prices remained
stable. Drastic change in fertilizer prices occurred with the reintroduction of
fertilizer subsidy in 10 October 1994 by the new government that came in to
power. Accordingly, four main fertilizers were subsidized by about 30 per
cent of their retail price.
The increasing trend in farm gate price of paddy that started in 1990
continued until 1993. With the increase in marketable paddy surplus in 1994,
the farm gate price of paddy dropped by 13 per cent. However, the drop in
paddy prices did not affect fertilizer consumption in 1994 since fertilizer prices
were reduced by the reintroduction of fertilizer subsidy during the fourth
quarter of the year.
Period III (1997-2004) – Subsidy Provided only for Urea

There was a drastic change in the fertilizer subsidy scheme in October 1997
as it restricted fertilizer subsidy only to urea. The objective of this change
was to provide a higher benefit to paddy farmers who were the main users of
urea. Past statistics indicate approximately 75 per cent of Urea used in the
country was utilized for the paddy sector, which is dominated by small-scale
farmers. Therefore, the government decided to confine the subsidy scheme
for urea only with a view to pass on the benefits of the scheme to small-scale
farmers. The subsidy rates were announced in rupee terms and it was decided
to review the subsidy amount on a seasonal basis maintaining the retail price
of Urea unchanged during the period. In 2002, the subsidy scheme was revised
with increased international prices. Accordingly a fixed sum of subsidy was
Figure 6 : Changes in Fertilizer Usage in Period III
Mt./Ha Changes in Fertilizer Usage

0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
2000
2001
2002
2003
2004
1997
1998
1999
Urea TSP MOP
84
given to urea importers allowing retail prices to be varied according to the

world market prices of fertilizer. However, in 2004 the new government came
in to power and it was decided to fix retail price of Urea again to avoid severe
fluctuations in retail price.
With the commencement of new subsidy scheme in 1997, use of fertilizer
for paddy decreased marginally over the previous year. This was entirely due
to the reduction in the use of TSP and MOP, since Urea usage increased by
1.6 per cent during the period. Use of fertilizer for paddy recovered in 1998
due to the increase in usage of highly subsidized Urea by 20.6 per cent over
that of 1997. However, use of MOP increased marginally while application of
TSP decreased over the previous year. In 1999, use of Urea for paddy increased
substantially to a record level, indicating that paddy farmers benefited most
from the fertilizer subsidy scheme. Use of TSP for paddy too increased in
1999, even though it was not proportionate to maintained the N:P:K ratio in
the overall fertilizer use.
In line with the decline in area cultivated in 2000, the total fertilizer use
for paddy dropped by 18 per cent. However, it was recovered in 2001 and
fertilizer usage for paddy increased by 10.6 per cent despite the drought
experienced during the year. This trend continued in 2002 as well. In 2003,
use of fertilizer dropped significantly mainly due to the rise in fertilizer prices
as a result of the revision of fertilizer subsidy scheme in October 2002 with
the increased price in the international market. Under the revised scheme a
fixed sum of Rs.6,000 per metric ton was given as a subsidy payment for Urea
importers. This scheme was continued until May 2004. The new government
reintroduced subsidy scheme with fixed retail price and as a result Urea usage
increased slightly over the previous year.
With the introduction of new subsidy scheme in 1997, there were major
price revisions during the years. The sale price of Urea was maintained at the
Figure 7 : Changes in Fertilizer Prices and Farm gate Price of paddy in Period III
Changes in Fertilizer Prices Changes in Farm gate Price of Paddy
Rs./mt Rs./mt
3 50 0 0 1 70 0 0
3 0 00 0 1 50 0 0
2 50 0 0
1 3 00 0
2 0 00 0
1 1 0 00
1 5 0 00
9 0 00
1 00 00
500 0 7 00 0
0 5 00 0
1997
1998
1999
2000
2001
2002
2003
2004
1998
1999
2000
2001
2002
2003
2004
1997
Urea MOP TSP
85
indicative price under the fertilizer subsidy scheme. Prices of other fertilizers
were determined by the market considering the factors such as appreciation
of US$, price fluctuations in the international market and freight rates. With
the increase demand for Urea owing to the subsidized price there was a stiff
competition among sellers and hence the actual price of Urea operated below
the indicative price of the government. As a result Urea price came down and
remain static until 2002 while prices of other fertilizers moved upward with
increase in import prices. In 2002 with the change in fertilizer subsidy scheme
annual average price of Urea increased to Rs.9,400 per metric ton in 2002
from Rs.7,000 per metric ton prevailed in 2001. The urea price was further
increased in 2003 and reached highest ever retail price of Rs.17,000 per metric
ton as at end 2003. However, with the reintroduction of fixed retail price in
2004, the price of urea dropped significantly over that of previous year.
Paddy price, which is mainly determined by the supply conditions,
heavily fluctuated during the period. In 1998, with the increased paddy
production, the average producer price of paddy declined over the previous
year. However, this price remained well above the guaranteed price of paddy.
In contrast to the normal behaviour, in 2000 the farm gate price of paddy
declined mainly due to rice importation even though the paddy production
also dropped. In 2001 the farm gate price increased and maintained at a higher
level until 2003. Paddy price dropped in 2003 with the record level harvest
and increased again in 2004 with the drop in paddy production due to severe
drought prevailed in the country. Fertilizer usage in 2003 was highly affected
by the dropped in paddy price and also due to increased fertilizer price under
the revised subsidy scheme.
According to the above analyse it reveals that fertilizer usage can be
influenced by fertilizer price, producer price of paddy and extent under paddy
cultivation. Hence to estimate the sensitivity of these variables to fertilizer
usage separate demand functions were employed by using 25 years of data.
B. Estimation of Fertilizer Demand Functions
Demand Function: Urea

Prior to engage in performing econometric analysis, all variables in log form
were subjected to Augmented Dicky Fuller (ADF) unit root tests to verify
their time series properties. The results were also tested by using Philips Perron
(PP) test.
86
Table 1 : Unit Root Test – Variables in Levels

Variable in levels ADF PP
UQ (Urea Usage) -1.114992 -0.820758
UP (Urea Price) -1.882988 -1.824195
PP (Paddy Price) -1.116489 1.681542
EX (Extent) -1.227110 -0.986910
* Significant at 5% level
Since all the variables were found to be non-stationary on levels, unit

root test were carried out on first differences. As indicated below all first
difference series were stationary at a very high levels of significance.
Therefore, it was concluded that all the series were I(I).
Table 2 : Unit Root Test – Variables in First Difference

Variables in 1st Difference ADF PP
UQ1 -4.890511** -5.319587**
UP1 -5.159817** -5.214621**
PP1 -5.324109** -5.359762**
EX1 -5.403066** 7.241751**
** Significant at 1% level
Given that urea usage, urea price, paddy price and extent under irrigation
were I(I), simple linear regression analysis was carried out to estimate size
and significance of those variables to urea usage. The model used is as follows.
Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D1) }

D1 – Dummy for policy changes
87
Table 3 : Results of the regression – Urea

Dependent Variable : UQ1
Method : Least Squares
Sample (adjusted) : 1981–2004
Included observations : 24 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
C -0.611588 2.888294 -0.211747 0.8346
UP1 -0.154291 0.087234 -1.768702 0.0930
PP1(-1) 0.330789 0.150557 2.197098 0.0406
EX1 2.532465 0.895402 2.828298 0.0107
DU 0.182554 0.071723 2.545274 0.0198
R-squared 0.913806 Mean dependent var 11.87652
Adjusted R-squared 0.895660 S.D. dependent var 0.287835
S.E. of regression 0.092976 Akaike info criterion -1.729909
Sum squared resid 0.164245 Schwarz criterion -1.484481
Log likelihood 25.75891 F-statistic 50.35847
Durbin-Watson stat 1.972436 Prob (F-statistic) 0.000000
Results show that all the selected variables are significant for urea usage
with theoretically expected signs. Further, these results indicate that fertilizer
demand is inelastic to own price, cross price and policy changes. However,
extent under irrigation is highly sensitive to urea usage, which implies that
increasing the total area under paddy under irrigation would have a higher
impact on the urea application. Previous researches also observed that the
input use is higher under irrigated conditions than in rain fed lands in the
country.
Inelasticity of urea usage to its own price, cross price and policy changes
suggests that unavailability of substitutes for urea encourage farmers to apply
urea with out considering price changes. According to the results a one-unit
increase in price of a metric ton of urea would reduce annual urea demand for
paddy by 0.15 units. If lagged farm gate price increases by one unit per metric
ton, urea demand will increase by 0.33 units. This information reveals that
fertilizer demand is more sensitive to change in paddy price than to change in
fertilizer prices. If fertilizer prices were increased holding paddy prices
constant a reduction in urea use would occur.
Demand Function: Muriate of Potash (MOP)

At first all the variables in log form that use to analyse MOP demand function
were tested for stationary. The ADF tests showed that all variables are
88
stationary in first difference, i.e., I(I) and the results were confirmed by Philips
Perron tests.
Table 4 : Unit Root Test – Variables in First Difference

Variables in 1st Difference ADF PP
MOPQ1 (MOP usage) -4.164773** -5.048492**
MOPP1 (MOP price) -6.095492** -6.095492**
PP1 (Paddy Price) -5.324109** -5.359762**
EX1 (Extent) -5.403066** -7.241751**
** Significant at 1% level
A Simple linear regression was carried out to estimate size and

significance of those variables to MOP usage. Since there is a tendency to use
of urea excessively than other fertilizers due to its lower price, in analyzing
MOP demand function, the Urea price is also taken as an independent variable.
The model used is as follows.
Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D2) }

Results show that except extent under irrigation all other selected
variables are significant for MOP Usage with theoretically expected signs.
Similar to the results of the Urea demand function, results of the MOP demand
function indicates that MOP demand is inelastic to own price, cross price and
policy changes.
Inelasticity of MOP usage to its own price, cross price and policy changes
implies that since there are no substitute available for MOP, farmers tend to
apply MOP without considering price changes. According to the results, a
one-unit increase in price of a metric ton of MOP would reduce its annual
demand for paddy by 0.21 units. If lagged farm gate price increases by one
unit per metric ton, MOP demand will increase by 0.79 units. These results
89
Table 5 : Results of the regression – MOP

Dependent Variable : MOPQ1
C 4.494407 2.981188 1.507589 0.1481
MOPP1 -0.212730 0.117575 -1.809312 0.0863
PP1(-1) 0.792315 0.235511 3.364243 0.0033
EX1 0.149572 0.978238 0.152900 0.8801
D2 0.248983 0.091279 2.727709 0.0134
also confirm that the fertilizer demand is more sensitive to change in paddy
price than to change in fertilizer prices. Further, in the case of MOP demand,
own price and cross price elasticities are relatively higher than in Urea demand.
Demand Function: Triple Super Phosphate (TSP)

Except quantity used and the fertilizer price, all the other variables were similar
to the previous estimation. Since the ADF tests confirmed that all variables
are stationary in first difference, i.e., I(I), a simple linear regression were
carried out to estimate TSP demand function. The Urea price was taken as an
independent variable for this analysis too. The Model and variables are given
bellow.
Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D2) }

90
Table 6 : Results of the regression – TSP

Dependent Variable : TSPQ1
C 1.336310 3.500426 0.381756 0.7069
TSPP1 -0.075420 0.151730 -0.497065 0.6248
EX1 1.365139 1.138181 1.199405 0.2451
PP1(-1) 0.414735 0.290252 1.428878 0.1693
D2 0.273064 0.110374 2.473980 0.0230
Even though results confirmed theoretically expected signs, most of the

variables are not significant for TSP demand. According to the results, changes
in fertilizer price, out put price and the cheapest fertilizer price could not
influence the TSP demand significantly.
V. Conclusion
Findings in section IV indicate that even though there are some similarities,
fertilizer demand functions could vary according to the type of fertilizer.
Selected main fertilizers are inelastic to its’ own price, output price and policy
changes with different significance levels. Therefore, it could be concluded
that the fertilizer demand is not significantly affected by changes in own price,
output price and related policies in the short run and hence fertilizer subsidy
is not a key determinant of fertilizer usage in paddy cultivation. These results
were also in line with the findings of related research in other countries. Since
the unavailability of substitutes for fertilizer, farmers tend to apply fertilizer
with whatever difficulties to ensure the optimum yield levels. However,
increase in fertilizer price could reduce farmers’ profit since it increases cost
of production.
Further, these results also revealed that own price elasticity of demand
for fertilizer was smaller than cross price elasticity between demand for
91
fertilizer and paddy price. Therefore, increasing paddy price would increase
demand for fertilizer at a faster rate than when decreasing fertilizer prices.
Increasing paddy prices or decreasing fertilizer prices are the alternatives
available to the government to provide incentives to the paddy farmers in the
short run. Even though increase in output price could play a greater role in
this regard, such a policy would spread the farmers’ problem over the entire
society.
Hence, policy makers should weigh pros and cons of two options
available before making a decision. However, a sudden withdrawal of the
subsidy will push farmers in low-income groups into worse situation than
they were in the past. At least there should be a way to group farmers according
to their income levels and to help low income group with subsidies. Therefore,
well-targeted subsidy scheme, which is subjected to gradual phase out, is
preferred in the short run. The government’s decision of restricting fertilizer
subsidy only for paddy farmers from 2006 is an important step taken for this
direction.
It is also revealed that farmers had not been using correct amount of
fertilizer to maintain N:P:K ratio of the field especially during the period III
(1997 – 2004) in which only urea was subsidized. This situation should also
be corrected through a sound agricultural extension programme. Further, it
will reduce cost of production by increasing average yield.
Meanwhile, public expenditure should also be focused on providing
infrastructure as well as institutional facilities that are required to improve
efficiency in paddy marketing to allow paddy price to be attractive. Once
productivity and profitability in paddy cultivation is increased to an acceptable
level fertilizer subsidy could be withdrawn completely.
92
References
Boyle, G.E., (1982), “Modelling Fertilizer Demand in the Republic of Ireland:

A Cost Function Approach”, Journal of Agricultural Economics, Vol. 33.
Burrell, A., (1982), “The Demand for Fertilizer in the United Kingdom”,
Journal of Agricultural Economics, Vol. 40:1-20.
Central Bank of Sri Lanka, Annual Reports, various issues.
Chandrapala, H.A. and De Silva, S., (1988), “Economics of Fertilizer use in
major Crop sectors of Sri Lanka”, Staff Studies, Central Bank of Sri Lanka
(CBSL).
Griliches, Z., (1958), “The Demand for Fertilizer: An Economic Interpretation
of a Technical Change”, Journal of Farm Economics, Vol. 40 (3).
Griliches, Z., (1959), “Distributed Lags, desegregation and Regional Demand
Function for Fertilizer”, Journal of Farm Economics, Vol. 41 (2).
Gunawardena, P. and Flinn C. J., (1987), “Supply Response and Fertilizer
demand in Rice sector in Sri Lanka”, Quarterly Journal of International
Agriculture, Vol. 26 (4).
Heady, E.O. and Yeh, M. H., (1959), “National and Regional Demand
Function for Fertilizer”, Journal of Farm Economics, Vol. 41 (2).
National Fertilizer Secretariat,“Fertilizer Review” various issues.
Rice Research Institute, (2004), “Fertilizer Recommendations for Paddy”,
Internal Circular.
93
Appendix 1 : Fertilizer Recommendations for Paddy*
To obtain 100 Bushels per Acre To obtain 120 Bushels per Acre To obtain 140 Bushels per Acre
Three months
variety Urea T.S.P. M.O.P. Urea T.S.P. M.O.P. Urea T.S.P. M.O.P.
Kg/Acre Kg/Acre Kg/Acre Kg/Acre Kg/Acre Kg/Acre Kg/Acre Kg/Acre Kg/Acre
Basel Dressing 05 25 15 05 35 15 05 45 15
1st top dressing 35 40 50
2nd top dressing 50 10 60 15 70 20
Total
Requirement 90 25 25 105 35 30 125 45 35
Source: Rice Research Institution, Batalagoda, 2004
94
* Fertilizer recommendations (Kg/Acre), for three-month paddy variety in Anuradhapura, Polonnaruwa, Moneragala, Hambanthota,
Puttlam, Mannar, Vavuniya, Ampara, Kilinochchi, Trincomalee, Batticalo, Kurunegala Districts.
Appendix 2 : Public Expenditure on Fertilizer Subsidy
Year Expenditure As a Percentage

(Rs.mn) of GDP
1987 600 0.27
1988 347 0.32
1989-1993 – –
1994 630 0.1
1995 1,345 0.2
1996 1,500 0.2
1997 1,895 0.2
1998 2,152 0.2
1999 1,390 0.1
2000 1,733 0.1
2001 3,650 0.3
2002 2,448 0.2
2003 2,191 0.1
2004 3,572 0.2
Source: Central Bank Annual Reports
Appendix 3 : Annual Average Retail Price of Fertilizer

Rs/Mt
Year Urea MOP TSP Year Urea MOP TSP
1981 2,463 2,565 2,375 1993 9,850 9,500 10,300
1982 2,785 2,900 2,685 1994 9,850 9,500 10,300
1983 2,850 2,750 2,850 1995 10,200 10,196 11,204
1984 2,850 2,750 2,850 1996 11,000 11,250 12,200
1985 2,850 2,750 2,850 1997 9,400 12,625 15,375
1986 2,850 2,750 2,850 1998 6,800 13,500 19,200
1987 2,850 2,750 2,850 1999 6,300 14,633 19,200
1988 3,650 3,550 3,650 2000 7,000 16,500 19,200
1989 3,650 3,550 3,650 2001 7,000 17,550 18,200
1990 7,900 8,200 9,650 2002 9,450 19,940 21,000
1991 10,000 9,367 10,050 2003 17,000 24,500 25,500
1992 9,850 9,500 10,300 2004 10,740 32,200 33,250
Source: National Fertilizer Secretariat
95
Appendix 4 : Annual Use of Fertilizer in Paddy Cultivation

000’Mt
Year Urea TSP MOP Year Urea TSP MOP
1980 116.2 18.3 17.1 1993 136.5 37.7 32.7
1981 94.6 17.7 16.4 1994 161.5 35.2 40.1
1982 97.3 22.2 21.8 1995 157.6 34.6 40.0
1983 97.0 21.8 23.0 1996 147.0 34.8 36.9
1984 112.6 25.1 24.5 1997 149.3 26.8 32.7
1985 125.0 28.0 27.8 1998 180.1 23.7 34.6
1986 139.7 36.3 32.4 1999 234.2 34.1 38.4
1987 133.0 28.8 31.9 2000 193.3 27.1 29.8
1988 131.0 28.6 33.8 2001 214.4 32.9 29.5
1989 139.1 32.0 32.0 2002 270.1 37.2 37.5
1990 99.2 20.5 26.0 2003 201.9 38.2 33.2
1991 98.2 22.8 29.8 2004 222.2 34.8 36.6
1992 118.9 24.2 32.3
Source: National Fertilizer Secretariat
96

The Impact of Fertilizer Subsidy On Paddy Cultivation in Sri Lanka

Uploaded by

Copyright:

Available Formats

The Impact of Fertilizer Subsidy On Paddy Cultivation in Sri Lanka

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Impact of Fertilizer Subsidy On Paddy Cultivation in Sri Lanka

Uploaded by

Copyright:

Available Formats

The Impact of Fertilizer Subsidy on Paddy Cultivation

The Government of Sri Lanka has been subsidizing fertilizer

application is needed to obtain respective yield levels. With respect to paddy,

Figure 1 : Annual Paddy Production and Fertilizer usage in Paddy Cultivation

Paddy Production & Fertilizer Usage in Paddy

Paddy Pro. Fertilizer. Consumption

Because of the importance of fertilizers in increasing agricultural output,

In order to analyse these issues this paper is organized as follows. The

A. Evolution of Fertilizer subsidy policy in Sri Lanka

As fertilizer is an essential input in agriculture strong fertilizer related policies

II. Literature Review

III. Data and Model used

As mentioned earlier, the objective of this paper is to analyse factors affecting

General analysis of Fertilizer Subsidy Scheme

Since, data is not sufficient to run statistical programme for each of

Estimation of Fertilizer Demand Function

2. Paddy Prices (Rs./mt)

B. Description of Econometric tools used

Unit root test

Before estimating the models, it is essential to verify whether the series

Qt = f { (P(f)(t)), (P (p)(t-1)), (E(p)(t)), (D) }

Qt – Quantity of fertilizer (mt) used in year t

IV. Results and Discussion

A. General analysis of Fertilizer Subsidy Scheme

Period I – subsidy provided for three main Fertilizers

Figure 2 : Changes in Fertilizer Usage in Period I

Mt./Ha Changes in Fertilizer Usage

Urea TSP MOP

The selling prices of fertilizer during this period is determined by the

Period II (1990-1994) – Period of Subsidy Removal

Figure 4 : Changes in Fertilizer Usage in Period II

Changes in Fertilizer Usage

fertilizer declined in 1990, in the wake of increase in fertilizer prices due to

Changes in Fertilizer Prices Changes in Farm gate Price of

Period III (1997-2004) – Subsidy Provided only for Urea

Figure 6 : Changes in Fertilizer Usage in Period III

Mt./Ha Changes in Fertilizer Usage

Urea TSP MOP

given to urea importers allowing retail prices to be varied according to the

Urea MOP TSP

B. Estimation of Fertilizer Demand Functions

Demand Function: Urea

Table 1 : Unit Root Test – Variables in Levels

Since all the variables were found to be non-stationary on levels, unit

Table 2 : Unit Root Test – Variables in First Difference

Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D1) }

Qt – Quantity of fertilizer (mt) used in year t

Table 3 : Results of the regression – Urea

Demand Function: Muriate of Potash (MOP)

Table 4 : Unit Root Test – Variables in First Difference

A Simple linear regression was carried out to estimate size and

Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D2) }

Qt – Quantity of fertilizer (mt) used in year t

Table 5 : Results of the regression – MOP

Demand Function: Triple Super Phosphate (TSP)

Qt = f { (P(f)(t)), (P(p)(t-1)), (E(p)(t)), (D2) }

Qt – Quantity of fertilizer (mt) used in year t

Table 6 : Results of the regression – TSP