Manufacturing Compensation in China

Manufacturing Compensation in China

Manufacturing earnings
and compensation in China
On the basis of published earnings data,
estimated compensation ratios, and estimated hours,
China’s manufacturing employees averaged
about 57 cents compensation per hour worked in 2002

W
Judith Banister ith by far the world’s largest manu- enterprises (TVE’s).3 Combining the published
facturing workforce, at more than 100 earnings figures and adjusted labor compensation
million,1 China is widely known to have figures for these two groups results in a reasonable
low labor costs. Statistics available for the approximation of average 2002 labor compensation
first time for the entire country for 2002 now permit per manufacturing employee in China. A national
the estimation of those costs with some degree of time series on compensation for China could not be
precision. Employees in China’s city manu- developed due to the lack of earnings data for the
facturing enterprises received a total compen- country’s noncity manufacturing workers prior to
sation of $0.95 per hour, while their noncity 2002; however, data on trends in real (price-
counterparts, about whom such estimates had adjusted) earnings for city manufacturing em-
not previously been generally available, aver- ployees from 1990 onward are available and show a
aged less than half that: $0.41 per hour. Alto- sharp upward trend since 1998.
gether, with a large majority of manufacturing Because China has not systematically collect-
employees working outside the cities, the aver- ed and reported adequate data on actual hours
age hourly manufacturing compensation esti- worked by manufacturing employees for the whole
mated for China in 2002 was $0.57, about 3 percent year 2002 or, indeed, for any full year, this article
of the average hourly compensation of manu- uses published partial labor force survey infor-
facturing production workers in the United States mation and a set of hypotheses to estimate annual
hours worked by city and noncity manufacturing
and of many developed countries of the world.
employees, thus calculating approximations of
Equally as striking, regional competitors in the
average 2002 hourly labor compensation in manu-
newly industrialized economies of Asia had, on
facturing for these two categories of manufacturing
average, labor costs more than 10 times those for
employees and for China as a whole. Labor compen-
China’s manufacturing workers; and Mexico and sation estimates are converted into U.S. dollars at
Brazil had labor costs about 4 times those for China’s the official exchange rate for 2002.
manufacturing employees. The article also assesses the probable biases in
This article evaluates the quality and usability of China’s statistics on manufacturing earnings and
China’s statistics on manufacturing earnings and total labor compensation. The analysis that follows
Judith Banister is a
consultant working
labor compensation for 2002—the most recent year argues that city manufacturing enterprises in partic-
with Javelin for which adequate data are available—and for the ular have powerful incentives to underreport earn-
Investments in Beijing, period since 1990. The analysis demonstrates that ings and other elements of the compensation pro-
China. She is former
head of the
China has released just enough relevant data on vided to their employees. The main purposes of
International average annual earnings and labor-related employer underreporting employee compensation are to
Programs Center at costs to derive 2002 estimates of annual labor avoid taxes and to minimize required employer and
the U.S. Census
Bureau. E-mail:
compensation for 30 million city manufacturing employee payments to social insurance and em-
Judith_Banister employees2 and 71 million noncity manufacturing ployee housing funds administered by urban
@yahoo.com employees—those working in town and village authorities.

22 Monthly Labor Review August 2005

 There is, however, a competing bias in city manufacturing among countries, and the official exchange rate is not a reliable
employment and earnings data. Indirect evidence indicates that indicator of the relative difference in prices between China and
many city manufacturing workers are not included in these other countries.4
numbers at all. In particular, the lower paid migrant manufacturing As will be demonstrated in the analysis, the numbers fre-
workers seem to be considerably underrepresented in the quently published in the global and U.S. popular media on the
reported urban employment data for cities, and the earnings of low compensation of China’s manufacturing workers ($0.40–
most of the comparatively poorly paid migrant workers in general $1.50 per hour) are within the realm of reasonable estimates.
also appear to be excluded from urban manufacturing earnings China is indeed a relatively low wage manufacturing environ-
data. Whether the net result of these competing biases is to ment, and the country also enjoys other advantages that give it
underreport or overreport earnings of the average urban manu- a competitive edge over many other manufacturing locations
facturing employee for 2002 is unclear; however, it is likely that around the world.
the exclusion of the more stagnant earnings of the rural-to-urban This article is the second of a two-part series on manufacturing
migrants leads to some exaggeration of the trend of rising average labor statistics in the People’s Republic of China (hereinafter,
earnings in city manufacturing for the 1990–2002 period. “China”).5 The earlier article6 focused on levels and trends of
The analysis that follows discusses the cost to employers of manufacturing employment; this one estimates average hourly
employee compensation and the competitiveness of Chinese labor compensation for China’s manufacturing employees. A
manufacturing in the global economy. For comparative pur- more detailed exposition of the analysis in the two articles is
poses, official exchange rates were used to convert compensa- found on the Bureau of Labor Statistics (BLS) Web site.7 Occa-
tion costs to U.S. dollars. The official exchange rate is the sionally, that report refers to terminology in Chinese because
appropriate conversion rate for compensation cost comparisons, the standard English translations of the terms are misleading or
because it reflects the cost in U.S. dollars that employers must ambiguous and, in some cases, because there is no succinct
actually pay for Chinese labor. Compensation costs converted and accurate English translation of the term. A complete glossary
with the use of commercial exchange rates do not, however, of Chinese terms used in this and the earlier article can be found
indicate relative living standards of workers or the purchasing at the end of the report on the BLS Web site.
power of their income, for at least two reasons. First, because
they include costs that are not paid directly to the worker, com-
pensation costs do not provide an accurate portrayal of worker
Background
income. Second, prices of goods and services vary greatly
The Bureau of Labor Statistics publishes estimates of hourly
compensation costs for production workers in manufacturing
for 31 economies on its Web site.8 Although most of the coun-
The Bureau of Labor Statistics has been a leader in compiling tries are developed countries with high-quality data, some
international comparisons of hourly compensation of manufac- developing countries with adequate data also are included. The
turing workers over a wide range of countries. Despite its large and
growing importance in world manufacturing, China has not been
Bureau is working to add countries, including China, to the
included in the comparisons because of difficulties in obtaining and published list, but BLS standards for the quality of statistics are
interpreting that country’s data and because of concerns about the high. Data for China are not yet in accord with BLS comparability
quality of the data. Although the two Monthly Labor Review articles definitions. (See box, this page.) This article assesses the quality
by Judith Banister have greatly facilitated understanding of Chinese and completeness of those statistics which are available on
employment and compensation statistics, many problems with
data availability, coverage, and reliability remain, as described in manufacturing earnings and compensation in China.
the articles. Therefore, the Bureau does not plan to include China in The subsequent analysis is based as much as possible on
its regular comparisons of hourly compensation costs at this time. information published by China’s official statistical organi-
These articles and the associated report on the BLS Web site, which zations. Most statistics for China are collected under the central
have been funded by the Bureau, are intended as first steps toward guidance of the National Bureau of Statistics (NBS) and often
developing the measures necessary to include China in the regular
comparisons series that currently includes 31 countries. Because
are published jointly with the Ministry of Labor and Social
of the widespread interest in expanded country coverage, the Bureau Security (hereinafter, Ministry of Labor). Collecting data on
is indeed considering providing data on China, along with data on manufacturing employment and earnings in TVE’s, however, is
some other countries, the quality of whose data is problematic, but the responsibility of the Ministry of Agriculture, and data on the
in a separate format with appropriate annotations. As better data earnings of noncity manufacturing employees were first pub-
become available, China and other countries could be moved into
the regular comparisons series.
lished for the year 2002.9
Focusing on 2002, the most recent year for which adequate
Division of Foreign Labor Statistics, Bureau of Labor Statistics data are available, the upcoming discussion tabulates infor-
mation on earnings, required social benefit payments, and other

Monthly Labor Review August 2005 23

Manufacturing Compensation in China

labor compensation and derives annual, monthly, and estimated China’s statistical authorities at the NBS also try to use an
hourly manufacturing labor compensation, in Chinese yuan, for internationally recognizable definition of employee com-
urban, TVE, and all-China manufacturing employees. These esti- pensation in the calculation of China’s gross domestic product.
mates are then calculated in U.S. dollars at the official exchange The NBS defines what it variously translates as “compensation
rate. of employees” or “laborers’ remuneration” (laodongzhe bao-
The annual data on labor compensation in manufacturing chou) as follows:
used in this article come from the annual yearend statistical
reporting system. (China’s censuses do not ask for earnings Laodongzhe baochou refers to the whole payment of
data.) In China’s cities and, to a lesser extent, outside the cities, various forms earned by the laborers from the productive
each enterprise, economic unit, small business, or self-employed activities they are engaged in. It includes wages, bonuses,
individual or group is required to report employment and earn- and allowances the laborers earned in monetary form and in
kind. It also includes the free medical services provided to
ings data each year according to the group’s “labor situation”
the laborers and the medicine expenses, transport sub-
the previous year and at the previous yearend. The data are
sidies, social insurance, and housing fund paid by the
then compiled upward in a statistical reporting chain to the
employers.11
national government. Accountants or those who report employ-
ment and earnings figures on behalf of their enterprises or other This passage suggests that China’s government either collects
work units (at least, those in urban areas) are given detailed data on these various components of worker compensation or
instructions on how to report monthly, quarterly, yearend, and at least estimates them for its calculations of China’s gross
average annual figures on employment and earnings. The domestic product.
instructions are based on regulations released by the NBS, The subsequent analysis begins with a description of Chinese
especially those released in 1990, with further clarifications in earnings statistics on manufacturing workers and then describes
1998 and 2002.10 the sources and methods of estimating the nonearnings portions
In reporting annual statistics on employment and earnings, of compensation—that is, the social insurance expenditures that
China’s NBS and Ministry of Labor use an administrative report- employers must pay on behalf of employees. Two issues that
ing system that ignores the progress China has made in the are relevant to the estimation of social insurance expenditures,
statistical definitions of “urban” and “rural” during the last namely, the difference by city in mandatory social insurance
several decades. As mentioned in the earlier companion piece to contribution rates and the likely underreporting of earnings to
this article, in statistical publications on China’s labor force, minimize tax and social insurance contributions, are discussed.
employment and earnings data labeled “urban” actually refer to The article then examines the difficult issue of estimating working
cities and exclude employees working outside narrowly defined time in manufacturing in order to construct estimates of compen-
city boundaries. Even factories located in suburbs, large sation on a per hour basis. Following an analysis of the compen-
industrial parks, and towns that have been officially established sation of manufacturing employees in export-oriented industries
as urban places since the 1980s are excluded from the so-called
and of migrant workers, the discussion touches on how
urban statistics on employment and earnings. In the tables and
manufacturing earnings in China have changed over time and
charts of the current article, statistics are faithfully shown as
how the compensation estimates in this article compare with
they were reported in official publications. In the text, the word
those published in other venues. Finally, the implications of the
“city” often is used to describe the “urban” data, simply because
those data actually refer to city employees and their earnings. current research results for China’s competitiveness are explored.
By contrast, the term “town and village enterprises” (TVE’s) Throughout the analysis, separate estimates are made for
seems to cover not only rural areas, but also factories in urban workers and TVE workers, because the data sources and
urbanized places outside narrow city boundaries. Accordingly, the working situations that relate to each group are different.
the text uses the word “noncity” to refer to TVE data. Where possible, national estimates combining the two groups
are made as well.
The concept of compensation
Reported manufacturing earnings in
The BLS measures of hourly compensation costs include both Chinese currency
data on hourly direct pay (which includes pay for time worked,
pay for vacations and holidays, bonuses, in-kind pay, and other Earnings and other compensation data for manufacturing
premiums) and data on employer social insurance expenditures workers in China are poorly and partially reported. The available
and other labor taxes (which include employer expenditures for data on “wages” or “earnings” come from the annual yearend
legally required insurance programs and contractual and private reporting system, and the fragmentary figures are published in
benefit plans, as well as other taxes on payrolls or employment). the China Labor Statistical Yearbook and, for TVE employees,

24 Monthly Labor Review August 2005

China Village and Town Enterprise Yearbook 2003.12 Average and workers” whose earnings that year averaged 11,001 yuan,
annual remuneration for manufacturing workers is called and 5 percent were the 740,000 “other” city manufacturing
“wages” (gongzi) when referring to staff and workers, but is workers, who averaged much higher earnings of 17,237 yuan in
called “earnings” or remuneration (laodong baochou) when 2002 (in part because this category includes foreign employees
referring to the other employees of urban manufacturing units. of China’s manufacturing companies and reemployed or still
The two terms appear to mean the same thing, and both are employed retirement-age workers with high seniority, and both
defined as follows: these groups probably get higher earnings than the average for
“staff and workers”).
The total wages and total earnings are calculated this way: The 11,152-yuan average annual earnings figure of the 30
They include whatever is paid to or for the workers in money million workers in manufacturing urban units masks a wide
or in kind according to relevant regulations, including range of earnings in different urban manufacturing subsec-
salaries paid for a certain time period or payments based on tors, as shown in table 2. For example, the lowest-paid group
piece work, bonuses, allowances, subsidies, overtime pay, of city manufacturing workers is the 3 million textile industry
and pay for dangerous or challenging duty.13 workers, whose earnings average 7,268 yuan per year. The 5
million city manufacturing workers in the subsectors of timber
In this article, the term “earnings” designates the wages or and bamboo products, food processing, nonmetal mineral
earnings of both urban and TVE manufacturing employees in products, paper products, furniture manufacturing, and “other”
cash and in kind, as reported to statistical and tax authorities. manufacturing also earn less than the average urban worker:
The term does not include the social insurance payments that their reported average annual earnings are less than 9,000 yuan.
employers are required to pay to city or county authorities on At the other end of the pay spectrum, the 7.5 million city
behalf of their employees or the welfare fund payments given to manufacturing workers in tobacco processing, electronics and
employees in the enterprises. The terms “compensation” and telecommunications, petroleum processing, ferrous metal
“total compensation” include earnings plus these other elements smelting, transport equipment manufacturing, and medical and
of total labor compensation in manufacturing. These definitions pharmaceutical products all have average annual earnings of
correspond to the definitions used by the Bureau of Labor 13,000 yuan or higher.
Statistics in its international report on hourly compensation The recorded 9 million laid-off manufacturing workers still
costs. nominally connected to their manufacturing units averaged a
Table 1 shows that the 30 million on-post employees of small annual living subsidy of 2,213 yuan. (See table 1.) This
manufacturing enterprises in China’s cities had average reported kind of payment might be considered similar to payments of
earnings of 11,152 yuan for the year 2002.14 Of these employees, unemployment compensation for laid-off or unemployed
95 percent were on-post (not laid-off or unemployed) “staff workers in developed countries.

Table3.1.
Table Published earnings of manufacturing employees in China, 2002

Category of manufacturing Total earnings Number of Average number Average earnings Average living
workers paid employees of employees per employee subsidy
(billions of yuan) (yearend, millions) (millions) (yuan) (yuan)

2
Manufacturing in urban units ..................... 334.39 29.81 29.98 11,152 —
On-post urban manufacturing staff and
workers ................................................ 321.90 29.07 29.26 11,001 —
Other urban manufacturing employment — .74 — 17,237 —
Laid-off urban manufacturing staff and
workers .................................................... — 9.13 — — 2,213
Manufacturing TVE’s1 .................................. 489.22 70.87 2
70.62 2
6,927 —
Large-scale manufacturing TVE ’s1 ........... 168.94 19.05 2
18.98 2
8,899 —

1
TVE’sare town and village enterprises. directly reported. All figures for manufacturing in urban units exclude self-
2
Derived from other numbers reported in the table or in the sources. employed individuals and small privately owned firms.
NOTES: Dash indicates data are not available or not applicable. In the SOURCES : China National Bureau of Statistics and China Ministry of
sources, remuneration for workers in urban manufacturing units and for other Labor, compilers, China Labor Statistical Yearbook 2003 (Beijing, China
urban manufacturing employees is called “earnings” (laodong baochou), whereas Statistics Press, 2003), pp. 29, 34, 46, 169, 171, 179, 230, 243, 249, 473;
remuneration for on-post urban manufacturing staff and workers is called China Ministry of Agriculture, TVE Yearbook Editorial Committee, ed., China
“wages”(gongzi). For manufacturing TVE’s, only the total 2002 expenditure for Village and Town Enterprise Yearbook 2003 [in Chinese] (Beijing, China
earnings (laodongzhe baochou) is reported; the average per employee is not Agriculture Publishing House, 2003), pp. 130–31.

Monthly Labor Review August 2005 25

Manufacturing Compensation in China

Table 2. Urban manufacturing employment and in 2002, 62 percent of the average earnings that year for em-
earnings by subsector in China, 2002 ployees of urban manufacturing units. Workers in large-scale
Average
manufacturing TVE’s had higher average 2002 earnings of 8,899
Urban
employees
earnings yuan, 80 percent of the average reported earnings for employees
Urban manufacturing subsector per employee
(yearend)
(yuan)
of urban manufacturing units.
What forms of remuneration are included in the average annual
Total manufacturing in urban units ... 29,984,619 11,152 earnings figures for China’s manufacturing employees? Exhibit 1
Tobacco processing .......................... 233,485 23,744 lists all the items whose value is required to be included in
Electronics and telecommunications .. 1,623,783 17,636 earnings data reported by enterprises in urban China for their on-
Petroleum processing and coking
products ......................................... 565,505 17,357 post manufacturing staff and workers, based on written in-
Smelting and pressing of ferrous structions to enterprise accountants and statistical personnel.
metals ............................................ 1,900,648 15,032
Transportation equipment Most forms of income, benefits, and subsidies in cash and in
manufacturing ................................ 2,319,421 14,409 kind are on this list. Cash salary and earnings payments, housing
Medical and pharmaceutical
products ......................................... 844,857 13,207 and transportation provided to workers, meals given to them,
Instruments and office machinery .... 464,762 12,720 and the value of income tax and social insurance payments
Smelting and pressing of nonferrous
metals ............................................ 755,646 12,491 deducted from earnings and remitted to the government on behalf
Electric equipment and machinery ... 1,441,399 12,405 of employees are all required to be included in the “total earnings”
Chemical fibers manufacturing .......... 263,378 11,404
Printing and record medium figure, based on relevant reporting regulations.
reproduction ................................... 493,497 10,863 One group of benefits that is provided by some of China’s
Ordinary machinery manufacturing ... 1,921,315 10,668
Special-purpose equipment manufacturing enterprises to employees, but that is specifically
manufacturing ................................ 1,400,594 10,406 excluded from the earnings figures, is the use of a company
Cultural, educational, and sport
products ......................................... 294,636 10,390 medical clinic or the payment of some employee hospital costs.16
It would seem that this is an important group of benefits which,
Chemical raw materials and
products ......................................... 2,213,256 10,359 conceptually, ought to be included in earnings data. But many
Plastic products ................................ 606,800 10,131 countries share this shortcoming in earnings statistics, with the
Metal products .................................. 897,455 10,075
Food products manufacturing ........... 621,757 10,064 result that the Bureau of Labor Statistics specifically excludes
Rubber products ................................ 377,633 10,055 the costs of medical clinics in plant facilities from its comparative
Beverage manufacturing ................... 740,250 9,619
Leather, furs, down, and related international data on labor compensation in manufacturing.17
products ......................................... 578,590 9,108 This article does not include any estimation of these particular
Garments and other fiber products ..... 1,336,191 9,066
Furniture manufacturing .................... 180,484 8,881 medical benefits which are missing from China’s earnings data.
Other manufacturing ......................... 601,416 8,781 One important difference between China’s earnings data
Papermaking and paper products ..... 592,400 8,668
Nonmetal mineral products ............... 2,116,034 8,123 shown in table 1 and the data used by the Bureau in its inter-
Food processing ................................ 977,439 7,965 national comparisons is that the Bureau data relate only to
Timber, bamboo, natural fiber and
straw products ............................... 267,666 7,339 production workers, while the Chinese data relate to all
Textile industry .................................. 2,841,565 7,268 employees—that is, both production and nonproduction
workers. Because production workers typically have lower wages
NOTES: These data refer only to urban manufacturing employment and
earnings.The subsectors listed here refer to 29.47 million of China’s urban than those of nonproduction workers, it is likely that the inclusion
manufacturing workers. Rural manufacturing workers in each subsector of both types of workers in the Chinese data leads to higher
undoubtedly have lower earnings than those displayed here. The earnings
figures shown do not include required employer social insurance payments earnings levels. However, the production worker data necessary
or other nonwage labor costs. to match the BLS concept are not available for China, so it is
SOURCES: China National Bureau of Statistics and China Ministry of
Labor, compilers, China Labor Statistical Yearbook 2003 (Beijing, China
unclear how much lower Chinese earnings for production workers
Statistics Press, 2003), pp. 179 and 218–25. would be.
The earnings data do not include figures for the comparatively
In years prior to 2002, earnings data were not published for small privately owned manufacturing groupings and the self-
manufacturing workers outside the cities. For the reported 71 employed manufacturing workers in China’s cities. These two
million manufacturing TVE employees in 2002, the Ministry of categories of workers together totaled 8.2 million (22 percent of
Agriculture published, for the first time, the total earnings China’s reported total of urban manufacturing workers) in 2002,
(laodongzhe baochou) paid out for that entire year in all according to China’s State Administration for Industry and
manufacturing TVE’s.15 Average annual earnings per worker are Commerce.18 This feature of China’s earnings data parallels the
derived in table 1 in the same way that the average annual same dearth in manufacturing earnings data from many countries.
earnings are calculated for urban manufacturing workers. TVE For reasons of practicality, if a country does not include earnings
manufacturing workers averaged 6,927 yuan in reported earnings for employees in small manufacturing units in its earnings data,

26 Monthly Labor Review August 2005

Exhibit 1. Components of Chinese urban earnings statistics

The statistical concept of wage (gongzi) or earnings for on-post urban “staff and workers” includes the following compo-
nents, whether the employees receive the earnings or benefits in money or in kind and whether the earnings or benefits are or
are not taxable items:

Monthly or annual salary income (including base earnings Housing subsidy (dormitory provided, or directly subsidized
and additions based on position, seniority, wage scale, rent or purchase of housing)
and so on) Individual income tax deducted from earnings and paid
Earnings during on-the-job training, probationary period directly by enterprise to government
Employee income paid on an irregular basis Social insurance funds (pension, medical, unemployment
Hourly payment for work performed insurance funds, and housing purchase fund) deducted
Piecework payment for work performed from the employee's wage and paid by the work unit to
Bonus payments
government on behalf of the employee
Incentive, performance-based payments
Money for rent, and utilities (electricity, water)
Overtime pay
Money given for fixed line or mobile phone
Hardship, danger pay
All kinds of subsidies in cash or in kind Clothing subsidy
Festival, holiday subsidy Subsidy compensating workers for lack of vacation time
Travel money, food allowance while traveling Earnings during approved leaves of absence, pay for time
Transport subsidy (car or shuttle bus provided, cash for bus not worked (regular vacation, compassionate leave, to visit
or taxi, and so on) relatives, family-planning operation, national or societal
Personal services such as baths, haircuts duty, study leave, leave due to sickness or injury)
Books, newspapers, magazines provided for employees Anything that has the nature or spirit of labor earnings, even
Meals provided, food allowance if it is not spelled out in the regulations

SOURCE: Laodong gongzi; tongji taizhang [Labor wages; statistical accounts ] (Beijing, Beijing Municipality Statisti-
cal Bureau, 2004), pp. 2–1 to 2–5.

the Bureau also excludes the employees and compensation for insurance funds and housing funds to which both employers
these units from its estimates of hourly labor compensation in and employees are required to contribute each month.20 There
manufacturing.19 Self-employed workers in manufacturing also are six kinds of funds: an old-age pension fund, a medical
are excluded from the Bureau’s estimates. Using data from insurance fund, an unemployment insurance fund, a workers’
manufacturing censuses, the Bureau has researched the effect compensation fund, a maternity leave fund, and a fund in
of excluding such earnings and found it to be small. which money is set aside for each worker by name—money
that the worker can use to help buy an apartment. These
Estimating total 2002 compensation in monthly payments by employers to city governments are
manufacturing mandatory, and stiff penalties are specified for noncom-
pliance,21 but noncompliance is rampant and penalties are
To estimate total compensation for China’s manufacturing rarely enforced.
employees, it is necessary to add to the reported earnings the The payments deducted from employee earnings for the six
other components of total compensation, including social public funds and remitted to city governments are included in
insurance payments paid by employers on behalf of employees, the reported earnings data (see exhibit 1), but the part paid by
as well as other payments to or for employees that are not employers is excluded.22 Legally required payments to gov-
included in the earnings data. ernment social insurance and employee benefit programs are
In the urban areas, employers pay considerable sums for included in the BLS concept of compensation,23 so, in order to
social welfare benefits on behalf of their employees, above adjust the reported manufacturing earnings to include legally
and beyond the employees’ earnings. China’s cities today required employer social insurance payments and other labor
have built, or are in the process of building, municipal social compensation costs, one needs to know the overall per-

Monthly Labor Review August 2005 27

Manufacturing Compensation in China

centage of the total earnings bill that urban manufacturing contribution to the social insurance system and, in some cities,
employers paid in 2002 for social insurance and required the home purchase fund varies from city to city.26 For example,
housing fund payments, as well as other employee benefit the following tabulation shows the additional amount, ex-
payments. China’s Ministry of Labor conducted a survey of pressed as a percentage of earnings, that manufacturing em-
11,704 urban enterprises in 51 large and medium-sized cities ployers in three cities are required to contribute:27
throughout the country and collected all relevant worker
compensation data from these organizations for the year Changshu City, Wuxi City,
2002.24 This article uses the results of that large survey to Jiangsu Jiangsu Beijing
estimate average labor compensation costs in urban manu- Contribution Province Province Municipality
facturing above and beyond the reported earnings data for
2002 given in table 1. On the basis of the results of this Labor Old-age pension fund ........ 16.5 22.0 20.0
Ministry survey, the reported 2002 annual earnings should be Medical insurance fund ..... 8.0 8.0 9.0
increased by an amount equivalent to 53.8 percent of earnings Unemployment insurance . 2.0 2.0 1.5
to estimate the following labor compensation costs (expressed Workers’ compensation
insurance ........................ .6–.8 — 1.0
as a percentage of urban earnings) actually paid by em- Maternity leave insurance . 1.0 — —
ployers:25 Employee housing fund ..... — — 8.0

Cost Percent Not only do the required employer contributions vary by munic-
Required employer social insurance ipality and city, but also, the amounts have been increasing over
payments to the government ............................... 28 time. Therefore, it is likely that the legally required employer
Required housing fund payments ............................ 4 contribution to the social insurance funds for the average manu-
Additional employee welfare costs facturing employee has increased since 2002.
not included in earnings ....................................... 12
Other labor-related costs The inclusion in total labor compensation of the amorphous,
not specified in detail ........................................... 10 vaguely reported categories of welfare costs and other un-
specified labor-related costs just discussed may help offset some
of the likely downward biases in the basic earnings data. To
In table 3, therefore, average 2002 total compensation for em- minimize individual and corporate taxes and required social
ployees of urban manufacturing enterprises is estimated to insurance payments, urban employers tend to underreport
be 17,152 yuan. earnings to the extent possible, neglecting to include some in-
Note that the amount China’s urban employers are required kind benefits in the reported earnings and offloading as many
by law to remit to the government every month as the employer employee subsidies and benefits as possible into the welfare

Table3.3.
Table Estimated labor compensation of manufacturing employees in China, 2002

Average
Average earnings Annual compensation Monthly compensation Hourly compensation
Category of manufacturing number of per per employee per employee per employee
workers employees employee
(millions) (yuan) U.S. U.S. U.S.
Yuan Yuan Yuan
dollars dollars dollars

Total for manufacturing urban units

and TVE ’s1 ..................................... 100.61 8,186 10,363 $1,252 864 $104 4.73 $0.57
Manufacturing urban units ............. 29.98 11,152 17,152 2,071 1,429 173 7.87 .95
On-post urban manufacturing
staff and workers .................... 29.26 11,001 16,920 2,043 1,410 170 7.76 .94
Other urban manufacturing
employment .............................. .72 17,237 26,511 3,202 2,209 267 12.17 1.47
Manufacturing TVE’s1 ...................... 70.62 6,927 7,481 904 623 75 3.40 .41
Large-scale manufacturing TVE’s1 18.98 8,899 9,611 1,161 801 97 4.37 .53

1
TVE’s are town and village enterprises. that TVE workers perform 2,200 hours per year. (See text for details.)
SOURCES: Table 1; China National Bureau of Statistics and China Ministry
NOTES : Total labor compensation for urban workers is 1.538 times of Labor, compilers, China Labor Statistical Yearbook 2003 (Beijing, China
earnings and for TVE workers is 1.08 times earnings. U.S. dollars are calcu- Statistics Press, 2003), pp. 29, 34, 46, 169, 171, 179, 230, 249, 473; China
lated at the 2002 prevailing commercial exchange rate: 8.28 yuan = U.S.$1. Ministry of Agriculture, TVE Yearbook Editorial Committee, ed., China Village
Hourly compensation is calculated under the assumption that urban and Town EnterpriseYearbook 2003 [in Chinese] (Beijing, China Agriculture
manufacturing employees perform 2,179 actual hours of work per year and Publishing House, 2003), pp. 130–31.

28 Monthly Labor Review August 2005

fund category or “other” labor compensation category. (Under- joint venture or merger with, a (usually state-owned) Chinese
reporting of urban manufacturing employment and earnings is company, the foreign company insists on engaging in a due
discussed shortly.) diligence process to determine whether the joint venture, merger,
For TVE manufacturing employees, there is ample evidence or acquisition is in the interests of its owners and shareholders.
that the reported earnings total may capture almost all of their The auditors and accounting companies frequently discover that
total compensation, because TVE workers do not have many of the target company has two sets of books: “Most domestic
the social insurance and other welfare benefits that urban enterprises keep separate sets of ‘management accounts’ and
employees often get. For example, by the end of 2002, the number ‘tax accounts.’”31 The “tax ledger” is the set of employee and
of rural and smalltown workers with any rural social pension financial data reported to the tax and other authorities, and
insurance was minuscule.28 China’s urban towns and rural areas the “administrative ledger” records a more accurate picture of
have very weak or nonexistent social benefit systems for pen- the numbers of employees, their actual earnings, the true costs
sions, medical insurance, unemployment insurance, workers’ and income of the company, its actual profits, and more. The
compensation, and the like. Pension and medical insurance tax ledger is designed to minimize tax exposure, particularly
systems paid into by employers and employees essentially do corporate income taxes, value-added taxes, personal income
not exist in China outside of cities today.29 A survey of large taxes for employer and employees, and required social benefit
manufacturing enterprises in Nanjing Municipality, the capital payments. It is believed that non-public-sector domestic
of Jiangsu Province on the country’s east coast, found that wel- Chinese enterprises avoid taxation and social benefit pay-
fare benefits for workers, above and beyond earnings, for the ments to an even greater extent than the state-owned and
years 1994–2001 averaged 36 percent of the earnings in urban collective-owned enterprises.
state-owned manufacturing enterprises, but only 16 percent of Such tax avoidance in the manufacturing sector probably
the earnings in unusually large manufacturing TVE’s in counties has a number of implications.32 First, many urban employees,
under Nanjing’s administration.30 Now, on the one hand, these especially those who are in-migrants and do not have city
TVE’s surely had an exceptionally high level of welfare benefits residence permits or those who are temporary or part-time
compared with those offered by all manufacturing TVE’s in workers, may be left off the books entirely, at least with regard
China during those years, both because TVE’s in counties to what is reported to authorities. When they are, their em-
near major cities have better social welfare benefits than TVE’s ployment is kept informal, and neither the employee nor his or
elsewhere and because large TVE’s have better benefits than her earnings, which are paid in cash, are reported. This means
average-sized TVE’s. On the other hand, average manufacturing that the employee can avoid paying income tax and any required
TVE worker welfare benefits in 2002 were very likely a higher social insurance deductions, while the employer can avoid
percentage of those workers’ total compensation than in earlier paying the required social insurance payments for the em-
years. Therefore, pending the discovery of better data for 2002, ployee. As a result, actual manufacturing employment may be
the average total of social insurance and other welfare benefits underreported in China’s statistics, especially in the urban
for China’s manufacturing TVE employees can be tentatively figures.33
estimated to be in the range from 0 percent to 16 percent of their Second, even when employment is reported to authorities,
total earnings. A reasonable estimate of such employee benefits both employer and employees tend to collude to minimize
for the average TVE employee in 2002 is 8 percent, the midpoint reported earnings. Employers in urban areas are required to remit
of the range. Table 3 estimates average annual total compensation to the city government social insurance and other payments that
for TVE employees at 7,481 yuan. are calculated as a percentage of the unit’s reported total earnings.
These required payments are high by international standards
Underreporting of urban manufacturing and have been increasing rapidly: “high contribution rates are
employment and earnings leading to high rates of evasion in the basic pension system,” as
well as evasion of other required social welfare payments.34 Many
China’s people and work units were unaccustomed to paying employers might perceive that the required payments are
income taxes, value-added taxes, corporate income taxes, or high squeezing their profits and are burdensome; they would
payments for social insurance during the Maoist decades from therefore have an incentive to underreport employee earnings.
1949 to 1978. The government extracted the money for its budget Some of the money actually given to employees (as bonuses,
in other ways, but not so visibly as the way taxes are taken out overtime pay, or financial subsidies of various kinds) may not
now. Individuals got benefits in both urban and rural areas, while be reported as earnings, instead getting shifted to the welfare
earnings were kept very low. Today, during the post-Mao eco- fund category or other unspecified labor-related cost cate-
nomic reform era, employers appear to have developed a culture gory; thus, it is important to include these labor cost cate-
of tax avoidance. For example, when foreign and multinational gories in a realistic estimate of urban manufacturing labor
companies come to China and attempt to acquire, or set up a compensation in China. It is also likely that many urban enter-

Monthly Labor Review August 2005 29

Manufacturing Compensation in China

prises underreport or leave out of reported earnings the value of Monthly labor compensation in
some benefits provided in kind to employees (for example, meals, manufacturing
housing, transportation, and food distributions). Therefore, it is
likely that even the earnings of urban manufacturing workers To calculate the monthly compensation of TVE manufacturing
whose employment is reported to authorities are systematically workers from their average annual labor compensation, it would
underreported. be helpful to know whether all or even most of the reported 71
Those employees whose employment is not reported to the million TVE manufacturing employees work most of the year and
what proportion are part-year or part-time workers. As noted
authorities at all, whether in urban or rural areas, are usually paid
earlier, it is likely that many unreported workers do not work year
lower wages than other employees. According to anecdotal
round. If the assumption is made that these 71 million reported
evidence, the going rate for an unskilled rural or migrant worker
workers represent year-round workers, then their average
in nonagricultural work in China today is about 500–600 yuan monthly total compensation was about U.S.$75. (See table 3.)
per month, plus whatever benefits it is essential to provide, Urban manufacturing employees are, generally speaking, year-
such as simple meals, dormitories, and emergency medical round, full-time employees. Monthly urban manufacturing labor
assistance. Some rural workers are paid as little as 300 yuan compensation was U.S.$173.
per month, while more desirable workers might get as much as
800 yuan monthly. If unreported workers in the manufacturing Annual hours worked in manufacturing
sector average cash pay of 550 yuan per month, and if their
simple accommodations and food cost another 200 yuan per To calculate the hourly labor compensation of China’s manu-
month, then their earnings total 750 yuan, or U.S.$91, per month, facturing employees in 2002 would require data on the average
but only when they are actually working. Thus, if, for 3 months of number of hours actually worked per employee during that year.
the year, they are not engaged in paid employment while planting Some data have been published on China’s urban manufacturing
and harvesting and while taking time off for holidays, illnesses, employees’ average hours worked in 2002. Specifically, China’s
and personal business, then their annual take-home cash plus NBS and Labor Ministry have been conducting a labor force

in-kind benefits would be 6,750 yuan per year. This estimate is survey for some years. Most results of this survey have not
close to the reported data that yield earnings of 6,927 yuan for been published, but data on hours worked by urban manu-
facturing workers during 2 reference weeks of 2002 have been
TVE manufacturing workers in 2002.
published. According to the survey, urban manufacturing em-
ployees in China actually worked an average of 44.86 hours
Annual dollar compensation for during the 7-day period from May 9 to May 15, 2002, and 46.0
manufacturing workers hours during the reference week of September 24–30, 2002.37
Averaging those two figures results in the estimate that, dur-
To translate reported average annual earnings for China’s
ing 2002, in the weeks when urban manufacturing employees
manufacturing workers into dollars (see table 3), the analysis
actually worked at all, they averaged 45.4 hours of work per
that follows uses official nominal exchange rates between U.S.
week.
dollars and Chinese yuan. The Chinese yuan was pegged to
The remaining problem is to estimate the average number of
the U.S. dollar at 8.28 yuan per dollar for a decade from weeks actually worked by urban manufacturing employees in
1994 to August 2005; this exchange rate is the correct one for China during 2002. Because urban employees are supposed to
2002 data.35 receive a total of 10 days of statutory holidays per year, it is
On the basis of reported earnings data only, China’s 30 reasonable to assume that urban manufacturing employees get 2
million employees of urban manufacturing units had average weeks of public holidays per year. It is also reasonable to assume
2002 earnings of 11,152 yuan, or U.S.$1,347, at the official ex- that urban manufacturing employees, on average, missed 1 week
change rate. China’s manufacturing workers in TVE’s averaged per year for some combination of illness, injury leave, and mater-
6,927 yuan, or U.S.$837, in reported annual earnings in 2002. nity leave and 1 week per year for personal leave plus work stop-
(See tables 1 and 3.) After adjusting reported earnings to pages and downtime due to equipment repair and shortages of
account for additional indirect and direct remuneration for electricity and manufacturing inputs. On the assumption that
employees, table 3 estimates that China’s urban manufactur- China’s urban manufacturing workers actually worked 48 weeks
ing employees received an average of about U.S.$2,071 in during 2002, averaging 45.4 hours per week, the average annual
annual labor compensation for 2002, while TVE manufacturing hours worked are estimated to be 2,179 hours.
employees got approximately U.S.$904. It is important to note, No data have been published or released on average hours
however, that TVE employment is highly desirable to China’s worked per week by rural or TVE manufacturing employees, even
rural workers because their TVE earnings are higher than the though such data were collected for September 24–30, 2002, in
earnings they can derive from agriculture.36 China’s October 2002 labor force survey.38 All of the calculations

30 Monthly Labor Review August 2005

that follow are therefore strictly hypothetical. Because labor laws 23,005 yuan, or $2,778. Then, if some city manufacturing
are more explicit and more enforced in cities than outside the employees worked 4,000 hours in 2002 for that income, hourly
cities, it is likely that, during each week that manufacturing compensation was $0.69 per hour. Outside Guangdong’s cities
employees actually are working, those in cities work fewer hours in Guangdong Province, reported 2002 average earnings in
than those outside the cities. Therefore, it is in this case reason- industry were 8,345 yuan. (See table 4 and the discussion that
able to assume that TVE manufacturing workers averaged 50 follows.) Increasing this figure by 8 percent to adjust for social
hours of work per week in 2002 during those weeks that they insurance payments on the part of employers results in a total
were working. Also, assuming that TVE manufacturing em- average labor compensation of 9,013 yuan, or $1,088, in 2002.
ployees took 2 weeks off for Chinese New Year and stopped For those factories whose workers put in 4,000 hours of
work for another 2 weeks for reasons such as illness, injury, fam- production work that year, per hour average labor com-
ily emergencies, personal leave, and factory downtime due to pensation was $0.27. This illustration emphasizes why it is
shortages and breakdowns, this would leave 48 weeks of actual important to determine the actual average number of hours
work per year. In addition, some TVE manufacturing employees worked in each year for both city and TVE manufacturing
who work in the same county as their home village also may be employees.
involved in agriculture during peak seasons. This assump- Data from China’s 2000 census confirm that, generally
tion is made because most TVE workers come from rural speaking, manufacturing employees in China work a lengthy
households that still grow crops, and farm households tend week; at least, they did during the last week of October 2000.
to need all the labor they can get for planting and harvesting. The census indicated that 58 percent of manufacturing work-
However, migrant manufacturing workers would not be able ers had worked 6 or 7 days the previous week; however, the
census may have classified tens of millions of part-year,
to get home to participate in agriculture, and some manu-
seasonal manufacturing workers from rural areas and small
facturing workers who live close to their family homes have left
towns as farmers.41 Such rural (probably called TVE) manu-
agriculture altogether. It is therefore reasonable to assume that,
facturing workers would put in far fewer hours in manu-
say, one-half of TVE manufacturing workers take leave from their
facturing per year than those counted in the census or those
manufacturing jobs for 2 weeks for peak planting time twice a
working year round in coastal-zone factories. Thus, the
year (assuming double-cropping, on average) and 2 weeks for
percentage of workers who worked 6 or 7 days probably was
each of two peak harvest seasons, thus working 40 weeks per
lower than 58 percent.
year in manufacturing, but that the other half of TVE manu-
It is not known whether manufacturing employees whose
facturing workers do not do agricultural work and, as a conse-
factories sell only to China’s domestic market work about the
quence, work 48 weeks in manufacturing each year. Under these
same number of hours per week, month, or year as does the
assumptions, TVE manufacturing workers would have averaged
average employee of export-oriented factories. Of China’s
44 weeks of actual factory work in 2002 at 50 hours per week,
reported 70.9 million TVE manufacturing employees in 2002, for
totaling 2,200 hours for the year.
example, only 13.4 million were reported to be producing for
It is possible that the estimate for the numbers of hours export, while 57.5 million were apparently producing only for the
worked, on average, per year by manufacturing employees in domestic market.42 An adequate estimate of average annual hours
city and noncity factories is too low. Some investigations in worked must take into account both of these categories of
China’s export zones in Guangdong and other coastal provinces manufacturing workers—those who produce for export and
have discovered many factories in which the employees typically those who produce for domestic sale.
work the entire year, with a 2-week holiday at Chinese New Year. For China, legal limits on working hours or overtime hours are
In many such export-oriented factories, employees usually work not likely to yield realistic estimates of actual hours worked.
6 or 7 days each week, totaling 60 to 80 hours per week in whatever Factories routinely report that they are abiding by the regulations
period constitutes the peak season for that manufacturing when, in fact, employees are working more hours per day, and
subsector.39 This season can last up to 8 months a year. Average many more hours per week or month, than the statutory limits.
yearly hours actually worked per employee might be as high as One purpose of the double bookkeeping in China’s factories is
4,000 hours in some China manufacturing enterprises. Suppose to report compliance with laws on minimum wages and maximum
that, in those hardworking Guangdong factories, the average permissible overtime hours when, in reality, the factory routinely
urban wage in 2002 was 14,958 yuan, as discussed shortly and violates the laws. Generally speaking, grassroots investigators
as reported in table 4, and suppose also that urban earnings report that the factories do not claim that they paid more total
must be increased by 53.8 percent to include all employer earnings per month or per day to the employees than they
social insurance payments, welfare costs, and other labor actually paid; rather, they underreport the actual hours worked
costs,40 giving an average annual labor compensation of to earn the reported monthly or daily income.

Monthly Labor Review August 2005 31

Manufacturing Compensation in China

Hourly labor compensation in data from the BLS series. Even compared with some of the lower
manufacturing cost countries in the series, Chinese costs are low: a quarter of
the cost level in Brazil and Mexico and less than a tenth of the
Despite the limitations on estimates of annual hours worked, average of Hong Kong, Korea, Singapore, and Taiwan.44
it is possible to produce reasonable estimates of hourly com-
pensation costs for manufacturing workers in China, as is Manufacturing labor compensation in key
shown in table 3. According to these estimates, compensation export regions
for employees of urban manufacturing units was about
U.S.$0.95 per hour of work and for TVE manufacturing China’s urban manufacturing earnings statistics are reported
employees was about U.S.$0.41 per hour.
by province, which facilitates estimating urban manufacturing
The analysis presented herein combines labor compen-
labor compensation for the leading export centers. Using the
sation estimates for the reported 71 million TVE manufac-
same ratio of additional compensation to earnings, namely,
turing employees and the 30 million manufacturing employees
of urban units to derive estimates for annual, monthly, and hourly 53.8 percent, as in table 3, table 4 adjusts the earnings of
labor compensation in China’s manufacturing sector. As urban manufacturing workers to derive annual, monthly, and
shown in table 3, these 101 million Chinese manufacturing hourly labor compensation for the city manufacturing workers
employees received an average of approximately U.S.$1,252 of four leading provinces in China’s manufacturing import
in labor compensation in 2002, a figure that works out to about and export trade. (Actual levels of additional compensation
U.S.$104 in monthly labor compensation and implies an hourly as a percentage of earnings vary by province and by munic-
labor compensation of around U.S.$0.57 for China’s manu- ipality, but data are not available to adjust earnings by using
facturing employees.43 different multipliers for the urban manufacturing workers in
How does that U.S.$0.57 compare internationally? Chart 1 different provinces.)
shows manufacturing hourly compensation costs in China in The three provinces of the Yangtze River Delta have a wide
relation to the same costs in several other countries. Chinese range of urban manufacturing earnings and labor compensation.
costs are 3 percent of those in the United States, according to As shown in table 4, Shanghai’s 1.3 million city manufacturing

Chart 1. Average hourly compensation costs of manufacturing workers, selected economies

and regions, 2002
U.S. = 100 U.S. = 100
($21.11) ($21.11)
120 120

100
100 94 100
88

80 80

60 60

40 40
33

20 20
12 12
3
0 1
0
United States China Brazil Mexico EU(15) Japan Asian NIE'S 2
1
EU(15) are the European Union member countries prior to the expansion to 25 countries on May 1, 2004.
2
Asian NIE's are the newly industrialized economies of Hong Kong, Korea, Singapore, and Taiwan.
SOURCE: Bureau of Labor Statistics, "International comparisons of hourly compensation costs for production workers in
manufacturing,1975–2003," Nov. 18, 2004; on the Internet at http://www.bls.gov/fls/home.htm. For China, data are from this article and
not from the BLS series. The data for China refer to all employees rather than just production workers.

32 Monthly Labor Review August 2005

TableTable
4. Compensation of urban manufacturing employees and TVE1 industry employees, Yangtze Delta provinces
and Guangdong, China, 2002

Adjusted annual labor Adjusted monthly labor Adjusted hourly labor

Annual compensation compensation compensation
Province earnings
(yuan) U.S. U.S. U.S.
Yuan Yuan Yuan
dollars dollars dollars
Urban manufacturing employees:
National average .......................... 11,152 17,152 $2,071 1,429 $173 7.87 $0.95
Shanghai municipality ..................... 21,957 33,770 4,078 2,814 340 15.50 1.87
Zhejiang province ............................ 13,435 20,663 2,496 1,722 208 9.48 1.15
Jiangsu province ............................. 11,731 18,042 2,179 1,504 182 8.28 1.00
Guangdong province ....................... 14,958 23,005 2,778 1,917 232 10.56 1.28

TVE1 industry employees:

National average .......................... 6,891 7,442 $899 574 $69 3.13 $0.38
Shanghai municipality ..................... 11,939 12,894 1,557 1,075 130 5.86 .71
Zhejiang province ............................ 10,188 11,003 1,329 917 111 5.00 .60
Jiangsu province ............................. 8,143 8,794 1,062 733 89 4.00 .48
Guangdong province ....................... 8,345 9,013 1,088 751 91 4.10 .49

1
TVE ’s are town and village enterprises. SOURCES: Table 3; China National Bureau of Statistics and China Ministry
NOTES: U.S. dollars are calculated at the 2002 prevailing commercial ex- of Labor, compilers, China Labor Statistical Yearbook 2003 (Beijing, China
change rate: 8.28 yuan = U.S.$1. Hourly wage estimates for urban workers Statistics Press, 2003), pp. 179, 473; China Ministry of Agriculture, TVE
are calculated under the assumption that urban manufacturing employees Yearbook Editorial Committee, ed., China Village and Town Enterprise
perform 2,179 actual hours of work per year and that TVE workers perform Yearbook 2003 [in Chinese] (Beijing, China Agriculture Publishing House,
2,200 hours per year. (See text for details.) 2003), pp. 156, 174.

workers are comparatively highly paid in the Chinese context. areas and U.S.$0.60 an hour in Zhejiang Province’s rural and
Their 2002 labor compensation averaged about U.S.$4,078, and industrial zones outside of its cities. Noncity industry workers
hourly compensation was approximately U.S.$1.87. Manu- in Jiangsu and Guangdong Provinces were not as well paid,
facturing workers in Zhejiang, Jiangsu, and Guangdong had lower receiving U.S.$0.48 and U.S.$0.49 per hour, respectively.
labor compensation than Shanghai, but still higher than the
national average. Earnings of migrant manufacturing workers
These city manufacturing earnings statistics for China’s
leading export-manufacturing regions do not yield a true picture In theory, if a worker has migrated from a village to a city and
of the earnings paid by manufacturing enterprises in those is employed in a manufacturing enterprise, the employer
provinces. In the first place, it is not certain that the earnings of should report the migrant’s job and earnings in the “manu-
most migrant manufacturing workers in the cities of the facturing staff and worker” category. But in practice, in most
aforementioned provinces are included in the urban manu- cities of China, migrants who do not possess permanent-
facturing earnings data. Second, no wage data are reported for resident documents are apparently not eligible for urban social
the so-called rural manufacturing workers by province, nor insurance and housing benefits:
are TVE manufacturing earnings figures reported by province.
However, reported earnings statistics are available by prov- Contracted rural migrant laborers are supposed to be
ince for TVE industry (gongye) employees. Nationally, 92.4 covered [in the social basic pension system] as well.
percent of TVE industry workers are manufacturing em- While the inclusion of rural migrant labor in urban areas
ployees, and wages of these manufacturing workers are would also reduce the dependency ratio because of the
concentration of migrant laborers in the young working
similar to those of other industry workers. Therefore, TVE
age groups, present weaknesses in administrative capac-
industry earnings by province can be used to estimate manu-
ity make it questionable whether these workers will ever
facturing earnings.
draw benefits, especially if they return to rural areas or
Table 4 also reports 2002 TVE industry earnings and
move on to other urban areas. In some cases, the pension
derives labor compensation for the same regions. Like their
contribution is simply an added tax from which the
urban counterparts, TVE industry workers in these regions
migrant will derive no benefits.45
have higher earnings than the national average. Shanghai
and Zhejiang TVE industry employees were the highest paid, There is increasing informal evidence that published urban
earning U.S.$0.71 per hour in the Shanghai suburban and rural earnings data exclude the pay of most migrant workers.46 The

Monthly Labor Review August 2005 33

Manufacturing Compensation in China

earlier companion piece to this article47 referred to published facturing earnings over several years, the data required to
2002 statistics on manufacturing employment in urban units, construct such series over time are sparse. Published data on
totaling 29.81 million, that included 4.59 million rural-to-urban earnings trends for the manufacturing sector are available
migrants whose household registration was still in rural areas. solely for urban manufacturing staff and workers. Table 5
Probably, their reported earnings were part of the published presents published information on annual percent changes in
average earnings data for urban manufacturing staff and workers, average real earnings for this subset of city manufacturing
but very likely, many millions more rural-to-city migrant manu- employees. Real living standards have been rising in China’s
facturing workers were not in the reported urban manufacturing cities, and real earnings have been rising for urban staff and
employment or earnings data. There are many possible reasons workers in manufacturing.49 The “staff and worker” component
for such exclusion, including the fact that many cities and of urban manufacturing workers is supposed to include manu-
municipalities in China do not consider rural-to-urban migrants facturing workers who migrated into cities from rural areas, but
to be real urban or municipal employees.48 It is not known the rising wages indicated in table 5 probably exclude data on the
whether these migrant manufacturing workers and their earnings earnings of most rural-to-urban migrant manufacturing workers.50
get picked up in the TVE manufacturing data. Reported urban manufacturing earnings rose rapidly in the early
It is reasonable to assume that TVE manufacturing employ-
1990s, slowly in the mid-1990s, and very rapidly at the end of the
ment and earnings data usually include the migrant manu-
1990s and on into the early 21st century. Tables 5 and 6 and chart
facturing workers in towns and rural areas. The reason is that,
2 show that these generalizations about city manufacturing
because of the much lower ratio of social insurance costs in
earnings trends also hold for manufacturing employees in state-
towns and rural areas, there is almost no incentive to leave these
owned units, collective-owned units, and “other” ownership
workers out of the data in those areas, in contrast to the situation
in cities, where the higher ratio of social insurance costs affords units (joint ventures, foreign-owned firms, multinational com-
a financial incentive to exclude migrant workers. There is no panies, and the like).
separate reporting of the earnings of migrant manufacturing Table 6 and chart 2 present trends in real annual earnings (not
workers either in the cities or outside urban areas. including required employer payments for social insurance plans
or other nonwage labor costs) for urban manufacturing staff and
Manufacturing earnings over time workers in China. In 1990, the 53 million urban manufacturing
staff and workers earned an average of 5,058 yuan (in constant
Most of the data in this article relate to the year 2002 only. 2002 yuan). As the number of urban manufacturing staff and
Although it would be revealing to analyze trends in manu- workers shrank to 29 million in 2002, the earnings of those

Table 5.TablAnnual percent change in average real (price-adjusted) earnings of urban manufacturing staff and workers
in China, selected years, 1979–2002

Urban state- Urban collective- Other urban

Year Total owned units owned units ownership units

1979 ............................................................ 9.1 7.4 4.4 —

1980 ............................................................ 5.4 5.2 7.5 —
1985 ............................................................ 4.1 3.4 6.9 17.9
1986 ............................................................ 7.1 8.6 4.3 7.5
1987 ............................................................ 2.2 2.6 .8 7.6
1988 ............................................................ –.1 .5 –2.5 14.0
1989 ............................................................ –4.5 –4.4 –5.7 .9
1990 ............................................................ 7.7 8.6 5.2 4.4
1991 ............................................................ 5.1 4.1 5.4 12.9
1992 ............................................................ 6.0 6.2 3.3 5.5
1993 ............................................................ 9.4 6.2 5.4 1.1
1994 ............................................................ 2.3 1.2 –.3 .1
1995 ............................................................ 3.3 1.6 3.5 1.8
1996 ............................................................ .3 –.4 –.9 .8
1997 ............................................................ 2.0 .5 –.3 2.3
1998 ............................................................ 5.1 2.3 2.4 –1.8
1999 ............................................................ 11.8 10.5 7.6 10.3
2000 ............................................................ 11.4 11.5 6.6 8.5
2001 ............................................................ 10.9 11.3 5.7 7.9
2002 ............................................................ 13.7 14.6 12.0 9.7

NOTE: Dash indicates data are not available. compilers, China Labor Statistical Yearbook 2003 (Beijing, China Statistics
SOURCE: China National Bureau of Statistics and China Ministry of Labor, Press, 2003), pp. 36, 39, 42, 45.

34 Monthly Labor Review August 2005

 Table3.6.
Table Average annual real earnings of urban manufacturing staff and workers in China, 1990–2002
[In constant 2002 yuan and constant 2002 dollars]
Urban state-owned Urban collective- Other urban ownership
Total
units owned units units
Year
U.S. U.S. U.S. U.S.
Yuan dollars Yuan Yuan Yuan
dollars dollars dollars

1990 ................................................... 5,058 $611 5,599 $676 4,149 $501 6,833 $825
1991 ................................................... 5,316 642 5,828 704 4,373 528 7,714 932
1992 ................................................... 5,635 681 6,189 748 4,517 546 8,138 983
1993 ................................................... 6,165 745 6,573 794 4,761 575 8,228 994
1994 ................................................... 6,307 762 6,652 803 4,746 573 8,236 995
1995 ................................................... 6,515 787 6,759 816 4,913 593 8,384 1,013
1996 ................................................... 6,534 789 6,731 813 4,868 588 8,452 1,021
1997 ................................................... 6,665 805 6,765 817 4,854 586 8,646 1,044
1998 ................................................... 7,005 846 6,921 836 4,970 600 8,490 1,025
1999 ................................................... 7,832 946 7,647 924 5,348 646 9,365 1,131
2000 ................................................... 8,724 1,054 8,527 1,030 5,701 689 10,161 1,227
2001 ................................................... 9,675 1,169 9,490 1,146 6,026 728 10,964 1,324
2002 ................................................... 11,001 1,329 10,876 1,314 6,749 815 12,027 1,453

NOTE: This table presents only the reported annual earnings, which have SOURCE: China National Bureau of Statistics and China Ministry of
not been adjusted to include other labor compensation costs, such as Labor, compilers, China Labor Statistical Yearbook 2003 (Beijing, China
required employer payments to municipal social insurance systems. Statistics Press, 2003), pp. 34–45.

remaining averaged 11,001 yuan, more than double the 1990 equivalent of $1.50 per hour in wages and benefits.”55 Table 2
average earnings. There was a shift in the composition of the indicates that China’s urban transportation equipment manu-
“urban manufacturing staff and workers” category over that 13- facturing workers had average 2002 earnings of 14,409 yuan,
year period.51 In 1990, the lowest-paid subgroup, urban collective which would translate into about 80 cents an hour for earn-
manufacturing workers, was large (18 million) and held down ings alone and $1.23 per hour for total compensation. There-
average real earnings, while the highest-paid subgroup, private- fore, the overseas reports of the compensation of auto work-
sector enterprises, was minuscule. By 2002, the highest-paid ers in China are compatible with the data presented in this
subgroup constituted more than half of urban manufacturing article.
staff and workers. This trend toward the better paid private sector One journal wrote, “China is already by far the biggest
raised average earnings among urban staff and workers in manu- garment exporter in the world, with average wages in the
facturing. industry of 40 cents an hour.”56 That figure is close to the 41
cents an hour that the foregoing analysis has posited for the
Estimates of manufacturing employee compensation of China’s TVE manufacturing employees.
compensation Garment workers outside the cities are paid less than that,
because they are among the lower paid manufacturing em-
Many media and other sources around the world have published ployees in China. Table 2 indicated that urban garment workers
very rough estimates of hourly or monthly earnings or total average 9,066 yuan per year, or approximately 50 cents per
compensation for manufacturing workers in China. A comparison hour, in earnings; their total compensation might be about 77
of their estimates with those in this article is instructive. For cents an hour. If so, then the estimate of 40 cents per hour is
example, one journal stated that manufacturing wages in China too low for China’s urban garment workers, but correct for
average about 60 cents an hour,52 very close to the 57 cents noncity employees in garment manufacturing.
estimated here for total compensation. One newspaper wrote, In general, global media-published estimates of manu-
“A Chinese factory worker earns the equivalent of less than facturing earnings or compensation in China are in the ballpark
$1 per hour,” 53 a statement supported by the preceding of reasonable estimates.
analysis, and one that holds true even for urban manu-
facturing workers, who are better paid than their counterparts Labor compensation costs and China’s
outside the cities. competitiveness
Regarding particular manufacturing sectors, a newspaper
article said that, in China, employees of auto-parts suppliers have It is widely agreed57 that low earnings and low total labor
average wage costs of 90 cents an hour.54 Another author said compensation costs make manufacturing production in China
that employees of big global automakers in China “make the competitive in the international market. One of the leading

Monthly Labor Review August 2005 35

Manufacturing Compensation in China

Chart 2. Average real earnings of urban manufacturing staff and workers in China, 1990–2002

Yuan Yuan
14,000 14,000

12,000 12,000

10,000 10,000
Other urban ownership units, real earnings
8,000 8,000
Urban state-owned units, real earnings

6,000 Urban total manufacturing staff and workers, real earnings 6,000

4,000 Urban collective-owned units, real earnings 4,000

2,000 2,000

0 0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
SOURCE: Table 6.

reasons that some of China’s own domestic manufacturing middle school education and therefore are basically literate and
industries can sell their products at home and abroad, and numerate. Also, millions of young and middle-aged workers from
that multinational and other foreign companies are moving rural areas are eager to get out of the countryside and therefore
their manufacturing operations to China, is the low cost of willing to work hard in a disciplined manner for pay that is low by
employing manufacturing workers there. international standards, but higher than they can earn in agricul-
The low cost of labor makes China particularly competitive ture. China also has many millions of university-educated young
in a number of manufacturing industries, including labor- adults who are especially competitive because they are good in
intensive, assembly, and reprocessing industries; industries engineering and technical fields, are hard working and motivated,
with low value added; those with simple repetitive steps in and work for a fraction of the salaries received by equally capable
the manufacturing process; and food-processing industries. young adults in developed countries. China now produces at
As one source puts it, “China has become an essential link in least half of the world’s cameras and photocopiers and one-
quarter of the world’s television sets and washing machines.61
the global production chain for many labor-intensive prod-
Indeed, China “is the new workshop of the world, producing
ucts...a manufacturing hub for the rest of the world in low-end
two-thirds of all photocopiers, microwave ovens, DVD players,
labor-intensive goods.” 58 Labor productivity (output per
and shoes, over half of all digital cameras, and around two-fifths
employee) is low by world standards in these kinds of Chinese
of personal computers.”62
factories, and earnings are correspondingly low.59 In the 1990s Labor compensation in China’s manufacturing sector is
and beyond, China’s employees experienced widening earnings higher than it was a decade or two ago. This means that some
inequality, as earnings rose for city workers, but basically other developing countries are now able to compete with
stagnated for the least skilled and least educated workers.60 China purely on the basis of earnings per manufacturing
China is not particularly competitive in capital-intensive or worker. Real living standards have been rising in China’s cities,
materials-intensive industries. and real earnings have been rising for urban staff and workers
However, China is beginning to compete successfully in some in manufacturing, as shown in tables 5 and 6 and chart 2.63
kinds of moderately skills intensive kinds of manufacturing. Large Why are urban manufacturing earnings rising rapidly in
proportions of China’s young adults now have at least a lower China? Some scholars argue that because labor productivity is

36 Monthly Labor Review August 2005

rising rapidly in China’s city factories, we would expect city producing and exporting countries because the cost of textile
manufacturing earnings also to rise.64 Among the forces driving production is generally lower there than in China.75 Of course,
the increase in urban manufacturing earnings are a sustained China remains highly competitive globally because of its
rise in the returns to education and skill, as well as a wage relatively low labor costs and many other favorable factors,76
premium for Communist Party members and others remaining in but rising labor compensation in China has begun to erode
protected state-owned enterprises.65 Rigidities in urban labor the country’s manufacturing price advantage.
markets also have forced earnings upward and impeded com-
petition.66 Other experts contend that the huge supply of surplus THIS ARTICLE HAS COMBINED EMPLOYMENT AND EARNINGS
urban and rural workers ought to keep their earnings down: DATA for China’s urban manufacturing workers and for the
“The coincidence of rising mass unemployment and rapid noncity TVE manufacturing workers in order to derive approxi-
increases in real wages in the late 1990s appears contrary to the mations of annual, monthly, and hourly labor compensation for
predictions of competitive labour markets.”67 The range of urban, noncity, and all-China manufacturing employees. Reported
earnings in Chinese manufacturing has indeed widened, and earnings and labor compensation data have been adjusted
the least educated unskilled workers have experienced near
separately to yield urban data and TVE data. As of 2002, the
stagnation in their real earnings “under the twin pressures of
latest year for which adequate earnings data are available, average
heavy migration from China’s villages and [the] intense pur-
labor compensation for 30 million of China’s urban manufacturing
suit of cost advantage from overseas buyers of labor-intensive
employees was approximately U.S.$0.95 per hour, while the
goods.”68
reported 71 million manufacturing employees in TVE’s outside
In addition to the earnings bill, required payments for other
the cities averaged about U.S.$0.41 in labor compensation per
urban employee benefits have increased.69 China is trying to
hour of work. Combining the labor compensation of manu-
build a viable system of pensions, medical benefits, unemploy-
facturing workers in cities and in TVE’s to derive an all-China
ment benefits, workers’ compensation, and housing benefits, at
estimate results in average labor compensation of approx-
least for its city population, as discussed previously. One source
imately U.S.$0.57 per hour of work for 101 million manufac-
argues that required employer payments for these urban social
safety net programs in China are now higher than they need to turing workers in China.
be—for example, substantially higher than in Malaysia, South The following items should have high priority for future
Korea, Taiwan, and Singapore.70 In some cities, the mandated data collection in China and future research on hourly labor
payments are still rising rapidly. For example, compensation in China’s manufacturing sector:

Average labor costs in Shanghai rose by 15% last year due 1. Data on hours worked. For the important goal of
to increases in welfare payments, healthcare subsidies, and calculating average hourly labor compensation in
housing subsidies. On average local companies paid 10,849 manufacturing in China, a high priority is to get better
yuan in fixed and optional welfare fees, up 22.4% [from the data on actual hours worked by employees in the
year before]. This rise was significantly higher than in cities manufacturing sector. China’s government could itself
such as Kunshan, Nanjing, Hangzhou, Suzhou, or Ningbo.71 gather and publish more systematic data on this impor-
tant measure, and scholars should also emphasize
As earnings and mandated social insurance payments in- gathering information on it.
crease, urban China becomes less competitive in the global
context and even in the domestic Chinese context. Shanghai, 2. National economic census. During the year 2005, with
for example, is beginning to become too expensive for many reference year 2004, China conducted its first national
manufacturing concerns.72 Some businesses are moving from census of the economy. This undertaking is expected
the city to the poorer inland province of Anhui.73 Cities to refine, correct, and update data on labor compen-
throughout China are much more expensive for manufacturing sation received in manufacturing. When results of the
than even their nearby suburbs. Factories can save a third in economic census become available starting in late 2005,
power costs and half in wage bills just by relocating a factory the new information should be used to update the
half an hour’s drive outside of Guangdong’s capital city of estimates in this article.
Guangzhou.74 Indeed, many manufacturing companies are
now choosing to move their production operations from 3. Noncity manufacturing labor compensation. Much
developed countries or from China to other developing more data collection and analytical research are needed
countries with lower labor costs. For instance, India, Pakistan, to fill in some of the missing information on rural and
and Vietnam are becoming competitive as textile and apparel town manufacturing earnings and total compensation.

Monthly Labor Review August 2005 37

Manufacturing Compensation in China

4. Labor force surveys. China needs to design, carry and publish data on earnings and total compensation.
out, and publish results of labor force surveys using China reportedly will begin a regular labor force survey
international standards and definitions. Such surveys in 2006, the results of which will subsequently be
should cover the whole country and should collect published.

Notes

ACKNOWLEDGMENTS: This article was written under contract to the example, the purchasing power parities used may not accurately reflect
U.S. Department of Labor, Bureau of Labor Statistics, in order to further the actual purchasing patterns of manufacturing workers, and the price
the knowledge of China’s manufacturing earnings and labor compensation data used to construct the parities may not correctly approximate the
statistics. The views expressed are those of the author and do not reflect relative prices of many goods and services. For a discussion of the
the views of the Bureau. This research project has benefited from the purchasing power of Chinese manufacturing worker incomes, see Judith
valuable feedback of colleagues in China and in other countries on China’s Banister, “Manufacturing Employment and Compensation in China,”
economy and Chinese business practices. In particular, economists Loraine on the Internet at http://www.bls.gov/fls/#publications.
A. West and Nicholas R. Lardy served as expert discussants at a November 5
2004 BLS seminar on an early draft of the full report on the BLS Web site. The analysis presented herein applies to the mainland of the
Official statistical organizations in China have helped to correct some People’s Republic of China and excludes statistics for Hong Kong,
errors and point toward missing pieces of information. BLS economists— Macao, and Taiwan.
in particular, Constance Sorrentino, Chris Sparks, Elizabeth Taylor, Aaron 6
Banister, “Manufacturing employment in China.”
Cobet, Susan Fleck, Marie Claire Guillard, Gary Martin, Ann Neff, and
7
Erin Lett—have provided their expertise and support. Patricia Capdevielle, Banister, “Manufacturing Employment and Compensation in China.”
formerly of the Bureau of Labor Statistics, provided expert advice and 8
comments. I would especially like to thank Xing Yan (LeLe), Xing Shuo, “International Comparisons of Hourly Compensation Costs for
Song Jintao, Xing Shuqin, Wang Jianping, Li Fang, Xue Jianwen, and Production Workers in Manufacturing,” on the Internet at http://
Robert Boyer for their dedicated research assistance. The opinions, www.bls.gov/news.release/ichcc.toc.htm.
analysis, and conclusions expressed in this report are solely mine, and any 9
See Banister, “Manufacturing employment in China,” for further
mistakes or errors remain my responsibility. background information about China’s statistical system.
1
The companion piece to this article, “Manufacturing employment 10
Examples are available of statistical reporting forms and instructions
in China” (Monthly Labor Review, July 2005, pp. 11–29), noted that
issued to city enterprises to use to report employment and earnings data
China’s official statistics reported 83 million manufacturing employees
for the calendar year 2003. A “labor situation form” [Laodong qingkuang
at yearend 2002, but a variety of other available statistics strongly
indicated that the actual number was more than 100 million. biao] was to be submitted to authorities by the end of February 2004.
Wage-reporting instructions were in the publication Laodong gongzi;
2 tongji taizhang [Labor wages; statistical accounts] (Beijing, Beijing
Banister, “Manufacturing employment in China,” noted that
Municipality Statistical Bureau, 2004), especially p. 2-1.
China’s official statistics reported 38 million city manufacturing
employees at yearend 2002. Data on earnings are not available for 8.2 11
China National Bureau of Statistics, China Statistical Yearbook
million manufacturing workers in the cities; of these workers, 2.6 2003 (Beijing, China Statistics Press, 2003), pp. 66, 84, 87, 90.
million are self-employed. The Bureau of Labor Statistics does not
include the self-employed in its comparative estimates of hourly 12
China National Bureau of Statistics and China Ministry of Labor,
compensation costs, which relate only to production workers. China's compilers, China Labor Statistical Yearbook (Beijing, China Statistics
data cover both production and nonproduction workers. Press, published annually); China Ministry of Agriculture, TVE Yearbook
3 Editorial Committee, ed., China Village and Town Enterprise Yearbook
TVE’s originally were established as collective economic units run by
2003 [in Chinese] (Beijing, China Agriculture Publishing House, 2003),
local governments in rural areas and towns. The purpose of TVE’s was, and
pp. 130–31.
still is, to employ small farmers and rural laborers in industrial or service
occupations in locations not far from their family homes. This effort 13
China Labor Statistical Yearbook 2003 (Beijing, China National
allows China’s vast countryside to become modernized without neces- Bureau of Statistics and China Ministry of Labor, compilers; China
sitating massive migration from the villages to cities. In the 1980s, and Statistics Press, 2003), pp. 630, 638.
especially from the 1990s to today, TVE’s shifted from public toward
private ownership, and many foreign-funded enterprises became classified 14
Chinese sources did not report earnings data for another 8 million
as TVE’s. Nowadays, in addition to including small local enterprises, the urban manufacturing employees: self-employed individual manufacturing
TVE category can include very large factories in industrial parks outside workers and the investors and workers in relatively small private manu-
cities, as well as suburban, town, and rural factories. Companies have facturing concerns. It is not known whether this group of city manu-
incentives to have their factories classified as TVE’s because required social facturing employees earns more or less than the “manufacturing em-
insurance payments are low, statistical reporting requirements are minimal, ployees in urban units.” However, some of the employers of these 8
and the companies receive many legal and tax benefits. million workers pay lower social insurance payments or none at all to city
governments.
4
To more closely approximate the purchasing power of Chinese
15
manufacturing worker incomes in U.S. dollars, some type of purchasing China Village and Town Enterprise Yearbook 2003, pp. 130–31.
power parity (that is, the amount of yuan required to purchase the 16
Wage-reporting instructions, 2004, p. 2-4.
equivalent of $1 of goods and services in China) would be needed. Although
17
purchasing power parities provide a better measure of differences in relative See BLS Handbook of Methods (Bureau of Labor Statistics, 1997),
price levels than do commercial exchange rates, there are still important Chapter 12, “Foreign labor statistics,” pp. 114–15; and Chris Sparks,
limitations in using them to construct comparisons of worker income. For Theo Bikoi, and Lisa Moglia, “A perspective on U.S. and foreign

38 Monthly Labor Review August 2005

compensation costs in manufacturing,” Monthly Labor Review, June Inequality, Labor Market and Welfare Reform in China, Australian
2002, pp. 36–50, especially p. 49. National University, Canberra, Australia, August 2004, Table 1, pp.
18
28–29; on the Internet at http://econrsss.anu.edu.au/pdf/china-
See Banister, “Manufacturing employment in China.” abstract-pdf/Dongpaper.pdf.
19
Sparks, Bikoi, and Moglia, “U.S. and foreign compensation costs,” 31
Kim Woodard and Anita Qingli Wang, “Acquisitions in China: A
p. 49.
View of the Field,” China Business Review, November–December 2004,
20
Xiaochun Qiao, China’s Aging and Social Security of the Elderly: pp. 34–38, and “Acquisitions in China: Closing the Deal,” China
With Reference to Japan’s Experiences, Japan External Trade Business Review, January–February 2005, p. 35.
Organization, IDE-JETRO Visiting Research Fellow Monograph Series 32
See Fox and Zhao, “China’s labor market reform.”
No. 388 (Chiba, Japan, Institute of Developing Economies, 2004).
33
21 See Judith Banister, “Manufacturing employment in China,”
“Shehui baoxianfei zheng jiao zanxing tiaoli” (“Provisional
Monthly Labor Review, July 2005, pp. 11–29, for a further explanation
regulations for payment of social insurance fees”), in Laodong he
of the underreporting of manufacturing employment and its
shehui baoxian zhengce xuanchuan cailiao (Materials on social
consequences.
insurance policy announcements), Beijing, Haidian District Labor and
Social Security Office, regulation number 259, promulgated Jan. 22, 34
Richard Jackson and Neil Howe, The Graying of the Middle King-
1999. dom (Washington, DC, Center for Strategic and International Studies
22 and Prudential Foundation, 2004), p. 14.
Wage-reporting instructions, p. 2-5.
35
23
Handbook, pp. 114–15; Sparks, Bikoi, and Moglia, “U.S. and
BLS David Hale and Lyric Hughes Hale, “China takes off,” Foreign Affairs,
foreign compensation costs,” p. 37. November–December 2003, p. 46; Nicholas R. Lardy, “United States-
24 China ties: reassessing the economic relationship,” testimony presented
All data in this paragraph are from China Ministry of Labor,
before the House Committee on International Relations, U.S. House of
Zhongguo laodongli shichang gongzi zhidao jiawei (2003 nian)
Representatives, Oct. 21, 2003; on the Internet at http://www.iie.com/
[China Labor Force Market Wage Guide 2003] (Beijing, China Labor
publications/papers/lardy1003.htm; and Henny Sender, “Self-interest
Social Security Press, 2004), p. 379.
may lead China to revalue yuan,” Wall Street Journal, Apr. 19, 2004, p.
25
Ibid., p. 379. A2.
26
Loraine A. West, “Pension reform in China: Preparing for the fu- 36
John Knight and Linda Yueh, “Urban Insiders Versus Rural Outsiders:
ture,” Journal of Development Studies, February 1999, p. 165. In some Complementarity or Competition in China’s Urban Labour Market?”
cities, the social benefit payment that the enterprise is required to pay the paper presented at the International Research Conference on Poverty,
government is not strictly a percentage of whatever the total gross salary Inequality, Labour Market and Welfare Reform in China, Australian
bill is. For example, in Shanghai for 2003, enterprises had to pay 43.5 National University, Canberra, Australia, August 2004; on the Internet at
percent of the total wage bill, subject to the following constraints: if the http://econrsss.anu.edu.au/chinaconfabstracts.htm.
reported total wage bill divided by the reported number of employees
averaged less than 60 percent of Shanghai’s average monthly salary for 37
Jianchun Yang, “China Working Time Statistics,” on the Internet
the first half of 2003, the enterprise still had to pay 43.5 percent of that at http://www.insee.fr/en/nom_def_met/colloques/citygroup/pdf/
minimum salary threshold; the maximum payment the enterprise was China-general.pdf; China Labor Statistical Yearbook 2004, p. 111;
required to remit was 43.5 percent of the total wage bill that would personal communication with NBS officials.
represent 3 times the average 2003 Shanghai wage. (See Lulu Zhang,
“Shanghai region: Updates on Shanghai social benefit affecting FIE monthly 38
Yang, “China Working Time Statistics,” p. 1.
overheads,” China Briefing; The Practical Application of China Business,
39
June 2004, p. 10.) This procedure is supposed to be applied nationwide, See “Excessive Overtime in Chinese Supplier Factories: Causes,
based on State Council Document Number 6. See also Loraine A. West and Impacts, and Recommendations for Action,” Verité Research Paper,
Daniel Goodkind, Pension Management and Reform in China, NBR September 2004, on the Internet at http://www.verite.org/Excessive
Executive Insight Series No. 15 (Seattle, National Bureau of Asian Research, %20Overtime%20in%20Chinese%20Factories.pdf; and Leslie T.
1999), p. 3. Chang, “At 18, Min finds a path to success in migration wave,” Wall Street
27 Journal, Nov. 8, 2004, p. A1.
Data for Changshu City are from Qiye shenbao shehui baoxian
40
jiaofei yewu zhinan (Business guide to enterprises on social insurance See earlier in this article, pp. 27–29.
payments), Jan. 15, 2004; on the Internet at http://www.changshu. 41
Banister, “Manufacturing employment in China.”
gov.cn/H/content/HQA0000000000002837.htm. Data for Wuxi City
are from Shehui baoxianfei jiaofei bili mingxi biao (Table of detailed 42
China Village and Town Enterprise Yearbook 2003, p. 219.
comparisons of required social insurance payments), 2858 fuwuwang
43
(2858 service Internet site) at http://www.wx2858.com/XCBST/jyzn/ Employment weights are used to calculate an estimate of national
shehuibaoxian.asp. total labor compensation in manufacturing.
28
China Labor Statistical Yearbook 2003, pp. 471, 575–81. China had 44
Note again that the data for China refer to all employees, while
21.3 million TVE’s of all kinds in 2002, but only 85,000 of them had any the figures for the United States and other countries refer to production
rural old-age pension insurance. By yearend 2002, a cumulative total of workers. Employees have higher compensation than production work-
54.6 million people had ever contributed to any rural social pension ers, so the data for China are overstated to an unknown degree for
insurance scheme, but during 2002, only 4.1 million contributed to such these comparisons.
a system.
45
29
Louise Fox and Yaohui Zhao, “China’s labor market reform: West, “Pension reform in China,” p. 172.
Performance and prospects,” background paper for the China 2002 46
Thomas G. Rawski, personal communication, May 28, 2004.
Country Economic Memorandum (Washington, DC , World Bank,
47
2002); Xiaochun Qiao, China’s Aging and Social Security. Banister, “Manufacturing employment in China,” p. 23.
30
Xiao-yuan Dong, “The Changing Wage-Structures in the 1990s: 48
Shanghai municipality, for example, excludes from its employ-
A Comparison between Rural and Urban Enterprises in China,” paper ment statistics data on in-migrant workers from other provinces. (See
presented at the International Research Conference on Poverty, Banister, “Manufacturing employment in China.”)

Monthly Labor Review August 2005 39

Manufacturing Compensation in China

49 65
See Nicholas R. Lardy, “Do China’s Abusive Labor Practices See Fox and Zhao, “China’s labor market reform.”
Encourage Outsourcing and Drive Down American Wages?” testimony 66
Knight and Yueh, “Urban insiders versus rural outsiders.”
presented before the Senate Democratic Policy Committee Hearing,
67
Mar. 29, 2004; on the Internet at http://democrats.senate.gov/dpc/ Simon Appleton and Lina Song, “The evolution of wage structure in
hearings/hearing14/lardy.pdf. urban China during reform and retrenchment,” paper presented at the
50
International Research Conference on Poverty, Inequality, Labor Market
Rawski, personal communication, May 28, 2004; Fox and Zhao, and Welfare Reform in China, Australian National University, Canberra,
“China’s labor market reform,” pp. 3, 22. August 2004, p. 2; on the Internet at http://econrsss.anu.edu.au/
51
Banister, “Manufacturing employment in China,” table 1. chinaconfababstractshtm.
68
52
“Is the wakening giant a monster?” The Economist, Feb. 15, Thomas G. Rawski, “Recent developments in China’s labour
2003, pp. 63–65. economy,” revised November 2003 from a report prepared for the
International Labor Office in January 2002, p. 17; see also Fox and Zhao,
53
George Stalk and Dave Young, “How China gets our business,” “China’s labor market reform,” pp. 3, 22; and the entire Rawski article.
Washington Post, Mar. 7, 2004, p. B3.
69
54
Jianchun Yang, “2002 nian zhongguo jiuye qingkuang” [“China
Norihiko Shirouzu, “China drives auto-parts shift,” Asian Wall 2002 employment situation”], Zhongguo renkou tongji nianjian 2003
Street Journal, June 10, 2004, p. A5. [China Population Statistics Yearbook 2003] (Beijing, China Statistics
55
Joseph Szczesny, “China an exporter by 2007? Will too many Press, 2003).
cars force Chinese automakers to begin selling outside the Middle 70
Rawski, “Recent developments,” p. 27; see also Bureau of Labor
Kingdom?”; on the Internet at http://www.thecarconnection.com/
Statistics, “International comparisons of hourly compensation costs
index.asp?article=7233.
for production workers in manufacturing, 1975–2003”; on the Inter-
56 net at http://bls.gov/fls/home.htm.
“Is the wakening giant a monster?” p. 63.
71
57
For a few examples, see Stalk and Young, “How China gets our Paul French, “Welcome to bubble town,” Asian Wall Street Journal,
business”; Szczesny, “China an exporter by 2007?”; and Chinese Academy May 27, 2004, p. A7.
of Social Sciences, Industry Economic Research Institute, Zhongguo 72
gongye fazhan baogao [China’s Industrial Development Report] (Beijing, Iain McDaniels, “A critical eye on Shanghai: Will the city’s
Economic Management Press, 2001), pp. 109, 547. extraordinary growth continue?” China Business Review, January–
February 2004, pp. 8–9, especially p. 8.
58
Hale and Hale, “China takes off,” p. 46. 73
French, “Welcome to bubble town.”
59
Lardy, “China’s Abusive Labor Practices.”
74
“String of pearls: China’s development,” The Economist, Nov.
60
Fox and Zhao, “China’s labor market reform.” 20, 2004, p. 44.
61
Matt Forney, “Tug-of-war over trade: As China becomes the 75
Mu Xin and Zhenpeng Liang, “Mei caigou shang xuejian Zhongguo
world’s factory, U.S. and European manufacturers are hurting,” Time fangzhi dingdan” [“U.S. purchasers have cut textile orders from China”],
International (Europe Edition), Feb. 23, 2004, p. 34. Xin kuai bao [New Express], Apr. 28, 2004; on the Internet at http://
62
“The dragon and the eagle,” The Economist, Sept. 30, 2004. www.ycwb.com/gb/content/2004-04/28/content_683077.htm.
63
See also Lardy, “China’s Abusive Labor Practices.” 76
See Banister, “Manufacturing Employment and Compensation,”
64
Nicholas R. Lardy, discussant, BLS seminar, Washington, DC, Nov. for further information on China’s many competitive advantages in
8, 2004. manufacturing.

40 Monthly Labor Review August 2005