Manufacturing in India

Labor costs in Indias organized

manufacturing sector

Compensation costs in Indias organized manufacturing sector

were 91 cents per hour for all employees in 2005; this amounted to
about 3 percent of hourly labor costs in the U.S. manufacturing sector,
but was above BLS estimates of labor costs in China

I
Jessica R. Sincavage, ndias important role in the global econ- ganized manufacturing sectorthe por-
Carl Haub, omy is perhaps best exemplified by its tion of the countrys manufacturing activity
and O.P. Sharma
membership in the G-20, the group that is formally registered with Indian state
that has replaced the G-8 as the major governments, making it subject to regula-
international economic forum. Although tion. BLS estimates that in 2005, the latest
India is the fourth-largest economy in the full year for which data were available at
world, accounting for 4.6 percent of the the time this article was written, employ-
worlds GDP, the value of Indias exports ers in Indias organized manufacturing sec-
in 2007 was only 1 percent of the worlds tor compensated employees at a mean rate
total exports.1 Many factors affect the level of $0.91 an hourapproximately 3 percent
of a countrys exports and the growth of its of the compensation level of manufacturing
GDP. The Government of Indias National employees in the United States. (All aver-
Manufacturing Competitiveness Council ages referred to in this article are means.)
has identified manufacturing as the main This article describes the Indian manu-
engine for economic growth and creation facturing industry and the differences be-
of wealth for the country.2 Currently, the tween the organized and unorganized sec-
Council believes that Indias export levels tors. However, it focuses primarily on the
are far below its potential. India has been organized manufacturing sector. This sector
identified as a potential manufacturing gi- produces over two-thirds of Indias manu-
ant by outsiders, as well, and has generated facturing output, and the firms in this sec-
Jessica R. Sincavage is a
supervisory economist in
interest in the global marketplace because tor are more comparable to enterprises in
the Division of Interna- of its low cost of labor and large population. advanced countries than are firms in the
tional Labor Compari-
sons, Bureau of Labor Because of Indias economic prominence, unorganized sector.3 The article also dis-
Statistics; Carl Haub is and in light of BLSs history of providing cusses Indias statistical system, features of
a senior demographer
and the Conrad Taeuber comparative statistics, BLS has undertaken the available Indian manufacturing industry
Chair of Information at
the Population Reference a research project to study the manufac- data, the procedure used by BLS to estimate
Bureau; and O.P. Sharma is turing industry in India, supported by the hourly compensation, and compensation
the former Deputy Direc-
tor of Census Operations expertise of coauthors Haub and Sharma. trends both in all manufacturing and in 18
in India. Email: sincav-
age.jessica@bls.gov or
This article presents, for the first time, BLS industries within manufacturing. Lastly, it
chaub@prb.org estimates of compensation in Indias or- addresses the commonly made comparison

Monthly Labor Review May 2010 3

Manufacturing in India

of Indian and Chinese manufacturing. The Indian statistical system

Background Unlike most developing countries, India has a long his-

tory of conducting surveys and maintaining statistics, and
The Bureau of Labor Statistics calculates and publishes its systems have evolved and remained relevant to chang-
hourly compensation costs in manufacturing for all em- ing economic and political conditions. Statistical systems
ployees in 32 countries and for production workers in in India can be traced back as far as the fourth century
34 countries.4 In recent years, BLS has added emerging BC, when rulers maintained information on population,
economies to these two series, which previously had con- land, and agricultural production primarily to serve their
tained only data from developed countries. Although In- own needs. In general, data collection was neither highly
dia has been recognized among developing economies for developed nor well coordinated until after India gained
the abundance and quality of its statistics, compensation its independence in 1947, when the need for more ad-
estimates for Indias manufacturing sector cannot yet be vanced economic planning arose.6 By the early 1950s, the
incorporated into the main BLS comparative compensa- country had established the Central Statistical Organisa-
tion series because of limitations such as a lack of timely tion (CSO), which coordinates the state statistical offices,
data publication, absence of data on recorded work hours, and the National Sample Survey Organisation, which
and a likelihood of many businesses reporting innacurate conducts large-scale sample surveys.7 These two entities
data. Instead, BLS hopes to present data for India as a spe- are currently housed under the Ministry of Statistics and
cial supplemental seriesan approach similar to that used Programme Implementation.
for China, another country for which BLS has identified a In the 1990s, Indias government and its markets under-
number of data quality issues, and a country to which In- went changes that put new pressures on the statistical sys-
dia is often compared.5 Because these two countries have tem. The closed economy, driven fundamentally by public
become important forces in the global economy, there is sector activity, began opening up and relying more heavily
value in studying the compensation data for both coun- on the private sector. In January of 2000, the government
tries, to the extent possible. created a formal bodythe Rangarajan Commissionto
This article presents, for the first time, BLS estimates of review the statistical system and all the official statistics
compensation in the organized sector on an estimated it produces.8 In response to the groups recommenda-
hourly basis in Indian rupees and in U.S. dollars for the tions, India has been working to create a system that is
period from 1999 to 2005. The limitations of the esti- more centralized, consistent, timely, credible, and reliable.
mates also will be discussed. The analysis in this article One major initiative is the India Statistical Strengthening
uses information published by Indias national statistical Project, which calls for creating and maintaining a nation-
organizations, the primary source being the Indian An- al business register to allow for more scientific periodic
nual Survey of Industries (ASI), which collects employ- business surveys, improve the training of employees who
ment and compensation data for the countrys organized work with statistics, and increase resources available to
manufacturing sector. the states.9 The experience and history that India has with

Publication of data from India

The Bureau of Labor Statistics has been a leader in com- hourly compensation costs at this time. This article is in-
piling international comparisons of hourly compensation tended as the first step toward developing the measures nec-
of manufacturing employees over a wide range of countries. essary to include India in the regular comparisons series that
Despite its large and growing importance in world manu- currently comprises 36 countries. Because of the difficulties
facturing, India has not been included in the comparisons in creating hourly compensation estimates for India, the
because of difficulties in obtaining and interpreting that short-term plan is to publish updates for this country, with
countrys data and because of concerns about the quality of appropriate annotations, separate from the regular series of
the data. Although this Monthly Labor Review article greatly international comparisons of hourly compensation. This is
facilitates understanding of Indian compensation statistics, similar to how BLS treats hourly compensation estimates de-
many problems with data availability, coverage, and reliabil- veloped for China. The final goal of moving India and China
ity remain, as described in the article. Therefore, the Bureau into the regular comparisons series would, of course, remain
does not plan to include India in its regular comparisons of intact.

4 Monthly Labor Review May 2010

respect to collecting data increases BLSs confidence in organized manufacturing sector for all employees and for
the credibility of the Indian statistical system as a reliable production workers for each fiscal year, which in India
source of data and information. Still, India acknowledges runs from April 1 to March 31.15 Although the survey has
opportunities for improvement and a need to respond to been conducted since 1960, the BLS hourly compensation
its rapidly changing economy. costs series for Indias organized manufacturing sector
does not begin until 1999, primarily because of industry
Organized sector versus unorganized sector classification changes that occurred before that year and
would have compromised historical comparisons.
Although detailed data are available for Indias organized Beginning with the ASI of 199899 (which is survey
sector, they are less plentiful for Indias unorganized sec- notation for the fiscal year from April 1, 1998 to March
tor. Understanding how these two sectors differ is impor- 31, 1999), data were classified according to the National
tant in analyzing Indias labor statistics. Industrial Classification (NIC) of 1998, which is based on
Indias organized and unorganized sectors generally the International Standard Industry Classification sys-
correspond with what economists call the formal and tem (ISIC Rev.3). In 2004, the NIC was modified, and its
informal sectors in other countries.10 The official distinction changes were captured in the ASI of 200405 (henceforth
between the organized and unorganized sectors lies in ASI 200405). However, BLS analysis shows that the
whether businesses register with the government and differences between NIC 1998 and NIC 2004 do not affect
regularly maintain prescribed records. According to the year-over-year comparisons between the BLS estimates
National Accounts Statistics for India, the organized sector for ASI 200405 and those for previous survey periods.
comprises enterprises for which statistics are available from Ultimately, BLS adjusts the Indian manufacturing data to
budget documents, reports, or other such documents. In make them comparable with data that were calculated in
contrast, the unorganized sector refers to those enterprises a manner consistent with the North American Industry
whose activities or collection of data is not regulated under Classification System (NAICS).
any legal provision or enterprises that do not maintain any The ASI is conducted every year by mail and covers 31
regular accounts.11 Not surprisingly, there are relatively few of the 35 states and union territories that make up In-
data series that cover the unorganized sector. Individual dia. The four areas not covered likely have little impact on
establishments tend to be small, typically employing fewer measurement because of their small size.16 Because the
than 10 persons, and many of these enterprises have no survey frame includes all establishments that have regis-
hired workers and operate primarily for family sustenance. tered with the Indian states, the ASI sample is believed
The two sectors also differ in how they contribute to to be representative of the organized manufacturing sec-
Indias thriving manufacturing industry, which accounted tor.17 Although the data are thought to be characteristic of
for approximately 16 percent of Indias real GDP from firms in the organized sector, there are important caveats.
2000 to 2006.12 When measured by output, the organized ASI survey data are presented in raw form without adjust-
sector dominates, producing approximately two-thirds of ments to the ways that employers reported them; there
the countrys manufacturing output.13 The organized sec- are no attempts to contact employers to fill in missing or
tors average annual rate of growth was stronger than that incomplete data or to correct for data that seem out of
of the unorganized sector, 13.1 percent compared with 9.9 line with other data. In addition, although participation
percent. When measured by employment levels, however, is compulsory by the Collection of Statistics Act of 1953,
the unorganized sector dominates. According to estimates penalties for noncompliance are not enforced frequently.18
from national data, close to 80 percent of manufacturing Because of the problem of nonresponse and because no
employees work in the unorganized sector.14 From either attempt is made to impute values for employers that do
perspective, the unorganized sector must be regarded as not respond, the results are dependent upon which estab-
an important part of Indian manufacturing, and BLS is lishments return the survey questionnaire. These problems
currently conducting additional research on it. This arti- cause the data to be less reliable than survey data that are
cles primary focus, however, is the organized manufactur- adjusted by the receiving statistical agency, or data that are
ing sector. weighted to be representative of the entire survey popula-
tion.
The Annual Survey of Industries The ASI covers manufacturing activities as defined by
the Indian Factories Act as any of the following five pro-
The ASI collects employment and earnings data from the cesses:

Monthly Labor Review May 2010 5

Manufacturing in India

(i) making, altering, ornamenting, finishing, packing, on whichever factories respond to the survey in any given
oiling, washing, cleaning, breaking up, demolishing year. General trends can be compared across years for all
or otherwise treating or adapting any article or sub- of manufacturing and for subsectors within manufactur-
stance with a view to its use, sale, transport, delivery ing, but ASI data on industries with 4-digit NIC codes
or disposal; or generally are not comparable from one year to the next.
(ii) pumping oil, water, or sewage; or
(iii) generating, transforming or transmitting power; or The growth of contract labor. In 200506, the most recent
fiscal year for which data from the ASI are available, 8.7
(iv) composing types for printing by letter press, lithog- million people were covered in the survey and reported
raphy, photogravure or [a] similar process, or bind- as employed in Indias organized manufacturing sector.22
ing [books]; or (See table 1.) As mentioned earlier, there are difficulties
(v) constructing, reconstructing, repairing, refitting, fin- in estimating trends in employment by use of data from
ishing or breaking up ships or vessels.19 the ASI because the survey results are not representative of
The manufacturing sector is defined differently in the the entire organized manufacturing sector. The National
BLS hourly compensation series. Under the 2007 NAICS, Sample Survey Organisation does not publish response
manufacturing comprises establishments engaged in the rates, and, as mentioned earlier, data from the ASI are not
mechanical, physical, or chemical transformation of ma- adjusted to account for nonresponse.23 Despite these limi-
terials, substances, or components into new products.20 tations, it is possible to discern from the data that some
The assembling of component parts for manufacturing is changes in the makeup of the Indian organized labor
considered manufacturing, except in cases in which the force are occurring.
BLS produces data for two groups of people in its in-
activity is classified in construction. In order to reconcile
hourly compensation costs calculated by use of the NA- ternational series on hourly compensation in manufactur-
ICS definition of manufacturing with those calculated by
ing: all employees and production workers.24 Production
use of the ASI definition, BLS must remove from the raw workers are defined as those employees who are engaged
Indian data all publishing activity as well as industries en- in fabricating, assembly, and related activities; material
gaged in items (ii) and (iii) of the Factories Act definition handling, warehousing, and shipping; maintenance and
of manufacturing.21 repair; janitorial and guard services; auxiliary production;
or other services closely related to the aforementioned ac-
Data features tivities. Working supervisors generally are included; ap-
prentices and other trainees generally are excluded. The
Knowledge of ASI data reporting practices and the salient category all employees comprises production workers as
features of the ASI data are important to understanding well as other workers employed full time or part time in
the estimates presented in this article and their limita- an establishment during a specified payroll period. Tem-
tions. Trends in employment, including the growth of porary employees are included. People are considered
contract labor in the organized manufacturing sector, will employed if they receive pay for any part of the speci-
be discussed, as will the lack of data on payment for over- fied pay period. Unpaid family workers, workers in private
time work. As noted earlier, ASI data are reported as they households, and the self-employed are excluded. Typi-
are collected and are not weighted to represent Indias en- cally, contract workers are excluded from BLS estimates of
tire organized manufacturing sector. The results are based hourly compensation, but for India, contract workers are

Table 1. Employment in India's organized manufacturing sector, 19982006

[Numbers in thousands]

Type of employees 199899 19992000 200001 200102 200203 200304 200405 200506
All employees............................................................................. 8,317 7,857 7,634 7,400 7,590 7,518 8,064 8,688
All production workers........................................................ 6,174 6,049 5,933 5,757 5,961 5,887 6,373 6,893
Directly employed................................................................ 5,213 4,857 4,725 4,507 4,591 4,440 4,685 4,920
Employed through contractors...................................... 960 1,192 1,208 1,249 1,369 1,447 1,688 1,973
Employees other than production workers.................. 2,143 1,808 1,702 1,643 1,629 1,631 1,691 1,800
SOURCE: BLS estimates made by use of Annual Survey of Industries data make data comparable with estimates that were calculated in a manner
from the Central Statistical Organisation of India. consistent with NAICS. Because of rounding, some sums of components do
NOTE: Data are not as originally published. Industries were removed to not equal their respective totals.

6 Monthly Labor Review May 2010

included in both the production workers and all employ- in the Indian labor market, workers continue to accept
ees series because their wages are reported together with these types of job arrangements because they tend to pre-
the earnings of other workers and cannot be separated. fer secure employment to employment opportunities that
According to ASI 200506 data, production work- are less secure, even if the less secure opportunities are po-
ers accounted for 79.3 percent of all employment in the tentially more lucrative. It is likely that large growth in the
organized manufacturing sector in India, an increase of number of contracted production workers has caused the
approximately 5 percentage points from 199899, when average compensation estimates published here for both
production workers accounted for 74.2 percent of total production workers and all employees to be lower than
organized manufacturing sector employment. (See chart they otherwise would be. Earnings of contract workers are
1 for information on the structure of employment). This included in the earnings data for all workers, but the ASI
increase in production workers share of employment was does not publish separate earnings data for contract work-
driven by an increase in the number of contractors em- ers. Therefore, it is not possible to determine directly the
ployed as production workers in the organized manufac- effect of contract work on earnings in India.26
turing sectora number that more than doubled over the
period in question. In 199899, contract workers account- Hours, part time, and overtime. BLS needs data on the
ed for only 15.6 percent of the employment of production number of hours that employees worked, as well as infor-
workers; by 200506, contract workers accounted for 28.6 mation on employers practices as regards compensating
percent of production workers employment. The increase employees. BLS estimates assume a 6-day, 8-hours-per-
in the proportion of contract workers in the organized day workweek on the basis of research and interviews, as
manufacturing sector has likely helped keep overall labor described in the following paragraphs.
costs lower over the period in question because employ- In the ASI, wages are based on gross amounts paid to
ing contract workers is a legal way for employers to avoid workers in general; no distinction is made between wages
many of the costs associated with hiring workers directly, paid to full-time workers and wages paid to part-time
such as the costs of social insurance and paid vacation. workers. This is a common limitation of earnings and
The use of contract labor has been cited as a global trend compensation data across countries. Additionally, be-
and a phenomenon by which, according to Amit K. Bhan- cause regular-time earnings and overtime earnings are
dari and Almas Heshmati, workers earn lower wages and combined when they are reported, average wage data in-
also are deprived of benefits like health, safety, welfare clude the effect of an unknown number of overtime hours,
and social security.25 Bhandari and Heshmati found that, which may be paid at a higher rate. Overtime is common

Chart 1. Structure of employment in Indias organized manufacturing sector

All employees

Production workers Salaried workers

Directly hired Contract production

production workers workers

Monthly Labor Review May 2010 7

Manufacturing in India

in Indian manufacturing, but no data on actual overtime do not receive the proper overtime pay, if they receive any
hours are available. Government regulations in India stip- additional pay at all. Overall, the respondents did report
ulate that workers be paid twice their regular earnings for that a 6-day, 8-hours-per-day workweek is the common
each hour of overtime worked.27 However, it is not clear practice, which is in line with the hours estimate used in
how many workers in the manufacturing industry actually the BLS calculations.
receive this increased wage for their overtime hours. For It is important to consider these cultural practices and
those who do receive it, it is not clear whether they receive data nuances when one interprets the hourly compensa-
the full amount to which they are entitled or only some tion figures presented in this article. The increase in con-
fraction of it. tract labor has likely suppressed the average hourly cost
The practice of ignoring regulations regarding hours of compensation in Indian manufacturing over time. Ad-
worked and overtime and the practice of using contract ditionally, it is not clear how much work is occurring off
labor to circumvent paying required amounts are wide- the books. The addition of pay for work done beyond the
spread in India; fortunately, some employers were will- number of hours in a standard workweek could cause the
ing to provide information on an anonymous basis dur- average hourly compensation estimate to be slightly in-
ing personal interviews and through a small, independent flated since those additional hours worked are not includ-
survey of manufacturing establishments administered by ed in the BLS estimates (and the pay for those hours would
coauthors Haub and Sharma in Faridabad, Haryana state, be estimated at a higher rate). Although earnings, hours,
an industrial suburb of Delhi, in July 2006 specifically and employment that are not documented by employers
for this article.28 A branch supervisor of a private print- likely affect the hourly compensation estimates presented
ing firm provided information on common practices.29 At in this article, no adjustments have been made because the
his firm, the normal workday is 8 hours, with overtime magnitude of the unrecorded data is not known. BLS esti-
worked as needed. He stated that his firm and others with mates are based on the data as they are reported in the ASI.
which he is familiar pay an overtime rate that equates to Lastly, there are a number of inconsistencies in the ways
the amount required by law, 2 times salary, but added that factories respond to some survey items in the ASI, which
he was also aware of printers who pay less than the legally reduces the level of detail that can be shown in the survey
required rate. He noted that most employment contracts reports. For example, although the ASI questionnaire in-
are arrived at orally, are typically cash transactions, and cludes columns titled contribution to provident & other
that the records kept by employers do not always reflect funds, workman & staff welfare expenses, and bonus,
reality. all broken down by type of worker, a substantial number
The supervisor also noted that 50 percent of workers at of respondents simply write in a lump sum for all work-
his firm were contract labor, a high proportion, and that ers. The Indian term for this practice in reporting data is
the hours worked do not matter (meaning that a persons clubbing, and, when it occurs, only aggregate expenses
salary will be the same whether he or she works regular for all employees are reported. For the BLS estimates, this
hours or long hours). Work that is somewhat irregular in does not present a problem. In the BLS hourly compensa-
nature is often contracted, and most contracted work is tion series, data on the structure of labor costs for all em-
not regulated. Employers and contracted workers negoti- ployees are frequently used to estimate the corresponding
ate a specific job, and the workers are paid a lump sum values for production workers.30 This common practice
for the work, regardless of the number of hours the job was adopted because of a lack of detailed data on produc-
eventually takes. tion workers for many countries. BLS analysis has shown
During other interviews, respondents provided less that in the manufacturing sector data on the structure of
specific information, but one theme was expressed re- labor costs for all employees tend to be similar with those
peatedlyenforcement difficulties are compounded by for production workers.
employee connivance in circumventing hours and over-
time pay regulations. Employees frequently wish to work Hourly compensation estimation procedures
additional hours and to earn more than the standard
hourly rate doing so, but employers often point out that BLS comparative measures of hourly compensation costs
they can simply hire additional workers who are happy include both data on hourly direct pay (which compris-
to work at the regular rate because there is a large num- es pay for time worked, pay for vacations and holidays,
ber of workers competing for jobs. As a result, workers bonuses, in-kind pay, and other premiums) and data on
who work beyond the standard number of hours usually employers social insurance expenditures and other labor

8 Monthly Labor Review May 2010

taxes (a category that comprises employers expenditures in compensation and what factors affect them.
for legally required insurance programs and contractual To estimate the amount of compensation attributable to
and private benefit plans, as well as other taxes on payrolls paid time off, a measure of hours or days paid was needed.
or employment). Estimating the number of days paid for but not worked
The concept of earnings as reported in the ASI for all is complicated by the fact that employers are not required
employees is nearly equivalent to the BLS concept of to pay all workers for vacations and holidays. The Facto-
total direct pay, except that there are no estimates of pay ries Act stipulates that production workers and salaried
in kind in the ASI data.31 The ASI also reports data on workers in organized manufacturing are entitled to 1 day
social insurance, such as employers contributions to the of earned leave for every 20 days worked in the previous
provident fund and other funds, and workmen and staff year.35 Also entering into the calculation are 10 national
welfare expenditures (that is, additional expenditures holidays in India during which employees do not work,
that promote the general well-being of employees.)32 but are paid.36 However, employers are only legally re-
In addition to earnings data, a measure of the number quired to provide paid leave to employees who were hired
of days or hours worked by employees in manufacturing directly. There is no legal obligation to provide paid time
is needed to calculate hourly compensation. The ASI does off for contract workers, although the contractor is sup-
not report the number of days or hours worked in man- posed to do so; however, anecdotal evidence indicates that
ufacturing, but does report the number of man-days. these workers often are not paid for time off. For this rea-
Man-days are days both worked and paid for during the son, BLS calculated an estimate of the number of paid days
accounting year. The number of man-days is calculated by worked and of the number of paid days not worked for
summing the number of paid employees working during three separate groups of workers in the Indian organized
each shift over all the shifts worked on all days. Man-days manufacturing sector: directly hired workers other than
include only days on which employees actually worked; production workers, directly hired production workers,
because of how they are defined and recorded by employ- and contract workers.
ers, man-days do not include days for which employees Man-days in the Indian organized manufacturing sector
were paid but on which they did not work, such as vaca- for salaried workers can be derived from data published
tion days and holidays. by the CSO for all employees and for production work-
Total hourly compensation can be obtained by a simple ers. Separate man-days data for directly hired and con-
division equation. The numerator is the sum of total direct tract production workers, respectively, are not available,
pay, or earnings (including bonuses), and social insurance so BLS allocated production worker man-days using the
as reported in the ASI. The denominator is aggregate hours ratio of people employed as directly hired employees to
worked, which is equal to man-days as reported in the those employed as contract workers. Then, paid leave days
ASI multiplied by the estimated number of hours worked for salaried workers and directly hired production work-
daily. In order to estimate average hourly earnings, the av- ers were calculated. The number of paid leave days for
erage number of hours worked daily is necessary. Unfor- contract production workers is assumed to be zero since
tunately, no data on hours worked are collected in the ASI employers have no legal obligation to pay them.37 (That
or from any other national source. Coauthors Haub and is, contract workers are removed from the calculation of
Sharma thus solicited information from the CSO on typi- man-days paid but not worked.) Paid leave excluding hol-
cal working practices in Indias organized manufacturing idays for non-contract employees is estimated to be 1 day
sector, conducted interviews with employers in Delhi, and for for every 20 days worked (because of the requirement
conducted the aforementioned survey in Faridabad in July in the Factories Act). The sum of paid holidays and paid
2006.33 All three of these sources indicated that a 6-day leave days excluding holidays is the total number of days
workweek lasting from 10 a.m. to 6 p.m. is very common. paid but not worked; this sum is added to the published
BLS thus estimates average daily hours worked at 8.34 number of man-days worked to get the total number of
To better understand ASI data on compensation in Indias paid man-days in manufacturing. All the aformentioned
organized manufacturing sector, BLS created estimates of calculations were done on a per-worker basis.
components of compensation not already reported in the The ratio of man-days worked to man-days paid can be
ASI: pay for time worked and pay for time not worked (pay multiplied by the earnings (without bonuses) figure re-
for vacation days and holidays). Having data on the vari- ported in the ASI to provide a rough estimate of aggre-
ous components of compensation and how they change gate pay for time workedor basic wages and salaries. All
over time allows for a greater understanding of the trends employees pay for time worked is the sum of production

Monthly Labor Review May 2010 9

Manufacturing in India

workers pay for time worked and salaried workers pay for available at the time this article was written was 2005.
time worked. To get average hourly earnings, this aggre-
gate is then divided by aggregate hours worked, or the Estimate of hourly compensation for production workers. The
product of man-days worked and estimated daily hours foregoing discussion relates to the procedures used to de-
worked. The value of pay for time not worked can also be rive estimates of hourly compensation for all employees in
calculated by subtracting aggregate pay for time worked manufacturing. BLS also constructed estimates of hourly
from earnings (without bonuses). compensation of production workers. Data on earnings of
Next, total compensation ratios were calculated by BLS. production workers are available from the ASI, but those
The total compensation ratio is a multiplicative factor data differ from the data for all employees in that bonuses
that, when applied to the average hourly earnings figure, are not included. In order to put the production worker
results in a product equal to total compensation. For In- estimates and the all-employee estimates on a comparable
dia, it was calculated by dividing aggregate total compen- basis, BLS derived an estimate of bonuses that was added
sation by aggregate total pay for time worked. Total com- to the earnings of production workers. Bonuses and so-
pensation was calculated by summing total direct pay (pay cial insurance have been redistributed among workers in a
for time worked, pay for time not worked, and bonuses) manner proportionate to their earnings; this procedure was
and aggregate annual social insurance costs. Aggregate recommended by the CSO as a method of estimating these
annual social insurance costs for all employees in Indian components of compensation.38 Under the assumption
manufacturing are equal to employers contributions to that all employees (including production workers) receive
the Provident Fund and other funds plus worker and staff bonuses in direct proportion to their wages, bonuses were
welfare expenses. estimated by applying the ratio of all employees bonuses
As noted earlier, data from the ASI are reported on a fiscal- paid to their nonbonus earnings. Like data on bonus-
year basis, from April 1 to March 31. In order to compare es, data for social insurance expenditures for production
the total compensation estimates created from fiscal-year workers are not available from the ASI. Thus, BLS applied
ASI data with the corresponding estimates from other coun- the ratio of social insurance to earnings for all employees to
tries in the BLS hourly compensation series, the data must production workers earnings in order to derive an estimate
be adjusted to conform to a calendar-year basis. To do this, of social insurance expenditures for production workers.
BLS used a weighted average of two sets of ASI fiscal-year Similar methods are used in the BLS series for a number of
data. For example, to obtain data for calendar-year 2005, countries for which the requisite production-worker data
BLS applied a weight of 0.25 to ASI 200405 estimates and are lacking. Research conducted by BLS in the past for sev-
a weight of 0.75 to ASI 200506 estimates. The 0.25 figure eral other countries has shown that this practice does not
represents the quarter of 2005 that is covered in ASI 200405 substantially affect the hourly compensation estimates.
( January 2005March 2005) and the 0.75 figure represents
the three quarters of 2005 that are covered in ASI 200506 Results
(April 2005December 2005). Under this system of estima-
tion, the most recent calendar year for which ASI data were Table 2 displays detailed estimates of Indias hourly com-

Table 2. Hourly compensation costs in India's organized manufacturing sector, 19992005

Mean hourly earnings
Total compensation Hourly compensation Hourly compensation in
in rupees Exchange
ratio in rupees USD
(hourly pay for time rate:
Year worked) rupees/
[2] [3]=[1] [2] [5]=[3] [4]
[1] USD
Production Production Production [4] Production
All employees All employees All employees All employees
workers workers workers workers
1999......... 20.68 15.97 1.423 1.423 29.43 22.72 43.06 0.68 0.53
2000......... 22.54 16.97 1.406 1.406 31.68 23.86 44.94 .70 .53
2001......... 23.77 17.57 1.416 1.416 33.65 24.88 47.22 .71 .53
2002......... 24.95 18.22 1.417 1.417 35.36 25.83 48.63 .73 .53
2003......... 26.58 18.98 1.417 1.418 37.68 26.91 46.59 .81 .58
2004......... 27.57 19.46 1.398 1.398 38.55 27.21 45.26 .85 .60
2005......... 29.10 20.06 1.375 1.376 40.02 27.60 44.00 .91 .63
SOURCE: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India.

10 Monthly Labor Review May 2010

Chart 2. Components of hourly compensation in Indias organized manufacturing sector, 2005

Total hourly compensation = 40 rupees

Pay for time worked (average hourly earnings) Other direct pay Social insurance

Pay for time worked (average hourly earnings) = 29 rupees Other direct pay = Social insurance =
4 rupees 7 rupees

SOURCE: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India.

pensation costs for all employees and for production cial insurance. (See chart 2.) From 1999 to 2005, average
workers. When measured in Indian rupees, total compen- hourly earnings increased 40.7 percent, other direct pay
sation of all employees in Indias organized manufacturing grew by 31.7 percent, and average social insurance expen-
sector increased by 36.0 percent from 1999 to 2005. From ditures per hour increased 20.7 percent; in 2004 and 2005,
1999 to 2003, total hourly compensation for all employees average social insurance expenditures actually decreased.
grew, on average, by 6.4 percent each year. The growth of Widespread pension reform has been occurring across in
hourly compensation slowed to 2.3 percent in 2004 and India over the past several years as many states move from
was 3.8 percent in 2005. defined benefit pension schemes to defined contribution
When measured in U.S. dollars the increase for all em- schemes, but it is unclear exactly what role this has played
ployees was slightly less (34.1 percent) over the same pe- in trends in social insurance expenditures.39 Typically, it
riod because of the depreciation of the rupee relative to takes some time for the effects of pension reform pro-
the dollar. Overall, the rupee depreciated slightly over the grams to show up in labor cost data, and many changes
19992005 period, but appreciated from 2002 through have been happening in India simultaneously. Longer
2005. Increases in hourly compensation were accompa- time series of data for India will likely provide more in-
nied by decreases in the value of the rupee against the U.S. sight into trends in social insurance.
dollar from 1999 to 2002which is evidenced by relative- Pay for time worked, or basic wages and salaries, ac-
ly small increases in the all-employees section of column 5 counted for the largest portion of total compensation in
during these years. Hourly compensation as measured in Indias manufacturing sector by far in 2005 (approximate-
U.S. dollars grew much faster from 2003 through 2005 as ly 73 percent). As noted earlier, this component of com-
the rupee appreciated against the dollar. pensation grew the fastest in comparison with other com-
The ratio of total compensation to average hourly earn- ponents of compensation over the 1999-to-2005 period.
ings rose or stayed the same every year from 2000 to 2003. For production workers, average hourly earnings in-
However, the ratio decreased slightly over the last 2 years creased by only 25.6 percent over the 19992005 period,
of the 19992005 period, declining from 1.417 in 2003 to compared with 40.7 percent for all employees, so total
1.375 in 2005. The total compensation adjustment ratio is compensation for production workers as measured in In-
obtained by dividing total compensation by pay for time dian rupees increased significantly less than it did for all
worked; for India, average hourly earnings are equal to pay employees over the same period (21.5 percent versus 36.0
for time worked. percent). Production workers total compensation as a per-
Changes in total compensation are affected by changes centage of all employees total compensation decreased as
in any component of compensation. The components on result. (See chart 3.)
which BLS has data for Indias organized manufacturing ASI data on employment and man-days show that, over
sector are the following: pay for time worked (average the 7-year period, the average employee in Indias orga-
hourly earnings), other direct pay (which for India con- nized manufacturing sector consistently worked about
sists primarily of pay for time off and bonuses), and so- 305 days a year, with the exception of 1999, for which

Monthly Labor Review May 2010 11

Manufacturing in India

Chart 3. Total hourly compensation of all employees and of production workers in Indias organized
manufacturing sector, 19992005
Production workers mean
compensation as a percent
of all employees mean Compensation, in
compensation rupees
78 44
Production workers mean compensation as a percent of all employees mean
compensation 42
76 All employees mean compensation 40
Production workers mean compensation
38
74
36
72 34
32
70 30
28
68
26

66 24
22
64 20
1999 2000 2001 2002 2003 2004 2005

SOURCE: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India.

the average was 289. This implies that, for the 2000-to- and the Philippines (3.6 percent of the U.S. level). The
2005 period, employees worked an average of just under a average hourly compensation cost for manufacturing pro-
6-day workweek, which is consistent with the information duction workers in Sri Lanka, a country for which BLS
received from the CSO and from interviews with Indian publishes hourly compensation cost data for production
employers. workers only, was 2.3 percent of the U.S. average hourly
compensation of all manufacturing production workers.
Comparisons with other countries Compensation costs were moderately higher in Mexico,
Brazil, the Eastern European countries, and in the coun-
Hourly compensation costs in India are among the lowest tries in East Asia excluding Japancountries that are
when compared with the 36 countries in the BLS hourly often thought of as having relatively low manufacturing
compensation series.40 In 2005, Indias average hourly compensation costs.
compensation cost for all employees in manufacturing When BLS hourly compensation estimates for Indias
($0.91) was approximately 3.1 percent of the level seen in production workers were compared with estimates of
the United States ($29.74) when measured in U.S dollars. hourly compensation of U.S. production workers, the
(See chart 4.) Over the period from 1999 to 2005, hourly analysis yielded results similar to the those obtained in
compensation costs for all employees in Indian manu- the analysis for all employees. The cost of employing 1
facturing fluctuated between 2.7 and 3.1 percent of the hour of production worker labor in India in 2005 ($0.63)
U.S. level. This fluctuation is due in part to changes in the was equal to 2.6 percent of the cost in the United States
rupee-to-dollar exchange rate. As seen earlier, measured ($23.81) as measured in U.S. dollars. (See table 2.)
in rupees, hourly compensation costs increased each year Historically, other countries in the BLS series have been
from 1999 to 2005. in comparatively low positions, similar to those of India,
Among the economies studied by BLS, the lowest hourly the Philippines, and Sri Lanka. In 1975, the initial year of
compensation costs for all employees in manufacturing in the BLS hourly compensation series, hourly compensation
2005 were found in India (3.1 percent of the U.S. level) costs for production workers in manufacturing in Korea

12 Monthly Labor Review May 2010

Chart 4. Mean total hourly compensation cost of manufacturing employees, selected countries and regions, 2005
Index Index
($29.74 = 100) ($29.74 = 100)

120 120

100 100

80 80

60 60

40 40

20 20

0 0
United India Brazil Mexico Euro area1 Eastern Japan East Asia Philip- Sri Lanka4
States Europe2 excluding pines
Japan3
1
Euro area refers to European Union member countries in the BLS series that have adopted the euro as the common currency as of
January 1, 2009. These countries are the following: Austria, Belgium, Finland, France, Germany, Ireland, Italy, the Netherlands, Portugal,
Slovakia, and Spain.
2
Czech Republic, Hungary, Poland, and Slovakia.
3
Republic of Korea, Philippines, Singapore, and Taiwan.
4
Data are for production workers only.
SOURCES: See ftp://ftp.bls.gov/pub/special.requests/ForeignLabor/ichccaesuppt01.txt for data on all employees, and see
ftp://ftp.bls.gov/pub/special.requests/ForeignLabor/ichccpwsuppt01.txt for data on production workers in Sri Lanka.

and Taiwan were equal to 5 percent and 6 percent of the employer labor costs are much higher, or much lower, than
U.S. level, respectively, when measured in U.S. dollars.41 in other subsectors. Also, some subsectors have high em-
As these countries became larger players in the global ployment relative to others. Compensation costs in sub-
marketplace, their compensation costs grew more sectors within manufacturing can provide insights that
quickly than those of the United States, whose global are useful for making international comparisons, because
manufacturing presence was already well established. individual subsectors generally play larger roles in some
By 1980, compensation costs in Korea and Taiwan had countries than in others. Data on all employees aggregate
increased to 10 percent and 11 percent of the U.S. level, earnings and on their aggregate social insurance paid, as
respectively. By 2005, the percentages had increased to 52 well as on their employment and man-days worked, are
percent and 27 percent. available for the subsectors.
In 2005, the lowest hourly compensation costs were
Subsectors within manufacturing in food, beverage, and tobacco manufacturing, and in
wood product manufacturing. (See chart 5.) Employees
Employment and earnings data are also available for 18 were most highly compensated in the petroleum and coal
industries within the manufacturing sector in India. For products manufacturing subsector; costs in this subsector
this analysis, the food manufacturing subsector (NAICS were more than twice the level faced by employers in all
311) and the beverage and tobacco product manufactur- manufacturing subsectors on average. However, because
ing subsector (NAICS 312) are considered together as one this subsector accounts for only 1 percent of total
industry. The same goes for the textile mills subsector (NA- employment in the organized manufacturing sector, these
ICS 313) and the textile product mills subsector (NAICS high compensation costs have little effect on the average
314). Each of the other 16 industries is a subsector. The compensation level for all of manufacturing.
level of total compensation in all manufacturing can mask Six subsectors make up about half of all manufacturing
important differences among the compensation levels in employment in Indias organized sector. The ASI 200506
the subsectors within manufacturing. In some subsectors, data show that organized-sector employment is highest

Monthly Labor Review May 2010 13

Manufacturing in India

Chart 5. Mean total hourly compensation in Indias organized manufacturing sector, by subsector, 2005

Subsector(s)
311 and 312, Food, beverage, and tobacco
321, Wood products All employees
316, Leather and allied products Production workers
315, Apparel
313 and 314, Textile and textile product mills
327, Nonmetallic mineral products
326, Plastic and rubber products
322, Paper
332, Fabricated metal products
31 to 33, All manufacturing (excluding publishing)
337, Furniture and related products
325, Chemicals
335, Electrical equipment, appliances, and components
331, Primary metals
333, Machinery
336, Transportation equipment
334, Computer and electronic products
324, Petroleum and coal products

0 20 40 60 80 100 120
Rupees per hour
SOURCE: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India.

in the following industries: food, beverage, and tobacco was 31 percent lower than average hourly compensation
manufacturing (two subsectors considered together, as for all employees. (See chart 5.) Within manufacturing,
previously mentioned); textile and textile product mills however, the ratio of the mean hourly compensation of
(two subsectors considered together, as previously men- production workers to that of all employees varied across
tioned); chemical manufacturing (NAICS 325); and pri- industries. Among the industries analyzed, the ratio was
mary metal manufacturing (NAICS 331).42 (See table 3.) the greatest in textile and textile product mills, where
Food, beverage, and tobacco manufacturing, and textile hourly compensation of production workers was equal
and textile product mills are among the lowest paid in- to 83 percent of the level of hourly compensation of all
dustries in Indias organized manufacturing sector and in employees. In the computer and electronic product manu-
200506 accounted for over 36 percent of all organized- facturing subsector (NAICS 334), the difference between
sector manufacturing employment. Their high employ- the hourly compensation of all employees and that of
ment share and low compensation levels drag down the production workers varied greatly; the average compen-
average compensation level for all of manufacturing. sation of production workers was only 52 percent of the
Data on employment of production workers in average compensation of all employees in the same sub-
manufacturing subsectors are reported in the ASI; however, sector. Generally, subsectors that required more technical
man-days for production workers in the subsectors are expertise tended to have greater differentials between all
not. Because man-days are directly linked to the level employees average hourly compensation and that of pro-
of employment in any given industry, BLS was able to duction workers.
estimate the number of man-days worked by production
workers in each of the manufacturing subsectors by use International comparisons of subsectors within manufactur-
of employment and man-days data for all employees and ing. As previously noted, when 2005 data from other
employment data for production workers. countries in the BLS series are compared with those from
In 2005, the average hourly compensation cost for pro- India, only the Philippines is found to have similar hourly
duction workers in Indias organized manufacturing sector compensation costs in the manufacturing industry as a

14 Monthly Labor Review May 2010

 Table 3. Employment in subsectors within India's organized they do in manufacturing as a whole.43
manufacturing sector, 200506
Comparisons of India with China
Percent
of total
manu- India and China are two countries that often have been
NAICS
code(s)
Subsector(s) facturing compared in terms of their manufacturing and develop-
employ-
ment ment potential. Even with the recent growth in Indias
(8,688) manufacturing activity, the manufacturing sector in India
3133 All manufacturing (excluding publishing)................ 100.0 is still considerably smaller than the manufacturing sector
311312 Food, beverage, and tobacco......................................... 20.9 in China. The $70 billion in manufacturing goods export-
313314 Textiles and textile product mills.................................. 15.3
325 Chemicals.............................................................................. 9.5 ed by India over the 2006 fiscal year is still only one-tenth
331 Primary metals.................................................................... 7.4 of the $700 billion in manufactured goods exported by
327 Nonmetallic mineral products....................................... 6.6 China in 2005.44 The difference in the magnitude of the
336 Transportation equipment.............................................. 6.4
315 Apparel................................................................................... 6.2
manufacturing sector can also be seen when one compares
333 Machinery............................................................................. 5.3 manufacturing activity with overall GDP for each country.
332 Fabricated metal products.............................................. 4.2 Over the period from 2000 to 2005, manufacturing ac-
326 Plastics and rubber products......................................... 3.6
counted for 32 percent of Chinas GDP, while account-
335 Electrical equipment, appliances, and 3.1
components.................................................................... ing for only 16 percent of Indias GDP.45 In 2005, 108.4
322 Paper........................................................................................ 2.3 million workers were employed in Chinas manufacturing
316 Leather and allied products............................................ 2.0
sector on average, while only 8.7 million were employed in
334 Computer and electronic products.............................. 1.6
324 Petroleum and coal products......................................... 1.0 Indias organized manufacturing sector, according to ASI
321 Wood products.................................................................... .6 200506.46 Even when workers in the unorganized sec-
337 Furniture and related products..................................... .4 tor are included, Indias total manufacturing employment
SOURCE: BLS estimates made by use of Annual Survey of Industries data is still dwarfed by employment in the Chinese manu-
from the Central Statistical Organisation of India.
facturing sector. For now, Chinas manufacturing sector
NOTE: The sum of the subsectors percents of total manufacturing em-
ployment does not equal 100 because of the exclusion from the table of outweighs Indiaseven when the unorganized sector is
certain subsectors whose data BLS does not publish. included.
In terms of population, India has been growing faster
than China, and it surpassed 1 billion people in the year
whole. International comparisons of hourly compensation 2000.47 In 1990, the population of India was equal to 73
costs in manufacturing subsectors also can be made. (See percent of the population of China. By 2008, Indias popu-
chart 6.) When hourly compensation costs are calculated lation had grown to equal 86 percent of the level in China.
as a percentage of those costs in the United States, labor Additionally, Indias population is younger than Chinas.
in India is found to be substantially less expensive than (See charts 7 and 8.) Because Indias population pyramid
labor in the Philippines in five industries: food, beverage, is currently bottom heavy, or concentrated in the younger
and tobacco manufacturing; textile and textile product age groups, over the next few decades the working-age
mills, chemical manufacturing; nonmetallic mineral prod- population will grow considerably. This larger labor supply
uct manufacturing; and transportation equipment manu- could serve as a source of growth for the manufacturing
facturing. Hourly compensation costs in these industries sector in India. Chinas population pyramid is different in
were at least 1.25 percentage points lower in India than in that the largest segment of the population is currently in
the Philippines when measured as a percentage of hourly the 3544 age range and the younger age groups contrib-
compensation costs in the United States. For countries ute less to the overall population. Thus, one would not ex-
with such low levels of labor costs, a difference of 1.25 pect the working-age population in China to experience
percentage points, or more, of the U.S. level is signifi- the same rate of growth as that in India.
cantin the food, beverage, and tobacco manufacturing The growing manufacturing sectors of India and China
industry, costs in the Philippines ($1.03) are actually dou- have attracted much interest in recent years. As regards
ble those in India ($0.51). Although these results can vary statistics, it was mentioned earlier that Indias statisti-
from year to year depending on currency exchange rates, cal system is already highly developed relative to that of
they do provide an example of labor costs within manu- many other developing countries, even as it strives to im-
facturing varying across countries to a greater extent than prove itself. In China, the private sector has been largely

Monthly Labor Review May 2010 15

Manufacturing in India

Chart 6. Hourly compensation costs in India and the Philippines as a percent of costs in the United States, measured in
. U.S. dollars, 2005
Subsector(s)
31 to 33, All manufacturing (excluding publishing)
311 and 312, Food, beverage, and tobacco
313 and 314, Textile and textile product mills
315, Apparel
316, Leather and allied products Philippines
321, Wood products India
322, Paper
325, Chemicals
326, Plastic and rubber products
327, Nonmetallic mineral products
331, Primary metals
332, Fabricated metal products
335, Electrical equipment, appliances, and components
336, Transportation equipment

0.0 1.0 2.0 3.0 4.0 5.0 6.0

Percent (U.S. costs = 100 percent)
SOURCE: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India;
.see ftp://ftp.bls.gov/pub/special.requests/ForeignLabor/aecountrytables.txt for Philippines data.

Chart 7. Population pyramid, India, 2006

Age Age

80+ 80+
7579 7579
7074 7074
6569 6569
6064 6064
5559 Males Females
5559
5054 5054
4549 4549
4044 4044
3539 3539
3034 3034
2529 2529
2024 2024
1519 1519
1014 1014
59 59
04 04
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7
Percent
SOURCE: Population Reference Bureau projections, based on the 2001 Census of India.

16 Monthly Labor Review May 2010

Chart 8. Population pyramid, China, 2007

Age Age

80+ 80+
7579 7579
7074 7074
6569 6569
6064 6064
5559 Males Females 5559
5054 5054
4549 4549
4044 4044
3539 3539
3034 3034
2529 2529
2024 2024
1519 1519
1014 1014
59 59
04 04
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7
Percent
SOURCE: U.S. Census Bureau, International Data Base.

neglected in statistics; the dominance of private sector tion estimation methods vary from those used in the se-
businesses in todays economy does not fit easily into the ries for India. Readers should refer to articles previously
theories and ideologies that prevailed in China in the published in The Monthly Labor Review for a comprehen-
recent past.48 During the most recent quarter century of sive description of the estimation methods used to calcu-
economic reform, China has been working to adopt bet- late hourly compensation costs for employees in Chinese
ter, internationally recognized statistical practices, with manufacturing.51
guidance from developed countries and from internation- For China, hourly compensation estimates can be bro-
al organizations such as The World Bank and the Inter- ken into three employment-based groups: all employees,
national Monetary Fund. However, much work remains employees in urban enterprises, and employees in town
to be done. and village enterprises. As discussed, the compensation
BLS has conducted extensive research on Chinas costs presented for India refer to all employees in the or-
manufacturing sector and published research on ganized sector. Compensation costs for employees in In-
employment and hourly compensation in Chinese dias unorganized sector are not presented here. Because
manufacturing.49 In November 2006, BLS published, for the employment groups are defined differently for each
the first time in a news release, a supplemental hourly country, and because of how difficult it can be to collect
compensation series for Chinese manufacturing; it reliable data on employment and compensation in both
covered the years 200204. To date, estimates for China India and China, there are limitations associated with
through 2006 are available from BLS.50 comparisons of hourly compensation costs between the
BLS now has estimates of hourly compensation for em- two countries. Nevertheless, BLS research on both coun-
ployees in manufacturing in both India and China. These tries indicates that the concept of all employees in the or-
estimates can be compared to gain insight into the rela- ganized manufacturing sector in India is similar enough
tive compensation costs in the two countries, but they are to the all employees concept for manufacturing in China
not derived by use of the same methods. The features of (estimates are calculated as the employment-weighted av-
the Chinese source data and the BLS hourly compensa- erage of Chinese urban and town and village enterprise

Monthly Labor Review May 2010 17

Manufacturing in India

Chart 9. Mean hourly compensation costs in the manufacturing sectors of India and China as a percent
of corresponding costs in the United States, 200205
Percent of Percent of
U.S. cost U.S. cost
3.5 3.5
Costs in India as percent of Costs in China as percent of
costs in U. S. costs in U.S.
3.0 3.0

2.5 2.5

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
2002 2003 2004 2005
NOTE: Mean hourly compensation costs for all manufacturing employees in China were the following: $0.57 in 2002, $0.62 in 2003, $0.67
in 2004, and $0.73 in 2005. The corresponding costs in India are reported in table 2.
SOURCES: BLS estimates made by use of Annual Survey of Industries data from the Central Statistical Organisation of India; International
comparisons of hourly compensation costs in manufacturing, 2007 (Bureau of Labor Statistics), March 26, 2009.

manufacturing) to allow for rough comparisons to be is estimated that the average manufacturer in India loses
made. 8.4 percent of its potential sales each year because of pow-
Organized-sector compensation costs in India and er outages, compared with less than 2 percent for the av-
compensation costs for all employees in Chinese manu- erage manufacturer in China.53 In 2005, annual spending
facturing were both very low in comparison with corre- on infrastructure as a share of GDP in India was 5.9 per-
sponding costs in the United States from 2002 through cent, compared with 14.6 percent in China.54 In addition,
2005. Chart 9 shows that costs in China were lower than the nature of manufacturing in India tends to be different
those in India each year. During this period hourly com- from that in China. Chinas factories tend to be very large
pensation costs increased by 25 percent in India and by 28 scale facilities that specialize in low-cost manufacturing of
percent in China as measured in U.S dollars. According to goods. In terms of value, the major items that are import-
preliminary BLS research, if data were available to create ed by the United States from China include the following:
a series on hourly compensation encompassing the total toys and sporting goods, miscellaneous household goods,
number of employees in Indian manufacturinginclud- computers and computer accessories, telecommunications
ing employees in both the organized and unorganized sec- equipment, video equipment, and cotton household fur-
torsthe estimate would be considerably lower because nishings and clothing.55 In India, extensive required pa-
workers in Indias unorganized sector earn substantially perwork, restrictive labor laws, and spotty power supplies
less than their organized-sector counterparts and greatly make large-scale factories less common than in China.
outnumber them. Instead of using big factories, a large portion of Indian
It has been reported that some manufacturers are find- manufacturing relies on a mix of technical skill and low-
ing labor shortages in China, a situation that is already cost labor to produce goods. India appears to have a com-
causing wages to rise and making goods costlier to pro- petitive advantage over China in the manufacture of such
duce.52 Businesses that choose India for offshore produc- items as cell phones, car parts, and apparel items that are
tion face challenges as well, many of them stemming from more complex to construct.56 In terms of value, the major
the current state of Indias infrastructure and labor laws. It manufacturing imports from India into the United States

18 Monthly Labor Review May 2010

are items such as jewelry; medicinal, dental and pharma- these years11.8 percent from 200506 to 200607 and
ceutical preparations; drilling and oil field equipment and 8.2 percent from 200607 to 200708. However, the glob-
platforms; and industrial machinery.57 al financial crisis that started in 2008 did not leave India
Although employers labor costs in Indian and Chinese untouched. Even though India is not a huge exporter and
manufacturing are currently at similar levels, a 2002 Con- has a large domestic market for its goods, growth slowed
federation of Indian Industry report created by McKinsey considerably, to 2.4 percent from 200708 to 200809 in
& Company indicated that the retail price of the average the organized sector. (When this article was authored,
Chinese product is about 30 percent lower than the retail only GDP figures for total manufacturing were available).
price of the same product produced in India, in spite of The global financial crisis also indirectly affected Indias
similar labor costs and other input costs.58 The Indian Na- growth potential because of the extent to which other
tional Manufacturing Competitiveness Council has gone countries around the globe were hit. Indias plan to invest
on record asserting that the key to improving Indias po- $500 billion in infrastructure improvements from 2008
sition in global manufacturing is to keep costs low.59 Of through 2012 may have to be revisited, since one-third
course, manufacturing involves many other costs as well, of that money was to come from the private sector. In
such as shipping, raw materials, and tariffs. The Council 2007, some of the worlds biggest banks and private-eq-
also strongly endorsed the Second National Labor Com- uity funds announced dedicated infrastructure funds with
missions recommendation that India harmonize its cur- India as a priority, and now, India is looking for those
rently scattered labor laws, stating that with the harmo- investors to begin building new roads. 62 As of April 2009,
nization not only will the flexibility improve in the orga- the National Highways Authority of India (NHAI) was
nized labor market, simultaneously better social security having difficulty finding bidders on its infrastructure proj-
provisions will also be made in the unorganized sector.60 ects. However, by April 2010, the NHAI had restructured
As more reforms are implemented and more resources in- its project plans and its bidding requirements to attract
vested, it will be of interest to the world whether India more bidders.63 In order for India to reach the level of
expands its share in global manufacturing. exports envisioned by Indias National Manufacturing
Competitiveness Council and for manufacturing to truly
Recent economic trends be the engine of growth that it envisions, infrastructure
growth in all formsroads, power sources, ports, and so
According to Indias Central Statistical Office, growth in forthlikely will be important. Of course, manufacturing
Indian manufacturing in fiscal year 200607 was strong. growth can be spurred by consumer demand as well. A
In the organized sector, at constant prices, the GDP growth recent BusinessWeek article states that domestic demand
rate from 200506 to 200607 was 11.6 percent.61 From accounts for two-thirds of the Indian economy and that
200607 to 200708, GDP growth slowed in the organized Indians can buy their way to growth.64 BLS will continue
sector, but was still impressive: 7.6 percent. In manufac- to make estimates and monitor trends in hourly compen-
turing overall, including both the organized and the unor- sation costs in Indias organized manufacturing sector as
ganized sectors, growth in GDP was slightly higher during updated ASI data are released by the CSO.65

Notes

ACKNOWLEDGMENTS: The authors thank Chris Sparks, Connie Table 3, Merchandise trade: leading exporters and importers, 2007,
Sorrentino, Bradley Nicholson, Elizabeth Zamora, Andrew Petajan, on the Internet at www.wto.org/english/news_e/pres08_e/pr520_e.
Jake Kirchmer and Marshall Carter, all of the BLS Division of Inter- htm#appendix_table3 (visited Apr. 26, 2010).
national Labor Comparisons, for their assistance in the preparation of
this article. 2
The National Strategy for Manufacturing (Government of India Na-
tional Manufacturing Competitiveness Council, March 2006), 1.1, p.
1
Table. PPP Conversion Factors and Share of Global Output, 2, on the Internet at http://nmcc.nic.in/pdf/strategy_paper_0306.
2007 (Washington, DC, International Monetary Fund, January 8, pdf (visited Apr. 26, 2010).
2008). Visit www.imf.org/external/pubs/ft/survey/so/2008/res018a.
htm and click on Link to PPP data under Related Links (visited Apr. 3
Statement 010. Summary of macro economic aggregates at constant
26, 2010); WTO: developing, transition economies cushion trade slowdown, (1999-2000) prices, 1950-51 to 2008-09 (Government of India, Min-
Press/520/Rev. 1 (World Trade Organization) Apr. 17, 2008, Appendix istry of Statistics and Programme Implementation, Central Statistical

Monthly Labor Review May 2010 19

Manufacturing in India

Organisation, National Accounts Division), on the Internet at www. manufacturing process or in cleaning any part of the machinery or
mospi.gov.in/mospi_nad_main.htm (visited Apr. 26, 2010). premises used for manufacturing process or in any other kind of work
incidental to or connected with the manufacturing process or the prod-
4
International Comparisons of Hourly Compensation Costs in Manu- uct. Workers engaged in repair and maintenance or production of fixed
facturing, News Release number USDL 090304, (Bureau of Labor assets for a factorys own use and workers employed in the production
Statistics, Mar. 26, 2009). of electricity or coal, gas, etc. are included. This definition is deemed
equal to the BLS definition of production workers, which is those em-
5
Judith Banister, Manufacturing earnings and compensation in ployees who are engaged in fabricating, assembly, and related activities;
China, Monthly Labor Review, August 2005, pp. 2240, on the In- material handling, warehousing, and shipping; maintenance and repair;
ternet at www.bls.gov/opub/mlr/2005/08/art3full.pdf; and Interna- janitorial and guard services; auxiliary production (for example, power
tional Comparisons of Hourly Compensation Costs in Manufacturing. plants); or other services closely related to the aforementioned activi-
ties. Working supervisors generally are included; apprentices and other
6
Report of the National Statistical Commission, section 1.1 (Govern- trainees generally are excluded. However, the ASI definition includes
ment of India, Ministry of Statistics and Programme Implementation, workers who do not receive wages. This inclusion of some additional
Sept. 5, 2001), on the Internet at http://mospi.gov.in/nscr/hp.htm workers is not believed to significantly affect the BLS estimates of
(visited May 11, 2010). hourly compensation costs.
7
Ibid. 16
The states of Arunachal Pradesh, Mizoram, and Sikkim, and the
union territory of Lakshadweep are not included in the geographical
8
Ibid, section 14.2. coverage of the ASI. The source of this information is Carl Haub and
O.P. Sharma, Hourly Compensation Costs for Workers in India, Novem-
9
Dr. Govindan Raveendran, Reforming the Indian Statistical System ber 2005, unpublished manuscript.
(Organisation for Economic Co-operation and Development), The
Statistics Newsletter, February 2006, on the Internet at www.oecd.org/ The Factories Act, 1948, Commercial Law Publishers (India) Pvt.
17

dataoecd/13/62/36132793.pdf (visited May 11, 2010). Ltd., Delhi, 2006.

See Key Indicators of the Labor Market (International Labour

10
18
The Collection of Statistics Act, 1953 (Government of India,
Organization), section 7, on the Internet at http://ilo-mirror.library. Ministry of Law), on the Internet at www.mospi.gov.in/mospi_stat_
cornell.edu/public/english/employment/gems/eeo/download/ act53.htm (visited May 3, 2010).
kilm07.pdf (visited May 12, 2010).
19
The Factories Act, 1948.
11
Informal Sector in India: Approaches for Social Security (Government
of India, Ministry of Labour), p. 2, on the Internet at http://labour. 20
For the 2007 NAICS definition of the manufacturing sector, visit
nic.in/ss/INFORMALSECTORININDIA-approachesforSo-
the BLS Web site at www.bls.gov/iag/tgs/iag31-33.htm (visited May
cialSecurity.pdf (visited Apr. 29, 2010). The unorganized sector in-
24, 2010).
cludes enterprises run by unincorporated businesses and partnerships,
in addition to cooperative societies (co-ops owned and managed by
and for the benefit of the customers or workers), trusts (corporations
21
For information on NIC 1998 see www.mospi.nic.in/nic_98.htm
organized to perform a fiduciary function), private companies (firms (visited May 3, 2010). Raw data for industries 0140, 1422, 2211, 2212,
not owned by the government) and limited companies (corporations 2219, and mining and utilities (industries 4000 to 4390) have been
with shareholders whose liability is limited by shares), all of which are excluded from the BLS estimates.
not included in the informal sector as defined by the International La-
bour Organization.
22
Note that the data published in this article do not match data
published by Indias CSO because of adjustments performed by BLS to
12
Statement 010. Summary of macro economic aggregates. make the data comparable with data calculated in a manner consistent
with NAICS.
13
Output is measured at factor cost. BLS was unable to locate reli-
able data that could indicate the portion of Indias manufacturing ex-
23
BLS has no information on the level of nonresponse to the ASI.
ports that are produced in the organized sector or the portion produced
in the unorganized sector. 24
International Comparisons of Hourly Compensation Costs in Manu-
facturing.
14
See the 19992000 Annual Survey of Industries (Government
of India, Ministry of Statistics and Programme Implementation), on 25
Amit K. Bhandari and Almas Heshmati, Wage Inequality and Job
the Internet at http://mospi.gov.in/mospi_asi.htm; and Employ- Insecurity among Permanent and Contract Workers in India: Evidence
ment and Unemployment in India, 1999-2000: Key Results, Report No. from Organized Manufacturing Industries, discussion paper no. 2097
455(55/10/1) (Government of India, Ministry of Statistics and Pro- (Institute for the Study of Labor, April 2006), p. 3, on the Internet
gramme Implementation, National Sample Survey Organisation), on at http://ideas.repec.org/p/iza/izadps/dp2097.html (May 3, 2010).
the Internet at http://www.mospi.gov.in/mospi_nsso_rept_pubn.
htm (visited May 11, 2010). The ASIs exact coverage of the manu- 26
Bhandari and Heshmati also point out that contract labor is not
facturing sector cannot be determined because the sample is drawn spread evenly across all industries within manufacturing. The ASI data
from the list of registered factories and not from a complete list of all support the claim of the Institute for the Study of Labor that labor-
manufacturing establishments in India. intensive industries like the tobacco industry hire a high percentage of
contract labor, whereas industries such as the pharmaceutical indus-
15
Indias ASI defines workers as all people employed directly or try that require more capital and highly skilled labor hire a relatively
through any agency, whether for wages or not, and engaged in any low percentage of contract labor. According to published ASI data, in

20 Monthly Labor Review May 2010

200506 contract workers accounted for 68.3 percent of all produc- tracted employees. For these reasons, BLS assumes no paid leave for
tion workers in Indias organized tobacco industry and only accounted contracted employees in the hourly compensation estimates presented
for 31.7 percent of the production workers in the chemicals industry in this article.
(which includes pharmaceuticals). As previously mentioned, contract
workers accounted for 28.6 percent of all production workers in Indias Haub and Sharma, Hourly Compensation Costs for Workers in India,
38

manufacturing sector in 2005. November 2005, unpublished manuscript.

27
The Factories Act, 1948, section 59 (1), states: Where a worker 39
Indias Pension Reform: Chronology of Events, Invest India Eco-
works in a factory for more than nine hours in any day or for more than nomic Foundation, on the Internet at www.iief.com/chronology.htm
forty-eight hours in any week, he shall, in respect of overtime work, be (visited May 5, 2010).
entitled to wages at the rate of twice his ordinary rate of wages.
40
In a 2006 paper, the Conference Board published an estimate of
28
Carl Haub and O.P. Sharma, Hourly Compensation Costs for Work- compensation per employee for India in 2002. See Bart van Ark, Judith
ers in India: 1989-1990 to 1997-1998 and 2003-2004, September 2006, Banister, and Catherine Guillemineau, Competitive Advantage of Low-
unpublished manuscript. Wage Countries Often Exaggerated, (The Conference Board, Executive
Action Series, No. 212, October 2006), p. 5. The estimate is for large-
29
Although publishing is not included in the NAICS definition of scale manufacturing only, which includes registered manufacturing
manufacturing, printing is included in manufacturing under NAICS enterprises onlythat is, those enterprises in the organized sector.
subsector 323: printing and related support activities. The Conference Board reports that Indian manufacturing employees
received compensation at a level equal to 2.5 percent of the level
30
Data on the structure of labor costs are used to analyze relationships of compensation in U.S. firms. BLS estimates put Indian hourly
among various components of labor costs. For example, structure-of- compensation at a level equal to 2.7 percent of the U.S. level in 2002.
labor-costs data can provide information on the percent of total labor The small difference between these numbers is likely due to differences
costs that is accounted for by the cost medical insurance. in estimation methods. One obvious difference is that the Conference
Board estimates measure the ratio of annual compensation per
31
BLS was unable to locate data to serve as a proxy for pay in kind. employee in India to annual compensation per employee in the United
BLS was unable to find conclusive evidence regarding what portion of States, whereas BLS estimates measure the hourly compensation ratio.
total compensation pay in kind represents for the organized manufac- The Conference Board estimates that large-scale manufacturing
turing sector, but it is believed to be small, and its exclusion should not employed 7.8 million employees in 2002which includes unpaid
significantly affect the estimates. family members, sole proprietors, etc. BLS omits this group of
workers and only considers paid employees when estimating hourly
compensation costs. See Judith Banister, India and China: Demography,
32
Employers in Indian manufacturing currently are not subject to
Human Capital, and Socioeconomic Transformations (The Conference
any taxes or subsidies linked to the level of employment in their firms;
Board, 2007), p. 27. Additionally, as mentioned earlier, BLS excludes
therefore, this component of total compensation is zero.
employment and compensation data from the ASI for industries
that do not fit within the NAICS definition of manufacturing. BLS
33
Interviews were conducted in New Delhi by Carl Haub and O.P. estimates that there were 7.5 million paid employees in the organized
Sharma during a 7-day period in July 2006. Out of the 120 employers manufacturing sector in 2002.
in Faridabad who were mailed survey forms, 10 employers returned the
completed form and 15 addresses were found to be invalid. Haub and 41
See Table 1. Production Workers: Indexes of hourly compensa-
Sharma, Hourly Compensation Costs for Workers in India: 1989-1990 to tion costs in U.S. dollars in manufacturing, 34 countries or areas and
1997-1998 and 2003-2004, September 2006, unpublished manuscript. selected economic groups, 1975-2007 (Bureau of Labor Statistics,
March 2009), on the Internet at ftp://ftp.bls.gov/pub/special.re-
34
BLS calculated hourly compensation using an average of 9 hours quests/ForeignLabor/ichccpwsuppt01.txt (visited May 5, 2010).
worked per day to see how the change in working time would affect
the estimate. The result was that the change in working time had little 42
The industries with the highest levels of employment are not nec-
effect. When measured in U.S. dollars, mean hourly compensation for essarily the industries that contribute the most to Indias position in
all employees was $0.81 in 2005 and was still equal to approximately 3 the global economy, however. According to The National Strategy for
percent of the U.S. level. Manufacturing, gems and jewelry, textiles and garments, engineering
goods, chemicals, and leather and leather goods account for approxi-
35
The Factories Act, 1948, chapter 8. mately 75 percent of Indias exports.

The 10 paid holidays included in BLS estimates are: New Years

36 43
For a full list of the BLS international hourly compensation cost
Day, Holi, Id-ul-Fiter, Raksha Bandhan, Guru Nanks birthday, estimates for both all employees and production workers, visit the BLS
Dusshera, Diwali, Ambedkar Jayanti, Krishnas birthday, and Christ- Web site at www.bls.gov/ilc/ (visited May 6, 2010).
mas. Some states observe more holidays than others; BLS chose to ac-
count for these major 10 paid holidays across all Indian states because 44
Michael Schuman, The Drive to Compete, Time, June 19,
they are those which function as paid holidays almost everywhere. 2006, on the Internet at www.time.com/time/magazine/print-
out/0,8816,1205526,00.html (visited May 6, 2010).
37
Given that a manufacturers responsibility for employees em-
ployed by contractors, as well as its need to keep records of these em- 45
Gordon H. Hanson and Raymond Robertson, China and the
ployees, ends when the contract is issued, it is not possible to estimate Manufacturing Exports of Other Developing Countries (Cambridge,
any amount of paid leave that contracted employees may receive. This is Mass., National Bureau of Economic Research, July 2007), on the In-
especially true of work delegated on short-term contracts. In addition, ternet at www.nber.org/books_in_progress/china07/cwt07/hanson.
it is not legally required that employers provide any paid leave to con- pdf (visited May 6, 2010).

Monthly Labor Review May 2010 21

Manufacturing in India

46
Erin Lett and Judith Banister, Chinas manufacturing employ- 56
Anand Giridharadas, India, Known for Outsourcing, Expands in
ment and compensation costs: 200206, Monthly Labor Review, April Industry, The New York Times, May 19, 2006, on the Internet at www.
2009, p. 32, on the Internet at www.bls.gov/opub/mlr/2009/04/art- nytimes.com/2006/05/19/business/worldbusiness/19factory.html
3full.pdf (visited May 6, 2010). (visited May 11, 2010).

See the U.S. Census Bureaus International Data Base at www.

47 57
U.S. Imports from India by 5-digit End-Use Code 2005-2009.
census.gov/ipc/www/idb/ (visited May 11, 2010), click on Data Ac-
cess, and select the country and years for which you would like to 58
This information was obtained from The National Strategy for
download data. Manufacturing, p. 20; the original source is listed as Learning from
China to unlock Indias manufacturing potential (CII-McKinsey,
48
Judith Banister, Manufacturing employment in China, Monthly October 2002). McKinsey & Company undertook a study on behalf
Labor Review, July 2005, p. 11, on the Internet at www.bls.gov/opub/ of the Confederation of Indian Industry (CII) in March 2002. The
mlr/2005/07/art2full.pdf (visited May 7, 2010). objective was to understand the drivers of Chinese competitiveness
in manufacturing and identify how India could put its manufacturing
49
Judith Banister, Manufacturing Employment and Compensation in sector on the path to high growth. BLS estimates indicate that in 2002
China (Bureau of Labor Statistics, November 2005), on the Internet at hourly compensation costs in China were 22 percent lower than those
www.bls.gov/fls/chinareport.pdf (visited May 7, 2010). in India, as shown in Chart 9. For reasons described in this article, es-
timates from China are not directly comparable with those from India.
50
Lett and Banister, Chinas manufacturing employment and com-
pensation costs: 200206, pp. 3038. 59
The National Strategy for Manufacturing, 3.3.2, p. 20.
51
Ibid; and Banister, Manufacturing Employment and Compensa-
tion in China, pp. 2647.
60
Ibid, 4.2.2.12, p. 64.

52
Barbara Demick and David Pierson, People, people everywhere
61
Statement 10. Summary of macro economic aggregates.
in China, and not enough to work, Los Angeles Times, Mar. 28, 2010,
on the Internet at http://articles.latimes.com/2010/mar/28/world/ 62
Geeta Anand, Indias Infrastructure Funds Fall, Wall Street Jour-
la-fg-china-labor28-2010mar28 (visited May 11, 2010). nal, Apr. 28, 2009.

53
The National Strategy for Manufacturing, 3.6.4, pp. 3435. See also 63
Sobia Khan, NHAIs new bid norms may speed up road proj-
The long journey, an article published in the June 3, 2006, issue of The ects, The Economic Times, Mar. 13, 2010, on the Internet at http://
Economist. On page 11, Vineet Agarwal of the Transport Corporation economictimes.indiatimes.com/news/economy/infrastructure/
of India describes the typical journey cargo must make between Kol- NHAIs-new-bid-norms-may-speed-up-road-projects/article-
kata and Mumbai. The 1,340 mile trip takes 8 days at an average speed show/5678094.cms (visited May 11, 2010); and NHAI to invite fresh
of less than 7 miles per hour. More than 32 hours are spent waiting at bids for 38 projects, Business Standard, Apr. 15, 2009, on the Internet
tollbooths and checkpoints. at www.business-standard.com/india/news/nhai-to-invite-fresh-
bids-for-38-projects/58608/on (visited May 11, 2010).
54
India urged to copy China in infrastructure spending, The China
Post, May 5, 2008, on the Internet at www.chinapost.com.tw/busi- 64
John Lee, Dont Underestimate Indias Consumers, BusinessWeek,
ness/asia/india/2008/05/05/155047/India-urged.htm (visited May Jan. 21, 2010, on the Internet at www.businessweek.com/print/
7, 2010). magazine/content/10_05/b4165084462859.htm (visited May 7,
2010).
55
U.S. Imports from China by 5-digit End-Use Code 2005-2009
(U.S. Census Bureau), on the Internet at www.census.gov/foreign-
trade/statistics/product/enduse/imports/c5700.html (visited May At the time this article was published, the CSO had released data
65

7, 2010). from ASI 200607 and ASI 200708.

22 Monthly Labor Review May 2010