©The Pakistan Development Review
48 : 4 Part II (Winter 2009) pp. 921–938
Efficiency Dynamics of Sugar
Industry of Pakistan
ABDUL RAHEMAN, ABDUL QAYYUM, and TALAT AFZA*
1. INTRODUCTION
Sugarcane is among the most valuable crops of Pakistan. It is a source of raw
material for entire sugar industry. At present, the sugar industry is second largest agrobased industry in Pakistan. The future of this industry in Pakistan is mainly attributed to
the poduction efficiency because of higher cost of production; increase in the imports and
due to declining competitiveness of the domestic sugar industry. Productive efficiency
can be improved by the adoption and development of new production technologies but at
present it is difficult due to limited income and credit to the out growers. Therefore, this
industry can improve the efficiency of its operations using currently available technology.
Measures of productivity, its growth and sources for the sugar industry of Pakistan
play a significant role for policy development. Productivity growth can be decomposed
into three components: technical change, scale effects, and changes in the degree of
technical efficiency [Coelli, et al. (2005)]. Technical change means progress in
technology not only physically in the form of improved machinery but also innovations
in the knowledge base. Regarding scale effects, it relate to economies in production. If
there exists increasing economies of scale it indicates that the production of additional
outputs will require a less than proportional increase in inputs. Improvements in the
degree of technical efficiency arise from situations where resources can be used more
efficiently by applying practices from the present stock of knowledge.
The most comprehensive measure of aggregate or sectoral productivity is Total
Factor Productivity (TFP). However, given the paucity of good data, this area of research
has remained quite limited in Pakistan [Ali (2004)]. There are some studies on
manufacturing sector of Pakistan which include Raheman, et al. (2008), where TFP and
its sources are estimated using Malmquist Productivity growth index for major
manufacturing industries of Pakistan using aggregate firm level financial data but sugar
industry is not among the industries analysed. The results of the study highlighted the
Abdul Raheman <abdulrehman@uaar.edu.pk> is PhD Scholar, Department of Management Sciences,
COMSATS Institute of Information Technology, Islamabad and Assistant Professor, University Institute of
Management Sciences, PMAS-Arid Agriculture University Rawalpindi. Abdul Qayyum <abdulqayyum@pide.org.pk>
is Professor at the Pakistan Institute of Development Economics, Islamabad. Talat Afza
<talatafza@ciitlahore.edu.pk> is Professor, Department of Management Sciences, COMSATS Institute of
Information Technology, Lahore.
922
Raheman, Qayyum, and Afza
role of efficiency change in the TFP growth while deficiencies in terms of technological
progress. Similarly, another study by Mahmood, et al. (2007) examined the efficiency of
the large scale manufacturing sector of Pakistan by using the stochastic production
frontier approach for periods 1995-96 and 2000-01. Afzal (2006) also analysed the TFP
for the large scale manufacturing sector from 1975 to 2001 using three different
approaches. There are no reported productivity efficiency studies for the sugar industry in
Pakistan.
This study attempts to fill this gap by estimating firm level efficiency and total
factor productivity growth and its components for a sample of twenty sugar firms in the
sugar industry and to assess the variations in TFP growth between firms and over Time.
The TFP growth is estimated for the period 1998 to 2007. This study, therefore, would
provide a fresh perspective on the growth of TFP in sugar sector for use in developing
appropriate policy responses towards this sector of Pakistan’s economy.
There are several techniques available, parametric and non-parametric, to estimate
total factor productivity. The most widely used example of a non-parametric technique is
DEA [Coelli (1995); Seiford (1996)]. Parametric techniques encompass stochastic
frontier techniques and Bayesian methods [Kalirajan and Shand (1999)]. In this paper we
employ DEA to estimate Malmquist TFP indices from panel data set. The reason for the
choice of DEA as the method of estimation is that the methodology has been employed
widely to conduct benchmarking analysis [for example, see Jaforullah and Whiteman
(1999)]. Most of the existing studies that employs panel data for estimation of efficiency
and productivity change reports estimates for the entire data period, while in the present
study our focus is on the annual estimates because we wish to examine how productivity
changes through time at the firm level.
The basic objective of this paper is to use the Data Envelopment Analysis (DEA)
as a tool for the measurement of TFP growth for sugar industry and sugar firms. The
objective/purpose is also to decompose TFP growth into technical change, efficiency
change and scale efficiency change in order to understand the source of productivity for
Pakistani sugar firms listed at Karachi Stock Exchange. This decomposition enables
policymakers to trace lagging productivity to particular factors. For example, if slowing
technical progress causes declining TFP growth, the production frontier can be shifted
upward through investment in research and development (R&D); if slow productivity
growth is traced primarily to deteriorating technical efficiency (TE), learning-by-doing
processes and managerial practices can be targeted for this purpose; if there will be
benefits from SE, production scales should be adjusted toward optimum values. The
specific objective of the study is to provide policy implications and strategies for
improvement in the production efficiency of sugar firms. Policymakers can recommend
policies that improve the productivity of firms only if they understand the sources of
variation in productivity growth.
Generally, studies at country level on productivity growth are based on the
overall or aggregate data; therefore, the results of those studies are average of the
overall economy which comprises of different sectors. Hence contribution in each
country’s productivity has different proportion of sectors. This study uses financial
data of sugar firms extracted from annual reports obtained from different sources.
This data allows examination of the TFP performance of individual firms, which was
not previously done.
Efficiency Dynamics of Sugar Industry of Pakistan
923
The structure of this article is as follows. In the following section, an overview of
sugar industry of Pakistan is presented followed by the third section which describes the
data used in the analysis and methodology opted for analysis including discussion of
input and output variables. Then the results of our Malmquist TFP estimates are
presented. In the final section we discuss the results presented and provide conclusions.
2. OVERVIEW OF SUGAR INDUSTRY OF PAKISTAN
Sugarcane is an important industrial and cash crop in Pakistan. Pakistan is an
important sugarcane producing country and is ranked fifth in terms of area under sugar
cultivation, 60th in yield and 15th in sugar production. Sugarcane is grown on over a
million hectares and provides the raw material for Pakistan’s 84 sugar mills which
comprise the country’s second largest agro-industry after textiles [Pakistan Annual Sugar
Report (2009)]. The sugar sector constitutes 4.2 percent of manufacturing. In size, the
sugar sector matches the cement sector. Sugar industry has an indirect socio-economic
impact in overall terms which is significantly larger than its direct contribution to GDP
because of it’s backward (sugarcane growers) and forward linkages (food processors) in
the economy.
The sugar cane yield for some important countries of the world is given in the
following Table 1.
Table 1
Country
Australia
Egypt
Brazil
USA
Colombia
Mexico
India
Pakistan
World Avg.
Sugarcane Yield of World
Cane Yield (T/ha)
Sugar Recovery (%)
100.4
13.8
110.8
11.5
68.4
14.5
80.2
11.7
80.5
11.5
79.5
11.6
66.9
9.9
49.0
9.2
64.4
10.6
Sugar Yield (t/ha)
13.85
12.74
9.91
9.38
9.26
9.22
6.64
3.54
6.82
Source: www.pakboi.gov.pk/word/Sugar%20.doc
According to the Table 1, Egypt is the highest in terms of sugarcane yield per
hector which is 110.8 tons per hector while the Pakistan is the lowest in terms of this
yield. As far as the sugar recovery is concerned, Brazil has the highest percentage and
again Pakistan is at the lowest. If we analyse the sugar yield from sugarcane, Australia
has the highest sugar yield in these countries and again Pakistan is at the lowest with 3.54
tons per hector. It indicates that in Pakistan, improvements can be made in terms of
sugarcane yield, sugar recovery and sugar yield.
The area under cultivation has increased more rapidly than any other major
crops. The Table 2 presents the area production and yield during period 1997-98 to
2007-08.
Raheman, Qayyum, and Afza
924
Table 2
Year
Pakistan Sugarcane Area and Yield
Produced 000
Area (000 Ha)
Tonnes
Yield per Hectare
Utilisation % by
Sugar Mills
1997-98
1,056.2
53,104
50.28
77.32
1998-99
1,155.1
55,191
47.78
77.90
1999-00
1,009.8
42,000
41.59
69.00
2000-01
960.0
43,620
45.40
67.47
2001-02
999.7
48,041
48.10
76.33
2002-03
1,099.7
52,049
47.30
80.28
2003-04
1,074.8
53,800
50.10
81.15
2004-05
966.4
43,533
45.00
73.74
2005-06
907.0
44,292
48.80
67.94
2006-07
1,033.0
54,871
53.12
73.78
2007-08
1160.0
61,503
53.02
–
2008-09
1045.0
55,385
53.00
–
Source: Pakistan Sugar Mills Association Annual Report: 2007, 2008.
During the year 2007-08 production of sugar was estimated at 61.5Million Metric
Ton (MMT), an increase of 12 percent over previous year due to increase in area under
cultivation and yield. While during 2008-09 sugar production is estimated at 55MMT a
decline of 10 percent over the previous year. According to press reports [Jang Weekly
News, August (2009)], Pakistan’s 2009-10 sugar production is expected around 3
millions tons as against 3.2 million tons in the last year. The annual consumption of sugar
varies in between 3.6 to 4.2 million tons, but according to the industry’s officials, it has
gone down since October due to economic slowdown and higher prices that resulted in
lower demand from industries like drink producers. With this scenario, Pakistan has to
import sugar which exposes it to the effects of shortage and rising prices in the world.
The consumption of sugar is showing an increasing trend for the last 15 years. In
1995-96, it was 2.89 million tons, which increased to 3.95 million tons in 2005-06. This
is mainly due to increase in the population growth of the country, which is now almost
170 million. According to a rough estimate, the country will need approximately 5.5
million tons of sugar to meet the local demand by year 2020. It will require about 1.5
million hectares of area under cultivation which is at present about 1 hector. The per
capita sugar consumption is around 25kg per year which is highest in the developing
countries. The demand of sugar will increase in the coming years at the rate of about 2.3
percent because of growth in the population which is about 2.3 percent.
The sugarcane production in terms of sugarcane crushed, sugar made and recovery
percentage is presented in the Table 3 for period 1997-98 to 2006-07.
Efficiency Dynamics of Sugar Industry of Pakistan
925
Table 3
Year
1997-98
1998-99
1999-2000
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
Sugarcane Production and Recovery
No. of
Cane Crushed
Sugar Made
Mills
Tonnes
Tonnes
71
41,062,268
3,548,953
71
42,994,911
3,530,931
69
28,982,711
2,414,746
65
29,408,879
2,466,788
69
36,708,638
3,197,745
71
41,786,689
3,652,745
71
43,661,378
3,997,010
71
32,101,739
2,922,126
74
30,090,632
2,588,176
77
40,483,977
3,516,218
Recovery
8.64%
8.21%
8.33%
8.39%
8.71%
8.74%
9.15%
9.10%
8.59%
8.69%
Source: Pakistan Sugar Mills Association Annual Report: 2007.
This table is showing an increasing trend in terms of sugarcane crushed and sugar
made except for years 2004-05 and 2005-06. During these two years Pakistan sugar
industry faced the crisis due to decline in area under cultivation which causes decline in
production and yield. Otherwise number of mills increased during this period.
After getting an overview of the sugar industry, we develop the methodology for
estimating productivity growth of sugar industry in Pakistan by examining this issue at
firm level.
3. METHODOLOGY
Total factor productivity growth and its sources are estimated using Data
Envelopment Analysis approach. Malmquist productivity growth indices are calculated
for twenty sugar firms and also for sugar industry. The Malmquist Productivity Index
also includes the sources of productivity growth for these firms.
3.1. Malmquist TFP Index
The Data Envelopment Analysis (DEA) methodology was initiated by Charnes, et
al. (1978) who built on the frontier concept started by Farell (1957). The methodology
used in this paper is based on the work of Fare, et al. (1994) and Coelli, et al. (1998) and
Raheman, et al. (2008). The DEA-Malmquist Index has been used to calculate the total
factor productivity growth of sugar firms listed at Karachi stock exchange where each
firm in the sugar industry is a Decision Making Unit (DMU).
This Malmquist productivity index can be decomposed into efficiency change,
technical change and total factor productivity growth. TFPG is geometric mean of
efficiency change and technical change. We have used the DEAP software developed by
Coelli (1996) to compute these indices. Following Fare, et al. (1994), the Malmquist
output-orientated TFP change index between periods s(the base period) and period t (the
subsequent period) is calculated as follows:
Raheman, Qayyum, and Afza
926
1
d s yt , xt
d t yt , xt 2
X 0t
m 0 y s , x s , y t x t 0s
d 0 y s , x s
d 0 y s , x s
…
…
(1)
In the above equation, d 05 ( y t , x t ) represents the distance from the period t
observation to the period s technology, y represents output and x represents input. Like
the DEA specification, each of the distance functions is calculated as a linear program.
While interpreting the Malmquist index, when mo is greater than 1 this indicates that the
TFP index has grown between periods t and s while mo less than 1 indicates that TFP has
declined. This productivity index can also be written in the following way.
1
m0 y s , x s , y t xt
d 0t
d 0s
yt , xt d 0s yt , xt X d 0s y s , x s 2
y s , x s d 0t yt , xt d 0t y s , x s
…
(2)
By re-expressing the Malmquist index in this way we have derived the following
components. The ratio outside the bracket measures the change in the output-oriented
measure of technical efficiency between period s and t. The other part of Equation 2
measures the technical change which is measured as a geometric mean in the shift in the
production technology between two periods evaluated at xt and xs.
In the above model efficiency change (catching up effect) and a technical change
(frontier effect) as measured by shift in a frontier over the same period. In this
methodology, we will use the output oriented analysis because most of the firms and
sectors have their objectives to maximise output in the form of revenue or profit.
3.2. Variables
We have applied the Data Envelopment Analysis (DEA) approach to the
revenue producing firms by converting the financial performance measures to the
firm’s technical efficiency equivalents. Ee have followed the methodology of
Raheman, et al. (2008) which is also based on Feroz, et al. (2003) and Wang
(2006), who have converted the financial performance measures to the firm’s
technical efficiency equivalent using DuPont Model.1 The DuPont model is a
technique for analysing a firm’s profitability using traditional performance
management tools. For enabling this, DuPont model integrates income statement
elements with balance sheet.
This process of measuring financial performance indicators can be converted into
output and input variables. Where, sales revenue can be used as output variable while
cost of goods sold, operating expenses, total assets and shareholder’s equity as input
variables. In this way long term resources total assets and equity and short term resources
cost of goods sold and operating expenses are used to produce output in the form of sales
revenue.
1
The Dupont formula and discussion regarding conversion of financial performance measures to firm’s
technical efficiency equivalents can be seen in Raheman, et al. (2008).
Efficiency Dynamics of Sugar Industry of Pakistan
927
3.3. Data
There are 38 sugar firms listed in the sugar and allied sector on Karachi stock
exchange. We have used the data only for those sugar firms which have performed the
operations and are among the listed firms on the Karachi Stock Exchange during the
study period 1998 to 2007. Furthermore, only those firms are included in the analysis
which have their shareholder’s equity positive because of the consideration of the
imitates of Data Envelopment Analysis Programme (DEAP) and their annual reports
(financial statements) are available for all the ten years. Hence, finally 20 firms are
selected for the analysis. Malmquist productivity Index has been used to calculate the
Total Factor Productivity Growth and its sources for these twenty sugar firms.
4. RESULTS AND DISCUSSION
The data of twenty sugar firms is used to construct a grand frontier using TFP
Index technique where each firm is compared to the frontier. We have calculated
Malmquist total factor productivity Index which shows TFP growth, efficiency change,
technical change, pure technical efficiency and scale change component for all the sugar
firms in the sample.
4.1. Total Factor Productivity Growth in Sugar Sector
Malmquist Index of firm means for efficiency change, technical change, pure
efficiency change, scale efficiency change and TFP growth are presented in Table 4.
Sugar industry experienced an overall negative TFP growth of –0.1 percent during 1998–
2007 which is insignificant. It means that during the study period there is no substantial
Table 4
Malmquist Index of Firm Means (1998–2007)
No. Firm
1 Adam Sugar Mills Limited
2 Al Abass Sugar Mills Limited
3 Al Noor Sugar Mills Limited
4 Chashma Sugar Mils Limited
5 Dewan Sugar Mills Limited
6 Faran Sugar Mills Limited
7 Habib Sugar Mills Limited
8 Haseeb Waqas Sugar Mills Limited
9 Husein Sugar Mills Limited
10 JDW Sugar Mills Limited
11 Kohinoor Sugar Mills Limited
12 Mirpurkhas Sugar Mills Limited
13 Noon Sugar Mills Limited
14 Sanghar Sugar Mills Limited
15 Shahtaj Sugar Mills Limited
16 Shakarganj Mills Limited
17 Sind Abadgar Sugar Mills Limited
18 Tandlianwala Sugar Mills Limited
19 The Frontier Sugar Mills and Distillery Limited
20 The Thal Industries Corporation Limited
Mean Sugar Sector
TE
Change
0.967
0.996
1.000
1.000
0.987
1.000
1.000
0.983
1.001
1.000
0.979
0.998
0.991
1.011
1.000
1.002
1.000
1.000
0.910
1.015
0.992
Tech.
Change
1.021
1.008
0.996
0.993
1.007
0.980
1.012
1.005
0.999
0.999
1.001
1.058
0.999
1.008
0.999
1.112
1.022
1.008
0.998
0.937
1.008
PE
Change
0.978
0.999
1.000
1.000
1.000
1.000
1.000
0.987
0.998
1.000
0.981
0.995
0.989
1.007
1.000
1.000
1.000
1.000
1.000
1.000
0.997
SE
Change
0.988
0.997
1.000
1.000
0.987
1.000
1.000
0.996
1.003
1.000
0.998
1.002
1.002
1.004
1.000
1.002
1.000
1.000
0.910
1.015
0.995
TFP
Change
0.987
1.004
0.996
0.993
0.993
0.980
1.012
0.988
0.999
0.999
0.980
1.056
0.990
1.019
0.999
1.114
1.022
1.008
0.908
0.951
0.999
Raheman, Qayyum, and Afza
928
increase or decrease in the total factor productivity growth. The analysis of sugar mills
revealed that TFP growth increased for seven out of twenty mills. The decline in
technical efficiency by 0.8 percent is offset by a same percentage increase in the technical
change which resulted in insignificant overall TFP growth. The technical change in 11
out of 20 firms is more than 1. Pure efficiency change and scale efficiency change results
in technical efficiency change. In case of pure efficiency change, it is one or more than
one in most of the firms but overall the pure efficiency of sugar industry declined by 0.7
percent while for scale efficiency change, value close to unity shows that most of the
firms are operating at optimum scale but again the scale efficiency of sugar industry
declined by 0.5 percent. Therefore, both scale efficiency and pure technical efficiency
have contributed to the decline in efficiency change.
In the above table, the comparison of total factor productivity change in different
firms shows that Shakarganj Mills Limited on average has the highest growth in TFP
(11.4 percent) during 1998 to 2007, followed by the Mirpurkhas Sugar Mills Limited that
has (5.6 percent) total factor productivity growth. The worst performer in terms of total
factor productivity growth is the Frontier Sugar Mills and Distillery Limited and the Thal
Industries Corporation Limited. Total factor productivity of these two mills decreased on
average by –9.2 percent and –4.9 percent respectively.
The results presented in Table 5 show that TFP growth has been volatile with little
apparent trend. The changes in TFP growth closely follow changes in technical progress
with changes in technical efficiency. The years 2002 and 1999 appear to be the years
where the total factor productivity growth was the highest at 5.3 percent and 5.2 percent
respectively. During years 2001 and 2007, the TFP growth is lowest at 4.7 percent and
4.4 percent respectively. If we analyse the efficiency change over period, it indicates that
during year 2003 the efficiency increased by 3.9 percent while it decreased by –5.9
percent during 2006. On the other hand the technological change increased by 8.7 percent
during year 2002 where the TFP growth is also maximum. Similarly technical change is
negative in the similar years where TFP growth was negative i.e. year 2001 and 2007.
Table 5
Year
Malmquist Index of Yearly Means of All Sugar Firm (1998-2007)
TE Change
Tech. Change PE Change
SE Change
TFP Change
1999
0.998
1.054
0.994
1.005
1.052
2000
0.957
1.036
0.970
0.986
0.991
2001
1.005
0.948
1.016
0.989
0.953
2002
0.969
1.087
0.965
1.004
1.053
2003
1.039
0.999
1.023
1.016
1.038
2004
1.024
0.960
1.015
1.009
0.983
2005
0.985
1.026
0.990
0.995
1.011
2006
0.941
1.022
0.985
0.956
0.962
2007
1.010
0.947
1.014
0.996
0.956
Mean
0.992
1.008
0.997
0.995
0.999
Efficiency Dynamics of Sugar Industry of Pakistan
929
These above results show an overall picture of TFP growth, efficiency change and
technical change for the sugar industry. For firm level analysis, these measures of
productivity need to be analysed at firm level during period 1998 to 2007.
4.2. Total Factor Productivity Growth
Yearly comparative results of TFP growth for individual firms during 1998–2007
are presented in Table 6 which provides a complete understanding about the performance
of these sugar firms.
During first year of analysis, The Thal Industries Corporation Limited performed
best among all the firms with TFP growth 24.2 percent followed by The Frontier Sugar
and Distillery Limited where the productivity increased by 19.9 percent. Habib sugar mill
is the worst performer with decline in TFP growth by –6.6 percent. This year was also the
most favourable for sugar industry where the TFP of 15 out of 20 firms increased and
TFP for sugar industry increased by 5.2 percent. During year 2000, the total factor
productivity of 10 out of 20 firms increased with the Husein sugar mills limited has the
highest TFP growth of 9.6 percent. In the next year 2001, the TFP declined for thirteen
sugar mills and the Chashma sugar mill was the worst performer in terms of TFP growth
which declined by 25.2 percent and the TFP declined by 4.7 percent for the overall sugar
industry which is the worst performance for the overall sugar industry during the study
period. The next three years 2002, 2003 and 2004 were relatively better years for the
sugar firms where the TFP increased for 12 out of 20 firms in all the three years.
Mirpurkhas sugar mill was the best performer during year 2002 while Faran sugar mill
was the best performer during year 2003 and Chashma sugar mill during 2004. TFP
growth for the sugar industry increased during 2002 and 2003 while declined during
2004. Shakarganj sugar mill played a leading role in total factor productivity growth with
highest (best performance) 76.6 percent during year 2005. Year 2006 was suitable for
nine sugar mills in terms of total factor productivity with highest TFP growth for Dewan
sugar mill at 35.9 percent. In this year the TFP for the sugar industry declined by 3.8
percent. Year 2006-07 was a crucial year for the sugar industry where the productivity
change for fourteen out of twenty firms declined and the TFP for the sugar industry
declined by 4.4 percent. In this year the best performer was the Chashma sugar mill with
a growth of 23 percent in total factor productivity. These results serve to show that firmlevel results can display a great deal of variations.
In terms of total factor productivity change, Shakarganj sugar mill has relatively
more stable results. In this firm TFP change in seven out of nine years is greater than
unity. Due to this reason, this firm topped in ranking in terms of total factor productivity.
As discussed earlier year 2006-07 was the most crucial year for most of the firms where
TFP declined for fourteen firms in the sample. Excluding this year from the analysis, the
overall TFP growth for the sugar industry would increase to 0.53 percent which is now
–0.1 percent including year 2007. The Frontier sugar mill is the worst performer in terms
of TFP growth followed by the Thal industries corporation limited which has negative
TFP growth for six out of nine years.
Two sources of total factor productivity named technical efficiency change and
technical change are presented in the next section.
Table 6
Comparative Total Factor Productivity Change in all Sugar Firms During (1998–2007)
Sector
1999
2000
2001
2002
2003
2004
2005
Adam Sugar Mills Limited
1.101 0.914 1.277 1.082 0.916 0.976 1.020
Al Abass Sugar Mills Limited
1.046 0.952 1.056 0.894 1.128 1.087 0.882
Al Noor Sugar Mills Limited
1.022 1.005 0.947 0.944 1.032 0.990 1.051
Chashma Sugar Mils Limited
1.118 0.984 0.748 1.199 0.769 1.222 0.966
Dewan Sugar Mills Limited
1.030 0.988 0.995 0.818 1.141 1.062 0.967
Faran Sugar Mills Limited
1.034 1.070 1.045 0.768 1.668 0.591 0.892
Habib Sugar Mills Limited
0.934 1.020 0.965 0.925 1.063 1.135 0.996
Haseeb Waqas Sugar Mills Limited
0.992 1.046 0.885 1.138 1.019 1.001 1.067
Husein Sugar Mills Limited
1.053 1.096 0.770 1.667 0.794 1.013 0.999
JDW Sugar Mills Limited
1.069 0.892 1.284 0.792 1.072 0.998 1.036
Kohinoor Sugar Mills Limited
1.079 1.023 0.832 1.154 0.888 1.082 1.040
Mirpurkhas Sugar Mills Limited
0.976 1.025 1.003 1.812 0.943 0.879 1.175
Noon Sugar Mills Limited
1.059 1.054 0.935 1.079 0.963 1.007 0.996
Sanghar Sugar Mills Limited
1.066 0.976 1.051 0.716 1.249 1.131 0.963
Shahtaj Sugar Mills Limited
1.062 0.893 0.966 1.164 0.921 0.985 0.964
Shakarganj Mills Limited
1.020 0.961 1.080 1.024 1.085 1.203 1.766
Sindh Abadgar Sugar Mills Limited
0.986 1.016 0.974 0.929 1.121 0.871 1.015
Tandlianwala Sugar Mills Limited
0.995 0.978 0.941 1.184 0.840 1.015 1.047
The Frontier Sugar Mills and Distillery Limited
1.199 1.005 0.762 1.146 1.124 1.202 0.855
The Thal Industries Corporation Limited
1.242 0.944 0.762 1.210 1.368 0.565 0.787
1.052 0.991 0.953 1.053 1.038 0.983 1.011
Mean
2006
0.865
1.016
1.012
0.852
1.091
1.359
1.125
0.822
0.874
0.994
0.979
1.064
0.851
1.213
0.979
0.984
1.298
1.013
0.387
0.970
0.962
2007
0.811
1.000
0.967
1.230
0.888
0.789
0.971
0.964
0.956
0.923
0.804
0.864
0.984
0.919
1.082
1.070
1.039
1.089
0.892
0.999
0.956
Mean
0.987
1.004
0.996
0.993
0.993
0.980
1.012
0.988
0.999
0.999
0.980
1.056
0.990
1.019
0.999
1.114
1.022
1.008
0.908
0.951
0.999
Efficiency Dynamics of Sugar Industry of Pakistan
931
4.3. Technical Efficiency Growth
Firm-wise technical efficiency movement is presented in Table 7 for
understanding the contribution made by technical efficiency in the productivity growth of
sugar firms.
The results in general suggest that technical efficiency is an important factor in
dampening the total factor productivity growth of the sugar industry. The average
efficiency change for eight mills is less than one while for nine firms it is equal to one
which means there is no change in the managerial efficiency during study period for these
firms. During year 1999, the technical efficiency change for eight firms is less than one
and Habib sugar mills the worst performer with a decline in efficiency change by –8.7
percent. In this year six mills did not show any change in their efficiency. Managerial
efficiency further declined in year 2000, where 14 mills have their efficiency change in
negative and three mills have no change in efficiency. During this year AL Abass sugar
mill was the worst performer with a decline in efficiency change by 13.8 percent. Year
2001 was relatively better for the sugar industry in terms of managerial efficiency where
thirteen mills were having their efficiency change equal to or more than one. The
efficiency change for sugar industry declined during years 2002, 2005 and 2006 by –3.1
percent, –1.5 percent and –5.9 percent respectively. The maximum decline in the
managerial efficiency for the sugar industry was during year 2006. On the other side
efficiency change increased during years 2003, 2004 and 2007.
The firm level changes in managerial efficiency shows that many mills remain
static as their efficiency change remain equal to one in most of the years. These firms
include Faran sugar mills, JDW sugar mills and Shahtaj sugar mills limited. Thal
industries corporation limited which is on top in ranking according to managerial
efficiency based on aggregate efficiency change is also more stable firm where efficiency
change is more than one in seven out of nine years.
4.4. Technology Adoption
The comparative technical change for twenty sugar firms during period 1998 to
2007 is presented in Table 8. Generally, the technical change can be seen in eleven firms
where Shakarganj mills limited at the top with 11.2 percent change followed by the
Mirpurkhas sugar mills limited with 5.8 percent. In year 1999, the comparative technical
change shows positive change where all mills have their technical change more than one
and Thal industries corporation top in ranking followed by the Chashma sugar mills
limited. In this year technical change increased by 5.4 percent for the overall sugar
industry. Year 2000 was also better in terms of technical change where it was positive for
sixteen mills and sugar industry overall recorded a 3.6 percent technical progress. In this
year Haseeb Waqas sugar mills limited was the best performer where technical change
increased by 13 percent while Shahtaj sugar mills limited was the worst performer with
decline in technical progress by 10.7 percent. Years 2001 and 2007 were the worst in
terms of technical progress where it declined by 5.2 percent and 5.3 percent respectively.
In these years only three to four mills were having their technical change in positive. The
best year according to technical progress was the year 2002 where the technical change
increased by 8.7 percent for the overall sugar industry and eighteen firms have their
technical change above one. In this year Mirpurkhas sugar mill was highest in ranking
Table 7
Comparative Efficiency (Managerial Efficiency) Change in all Sugar Firms during (1998–2007)
Sector
1999
2000
2001
2002
2003
2004
2005
2006
Adam Sugar Mills Limited
1.000 0.981 1.019 1.000 0.966 0.990 0.996 0.884
Al Abass Sugar Mills Limited
0.992 0.862 1.169 0.839 1.158 1.030 0.857 1.057
Al Noor Sugar Mills Limited
1.000 0.985 0.995 0.891 1.076 0.951 1.025 1.052
Chashma Sugar Mils Limited
1.000 1.000 0.886 1.128 0.814 1.227 0.945 0.862
Dewan Sugar Mills Limited
0.984 0.948 1.071 0.789 1.115 1.026 0.936 1.085
Faran Sugar Mills limited
1.000 1.000 1.000 0.825 1.212 1.000 1.000 1.000
Habib Sugar Mills Limited
0.913 0.920 1.083 0.862 1.094 1.115 0.917 1.085
Haseeb Waqas Sugar Mills Limited
0.954 0.925 1.035 1.063 0.987 1.005 1.041 0.820
Husein Sugar Mills Limited
1.016 1.071 0.819 1.221 0.956 1.038 0.967 0.953
JDW Sugar Mills Limited
1.000 0.882 1.134 0.951 1.052 1.000 1.000 1.000
Kohinoor Sugar Mills Limited
1.038 0.947 0.930 1.082 0.913 1.075 0.984 1.011
Mirpurkhas Sugar Mills Limited
0.919 1.012 1.053 1.072 0.952 0.843 1.136 1.097
Noon Sugar Mills Limited
1.029 0.961 1.049 1.000 1.000 1.000 0.964 0.874
Sanghar Sugar Mills Limited
1.042 0.935 1.116 0.664 1.292 1.127 0.983 1.066
Shahtaj Sugar Mills Limited
1.000 1.000 0.977 1.023 1.000 1.000 1.000 1.000
Shakarganj Mills limited
0.965 0.912 1.155 0.968 1.033 1.000 1.000 1.000
Sind Abadgar Sugar Mills Limited
0.944 0.992 1.025 0.878 1.143 0.924 1.122 1.000
Tandlianwala Sugar Mills Limited
0.961 0.923 1.011 1.115 0.870 0.987 1.008 1.025
The Frontier Sugar Mills and Distillery Limited
1.135 0.890 0.870 1.074 1.156 1.213 0.871 0.396
The Thal Industries Corporation Limited
1.097 1.013 0.810 1.136 1.119 1.000 1.000 0.914
0.998 0.957 1.005 0.969 1.039 1.024 0.985 0.941
Mean
2007
0.877
1.061
1.034
1.229
0.968
1.000
1.051
1.043
1.012
1.000
0.857
0.933
1.052
1.000
1.000
1.000
1.000
1.127
0.935
1.092
1.01
Mean
0.967
0.996
1.000
1.000
0.987
1.000
1.000
0.983
1.001
1.000
0.979
0.998
0.991
1.011
1.000
1.002
1.000
1.000
0.910
1.015
0.992
Table 8
Comparative Technical Change in all Sugar Firms during (1998-2007)
Sector
Adam Sugar Mills Limited
Al Abass Sugar Mills Limited
Al Noor Sugar Mills Limited
Chashma Sugar Mils Limited
Dewan Sugar Mills Limited
Faran Sugar Mills Limited
Habib Sugar Mills Limited
Haseeb Waqas Sugar Mills Limited
Husein Sugar Mills Limited
JDW Sugar Mills Limited
Kohinoor Sugar Mills Limited
Mirpurkhas Sugar Mills Limited
Noon Sugar Mills Limited
Sanghar Sugar Mills Limited
Shahtaj Sugar Mills Limited
Shakarganj Mills limited
Sind Abadgar Sugar Mills Limited
Tandlianwala Sugar Mills Limited
The Frontier Sugar Mills and Distillery Limited
The Thal Industries Corporation Limited
Mean
1999
1.101
1.054
1.022
1.118
1.047
1.034
1.024
1.039
1.036
1.069
1.039
1.062
1.030
1.024
1.062
1.057
1.044
1.036
1.056
1.132
1.054
2000
0.932
1.104
1.021
0.984
1.042
1.070
1.109
1.130
1.023
1.012
1.080
1.013
1.097
1.043
0.893
1.054
1.024
1.059
1.129
0.931
1.036
2001
1.252
0.903
0.952
0.844
0.929
1.045
0.891
0.855
0.941
1.132
0.895
0.953
0.892
0.941
0.988
0.935
0.950
0.931
0.876
0.940
0.948
2002
1.082
1.066
1.059
1.063
1.038
0.931
1.073
1.071
1.365
0.833
1.067
1.691
1.079
1.079
1.137
1.058
1.058
1.063
1.067
1.065
1.087
2003
0.948
0.974
0.959
0.945
1.023
1.376
0.972
1.032
0.831
1.019
0.972
0.990
0.963
0.966
0.921
1.050
0.981
0.965
0.972
1.223
0.999
2004
0.986
1.056
1.041
0.996
1.035
0.591
1.018
0.996
0.976
0.998
1.007
1.043
1.007
1.004
0.985
1.203
0.942
1.028
0.991
0.565
0.96
2005
1.024
1.030
1.026
1.022
1.033
0.892
1.086
1.025
1.033
1.036
1.056
1.035
1.033
0.980
0.964
1.766
0.904
1.040
0.982
0.787
1.026
2006
0.979
0.961
0.962
0.988
1.005
1.359
1.037
1.003
0.917
0.994
0.967
0.970
0.974
1.138
0.979
0.984
1.298
0.989
0.976
1.061
1.022
2007
0.925
0.943
0.935
1.001
0.917
0.789
0.924
0.925
0.945
0.923
0.938
0.926
0.936
0.919
1.082
1.070
1.039
0.966
0.954
0.916
0.947
Mean
1.021
1.008
0.996
0.993
1.007
0.980
1.012
1.005
0.999
0.999
1.001
1.058
0.999
1.008
0.999
1.112
1.022
1.008
0.998
0.937
1.008
3
934
Raheman, Qayyum, and Afza
with a progress of 69 percent followed by Husein sugar mills limited with 36.5 percent.
JDW sugar mill was the worst performer where the technical change declined by 16.7
percent. Shakarganj sugar mill was the leading one during year 2004 and 2005, where the
technical progress increased by 20.3 percent and 76.6 percent. Further, increase of 76.6
percent is the maximum increase in any mill in a year during period 1998 to 2007.
The ranking of all sugar firms in terms of total factor productivity growth,
technical efficiency change and technical change is presented in Table 9. According to
the ranking, Shakarganj mills limited is top in ranking according to TFP growth and
technical change while at number three according to efficiency change. Mirpurkhas sugar
mill is although next in ranking according to TFP growth and technical change but at
number thirteen according to managerial efficiency change. Similar type of ranking is for
the Sind Abadgar sugar mill which is at third in ranking as per TFP growth and technical
change but at number eleven according to efficiency change. This indicates that technical
change is the major factor which affects the total factor productivity growth for the sugar
firms. The Frontier sugar mills and distillery limited is the laggard firm according to
efficiency change and technical change. The other laggard firm is The Thal Industries
Corporation limited according to TFP growth and technical change but highest in ranking
according to efficiency change. This also indicates that for sugar firms technical change
is the major source of total factor productivity.
5. CONCLUSION
Research on productivity growth is very important because economic growth
cannot be sustainable without improvement in the Total Factor Productivity. From a
policy point of view, the assessment of TFP growth is important as it serves as a guide for
resource allocation and investment decisions. In this paper we have applied Data
Envelopment Analysis approach for estimating TFP growth, efficiency change and
technological progress in Pakistan’s sugar industry using data for twenty sugar firms
from 1998 to 2007. Productivity Growth is estimated using Malmquist productivity
index. The decomposition of TFP growth also helped us to identify improvement in
efficiency and contribution of technological progress and innovation to productivity
growth in sugar industry. Most of the studies of productivity growth efficiency which are
based on panel data discuss the estimates of overall sample or sector. However, we have
presented the estimated TFP growth, efficiency change and technical change at each firm
level and for each year during 1998 to 2007 which shows that these estimates varies
widely at firm level during the data period.
The empirical estimates on the performance of sugar industry yielded several
striking results. The Malmquist TFP results reflect a tormenting picture for the sugar
industry. Overall sugar industry improved technological progress by 0.8 percent while
managerial efficiency change declined by a same percentage. Due this reason the overall
TFP growth during 1998–2007 remained almost static with a decline of 0.1 percent.
The results of TFP growth and its components also presents divergent trend in the
individual years for the overall sugar industry. The efficiency change declined for nine
sugar firms and remained equal to one for nine sugar firms during period 1998 to 2007,
while the technical change is positive for eleven out of twenty sugar firms. Therefore, the
result shows static TFP Growth. It suggests that sugar industry is lacking in terms of
Table 9
Ranking of Sugar Firms Based on Malmquist TFP and its Components
Ranking
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Industry
Shakarganj Mills Limited
Mirpurkhas Sugar Mills Limited
Sind Abadgar Sugar Mills Limited
Sanghar Sugar Mills Limited
Habib Sugar Mills Limited
Tandlianwala Sugar Mills Limited
Al Abass Sugar Mills Limited
Husein Sugar Mills Limited
JDW Sugar Mills Limited
Shahtaj Sugar Mills Limited
Al Noor Sugar Mills Limited
Chashma Sugar Mils Limited
Dewan Sugar Mills Limited
Noon Sugar Mills Limited
Haseeb Waqas Sugar Mills Limited
Adam Sugar Mills Limited
Faran Sugar Mills Limited
Kohinoor Sugar Mills Limited
The Thal Industries Corporation Limited
The Frontier Sugar Mills & Distillery Limited
TFP
Change
1.114
1.056
1.022
1.019
1.012
1.008
1.004
0.999
0.999
0.999
0.996
0.993
0.993
0.990
0.988
0.987
0.980
0.980
0.951
0.908
Industry
The Thal Industries Corporation Limited
Sanghar Sugar Mills Limited
Shakarganj Mills Limited
Husein Sugar Mills Limited
Al Noor Sugar Mills Limited
Chashma Sugar Mils Limited
Faran Sugar Mills Limited
Habib Sugar Mills Limited
JDW Sugar Mills Limited
Shahtaj Sugar Mills Limited
Sind Abadgar Sugar Mills Limited
Tandlianwala Sugar Mills limited
Mirpurkhas Sugar Mills Limited
Al Abass Sugar Mills Limited
Noon Sugar Mills Limited
Dewan Sugar Mills Limited
Haseeb Waqas Sugar Mills Limited
Kohinoor Sugar Mills Limited
Adam Sugar Mills Limited
The Frontier Sugar Mills & Distillery Limited
TE
Change
1.015
1.011
1.002
1.001
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.998
0.996
0.991
0.987
0.983
0.979
0.967
0.910
Industry
Shakarganj Mills Limited
Mirpurkhas Sugar Mills Limited
Sind Abadgar Sugar Mills Limited
Adam Sugar Mills Limited
Habib Sugar Mills Limited
Al Abass Sugar Mills Limited
Sanghar Sugar Mills Limited
Tandlianwala Sugar Mills Limited
Dewan Sugar Mills Limited
Haseeb Waqas Sugar Mills Limited
Kohinoor Sugar Mills Limited
Husein Sugar Mills Limited
JDW Sugar Mills Limited
Noon Sugar Mills Limited
Shahtaj Sugar Mills Limited
The Frontier Sugar Mills & Distillery Limited
Al Noor Sugar Mills Limited
Chashma Sugar Mils Limited
Faran Sugar Mills Limited
The Thal Industries Corporation Limited
Tech.
Change
1.112
1.058
1.022
1.021
1.012
1.008
1.008
1.008
1.007
1.005
1.001
0.999
0.999
0.999
0.999
0.998
0.996
0.993
0.980
0.937
936
Raheman, Qayyum, and Afza
managerial efficiency which could be explained by a general reduction in the quality of
managerial decision-making among the best practice firms. Regardless of the reason for
this decline, it has potentially serious implications for the longer-term financial viability
of these sugar firms. Except few firms which are relatively stable include Shakarganj
mills limited and Al Abass sugar mills limited, all sugar firms have a mix trend over
1998–2007 which affects the productivity and ranking of firms.
The pattern of TFP growth tends to be driven more by technical change (or
technical progress) rather than improvements in technical efficiency. Shakarganj mills
limited has highest technical change and also better performance in terms of managerial
efficiency change which lead it top in ranking in terms of TFP. This firm has also
performed better in terms of stability over the period 1998 to 2007, where the TFP
increased for seven out of nine years. The major source for Mirpurkhas sugar mill is the
technical change, which lead it to next in ranking. The technical change is also a main
source of relatively better performance for Sind Abadgar sugar mill and Habib sugar mill
while Sanghar sugar mill is also among the top ranking firms where the main sources is
managerial efficiency. The Frontier sugar mill is among the worst performers in terms of
productivity over 1998 to 2007 where the problem lies in managerial efficiency and also
non adoption of new technologies. Similarly, The Thal Industries is also one of the
laggard firms in terms of TFP where the major source is non adoption of new
technologies although top in ranking in terms of efficiency change.
The research suggests that the Pakistani sugar industry is facing serious
productivity growth problems where no increase is recorded in total factor productivity
during 1998 to 2007. Therefore, this industry must increase total factor productivity in
most of the firms and efforts must be made to provide a stable pattern to the productivity
growth. The improvement is needed in both technical efficiency and technological
progress in the sugar industry. For increasing technical efficiency, efforts are needed to
improve the quality of inputs like capital and labour. On the other side the management
aspect cannot be ignored and it is also very important in terms of capital. Furthermore,
the research and development (R & D) activities can also play a vital role in bringing
technological progress. Although there is very little increase in the technical change but
for further considerable increase in the productivity, efforts could be made to increase the
research and development (R & D) activities in this industry. Therefore, firms in the
sugar industry need greater investment in (R & D) activities and adoption of new
technologies.
REFERENCES
Afzal, M. (2006) Some New Production Measurement Methods for Large-scale
Manufacturing Sector of Pakistan. PhD Dissertation (unpublished) submitted to
National College of Business Administration and Economics. Agricultural Economics
78:2, 331–8.
Ali, S. (2004) Total Factor Productivity Growth in Pakistan’s Agriculture: 1960–1996.
The Pakistan Development Review 43: 4, 493–513.
Burki, A. A. and K. Mahmood ul Hassan (2005) Effects of Allocative Inefficiency on
Resource Allocation and Energy Substitution in Pakistan’s Manufacturing. Lahore
University of Management Sciences. (CMER Working Paper No. 04-30).
Efficiency Dynamics of Sugar Industry of Pakistan
937
Charnes, A., W. W. Cooper, and E. Rhodes (1978) Measuring the Efficiency of Decision
Making Units. European Journal of Operations Research 2, 429–444.
Coelli, T. (1996) A Guide to DEAP Version 2.1: A Data Envelopment Analysis
(Computer) Programme. Centre for Efficiency and Productivity Analysis. Armidale,
NSW, Department of Econometrics, University of New England, Australia. (Working
Paper 96/08).
Coelli, T. J., D. S. P. Rao, C. J. O’Donnell, and G. E. Battese (2005) An Introduction to
Efficiency and Productivity Analysis (2nd ed.). New York: Springer.
Coelli, T., D. S. P. Rao, and G. E. Battase (1998) An Introduction to Efficiency and
Productivity Analysis. Boston: Kluwer Academic Publishers.
Coelli, T. J. (1995) Recent Developments in Frontier Modelling and Efficiency
Measurement. Australian Journal of Agricultural Economics 39, 219–45.
Fare, R., S. Grosskopf, M. Norris, and Z. Zhang (1994) Productivity Growth, Technical
Progress, and Efficiency Change in Industrialised Countries. The American Economic
Review 84, 66–83.
Farrell, M. J. (1957) The Measurement of Productive Efficiency. Journal of the Royal
Statistical Society 120, 253–81.
Feroz, E. H., S. Kim, and R. L. Raab (2003) Financial Statement Analysis: A Data
Envelopment Analysis Approach. Journal of Operational Research Society 54, 48–
58.
Jaforullah, M. and J. Whiteman (1999) Scale Efficiency in the New Zealand Dairy
Industry: A Non-parametric Approach. Australian Journal of Agricultural and
Resource Economics 43, 523–42.
Kalirajan, K. P. and R. T. Shand (1999) Frontier Production Functions and Technical
Efficiency Measures. Journal of Economic Surveys 13, 149–72.
Mahmood, T., E. Ghani, and M. Din (2007) Efficiency of Large Scale Manufacturing in
Pakistan: A Production Frontier Approach. Pakistan Institute of Development
Economics. (Working Paper 27).
Malmquist, S. (1953) Index Numbers and Indifference Curves. Trabajos de Estatistica 4,
1, 209–42.
Pakistan Sugar Mills Association of Pakistan (2007) Annual Report -2007.
Pakistan Sugar Mills Association of Pakistan (2008) Annual Report -2008.
Raheman, A., Talat A., Abdul Q. and A. B. Mahmood (2008) Estimating Total Factor
Productivity and Its Components: Evidence from Major Manufacturing Industries of
Pakistan. The Pakistan Development Review 47:4.
Seiford, L. M. (1996) Data Envelopment Analysis: The Evolution of the State of the Art.
Journal of Productivity Analysis 7, 99–137.
Squires, D. and C. Reid (2004) Using Malmquist Indices to Measure Changes in TFP of
Purse-Seine Vessels While Accounting for Changes in Capacity Utilisation, The
Resource Stock and the Environment. SCTB17 Forum Fisheries Agency. (Working
Paper, pp. 1–15).
Wang, J. C. (2006) Corporate Performance Efficiency Investigated by Data Envelopment
Analysis and Balanced Scorecard. Journal of American Academy of Business 9:2,
312–18.
Comments
The paper titled ‘Efficiency Dynamics of Sugar Industry of Pakistan’ is interesting
and analytical technique used in this paper is latest one. However, write up of this paper
needs some editing. For example, in abstract and introduction of this paper, it is stated
that total factor productivity (TFP) in sugar industry will be decomposed in 3 categories;
technical, scale and managerial. But in Table 4, Malmquist indices have been worked out
for technical efficiency change, technical change, production efficiency change, scale
efficiency change and TFP. Furthermore, only 3 of these indices have been discussed in
Sections 5.3 and 5.4.
In abstract of the paper it is mentioned that there are 81 sugar mills in Pakistan
whereas on page 4, the number changes to 84. Also subheading 5.1 is exactly same as
5.2 that should be avoided. Similarly, in Table 1 in ‘Overview of Sugar Industry’, sugar
yield in Pakistan is reported as 3.54 while its correct figure comes out 4.51. Column 4 in
Tables 2 and 3 of this section are not commented anywhere in the text. Furthermore, the
first sentence in paragraph 2 at page 5 states that area under sugarcane cultivation has
increased but data in Table 2 and the last sentence in first paragraph at page 6 do not
support it. Calculation of Malmquest indices on pages 8 and 9 is not properly explained.
I am sure that careful editing of this paper will improve its reading and worth.
M. Mazhar Iqbal
Quaid-i-Azam University,
Islamabad.