Are Chinese Workers Paid the Correct Wages? Measuring Wage Underpayment in the Chinese Industrial Sector,2005-2010

Abstract

This paper examines the labor compensations of the Chinese industrial sector for the period 2005-2010. We find that both the state owned enterprises and the non-domestic enterprises pay more than the living wage. But the domestic private enterprises pay substantially less than the living wage. We also find that all types of Chinese industrial enterprises pay the workers with wages that are substantially less than their marginal product of labor.

Keywords

living wage China marginal product of labor

1. Introduction

The Chinese economy has grown rapidly for about three decades, with an average annual growth rate of about 10 percent from 1980 to 2010. One of the key factors behind the economic surge is the healthy and educated but relatively cheap labor force in China. The low labor cost contributed to capital accumulation and economic growth, which in turn created job opportunities for the workers. Despite years of rapid economic growth, hundreds of millions of Chinese workers still work under sweatshop conditions, with little effective protection from the government and the official labor unions (Hart-Landsberg and Burkett 2005; Pun 2005; Weil 2008).¹ Although the average nominal wage for formal sector workers has increased by 14 percent per year from 2000 to 2010,² recent surveys find that about one-third of Chinese workers receive no more than official minimum wages, more than two-thirds of workers work overtime, and only 25 percent of them receive compensations for overtime work.³

Overall, China’s current growth regime, based on the exploitation of the cheap labor force, natural resources, and high investment, has been highly effective and successful over the past three decades. However, there has been growing recognition that the regime will no longer be sustainable (Palley 2006; Kuijs and Wang 2006; Piovani and Li 2011). In recent years, the Chinese economy relied heavily upon investment and exports to sustain expansion of aggregate demand (Zhu and Kotz 2011). Meanwhile, domestic consumption has grown less rapidly than the overall economy due in part to rising income inequality and falling labor income shares. China’s Gini coefficient has increased from around 0.3 in 1978 to 0.45 in 2006 (Sutherland and Yao 2011), and a recent estimate suggests that it is close to 0.5.⁴ Labor share in GDP decreased from 50 percent in 1990 to 37 percent in 2005 (Piovani and Li 2011).

Since the 2008-2009 global economic crisis, Western capitalist economies have suffered from persistent stagnation. In the coming years, the Chinese economy will have to undergo a major rebalancing, with the economy led by domestic consumption rather than investment and exports.⁵ Income redistribution from capitalists to workers may be a pre-condition for the expansion of domestic consumption (Zhu and Kotz 2011).

Given the changing circumstances, an economic growth regime based on the exploitation of cheap labor would not only prevent the expansion of domestic consumption, but also undermine the economy’s long-term growth potential.

In the neoclassical growth literature, economic growth depends on physical capital as well as human capital. Some use education as a proxy for human capital. Mankiw et al. (1992) estimates the human-capital-augmented Solow growth model and concludes that human capital contributes significantly to economic growth. Some consider the population’s health conditions as a component of human capital. Bloom et al. (2004) finds that a 1 percent increase in life expectancy may lead to a 4 percent increase in output. An economic regime based on low wages may undermine the accumulation of human capital in the long run.

In the Marxian literature, economic growth is driven by capital accumulation and technical change. Sustainable capital accumulation depends on normal reproduction of labor power that meets the basic needs of workers and their families. Moreover, the basic needs contain a historical and moral element that depends on the level of economic development (Marx 2004: 275).

The general perception is that Chinese workers are often paid wages that fail to meet their basic needs, or to fulfill the requirements of normal reproduction of labor power.⁶ Unless Chinese workers’ wages start to catch up with overall economic growth, normal reproduction of labor power may be undermined.

As Chinese society becomes urbanized and a growing proportion of the labor force becomes wage workers working in the modern industrial and services sectors, Chinese workers may demand not only higher wages but also a growing range of economic, political, and social rights. If China fails to accommodate the workers’ growing economic, political, and social demands, growing conflicts between workers and capitalists, as well as between the government and the general public, could lead to major social instabilities and potentially result in the demise of China’s existing model of development (Zhao and Huang 2009; Pei 2009; Li 2011).

Given these considerations, it is important to study China’s wage levels and examine policies that may help to raise Chinese workers’ wages to economically and socially appropriate levels. This paper examines the wage rates of the Chinese industrial sector for the period 2005-2010. The observed labor compensations are evaluated against the estimated living wages as well as the neoclassical marginal product of labor, as two alternative measures of “correct wages.” To the degree that Chinese workers’ labor compensations are below the “correct wages,” one may argue that Chinese workers are underpaid relative to socially or economically appropriate wages.

The rest of the paper is organized as follows. The next section briefly reviews the alternative wage theories from the neoclassical, Keynesian, and Marxist perspectives, followed by a discussion of the living wage concept and its relevance in China. The third section explains the measurement of labor compensations for different types of enterprises in the Chinese industrial sector. The following two sections compare the observed labor compensations with the estimated living wages and the estimated marginal product of labor in the Chinese industrial sector. The last section discusses policy implications and concludes the paper.

2. What Wages Are Correct: Alternative Theories

Different economic theories have different understandings of what wages are “correct,” or what wages are economically or socially appropriate.

According to Marx, in the long run, wages were determined by the value of labor power (the socially necessary labor time embodied in the workers’ means of subsistence) (Marx 2004: 274). The means of subsistence is the normal living standard at a given time and place, which is affected by various factors, including technological and social norms. But in the short and medium term, Marx emphasized the importance of the reserve army of labor: “the general movements of wages are exclusively regulated by the expansion and contraction of the industrial reserve army” (Marx 2004: 790). Marx’s theory of “reserve army of labor” and its impact on workers’ bargaining power has influenced modern heterodox economists (Kalecki 1943; Pollin 1998; Bowles 2004: 271; Glyn 2007: 3-9).

In neoclassical economics, workers are supposed to be paid by their marginal product of labor in a competitive market (for a textbook illustration, see Varian 1992: 26). If a worker receives less (or more) than his or her marginal product, the profit-maximizing capitalist will increase (or decrease) demand for labor so that the wage rate should converge towards the equilibrium level. Any discrepancy between the wage rate and the marginal product of labor implies labor market imperfection and social welfare loss.

Keynes criticized neoclassical economics by arguing that the level of employment was determined by effective demand rather than labor market equilibrium. However, Keynes maintained that the neoclassical theory of marginal productivity remained valid. Keynes assumed that workers and capitalists would negotiate over nominal wages and capitalists would set the prices so that the real wage would equal the marginal product of labor. But Keynes did not explain how the nominal wage would be determined (Keynes 1964: 4-22).

Since the early 1990s, the literature on “living wages” has received growing attention (Pollin et al. 2008: 14). A living wage is generally understood as the minimum income required for a worker to meet his or her basic needs for an extended period of time. In the United States, a living wage was sometimes defined as a full-time worker’s compensation that is enough to cover the basic living expenses of a three-person family (Pollin 2001). A living wage can be more broadly defined as the wage level that allows the worker to “participate in the civic life” of a nation (Glickman 1997: 66).

The concept of living wage has important economic, social, and moral implications. Living wage may be seen as the minimum wage required for long-term reproduction of labor power. If a country’s workers generally receive lower than living wages, workers’ nutrition, health, education, and skills may be undermined, lowering the country’s long-term economic growth potential. Low wages would also undermine workers’ purchasing power, depress mass consumption, and limit expansion of effective demand.

In social and moral terms, if workers receive less than living wages, the prevailing economic system may be seen as morally unjustified. Currently, the principle of living wages is endorsed by the United Nations’ Universal Declaration of Human Rights, constitutions of various countries, and the International Labor Organization (Anker 2011). Persistent failures to provide living wages may undermine a society’s legitimacy and lead to political instability.

This paper evaluates the Chinese industrial sector’s labor compensations by comparing the observed labor compensations with the estimated living wages.⁷ If observed labor compensations are lower than estimated living wages, it could suggest China’s current growth model is unsustainable given its failure to meet workers’ basic needs or provide normal reproduction of labor power.⁸

From the neoclassical perspective, observed low wages may be attributed to workers’ low productivity. Even if a worker receives lower than the living wage, a neoclassical economist could argue that so long as the worker is paid marginal product of labor and the worker works for the wage voluntarily, the net result of the labor market transaction remains welfare-enhancing. To take this argument into account, this paper also estimates the marginal product of labor for different types of Chinese industrial enterprises and compares it with observed labor compensations.

3. Measuring the Labor Compensations: Data Sources and Construction

All data used in this paper are from the China Statistical Yearbook published by National Bureau of Statistics of China (2006-2011).

The industrial sector (including mining, manufacturing, and utilities) is China’s leading economic sector, accounting for about 40 percent of China’s GDP and one-quarter of China’s total employment. The “Industry” section of the China Statistical Yearbook provides the main economic indicators of “industrial enterprises above designated sizes.” Currently, “industrial enterprises above designated sizes” refers to all industrial enterprises with annual sales greater than 5 million yuan.

There are no available statistics about the urban or rural employment share in industrial enterprises above designated sizes. But according to tables 4-6, 4-7, and 4-8 in the China Statistical Yearbook, 2011, there were about 86.8 million workers in the entire industrial sector, and about 25 percent of them (20.2 million workers) worked in rural areas. Considering that rural industrial enterprises are likely to be smaller, the rural share of the above designated size enterprises is likely to be lower than 25 percent. Without further information, this paper mainly considers urban industrial workers (accounting for at least 75 percent of total employment in industrial enterprises above designated sizes).

Within the industrial sector, there are several distinct forms of enterprises based on their ownership status: state-owned enterprises (including state-owned and state-holding enterprises, to be explained below), domestic private enterprises, non-domestic enterprises, collectively-owned enterprises, and enterprises of unspecified ownership types.

In 2010, state-owned and state-holding enterprises accounted for 27 percent of the gross output value of all industrial enterprises above designated sizes and 19 percent of employment. Domestic private enterprises accounted for 31 percent of gross output value and 35 percent of employment. Non-domestic enterprises (including enterprises funded by foreign investment and enterprises funded by investments from Hong Kong, Macau, and Taiwan) accounted for 27 percent of the gross output value and 28 percent of employment.

The above three types of enterprises accounted for 85 percent of the gross output value of all industrial enterprises above designated sizes and 82 percent of employment. In addition, enterprises of unspecified ownership types accounted for the remaining 15 percent of gross output value and 18 percent of employment. The rest of this paper focuses on the three dominant types of enterprises for measurements and estimations.

There are significant differences between the three types of enterprises in terms of labor compensation and labor relations. State-owned enterprises were mostly created before market-oriented reform that started in the 1980s. Before market-oriented reform, workers in state-owned enterprises had tenured jobs and enjoyed a package of welfare and benefits. In the 1990s, many state-owned enterprises were privatized and tens of millions of state sector workers were laid off. Nevertheless, workers in the remaining state-owned enterprises tend to be better paid than workers in the private sector, and state-owned enterprises typically conform to the official labor laws and regulations.⁹

By comparison, Chinese government officials often fail to enforce labor laws and regulations on private enterprises. Within the private sector, non-domestic firms generally pay higher wages than domestic private enterprises.¹⁰

3.1. State-owned enterprises

For the rest of this paper, the term “state-owned enterprises” include “state-owned” enterprises and “state-holding” enterprises. State-holding enterprises refer to those mixed-ownership enterprises where the state has a larger share of the equity capital than any other shareholder.¹¹

The China Statistical Yearbook (“Employment and Wages,” Table 4-15) provides average annual wages of “state-owned” enterprises in mining, manufacturing, and utilities. These are weighted by employment (“Employment and Wages,” Table 4-5) to derive the average annual wages of state-owned industrial enterprises.

The “Employment and Wages” section also provides average annual wages for “other economic units” in the industrial sector (“Employment and Wages,” Table 4-15), which mainly include share-holding companies (including both state-holding and non-state-holding enterprises) and non-domestic enterprises.¹² In this paper, we assume that workers in “state-holding” industrial enterprises are paid the same average annual wages as workers in “other economic units.”¹³

Average annual wages of state-owned industrial enterprises are calculated by weighting “state-owned” sector wages and “state-holding” sector wages by their employment.

3.2. Domestic private enterprises

The China Statistical Yearbook (“Employment and Wages,” Table 4-17) provides average annual wages of domestic private enterprises in mining, manufacturing, and utilities. These are weighted by employment (“Industry,” Table 14-10) to derive average annual wages of domestic private industrial enterprises.

3.3. Non-domestic enterprises

In the China Statistical Yearbook, non-domestic enterprises are included in “other economic units” and the China Statistical Yearbook does not provide average annual wages for the non-domestic enterprises in the industrial sector. However, it does provide economy-wide average annual wages of non-domestic enterprises which may be used to estimate industrial sector average wages.

In this paper, we estimate the average annual wages of non-domestic industrial enterprises as follows:

Average Annual Wage of Non-domestic Industrial Enterprises = Average Annual Wage of Other Economic Units in the Industrial Sector * (Average Annual Wage of Non-domestic Industrial Enterprises in the Total Urban Economy) / (Average Annual Wage of Other Economic Units in the Total Urban Economy).¹⁴

3.4. Labor compensation

In this paper, labor compensation is defined as the sum of wages and employers’ social insurance contributions. The “Health Care and Social Services” section of the China Statistical Yearbook provides data on social insurance contributions. Using the data, we calculate average annual labor compensations for industrial enterprises of different types.

Chinese employers are required by law to make various social insurance contributions for their employees. The legally required social insurance contribution to be paid by the employer is around 30 percent of an employee’s wage. However, the actual coverage ratio varies greatly across different ownership types.

In 2010, there were 302 million workers employed in the urban sector (excluding the self-employed). But only 194 million urban workers were covered by the basic social insurance scheme.¹⁵ “State-owned” economic units (including the government and “state-owned” enterprises) normally provide social insurance coverage for all of their employees. Excluding 65 million workers employed by “state-owned” economic units, the implied average coverage ratio for workers employed by non-state-owned economic units is 54.4 percent ((194-65) / (302-65) = 0.544).

In this paper, we assume that workers in “state-owned” industrial enterprises are fully covered by the basic social insurance schemes, and state-owned enterprises’ social insurance contributions are assumed to be 30 percent of workers’ wages.

For other industrial enterprises, including “state-holding” enterprises, domestic private enterprises, and non-domestic enterprises, we assume that employers’ contributions are 30 percent of workers’ wages multiplied by the average social insurance coverage ratio in the non-state-owned sector. For 2010, the effective social insurance contribution ratio (the ratio of social insurance contributions to wages) for non-state-owned industrial enterprises is estimated to be 30 percent * 0.544 = 16.3 percent.

Table 1 shows the average annual labor compensations of different types of industrial enterprises from 2005 to 2010.

Table 1.

Average Annual Labor Compensations of the Chinese Industrial Enterprises (Yuan).

Year	State	Domestic Private	Non-Domestic
2005	22,694	n.a.	22,498
2006	26,551	n.a.	25,219
2007	30,997	n.a.	28,403
2008	36,014	n.a.	32,375
2009	39,462	20,212	34,858
2010	45,229	23,478	39,639

Sources: National Bureau of Statistics of China (2006-2011). See the section on “Measuring the Labor Compensations” for further explanations.

4. Can the Chinese Workers “Live”?

In a study on internationally comparable living wages, Anker (2006) defined living wage as “hourly pay rate a full-time worker would need to earn to support a family of four at the poverty line (i.e. an acceptable minimum living standard) for the country.” Anker identified a model diet based on World Health Organization recommendations and national specifics, and used comparable food prices to estimate total food costs. Anker inferred non-food costs from the Engel curve. The paper then combined food and non-food costs to calculate the total costs of living and the implied poverty line.

According to Anker (2006), a Chinese worker’s living wage in 2000 was estimated to be 2.86 yuan per hour. Anker’s estimate was based on the assumption that a typical Asian full-time worker worked for 48 hours a week and 50 weeks a year. The hourly living wage rate translated into an annual living wage of 6,864 yuan for a four-person family for year 2000, which implied an income of 1,716 yuan per person.

The average per person annual income of the poorest 5 percent of China’s urban households was 2,351 yuan in 2000.¹⁶ Thus, Anker’s living wage was only about 73 percent of the average per person income of the lowest 5 percent of urban households. This is far below what one may expect from any reasonable definition of a living wage.

The following two factors might have contributed to Anker’s underestimation. First, Anker used food prices data from both urban and non-urban areas in China. As non-urban food prices were usually lower than urban food prices, Anker’s method would underestimate living costs for urban workers. Second, Anker assumed that the Engel coefficient (the share of spending on food in the total household budget) in China was 60 percent in 2000. But according to the China Statistical Yearbook, the actual Engel coefficient in China was 40 percent in 2000. Since Anker used the Engel coefficient to estimate non-food expenditures, non-food living costs were underestimated. If one takes Anker’s estimate of food costs but assumes an Engel coefficient of 40 percent, China’s annual living wage in 2000 would be 10,275 yuan.¹⁷

A more recent survey by the Fair Wear Foundation (2009) conducted interviews with workers in the garment industry in four Chinese cities, and found that legal minimum wages were less than 70 percent of the actual monthly expenses of workers. The study also found that in 2009, a worker’s monthly take-home pay needed to be between 2,000 and 2,500 yuan to cover the workers’ basic needs. This corresponds to an annual take-home pay of between 24,000 and 30,000 yuan, or an annual gross pay between 31,200 and 39,000 yuan if adjusted for the employer’s social insurance contribution (assuming that the social insurance contribution is 30 percent of take-home pay).¹⁸

In 2010, the average per person annual income of the poorest 20 percent of China’s urban households was 8,475 yuan (China Statistical Yearbook 2011: Table 10-7). Thus, an annual income of 33,901 yuan would allow a four-person household to live on the lowest living standards in Chinese cities in 2010.¹⁹ This paper assumes that the living wage equals four times the average per person annual income of the poorest 20 percent of urban households.²⁰

Similarly, we estimate the living wages from 2005 to 2010.²¹ Table 2 reports the estimated living wages from 2005 to 2010 as well as the calculated “living wage premiums.” The “living wage premium” is defined as the difference between the average annual labor compensation and the living wage as a ratio over the living wage.

Table 2.

Living Wages and the Living Wage Premiums.

		Living Wage Premiums
Year	Living Wage	State-Owned	Domestic Private	Non-Domestic
2005	17,160	32.3%	n.a.	31.1%
2006	19,635	35.2%	n.a.	28.4%
2007	23,193	33.6%	n.a.	22.5%
2008	26,241	37.2%	n.a.	23.4%
2009	29,815	32.4%	−32.2%	16.9%
2010	33,901	33.4%	−30.7%	16.9%

Sources: National Bureau of Statistics of China (2006-2011). See the section on “Can the Chinese Workers Live” for further explanations.

State-owned enterprises had consistently paid about one-third more than living wages and domestic private enterprises had paid one-third lower than living wages. In 2010, the total wage underpayment (relative to the living wage) by domestic private industrial enterprises amounted to 345 billion yuan, or 23 percent of the domestic private industrial enterprises’ total profits.

Non-domestic enterprises paid 20-30 percent higher than living wages before the recent global economic crisis. But in 2009 and 2010, for workers employed in non-domestic enterprises, the living wage premium fell to 17 percent.

5. Are the Chinese Workers Rewarded by Their Productivity?

According to neoclassical economics, workers are paid their marginal product of labor. From the neoclassical perspective, the fact that a worker receives less than the living wage may be explained by the argument that the worker’s marginal productivity is very low. Thus, the observed wage would remain economically efficient even though its moral foundation might be questioned. To take into account this argument, in this section we examine whether Chinese workers are paid by their marginal product of labor.²²

In the neoclassical Cobb-Douglas production function, below, Y is output, K is capital input, L is labor input, the labor elasticity of production is denoted by α.

Y = A L^{α} K^{β}

The marginal product of labor (∂Y/∂L) is the product of the average product of labor (Y/L) and the labor elasticity of production (α), since

\frac{\partial Y}{\partial L} = α A L^{α - 1} K^{β} = α * (\frac{Y}{L})

Thus, to calculate the marginal product of labor, we need to first estimate the labor elasticity of output (labor elasticity hereafter). Chow (1993) estimated the labor elasticity in the Chinese industrial sector using the Cobb-Douglas production function. The estimation covered the period from 1952 to 1980. The estimated labor elasticity ranged from 0.3 to 0.32. Jefferson et al. (1992) estimated that in the mid-1980s, labor elasticity for state- and collectively-owned industrial enterprises was between 0.13 and 0.16. Yang and Zhou (1999) used provincial data between 1987 and 1992 and found that the labor elasticity of state industrial enterprises was 0.38.

In a later study, Chow and Li (2002) extended Chow’s estimation to 1998 and found that labor elasticity had increased to 0.41. Using provincial data from 1978 to 2001, Zhang and Tan (2007) estimated the labor elasticity of the urban industrial sector to be 0.82-0.85. Jefferson et al. (2006: 19) noted in a footnote that their estimate of the labor elasticity of the Chinese industrial sector is 0.64, under the restriction of constant returns to scale.

In this section, we estimate the marginal product of labor of Chinese industrial enterprises of different ownership types and use the estimated results to calculate the marginal product of labor. We estimate a basic Cobb-Douglass production function using two-way fixed effects. The above literature review suggests that previous estimates of labor elasticity in the Chinese economy have produced a wide range of results, depending on their specific data coverage and methodology. One lesson might be that there is no single answer for “true” labor elasticity, especially for such a diverse economy like China. To partly address the concern on the wide range of estimates, we utilize two different types of panel datasets to provide a more balanced picture. One uses province-level data and the other uses sector-level data. ²³

\ln (Y_{i j}) = β_{0} + β_{1} * \ln (K_{i j}) + β_{2} * \ln (L_{i j}) + \sum_{0}^{29} D_{i} + \sum_{0}^{5} T_{i} + ε_{i j}

where Y is output, K is capital input, L is labor input, D is the dummy variable capturing provincial or sectoral differences, and T is the year dummy.

Output is measured by the real value added of industry (calculated as the value added of industry deflated by the industrial sector’s producer price index). Capital input is measured by the real net value of fixed assets (net value of fixed assets deflated by the fixed investment price index). Labor input is measured by the annual average level of employment.²⁴ All data are from the China Statistical Yearbook, various years.²⁵

Table 3 reports the regression results. All estimated labor elasticity coefficients are statistically significant. Using the provincial level data, the estimated labor elasticity ranges from 0.39 to 0.81. Using the sectoral level data, the estimated labor elasticity ranges from 0.53 to 0.95. The estimated labor elasticity of domestic private enterprises using sectoral level data is 0.95. This is unlikely to be realistic. Other results are broadly consistent with the labor elasticity estimates reported by more recent studies.

Table 3.

Estimating the Cobb-Douglas Production Function for the Chinese Industrial Sector, 2005-2010.

	Provincial Level			Sectoral Level
	State-Owned	Domestic Private	Non-Domestic	State-Owned	Domestic Private	Non-Domestic
Labor	0.392*	0.569*	0.808*	0.527*	0.946*	0.669*
	(0.093)	(0.104)	(0.105)	(0.059)	(0.045)	(0.122)
Capital	0.260*	0.503*	0.256*	0.454*	0.024	0.452*
	(0.073)	(0.070)	(0.059)	(0.044)	(0.028)	(0.092)
Constant	−0.254	−1.457*	−1.787*	−3.807*	−6.549*	−5.111*
	(0.336)	(0.312)	(0.249)	(0.562)	(0.506)	(1.077)
Year Dummies	Yes	Yes	Yes	Yes	Yes	Yes
P/S Dummies	Yes	Yes	Yes	Yes	Yes	Yes
Observations	180	180	181	233	234	232
R-Squared	0.882	0.980	0.961	0.939	0.964	0.898

Notes: Numbers in parentheses are standard errors; the coefficients of provincial/sectoral (P/S) dummies and year dummies are not reported.

Statistically significant at 1% level.

The marginal product of labor can be calculated as the product of average labor productivity (output divided by labor input) and the estimated labor elasticity. The calculated marginal product of labor can be compared with average annual labor compensations. Table 4 reports the “wage-productivity gaps” for industrial enterprises of different ownership types.

Table 4.

The Wage-Productivity Gaps.

	Provincial Level Dataset			Sectoral Level Dataset
	State-Owned	Domestic Private	Non-Domestic	State-Owned	Domestic Private	Non-Domestic
2005	−47.9%	n.a.	−61.5%	−37.9%	n.a.	−49.9%
2006	−51.0%	n.a.	−60.7%	−43.5%	n.a.	−48.9%
2007	−55.0%	n.a.	−60.6%	−50.5%	n.a.	−48.6%
2008	−55.0%	n.a.	−56.6%	−50.2%	n.a.	−42.9%
2009	−48.7%	−62.8%	−55.6%	−39.4%	−80.5%	−41.0%
2010	−52.4%	−63.6%	−56.1%	−46.2%	−81.0%	−41.6%

Source: Based on results from Table 1 and Table 3. See the section “Are the Chinese Workers Rewarded by Their Productivity” for further explanations.

The “wage-productivity gap” is defined as the difference between the average annual labor compensation and the marginal product of labor as a ratio over the marginal product of labor. State-owned enterprises paid workers about half of their marginal product of labor. Domestic private enterprises paid workers between one-fifth and one-third of their marginal product labor. Non-domestic enterprises paid workers between two-fifths and three-fifths of their marginal product of labor.

These results suggest that all types of Chinese industrial enterprises pay workers with substantially less than the marginal product of labor. To the extent that neoclassical theory expects that marginal product of labor is a useful measure of worker productivity, what the results show is that the very low wage (lower than the estimated living wage) received by workers in private enterprises cannot be explained by low marginal productivity of labor.

6. Conclusion and Policy Implications

In this paper, we examine the question of whether Chinese workers are paid the “correct wages” by comparing average annual labor compensations in the Chinese industrial sector with the estimated living wages and the estimated marginal product of labor.

We find that both state-owned enterprises and non-domestic enterprises pay workers more than living wages. But domestic private enterprises pay workers substantially less than living wages. Moreover, workers’ low wages cannot be explained by their low productivity. We find that all types of Chinese industrial enterprises pay the workers with wages that are substantially less than the marginal product of labor.

From the Marxian perspective, the finding that workers receive lower wages than marginal product of labor may simply suggest that the neoclassical concept of marginal product is theoretically meaningless and empirically invalid. From the Keynesian perspective, the finding should suggest that the observed employment level falls short of the practically attainable full employment, implying insufficient effective demand. From the neoclassical standpoint, the finding may suggest that there are certain labor market imperfections, implying social welfare losses.

This paper’s findings suggest that despite the relatively rapid wage growth in recent years, a significant portion of Chinese workers are substantially underpaid based on both economic and social-moral criteria. These findings provide strong arguments for the Chinese government to adopt policies that will help Chinese workers (especially in the private sector) to substantially increase their wages.

One policy the Chinese government may consider is to undertake a large-scale public employment program. The official urban registered unemployed was near 10 million in 2010. But there were more than 200 million migrant workers who mainly worked in the informal sector with insecure jobs and very low wages.²⁶ The very large pool of migrant workers lowers the overall bargaining power of the Chinese working class.

Suppose the Chinese government commits an annual spending of 1.2 trillion yuan (equaling about 200 billion dollars) to a program of public employment. China’s current capital-output ratio is about 2.5 to 1 (see the calculations made by Piovani and Li 2011). Assuming a capital-wages ratio of 2.5 to 1, about 340 billion yuan may be used directly for wages and 860 billion yuan may be used for capital expenditures.

If the workers are paid a living wage of 34,000 yuan a year, the program can directly create approximately 10 million jobs which could potentially fully absorb the 10 million officially unemployed urban workers. In addition to directly creating 10 million jobs with living wages, 860 billion yuan of capital expenditures can indirectly create tens of millions of jobs. The public employment program can help to increase workers’ bargaining power by dramatically reducing the pool of unemployed and underemployed workers.

A spending program of 1.2 trillion yuan would cost about 2 percent of China’s GDP and is within the Chinese government’s fiscal capacity. By comparison, in response to the recent global economic crisis, the Chinese government undertook a four-trillion-yuan program of fiscal and financial stimulus. The workers hired by the public employment program may be productively employed to build social housing for low-income urban residents or undertake environmental cleaning projects.

Currently, the Chinese government spends less than 8 percent of its total expenditure on education, health care, and social welfare programs.²⁷ The Chinese government should increase social spending (in health care, education, and pensions) that helps to improve not only workers’ living standards but also China’s long-term economic growth potential.

Revenues from individual income taxes account for only 5.8 percent of China’s total fiscal revenue.²⁸ Moreover, state-owned enterprises pay disproportionately more taxes than private enterprises. In 2010, state-owned industrial enterprises paid 49 percent of the total taxes paid by all industrial enterprises (by comparison, state-owned enterprises accounted for 27 percent of output value and 19 percent of employment).²⁹ Domestic private enterprises paid 23 percent of the taxes paid by industrial enterprises and non-domestic enterprises paid 20 percent.³⁰

To finance spending on public employment and social programs, the Chinese government may consider raising taxes on wealthy individuals and private enterprises. These policies will not only benefit Chinese workers, but also contribute to a more sustainable pattern of economic growth and a more stable society.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

1

The miserable conditions of the Chinese working class have been widely documented even in the mainstream media; see a recent report in the New York Times on the brutal nature of Chinese capitalism: “In China, Human Costs Are Built Into an iPad,” .

2

See China Statistical Yearbook 2011, Table 4-12 for the growth in average urban wages.

3

These numbers are reported in official Chinese newspapers and websites; see Li Bingsheng’s report, ; and Li Bingsheng’s article in People’s Daily, 2010/05/19, “Wanshan Zuidi Gongzi Biaozhun Tiaozheng Jizhi (Improving the Adjustment Mechanisms of Minimum Wages.”

4

See “China’s Hidden Wealth Feeds an Income Gap,” , accessed on July 17, 2012.

5

This already becomes a consensus among the Chinese policy makers; see “Domestic Consumption Takes Driver’s Seat,” China Daily, 03/22/2012, http://www.chinadaily.com.cn/bizchina/2012-03/22/content_14886188.htm. The World Bank report China 2030 argues that inequality and insufficient consumption may undermine the sustainability of the Chinese model ().

6

The observations that the Chinese workers’ wages are insufficient to meet their basic needs have been made by the media, non-governmental organizations, as well as the Chinese official labor union. See the reports published by the China Labor Watch, http://www.chinalaborwatch.org/reports.html; the New York Times op-ed, “Can China Escape the Low-Wage Trap?,” http://www.nytimes.com/2012/05/27/opinion/sunday/can-china-escape-the-low-wage-trap.html; and the report by the All-China Federation of Trade Unions, “On Problems Faced by the Chinese Working Class,” http://yxlw.acftu.org/template/6/file.jsp?cid=16&aid=264.

7

The legal minimum wage is supposed to provide workers a living wage, but it is rather arbitrarily set and often significantly less than basic living needs ().

8

It might be argued that China has already achieved sustained growth despite relatively low wages in the past decades. However, as human capital becomes more and more important, future growth might increasingly rely on improved education and health care and normal reproduction of labor power.

9

In recent years, the mainstream media has been continuously attacking state-owned enterprises, even arguing that the higher wages in the public sector has caused serious inequality in China. As we are writing this, more than 100 workers in a private slaughterhouse died in a fire because the boss locked the gate to prevent them from “shirking”; see report in the Guardian, , accessed on June 8^th, 2013.

10

This does not suggest that non-domestic enterprises necessarily treat workers better than domestic enterprises. The famous Foxconn serial worker suicides in recent years provide a good illustration of the problems; see , accessed on June 8^th, 2013.

11

See China Statistical Yearbook, 2011, “Explanatory Notes on Main Statistical Indicators” in the Industry section.

12

Other economic units include share-holding companies, corporations with limited responsibilities, enterprises with non-domestic investments (that is, enterprises with Hong Kong, Macau, Taiwan, and foreign investments), and various economic units of mixed ownership. In 2010, share-holding companies and corporations with limited responsibilities accounted for 61 percent of total employment in “other economic units,” non-domestic enterprises accounted for 31 percent, and other enterprises with mixed ownership accounted for 8 percent.

13

The China Statistical Yearbook does not provide average annual wages of state-holding industrial enterprises. However, the “Employment and Wages” section (Table 4-17) of the China Statistical Yearbook provides economy-wide average annual wages for various types of enterprises. In 2010, the economy-wide average annual wage was 35,801 yuan for all workers in “other economic units” and 35,986 yuan for workers in share-holding and limited liability corporations (the two categories that include state-holding enterprises). This suggests that the average annual wages of state-holding enterprises are likely to be similar to that of “other economic units.”

14

In the years 2005, 2006, 2007, 2008, 2009, and 2010, the ratio between the economy-wide average annual wage of non-domestic enterprises and the average annual wage of “other economic units” was 1.146, 1.122, 1.095, 1.080, 1.061, and 1.051, respectively.

15

China Statistical Yearbook, 2011, Table 21-39.

16

China Statistical Yearbook (2001: Table 10-5).

17

The calculation is as follows: 6,864*60 percent/40 percent=10,275.

18

One possible reason of the difference between Anker’s result and the Fair Wear Foundation result is the data source. The Fair Wear Foundation study was based on interviews and actual household budgets in four major cities in China; therefore the results include more geographical specificities while was calculating a national average value.

19

The four-person family assumption is based on our understanding of the demographic characteristics of China. According to the China Statistical Yearbook, 2011, the typical urban core family consists of 2.88 persons, and the poorest 20 percent normally have 3.2 to 3.3 members. At the same time, it is common for a core family to support at least one elderly person in the extended family, which is partly due to underdevelopment of the social security network.

20

This threshold is not entirely arbitrary. In the official statistics, the lowest 20 percent is often referred to as “low income group,” while the “middle income group” refers to the 20 to 80 percentile in the distribution. Ideally, the living wage should provide at least a “middle income”; we choose “low income group” as a starting point to investigate the living wage issue, but this does not imply “low income” should be the goal for the living wage movement.

21

Our estimated living wage is in about the same range as previous studies. The adjusted Anker’s living wage is 10,275 yuan in 2000, and that figure suggests a living wage of about 38,000 yuan (10275*1.14¹⁰ = 38,092) in 2010 given an average annual growth rate of nominal wages of 14 percent during the period (see footnote 2). The Fair Wear Foundation estimates for 2009 ranged between 31,200 and 39,000 yuan, which are also close to our estimated living wage.

22

The limitations of neoclassical marginal productivity theory are well known to heterodox economists. It is well known that neoclassical economists fail to provide a meaningful and measurable concept of capital. The failure undermines the entire marginal productivity theory. Nevertheless, most economic institutions continue to teach neoclassical marginal productivity theory and most economists apply the marginal productivity theory uncritically. In this section, we try to provide empirical evidence that neoclassical marginal productivity theory cannot explain Chinese workers’ wages.

23

The sectoral level dataset uses data from 39 sub-sectors within the industrial sector. For the province-level dataset, Tibet is excluded due to lack of data.

24

Ideally, labor input is measured by labor effort or actual work hours. But due to data availability, the existing literature mostly uses employment level as a measure of labor input.

25

For 2008, 2009, and 2010, the China Statistical Yearbook did not publish data for value added of industry. We assume that for these years, the ratio between value added of industry and the gross output value of industry was the same as the average ratio from 2005 to 2007.

26

The unemployment level is reported in the China Statistical Yearbook 2011: Table 4-1; we calculate the size of the informal economy in 2010 using the method in .

27

See China Statistical Yearbook 2011: Table 8-6.

28

See China Statistical Yearbook 2011: tables 8-1 and 8-2.

29

See China Statistical Yearbook 2011: tables 14-2 and 14-6.

30

See China Statistical Yearbook 2011:Table 14-2, Table 14-10, and Table 14-14.

Author Biographies

Zhun Xu is assistant professor of economics at Renmin University of China. His recent research focuses on social change in China, political economy of agrarian change, and alternatives to the neoliberal economic order.

Ying Chen is a PhD candidate in Economics at the University of Massachusetts Amherst. She is currently completing her dissertation titled “Three Essays on Sustainable Development in China: Social, Economic, and Environmental Aspects” under the supervision of Professor David Kotz and Professor Robert Pollin.

Minqi Li teaches economics at the University of Utah. He is the author of The Rise of China and the Demise of the Capitalist World Economy and Peak Oil, Climate Change, and the Limits to China’s Economic Growth.

References

Anker

2006. Living wages around the world: A new methodology and internationally comparable estimates. International Labour Review 145: 309-338.

Anker

2011. Estimating a living wage: A methodological review. Conditions of Work and Employment Series 29: ILO.

Bloom

D. E.

Canning

Sevilla

2004. The effect of health on economic growth: A production function approach. World Development 32: 1-13.

Bowles

2004. Microeconomics: Behavior, institutions, and evolution. Princeton University Press.

Chow

G. C.

1993. Capital formation and economic growth in China. The Quarterly Journal of Economics 108: 809-842.

Chow

G. C.

Kui-Wai

. 2002. China’s economic growth: 1952-2010. Economic Development and Cultural Change 51: 247-56.

Fair Wear Foundation. 2009. Study on wages in China. http://www.fairwear.org/ul/cms/fck-uploaded/archive/2010-08/fwf_-_study_on_wages_in_china_en_-_2010.pdf.

Glickman

L. B.

1999. A living wage: American workers and the making of consumer society. Cornell University Press.

Goodwin

R. M.

1967. A growth cycle. In Socialism, capitalism and economic growth, ed. Feinstein

C. H.

Cambridge: Cambridge University Press.

10.

Hart-Landsberg

Burkett

2005. China and socialism: Market reforms and class struggle. Monthly Review Press.

11.

Huang

P. C. C.

2009. China’s neglected informal economy: Reality and theory. Modern China 35: 405-438.

12.

Jefferson

G. H.

Hu Albert

G. Z.

Jian

. 2006. The sources and sustainability of China’s economic growth. Brookings Papers on Economic Activity 37: 1-60.

13.

Jefferson

G. H.

Rawski

T. G.

Yuxin

Zheng

. 1992. Growth, efficiency, and convergence in China’s state and collective industry. Economic Development and Cultural Change 40: 239-66.

14.

Kalecki

1943. Political aspects of full employment. The Political Quarterly 14: 322-330.

15.

Kuijs

Wang

Tao

. 2006. China’s pattern of growth: Moving to sustainability and reducing inequality. China & World Economy 14: 1-14.

16.

Minqi

. 2011. The rise of the working class and the future of the Chinese Revolution. Monthly Review 63.

17.

Mankiw

Gregory

D. R.

Weil

D. N.

1992. A contribution to the empirics of economic growth. The Quarterly Journal of Economics 107: 407-437.

18.

Marx

2004. Capital: A critique of political economy. Penguin Books Limited.

19.

National Bureau of Statistics of China. 2006-2011. China statistical yearbook. http://www.stats.gov.cn.

20.

Palley

T. I.

2006. External contradictions of the Chinese development model: Export-led growth and the dangers of global economic contraction. Journal of Contemporary China 15: 69-88.

21.

Pei

Minxin

. 2009. Will the Chinese Communist Party survive the crisis? Foreign Affairs. http://www.foreignaffairs.com/articles/64862/minxin-pei/will-the-chinese-communist-party-survive-the-crisis.

22.

Piovani

Minqi

. 2011. One hundred million jobs for the Chinese workers! Why China’s current model of development is unsustainable and how a progressive economic program can help the Chinese workers, the Chinese economy, and China’s environment. Review of Radical Political Economics 43: 77-94.

23.

Pollin

1998. The “reserve army of labor” and the “natural rate of unemployment”: Can Marx, Kalecki, Friedman, and Wall Street all be wrong? Review of Radical Political Economics 30: 1-13.

24.

Pollin

2001. Time for a living wage. Challenge 44: 6-18.

25.

Pollin

Brenner

Luce

Wicks-Lim

2008. A measure of fairness: The economics of living wages and minimum wages in the United States. Cornell University Press.

26.

Pun

Ngai

. 2005. Made in China: Women factory workers in a global workplace. Duke University Press Books.

27.

Sutherland

Yao

Shujie

. 2011. Income inequality in China over 30 years of reforms. Cambridge Journal of Regions, Economy and Society 4: 91-105.

28.

Varian

H. R.

1992. Microeconomic analysis. W. W. Norton.

29.

Weil

2008. City of youth: Shenzhen, China. Monthly Review 60.

30.

The World Bank and Development Research Center of the State Council. 2012. China 2030: Building a modern, harmonious, and creative high-income society. World Bank.

31.

Yang

D. T.

Hao

Zhou

. 1997. Rural-urban disparity and sectoral labor allocation in China. Duke University, Department of Economics.

32.

Zhao

Litao

Huang

Yanjie

. 2009. Unemployment and social instability in China: Will they run out of control? East Asian Institute Background Brief No. 439.

33.

Zhang

Xiaobo

Tan

Kong-Yam

. 2007. Incremental reform and distortions in China’s product and factor markets. The World Bank Economic Review 21: 279-299.

34.

Zhu

Andong

Kotz

D. M.

2011. The dependence of China’s economic growth on exports and investment. Review of Radical Political Economics 43: 9-32.