Forms
of Dualism: An Analysis of the Structure of India's
Unregistered Manufacturing Sector Based on the
56th Round Results
Despite more than five decades of state-led industrial
development, 'dualism' even 'multi-structuralism', has
been an abiding characteristic of India's industrial
sector. Early analyses of the structure of industry at
independence had pointed to the domination of 'lower
forms of production', or production characterized by the
use of non-power-driven techniques, and the absence of
hired labour in predominantly rural and semi-urban
areas. Thus, according to the national income data
relating to 1948–49, 'factory establishments' accounted
for just 6.34 per cent of national income, while 37 per
cent was generated in the mining and manufacturing sectors1.
Further, going by the 1951 Census,
factory establishments accounted for just 26 per cent of
the employment in mining and manufacturing, which in
turn amounted to only 9.3 per cent of total
employment2.
Although organized sector units, or those that
correspond to the registration criteria set by sections
2m(i) and 2m(ii) of the Factories Act of 1948, are not
always units that reflect the attributes of modern
industrial establishments, this is the closest one can
get to making a division between the unorganized and
organized, or informal and formal, a classification that
must necessarily be arbitrary. However, since by law
these units have to meet the requirements of labour
legislation in the country, this is at least one sense
in which they may be termed formal or organized.
Conventional wisdom based on the experience of developed
countries would have it that industrialization is
process in which an economy witnesses the gradual demise
of excessively small units and units based on primitive
techniques and primitive forms of organization, which
give way to large or larger units working with hired
labour and more advanced techniques and increasingly
adopting the impersonal joint-stock company form of
organization. It hardly bears stating that state
intervention to ensure minimum wage payments, a 'normal'
working day and reasonable conditions of work, by
encouraging productivity-enhancing technical change,
contributes to the above-described transition.
This would imply, as a corollary, that the implicit
'exemption' from regulation of units that do not meet
the criteria set by sections 2m(i) and 2m(ii) would
itself slow the pace of 'modernization' and contribute
to some degree of persistence of dualism. This is a
factor that favours the use of the Factories Act
definition to draw the dividing line between the
organized and unorganized sectors. In addition, since
historically the statistical system has sought to cover
registered units through the Annual Survey of Industries
(ASI), and unregistered units through the Economic
Censuses and the follow-up surveys undertaken by the
National Sample Survey Organization (NSSO), a body of
statistical evidence exists to assess the structure and
evolution of the unregistered sector.
Needless to say, dualism is not just a function of the
implementation of the Factories Act, allowing for wage
differentials between the 'organized' and 'unorganized'
sectors. Rather, as analysts have often pointed out, a
number of other factors on the demand side, such as the
persistence of 'traditional tastes' that can be catered
to only by more primitive techniques, and the existence
of poverty and inadequate market integration, which
generate a different kind of 'niche' market for cheap
but primitive products, have also encouraged dualism.
Further, inasmuch as poverty-inducing factors like
unemployment and underemployment result in a constant
search by the poor for supplementary work, and since
barriers to entry into many areas of unorganized
production are limited, the supply of commodities from
these unorganized sector units tends to be elastic, with
production occurring at extremely low wages and prices
that reflect small or negligible 'margins', as when the
above-mentioned demands manifest themselves. In the
event, the unregistered manufacturing sector would be
characterized by three kinds of 'instability'-that
stemming from the fact that some of its activities are
seasonal; that resulting from the high rate of mortality
of unregistered units, given low wages, small margins
and extreme competition, and that resulting from the
precariousness of the demands it caters to.
Besides all these, backward techniques and backward
forms of organization survive also because of the
subordination of these activities by merchant or trading
capital, which is able to earn for itself a suitable
margin even while remaining competitive with production
based on modern techniques because of its ability to
exploit the benefits of the low wages associated with
production in the unorganized or unregistered sector.
This is particularly true of primitive production units
that cater to national markets, like handloom, beedi and
match production units. In fact, the irrational
concentration of match production in a few contiguous
districts of Tamil Nadu, accounting for an overwhelming
share of the matches produced in the country and
catering to markets nationwide, can only be explained by
a trading chain dominated by merchant capital that
controls a complex, historically evolved production
system involving home- and unit-based female and child
labour and extremely low costs of production.
Thus 'dualism' or 'multi-structuralism' is not
determined primarily by the existence or otherwise of
factory legislation, but is a consequence of the larger
socio-economic context and its evolution. The Factories
Act merely provides an arbitrary dividing line to be
drawn between the organized and unorganized sectors, and
an ambiguous contribution to the differences between the
two sectors. 'Ambiguous' because, evidence regarding the
enforcement of the Factories Act does not inspire
confidence in the ability of the implementing mechanism
to ensure minimum wages and reasonable conditions of
work in the factories that are formally registered with
the Chief Inspector of Factories.
This note has a limited objective: to delineate the
structure of the unregistered manufacturing sector on
the basis of data yielded by the 56th Round
of the NSS, conducted during July 2000 and July 2001.
The 56th Round covered all unorganized
manufacturing enterprises (UMEs) under the two digit
codes 15 to 37 and enterprises under cotton-ginning,
cleaning and
baling3.
Out of these 5,586 rural and 8,942 urban units were
finally surveyed. The survey covered manufacturing
enterprises (MEs) not covered in the
(ASI)4
(i.e. those not registered under Section 2m(i) and
2m(ii) of the Factories Act 1948), manufacturing
enterprises registered under Section 85 of the Factories
Act, 1948, and enterprises manufacturing beedis and
cigars that are not covered under the ASI.
The NSS estimates the total number of workers employed
in the unregistered manufacturing sector in 2000–01 at
3.71 crore. If we exclude industries with NIC codes
01405 and 37, and compare the remaining industries with
the same set of 22 two-digit industries in the ASI, we
find that the 3.71 crore workers in the unregistered
sector in 2000–01 compare with 79 lakh workers in the
registered factory sector in 1999–2000. Assuming no
change in the number of workers between 1999–2000 and
2000–01 in ASI factories5,
the share of the unregistered sector in total
manufacturing employment thus works out to 82 per cent.
Similarly, gross value added in the unregistered sector
of the 22 two-digit industries mentioned above in
2000–01 by the product approach works out to 24.3 per
cent of the combined gross value added in registered
manufacturing in 1999–2000 and unregistered
manufacturing in
2000–016.(Interestingly,
this is substantially different from the contribution of
the unregistered sector to manufacturing GDP, which is
placed by National Accounts Statistics at 35 per cent in
2000–01.) Clearly, therefore, if the figures relating to
the immediate post-independence years are reliable, the
importance of the unregistered manufacturing sector has
increased significantly in employment terms, even if its
contribution has fallen in terms of value added shares.
While the unregistered sector accounts for a significant
share of manufacturing employment and production, the
structure of the sector remains heavily biased in favour
of small and organizationally 'primitive' units. Across
rural and urban areas, own account manufacturing
enterprises (OAMEs), or those that employ no hired
labour on a 'regular basis', account for an overwhelming
86 per cent of all enterprises. Though the presence of
OAMEs is greater in rural areas, their share in total
enterprises in urban areas too amounts to a remarkable
71 per cent. Further, both in rural and urban areas,
more than 70 per cent of the units employing hired
workers are non-directory manufacturing establishments,
or those employing less than six workers.
Table 2 provides a picture of the
two-digit industrial categories in which OAMEs account
for less than 80 per cent of all enterprises in rural
areas, and 66 per cent of all enterprises in urban
areas. While in the rural areas there are just 7 of the
24 industries covered by the 56th Round that
meet these stringent criteria, there are 14 out of 24 in
the urban areas in which less than two-thirds of the
units are OAMEs. Further, these sectors account for less
than 1 and 15 per cent, respectively, of all enterprises
in rural and urban areas. Clearly, the spread of
production based on use of hired labour on a fairly
regular basis is limited within the unregistered sector
in both rural and urban areas.
In addition to the persistence of small-sized units and
non-hired labour-based forms of organization, the
structure of the unregistered sector points to a
relatively high degree of specialization of activity
when analysed in terms of two-digit categories. Thus as
far as OAMEs are concerned as many as 60 per cent of the
total are food, tobacco, textile and garment units, in
both rural and urban areas. In the rural areas, if we
add wood and wood product units, the cumulative share
rises to as much as 83 per cent. Thus, small OAMEs are
concentrated in traditional areas such as food products,
textiles and wood products, all of which are agro-based
rather than chemical or metal-based. Even NDMEs and DMEs,
account for 60 to 75 per cent of the total in rural
areas and around 45 per cent in urban areas. The other
two areas with a significant number of enterprises are
non-metallic mineral products, especially in the DME
category in rural areas, and fabricated metal products,
excluding machinery and equipment. The former, we must
recall, includes activities such as manufacture of glass
and glass products (including glass bangles), ceramic
ware, bricks, roofing tiles and lime and plaster, all
activities that fall within the broad definition of
'traditional' and can be seen as predominantly catering
to local markets.
The concentration of activity in
these areas is visible even when assessed in terms of
the share of employment in unregistered units in
different two-digit industrial categories. Food,
tobacco, textile and wood products account for between
50 and 82 per cent of employment in OAMEs and NDMEs, and
if we include non-metallic mineral products and
fabricated metal products, the ratio rises to 85 to 90
per cent for all kinds of units.
The primitive nature of a
substantial segment of the unregistered manufacturing
sector is also reflected in the fact that 70 per cent of
all units are typically 'household units', in the sense
that they are located in household premises. This high
proportion is of course explained primarily by OAMEs
dominating the sector in terms of number of enterprises
and as many as 80 per cent of OAMEs in rural areas and
70 per cent in urban areas being located in household
premises. But what is noteworthy is the fact that 35 and
22 per cent of NDMEs in rural and urban areas
respectively, and 26 and 20 per cent of DMEs, are
household units. If operating out of permanent,
non-household premises is taken as a minimal
prerequisite for a unit being a 'modern' small unit in
the unregistered sector, then, only a little more than a
half of the small population of NDMEs and DMEs in rural
areas and about three-quarters of NDMEs and DMEs in the
urban areas would qualify as modern small units or their
precursors.
Another way to situate and assess
the character of the unregistered manufacturing sector
is to examine the degree and nature of its integration
with the formal economy. This integration can occur
either through financial, input–output or marketing
linkages. These linkages may in fact complement each
other, with financially-dependent small unregistered
units obtaining inputs from and supplying their outputs
to large players who provide financial assistance. With
nearly 50 per cent of the units surveyed in the 56th
round reporting 'shortage of capital' as the problem
confronting them, these kinds of inter-linkages across
finance, input and product markets is a real
possibility.
It needs to be made clear, however, that none of these
needs to be always positive from the point of view of
units in the unregistered sector. As mentioned earlier,
financial dependence could subordination to medium or
large finance capital, especially in industries catering
to large, state or national markets, which could have as
a corollary relatively small margins, as is known to be
true in the production of matches, beedis and handloom
textiles. Input dependence on large oligopolistic
suppliers could also imply high input prices that
squeeze the margins in a sector where demand growth may
be sluggish and the competition intense. And dependence
on the supply of outputs as inputs and products for
marketing by large units can make the small units bear
the brunt of any downturn (through delayed payments, for
example) or hard bargaining by oligopsonistic buyers who
can squeeze margins substantially.
As Table 6 indicates, evidence
yielded by the 56th Round suggests that an
inter-linkage between the formal and informal sectors is
indeed present, even if not overwhelming. Thus, 28 per
cent of units in rural areas and 38 per cent of units in
urban areas work on contracts, which are likely to be
with units or capital from the formal sector. What is
noteworthy is that in the rural areas and, to a smaller
extent, in the urban areas, the contract system is more
prevalent among OAMEs than establishments, pointing to
the possibility of penetration by merchant capital in
search of cheap home-based production sources that
ensure the required returns. This presumption is
supported by the evidence that, on average, 80 per cent
of the enterprises working under the contract system
enter into contracts solely with a master
contractor/enterprise.
Contractual links notwithstanding, it does not appear
from the evidence that these have dominated the choice
of product markets for unregistered units: 65 per cent
of rural units and 57 per cent of urban units reported
that they sold some of their final output to private
individuals or households. However, in the case of NDMEs
and DMEs, sale of final output to private enterprises or
contractors/middlemen was reported. About 37 and 55 per
cent of NDMEs and DMEs respectively in rural areas, and
45 and 71 per cent respectively in urban areas reported
such sales. Therefore, some kind of input–output linkage
and ancillarization do seem to be widely prevalent.
In sum, we can think of two kinds
of 'dualism' in the unregistered sector. First, dualism
with respect to the formal sector, as reflected in the
persistence of household units with primitive techniques
even while modern industry progresses, even if not in
terms of employment generated. This kind of dualism need
not, however, imply the lack of any linkages with the
formal sector. Rather, financial, input–output and
marketing linkages can exist. The persistence of
backwardness may not reflect just the existence of the
peculiar niche markets that low per capita incomes and
poverty create but also the subordination of backward
forms by capital from the formal sector, which treats a
segment of the unregistered sector as a source of
surplus even if that is at the expense of extremely low
wages. Second, dualism reflected by the signs of
coexistence of backward units with other units,
especially among the NDMEs and DMEs, which are taking on
characteristics of ancillaries that are typical of any
modern industrial environment. Analysts argue that the
growth of these kinds of small units and their
integration with the large-scale sector through a
process of 'ancillarization' is positive from the point
of view of generating a modern, well-managed small
industrial sector.
The point to note is that while there is a strong
positive relationship between the rank of a two-digit
industrial sector in terms of estimated number of
enterprises and estimated number of workers (with the
rank correlation coefficient exceeding 0.9 in categories
of units excepting NDMEs and DMEs in rural areas), there
is virtually no relationship or at best an extremely
weak relationship between the rank of an industry in
terms of estimated number of workers and in terms of
value added per worker. While it could be argued that
this should be 'expected', what it does suggest is that
in terms of the number of enterprises and estimated
workers, it is not the more productive units that
predominate. Features of this kind that emerge from this
preliminary analysis of the 56th Round
results suggest that the Indian context is surprising
inasmuch as the long experience with industrialization
has not undermined the former type of units, which in
fact not only persist but even appear to dominate the
landscape of the world of unregistered units.
[1] Computed from CSO
figures quoted in Reserve Bank of India (1956),
Report on Currency and Finance, RBI: Bombay, p. 127.
[2] Ref. Shirokov, G.K. (1973),
Industrialization of India, Moscow: Progress
Publishers.
[3] The Survey covered almost the
entire country except Leh and Kargil districts in Jammu
and Kashmir, villages beyond five kilometers of bus
routes in Nagaland, inaccessible villages in Andaman &
Nicobar Islands, and some first stage units where EC
1998 couldn't be undertaken. A total of 14788 first
stage units (5696 villages and 9092 urban blocks) were
selected for the survey.
[4] Enterprises with power with ten
people or less, or those without power with 20 people or
less are left out of the purview of the Factories Act
1948.
[5] We must recall that the total
number of workers in ASI factories fell from 85.5 lakh
to 81.7 lakh between 1998-99 and 1999-2000.
[6] Inflating the 1999-2000 ASI GVA
figure using the rate of growth of that magnitude
between 1998-99 and 1999-2000 to arrive at an estimate
for 2000-01 and using that figure changes the ratio to
22.9 per cent as opposed to 23.4 per cent.