structure, employment and productivity growth in the ...cmdr.ac.in/editor_v51/assets/mono-50.pdfan...

External Assistance For Health Sector : 1Structure, Employment and Productivity Growth in the Indian Unorganized Manufacturing Sector:

CMDR Monograph Series No. - 50

Structure, Employment and ProductivityGrowth in the Indian Unorganized

Manufacturing Sector:

An Industry Level Analysis

Rajesh Raj S. N.Associate Fellow

CENTRE FOR MULTI-DISCIPLINARY DEVELOPMENT RESEARCHAlur Venkatrao Circle, DHARWAD-580001, Karnataka, India

Ph : 091-0836-2447639, Fax : 2447627E-mail : [email protected]

2 CMDR Monograph Series No. - 50

The paper analyses the size, growth andproductivity performance of theunorganized manufacturing sector inIndia during 1978-79 – 2000-01. Thestudy shows evidence of increase in sizewith a slowdown in the reforms period.Evidence indicates that the rate of growthvaries widely across the two-digitindustries but the variation in growth rateis smaller in the 90s. Textiles andmachinery goods were the fastestgrowing segments of Indian unorganizedmanufacturing sector in the reformsperiod. Both the partial factorproductivity approach and total factorproductivity approach reflect that

productivity of the sector has improvedduring the period under study. Thedecomposition of productivity growth intotechnical change and efficiency changereveals that the latter has been the majorcontributor to TFPG during the periodunder study. It is also found that capitalintensity and wage rate are essentialfactors for augmenting labourproductivity levels in the sector.

JEL Classification: D24; O47; R11

Keywords: Unorganized Manufacturing;Reforms; Productivity Growth

Structure, Employment and Productivity Growth in theIndian Unorganized Manufacturing Sector:

An Industry Level Analysis*

Rajesh Raj S. N.Associate Fellow

Centre for Multi-Disciplinary Development Research (CMDR)Dharwad

Karnataka

e-mail: [email protected]

Abstract

*This study forms a part of my Ph.D thesis submitted to IIT Madras. I am thankful to my supervisorMalathy Duraisamy for valuable insights and advice. Helpful comments from D Jayaraj, Uma Rani, P RPanchamukhi, Suresh Babu and Mihir K Mahapatra are gratefully acknowledged. The usual disclaim-ers apply.


1. Introduction

Manufacturing is an important sector in theIndian economy accounting for about 31percent of the non-agricultural sector and25 percent of the overall GDP in India(Kalirajan and Bhide, 2004). This sector witha growth rate of 9.3 per cent was a fastgrowing segment of India’s domesticeconomy in the 1990s (Thomas, 2002).India’s manufacturing sector has a largeunorganized component (comprised of unitswith less than 10 employees using powerand those units with 10 to 19 employees notusing electric power) employing about 3/4thof the manufacturing workforce andcontributing to 17 percent of the total NDPof the unorganized non-agricultural sector(Kulashreshta and Singh, 2001). Aninteresting observation is that in the post1997-98 period, output in the organizedsector has grown at a slower rate than inunorganized manufacturing and this is dueamong other things to emergence of flexibleproduction systems and substantial increasein outsourcing by the organized sector(Kalirajan and Bhide, 2004). Urbanisationand rural to urban migration are the otherreasons for the rapid growth of unorganizedmanufacturing sector. Given the crucial roleof the manufacturing sector in India’seconomic growth, more attention needs tobe directed to the sector and this implies thatthe large and burgeoning unorganizedsegment of the sector cannot be ignored.Thus there is a need to study the size,structure and performance of unorganized

manufacturing sector in India.

There is a large body of literature on growth,productivity, and other aspects of theorganized manufacturing sector. Surprisingly,only a small number of studies haveaddressed these issues in the context of theunorganized manufacturing sector. This biasis even clearly visible in the Indian statisticalsystems (Kundu, 1998)1 . Breman attributesthis neglect “ both to lack of knowledgeregards the lower level of the urban economyand the lack of affinity with methods ofresearch that could increase that knowledge”(Breman, 1999: 409). Given this lacuna, itbecome crucial to analyze the growth andproductivity performance of the unorganizedmanufacturing sector which forms the subjectof inquiry of the present study.

The present study is an attempt to analyzethe structure, employment, and levels andchanges in factor productivity and intensityin the unorganized manufacturing sector atthe aggregate and disaggregate level, for theperiod 1978-79 to 2000-01. Specifically,the study addresses the following objectives:

1. To analyze the size, structure andgrowth of the unorganizedmanufacturing sector during the pre-reforms and reforms periods

2. To study the trends in factorproductivities and factor intensity inthe unorganized manufacturingsector during the pre-reforms andreforms periods


3. To determine the sources of labourproductivity growth in theunorganized manufacturing sector

4. To examine inter-industry variationin total factor productivity growthand its components namely,technical change and technicalefficiency change

The scheme of the paper is as follows.Section 2 presents a selective review ofliterature on the unorganized manufacturingsector in India. The database, variables andmethods used in the study are discussed insection 3. Section 4 examines the size,structure and growth of the sector. The levelsand changes in factor productivities andfactor intensity are analyzed in section 5. Insection 6, the determinants of labourproductivity growth are analyzed. The totalfactor productivity growth is decomposedinto technical change and efficiency changeand this is presented in Section 7. Lastsection summarizes the major findings of thestudy.

2. Review of Literature

Most of the studies on the unorganisedsector have considered the size and growthof the sector and its contribution to GDP,etc. (Visaria and Jacob, 1996; Kulshreshtaand Singh, 1998; Kulshreshta and Singh,2001; Mitra, 2001; Kabra, 2003). It hasbeen shown that over the years the valueaddition by the sector has dwindled fairlymarkedly but the number of people

employed in the sector has increased.Recently there have been some attempts tomeasure the productivity performance in thesector (Duraisamy, 2000; Unni et al., 2001;Kundu et al., 2001; Bhalla, 2001; Raj andDuraisamy, 2005). Some studies have founda large part of the informal sector units to befunctioning at a low level of labourproductivity (Duraisamy, 2000; Kundu,2001). Duraisamy (2000) has found strikinggrowth in the productivity of capitalemployed in the sector but the units are foundto be operating with low level of capitalassets (Kundu, 2001). Bhalla (2001)computed TFP growth in the unorganizedmanufacturing sector for two periods, 1984-85 to 1989-90 and 1989-90 to 1994-95using growth accounting method. The studyreported a positive TFP growth in NDMEsin the rural areas and a high negative TFPgrowth in DMEs in the urban areas in thefirst period. The growth in TFP in the secondperiod was negative in both types ofenterprises in urban and rural areas. Unni etal. (2001) found that the unorganised sectorhas witnessed positive labour and capitalproductivity only during the late 80’s. Theyobserved that both the organized andunorganized sector in India witnessed adecline in TFP during the period under study.The TFP growth was higher in the pre-reforms period but appeared to decline inthe reforms period, especially in theunorganized sector. However, this study hasclassified the unorganised manufacturingsector into certain broad categories like basicgoods, intermediate goods, capital goods,


consumer goods, etc. and has not takenaccount of the industries at a moredisaggregate two-digit level. Raj andDuraisamy (2005) examined the productivityperformance of the sector for the period1984-85 to 1994-95 using partial factorproductivity ratios. The study has consideredonly the OAMEs and NDMEs. It has beenfound that the sector has recorded a betterperformance during the period in terms ofvalue added per unit, labour productivity andcapital productivity, and the performancewas more striking in labour intensiveindustries.

Available studies on the unorganizedmanufacturing sector are mostly at theaggregate level, or refer to specific region,or sector. Further, the studies have mostlyconsidered the period up to 1994-95. Thereis paucity of research on Total FactorProductivity (TFP) growth in the sector,sources of TFP growth – technical efficiencychange and/or technical change - particularlyat the disaggregate industry level. Thepresent study makes an attempt to fill thesevisible gaps in the literature.

3. Data, Variables and Methods Data

The study is based exclusively on secondarydata. The data are drawn from the largenational surveys conducted by the NationalSample Survey Organization (NSSO) in its33rd (1978-79), 40th (1984-85), 45th

(1989-90), 51st (1994-95) and 56th (2000-01) rounds. In order to obtain the figures

for the unorganized sector as a whole, datafor each enterprise type (OAMEs, NDMEsand DMEs) and by location (rural and urban)have been added. In order to examine theimpact of reforms, the entire time period(1978-2001) has been sub-divided intoPre-Reforms period (1978-79 to 1989-90)and Reforms period (1989-90 to 2000-01)2 .

A major issue that needs to be solved whilecomparing the NSSO rounds is the choiceof National Industrial Classification (NIC).The NSSO has used different NIC in itsdifferent survey rounds. The report for the33rd and 40th rounds provides data as perNIC 1970, those for 45th and 51st as perNIC 1987 and the latest 56th round reportas per NIC 1998. While concordance ofNIC 1987 with 1970 required only theinterchanging of divisions 30 and 31,matching of NIC 1987 with NIC 1998requires a greater degree of approximationby relevant grouping. The exactconcordance between 2-digit industrygroups of NIC 1987 with that of NIC 1998requires data on some 3- and 4-digitindustrial divisions which are not readilyavailable in the published report for 56th

round (Report Numbers 478 and 479).Therefore, one has to depend on the unitlevel data for the 56th round available fromthe NSSO. In this study, the nineteen twodigit industries are reclassified into 15 industrygroups to enable comparison across variousrounds following the NIC 1987. Details of15 industry groupings clubbed for the


purpose of this study (which may differ withthe groupings made in other studies) are givenbelow.

NIC 1987(Two digit)

Manufacture of food products 20 - 21

151 + 152 + 153 + 154 +

15544Manufacture of beverages, Tobacco and related

products 22 (155 - 15544) + 16Manufacture of cotton textiles, wool, silk and man-

made fibre textiles and jute and other vegetable fibre

textiles 23 + 24 + 25 171 + 01405Manufacture of textile products (including wearing

apparel) 26 172 + 173 + (181 – 18104)Manufacture of wood and wood products; furniture

and fixtures 27 20 + 361Manufacture of paper and paper products and

printing, publishing and allied industries 28 21 + 22Manufacture of leather and products of leather, fur

and substitutes of leather 29 18104 + 182 + 19Manufacture of basic chemicals and chemical

products (except products of petroleum and coal) 30 24Manufacture of rubber, plastic, petroleum and coal

products; processing of nuclear fuels 31 23 + 25

Manufacture of non-metallic mineral products 32 26Basic metal and alloys industries 33 27 + 371Manufacture of metal products and parts except

machinery and equipment 34 2811 + 2812 + 289Manufacture of machinery and equipment other than

transport equipment 35 - 36

2813 + 29 + 30 + 31 + 32 +

3311

Manufacture of transport equipment and parts 37 34 + 35

Other manufacturing industries 38 (33 – 33111) + 369

Name of the Industry NIC 1998

Table 1 - Concordance Table

Variables

The size and structure of the sector areanalyzed using four indicators/variablesnamely, number of enterprises, employment,fixed capital stock and gross real value


added. The definition and measurement ofthese variables are as follows.

Gross Real Value Added

Gross value added figures have been usedto represent output. Use of gross valueadded at constant prices to represent outputis a common practice in the Indian empiricalliterature (Goldar, 1986; Ahluwalia, 1991;Balakrishnan and Pushpangadan, 1994;Balakrishnan and Pushpangadan, 1998).Studies on the unorganized manufacturingsector in India have by and large used thesingle deflation method to deflate the grossvalue added (Unni et al., 2001; Rani andUnni, 2004). The implicit deflators of grossdomestic product of the unorganizedmanufacturing available at the two-digitindustry group level have been used to deflategross value added at the industry level. Theadvantage of using this deflator is that it notonly takes care of the general changes in theprice level in the economy, but also thechanges in the prices of those goods andservices that constitute the value added,namely, the output and inputs. In this sense,this deflator has the advantage of the doubledeflation method. Since the values of grossdomestic product were expressed in threedifferent bases (1970-71, 1981-82 and1993-94), they have been arithmeticallybrought to a common base year (1993-94)through splicing method.

Fixed Capital Stock

The measurement of capital input is rather

problematic, and has been a controversialtopic in theoretical as well as empiricalliterature (for a discussion on capital stockmeasurement see Raj and Mahapatra,2006). Despite its limitations, most studiesin the unorganized manufacturing sector inIndia have used book value of total fixedassets owned by the unit on the closing dateof the accounting year to represent capitalinput (Unni et al., 2001; Rani and Unni,2004). Fixed assets include land, buildingsand other constructions, plant and machinery,transport equipment, tools and other fixedassets that have a normal economic life ofmore than one year from the date ofacquisition. The use of gross figures torepresent the capital stock can be justifiedin the case of developing countries such asIndia in general, and unorganizedmanufacturing sector in particular, on theground that capital stocks are more oftenused at approximately constant levels ofefficiency for a period far beyond theaccounting life measured by normaldepreciation until they are eventuallydiscarded or sold for scrap (Salim andKalirajan, 1999; Hossain and Karunaratne,2004). In essence, the value of old machinemay decline but it need not lead to anydecline in the current services of the capitalequipment. Due to the absence of data onfixed capital stock formation at the industrylevel, the present study used gross fixedcapital stock formation by unregisteredmanufacturing sector at the all India level todeflate fixed capital stock in the unorganizedmanufacturing sector. The values areexpressed in 1993-94 prices.


Employment

Total number of persons engaged is takenas the measure of labour input. As bothworkers, working proprietors andsupervisory/managerial staff can affectproductivity, so number of persons engagedis preferred to number of workers. Workersinclude full-time, part-time, hired and otherworkers.

Number of Enterprises

The unorganized manufacturing sector iscomprised of three types of enterprises,namely, Own Account ManufacturingEnterprises (OAMEs), Non-DirectoryManufacturing Enterprises (NDMEs), andDirectory Manufacturing Enterprises(DMEs). OAMEs employ only family labourwhile NDMEs and DMEs employ hiredlabour. The number of workers is less thansix in case of NDMEs and more than orequal to six in case of DMEs. The data ontotal number of enterprises were notavailable for the unorganized sector as awhole. In order to arrive at it, data for eachenterprise type (OAMEs, NDMEs andDMEs) and by location (rural and urban)have been summed.

Methods

The performance of the sector is examinedusing growth in value added and changes inemployment and share of investment duringthe pre-reforms and reforms period. Growthrates, shares and factor ratios are computed.

The productivity estimates are obtained usingpartial and total factor productivity methods.Econometric technique is used to estimatethe determinants of labour productivitygrowth. Non-parametric method namelyDEA and Malmquist index are used toobtain estimates of TFP and its components.The methods are discussed in relevantsections.

4. Size, Structure and Growth of theSector

Size of the Sector

The size of unorganized manufacturing sectorand changes in it over the last two decadesare analyzed using indicators such as numberof enterprises, employment, fixed capitalstock and gross value added. Table 2 showsthe basic features for the period 1978 - 2001.It may be observed that the number ofenterprises and workers in the unorganizedmanufacturing sector have almost doubledin about two decades since 1978-79. Allthe indicators witnessed positive averageannual growth during the pre-reforms (1978-79 to 1989-90) and reforms period (1989-90 to 2000-01). Surprisingly, in the latterperiod, there has been a decline in growthespecially in number of enterprises andemployment.


Pre-reforms Period

Enterprises (in lakhs) 84.69 170.24 6.77 0.4

Employment (in lakhs) 182.14 370.81 6.89 0.4

Fixed Capital (in billion) 71.16 495.98 14.14 5.79

Gross Value Added (in

billion) 86.57 412.28 12.23 3.77

Reforms

Period

Table 2 Size and Growth of the Unorganized Manufacturing Sector

Variable 1978-79 2000-01

Gro

wth

Rat

e

Note: (i) Pre-reforms period corresponds to 1978-79 to 1989-90 and Reforms periodcorresponds to 1989-90 to 2000-01.

(ii) Real Annual Average Growth has been estimated usingSource: National Sample Survey Organization (NSSO) surveys on the unorganizedmanufacturing sector, various rounds.

A similar pattern of decline in growth rate is noticed across different types of enterprises,namely OAMEs, NDMEs and DMEs, during the reforms period (Table 3). In otherwords, compared to the pre-reforms period, the growth of number of enterprises,employment, fixed capital and gross value added witnessed a fall irrespective of types ofenterprises and location (rural and urban).

Pre-Reforms Reforms Period

(GR) (GR)

71.78 146.65 6.99 0.36

9.56 17.12 5.46 0.46

3.35 6.47 5.4 1.21

63.68 119.35 6.76 -0.23

21.01 50.9 6.8 2.09

OAME 116.14 250.61 7.76 0.2

NDME 27.86 55.62 6.26 0.77

DME 38.13 64.58 4.37 0.91

Rural 125.04 239.86 6.94 -0.18

Urban 57.11 130.95 6.79 1.58

OAME 29.05 194.28 15.92 3.92

NDME 19.72 134.17 15.55 4.38

DME 22.39 167.52 9.6 10.47

Rural 38.24 190.34 12.92 3.61

Urban 32.92 305.64 15.41 7.49

OAME 34.25 174.3 13.35 3.46

NDME 17.62 103.15 14.76 3.61

DME 34.7 134.82 9.33 4.32

Rural 38.69 182.68 12.25 3.67

Urban 47.88 229.6 12.21 3.85

Sector

Gross Value Added (in Rs. Billion)

Types

Sector

Types

Sector

Fixed Capital (in Rs. billion)

Types

Sector

Rural

Urban

Employment (in lakhs)

Types

OAME

NDME

DME

Variable 1978-79 2000-01

Enterprises (in lakhs)

Source: NSSO surveys on the unorganized manufacturing sector, various rounds

Table 3: Size and Growth of the Sector by Types of Enterprises and Location


Analyzing the growth of the sector byenterprise type in the pre-reforms period, itis found that OAMEs and NDMEsdemonstrated higher growth rates in all theindicators where as in the reforms periodDMEs overtook these two enterprise types(Table 3). Similarly, comparing the rural andurban areas it may be seen that there wasnot much difference in the pre-reforms periodbut in the reforms period the growth rate inall the indicators is higher in urban comparedto rural areas. In fact, growth of enterprisesand employment is negative in rural areas.

Structure of the SectorFor a better understanding the structure ofthe sector is examined in terms of fourindicators at the dis-aggregate two-digitindustry level. The NSSO provides data for19 two-digit industries. For the purpose ofthe study, they are grouped into 5 broadindustry groups: Food and related (21, 22);Textiles and related (23, 24, 25, 26); Wood,paper and leather (27, 28, 29); Minerals andmetal (32, 33, 34); and others (30, 31, 35,36, 37, 38). The structure of the sector isanalyzed by examining the changes in theshare of each industry group in the fourindicators, number of enterprises,employment, fixed capital stock and grossvalue added between two different timepoints, 1978-79 and 2000-01. Figures 1 to4 show the share of each industry group intotal number of enterprises, employment,fixed capital stock and value added in 1978-79 and 2000-01.

Share in EnterprisesIn terms of share in enterprises to total, themanufacture of textiles and cotton goods(30.7 per cent) is the major industrial activity,closely followed by the manufacture of foodand beverages (30 per cent) in 2000-01.These two industry groups together accountfor about 2/3rd of the total enterprises in theunorganized manufacturing sector.Comparing the share in enterprises in 1978-79 and 2000-01, a striking finding is thesignificant increase in the share of food andrelated, category which includes themanufacture of food and beverages. Duringthe same period, all categories except thecategory representing ‘other’ manufacturingindustry groups witnessed decline in the sharein enterprises (Figure 1).

Share in EmploymentTextiles and food products are the majorindustries that employed largest number ofworkers (57 percent) in the sector in 2000-01. However, comparing the share inemployment between 1978-79 and 2000-01, the manufacture of food and relateditems shows evidence of considerableimprovement in relative importance from23.6 percent to 27.6 percent compared totextiles, which recorded considerableerosion in its share from 35.8 percent to 29.1percent (Figure 2).


23.6

35.8

18.813.3

8.5 11.2

27.6 29.1

12.9

19.2

05

10152025303540

Food and

related

Textiles

and related

Wood,

Paper and

Leather

Minerals

and Metals

Others

Sh

are

1978-79 2000-01

22.5

34.1

24.2

11.87.4

30.0

9.28.9

30.7

21.2

05

10152025303540

Food andrelated

Textilesand related

Wood,Paper andLeather

Mineralsand Metals

Others

Sh

are

1978-79 2000-01

17.6

33.1

22.7

15.612.1

19.2

28.7

20.117.815.1

05

101520253035

Food andrelated

Textilesand related


Mineralsand Metals

Others

Sh

are

1978-79 2000-01

35.3

15.9 15.221.5

25.2

14.1 11.9 13.4

26.8

20.6

05

10152025303540

Food andrelated

Textilesand

related


Mineralsand

Metals

Others

Sh

are

1978-79 2000-01

Figure 1 Industry Category wise Share of Number of Enterprises in the Unorganized Manufacturing Sector Source: 33rd and 56th rounds of NSSO Surveys

Figure 2 Industry Category wise Share of Number of Workers in the Unorganized Manufacturing Sector Source: 33rd and 56th rounds of NSSO Surveys

Figure 3 Industry Category wise Share of Fixed Capital Stock in the Unorganized Manufacturing Sector Source: 33rd and 56th rounds of NSSO Surveys

Figure 4 Industry Category wise Share of Gross Value Added in the Unorganized Manufacturing Sector Source: 33rd and 56th rounds of NSSO Surveys


Share in Fixed Capital StockThe manufacture of textiles and relatedgoods has maintained its supremacy in fixedcapital stock as well. While most of theindustry groups have increased their sharein fixed capital stock over the period,manufacture of food and beverages aloneemerges as the category that has registereda marked decline in its share from 35.3percent in 1978-79 to 20.6 percent in 2000-01 (Figure 3).

Share in Gross Value Added

In the case of value added too, manufactureof products related to textiles and food arethe best performers. However, themanufacture of textiles and related items hasregistered a decline in its share from 33.1percent to 28.7 percent where as food andrelated category has marginally improved itsshare. Besides textiles, the manufacture ofwood, paper and leather has also registereda significant decline in its share from 22.7percent to 15.1 percent (Figure 4).

Increase in the Number of Enterprises,Employment, Fixed Capital Stock andGross Real Value Added

In order to have more insights on the abovetrends discussed for the broad industrycategories, the analysis is repeated at thetwo-digit industry level which yields aninteresting picture. Table 4 shows thatnumber of enterprises, employment, fixedcapital stock and value added have increasedin all the industries during the period understudy. It is apparent from the table thatbeverages industry has primarily contributedto the growing importance of food andrelated category. At the same time, themanufacture of cotton and textiles registeredan absolute increase in the four variables butthe performance was still below that ofbeverages, contributing to the decline in shareof the textiles category (Table 4). Further,the increase in capital stock was three foldcompared to the increase in employment inmost of the industries. Perhaps this isindicative of a shift towards more capital-intensive production process in theseindustries.


Fixed Capital Gross Value Stock Added

Food 187 187 365 418

Beverages 1078 692 918 656

Cotton 128 114 406 215

Textiles 249 246 1087 465

Wood 211 237 689 212

Paper 390 321 1088 528

Leather 70 100 441 443

Chemicals 381 399 656 384

Rubber 216 198 679 299

Minerals 159 216 791 597

Basic Metal 144 132 394 256

Metal Products 188 225 1123 382

Machinery 326 336 1008 677

Transport 85 189 1322 446

Others 286 285 1611 943

All Industries 225 220 672 394

Change over 1978-79

Industry Class

1978-79 = 100

Enterprises Employment

Source: Computed using NSSO survey data.

In sum, our analysis shows evidence ofincrease in size with a slowdown or declinein the reforms period. The analysis at thedisaggregate industry level points to thedeclining importance of textiles and relatedindustry in terms of the four indicators ofsize and structure used in the study.

Growth of Unorganized ManufacturingSectorGrowth in Real Value AddedThe gross value added (GVA) by theunorganized manufacturing sector registereda growth of 9.4 percent and 4.4 percentrespectively in the pre-reforms (1978-79 to1989-90) and reforms period (1989-90 to

2000-01). The rates of growth, however,showed marked variation across the two-digit industries and, for the same industry,between the two time periods, which isshown in the accompanying chart andexplained below.

Figure 5 shows rate of growth of value addedin the two-digit industries, with rate of growthin the pre-reforms period plotted on the X-axis and those in the reforms period on theY-axis. The origin in the scatter plotcorresponds to the growth rates for the totalunorganized manufacturing sector as a whole- 9.4, 4.4 respectively in the pre-reformsand reforms period. The first quadrant

Table 4 Indices of Enterprises, Employment, Fixed Capital Stock and GrossValue Added in the Unorganized Manufacturing Sector


represents those industries whose growthrates are above the average for theunorganized manufacturing sector as a wholein the pre-reforms and reforms periods; andthe second quadrant includes those industries

with growth rates above the average for thereforms period but below the average forthe pre-reforms period. The other twoquadrants are similarly defined and shownin the diagram.

Wood

Chemicals

Beverages Paper

Food Transport

Rubber Cotton

Basic Metal

Metal Products Leather Others

Minerals

Machinery

Textiles

-4

-2

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10 12 14 16 18 20

I II

III IV

Higher growth in the pre-reforms and reforms period

Lower growth in the pre-reforms and reforms period

Lower growth in the pre-reforms period Higher growth in the reforms period

Figure 5 Rates of Growth of Gross Value Added in Two-Digit Industries,Unorganized Manufacturing Sector in India

Note: All value figures are at constant 1993-94 prices.Source: Computed using NSSO survey data.


Rates of Growth in Value Added acrossIndustriesThe estimated growth rates of GVA showsthat in the pre-reforms period growth ratesvaried from a high of 18.9 percent in themanufacture of beverages to a low of 0.4percent in the textiles industry. However, thiswide variation in growth rates witnessed inthe pre-reforms period, slightly declined inthe reforms period, judging from the lowergrowth rates registered by the industriesduring the reforms period. The manufactureof textiles, machinery goods and mineralsregistered very high growth rates of 14.5percent, 9.4 percent and 8.1 percentrespectively during the reforms period whilethe manufacture of chemicals and woodproducts registered negative growth rates of–1.5 percent and – 2.8 percent respectively

in the same period. Overall, the inter-industryvariation in output growth was smaller inthe reforms period.

The manufacture of machinery goods andminerals have registered growth rates abovethe average for the unorganizedmanufacturing sector as a whole both in thepre-reforms and reforms periods (see Figure5, quadrant I). In contrast, three industries,namely basic metal, cotton and rubber (seeFigure 5, quadrant III) have witnessed outputgrowth below the average for theunorganized manufacturing sector in the twoperiods (Box 1). It is also observed that thegrowth of output in these industries furtherslowed down in the reforms period ascompared to the pre-reforms period.

Industry Performance

Machinery, Minerals Higher growth in the pre-reforms

and reforms period

Basic Metal, Cotton, Rubber Lower growth in the pre-reforms and

reforms period

Textiles, Metal Products, Leather Upswing in growth in the reforms

period

Wood, Chemicals Slump in growth in the reforms

period

Box 1 Industry Performance


Another interesting finding is that the ratesof growth were less than the average (forthe unorganized manufacturing sector) forthree industries namely, metal products,leather goods and textiles in the pre-reformsperiod but were among the few industriesthat have registered growth rates aboveaverage for the sector in the reforms period(see Figure 5, II quadrant). In particular,performance of textile industry is noteworthy.The rate of growth of value added by theindustry experienced a turnaround fromlowest growth in the pre-reforms period (0.4percent) to highest among all the industriesin the reforms period (14.5 percent). Itappears that these industries have benefitedthe most from the reforms initiated in the 90s.In contrast, wood products, chemicals,paper products, beverages, food productsand transport goods (see Figure 5, IVquadrant) witnessed a reversal in theirperformance in the reforms period. Theyrecorded very low growth rates in thereforms period, after an impressiveperformance in the previous decade.

Investment and Growth of Employment

This section presents a discussion onchanges in investment, defined as additionto fixed capital stock and employmentacross industries in the unorganizedmanufacturing sector during the pre-reformsand reforms periods. From Table 5, it isevident that the sector witnessed an overallincrease of about 160 lakh workers in thepre-reforms period and 40 lakh workers in

the reforms period. The growth ofemployment was however not quiteencouraging during the reforms period. In10 out of 15 industries, employmentincreased in both the periods but the growthof employment declined substantially duringthe reforms period especially in industriesproducing food products and beverages(Table 5). In terms of investment, allindustries registered a consistent growth inboth the periods. However, the share of mostof the industries in total investment declinedin the reforms period.

In the pre-reforms period, only themanufacture of textiles recorded a declinein employment of 70,000 workers. On theother hand, industries such as woodproducts, beverages and food productsregistered the largest increase in employmentin the same period. These three industriestogether accounted for around 60 percentof the total employment generated in thesector (94,10,000 out of 1,59,60,000).

Textiles industry that showed a decline inwork force in the pre-reforms periodrecorded huge increase in employment in thereforms period (43,20,000) – which is 63percent of total employment generated in thesector. As against this, the manufacture ofcotton, wood products, basic metal andleather goods registered a considerabledecline during the said period. In the ‘jobless’decade of reforms period, 11,30,000employees lost their jobs in the unorganizedsegment of the cotton industry (48 percent


of the total jobs lost); 6,50,000 workerswere similarly affected in wood industry (27percent of the total jobs lost), 4, 60, 000 inbasic metal industry (19 percent of the totaljobs lost) and 1,30,000 in leather goodsindustry (5 percent of the total jobs lost). In

total, 23,70,000 jobs were lost as against63,50,000 new jobs generated by the sectorin the reforms period – number of jobs lostwas 37 percent of the total jobs generatedin the sector.

Pre-reforms Pre-reforms

Reforms

Period

Period Period

Textiles -7 432 Textiles 3.2 28.9

Food 234 62 Food 19.6 13.2

Metal

Products 46 41 Others 9.2 9.5

Machinery 28 27

Metal

Products 8.8 8

Beverages 301 23 Minerals 4.4 7.8

Minerals 144 19 Machinery 5.6 7.5

Paper 38 11 Wood 10.9 7

Others 135 11 Cotton 12 4.9

Rubber 10 7 Transport 0.9 3

Transport 6 2 Rubber 2.7 2.7

Chemicals 42 0 Paper 14.4 1.9

Leather 13 -13 Leather 1 1.6

Basic Metal 50 -46 Beverages 4.4 1.4

Wood 406 -65 Chemicals 2.6 1.4

Cotton 156 -113 Basic Metal 0.6 1.2

All Industries 1603 398 All Industries 100 100

Increase in Employment (in ‘0000s) Increase in Share of Investment (in

Industry

Reforms

Period Industry

Notes: (i) Pre-reforms period corresponds to 1978-79 to 1989-90 and Reforms periodcorresponds to 1989-90 to 2000-01.



Asymmetric Character of Investment

In the reforms period, the manufacture oftextiles and food products and ‘others’industry group together accounted for amajor share in investment in the Indianunorganized manufacturing sector. In all,around 51 percent of the total investmentwas into these three industries. Of these,food products and others accounted for aconsistently larger share in investment. Recallthat textiles and food products are the majoremployment providers in the sector (Figure2). The very low shares of investment – ofless than two percent – in industriesproducing basic metal, chemicals,beverages, leather goods and paper products(these five industries together received 7.5percent of the total investment) is anothersignificant feature of the unorganizedmanufacturing sector. Leather goods,chemicals and basic metals received very lowinvestment in both the decades.

Gainers and Losers in Output,Employment and Investment

The manufacture of textiles and metalproducts which as shown earlier, registeredrelatively high growth in value added, alsogenerated largest number of jobs, andwitnessed a sizeable share in investment inthe reforms period (Box 2). The textileindustry received the largest share ininvestment (28.9 percent) and employment(43, 20, 000 out of 63, 50, 000 jobsgenerated) during the said period. Themanufacture of metal products, on the otherhand, registered a consistent increase inemployment (4, 60, 000 in the pre-reformsperiod and 4, 10, 000 in the reforms period)and maintained a consistent share ininvestment (8.8 percent in the pre-reformsperiod and 8.0 percent in the reformsperiod). Similarly, machinery goods andminerals also registered a consistent increasein employment, investment and value addedin the pre-reforms period and reformsperiod.

Industries Performance

Textiles, Metal Products High employment generation, high investment and high value added growth

Beverages, Paper Products High employment generation, low investment and low value added growth

Wood Products, Cotton Goods Employment decline, high investment and negative value added growth

Textiles, Food Products Major employment generators and investors

Box 2 Industry Performance: Gainers and Losers


Beverages and paper products thoughgenerated more jobs but recorded lowinvestment and low value added growth inthe reforms period. But compared to thepre-reforms period, these industriesgenerated less number of jobs in the reformsperiod. On the other hand, manufacture ofwood products and cotton goods had shedtheir work force and made more investmentin capital stock but ended up producingbelow the average production (for theunorganized manufacturing sector) during thereforms period.

In sum, our analysis shows that the growthof value added has declined in this period,implying that the unorganized manufacturingsector failed to sustain the growth momentumattained in the earlier period. This has beenprimarily on account of slow growth ofemployment and investment during the saidperiod. At the disaggregate industry level,manufacture of textiles and machinery goodswere the fastest growing segments of Indianunorganized manufacturing sector during thereforms period. In contrast, performance ofsome of the major employment providers inthe sector, particularly the manufacture ofwood products, cotton products andbeverages, is quite disappointing during thesame period. Next, we examine the factorproductivities and factor in the unorganizedmanufacturing sector.

5. Levels and Changes in Partial FactorProductivity and Factor Intensity in theUnorganized Manufacturing Sector

Productivity growth has long been recognizedas an important driver of economic growthand a determinant of internationalcompetitiveness of a country relative toothers. According to Kuznets (1966), anessential element in the development andstructural transformation of the developedeconomies was the fast growth in industrialproductivity (Duraisamy, 2000).

To examine the productivity in IndianIndustry various methods have beenemployed. These include partial factorproductivity or single factor productivity(ratio of output to a single input), multi-factorproductivity (ratio of output to more thanone input) or total factor productivity (ratioof output to all the inputs used in theproduction process) methods. The partialfactor productivity method considers onlyone factor of production at a time whileassuming the contribution from other factorsof production constant. Therefore, it fails tocapture the contribution of all the factors asa whole in the total output. In order tocircumvent this problem, total factorproductivity approach is suggested.However, it is not rational to ignore partialfactor productivity approach. Balakrishnan


(2004) argues that labour productivity meritattention in its own right and serves adifferent purpose for which the total factorproductivity is not a substitute. It is arguedthat labour productivity is a measure ofpotential consumption and a steady rise inthe productivity of labour is necessary for asustained increase in the standard of livingof a population. An attempt to ignorechanges in labour productivity reflects aninadequate concern for potential increase inconsumption. Thus, there is a strong casefor measuring labour productivity particularlyin the Indian context (Balakrishnan, 2004).Taking cognizance of it, an attempt is madein this paper to capture the levels and trendsin both partial and total factor productivityin Indian Industrial sector.

The present section captures the productivityperformance of the sector using select factorratios. The study considered the followingstructural ratios:

1. Enterprise productivity – Gross realvalue added / Number of enterprises

2. Labour productivity – Gross realvalue added / Number of workers

3. Capital productivity – Gross realvalue added / Real fixed assets

4. Capital intensity – Real fixed assets/ Number of workers

5. Emoluments per worker – Realemoluments / Number of workers

Total emoluments primarily constitute wagesto workers, bonus paid and other benefitsand so on. To obtain real emoluments, thenominal value has been deflated usingConsumer Price Index (CPI) with base1981-82. The definition and measurementof other variables namely, number ofenterprises, employment, fixed capital stockand gross real value added are as discussedbefore. The variation across industries, anaverage worker in each industry andvariation across the two time periods areanalyzed in terms of these structural ratiosand shown in Table 6.


Ind

ust

ria

lG

rou

p

VA

UC

LR

VA

LV

AK

EM

OL

CL

RV

AL

EM

OL

VA

UC

LR

VA

LV

AK

EM

OL

Fo

od

98

10

59

38

95

81

00

53

32

23

19

52

23

11

51

36

Bev

erag

es2

92

33

81

65

12

10

09

83

61

13

39

57

21

4C

ott

on

12

08

89

31

06

12

61

00

63

71

68

35

61

88

53

21

8

Tex

tile

s8

59

41

02

10

89

41

00

64

51

87

44

21

89

43

19

4W

oo

d5

45

26

31

20

45

10

07

14

10

02

91

89

31

16

2

Pap

er2

34

30

61

63

53

24

81

00

32

41

35

33

91

64

49

12

0L

eath

er1

70

13

11

62

12

41

60

10

07

46

63

54

42

44

31

00

60

6

Ch

emic

als

10

61

20

90

75

14

31

00

44

61

01

16

49

65

81

07

Ru

bb

er3

14

28

51

99

70

31

01

00

41

51

38

34

31

51

44

21

5

Min

eral

s1

88

77

11

01

42

14

31

00

84

93

76

36

62

76

75

26

0B

asic

Met

al3

74

26

02

68

10

32

87

10

06

16

17

72

99

19

46

51

51

Met

al

Pro

du

cts

20

11

82

17

89

81

84

10

05

85

20

45

00

17

03

42

40

Mac

hine

ry4

24

30

42

81

93

35

11

00

55

62

08

30

02

02

67

15

9T

ran

spor

t5

82

45

92

86

62

42

41

00

37

55

24

69

92

35

34

16

2

Oth

ers

15

21

38

14

11

02

14

71

00

61

53

30

56

63

31

59

13

0A

ll

Ind

ust

ries

10

01

00

10

01

00

10

01

00

59

51

76

30

61

79

59

16

7

Un

org

an

ized

Ma

nu

fact

uri

ng

= 1

00

CL

R =

10

01

97

8-7

9 =

10

0

12

3

Dif

fere

nce

bet

wee

n i

nd

ust

ries

, 2

00

0-0

1A

ver

ag

e p

er w

ork

er,

Incr

ease

ov

er 1

97

8-7

9

Tab

le 6

- I

nd

ices

of

Cap

ital

In

ten

sity

, Val

ue

Ad

ded

per

Un

it, L

abo

ur

Pro

du

ctiv

ity,

Cap

ital

Pro

du

ctiv

ity

an

d E

mo

lum

ents

per

Em

ploy

ee, 2

000-

01

Not

es: C

LR

= f

ixed

cap

ital

sto

ck p

er e

mpl

oyee

(ca

pita

l to

labo

ur r

atio

); V

AU

= v

alue

add

ed p

er e

nter

pris

e (e

nter

pris

e pr

oduc

tivi

ty);

VA

L=

val

ue a

dded

per

e

mpl

oyee

(la

bour

pro

duct

ivit

y); V

AK

= r

atio

of

valu

e ad

ded

to f

ixed

cap

ital

sto

ck (

capi

tal p

rodu

ctiv

ity)

; and

EM

OL

= e

mo

lum

ents

rec

eiv

ed p

er e

mp

loy

ee. A

ll v

alu

e fi

gu

res

are

at c

on

stan

t 19

93

-94

pri

ces

for

two

-dig

it in

du

stri

es in

In

dia

.S

ourc

e: C

ompu

ted

usin

g N

SS

O s

urve

y da

ta.


Difference between Industries in FactorRatios, 2000-01

The value added per enterprise varied widelyacross the various two-digit industries (Table6, major column 1). The ratio was highest inthe manufacture of transport goods (582)and lowest in the manufacture of beverages(29) for the period 2000-01. The averagevalue added per unit in the unorganizedmanufacturing sector is set at 100 for theyear 2000-01. A value over 100 indicatesthat the industry has outperformed theaverage for the unorganized manufacturingsector as a whole and vice versa, if the valueis below 100 for any industry. It can be seenthat capital-labour ratio also shows widevariation across the different industries. In2000-01, the manufacture of transportgoods was the most highly capital intensiveindustry whereas the manufacture ofbeverages was the least capital intensive one.Labour productivity ranged from 286 in thetransport industry to 38 in the manufacturingof beverages. Capital productivity did notshow much variation across industries incomparison with other ratios. Interestingly,beverages industry was more productive interms of capital as compared to otherindustries in 2000-01. Emoluments peremployee, on the other hand, showed widevariation across the industries with the highestin transport industry (Rs. 424) and lowest inthe manufacture of beverages (Rs. 12). Thevalue added per unit, capital-labour ratio,labour productivity, capital productivity andemoluments per employee are above theunorganized manufacturing sector average

in the manufacture of leather goods and basicmetals and ‘others’ industrial category. It isfound that industries with relatively highcapital intensity such as transport,machinery, paper, rubber and basic metalhave registered high levels of value addedper enterprise, labour productivity andemoluments per employee but low capitalproductivity. On the other hand, thoseindustries that were relatively low capitalintensive turned out to be more productivein capital. Perhaps the capital is lesssufficiently utilized in the former set ofindustries compared to the latter. The findingalso emphasizes the need to make moreinvestment in the sector so that theproductivity level can be substantiallyenhanced.

Average per Worker

In this subsection, we discuss whetheraverage employee in an industry is receivingemoluments proportionate to his/hercontribution in the value added. Assume thatan employee is supplied with a fixed capitalstock of Rs. 100, then what would be hiscontribution to value added and how muchhe would get back as emoluments in thepresent situation. An average employee inthe beverages industry would have added avalue of Rs. 98, if he were provided with afixed capital stock of Rs. 100. On the otherhand, the value added by a worker wouldbe a mere Rs. 32, if the worker wereemployed in the paper industry (See Table6, major column 2). Clearly capital


requirement per unit of value added is veryhigh in the paper industry. As expected,emoluments received by an employee isapparently very low in the unorganizedmanufacturing sector, and our estimates showthat the workers in most industries do notreceive emoluments commensurate with theirproductivity or the contribution they madein the value added. Emoluments receivedand value added by an employee is higherthan the all industry average in cotton, leather,minerals and metal industries. In the mineralsindustry, if fixed capital sock per employeeis Rs. 100, value added by him is Rs. 84and emoluments received is Rs.9 (Table 6,major column 2). In the cotton industry, oneof the major employment providers in thesector, an average worker received Rs 7 asemoluments, one of the highest in the sector,when he added value equivalent to Rs 74using capital worth Rs. 100. This is not thecase with all employment providers in thesector. For instance, the beverages industry’scontribution to value added per worker isthe highest (98) but emoluments per workeris the lowest. In a capital-intensive industrysuch as manufacture of paper products, anaverage employee received emoluments (Rs.4) less than the all industries average andadded value of only Rs. 32. An employee inthe manufacture of machinery goods – acapital intensive industry with the bestgrowth performance in the reforms period –received emoluments per worker slightlyhigher than the all industries average (Rs. 6)and added value slightly lower than theaverage compared to a similar employee inthe manufacture of paper goods.

Difference between Time-Periods:1978-79 to 2000-01For the unorganized manufacturing sector asa whole, there has been an increase inenterprise productivity, labour productivity,capital to labour ratio, emoluments perworker and decline in capital productivity in2000-01 over 1978-79 (see Table 6, majorcolumn 3). In all the industries exceptbeverages, value added per enterprise hasincreased during the same period. Themanufacture of leather products andtransport goods registered the largestincrease in enterprise productivity. Between1978-79 and 2000-01, capital-labour ratiohas increased in all the industries. Transportgoods, others, metal products, leather goodsand textile products have witnessed thelargest growth in capital-labour ratio. In 12out of 15 industries, labor productivity hasincreased during the period under study.Value added per labour has declined in themanufacture of beverages, wood productsand chemicals. Capital productivity hasdeclined in all except food and leatherindustries. Emoluments per employee hasdeclined only in the manufacture ofbeverages. An average employee in theleather industry received the largest increasein emoluments during the period under study.Interestingly, barring a few exceptions, therelatively more capital-intensive industrieshave recorded a better performance in termsof labour productivity, value added per unitand emoluments per employee over theperiod, 1978-79 to 2000-01. This possiblypoints to the importance of investing morein the fixed capital stock in the sector in order


to make the sector more productive andefficient.

6. Determinants of Labour ProductivityGrowthFactors such as economies of scale,increasing capital labour ratio and increasein wage rate have considerable influence onlabour productivity growth (Salter, 1960;Hahn and Mathews, 1964). The previousanalysis has shown that relatively morecapital-intensive industries are moreproductive and efficient than the less capital-intensive ones. Even though the emolumentsper employee has increased considerablyduring 1978-79 to 2000-01, the workers inthe sector did not receive emolumentscommensurate with the contribution theymade to value added. It is important to seehow changes in these factors affect theproductivity of labour in the unorganizedmanufacturing sector. A double logarithmicfunction is specified and estimated usingmultiple regression technique. Theproductivity growth function is expressed as:

ln (APLG) = á + ln (CAPG) + ln (EMOLG)+ ln (GVAG) + u (1)

where

APLG = growth of labour productivity

CAPG = growth of capital intensity

EMOLG = growth of emoluments peremployee

GVAG = growth of gross value added

u = error term

Labour productivity growth is regressed ongrowth of value added, capital labour ratioand emoluments per worker. A positive andsignificant relationship is expected betweengrowth of labour productivity and valueadded. Growth in labour productivity canalso be due to increase in capital intensitythrough the substitution of capital for labouror the availability of more machines perworker1 . Increase in growth of emolumentsper worker could positively influence theproductivity of labour, particularly in sectorssuch as unorganized manufacturing sectorwhere emoluments paid is very low. Theequation is estimated for pre-reforms andreforms period. For the pre-reforms period,we have taken observations for 15 industriesfor three time periods 1978-79, 1984-85and 1989-90. Data on 15 industries for theperiods 1989-90, 1994-95 and 2000-01are considered for the reforms period. Theregression results are shown in Table 7.


Pre-reforms

period

-5.508* -2.375*

(-3.969) (-2.673)

0.842* 0.349*

-8.201 -4.026

-0.151* 0.488*

(-1.967) -5.202

0.623* 0.173

-5.279 -1.479

0.791 0.787

N 30 30

Growth in Capital Intensity

Growth in Emoluments per Employee

Explanatory Variables

Reforms

period

Intercept

Growth in Value Added

2R

Table 7 OLS estimates of Determinants of Labour Productivity Growth in India’s UnorganizedManufacturing Sector

Note: * significant at 10 % level or above Source: Computed using NSSO survey data.

The coefficient of the growth of value addedis positive and statistically different from zeroin both the pre-reforms and reforms periods.A 10 percent increase in growth of valueadded is associated with an 8 percent and 3percent increase in labour productivitygrowth in the pre-reforms period andreforms period respectively. Growth ofcapital intensity or capital-labour substitutionis an important determinant of labourproductivity growth in the reforms period asobserved from the positive and statisticallysignificant effect of growth of capital intensity.A 10 percent increase in growth of capitalintensity, other things remaining the same, inthe reforms period is associated with a 5percent increase in growth of labourproductivity. The positive sign of the estimateof the coefficient of growth of emolumentsper employee shows that the growth in the

wage rate would enhance the productivityof labour. However, the coefficient of growthof emoluments per employee is notstatistically significant in the reforms period.In sum, this exercise shows that therealization of economies of scale has becomeincreasingly important over the decades asa determinant of productivity growth inIndia’s non-factory sector. Anotherimportant finding is that labour productivitygrowth can be improved by increasing thewages paid to the employees. Again, theavailability of more machines per worker,holding other factors constant, hassignificantly contributed to labour productivitygrowth reflecting the complementary roleplayed by the capital input in improving theefficiency of labour input. On the whole, thecoefficient of determination (R2)reported in


Table 7 reveals that these variables explainabout 80 percent of total variation of labourproductivity growth in the unorganizedmanufacturing sector during pre-reforms andreforms periods.

7. Total Factor Productivity Growth andits Decomposition

In the empirical exercise attempted in thissection, we report Malmquist total factorproductivity growth, which is estimated usingData Envelopment Analysis (DEA) method.DEA envelopes observed input-output datawithout requiring a priori specification offunctional forms that turns out to be its majoradvantage. In contrast, different specificationsof the production function under theparametric approach provide differentresults and this remains a methodologicalproblem. Secondly, DEA is more appealingthan the econometric model as inefficiencyis likely to be correlated with the inputs(Gong and Sickles, 1992). However, DEAis not free from drawbacks, either. Thesedrawbacks include the following. First,measurement error and statistical noise areassumed to be nonexistent. Second, it doesnot allow for statistical tests typical of theparametric approach.

The Malmquist TFP Growth

The Malmquist index is defined usingdistance functions. In this study, an outputdistance function is used to consider amaximum proportional expansion of theoutput, given the inputs. To be specific, the

Malmquist TFP index measures the TFPgrowth change between two data points bycalculating the ratio of the distances of eachdata point relative to a common technology.Following Färe et al. (1994), the output-oriented Malmquist TFP change indexbetween period s (the base period) andperiod t (the terminal period) is given by

where the notation represents thedistance from the period t observation tothe period s technology. A value of m

0 greater

than one will indicate positive TFP growthfrom period s to period t while a value lessthan one indicates a TFP growth decline. Inequation (1), the term outside the squarebracket measures the output-orientedmeasure of Farrell technical efficiencybetween period s and period t and the terminside measures technical change, which isthe geometric mean of the shift in thetechnology between the two periods. In otherwords, TFP growth can be decomposed as,

TFP Growth = Technical Efficiency Change(Catching up Effect) × Technical Change(Frontier Effect)An important issue that has to be addressedwhile measuring TFP growth is the returnsto scale properties of the technology in use.The present study uses a CRS technologybecause the estimates based on theassumption of VRS technology may not

0

0 0 0

0 0 0

1/2( , ) ( , ) ( , )

( , ) ( , ) ( , )( , , , )

s s t t

s s st t t t s s

s t ts s t t s s

d y x d y x d y x

d y x d y x d y xm y x y x

0( , )s

t td y x


properly reflect the TFP gains or losesresulting from scale effects (Grifell-Tatjé andLovell, 1995). Hence it is important thatCRS be imposed upon any technology thatis used to estimate distance functions for thecalculation of a Malmquist TFP index. We

1

0 ( , ) max ,

0,

0,

0,

tt t

it t

it t

d y x

st y Y

x X

1

0 ( , ) max ,

0,

0,

0,

ss s

is s

is s

d y x

st y Y

x X

1

0 ( , ) max ,

0,

0,

0,

ts s

is t

is t

d y x

st y Y

x X

have used the linear programming (LP)technique called DEA to calculate thedistance functions1 . This requires solving offour LPs for each industry. The LPs are:

1

0 ( , ) max ,

0,

0,

0,

st t

it s

it s

d y x

st y Y

x X

and


where yit is a MXI vector of output quantities for

the i-th industry in the t-th year;

xit is a KXI vector of input quantities for the i-th

industry in the t-th year;

Yt is a NXM matrix of output quantities for all N

industries in the t-th year;

Xt is a NXK matrix of input quantities for all N

industries in the t-th year;

is a NXI vector of weights and f is a scalar.

The Malmquist TFP growth rates arereported in Table 8 and diagrammaticallyrepresented in Figure 8. The TFPG rates

measured using the Malmquist index showsthat the productivity grew at a rate of 0.07percent per annum (hereafter, pcpa) duringthe period, 1978-79 to 2000-01. In 8 outof 15 industries, the TFP reported positivegrowth rates. The manufacture of leatherproducts and food products registered thehighest TFP growth rate of 4.24 pcpa and2.48 pcpa respectively. In contrast,manufacture of wood products and textiles,largest employment providers in the sector,registered the largest decline in TFP.

Efficiency Technical TFP

Change Change Change

Food 4.04 -0.82 2.48

Beverages 3.15 -2.5 -1.08

Cotton 1.32 -1.58 -0.72

Textiles 0.64 -1.6 -1.19

Wood -0.83 -2.14 -2.57

Paper -0.87 1.54 0.37

Leather 6.85 -1.04 4.24

Chemicals -0.84 -0.28 -1.07

Rubber -1.17 1.36 -0.16

Minerals 4.1 -1.73 0.82

Basic Metal 0 1.32 1.32

Metal Products -0.59 -0.52 -1.05

Machinery 0 1.77 1.77

Transport 0.08 1.15 1.25

Others 2.79 -1.06 1.08

All Industries 0.85 -0.65 0.07

Industry

Group

Table 8 Growth in Technical Efficiency, Technical Change andTotal Factor Productivity



It would be useful to find out sources ofTFPG which can come about due toimprovement in technical efficiency (catch-up) and/or by improvement in productiontechnology (boundary shift). UsingMalmquist index, total factor productivitygrowth is decomposed in componentsnamely, technical efficiency change andtechnical change. Our analysis shows thatefficiency has been the main contributor to

the growth in total factor productivity duringthe period under study (Table 8). Notably,most of the industries have registered apositive change in technical efficiency and inthose industries where TFP has declined, thedecline is mostly attributed to technicalregress. Only 5 out of 15 industries namely,

paper, rubber, basic metal, machinery andtransport have registered technical progressbetween 1978-79 and 2000-01.Interestingly, these are the industries withhighest capital intensity in the sector in 2000-01 (Table 6). Further, paper, machinery andtransport industries have made considerableinvestment in the fixed capital stock in 2000-01 over 1978-79 (Table 4). This suggeststhat, in the achievement of high rate of output

growth in the unorganized manufacturingsector, the principal difficulty is technology,improvement in which can lead to change orshift in the production frontier. As a result,there is a need for sustained improvement inand adoption of better technology as growthin output depends on technological progress.

-4

-2

0

2

4

6

8

foo

d

bev

erag

es

cott

on

tex

tile

s

wo

od

pap

er

leat

her

chem

ical

s

rub

ber

min

eral

s

bas

ic m

etal

met

al p

rod

uct

s

mac

hin

ery

tran

spo

rt

oth

ers

To

tal

TFPG Effch Techch


8. Summary and ConclusionThis study has examined the size, growth andperformance of unorganized manufacturingsector in India at the two-digit industry level.Our analysis shows evidence of increase insize with a slowdown or decline in thereforms period. The growth of value addedhas declined from 9.4 percent in the pre-reforms period to 4.4 percent in the reformsperiod, implying that the unorganizedmanufacturing sector failed to sustain thegrowth momentum during the period after1990. This has been primarily on account ofslow growth of employment and investmentduring the said period. Compared to the pre-reforms period, growth of the sector was‘employment declining’ in the reformsperiod.

‘Food and related’ industry group improvedits position in relation to other industry groupsand emerged as a major contributor toemployment (from 23.6 percent in 1978-79 to 27.6 percent in 2000-01) and valueadded (from 17.6 percent in 1978-79 to20.1 percent in 2000-01) in the sector,during the two decades under study. On theother hand, textile and allied industrieswitnessed erosion in their shares inemployment (from 35.8 percent to 29.1percent between 1978-79 and 2000-01)and value added (from 33.1 percent in 1978-79 to 28.7 percent in 2000-01); but theycontinue to be the largest source ofemployment generation and value additionin the sector. Manufacture of machinerygoods and minerals registered high rates of

value added growth in both pre-reforms andreforms period. Their share in employmenthas considerably increased while share ininvestment improved in the reforms period.In contrast, basic metal and cotton industries,each of which had low rate of value addedgrowth in the reforms period, reported aconsiderable decline in employment and hadrelatively very low share in investment duringthe said period.

The fast growth in the manufacture of textilesin the reforms period is probably a reflectionof the recent boom in demand for textilegarments. A large share of investment (7.5percent) has moved into the production ofmachinery goods during the said period.These two industries were the fastestgrowing segments of Indian unorganizedmanufacturing sector in the reforms period.Both significantly added to total employmentin the sector (72 percent of the total jobsgenerated), and the performance of the textileindustry (68 percent) in this respect isnoteworthy. In contrast, the performance ofsome of the major employment providers inthe sector, particularly the manufacture ofwood products, cotton products andbeverages, was quite disappointing. Theirgrowth rates fell considerably in the reformsperiod from their impressive performance ofthe previous decade.

The analysis of productivity in theunorganized manufacturing sector points toa different picture. Both the partial factorproductivity approach and total factor


productivity approach reflect thatproductivity of the sector has improvedduring the period under study. However, inproductivity growth, a wide variation isnoticed across the industries. The low growthperformance of wood products, cottonproducts and beverages was also gotreflected in their productivity performance.These industries registered the largest declinein total factor productivity during the periodunder study. Another worrisomephenomenon is the decline in total factorproductivity (-1.19) reported by the textilesindustry despite its impressive performancein terms of gross value added, employmentand investment. This implies that the textileindustry has the potential to increase theoutput much beyond the present level.

The decomposition of productivity growthinto technical change and efficiency changereveals that the latter has been the majorcontributor to TFPG during the period understudy. This suggests that technologicalupgradation needs to be prioritized so as toimprove growth in output of the unorganizedsector. It is also found that capital intensityand wage rate are essential factors foraugmenting labour productivity levels in thesector.

Policy SuggestionsThe industry level analysis showed widevariation across industries in productivitygrowth and efficiency change. In order tobridge the gap in performance of industries,industry specific or product specific policies

may be developed. The vibrant industrygroups should be treated with certain pushpolicies aimed at strengthening their linkagewith the organized sector, assisting them withmarketing support, and encouraging theircommercial operation. On the other hand,efforts may be made to improve theproductivity and efficiency levels in the lowperforming industries. A two pronged policyregime that addresses application of propertechnology would fulfill the resource needsfor productivity improvement and promoteformal-informal linkages which seem to bethe need of the hour.

According to this study, technical efficiencychange is a significant determinant ofproductivity and growth of the unorganizedmanufacturing sector and it outweighs theeffect of technological progress. In fact,results indicate technical regress in allindustries and a turnaround in selectindustries like paper, rubber, basic metal,machinery and transport. Appropriate andbetter technology deserves priority toimprove the growth of output in theunorganized sector. Additionally, there shouldbe emphasis on the evolution of indigenoustechniques so that existing resources can beused in a more efficient manner. Effortsshould also be made to enhance technicalefficiency in the sector especially for theindustry groups that show evidence ofdecline in efficiency. This can perhaps beachieved by improvement in managerialinput, organization and skill of the workforce.Consolidation of tiny firms may also help in


raising the efficiency level of the industry asa whole.

Another policy-relevant lesson from theempirical analysis of the unorganizedmanufacturing sector in India relates to theneed of making more investment in capitalstock. But it is widely known that the firmsin the sector are facing constraints on thefinancial side. In other words, credit non-availability is considered as one of the majorproblems faced by firms in the sector.Therefore, policies that promote a businessenvironment in which unorganized

manufacturing enterprises can have easyaccess to credit at affordable rate areneeded. Similar view is expressed by DeSoto (1989), according to whom excessivetransaction costs and bureaucracies keepinformal sector firms out of the formaleconomy. Further, low cost access tomarkets may also help in increasingefficiency. Strengthening the link betweenorganized and unorganized sector is alsoimportant in view of the fact thatsubcontracting play an important role inenhancing the growth performance of thesector.


Major Data Sources

1. Handbook of Statistics on IndianEconomy, Reserve Bank of India,Mumbai (Various Issues).

2. National Account Statistics, 1950-51 to 2002-03, Economic andPolitical Weekly ResearchFoundation, Mumbai, December2004.

3. Survey on the UnorganizedManufacturing Sector, NationalSample Survey Organization,Ministry of Statistics andProgramme Implementation,Government of India (VariousIssues).

References

Ahluwalia, I. J. Productivity Growth inIndian Manufacturing, Oxford UniversityPress, New Delhi, 1991.

Balakrishnan, P. (2004) Measuringproductivity in manufacturing sector,Economic and Political Weekly, 39, 1465-71.

Balakrishnan, P. and Pushpangadan,K. (1998), What do we know aboutproductivity growth in Indian industry?Economic and Political Weekly, 33,2241–46.

Balakrishnan, P. and Pushpangadan, K.(1994) Total factor productivity growth in

manufacturing industry: a fresh look,Economic and Political Weekly, 29, 2028-2035.

Bhalla, S. (2001) Assessing the quality ofemployment growth using National SampleSurvey Data, Conference Proceedings,Understanding Human Development throughNational Surveys, Pune, India.

Breman, J. (1999) The study of industriallabour in post-colonial India—The informalsector: A concluding review, Contributionsto Indian Sociology, 33, 407-431.

Coelli, T (1996) A guide to DEAP version2.1: A data envelopment analysis (Computer)program, Working Paper No. 96/08,Centre for Efficiency and ProductivityAnalysis (CEPA), University of NewEngland, Australia.

De Soto, H. The other Path: The InvisibleRevolution in the Third World, HarperRow, New York, 1989.

Duraisamy, M. (2000) Growth andproductivity in the unorganisedmanufacturing sector in India, 1984-1990,paper presented to a conference at NewDelhi.

Fare, R., Grosskopf, S., Norris, M. andZhang, Z. (1994) Productivity growth,technical progress and efficiency change inindustrialized countries, AmericanEconomic Review, 84, 66-83.


Goldar, B. Productivity Growth in IndianIndustry, Allied Publishers, New Delhi,1986.

Gong, B. H. and Sickles, R. C. (1992)Finite sample evidence on the performanceof stochastic frontiers and data envelopmentanalysis using panel data, Journal ofEconometrics, 51, 84-258.

Grifell-Tatje, E. and C.A.K. Lovell(1995) A Note on the Malmquist productivityindex, Economic Letters, 47, 169-175.

Hahn, F. H. and Mathews, R. C. O(1964) The theory of economic growth: asurvey, The Economic Journal, 74, 779-902.

Hossain, M. A. and Karunaratne N. D.(2004) Trade liberalization and technicalefficiency: evidence from Bangladeshmanufacturing industries, The Journal ofDevelopment Studies, 40, 87-114.

Kabra, K. N. (2003) The unorganisedsector in India: some issues bearing on thesearch for alternatives, Social Scientist, 31,23-46.

Kalirajan, K. and Bhide, S. (2004) Thepost-reform performance of themanufacturing sector in India, AsianEconomic Papers, 3, 126-157.

Kulashreshta, A. C. and Singh, G. (1998)

Contribution of unorganised sector in theIndian economy, Manpower Journal, 34,1-10.

Kulshreshta, A. C. and Singh, G.Informal sector in India: its coverage andcontributions, pp. 48-82. In Kundu, A.and Sharma, A. N. (eds.) Informal Sectorin India: Perspectives and Policies, IHD,New Delhi, 2001.

Kundu, A. (1998) Introduction, The IndianJournal of Labour Economics, 41, 439-440.

Kundu, A., Lalitha, N. and Arora, S. L.Growth dynamics of informalmanufacturing sector in urban India: ananalysis of interdependence pp. 93-123.In Kundu, A. and Sharma, A. N. (eds.)Informal Sector in India: Perspectives andPolicies, IHD, New Delhi, 2001.

Mitra, A. Employment in the informalsector, pp. 85-92. In Kundu, A. andSharma, A. N. (eds.) Informal Sector inIndia: Perspectives and Policies, IHD, NewDelhi, 2001.

Raj, R. and Duraisamy, M. (2005)Structure, employment and productivitygrowth: evidence from the unorganizedmanufacturing sector in India, 1984/5-1994/5, Journal of International Business andEntrepreneurship, 11, 68-84.

Raj, R. and Mahapatra, M. K. (2006)


Economic reforms and productivity growth:evidence from Indian industrial sector withfocus on selected states, Journal ofInternational Business and Entrepreneurship,12, 35-53.

Rani, U. and Unni, J. (2004) Unorganizedand organized manufacturing in India:potential for employment generating growth,Economic and Political Weekly, 39, 4568-4580.

Salim, R. A. and Kalirajan, K. P. (1999)Sources of output growth in Bangladeshfood processing industries: A decompositionanalysis, The Developing Economies, 37,247-269.

Salter, W. Productivity and TechnicalChange, Cambridge University Press,Cambridge, 1960.

Thomas, J. J. A review of Indianmanufacturing, pp. 84-101. In Kirit, P.and Radhakrishna, R. (eds.) IndiaDevelopment Report, Oxford UniversityPress, New Delhi, 2002.

Unni J, Lalitha, N. and Rani U. Economicreforms and productivity trends in Indianmanufacturing, Working Paper No.119,Gujarat Institute of Development Research(GIDR), Ahmedabad, 2000.

Unni, J., Lalitha, N. and Rani, U. (2001)Economic reforms and productivity trendsin Indian manufacturing’, Economic andPoliticalWeekly, 36, 3915-22.

Visaria, P. and Jacob, P. The informalsector in India: estimates of its size, needsand problems of data collection, pp. 138-160. Herman, B. and Stoffers, W. (eds.)Unveiling the Informal Sector: More thanCounting Heads, Aldershot, Avebury, 1996.

1. The survey organized for gathering information on unorganized activities have often lacked tempo-ral comparability due to non-standardization of concepts, changes in the format of tabulation, etc.Sometimes, paucity of staff and other resources, made available for this purpose, have also becomeimportant hindrances (Kundu, 1998: 439-440).2. The choice of cut off year was based on the availability of data. Nonetheless, the year 1989-90 canbe considered as the terminal year of the pre-reforms period.3. According to Salter (1960), factor substitution – along with economies of scale and technicalprogress – accounted for differences in productivity growth across British industries. For Hahn andMathews (1964), productivity per man depends on capital intensity of all the machines in use and ontheir average age.4. For estimation, we have used DEAP 2.1, a program for data envelopment analysis developed byCoelli (1996).

End Note

structure, employment and productivity growth in the ...cmdr.ac.in/editor_v51/assets/mono-50.pdfan...

Documents