TOTAL FACTOR PRODUCTIVITY AND EFFICIENCY IN INDIAN MEAT PROCESSING INDUSTRY

Jabir Ali1

Abstract
This study evaluates the performance of meat processing industry and role of technology in acceleration of growth. Productivity and efficiency improvement of the processing industry is the key for sustainable growth. Malmquist TFP index is used for measuring productivity change in Indian meat processing industry. Malmquist productivity index is defined as the ratio of two output distance functions !aves et al." #$%&'. Input oriented variable returns to scale ()*' +,- model has been used for measuring technical and scale efficiency. The input.output variables used include capital" labour" raw material consumed" fuel consumed and gross value of output. Malmquist TFP index and efficiency scores have been obtained by using +,-P software version &.#' developed by !oelli #$$/'. -lthough the level of meat processing is extremely low" it has been increasing. The growth in processed meat segment has been drastic during #$$0s #&.%1' as compared to #$%0s 2.21'. Most of this occurred due to input growth. The contribution of technology was negligible during #$%0s as well as #$$0s. 3n an average TFP grew at a rate of #.0# percent during #$%0.%# to #$$$.&000. The average technical efficiency score is estimated to be 0.4$ under !)* model and 0.$2 under ()* model. The efficiency indices values equal to unity imply that the industry is on frontier while values below unity imply that the industry is below the frontier or technically inefficient. 3n the other hand" average scale efficiency for the entire period is 0./5. There was considerable under utili6ation of input resources during #$%0s. 7evertheless" over time resource . utili6ation has improved perhaps due to rising market trends. This had significant positive impact on labour absorption as well as labour productivity. 8hile the capital investment in industry improved" capital productivity has remained stagnant.

JELClassification: L66, N55, O30, O47, Q1 Keywords: Technical Efficiency, TFP, Meat Processin , !E", #n$ia

-ssistant Professor" -griculture Management !entre -M!'" Indian Institute of Management" 9ucknow : &&/ 0#2 ,mail; <abirali=iiml.ac.in

TOTAL FACTOR PRODUCTIVITY AND EFFICIENCY IN INDIAN MEAT PROCESSING INDUSTRY

1. Intr !"cti n
Food processing is a sunrise industry of the Indian economy and has been identified as thrust area for development. Food processing sector covers a wide range of items like fruits and vegetables> meat and poultry> milk and milk products" alcoholic beverages" fisheries" plantation" grains" confectionery" chocolates and cocoa products" mineral water" high protein foods etc. ?ased on the basic raw material usage" food industry can broadly be claasified into plant based and animal based. Meat industry is one of the important segments of food processing industry in general and livestock@animal based industry in particular. India has immense potential for production" consumption and export of meat due to sufficient resources" available markets and huge livestock population. 3ver the last two decades" the value of meat output has been increasing at a rate of about / percent a year. )ising demand for meat has been the driving force behind it. ?etween #$%0 and &000" while per capita consumption of foodgrains increased by 5 percent" consumption of milk and meat increased by 40 percent and &4 percent respectively. In quantitative terms" per capita milk consumption increased from 50 kgs in #$%0 to // kgs in &000" and meat consumption increased from 5 kgs to 4 kgs during this period. Most of the meat output $/1' is consumed domestically" yet per capita meat consumption in India is much less compared to many developed AA kgs' and developing &A kgs' countries. The demand for meat is expected to grow faster with sustained economic growth" rising per capita incomes" strengthening urbani6ation trends and increasing awareness of the nutritive value of meat and meat products Bumar" #$$%> ?halla and Ca6ell" #$$% and +elgado" #$$$'. These opportunities can be capitailised for the benefit of producers as well as consumers and would largely be determined by the pace of development and diffusion of the technologies in processing of livestock based products Mishra" #$$4'.

&

The increase in demand has been accompanied by increase in production. Total meat production increased from &.A million tonnes in #$%0 to 5.A million tonnes in &000 with annual growth of 2.5# percent. The growth in meat production has largely been number driven" as yield growth is negligible in case of almost all the species. !attle" buffalo" goat" sheep" pigs and poultry are important meat species. 8hile goat" sheep" pig and poultry are exclusive meat animals" cattle and buffalo provide meat as an ad<unct to milk. -nimals slaughtered are of poor quality. The structure of meat production" however" is undergoing a gradual shift from ruminant to non.ruminant pig and poultry' meat production. The share of non.ruminant increased from #4 percent in #$%0 to &2 percent in #$$$. The growth of meat industry is constrained by a number of socio.cultural and economic factors at different levels of production" processing" handling and marketing. Meat production is mainly constrained due to lack of productivity augmenting technologies since the ma<or quantity of meat is being produced in unorgani6ed slaughterhouses" where resource utili6ation is very limited. These slaughterhouses are old" unhygienic and lack basic facilities like water" light" ventilation" drainage" waste disposal and effluent treatment which contribute to poor meat quality and low recovery of various by.products such as hides" blood" bonemeal" internal organs and trimmings. Therefore" technology is the key to improvement in growth and efficiency in meat processing sector. ,mpirical evidences on contribution of technology to growth of meat processing industry in India are scarce. Cowever" the evidences from food industry as a whole indicate varied contribution of technology to growth of food processing industry Doldar" #$%/> -hluwalia" #$$#> Mitra et al." #$$%> Mitra" #$$$> Doldar and Bumari" &00&> Trivedi et al." &00& and Pattnayak and Thangavelu" &002'.

#. Data an! M$t% ! l &'

The data on input and output related to registered@ organi6ed meat manufacturing units is compiled from -nnual *urvey of Industries published by the !entral *tatistical 3rganisation" Minstry of *tatistics and Programme Planning" Dovernment of India. The study evaluates the performance of meat processing industry and role of technology in acceleration of growth of this industry. Malmquist TFP index

is used to measure productivity change in Indian meat processing industry" which is defined as the ratio of two.output distance functions !aves et al." #$%&'. Input oriented variable returns to scale ()*' +,- model is used to measure technical and scale efficiency in Indian meat processing industry. The input.output variables used include capital" labour" raw material consumed" fuel consumed and gross value of output. Malmquist TFP index and efficiency scores are obtained by using +,-P software version &.#' developed by !oelli #$$/'. #.1. 2.1.1 Anal'tical A((r ac% Total Factor Productivity (TFP C!an"e The simplest indicators of productivity are partial productivity measures derived by dividing the output by relevant input. Most commonly used measures are labour productivity i.e. output@ labour ratio and capital productivity i.e. the output@ capital ratio. Cowever" these ratios can be misleading as improvement in productivity cannot be attributed to any single factor input individually. Therefore" an integrated model for measuring productivity is desirable which considers all the factor inputs in aggregate and explains interacting economic relationship. In the present study" Malmquist TFP index is used to measure productivity change in Indian meat processing industry. Malmquist productivity index is defined as the ratio of two output distance functions !aves et al." #$%&'. +istance functions are functional representations of multiple.output and multiple.input technology which requires data only on input and output quantities. Malmquist index has several advantages over Fisher and Tranquist index as it does not require assumptions regarding market structure and economic behaviour. Malmquist TFP index decomposes productivity change into technical change and technical efficiency change. Fare et al #$$5' specifies an output based Malmquist productivity change index as;

$ t ( % , y ) $ t +# ( % , y ) M 0 ( y t +# , %t +# , y t , %t ) = 0 t t +# t +# 0 t +# t +# t +# $ 0 ( %t , y t ) $ 0 ( %t , y t )

## &

#'

This represents the productivity of the production point ( %t +# , yt +# ) relative to the production point ( %t , yt ) . - value greater than one will indicate positive TFP growth from period t to tE#. This index is the geometric mean of two output based Malmquist TFP indices. The input thus employs distance functions from two different
t t periods or technologies" $ 0 ( %t , yt ) and $ 0 ( %t +# , yt +# ) > and two pairs of input.output

vectors"

( %t , yt )

t and ( %t +# , yt +# ) . !aves et al. #$%&' assume that $ 0 ( %t , yt ) F

t $0 ( %t +# , yt +# ) implying that own.period observations are technically efficient in the

sense of Farrell #$4A'. The Malmquist index can be decompose into two components namely technical efficiency change ,FF!C' and technical change T,!C!C'" defined as;
t ( %t , yt ) $ t ( % , y ) $ t ( %t +# , yt +# ) $0 M 0 ( yt +# , %t +# , yt , %t ) = 0 t t +# t +# t0 +# t +# $ 0 ( %t , yt ) $ 0 ( %t +# , yt +# ) $ 0 ( %t , yt ) ## &

&'

where the ratio of outside the square bracket measures the change in relative efficiency between t and tE#. The geometric mean of the two ratios inside the square bracket captures the shift in technology between the two periods. These may be given as;
t ( %t +# , yt +# ) $0 EFF&' = t ( %t , y t ) $0

2'

t t $0 ( (% ,y ) %t +# , y t +# ) $0 TE&'&' = t +# t +# t t $ 0 ( %t +# , y t +# ) $ 0 ( %t , y t )

## &

5'

The Malmquist index can further be explained in diagrammatic form Figure #'. In the figure" *t and *tE# denote the technologies in period t and tE# respectively. The input.output vectors

( %t , yt )

and ( %t +# , yt +# ) are feasible in their own periods" but

( %t +# , yt +# )

t t does not belong to *t. In the figure" $ 0 ( %t +# , yt +# ) F3a@3b and $ 0 ( %t , yt )

F3d@3e. Thus the term outside the square bracket in equation & equals;

EFF&' =

Oa Oe O( O$

4'

Fi&"r$ 1) Mal*+"ist O"t("t , -as$! TFP In!$. K b KtE# Fa c f e KtFd 3 Jt JtE# J *t *tE#

Source: Hossain and Bhuyan (2002)

*imilarly" the term inside the square bracket in equation & is given as;
Oa O( O$ Of TE&'&' = Oc Oa Oe O$
## &

O( Of = Oc Oe

## &

/'

The last expression shows that the ratio of term inside the square bracket in equation / measures shift in technology at input levels x t and xtE# respectively. This indicates technical change as the geometric mean of two shifts" which is of the same form as Fisher Ideal Index Cossain and ?huyan" &00&'. 2.1.2 Tec!nical and $cale Efficiency The non.parametric approach introduced as +ata ,nvelopment -nalysis +,-' by !harnes" !ooper and )hodes #$A%' is a method of measuring efficiency of +ecision Making Gnits +MGs'@ firms through linear programming techniques" which HenvelopI observed input : output vectors as tightly as possible ?oussofiane et al." #$$#'. The +,- is a methodology directed to frontiers rather than central tendencies *eiford and Thrall" #$$0'. The +,- is also capable of handling multiple inputs and outputs at the same time. This study employs input oriented variable returns to scale

()*' +,- to measure technical and scale efficiency in Indian meat processing industry using +,-P computer software version &.#' developed by !oelli #$$/'. The input.output variables used include capital" labour" raw material consumed" fuel consumed and gross value of output.

The original model developed by !harnes" !ooper and )hodes !!) model' was applicable when technologies were characteri6ed by constant returns to scale !)*'. It is assumed that there are H7I !M)s with B inputs and * outputs on each +MG. That is" produces

!M) ( * = 0"#"......." N ) consumes %

*

*i

amount of input i and

*r

amount of output r" where % *i 0 and y *r 0 . The mathematical

programming involves the selection of optimal weights that maximi6e the ob<ective function of the ratio of outputs to inputs for each +MG being evaluated. The constant returns to scale !)*' +,- model is only appropriate when firm is operating at an optimal scale !oelli et al." #$$%'.

In the method originally proposed by !harnes" !ooper and )hodes #$A%' relative efficiency of the !M)s can be measured by input oriented +,- model as;

min "
"

sub<ect to

+ +
r* ,

r0

i0

-
i*

0
A'

8here>

i*

F the amount of the ith input at !M)*,

 F the input technical efficiency T,' score"

!M)* The value of gives efficiency score for a particular !M)" which satisfies
0 # . The !M)s for which L# are inefficient while for F# are on frontiers

r*

F the amount of rth output from !M)*,

F vector of weight which defines the linear combination of the peers of

and hence efficient.

Imperfect competition may cause a !M) not to operate at optimal scale !oelli" #$$/'. ?anker" !harnes and !ooper #$%5' extended the !!) model to account for technologies that show variable returns to scale ()*'. The ?anker" !harnes and !ooper ?!!' model can be developed by adding the convexity constraint to the constant returns to scale !)*' linear programming problem. i.e.

* =#

=#

%'

The !)* technical efficiency scores can be decomposed into pure technical efficiency and scale efficiency. This can be done by applying both !)* and ()* +,on the same model. The difference between !!) model and ?!! model can be illustrated as follows. 8e shall assume one input and one output situation. The !)* and ()* frontiers have been drawn in Figure &.

Fi&"r$ #) CRS/ VRS an! Scal$ E00ici$nc' K !)*

s r c q m b

()*

3
Source: Coelli, 1996

Xt

The inefficient !M) is represented by the point P. Gnder input orientation model" the technical inefficiency of !M) HPI is ./ in !)* and (/ in ()*. The difference between these two measures is expressed as scale inefficiency *,'. In ratio form" technical efficiency in !)* is 0.10/ and in ()* it is 0(10/. *cale efficiency is 0.10(. Further"

TE

&42

= TE 342

2E

$'

Thus" technical efficiency T,' obtained from !)* can be decomposed into HpureI technical efficiency and scale efficiency. The point such as HcI on the frontier is scale efficient. The concept of scale efficiency constitutes two technologies i.e. constant return to scale !)*' and variable return to scale ()*'. The ()* technology in the figure & is presented by the single input . single output production function. The scale efficiency measure corresponding to input Jt is given by;
O0 @ Or ' Os = O0 @ Os ' Or

*cale ,fficiency =

#0'

1. In!ian M$at In!"str') An O2$r2i$3

The structure of meat industry is highly unorgani6ed and only a meagre quantity of meat is processed for value addition. Most of the meat produced in the country comes from traditional slaughterhouses. There are about ten thousand slaughterhouses in the country of which /0 percent are unregistered. Most of these slaughterhouses have poor hygiene and sanitation facilities resulting in poor meat quality and environmental degradation. The organi6ed sector of meat industry constitutes very few modern meat processing units in the country. The country has $ modern abattoirs and #A# meat processing units licensed under Meat Products 3rder. -nnual *urvey of Industries -*I' data shows that only 2A meat processing units are registered under Factories -ct. - few modern pork processing plants are also coming up in the country. Poultry processing is still in its infancy. There are only seven modern integrated poultry processing plants. Cowever" there are a good number of small poultry processing units engaged in production of poultry meat products.
Tabl$ 1) Ma4 r *an"0act"r$rs 0 (r c$ss$! *$at (r !"cts in In!ia C *(an' Frigo )efico -llana 9imited" Bulaba" Mumbai Frigo )efico -llana 9imited" Bulaba" Mumbai Cind Industries 9imited" 7ew +elhi Cind Industries 9imited" 7ew +elhi -lkabeer ,xports 9imited" Mumbai -lkabeer ,xports 9imited" Mumbai P.M.9. Industries" !handigarh G.P. Pashudhan Gdyog 7igam 9td. Gttar Pradesh G.P. Pashudhan Gdyog 7igam 9td. Gttar Pradesh -.P. Meat M Poultry !orporation" Cyderabad Pigpo" Norbagh Market" 7ew +elhi M-F!3" Mumbai )anchi ?acon Factory" )anchi )a<asthan Meat and 8ool Marketing Federation" -lwar (enkateshwara Catcheries" Pune +eeNay" ?angalore Ma4 r Pr !"cts Fro6en buffalo meat !anned meat Fro6en buffalo meat !hilled@Fro6en sheep and Doat meat Fro6en buffalo meat !hilled@Fro6en sheep and Doat meat Fro6en buffalo meat Pork and other meat products !anned meat manufactures Pork and other meat products Pork and other meat products Pork and other meat products Pork and other meat products !anned meat Manufactures Poultry products Poultry products -ran!s -llana -llana *ibaco" ,atco *ibaco" ,atco -lkabeer -lkabeer PM9 !+F !+F -P*MP! Pigpo M-F!3

(enkyIs Food

S "rc$) Ministry of Food Processing Industries" D3I

The performance of Indian meat indurstry has been measured by expotential growth rate in terms of production" value of output from meat" domestic counsumption

#0

and export earnings during last two decades. Meat production in India has increased significantly over the last two decades at a rate of 2.5# percent a year. The growth in contributions from different species" however" varied widely. Maximum growth occurred in poultry meat #0.05 percent' followed by pork 5.05 percent'" beef and veal &.$% percent'" buffalo meat &.$% percent'" goat meat &.2# percent'" mutton and lamb #.A4 percent' and processed meat #.5# percent'. The growth in meat production was higher during #$%0s as compared to #$$0s. Drowth in total meat production has improved slightly in recent years mainly because of acceleration in growth of contributions from buffalo" sheep and poultry.

In India" slaughter of cattle is banned in ma<ority of the states except in the states of Berala" 8est ?engal and some 7ortheastern states'. !attle are considered to be sacred by the ma<ority Cindu population. ?uffaloes are not sub<ected to any religious sensitivity and are slaughtered at a variety of weights and ages. 8ith increasing mechani6ation of agriculture sector" the demand for draught animals are going down drastically resulting in surplus of male cattle and buffaloes" which are slaughtered at birth or at low ages or when they becomes weak and unproductive. The slaughter of these young male calves is a waste of productive national wealth. Drowth in meat production is largely due to growth in number of animals slaughtered" as yield growth was negligible in case of almost all the species.
Tabl$ #) P$r0 r*anc$ 0 *$at s$ct r in In!ia/ 1567,1555 89: S($ci$s
1

Pr !"cti n 8777; t n$s:

?eef and (eal &.$% &.44O ?uffalo Meat &.$% .. Doat Meat &.2# &.02OO Mutton and 9amb #.A4 .. Pig meat 5.05 2.$0 Poultry Meat #0.05 #0.5# Processed Meat #.5# .. Meat" Total 2.5# 2.55 O -lso includes buffalo meat OO -lso includes mutton and lamb S "rc$) #F-3 Production M Trade Kearbook various issues' & 7ational -ccount *tatistics various issues'" !*3" Ministry of *tatistics and Program Implementation" Dovernment of India

Ani*al Sla"&%t$r$! 8N s.: #.%# &.A4 #.A4 &.0A 2.%5 %.$# .. ..

Val"$ 0 "t("t 8Rs.: 2.2#O .. 5.%#OO .. /.45 4.A$ 2./2 4.#2

C ns"*(ti n 8777; t ns:

E.( rt Earnin&s 8US<: #5.&5 /.%# $.0% &.#A #5.&& .#.0A .. /.54

##

The annual growth in meat output is greater than the growth in livestock and agriculture sector as a whole during the last two decades. The ma<or opportunity for growth of livestock sector lies in the increasing demand for animal food products. Per capita consumption of meat and meat products in India is extremely low. Cowever" demand for meat is expected to grow faster with sustained economic growth" rising per capita incomes" strengthening urbani6ation trends and increasing awareness of the nutritive value of meat and meat products ?halla and Ca6ell" #$$%> Bumar" #$$% and +elgado" #$$$'. It is not only the income factor which affects its consumption level in the country" but also the social and religious factors which play a crucial role in consumption of meat and meat products. Meat is not considered to be a regular food item in ma<ority of diet. !onsumption of meat in India is seasonal in nature and is influenced by various socio.religious practices and varies across regions. In some cases" religious practices prohibit meat consumption for specified periods and in others" celebrations and festivals lead to increase in meat demand 9andes et al" &005'. These seasonal swings in demand contribute to fluctuation in monthly market prices of meat. ,xport of meat and meat products showed a promising performance over the last two decades. It experienced significant annual growth of /./4 percent over the period of #$%0.$$. The export of meat products witnessed significant annual growth during the post.liberali6ed period and increased at the rate of A.$& percent during #$$0. $$. Cowever" specie.wise growth in export varies over the period. -nnual growth of bovine meat export was 4./& percent during #$%0.$0" which increased to %./# percent during #$$0.$$. 3n the other hand" growth in ovine meat exports was 2.52 percent during #$%0.$0" which substantially declined and has become negative to the extent of &.5# percent during #$$0.$$. The scope for export of sheep" goat and poultry meat is constrained by high domestic demand and prices )avishankar and ?irthal" #$$$'. The export of pig meat has also increased significantly but poultry meat export has declined. The annual growth in value of export from meat and meat products increased from 5.52 percent during #$%0s to %.40 percent during #$$0s" which is higher than quantity growth during the same period. It shows that India received higher export prices of meat during #$$0s. The share of different species in total meat export in value terms also reveal similar trends as of volume.

#&

=. Pr !"cti2it' an! E00ici$nc' in M$at Pr c$ssin& In!"str'

The technological change in production system is the key to overall productivity growth Cossain and ?huyan" &000'. The productivity and efficiency in Indian meat processing industry is measured using a non.parametric +ata ,nvelopment -nalysis +,-'. Malmquist TFP index is used for measuring productivity change in Indian meat processing industry. The TFP index interprets the change in output that is not accounted for by change in input" but is due to change in efficiency or technology or returns to scale or a combination of these three factors. Thus" changes in TFP can be decomposed into three components; i' technological change" ii' changes in technical efficiency" and iii' changes in scale efficiency. Input oriented variable returns to scale ()*' model has been used for measuring technical and scale efficiencies in meat processing industry. =.1. T$c%nical C $00ici$nt an! Pr !"cti2it' Rati s Table 2 shows important characteristics of meat processing industry in India in terms of gross value added" gross capital assets and employment. These indicators show that the meat industry is not operating at large scale on an average. The per unit gross value added increased at a moderate rate during #$%0s" which picked up considerably in early #$$0s. The trends in gross fixed assets also show similar trends. The growth in employment rate was negative during #$%0s. This significantly increased during #$$0s. Though meat processing operations at different stages are largely handled by labour" per unit employment is not very large.
Tabl$ 1) Tr$n!s in i*( rtant c%aract$ristics 0 *$at in!"str' 8($r "nit: Y$ar Gr ss Val"$ Ca(ital 8Rs. in A!!$! 8Rs. in la>%s: la>%s: &A ##/ &$ #05 #5 $2 &5 #22 #4 #2A #2 %A && #25 &A #22 &4 ##/ 2/ #2/ &$ #0& 2% ##% /4 #%2 T tal E* l"*$nts 8Rs. in la>%s: #5 #& #& #5 #2 #0 #2 #5 #$ #4 #& #& #2 E*(l '*$nt 8N"*b$r: #4A #25 #&% #5$ #22 #0$ ##0 #0% ##5 #&A ##5 $$ ##5

#$%0.%# #$%#.%& #$%&.%2 #$%2.%5 #$%5.%4 #$%4.%/ #$%/.%A #$%A.%% #$%%.%$ #$%$.$0 #$$0.$# #$$#.$& #$$&.$2

#2

#$$2.$5 2A #$$ #$$5.$4 #04 &&% #$$4.$/ ##0 &A5 #$$/.$A #5$ 2A4 #$$A.$% ### 2%4 #$$%.$$ ##$ 4%/ #$$$.00 %A 40A Gr 3t% Rat$ 89: #$%0.$0 2.&A 0.$$ #$$0.$$ #&.%0 #$.04 #$%0.$$ ##.20 %.5A S "rc$) !alculated from -*I data various issues'

#2 #/ #% &5 && &0 &2 #.5$ %.%% 2.05

$& #25 #20 #A4 #4& #A/ #%2 .&.A% A.0A 0.$&

The analysis of partial productivity shows that labour productivity in meat industry has increased at an annual rate of #.% percent during #$%0s and /.& percent during #$$0s. 3n the other hand capital productivity during this period declined. shift in capital to labour productivity is also visible during the last two decades Table 5'. !hange in partial productivities is largely influenced by improvement in capital intensity i.e. capital@ labour ratio I!I!I" #$$5'. The capital intensity in meat processing industry increased drastically during #$$0s due to substantial increase in investment level Figure 2'.
Tabl$ =) Pr !"cti2it' Rati s in M$at Pr c$ssin& In!"str' Y$ar Partial Pr !"cti2it' Ca(ital Int$nsit' Lab "r Ca(ital Ca(ital?lab "r rati #$%0.%# #.$A 0.&2 %./& #$%4.%/ #.25 0.#4 %.%# #$$0.$# &.5A 0.&% %.%# #$$#.$& 2.#% 0.2& $.%$ #$$&.$2 4.05 0.2/ #5.#% #$$2.$5 &.$0 0.#% #4./$ #$$5.$4 /.4% 0.5/ #5.22 #$$4.$/ /.&5 0.50 #4.44 #$$/.$A /.&5 0.50 #4.A# #$$A.$% 4.0# 0.&$ #A.55 #$$%.$$ 4.A$ 0.&0 &%./# #$$$.00 2.AA 0.#A &#.$% Gr 3t% Rat$ 89: #$%0.$0 #.A% &.&% .0.4# #$$0.$$ /.&0 .2.$A #0.#A S "rc$) !alculated from -*I data various issues'

#5

Fi&"r$ 1) Tr$n!s in (atrial (r !"cti2it' rati s

24.00

20.00

&4.00 y F /.444%e0.0452x )& F 0./%2#

&0.00 )atio #4.00

#0.00

y F #.#5&e0.0%&/x )& F 0.//55 y F 0.#A5&e0.0&%2x )& F 0.#$&A #$$/.$A #$%0.%# #$%#.%& #$%&.%2 #$%2.%5 #$%5.%4 #$%4.%/ #$%/.%A #$%A.%% #$%%.%$ #$%$.$0 #$$0.$# #$$#.$& #$$&.$2 #$$2.$5 #$$5.$4 #$$4.$/ #$$A.$% #$$%.$$ #$$$.00

4.00

0.00

9abour

!apital

!apital intensity

=.#.

C st Str"ct"r$ 0 M$at Pr c$ssin& In!"str' The cost of meat processing is influenced by the level of product processing and

packaging. The production cost of meat processing industry is broadly categori6ed into four inputs namely capital" labour" raw material and fuel. The share of various inputs show that raw material accounts for the ma<or share of more than %0 percent followed by fuel 4.51'" labour 5.51' and capital 2.#1'.

Tabl$ @) C st c *( siti n 0 *$at (r c$ssin& in!"str' at c nstant (ric$s Y$ar Ra3 Mat$rial C ns"*$! #$%0.%& 2.# $.5 %#.% #$$0.$& &.0 ##./ A$.& #$$A.$$ 2.# 5.5 %A.& S "rc$) !alculated from -*I data various issues' Ca(ital Lab "r F"$l C ns"*$! 4.A A.& 4.5

=.1.

Pr !"cti2it' an! E00ici$nc' Figure 5 shows increasing trends in indices of aggregate inputs and output. The

growth in aggregate input use shows faster trends than output growth. Indices of capital and labour in meat processing industry show increasing trends Figure 4'. The growth in capital related investment was stable during #$$%0s. This substantially increased during #$$0s. The increase in capital may be attributed to the rapid increase in investment after the economic reform. The payment for labour input also increased during this period but the rate of growth was lower.
#4

Fi&"r$ =) In("t an! O"t("t tr$n!s in *$at (r c$ssin& in!"str'

%.0 A.0 /.0 4.0 Index 5.0 2.0 &.0 #.0 0.0 #$%#.%& #$%5.%4 #$%/.%A #$%$.$0 #$$#.$& #$$5.$4 #$$/.$A #$$%.$$ #$$%.$$ #$$$.00 #$$$.00 #$%0.%# #$%&.%2 #$%2.%5 #$%4.%/ #$%A.%% #$%%.%$ #$$0.$# #$$&.$2 #$$2.$5 #$$4.$/ #$$A.$% #$$A.$% y F 0./044e0.0$$%x )& F 0.%0&

y F 0./&4&e0.#0&$x )& F 0.%405

3utput

Input

Fi&"r$ @) Tr$n!s in ca(ital an! lab "r "s$ in *$at (r c$ssin& in!"str'
%.0 A.0 /.0 4.0 Index 5.0 2.0 &.0 #.0 0.0 #$%0.%# #$%#.%& #$%%.%$ #$%$.$0 #$$0.$# #$$#.$& #$$&.$2 #$$2.$5 #$$5.$4 #$%&.%2 #$%2.%5 #$%5.%4 #$%4.%/ #$%/.%A #$%A.%% #$$4.$/ #$$/.$A y F 0.%##Ae0.05%2x )& F 0.A%/$ y F 0./#A5e0.#0&/x )& F 0.%5$2

!apital

9abour

The growth in capital use in meat processing industry is twice of the growth in labour use. The indices of raw material used and fuel consumed are also showing increasing trends Figure /'. The use of raw material which accounts for more than %0 percent of aggregate inputs is increasing at an annual rate of $ percent as compared to around ## percent of fuel consumption. The trends in capital and fuel consumption show that the industry is moving towards mechani6ation and becoming more and more capital intensive over time.

#/

Fi&"r$ A) Tr$n!s in ra3 *at$rials an! 0"$ls c ns"*$! in *$at in!"str'

$.0 %.0 A.0 /.0 4.0 5.0 2.0 &.0 #.0 0.0 #$%#.%& #$%&.%2 #$%2.%5 #$%4.%/ #$%/.%A #$%A.%% #$%%.%$ #$$0.$# #$$#.$& #$$&.$2 #$$2.$5 #$$4.$/ #$$/.$A #$$A.$% #$%0.%# #$%5.%4 #$%$.$0 #$$5.$4 #$$%.$$ #$$$.00 y F 0.$&55e0.0$x )& F 0.%%22 y F 0.4%$2e0.#0$Ax )& F 0.A$A2

Index

)aw Materials

Fuels

Malmquist TFP index measures productivity change over period t to period tE#. This output.based index explains the change in productivity level in given level of inputs. The TFPD in a firm occurs either due to technological progress" i.e." due to shift in the production function or due to efficiency improvements in the firm. productivity value index larger than one indicates a productivity improvement and a value less than one indicates productivity decline. Total Factor Productivity TFP' change over the year is given in table /. Meat processing in India shows mixed trends in TFP change. 3n an average TFP grew at a rate of #.0# percent during #$%0.%# to #$$$.&000. +uring #$$0s" there was no growth in TFP and output growth was mainly due to rapid growth in input use. The deviation in TFP index from unity shows that in most of the years" there was positive change in TFP. ?ut its magnitude seems to be very small.

#A

Fi&"r$ B) TFP C%an&$ in In!ian M$at Pr c$ssin& In!"str'

#.50

#.&0

y F #.0#2&e.0.00#&x )& F 0.002

#.00

TFP !hange

0.%0

0./0

0.50

0.&0

0.00 #$%#.%& #$%&.%2 #$%4.%/ #$%/.%A #$%A.%% #$%$.$0 #$$0.$# #$$#.$& #$$5.$4 #$$4.$/ #$$/.$A #$$%.$$ #$$$.00 #$%2.%5 #$%5.%4 #$%%.%$ #$$&.$2 #$$2.$5 #$$A.$%

The performance of meat processing industry is measured in terms of technical and scale efficiency Table /'. !!) Model estimates efficiency relative to !)* technology. *ince !)* technology is scale neutral" it is implicitly assumed that all +MGs are operating at optimum scale of operation. 3n the other hand" ?!! Model measures efficiency under ()* technology and allows the possibility that inefficiency may be due to +MGs deviating from respective scale of operation as well as due to pure technical inefficiency. The values of efficiency indices equal to unity imply that the industry is on best.practice frontier while values below unity imply that the industry is below the frontier or technically inefficient.

Tabl$ A) E00ici$nc' sc r$ 0 r In!ian *$at (r c$ssin& in!"str' Kear #$%0.%# #$%4.%/ #$$0.$# #$$#.$& #$$&.$2 #$$2.$5 #$$5.$4 #$$4.$/ #$$/.$A #$$A.$% #$$%.$$ #$$$.00 !!) Model T,' 0./0 0.24 0.4A 0./A 0.A/ 0.5# #.00 #.00 #.00 0.%5 0.A5 0.4# ?!! Model Pure T,' #.00 #.00 #.00 #.00 0.$$ #.00 #.00 #.00 #.00 0.%4 0.A5 0./2 *, 0./0 0.24 0.4A 0./A 0.AA 0.5# #.00 #.00 #.00 0.$$ #.00 0.%# TFP .. #.#A #.0/ 0.$/ #.&2 0.$5 #.0& #.02 #.05 0.$A #.00 0.%5

#%

A2$ra&$ #$%0.$0 0.55 0.$5 0.5/ #.0# #$$0.$$ 0.A4 0.$& 0.%& #.00 #$%0.$$ 0.4$ 0.$2 0./5 #.0# S "rc$) !alculated from -*I data various issues' 7ote; T,FTechnical ,fficiency" *,F *cale ,fficiency and TFPFTotal Factor Productivity !hange

The average technical efficiency score is estimated to be 0.4$ under !)* model and 0.$2 under ()* model. The average scale efficiency for the entire period is 0./5. +uring #$%0.%# and #$$0.$#" the average efficiency under the !)* and ()* technologies was 0.55 and 0.$5 respectively. +uring #$$0.$# to #$$$.&000" efficiency under !)* model improved to 0.A4" whereas in case of ()* model the efficiency slightly declined to 0.$&. The scale efficiency also improved from 0.5/ to 0.%& during this period. The performance scores based on !)* model are equal to one during the years #$$5.$4" #$$4.$/ and #$$/.$A. -ll the other years recorded the efficiency score less than one indicating inefficient use of resources. Cowever" the efficiency score based on ()* model indicate that performance scores are equal to one during more number of years than the !)* model. Meat processing industry was scale inefficient during most of the years except #$$5.$4" #$$4.$/" #$$/.$A and #$$%.$$. *cale efficiency scores suggest si6eable deviation from scale of operation but it has approached to unity over time.

Fi&"r$ 6) Tr$n!s in $00ici$nc' sc r$ 0 r *$at in!"str'

#.&0 y F 0.$%A%e.0.00/Ax )& F 0.0%%$

#.00

0.%0 ,fficiency score

y F 0.2244e0.0422x )& F 0.4A2&

0./0

y F 0.22#5e0.05/4x )& F 0.5&/&

0.50

0.&0

0.00 #$%0.%# #$%#.%& #$%&.%2 #$%2.%5 #$%5.%4 #$%4.%/ #$%/.%A #$%A.%% #$%%.%$ #$%$.$0 #$$0.$# #$$2.$5 #$$5.$4 #$$#.$& #$$&.$2 #$$4.$/ #$$/.$A #$$A.$% #$$%.$$ #$$$.00

!!) Model T,'

?!! Model Pure T,'

*,

Tabl$ B) Tar&$t in("ts an! $sti*at$! slac> in("ts in *$at (r c$ssin& in!"str' 8at 1561,6# (ric$s: Kear !apital ,mployees Materials consumed Fuel consumed )s. in lakhs' 7os.' )s. in lakhs' )s. in lakhs' Target *lacks Target *lacks Target *lacks Target *lacks

#$

#$%0.%# ##/.0 0.0 #4A.0 #$%4.%/ %A.0 0.0 #0$.0 #$%/.%A #04.A &5.& #0/./ #$%A.%% #22.0 0.0 #0%.0 #$%%.%$ ##&.% 0.4 ###.5 #$%$.$0 ##$.4 2.A ##4.0 #$$0.$# #0&.0 0.0 ##5.0 #$$#.$& ##%.0 0.0 $$.0 #$$&.$2 #/$.% ##.5 ##&.$ #$$2.$5 #$$.0 0.0 $&.0 #$$5.$4 &&%.0 0.0 #25.0 #$$4.$/ &A5.0 0.0 #20.0 #$$/.$A 2A4.0 0.0 #A4.0 #$$A.$% &A#.$ 45.A #&%.$ #$$%.$$ &A5.0 #4%.% #20.0 #$$$.00 &&A.% $&.& ##4.4 S "rc$) !alculated from -*I data various issues'

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

#0&.0 4$.0 $2.# $/.0 %#.# %%.# #&0.0 #A4.0 #%#.& &0#.0 #%4.0 &%0.0 24#.0 &AA.% &%0.0 &5&.2

0.0 0.0 0.0 0.0 0.0 &A.% 0.0 0.0 0.0 0.0 0.0 0.0 0.0 A$.2 55.2 5$.$

/.0 A.0 %.5 $.0 %.5 $.# #0.0 ##.0 #2.2 #5.0 #/.0 #%.0 &/.0 #A.$ #%.0 #4.%

0.0 0.0 0.5 0.0 0.5 0.0 0.0 0.0 0./ 0.0 0.0 0.0 0.0 /.A &.A 0.0

Table A provides results on target inputs and the estimated slack inputs in Indian meat processing industry. Target inputs refer to what a particular !M) ought to have consumed if it was on the efficient frontier. The slack inputs are excess inputs. The slack is calculated as the difference between actual inputs consumed minus the target input a !M) ought to have consumed. -n efficient !M) will have 6ero input output slacks. In meat processing units" slacks in capital use are showing mixed trends. The highest slack in capital use was recorded in #$$%.$$ to the extent of )s. #4%.% lakhs. If the industry was to qualify for an efficient !M) in #$$%.$$" )s. #4%.% lakhs of capital had to be reduced. 9abour inputs are efficiently used in meat processing units. 7o slacks in labour used were recorded except in #$%&.%2" where about ## labours were excessively employed. The slack in materials and fuel consumed also showed mixed trends but in most of the years these inputs were efficiently used.

@. C ncl"si ns an! R$c **$n!ati ns

Meat industry in India experienced tremendous growth during last two decades. The growth is largely number driven causing doubts on its sustainability. The processing of meat for value addition is meager and most of the production takes place in unorgani6ed slaughterhouses. The empirical analysis of productivity and efficiency in meat processing units indicate that there are significant possibilities of enhancing the performance of these units. Total Factor Productivity TFP' change is negligible and the increase in meat output is basically due to increase in input use.

&0

The average technical efficiency score is estimated to be 0.4$ under !)* model and 0.$2 under ()* model. The value of efficiency indices equal to unity implies that the industry is on frontier while values below unity show that the industry is below the frontier or is technically inefficient. The analysis indicates that the average technical inefficiency could be reduced by 5# percent under constant returns to scale and A percent under variable returns to scale. The average scale efficiency for the entire period is 0./5 which shows that the potential of increasing scale efficiency in meat processing units is to the extent of 2/ percent. There was considerable under utili6ation of input resources during #$%0s" which has improved overtime. This had significant positive impact on labour absorption as well as labour productivity. 8hile the capital investment in industry improved" capital productivity has remained stagnant. The analysis of input slacks in meat processing industry suggest that the industry is labour intensive and the effects of expansion of meat industry on labour employment and productivity appears to be favourable. The raw material which constitutes more that %0 percent of the production cost is raw meat which is also inefficiently used. Market infrastructure for live animals used for slaughtering is poor and unorgani6ed. There are a number of intermediaries acting between the producers and slaughterhouses@ processors. Meat processing units are often located in the urban areas and thus the transportation cost of live animals is high. -nimals are transported from distant markets" causing weight loss and hence meat yield is low. The meat processing industry has not taken much initiative to strengthen backward linkages with the farmers@ producers and largely depends on the intermediaries for its requirement of raw materials. This leads to irregular supply of animals for slaughter. Therefore" proper methods of sourcing quality animals for meat production should be adopted to shorten the supply chain of meat processing industry. Meat processing industry has also been scale inefficient due to slack in capital use in recent the years. In order to improve the industryIs productivity and efficiency" there is a need for investment in technological know.how and improvement in managerial capabilities of meat processors. This would help to improve capacity utili6ation and expansion. The scale inefficiency in meat processing shows that a proper coordination@ combination of inputs is required which is lacking at present. 7evertheless the industry is becoming more capital intensive. Thus while introducing
&#

meat processing technologies it should be kept in mind that the technological change should have minimum adverse effect on employment.

&&

A. R$0$r$nc$s
-hluwalia" I. N. #$$#'; 5Pro$6cti7ity an$ 8ro9th in #n$ia Man6fact6rin 5" 3xford Gniversity Press" +elhi. ?anker" ). +.> !harnes" -. and !ooper" 8. 8. #$%5'; :2o.e Metho$s for Esti.atin Technical an$ 2cale #nefficiencies in !ata En7elo/.ent "nalysis;, Management *cience" (ol.20" 7o.$" pp.#0A%.#0$&. ?halla" D.*. and Peter Ca6ell #$$%'; 5Foo$ rains !e.an$ in #n$ia to <0<0= " Preli.inary E%ercise5" ,conomic and Political 8eekly" (ol. 2& 4&'; -#40.-#45. ?oussofiane" -.> ). +yson and ,. Thanassoulis #$$#'> :"//lie$ !ata En7elo/.ent "nalysis;" ,uropean Nournal of 3perational )esearch" (ol. 4&" pp. #.#4. !aves" +.8." 9.). !hristensen and 8.,. +iewert #$%&'; :The Econo.ic Theory of #n$e% N6.(ers an$ the Meas6re.ent of #n/6t, O6t/6t an$ Pro$6cti7ity; " ,conometrica" (ol. 40" pp #2$2.#5#5. !harnes" -.> !ooper" 8. 8. and )hodes" ,. #$A%'; :Meas6rin the Efficiency of !ecision Ma>in )nits;" ,uropean Nournal of 3perational )esearch" (ol. &" pp.5&$. 555. !oelli" Tim #$$/'; :" 6i$e to !E"P 3ersion <?1= " !ata En7elo/.ent "nalysis @&o./6terA Pro ra.;" !,P- 8orking Paper $/@0%" !entre for ,fficiency and Productivity -nalysis" +epartment of ,conometrics" Gniversity of 7ew ,ngland" -ustralia. +elgado" !." M. )osegrant" C. *teinfeld" *. ,hui and !. !ourbois #$$$'; :Li7estoc> to <0<0= The Ne%t Foo$ 4e7ol6tion;, Food" -griculture and the ,nvironment +iscussion Paper &%" International Food Policy )esearch Institute" 8ashington" +.!." Food and -griculture 3rganisation" )ome" and International 9ivestock )esearch Institute" -ddis -baba. Fare" )." *. Drosskopf" M. 7orris and P. Phang #$$5'; :Pro$6cti7ity 8ro9th, Technical Pro ress an$ Efficiency &han e in #n$6strialiBe$ &o6ntries; -merican ,conomic )eview" (ol. %5 #' pp //.%2. Doldar" ?.7. #$%/'; 5Pro$6cti7ity 8ro9th in #n$ian #n$6stry5" -llied Publishers" +elhi. Doldar" ?.7. and -nita Bumari &00&'; :#./ort Li(eralisation an$ Pro$6cti7ity 8ro9th in #n$ian .an6fact6rin #n$6stries in the 1CC0s;" 8orking Paper *eries 7o. ,@&#$@&00&" Institute of ,conomic Drowth" +elhi : ##0 00A. Cossain" Ferdaus and *an<ib ?huyan &000'; :"n "nalysis of Technical Pro ress an$ Efficiency in )?2? Foo$ #n$6stries;" a paper presented in the 7ational !onference on -merican !onsumer in the !hanging Food *ystem" organi6ed by the ,)*@G*+-" May 2.4" &000" 8ashington +.!.

&2

I!I!I #$$5'; :Pro$6cti7ity in #n$ian .an6fact6rin = Pri7ate &or/orate 2ector 1C7<D 73 to 1CC1DC<;, The Industrial !redit and Investment !orporation of India" Mumbai. Bumar" P. #$$%'; 5Foo$ !e.an$ 26//ly Pro*ections5, Indian -gricultural )esearch Institute I-)I'" 7ew +elhi. 9andes" Maurice> *uresh Persaud and Nohn C. +yck &005'; :#n$iaEs Po6ltry 2ector= !e7elo/.ent an$ Pros/ects;" 8)*.05.02" ,conomic )esearch *ervice" Gnited *tates +epartment of agriculture" 8ashington +.!. Mishra" *.7. #$$4'; 5#n$iaEs Li7estoc> Econo.y= " Pers/ecti7e on 4esearch5 " Indian Nournal of -gricultural ,conomics" 50 2'; &44.&/2. Mitra" -. #$$$'; 5Total Factor Pro$6cti7ity 8ro9th an$ Technical Efficiency in #n$ian #n$6stries5" ,conomic and Political 8eekly" )eview of Industry and Management" Nuly 2#.-ugust /. Mitra" -> -. (aroudakis and M. -. (egar6ones #$$% A= 52tate #nfrastr6ct6re an$ Pro$6cti7e Perfor.ance in #n$ia Man6fact6rin 5" Technical Paper 7o. #2$" 3rganisation for ,conomic !o.operation and +evelopment" +evelopment !entre" Paris. Pattnayak" *an<a *. and *.M. Thangavelu &002'; :Econo.ic 4efor. an$ Pro$6cti7ity 8ro9th in #n$ian Man6fact6rin #n$6stries= "n #nteraction of Technical &han e an$ 2cale Econo.ies;" 8orking Paper 7o. 020A" +epartment of ,conomics" 7ational Gniversity of *ingapore. )avishankar" -. and Pratap *. ?irthal #$$$'; :Li7estoc> sector in #n$ia= " en$a for the F6t6re;" Policy ?rief 0A" 7ational !entre for -gricultural ,conomics and Policy )esearch 7!-P'" 7ew +elhi. Trivedi" P." -. Parkash" and +.*inate &000'; :Pro$6cti7ity in Ma*or .an6fact6rin #n$6stries in #n$ia= 1C73D74 to 1CC7DCF;" +evelopment research Droup *tudy 7o. &0" +epartment of economic -nalysis and Policy" )?I" Mumbai.

&5