1
Indian Gems and Jewellery Industry: A Search for Porter Effect
Sudeshna Chattopadhyay1 Bidhannagar College, India
Sarmila Banerjee University of Calcutta, India
Abstract
This paper attempts to investigate whether the competitive edge of India’s gems and jewellery industry, one of the largest earners of foreign exchange, is based on a strong production base or merely reflects pseudo comparative advantage based on structural weaknesses related to social standards. The analysis of the available secondary data reveals that apparently the sector’s competitiveness has been grounded on the organizational peculiarities of the sector like extensive formal informal linkage in production chain that creates opportunity for the formal sector to retain its price advantage in the international market by outsourcing labour-intensive stages of production to the informal sector. In fact, no significant improvement of total factor productivity and technical efficiency was noted to lend support to this surmise. A primary survey conducted on the gems and jewellery manufacturing units of Kolkata revealed that the status of compliance to labour norms in these units is abysmally low and there is no significant sign of technological up-gradation in recent past.
JEL codes: D2; L2; J3;
Keywords: pseudo comparative advantage, labour norms, production chain, outsourcing, total
factor productivity, technical efficiency.
1 Corresponding author
Address for communication: Jamuna, Flat-8, Saha Abasan,1/AF Bidhannagar, Kolkata 700064
2
Indian Gems and Jewellery Industry: A Search for Porter Effect
I. Introduction
The gems and jewellery industry has witnessed spectacular export growth in the post
liberalization period and has emerged as one of India’s leading foreign exchange earners (see
www.rbi.org, www.ibef.org). The question that we would like to address in this paper is whether
this spectacular export growth has been prompted by increases in productivity and technical
efficiency of the sector or whether the export growth is driven by pseudo-competitiveness based
on in-built weaknesses lying in the inherent features of underdevelopment like existence of lax
labour standards.
In the initial stage of globalization the competitiveness of developing countries is often
determined by factors like existence of weak social standards. However, competitiveness based
on such initial conditions is generally not sustainable. Porter (1992) has pointed out that genuine
competitiveness can be achieved only by overcoming these initial bottlenecks. The process of
development must target to change these structural weaknesses and create new conditions to
guarantee self sustained long-term growth. According to Porter the necessary pre-requisite for
acquiring genuine i.e. long term competitiveness is higher productivity. Higher productivity can
be achieved through innovation and technical progress. In this case expansion of export from
these sustainable sectors will not only lead to higher growth but can also ensure improved living
standard.
In this backdrop before getting into a formal analysis of productivity performance and efficiency
of the gems and jewellery producing units of India it will be useful to recall a few special
features of the sector. The informal sector accounts for a large percentage of the total production
of gems and jewellery manufactured in the country. The industry is dominated by family
jewellers, who constitute nearly 96 per cent of the market (www.rncos.com). The informal
jewellery firms are often connected with each other and with the larger formal sector firms
through a long production chain where different firms specialize at different stages of the (same)
production process. Since the informal sector firms are by definition outside the purview of
formal regulation, in order to save labour costs the formal sector firms might have an incentive to
let out labour intensive stages of production to the informal sector firms through subcontracting.
3
Hence, in case of developing countries like India, vertical specialization in production chain
might be a manifestation of weaker regulatory practices and not stronger production base. In fact
rapid expansion of production and exports of a number of labour intensive low technology
industries like textiles, footwear etc in developing countries like India and China has been
largely due to the incorporation of the firms from these countries in national and global value
chains following rapid expansion of subcontracting activities. Globalization and improvement of
transport and communication technology has led to decoupling of labour intensive
manufacturing activities from marketing and retailing activities and relocation of the labour
intensive stages of production to the low wage developing countries. In case of apparel
production for example most firms with brand labels do not take part directly in any
manufacturing activity but focus on design, marketing and retailing. The labour intensive
manufacturing activities are mostly outsourced to firms in the developing countries to take
advantage of the cheap labour in these countries. Specialization in these labour intensive low
value added services has no doubt led to growth of jobs and foreign earnings in the less
developed countries but the consequent increase in employment has mostly taken place in the
informal sector which is marked by low wages, unskilled work and sweat shop conditions of
employment (see Christerson et al 1995, Scott 2006).
In this paper we are interested in exploring the reflection of a buoyant export performance of the
industry on specific factor productivity, total factor productivity and technical efficiency of the
sector. If strong production base is the right kind of explanation then one should observe
satisfactory outcome of such analysis. If not, we have to look into other institutional
characteristics of the sector leading to this promising trade pattern. Here we would study the
morphology of the production nexus between the formal and the informal sector firms to find out
whether the interconnectedness between them with some underlying strategic motives like
evasion of labour laws, may provide some satisfactory explanation to the apparent
competitiveness in the international market.
The rest of the paper is organized as follows: In section II an analysis of trend, growth and
productivity performance of the Gems and Jewellery sector of India is reported by using the
Annual Survey of Industries (ASI) data between 1984-85 and 2004-05. In section III firm level
ASI data have been used at 5-digit level of disaggregation to carry out an analysis of the
changing pattern of efficiency of the firms in this sector across different policy regimes. How far
4
investment in R&D is leading to this productivity growth is an interesting issue to explore. To
study the linkages among production, R&D and export, the ASI data seem inadequate as here the
export figures are not reported at the firm level. Hence, we have identified the major gems &
jewellery exporting firms from the PROWESS database of CMIE and studied the interlinkages in
section IV. In the following section (section V) we look into the relevant literature to understand
the morphology of gems and jewellery production in India. In section VI we report the results of
a primary survey conducted on the gold and silver jewellery manufacturing units of Kolkata to
get deeper insights into the organizational characteristics and also to assess the extent of trickle-
down effect on labor, if any. Section VII concludes the paper.
II. Trend and Productivity at the Industry level
An analysis of trend, growth rate, input wise productivity as well as total factor productivity of
the gems and jewellery manufacturing sector has been carried out for a twenty year period from
1984-85 to 2004-05. For the analysis, data from different rounds of Annual Survey of Industries
conducted by the Central Statistical Organization (CSO) has been used. For the industry level
analysis, related to specific and total factor productivity as well as returns to scale, ASI data at 3-
digit level of aggregation has been taken for the years 1984-85 to 1997-98 and at 4 digit level of
aggregation has been taken for the years 1998-99 to 2004-052.
To study the trend of gems and jewellery production we regressed the annual gross value added
( tY ) expressed at 1993-1994 prices, against time: tYt βα += . The growth rate is obtained by
regressing the natural logarithm of gross value added ( tYln ) against time: tYt βα +=ln .
2 For the years 1984-85 to 1997-98, ASI data is reported for NIC -87 category while for 1998-99 to 2003-
04, NIC- 98 has been used. Again from 2004-05 onwards NIC -04 classifications have been adopted. Since for the gems and jewellery sector 3 digit level classification of NIC -87 matches with 4 digit level classification for the subsequent NIC tables, hence , for our analysis we have used data at 3 digit level aggregation for the years 1984-85 to 1997-98 and at four digit level of aggregation for1998-99 to 2004-05.
5
Table 1
Trend & Growth of Gems & Jewellery Production: 1984-85 to 2004-05
Equation Coefficient t-value p-value 2R df
tYt βα += 5915.61 9.41 0.00 0.81 20
tYt βα +=ln 0.01 14.53 0.00 0.91 20
Source: Calculated from ASI data
The results reported in the Table 1 indicate statistically significant and positive trend as well as
growth, suggesting a good overall performance for the industry. A Cobb-Douglas type
production function is estimated next to study the nature of factor productivities experienced in
this sector. Two factors taken are labour ( L ) and Capital ( K ) with L represented by total
number of persons engaged and K by gross investment of the year where this gross investment
is defined as follows: }){(/1 1 ttttt DKKPI +−= −
where tI is the gross investment, tK is the capital stock of period t, tD is the depreciation and
tP is an appropriate deflator. To estimate tK , the perpetual inventory method3 has been used and
since the gems and jewellery production is reported in a residual industrial category along with a
number of other non-specific products, hence, the GDP deflator is used as the relevant deflator to
convert the series in constant prices.
The part of the change in gross value added that cannot be explained by specific factor
productivities like MPk and or MPL is ascribed to a change in the total factor productivity (TFP),
where ),( LKAFY = and TFP =A
Y
∆
∆. To verify hypothesis related to specific factor productivities
and returns to scale a typical Cobb-Douglas production function has been estimated with the
additional constraint that the sum of labour and capital elasticities equals one. To study the effect
of total factor productivity we have incorporated a time–variable in our regression equation. This
variable is expected to capture the residual contribution of the temporal improvement in
3 tK is obtained by the perpetual inventory method as follows:
0KK t = +∑−
=
−
1
1
t
i
itI , where oK is the capital stock in the benchmark year. Here we have selected 1973-74
as the benchmark year and 1993-94 as the reference period for GDP-deflator.
6
technology and production environment leading to an enhancement of the total factor
productivity.
The value of tY is regressed once on tL andtK and again on tL , tK and t . In both regressions
labour productivity is highly significant while capital productivity is insignificant. Coefficient of
t, representing total factor productivity, also turns out to be insignificant (see regression 2 of
Table-2). The regression results suggest that the growth of gross value added during the
reference period has been mostly driven by increases in labour productivity. No significant effect
of capital productivity and/or total factor productivity is observed. This isolates labour as more
vibrant factor of production and it becomes apparent from the poor performance of capital that in
spite of its spectacular export performance the sector has not experienced any discernible
technological up-gradation.
Table-2
Regression Results: Factor Productivities & Returns to Scale
***Statistically significant at less than 1% level. Source: Calculated from ASI data;
Since an explicit calculation of tA is also possible, hence, in the next step an attempt has been
made to decompose productivity growth into different components like tL , tK and tA , by using
Solow Index. Under the assumption of a CRS production function, the growth of gross value
added now can be decomposed into the contribution of specific factors as well as that of the total
factor productivity. So, t
tt
t
tt
t
t
t
t
K
Kr
L
Lw
A
A
Y
Y ∆+
∆+
∆=
∆ where
=∆
t
t
Y
Yrate of change of gross value added;
t
t
L
L∆= rate of change of labour force;
Study Var tYln Regression 1 Regression 2
tLln 1.17*** 1.27***
tKln 0.17 0.40
t -0.06
Constant -3.42 -6.10 2
R 0.96 0.96
Df 18 17
7
=∆
t
t
K
K rate of change of fixed capital; tw = share of labour in gross value added; =tr share of
capital in gross value added and t
t
A
A∆ rate of change of total factor productivity. From the ASI
data information on all variables except tA can be gathered. The series of tA values would
follow as a residual category.
By Solow index −∆
=∆
t
t
t
t
Y
Y
A
A [ ]t
t
t
t
t
tK
Kr
L
Lw
∆+
∆ and
∆+=+
t
ttt
A
AAA 11
.
If tA in 0=t is taken as unity, then the series for tA can be generated.
To construct the Solow Index, share of labour in gross value added has been taken as the share of
wages and salaries4 in gross value added. The share of capital has been obtained by subtracting
the share of labour from unity. Capital stock has been estimated by using the perpetual inventory
method. The estimated trend of the tA series is obtained as:
tA = -0.860 - 0.04 t , ,67.02 =R df=20.
(-9.56*) (-5.71*)
The trend is negative and statistically significant indicating that there is no sign of temporal
improvement in total factor productivity. Thus, the possibility of overall technological up-
gradation in the production front is not confirmed by the available data.
III. Technical Efficiency at the Firm Level
To examine whether there has been any significant change in the output-oriented technical
efficiency in the sector we have derived the firm level technical efficiency scores using Data
Envelopment Analysis (DEA) 5. The analysis has been done for the years 1989-90, 1993-94,
2000-01 & 2004-05 with unit level ASI data at 5-digit level of disaggregation. The number of
firms for which all relevant information was available for the year 1989-90 was 42 whereas the
number increased to 188 by 2004-056. Table 3 presents the technical efficiency scores for the
4 GDP deflator (base year 1993-1994) has been used has been used to express both gross value added and wages
and salaries at 1993-1994 prices. 5 Roy (2003) and Coelli (1998) present comprehensive discussion of DEA technique. 6 A firm is retained for DEA analysis provided information on all relevant variables like input-used, output-produced
etc. is available on it. In ASI unit level data, sometimes some of the relevant information on the reported firms is missing.
8
selected years in terms of size classes defined with 0.10 class widths. Table 4 reports the
corresponding descriptive statistics. Figure 1 shows the bar-diagram of the relative frequencies
of these efficiency scores whereas Figure 2 depicts the stacked bars showing the change in
relative frequency of firms with technical efficiency scores below 0.50 for the selected years.
Table -3
Distribution of Output-oriented Technical Efficiency Scores
Class Boundaries
Frequency of Firms Cumulative Relative Frequency (%)
1989-90 1993-94 2000-01 2004-05 1989-90 1993-94 2000-01 2004-05
0.0-0.1 0 7 32 90 0 8.75 31.07 47.87
0.1-0.2 1 15 16 33 2.38 18.75 15.53 17.55
0.2-0.3 1 9 20 23 2.38 11.25 19.42 12.23
0.3-0.4 0 15 9 10 0 18.75 8.74 5.32
0.4-0.5 1 8 7 7 2.38 10 6.8 3.72
0.5-0.6 0 4 1 0 0 5 0.97 0
0.6-0.7 5 2 1 3 11.9 2.5 0.97 1.6
0.7-0.8 5 5 1 1 11.9 6.25 0.97 0.53
0.8-0.9 8 2 1 1 19.05 2.5 0.97 0.53
0.9-1.0 21 13 15 20 50 16.25 14.56 10.64
Total 42 80 103 188 100 100 100 100
Figure-1
Bar Diagram of Relative Efficiency Score
9
Figure - 2
Stacked Bars of Relative Efficiency Scores
Table -4
Descriptive Statistics of Relative Frequency Distribution
Summary Statistics 1989-90 1993-94 2000-01 2004-05
Mean 0.843 0.442 0.32 0.235
Median 0.902 0.357 0.225 0.105
Std. Dev 0.203 0.311 0.321 0.3
Skewness -1.92 0.72 1.25 1.78
Minimum 0.16 0.042 0.001 0.003
Maximum 1 1 1 1
Count 42 80 103 188
Our analysis suggests that there has been an overall decline in the technical efficiency in the
sector over the period of our study. This is evident from figure 1 as the concentration of firms in
the lower range of efficiency score has increased while that in the upper range of the efficiency
scores has decreased. This is also clear from figure 2 which shows that the relative frequency of
the firms with technical efficiency score below 0.5 has steadily increased over time. The increase
in concentration of firms in the lower end of the efficiency score is also supported by the
increase in skewness of the relative frequency distribution over time. It was negative (-1.92) in
1989-90 and turned positive in the subsequent years, gone up to 1.78 in 2004-05. The mean
value has gone down from 0.843 in 89-90 to 0.235 in 2004-05, while the median has gone down
from 0.902 to 0.105 during the same period (see table 4). In spite of fabulous export performance
this fall in technical efficiency is indicating an interesting possibility: before reform and
10
globalization only efficient firms participated in production; however, after reform the scope of
becoming cost-competitive in the international market has been enhanced for the relatively
inefficient new entrants also, who are mostly organizing production through procuring export-
contracts, importing raw materials and out-sourcing the manufacturing process to the informal
units to make quick money from the entire transaction. This possibility, if confirmed, will raise
doubt regarding the sustainability of the recently experienced fortune of the gems and jewellery
industry.
However, the ASI database is not suitable for exploring this line of explanation as here no export
data is reported at the firm level and the information on imported inputs and R&D expenditures
are not always adequate. To overcome this limitation we have utilized the database on listed
gems & jewellery making companies consolidated by the Centre for Monitoring Indian Economy
(CMIE) in its PROWESS database for the post-reform period. The following queries may be
formulated to be verified on the basis of CMIE data:
1. The contribution of R&D in explaining both sales-performance and export-
performance of the firms;
2. Being a labour intensive industry (as most of the jewellery making process in
India is hand-crafted) whether the promise of the sector has influenced the fortune
of the workers or not;
3. The importance of imported raw materials in enhancing sales as well as exports of
the firms in the face of global competition;
IV. Profile of Major Exporting Firms
The CMIE firm level data reports information on a number of heads like income, export earning,
expenditure on raw material and labour, value of fixed assets etc. It may be noted that the
number of firms under gems and jewellery industries for which figures are reported in the CMIE
database varies from year to year. For example in 1991 the number of firms for which data is
reported is 9 while the number increased to 82 in 2005. The average export intensity (i.e. export
as a % of total sales) of the firms reported here is close to 50% or more for all the years under
11
consideration7. However, in spite of the high export intensity of the firms, investment in R&D is
very low. None of the firms reported any R&D expenditure during 1991-1995. Even in the post
1995 period very few firms reported R&D and the average of the share of R&D expenditure in
total expenses of the firms never exceeded 0.01%. Thus, it seems that the spectacular success of
the gems and jewellery industry has not been influenced by increased investments in R&D.
Gems and jewellery making being labour intensive, expenditure on labour is expected to shoot
up with this rapid growth of the sector. However, contrary to our expectation the average share
of compensation to employees in total expenses of the firm is rather insignificant. The gems and
jewellery industry reveals dependence on the imported inputs in the form of good quality gold
and uncut precious and semi precious stones. This is apparent from the fact that the average
share of imported raw materials in total expenses of the firms is fairly high in most of the years
under consideration and it is always much more high than the share of compensation to
employees.
We identified 20 listed firms from the CMIE database on which all relevant information like
total sales (TS), total exports (TEXP), total expenses (TC), R&D expenditure (TRD),
expenditure on compensation to employees (TW) and import of raw materials (TIMR). was
available for four selected years, viz., 1994 (the year immediately preceding the introduction of
WTO), 1996 (to study the early effects of the emergence of this supra-national institution on this
industry), 2000 (to capture the effects of newer policies like SPS and TBT agreements) and 2005
(when the effect of the new world economic order would be prominently felt on all important
transactions). This helped us to generate a panel of 80 observations. Since the within group
variations for all the selected variables are small while the between group variations are large,
our panel appears to be balanced and consistent. It is interesting to note that the mean of
compensation to employees is only 2.92% of the average sales figure while the average value of
imported raw materials is 42.07% of the value of average sales. It may also be noted from Table
5, which reports the pair-wise correlation coefficients between the relevant variables for all the
selected years, that for all years the share of imported raw materials are highly connected with
the volume of export of the firm and here the correlation coefficient varies between 0.93 and
0.96 with a very high level of statistical significance. The relation between total sales and export
though positive and statistically significant for all years, here a declining tendency is observed
7 The analysis in this section is carried on the basis of a panel of firms spread over fifteen years (1990-91 to 2005-
06).
12
since 1996. On the other hand for both compensation to employees and R&D the correlations are
very weak, statistically insignificant and for the former it changed from a positive to a negative
coefficient from 2000 onwards. For R&D the value of the coefficient, if non-trivial, is always
negative. These observations endorse our earlier surmise of the presence of pseudo comparative
advantage for this sector in a globalized world coming in the guise of weak regulatory control for
social standards. At the next step we will carry out a panel regression to confirm this position on
a stronger statistical footing.
In the regression analysis value of sales (TS) and Exports (TEXP) are taken as two dependent
variables and only for EXP time-dummies are incorporated to capture the temporal effects across
different policy regimes, if any. For the regression of TS (i.e., regression 1), R&D (TRD),
Compensation to employees (TW), import of raw materials (TIMR) and export (TEXP) are taken
as independent variables. In regression 2 the dependent variable is TEXP and the independent
variables are the different components of cost. Regression 3 is same as regression 2 with the only
difference that here three time dummies have been explicitly introduced. The results are reported
in Table 6. For all regressions, both fixed effect and random effect models are estimated and on
the basis of Hauseman ( )2χ test8 the appropriate method of estimation has been selected.
For sales (confer regression 1), all the explanatory variables like TW, TIMR and TEXP are
important factors; however, TRD does not appear to be statistically significant. Here, in terms of
Hauseman’s 2χ test the Random Effect model has been accepted and the performance of S has
been strongly influenced by TEXP. So, at least, a part of the growth of the sector is export-led.
8 with Random Effect as the Null Hypothesis;
13
Table 5
Pair-wise Correlation Coefficients at Firm-level
1994 1996
TS TEXP TC TW TRD TIMR TS TEXP TC TW TRD TIMR
TS 1.00 1.00
TEXP 0.90* 1.00 0.76* 1.00
TC 0.99* 0.89* 1.00 0.99* 0.71* 1.00
TW 0.52* 0.11 0.54 1.00 0.58* 0.15 0.59* 1.00
TRD -- -- -- -- 0.42 -0.04 0.43 0.96* 1.00
TIMR 0.89* 0.97* 0.90* 0.21 -- 1.00 0.84* 0.97* 0.81* 0.23 0.03 1.00
2000 2005
TS TEXP TC TW TRD TIMR TS TEXP TC TW TRD TIMR
TS 1.00 1.00
TEXP 0.68* 1.00 0.73* 1.00
TC 0.99* 0.69* 1.00 0.99* 0.69* 1.00
TW 0.51* -0.03 0.49* 1.00 0.64* -0.03 0.49* 1.00
TRD 0.47* -0.07 0.45* 0.99* 1.00 0.61* -0.07 0.45* 0.99* 1.00
TIMR 0.79* 0.95 0.79* 0.05* 0.01 1.00 0.79* 0.93* 0.79* 0.08 0.05 1.00
Table 6
Regression Analysis of Firm-level Performance
Y-variable Sales (TS) Export (TEXP)
Regression 1 2 3
X-variables
TEXP 0.34***
TW 7.78*** 5.13* 0.93
TRD 25.61 -147.64** -64.55
TIMR 0.99*** 1.28*** 1.39***
D1 0.29
D2 -7.06
D3 40.01**
0.89
Hauseman 0.23(RE) 7.94**(FE)
0.51 0.49
F 67.34*** 107.85***
Wald-
779.26***
No. of groups 20
No. of observations 80
***: 1%, **: 5% and *10% level of statistical significance;
2
R
2χ
ρ
2χ
14
In regression of TEXP on the different cost components (confer regression 2) imported raw
materials (TIMR) appeared to be most significant followed by TRD, though for this latter
influence the sign of the coefficient is highly disturbing (-147.64). Compensation to employees
(TW) imparted only weakly significant effect as the coefficient here is positive but statistically
significant only at less than 10% level. So, the export performance is neither explained by
improved technology represented by higher R&D nor by the improved lot of the workers. Here
the Fixed Effect model turned out to be the more appropriate one.
Finally, to explore the impact of policy changes on export of the sector the dummy variable
regression is reported (confer regression 3). Here, TRD lost its statistical significance and so is
W. However, the dependence on imported raw materials is still very strong and the time dummy
D3 exhibited strong influence on the TEXP indicating a strong impact of globalization on the
sector in recent time.
The regression results from CMIE data indicate that while the labour input significantly affect
sales of gems and jewellery the effect of this input on exports of gems and jewellery is
insignificant. Our analysis in terms of the ASI data had also revealed that labour significantly
affect the production of gems and jewellery. The reason behind the difference in the role of
labour on sales and exports might lie in the fact that the predominately exporting firms function
differently from the firms which mainly target the domestic market. It is possible that the
exporting firms function in a putting out system and work as a part of a global production chain
where the main function of the firms is to import raw materials, get the production stages done
by the informal sector units and re export the finished product. For example, the diamond
processing firms import rough diamonds and after processing re export the processed diamonds.
In fact 85% of the world’s diamonds are processesed in India and the diamond cutting firms in
India has strong production links with the leading diamond retailers and manufacturers like De
Beers, Alrosa, BHP Billiton (www.afdiamonds.com.,http://catchef.ft.com)
The exporting firms may function as intermediaries between large multinational jewellery firms
and small informal production units. This means that the exporting firms are neither engaged in
production nor in the sales of the products to the final consumers. The firms involved in
domestic sales on the other hand are probably directly involved in marketing of the final product
for the domestic consumers. These promotional activities are generally much more labour
intensive. This might explain why the labour coefficient is significant for sales (a part of which
15
represent domestic sales) while it is insignificant for exports. The practice of subcontracting the
manufacturing stages to the informal units might also explain the low propensity of R&D and
lack of any significant improvement of factor productivity or technical efficiency among the
jewellery manufacturing units. For example, for Titan Industries Limited the average export
intensity of export over our study period was 7.46% and that of Su-Raj Diamond and Jewellery
Limited the corresponding figure was 97.28% identifying the former as a domestic market
oriented firm and the latter as an export-oriented one. For Titan the share of employees’
compensation in total cost is 9.45% and for Su-Raj it is only 0.85%. Moreover, for Titan the
R&D expenditure accounts for 0.24% of total cost and that for Su-Raj is absolutely zero. Cost
share of imported raw material for Titan is only 11.68% and for Su-Raj it is as high as 69.18%.
The incidence of lower R&D propensity among the export oriented firms is consistent with the
findings reported in the recent value chain literature which suggest that firms working in global
value chains generally have little scope to innovate as they have to produce according to
specification of the lead firms. Working in national value chains on the other hand can be more
conducive towards the development of innovation capabilities as the domestic firms have more
symmetric power relationship with the domestic customers and hence enjoy the freedom to
experiment and innovate (see Altenburg et al 2008). Combining all these observations together it
seems to us that the more export oriented production unit is importing raw material, out-sourcing
the follow-up processes to the informal sector and re-exporting the finished product.
This possibility could become the reality provided the production nexus of the sector would
reveal close connectivity between formal and informal units. So, in the following section we will
look into the relevant literature to investigate whether the formal and informal sector firms in the
gems and jewellery industry are interlinked through the subcontracting chain, identify the factors
which might form the basis of such linkage and examine whether the inter connectedness
between the formal and the informal sector explain the pseudo competitiveness of the industry in
international market.
V. Morphology of Production Nexus
An interesting feature of the gems and jewellery manufacturing industry in India is that despite
its huge size, the sector is largely unorganized9.
9 Organized players such as Tata with its Tanishq brand constitute only 4 percent of the market
(www.commodityonline.com/news )
16
The jewellery that finds a market across the country and is exported to other parts of the world is
often crafted in tiny informal workshops. These informal jewellery firms carry out different stages
of the manufacturing process on behalf of the large formal sector firms in return of making
charges. Thus, an important feature of the gems and jewellery industry in India is the
interconnectedness between the formal and the informal sector firms through the production
chain. Though it is difficult to get exact figures due to the informal nature of the units, the
contribution of these small informal sector units in total export of jewellery is quite high. As per
figures from the Gem and Jewellery Export Promotion Council, exports during 2000-01 from the
Domestic Tariff Area (DTA) in Mumbai amounted to $266 million. While the council was unable
to put in percentage terms the quantum of exports from Zaveri Bazaar which is an area where the
small informal jewellery manufacturing units of Mumbai are concentrated, the council stated that
the area accounted for a major portion of this export (Katakam 2001). Mumbai is no exception. In
fact, in all the regions of India which specialize in crafting of jewellery or polishing of precious
and semiprecious stones there is huge concentration of small informal sector firms accounting for
major portion of employment, production and export. For example in Jaipur, which is the second
most important gem exporting centre in India and number one in terms of sales to the foreign
tourists, approximately 90% of the 0.2 million workers employed in gem polishing industry there
work in unorganized units. (www.lac.org.hk).The informal artisans not only get meager payments
for their work but are also exposed to serious occupational health hazards as in most of the
informal units labour safety and pollution prevention norms are not followed.
The most distressing feature of the gems and jewellery industry is that a large percentage of the
workers employed in the industry especially in the gem cutting and polishing segment are
children. The children are preferred here because of their lower wage and tender fingers especially
suitable for cutting and polishing the smaller stones. In the flourishing gems polishing industry at
Jaipur nearly 20% of the workforce are children (Burra 1988, Kruijtbosch 1996). In fact, in the
gem processing units of Jaipur there has been an increase in the number of child labour following
increases in international demand for gem stones in the post liberalization period (Kruijtbosch
op.cit).
17
One of the main reasons why India is able to compete in the international market in spite of not
getting the best quality raw material is this availability of cheap labour who can be subjected to
inhuman working conditions. While the artisans get meager payments and exposed to serious
health hazards the profit that the traders and the exporters make on the fine craftsmanship is over
a thousand percent (Burra, op.cit.).
From the above discussion it is apparent that the two main features of the gems and jewellery
manufacturing industry in India that may be held responsible for its spectacular success in the
international market are extensive inter linkage between the formal and informal sector through
the production chain and weak regulatory pressure of compliance to labour standards. So, hidden
behind the dazzling success story of Indian gems and jewellery sector in the international market
there seems to be a real story of exploitation of the workers. Hence to get further insights into the
organizational set up of the industry and to assess the effect of the spectacular growth of the
industry on the condition of labour a primary survey on the jewellery-making units of Kolkata
has been conducted in 2008-09. The findings of that survey are reported in the following section.
VI. Primary Survey
We have chosen Kolkata as our area of survey10 as a number of both registered and unregistered
gems and jewellery producing firms are located here. Moreover, jewellery exports from the
eastern region have grown at over 21% compared to an all India growth rate of 6% and Kolkata
accounts for almost all of the jewellery exports from the eastern region (see www.wbidc.com)
Kolkata mainly manufactures plain gold and silver jewellery and most of jewellery produced
here is crafted in the small informal gold and silver jewellery manufacturing units.
In Kolkata most of goldsmiths are concentrated in a few localities: Bowbazar & Taltala in
Central Kolkata, Garanhata, Sovabazar & Sinthi in North Kolkata and Bhowanipur-Kalighat,
Ballygaunge-Dhakuria & Behala in South Kolkata. We have selected the following three sites for
our field survey; Bowbazar from Central and Garanhata and Sinthi from North. Bowbazar
represents the central business district where majority of transaction takes place. Garanhata is the
oldest location where the trade concentrated even in pre-independence days and finally Sinthi in
10
The primary survey has been conducted as a part of the research project titled “Environmental Regulation and Process Sub-Contracting: A Study of Gems and Jewellery Manufacturing Units in Kolkata”, UGC Sanction No: PWH 007/06-07(ERO) sponsored by the University Grants Commission (UGC) under the UGC Minor Research Project scheme.
18
the North has become a major production centre over the last quarter of the century, especially
after the influx of refugees from Bangladesh. A number of the immigrants were connected with
gold smithies in Bangladesh and most of these families settled down along the railway network
connected through Dumdum junction. Sinthi being close to Dumdum, flourished over recent
time. There are approximately 20000 goldsmiths in Bowbwzar11. In Garanhata there are about
3000 goldsmiths while in Sinthi there are about 25000 goldsmiths.
The gold jewellery manufacturing process can be divided into the following major stages:
a) Refining: If the gold (silver) is of inferior quality or if the gold (silver) is recycled gold
(silver) from old jewellery then the gold (silver) has to be refined before it can be used
for ornament making.
b) Crafting or making of jewellery: The Indian goldsmiths are famous their jewellery
crafting skills. Most of the jewellery produced in India are meticulously hand crafted.
c) Finishing: Finishing includes polishing of jewellery and chila work
d) Galai or recycling: Some amount of gold is wasted during the crafting, polishing and
refining processes. During the crafting or polishing of gold jewellery fine gold particles
come out from the ornament and gather on the floor of the workshop. Some amount of
the gold also gets dissolved in the acid that is used for polishing ornaments and refining
gold. The shop owners collect the gold particles mixed with dust and other impurities and
take them to some special units referred to as Gold Galai units that are especially trained
to separate the gold from the impurities.
Initially we planned to collect information of 300 units each, from both Bowbazar and Sinthi and
another 200 units from Garanhata. However, we had to remain satisfied with 586 observations in
all. In Bowbazar and Garanhata no major problem was faced in conducting the survey as per our
plan. Though the goldsmiths were initially reluctant to give interviews we were successful in
persuading them to cooperate and we could collect information from 301 units in Bowbazar and
200 units in Garanhata. However, in Sinthi due to extreme reluctance and active resistance from
the goldsmiths we could only cover 85 units12.
11 There statistics on the number of goldsmiths in different locations in Kolkata has been provided by Bangiya Swarna Silpi Samiti which is a major association of gold and silver jewellery manufacturers in West Bengal. 12 The reluctance of the goldsmiths to face the interviewers might be due to the fact that most of the units did not have the necessary licenses and did not follow the laws especially those related to labour and environmental standards. The goldsmiths were apprehensive that due to increased public attention through media coverage they
19
The salient features of the majority of units surveyed are their informal status and extensive inter
linkage through the production chain among themselves and with the larger formal sector firms.
Other important features relate to the poor compliance of the informal units to labour norms and
lack of investment in infrastructure, technology and human capital. These features of the
informal jewellery units are briefly discussed below:
Inter linkage between the formal –informal units:
A unit to be labeled as a formal unit, must be registered with a number of official bodies and
obtain a host of licenses like the Trade license, Manufacturing license, etc. Some of these
licenses are obtained at an individual level (required for both firm-owners and self-employed
persons) and some others are at the firm level. Most of the units surveyed did not have all the
required licenses.
Table-7
License Status of the survey units
Name of the License(s) No. of units in Bowbazar (%)
No. of units in Garanhata (%)
No of units in Sinthi (%)
Total no. of units (%)
NOC 91 (30.2) 51 (25.5) 4 (1.3) 146 (24.9)
Trade License 292 (97.0) 199 (99.5) 83 (99.8) 574 (97.9)
Karigari License 108 (35.8) 47 (23.4) 19 (22.4) 174 (29.7)
Health License 190 (63.1) 104 (52.0) 27 (31.8) 321 (54.8)
NOC+Trade License 90 (29.9) 51 (25.5) 4 (1.3) 145 (24.7)
NOC+Karigari License 31 (10.2) 3 (1.5) 0 (0.0) 34 (5.8)
NOC+Trade +Karigari+Health 30 (10.0) 3 (1.5) 0 (0.0) 33 (5.6)
Total 301 (100%) 200 (100%) 85 (100%) 586 (100%)
Source: Primary Survey
It is interesting to note that out of the total 586 units surveyed as many as 416 units (71 %) either
did not have any name or were reluctant to disclose it. Since Consent to Operate license from the
Pollution Control Board (NOC) or the manufacturing (Karigari) license from the KMC is issued
in the name of the individual manufacturing units, out of these 416 units 286 have either NOC or
might be subjected to serious regulatory monitoring. The resistance from the goldsmiths was so hostile that we were forced to abandon our survey after interviewing only 85 units there. Sinthi is reportedly infamous for clandestine cross border trade in gold. This might be the reason why we had to face extraordinarily active resistance from the goldsmiths.
20
Karigari license or both and the remaining 130 do not have any license at all (see Table-7).
Hence, these 130 are purely informal unregistered units and the remaining 456 are at most partly
registered. In fact only, 33 manufacturing units claimed to have all kinds of licenses, and,
therefore, formal in nature. However, it will be shown later that even these units are employing
labour on an informal basis.
Since the jewellery making process is divided into a number of operations, each demanding
some specialized craftsmanship, most of the units surveyed concentrate in a few of them. Table 8
reports the nature of specialization of the survey units. Out of 586 units surveyed 58.2% are
spcializing in any particular stage of operation (refining, making, finishing or recycling) and the
rest is involved in more than one stage. However, only 4% are carrying out the entire process
from cradle to grave under the same roof. Majority of units (69%) are involved in crafting the
ornaments while refining and finishing jobs are concentrated in fewer specialized hands. The
small informal units get manufacturing orders from the large jewellery shops on the basis of a
putting out type contract. Often, the dealers act as intermediaries between the large firms and the
small informal firms. The dealer takes the order from the large firm and then gets the entire
process done from the different units specialized in the respective stages. Thus, the small units
are interconnected with one another and the larger firms through a long subcontracting chain.
Table- 8
Nature of Process Specialization of the Survey Units
Name of the processes Numbers Percentage
All 24 4.1
Refining(R) 55 9.4
Making(M) 163 27.8
Finishing(F) 117 20.0
R+M 148 25.3
R+F 4 0.7
M+F 69 11.8
Galai (Recycle) 6 1.0
Total 586 100
Source: Primary Survey
Out of the total 586 firms surveyed only 42 firms, i.e., 7.2% were not dependent on the larger
firms and sold their manufactured items directly and exclusively to the customers. The majority
21
of the firms (73.7%) handed over all their products produced on a putting-out basis to other
larger firms or the dealers and thus had no direct transaction with the customers. In fact, the
remaining 19.1% of firms are engaged in both types of dealings. Thus, almost all the production
stages are carried out in the different informal units while the basic metal is generally supplied
by the registered firms. The formal sector firms procure orders, organize production through sub-
contracting chains and also take care of the marketing and related promotional activities.
Social Security Provisions:
The small informal units are generally owned by one or more goldsmiths. Tiny ill ventilated
rooms generally hired on a monthly rental basis serve as the workshop for the goldsmiths. The
owner goldsmiths hire workers to work under him. The owners procure orders from large
jewellery shops either directly or through middlemen or dealers and get the job done by his
workers. In most cases the owners also participate in the manufacturing process. Often the
workers are not paid in cash. Instead, the worker gets a certain percentage of the gold that is
wasted during the manufacturing process. Out of the 586 firms surveyed we were able to collect
information on wages from 495 firms. The average wage of the worker is about Rs.2000 per
month. Thus the workers on an average are forced to live on less than Rs.67 per day which is
lower than the national floor of minimum daily wage of Rs 100. (www.paycheck.in). The
workers are not only poorly paid they are also denied the benefits of statutory provident fund and
employees state insurance schemes. Not a single firm reported to have these facilities for the
workers. There are at least two major Trade Unions of the goldsmiths in West Bengal with their
Head Offices in Kolkata, viz., Bangiya Swarna Silpi Samiti and Akhil Bharatiya Swarnakar
Sangha. But the members of these organizations are generally the owner goldsmiths and so it
seems that the interests of the worker engaged in gold smithies do not get enough attention13.
Information on the monthly income of the owners of the informal units could be obtained from
543 firms. The average monthly income of the owners is stated to be about Rs.4800.00 per
month the range being Rs.1500 to Rs.12000. While the income of the owners is generally higher
than the workers it is much lower than the profit made by the dealers or the larger formal sector
firms. The workers are not only ill paid but are also regularly exposed to serious pollution
hazards. The fumes emitted from the acid used during refining, polishing and galai operations are
extremely toxic. Another source of pollution in gold manufacturing emanates from the use of
13
During our survey most of the owners were reluctant to let their workers face the interviewers independently and most of the survey questionnaires were filled up on the basis of the interviews given by the owners alone.
22
Cadmium as a soldering agent. According to the World Gold Council, Cadmium fumes are
highly toxic. There is another safer alternative agent like indium soldering, but the artisans are
forced to use Cadmium to retain their cost-competitiveness. The Bureau of Indian Standards has
introduced hallmarking and there is a rule which forbids issuance of hallmark to the jewellery
with Cadmium. However, there are umpteen instances of evasion in this case (see Basu 2007). In
our survey we have found that out of 398 firms involved in the making of the ornament about
98% use Cadmium. Moreover, the small manufacturers believe that use of Cadmium is a must in
obtaining the Hallmark14 sign. So, not only they are flouting norms, they are bearing wrong
impression regarding the recommended practices to follow.
Thus, in comparison to the huge profit that the large formal sector jewellery firms and exporters
make out of their craft both the small owners and workers involved in this trade are ill paid and
are often exposed to serious pollution in return of their hard work and exquisite artistic talent.
Up-gradation of Physical and Human Capital:
Improvement and up gradation of technology, infrastructure and human capital is indispensable
for accelerated growth of any sector. In the informal units surveyed, however, investments in all
these areas were found to be abysmally low. None of the firms reported to make any investment
in infrastructure development in the last ten years. Only 77 out of the 586 firms surveyed
reported purchase of some machineries like digital weighing machine, newer variety of polishing
machine popularly referred to as magnet machine, etc., over the last five years. None of the firms
reported any significant change in the ornament making techniques followed by them over the
decades. The only change seemed to be the introduction of Cadmium as a soldering agent in
place of the traditional metal solders. However, as we have already noted that the exposure to
cadmium fumes leads to serious health hazards
Even though the artisans form the backbone of the gems and jewellery industry the
manufacturing units not only show reluctance to invest in physical capital but in human capital
as well. Very few of the artisans working in the informal units have completed school education
and none of them has received any formal training in jewellery making. A new entrant to the
jewellery making profession generally has to work as an apprentice in the workshop of a
goldsmith for a few months before he starts working as a full fledged goldsmith. In fact, for the
informal artisans apprenticeship in the workshop of a fellow goldsmith offers the only scope for
14
Hallmark is a stamp specifying the purity of the gold ornamnet
23
training. As a result the goldsmiths miss the opportunity to supplement their traditional skills
with modern knowledge. There are few institutes like Shingar Institute in Howrah, Jesani
Institute in Mumbai, National Institute of Fashion Technology with branches in Delhi and
Kolkata, which offer jewellery designing and making courses but most of the informal artisans
do not satisfy the minimum educational qualification necessary for being eligible to enroll for
those professional courses.
VII. Concluding Observations
It is shown in the paper that in spite of the commendable performance of the gems and jewellery
sector at the domestic and international level the capital or total factor productivity of the sector
has not shown any improvement in the post liberalization period15. This suggests that the sector
has not experienced any discernible technological up gradation during our period of study. Our
analysis also revealed that the technical efficiency of the sector has declined during the period
1989-90 to 2004-05. ASI data does not report the export figures of the firms. So, to investigate
the factors that influenced the magnificent export performance of the gems and jewellery sector
we conducted an analysis of the firm level CMIE data. Our analysis of the CMIE data revealed
that the excellent growth of the gems and jewellery industry cannot be explained by factors like
investments in R&D. Interestingly, even though the Indian gems and jewellery industry is labour
intensive no significant contribution of labour on the exports of the industry could be identified
on the basis of CMIE data. Rather, the results suggested that the superior performance of the
sector has been attained mainly on the basis of imported inputs mostly outsourced to the
innumerous informal sector units where production is carried out using a rather traditional labour
intensive process with the help of lowly paid cheap labour. To confirm this hypothesis we
conducted a primary survey on the gold and silver jewellery manufacturing units in Kolkata.
Our primary survey, suggested that the formal and the informal sector firms are interlinked
through an extensive subcontracting chain. The results of our survey also revealed that the
informal jewellery making firms have all the features of a subsistence sector like low wage, low
investment in physical and human capital formation, poor working conditions and lack of
arrangement for worker safety. Thus, through inter-linkage the formal sector is getting access to
15
In fact, the Indices of revealed comparative advantage show very high RCA values of gems and jewellery exports from India amounting to 8.39 for Jewelry, goldsmiths' and silversmiths' wares, etc. (SITC897) and 16.49 for Pearls, precious and semi-precious stones in (SITC667) for the year 2002. (Chattopadhyay 2010).
24
the cheap labour in the informal sector and gaining competitiveness by flouting all labour related
norms without showing any sign of technological up-gradation.
According to Porter a country has to pass through several stages before it can acquire long run
competitiveness. In the first stage, competitiveness is essentially cost based competitiveness and
as Sanyal (1993) explains, in case of developing countries, competitiveness at this stage is based
on factors like surplus labour or absence of environmental concern. But such competitiveness
cannot be sustained in the long run as it makes the country dependent on low productivity
activities and, hence, cannot ensure sustained increases in standard of living. Globalization on
the basis of such competitiveness cannot put the country on the path of sustainable development
unless the process is accompanied by technology transfer and investment driven restructuring of
the economy.
Our analysis suggests that the competitiveness in gems and jewellery manufacturing industry is
still in the early stage of development as the source of this competitiveness is lying in the
existence of extensive formal informal linkage in production creating scope for regulatory
leakages. If the sector continues to grow only on the basis of temporary advantages derived from
‘weaknesses’ of the economic structure then such a growth process would merely lead to
expansion of low productivity employment in the informal sector with eventual deterioration of
the quality of life. Moreover as Christerson et al (1995) rightly pointed out, the competitive
advantage of low technology, labour intensive industrial clusters are also easily contestable as
these clusters are under constant threat of being out-competed by newer cheaper wage locations.
Hence, technological and qualitative up gradation of the gems and jewellery sector is essential
pre-condition to ensure sustainable growth of the sector.
25
REFERENCES
Altenburg T., Hubert Schmitz and Andreas Stamm (2008): Breakthrough? China’s and India’s
Transition from Production to Innovation in World Development Vol. 36, No. 2 Basu R., (2007): Jewellers at Toxic Risk The Telegraph 7th Nov, p. 25. Burra, N., (1988): Exploitation of Children in Jaipur Gem Industry I: Structure of Industry in Economic and Political Weekly Vol. 23, No: 3, January 16 Burra, N., (1988): Exploitation of Children in Jaipur Gem Industry II: Health Hazards of Gem
Polishing in Economic and Political Weekly Vol. 23, No: 4, January 23 Coelli, T. (1998): A Multi-Stage Methodology for the Solution of Orientated DEA Models, CEPA Working Paper No: 1/98 Department of Econometrics, University of New England, Arimdale, Australia. Chattopadhyay, S. (2010): Globalization and Environment: A Study on Gems and Jewellery Units of Kolkata, unpublished Ph.D dissertation, University of Calcutta. Christerson B., Richard P. Appelbaum (1995): Global and Local Subcontracting: Space,
Ethnicity, and the Organization of Apparel Production in World Development Vol.23, No. 8 Katakam, A., (2001): Sweat and Gold, Frontline Volume 18, Issue 22, November 9
Kruijtbosch, M.(1996): Child and Adult Labour in Export-Oriented Garment and Gem Polishing
Industry Of India with Case Studies From Tirupur, Bangalore, Jaipur and Trichy. Indian Committee of the Netherlands, Utrecht, November 1996, 2nd version (www.indianet.nl/chadlab.html) Porter M. E., (1992): The Competitive Advantage of Nations, Macmillan, London. Ray,S.C., (2004): Data Envelopment Analysis, Cambridge University Press; Sanyal, K., (1993): Paradox of Competitiveness and Globalization of Underdevelopment. Economic and Political Weekly June 19. Scott A. J. (2006): The Changing Global Geography of Low-Technology, Labour-Intensive
Industry: Clothing, Footwear and Furniture in World Development Vol.34 No.9