A stochastic frontier production function approach to Indian textile industry.
Manonmani, M.
India's Manufacturing Sector
Manufacturing sector is the backbone of any economy. It fuels
growth, productivity, employment, and strengthens agriculture and
service sectors. Astronomical growth in worldwide distribution systems
and IT, coupled with the opening of trade barriers, has led to
stupendous growth of global manufacturing networks, designed to take
advantage of low-waged yet efficient Indian work force. Though
agriculture has been the main pre-occupation of the bulk of the Indian
population, the founding fathers saw India becoming a prosperous and
modern state with a good industrial base. Programs were formulated to
build an adequate infrastructure for rapid industrialization. India is
fast emerging as a global manufacturing hub. Be it automobiles or
computer hardware, consumer durables or engineering products, all are
being manufactured by multinationals in India. India's cheap,
skilled manpower is attracting a number of companies, planning diverse
industries, making India a global manufacturing powerhouse (Adhikary
Maniklal & Ritwik Mazumder, 2009). Indian Textile industry is
presently one of the largest and most important industries in the Indian
economy in terms of output, foreign exchange earnings and employment
generation. The diversity and richness of Indian culture reflects in its
textile products in terms of variety, colours and patterns it offers to
the world. India has a diverse and rich textile tradition. Contemporary
Indian textiles not only reflect the country's rich and splendid
past, but also cater to the demands of the modern day. In fact, today
India is one of the world's leading manufacturers of man-made
textiles. India is the world's second largest producer of textiles
and clothing after China. The textile and clothing industry forms a
major part of India's manufacturing sector and has contributed
enormously to the country's impressive economic development in
recent years. Furthermore, India has a huge and growing domestic market
which is expected to be worth US$140 billion in 2020 as the population
increases in size and consumers become wealthier. This huge growth could
provide significant opportunities for foreign exporters to India and
potential foreign investors in the country, as well as for the Indian
textile and clothing industry itself. This report looks at the
development of the textile and clothing industry in India and its size
and structure as well as textile and clothing production and
consumption.
With the introduction of economic reforms since July, 1991, many
changes have come upon industrial structure in India. Introduction of
various reforms and gradual liberalisation of both domestic and
international trade marked the beginning of the end of the earlier
regulatory regime and recognition of the urgency on the part of the
Indian industries to become efficient so as to be able to withstand
successfully the pressure of foreign competition. Over the years several
measures have been taken by the government to help domestic industries
achieve efficiency. These include not only the fiscal and financial
measures such as rationalisation of excise duties, liberalisation of tax
laws and rates, reduction in interest rates and so on, but also such
physical measures as those meant to remove infrastructural constraints
in power, transport and telecommunications sectors.
India's Textile Industry
Indian textile industry largely depends on textiles manufacturing
and export. It also plays a major role in the economy of the country.
India earns about 27 per cent of its total foreign exchange through
textile exports. Further, it contributes about 14 per cent to industrial
production, 4 per cent to the gross domestic product (GDP), and 17 per
cent to the country's export earnings. The sector is the second
largest provider of employment after agriculture. It not only generates
jobs in its own industry, but also opens up scope for other ancillary
sectors. The industry currently generates employment to more than 35
million people.
The textiles sector has witnessed a spurt in investment during the
last five years. The main engine of investment has been the Technology
Upgradation Fund Scheme (TUFS). The increased investment will help to
upgrade technology, strengthen infrastructural facilities and also
increase the installation of additional spindles and looms. Besides, it
will provide a fillip to the garment, technical textiles and processing
segments of textiles industry, which have great potential for value
addition and employment generation. The industry attracted foreign
direct investments (FDI) worth [??] 61.36 crores (US$ 11.02 million) in
the month of May 2012 as compared to 24.75 crores (US$ 4.44 million)
during the corresponding month in 2011.The Indian textile industry saw
three mergers and acquisitions (M&A) deals worth US$ 455 million in
the month of July 2012.
The Indian textile sector is also well placed globally, in terms of
installed capacity of spinning machinery, it ranks second after china
while in weaving its ranks first in plain handlooms and fourth in the
shuttle looms. India has around 40 million spindles (23% of world) and
0.5 million rotors (6% of world capacity). India has 1.8 million shuttle
looms (45% of world capacity), 0.02 million shuttle less looms (3% of
world capacity) and 3.90 million handlooms (85 % of world capacity).
Organised sector contributes to almost 100% of spinning but hardly 5 %
of weaving of fabric. Cotton products are the stronghold of India. The
Indian textile industry is also globally well placed, in teams of
installed capacity of spinning machinery, it ranks second after china,
while in weaving it ranks first in plain handlooms and fourth in the
shuttle looms. In terms of all these positive aspects of textile sector
in India, it is imperative to study the efficiency aspect of this
sector.
Production Efficiency
The efficiency term describes the maximum outputs attainable from
utilizing the available inputs. A production is efficient if it cannot
improve any of its inputs or outputs without worsening some of its other
inputs or outputs. Efficiency can be increased by minimizing inputs
while holding output constant or by maximizing output while holding
inputs constant or a combination of both may increase efficiency (Alias
Radam et al, 2010). Productive efficiency (also known as technical
efficiency) is defined as a situation in which the most production is
achieved from the resources available to the producer. It occurs when
the economy is utilizing all of its resources efficiently, producing
most output from least input.
Productive efficiency can be determined by estimating the
best-practice production frontier and individual industries gives the
measure of inefficiency. In view of the growing high production costs
productive efficiency and profitability will become increasingly
important determinants of the future of Indian industries. In addition
to developing and adopting new production technology, the industries can
maintain their economic viability by improving efficiency of existing
operation with a given level of technology. In other words the
industry's total costs can be reduced and the total output can be
increased by making better use of available inputs and technology.
This study examined the industry level efficiency so as to identify
the sources where improvement can be made. The study will provide vital
information to help individual industries in using their resources more
efficiently and to assist the industries in becoming more competitive
and maintaining its long term survival. The determination of frontier
technology and knowledge of productive efficiency and its relationship
with firm size can provide important insights into future Indian
industries. Furthermore, the relationship between efficiency levels and
various industry--specific factors can provide useful policy--relevant
information. A comparison of industry's frontier or "best
practice" function and its average practice function will produce
useful information about possible future structural adjustments for the
industries.
Methodology
Net Value Added (NVA) was taken as output. Labor input (L)
consisted of both workers directly involved in production and persons
other than workers like supervisors, technicians, managers, clerks and
similar type of employees. The invested capital (K) was taken into
account as capital. Wages included remuneration paid to workers. The
basic data source of the study was Annual Survey of Industries (ASI)
published by Central Statistical Organization (CSO), Government of India
covering the period from 1991-92 to 2009-10. All the referred variables
were normalized by applying Gross Domestic Product (GDP) deflator. The
GDP at current and constant prices were obtained by referring to
Economic Survey, published by Government of India, Ministry of Finance
and Economic Division, Delhi.
Stochastic Frontier Production Function
The stochastic frontier production function as proposed by Battese
and Coelli (1992) is defined as :
[Y.sub.i] = f([X.sub.i],[beta])[[epsilon].sup.ei]
Where Yi, is the output vector for the ith firm, [X.sub.i] is a
vector of inputs, [beta] is a vector of parameter and e is an error
term. In this model, a production frontier defines output as a function
of a given set of inputs, together with technical inefficiency effects.
Furthermore, this model allows some observations to lie above the
production function, which makes the model less vulnerable to the
influence of outliers than with deterministic frontier models.
The stochastic frontier is also called composed error model,
because it postulates the error term [[epsilon].sub.i] as two
independent error components:
[[epsilon].sub.i] = [v.sub.i] + [u.sub.i]
When a symmetric component is normally distributed, [v.sub.i] ~ (N,
[[sigma].sup.2.sub.v]), represents any stochastic factors that is beyond
the firm's control affecting the ability to produce on the frontier
such as luck or weather. It can also account for measurement error in Y
or minor omitted variables. The asymmetric component, in this case
distributed as a halfnormal [u.sub.i]~(N,[[sigma].sup.2.sub.v]),
[u.sub.i] > 0 can be interpreted as pure technical inefficiency. This
component has also been interpreted as an unobservable or latent
variable; usually representing managerial ability.
The parameters of v and u can be estimated by maximizing the
following log-likelihood function:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Where,
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
F = the standard normal distribution function
N = Number of observations
Given the assumptions on the distribution of v and u, Jondrow et
al. (1982) showed that the conditional mean of u given [epsilon] is
equal to
E([u.sub.i]|[[epsilon].sub.1])
[[s.sub.u][s.sub.v]/s][f/([epsilon]ils)/1-f([epsilon]ils)] -
[[epsilon]il/s]]
Where f and F are the standard normal density and distribution
functions evaluated at [[epsilon].sub.i][lambda]/[sigma]. Measures of
technical efficiency (TE) for each firm can be calculated
[TE.sub.i] = exp (-E[[u.sub.i]/[epsilon]]) so that 0 [less than or
equal to] TE [less than or equal to] 1
The Cobb-Douglas stochastic frontier production function in
logarithm form is as follows:
In VAi = in [[beta].sub.0] + [[beta].sub.1] in C + [[beta].sub.2]
in [L.sub.i] + [[beta].sub.3] in [E.sub.i] + [[epsilon].sub.i]
Where VA represents Net value added per year. Independent variables
are: C (capital) and L (number of laborers). Parameters [[beta].sub.0]
denotes the technical efficiency level and [[beta].sub.1] is
elasticities of the various inputs with respect to the output level.
Results
The productive efficiency of the industry was calculated by
applying the stochastic frontier production approach. The results show
the summary statistics of the variables, maximum likelihood estimates
and technical efficiency for Indian textile industry for the reference
period. As for primary investigation the summary statistics results of
the selected variables of the industry are presented in the Table 1.
Mean values of input variables indicate that the industry's
main factors of production were both capital and labor since there were
not much differences in their mean values. The magnitude of variability
(C.V) also substantiated this point since the coefficients are less for
both the inputs.
Table 2 shows the maximum likelihood estimates in the context of
its productive efficiency.
The maximum likelihood estimates for productive efficiency show
that in the single output case, parameter of capital input is positive
and statistically significant. Hence capital is the main input factor
for this industry as its value was higher than labor. The coefficients
of [[sigma].sup.2] and [lambda] were statistically significant though
their signs differ. It reveals that the estimated levels of output
considerably differ from their potential levels due to factors, which
are within the control of the industry. The estimated value of [lambda]
indicated the absence[right arrow] of efficiency gap that exists between
the actual and potential level of performance which is mainly due to
technical efficiency of the industry. The statistically not significant
co-efficient of [mu] term indicated that it followed a normal
distribution and positive and statistically significant. Coefficient of
[eta] indicated that efficiency increases in getting production
overtime. The summation of the elasticities of factors of production
indicated return to scale of 1.8419. The value of return to scale
greater than unity suggested that conditions of increasing returns to
scale prevail. One percent increase in inputs (labor and capital)
resulted in an increase 1.84 percent in output level for the stochastic
frontier.
Table 3 presents the year-wise technical efficiency during the
reference period from 1991-92 to 2009-10 in the industry.
In terms of technical efficiency, the industry recorded an average
efficiency of 0.941 (94.1 percent). The table also revealed that the
technical efficiency has not shown any decline but showed mixed trend.
The average technical inefficiency was observed as 0.020, which was
negligible. The magnitude of variability was 4.25 in the growth of
technical efficiency of the industry during the reference period. In
other words it varied at the rate of 4.25 percent per annum in its
growth.
Conclusion
Based on the results it is recognised that the textile industry in
India had performed well in terms of efficiency though the efficiency
scores were mixed in nature over the reference period. For future
development, productive efficiency and technical change in industries
specifically in textile industry have been prominent issues in
discussions on the regional diversity of output and employment growth in
the industrial sector in developing countries like India. Without
improving technology and efficiency, however, the growth performance of
the manufacturing sector as of the other sectors of an economy is likely
to be limited.
References
Alias Radam & Ismail Latiff (2000), "Technical Efficiency
and Productivity Performance of Malaysian Manufacturing
Industries", The Asian Economic Review, 42(2): 249-62.
Adhikary Maniklal & Ritwik Mazumder (2009), "Economic
Reforms & Manufacturing Sector Productivity in West Bengal", in
Economic Reforms and Productivity Changes in Selected Indian Industries,
Abhijeet Publication, New Delhi.
James Jondrow, Knox Lovell, C.A & Ivan S. Mathew (1982)
"On the Estimation of Technical Inefficiency in the Stochastic
Frontier Production Function Model", Journal of Economics,
19:233:38.
M.Manonmani is Professor in Economics, Avinashilingam Institute for
Homescience & Higher Education for Women University, Coimbatore
641043. E-Mail: anomyil@yahoo.com
Table 1 Summary Statistics of the Variables
Variable Mean Std Minimum Maximum C.V
Deviation
Net Value Added (NVA) 3.2454 0.0895 2.50 3.67 2.77
Invested Capital (K) 3.1152 0.0717 2.99 3.40 2.30
Number of workers (L) 3.0393 0.0993 2.97 3.18 3.27
Source: Calculations based on ASI Data
Note: C.V-Co--efficient of variation
Table 2 Maximum Likelihood Estimates of Stochastic
Frontier Production Function
Variable Co-efficient Std-error t-ratio
Intercept -2.2175 1.6384 -0.9658
Ln K 1.4852 *** 0.6383 1.990
LnL 0.3567 1.6094 0.1799
[[sigma].sup.2] 0.0006 *** 0.0009 1.8959
r 0 9999 *** 0.00006 1.8003
L 0.0267 0.0496 0.4438
M 0.1118 ** 0.0459 2.7969
Source : Calculations based on ASI Data
Note: **--Significant at 5 % level
***--Significant at i0 % level
Table-3 Technical Efficiency Scores
S. Year Efficiency
No Scores
1 1991-92 0.889
2 1992-93 0.910
3 1993-94 0.919
4 1994-95 0.899
5 1995-96 0.965
6 1996-97 0.990
7 1997-98 0.888
8 1998-99 0.879
9 1999-00 0.935
10 2000-01 0.922
11 2001-02 0.893
12 2002-03 0.962
13 2003-04 0.980
14 2004-05 0.999
15 2005-06 0.985
16 2006-07 0.989
17 2007-08 0.939
18 2008-09 0.967
19 2009-10 0.963
20 Mean 0.941
21 Standard deviation 0.040
23 Co-efficient of 4.25
variation(Standard
Deviation/Mean)
24 Average inefficiency score 0.020
Source: calculations based on ASI data
