Digital Labour in Chinese Internet Industries

Bingqing Xia

Institute of Communications Studies, University of Leeds, UK, csbx@leeds.ac.uk

Abstract: Digital labour has been the subject of considerable research in recent years (Van Dijck 2009, Manzerolle 2010, Dyer-Witheford 2010). But relatively little research has considered professional workers in digital media. This research addresses this gap by focusing on professional workers in the Chinese Internet industries. This paper asks: How are these digital labourers involved in the digital media production? To what extent should we criticise this involvement?

Based on detailed empirical research in China, I argue that the rapid growth of the Internet industries depends on exploiting these Internet workers, such as the workers in Chinese Internet industries—the new ‘sweatshop’ of the digital era. Chinese Internet workers have been subsumed in the global capitalist system as the new ‘sweatshop workers’.

This paper shows that Chinese Internet workers suffer very poor working conditions, and argues that these working conditions are the result of exploitation, a concept explored via using Eric Olin Wright’s schema. This paper also argues that most of the Chinese Internet workers are in the lower middle-class class position, in which they are exploited by the upper classes. Their working conditions have seriously deteriorated and they are victims of inequality and injustice—although they also are able to exercise agency and resistance. This paper therefore calls urgent attention to the working conditions of these digital labourers.

Keywords: Quality of Working Life, Exploitation, Chinese Internet Workers, Working Conditions, Digital Labour

Acknowledgement: I express sincere thanks to my PhD supervisors, Prof. David Hesmondhalgh and Dr. Helen Kennedy, for their continuous encouragement, advice and critique, which have greatly contributed to the progress of this paper.

1. Introduction: The Existing Discussion of Digital Labour

Digital labour has been discussed from various perspectives in recent years. For example, some theorists regard Internet users as a form of digital labour and highlight productivity and creativity of these users (Barbrook 2005, Hills 2002, Jenkins 2008, Ornebring 2008 and Wang 2008). Terranova (2004) defines active users in the “digital economy” as “free labour”, who build a community without great financial rewards and in return, obtain “the pleasures of communication and exchange” (91). The people who carry out “free work” are involved in work such as “building websites, modifying software packages, reading and participating in mailing lists and building virtual spaces” (74). Terranova regards these free labourers as a new productive force of capitalist production, as well as believing that capitalism is increasingly relying on free labour with its emergence.

Barbrook (2005) highlights the contribution of Internet users to the digital economy by analysing the paradoxical relationships between the new form of digital labour and capitalist production. He divides digital economy into three parts: The public element, the commercial sector, and the gift economy. To Barbrook, the gift economy is a free economy based on Internet users’ free exchange of information—“anarcho-communist” (2) participation in his term. This gift economy is an alternative to existing capitalism, as it tends to build ‘anarcho-communism’ via mutual collaboration with the commercial sector: Free exchanges of information between users rely on the “capitalist production of computers, software and telecommunications” (5), and capitalist production depends on the “increasing numbers of people participating within the hi-tech gift economy” (5). In other words, Barbrook believes that the gift economy, based on Internet users’ activities, would develop via its collaboration with the commercial sector, and ultimately becomes an alternative to existing capitalism. 

Terranova (2004) criticises Barbrook for his optimistic emphasis on the autonomy of the gift economy from capitalism. Nevertheless, Barbrook usefully criticises the commodification of hi-tech gifts and anticipates what people today called the “capitalism of communism”, where the “communism of capital” (the elements that stem from information production that go beyond capitalism) are reversed and the economy of free gifts becomes a new capital accumulation strategy. As an alternative, Terranova (2004) emphasises the absorption of free labour into capitalist production. Both Terranova and Barbrook highlight the productivity of Internet users, via focusing on their online activities and participation, although they evaluate such activities in very different ways. Their discussions highlight dynamics between Internet users and capitalist production in the new media era. But these theorists do not conceive Internet work in the wider context of capitalist-labour production. They both fail to capture the ambivalent, complex, and dynamic relations amongst labour and capitalism, though the question of ‘unpaid labour’ is an important issue.

By contrast, some theorists pay attention to the question of professional labour in the digital era, by focusing on professional workers in new media industries, such as web designers and Internet workers (Kennedy 2012, Gill 2002). For example, Kennedy (2012) discusses ethics and values in web designers’ working experiences via examining work practices and working conditions of web designers in the UK. She highlights that web designers, who are ethically motivated, make efforts to include web users with disabilities, in order to develop the accessible and perceived web for all people, especially people with disabilities. For instance, she argues that self-regulation in web designers’ working experiences provides a different model from self-regulation in other cultural workers’ experiences: Self-regulation in some cultural workers’ experiences is problematic, because it results in individualisation; by contrast, self-regulation in web designers’ experiences is ethically motivated, because it is driven by “a commitment to the founding ideals of the web as open, interoperable and accessible” (20).

Gill (2002) investigates poor working conditions in new media industries by highlighting certain issues, such as pervasive insecurity, low pay, and long working hours. She particularly links these poor working conditions to gender inequalities in new media work, by arguing that female workers in new media industries experience inequalities in terms of education, access to the work, autonomy, flexibility, and pay.

Both theorists’ work suggests a need to focus on professional labourers in new media industries, such as examining how these digital labourers are working. Based on such “turn to labour” tendency in the research of digital labour, this paper pays particular attention to professional digital labourers in the Chinese context: Professional workers in the Chinese Internet industries.

According to Noon and Blyton (2002, 5), there are several criteria to classify people’s work, such as the way jobs are undertaken, the main purpose of the work, job status, temporal pattern, and work location. I classify the Internet workers by the criterion of temporal pattern, such as full-time or part-time and permanent or temporary. Therefore, my PhD project classifies Chinese Internet workers into full-time workers, interns, and agency workers.

However, in this paper, Internet workers particularly refer to the full-time workers, who are paid regularly and hold formal positions in Internet companies. These workers are the main contributors to the creation of cultural products. Some of these workers possess high skill, while some have low skill, though none of them only sell manual labour as manual workers did during industrialist capitalism. Thus, some full-time workers here are understood as technical workers, conducting technical-related work, such as programming, web design and APP (Application Software) developing, etc. Other full-time workers are non-technical workers, who are mainly involved in administrative and routine work, such as HR (Human Resources) recruiting and training employees; marketing workers doing promotional work; and PR (Public Relations) personnel maintaining relationships with government officials, etc.

By clarifying who are these Internet workers, this paper asks: How are the Chinese Internet labourers involved in the digital media production? To what extent should we criticise this involvement?

The first step to answer these two questions is to examine how Chinese Internet industries have developed in recent years, and to what extent it links to the global economic system. Therefore, in section two, I discuss how Chinese ICT industry and Internet industries have developed in recent years, and how they contribute to Chinese economy. In section three, I explain the methodology of this research. Section four presents the empirical data I collected from fieldwork, in which I discuss certain issues that answer how Internet workers are involved in the digital media production, such as working hours, pay, lay-off, and pension problems. Sections five and six are the theoretical parts that answer to what extent we need to criticise such involvement. Section seven, the conclusion, I indicate a sort of worker agency as a hope in Chinese Internet industries. Below, I discuss the development of Chinese ICT industry, Internet industries, and Chinese economy.

2. A New Economic Growth Field: Chinese Internet Industries

2.1. The Chinese Economy

David Harvey (2005) defines neoliberalism as “a theory of political economic practices that proposes that human well-being can best be advanced by liberating individual entrepreneurial freedoms and skills within an institutional framework characterized by strong private property rights, free markets, and free trade” (2). This concept had a strong impact to Chinese society in the 1980s with the introduction of market principles, which had close relationships with neoliberalism. Neoliberals appreciate the reform and opening up policy because they believe it emancipates the Chinese market and develops the Chinese economy. For example, Huang (2008) highly appreciates the move towards a more market-oriented economy in 1980s Chinese society, as it improved social welfare. Mok and Lo (2007) point out that, “the policy of decentralization and marketization being adopted to reform the social policy domain has significantly reduced the state provision and financing in social service and social provision” (2).   

Admittedly, the Chinese economy has seen a rapid growth since the issuing of the reform and opening-up policy in 1978. Foreign investors have rushed into the Chinese market since China’s accession to the WTO (World Trade Organisation) in 2001. However, social problems and tensions are generated from the ongoing economic reform: Inequality between the rich and poor, and injustice between the bureaucratic capitalists and workers have expanded. For example, Zhao (2003) discovers inequalities within the Chinese ICT industry by discussing different access to media between Chinese urban middle-class and rural peasants: “While the rising business and urban middle classes are increasingly using the media to articulate their interests and shape state policies toward their preferred ends, the rally cries of tens of thousands of Chinese workers and farmers in their struggles for economic and social justice, for example, have simply fallen on deaf ears in the Chinese media system” (63). Zhao states that the rising business and urban middle classes increasingly enjoy better quality of life, such as gaining more prestige, better education, and better health care; but workers and farmers, by contrast, are still struggling at the bottom of the social hierarchy. Moreover, Zhao (2007) explains the uneven regional development by quoting Hu, Zhou and Li’s work (2001), which depicts China as “one country, four worlds” (102) because of the fragmentation and polarization of “class, region, gender, ethnicity and other cleavages” (101). Here, Zhao highlights inequalities between Chinese people due to the ongoing economic reform.

Some theorists explore the Chinese economy by highlighting the role played by the state. For example, Wang Hui (2003) states that neoliberalism is problematic in the context of China because it denies “the close relationship between the market and the political process” (100). He claims that the state plays a significant role in the Chinese economic system, since the political system in China is highly centralised. As Wang Jing (2008) suggests, the Chinese market is still controlled by the “party-state”. Wang Hui (2003) then enriches the concept of neoliberalism with Chinese characteristics: It is “a combination of notions of market extremism, neo-conservatism, and neo-authoritarianism” (81). Here, neoliberalism is understood to accelerate the process of delegating economic and political power from the central government to regional governments in a stable manner, to build an authority to guarantee the process of marketization, as well as to help the retreat of the state in the process of globalisation.

As a supplement, some scholars use the concept “crony capitalism” (Andreas 2010) or “crony communism” (Dickson 2011) to understand the close relations between the Chinese state and the market. Andreas (2010) claims that contemporary Chinese society is the one labelled as a “state-led urban decade” (65). Dickson (2003) unpacks the reliance of a capitalist economy on close relationships between business and the state by the concept of crony communism. He argues that crony communism in China is different from other contexts, because the political hierarchy is dominated by all levels of officials, rather than a ruling family or central leader as in other East Asian countries, such as Burma. Consequently, the ruling officials are titled ‘red capitalists’, as many of them are involved in the economic system: “Many of the most wealthy entrepreneurs formerly held high-level party and government posts, and some are even the offspring of China’s leaders; a far larger number of private entrepreneurs are former mid-level officials, or simply rank-and-file party members who did not hold formal posts but left their previous jobs to go into business […]” (Dickson 2003, 4–5). Meanwhile, Chinese crony communism is also distinctive because capitalists are subsumed into the group of officials: “[…] another group […] [that refers to] those who were co-opted into the party after demonstrating their entrepreneurial skills and business success’ (Dickson 2003, 4-5).

Such discussion of the Chinese economy is helpful to grasp a sense of how the Chinese economy has developed in recent years, and is a useful way to understand the context in which Chinese Internet industries have developed. Below I introduce the recent development of Chinese Internet industries, and how they reflect certain characteristics and problems of Chinese economy.

2.2. The Chinese Internet Industries

In this study, the Internet market is divided into four parts: The hardware market (including companies producing computer hardware, such as Dell); the software market—including companies producing computer software, such as Oracle; the service market—including companies providing Internet services, such as Google; and the content market—including companies producing contents or converging contents provided by Internet users, such as Facebook. The Internet industries I focus on in this paper are part of the emerging content market, like Facebook and YouTube, as well as Chinese equivalents like Sina Weibo and Youku, which accumulate massive economic capital by providing online content.

The Chinese Internet content market has developed exponentially since the end of 2002, when the market was revived from the dot-com crash in 2001. At that time, some portals in the industries, such as Sina, Netease and Sohu began to make profits and to grow significantly. In 2003, the market was developed with the blooming of varied content services, such as search engines (Baidu), online gaming (SNDA), instant messaging (Tencent), and online commerce (Alibaba). According to a research report by the Boston Consulting Group, the Internet industry economy made up 5.5% of China’s GDP in 2010 (The Boston Consulting Group 2012). In 2012, the annual market value had reached 385.04 billion RMB (£38.5 billion), an increase of 54.1% from 2011 (iResearch 2013). As a report from Xinhua News indicates, Internet-related consumption of information and services would be one of the biggest drivers of China’s economic growth in the next ten years (iResearch 2013). The development of the Chinese Internet industries can be acknowledged from Figure 1, which indicates the fast economic accumulation of the Internet industries in one minute.

Figure 1: What happens in Chinese Internet industries in 60 seconds? (Translated from Sohu IT 2013)

This fast capital accumulation largely depends on the labour efforts of Internet workers, as huge numbers of workers are involved in the production process. The number of Chinese Internet workers had increased to 12.3 million by the end of 2009 (Liaoning Research Institute of Industry and Information Sciences 2013). Little academic attention has been paid to the class formation of these workers, such as their educational background, socioeconomic status, and social locations, nor to other dynamics and practices, such as their struggles in the industries and society. So how do Internet workers contribute to the fast growth of the Chinese Internet industries? What is their working life like in the industries? How class analysis can help understand Chinese Internet workers?

The Chinese Internet industries have been subsumed into the global value chain via not only its large number of Internet users (for example, Internet users in China had reached 590 million in the first half of 2013, which is more than the number of users in the whole of the Western Europe) (Xinhua News 2013), but also the close relationships between the Internet industries and the Chinese economy: The Chinese Internet industries have become an important economic growth field in contemporary Chinese economic system. As China has become one of the most important players in the global economic system, Chinese Internet industries also have been subsumed into the global economic system.

As I stated earlier, some theorists argue that Chinese economy develops in a different way than other social contexts, because of the close relationships between the market and the state. Likewise, Chinese Internet industries are also distinguished from ones in other geographical contexts because the state holds high authority and intervenes in the production process. For example, Ross (2005) recognises that Chinese government officials are still playing a strong role in the Chinese high-tech industries, even though the central government has not been planning the economy in detail since the 1980s.

Figure 2 shows a national conference, Constructing Healthy Network Cultures (gongjian gongxiang wangluo wenhua), organised by the Party, which all CEOs of the dominant Internet companies were required to attend. At this conference, the state intended to ask these Internet leaders to follow its plan for constructing the Internet industries. For example, as the first picture in Figure 2 points out, the state required Internet companies to work in conjunction with it to build a “healthy Internet space”, where information security, such as the filtering of all information against the state, would be guaranteed. In this case, the Internet leaders were ‘re-educated’ about the state’s plan for constructing online freedom.

Figure 2: A conference in relation to Internet industries organised by CCP (From an interviewee’s personal website on Campus)

According to iResearch, a leading company focusing on in-depth research on Chinese Internet industries, all the top 21 Internet companies in Chinese market are owned by individuals, who are widely known for pulling themselves up by their own bootstraps (iResearch 2013). These leaders were mostly not part of the bureaucratic capitalist class (I explain classes in section five) before their companies became the “large entities” in the industries. The state quickly made attempts to subsume leaders of the large entities by organising such “educational conferences”, in which leaders were required to follow the state’s rules and plans concerning the Internet industries, in order to realise its slogan “control the big, let go the small" (Conner and Gu 2012, 4).

It would seem that the Internet leaders are not keen to study the state’s plan and rules about the Internet industries, as the final picture in Figure 2 indicates—some Internet leaders, such as the leader of Alibaba, the largest online commerce company that has Taobao as its constituent company, and the leader of Netease, one of the largest portals, fell asleep in the conference. But this does not mean that they reject being subsumed into the bureaucratic capitalist class. As an alternative, most times, these leaders choose to stand with the state, because their projected benefits are the same—both seek to maximise their economic benefits from the Internet industries. These benefits are mostly based on the labour efforts of Internet workers.

However, it is not only the relevant industrial policies, through which the central government intervenes in the industries, but also certain direct intervention in Internet workers’ daily practices that influence the workers’ experiences in the industries. For example, according to Leo, one of my interviewees who worked in the open platform department at Campus in 2010, officials from the State Administration of Radio, Film, and Television visit Campus every month in order to have regular meetings with workers in the Security Department. In these meetings, officials inform the workers about recent sensitive issues, which are usually related to politics and pornography, and ask the workers to delete references to these issues from all Campus’ online products, such as forums.

Take the 1989 Tian’anmen Square Protests as an example: According to Alex, another interviewee who also worked in the Open Platform Department at Campus in 2010, at the end of April every year, the company starts to organise a number of meetings to prepare for the coming June 4, where a lot of Internet users usually organise various online activities for the anniversary of the protests. The company makes great efforts to stop these online activities, obeying the rules set by the state. Normally, there are two ways the company filters information relating to this sensitive issue: Filtering key words using censorship technologies and filtering pictures by manual examination.

The user-generated content department and the security department take charge of filtering key words, such as “explosion” and “bomb”. The state provides a list of sensitive words as a guideline, which includes millions of words relating to the Tian’anmen Square Protests, and asks the company to delete them from its website. Alex’s roommate, who worked in the security department, told him that such lists included 7,300 words relating to AV (Adult Video) actresses, not to mention numerous words relating to the Tian’anmen Square Protests, which remains one of the most sensitive political issues in contemporary China. Moreover, such prohibited words on the list include not only those written in Chinese, but also words from many other languages.

The manual examination of sensitive pictures requires efforts from a lot of workers. According to Alex, almost all workers, including full-time workers and interns, and even the boss of Campus, work day and night around 4^th June, in order to filter pictures posted by Internet users and delete the sensitive ones. Normally, online pictures are examined after users have posted them, but, during this special period, these pictures need to be examined before being shown on the website. Thus, more workers are required to have excessive working hours in order to filter all the pictures. The workers, who usually work in relays, are required to work together at the same time around 4^th June, in order to deal with the mountain of work. In other words, the workers are forced to have excessive working hours because of the state’s requirements concerning sensitive issues.

Sometimes, the state directly stops workers’ programs or products, because they may still include certain sensitive issues after the workers have filtered the information. For example, Leo said that the popular online game, Godfather, created by the department in 2010 and ranked as one of the top 10 most popular online games by users, was stopped by the Ministry of Culture, because it was perceived to contain sinister gang and gambling-related content. In this case, the workers’ creativity was directly intervened with by the state, when they stopped the product because of sensitive issues.

Put simply, the state assigns the responsibility of filtering sensitive issues to Internet workers, by requiring them to delete sensitive words and pictures in relation to certain issues without giving standards. This increases workers’ work intensity. When the state is unsatisfied with workers’ practices or programs, it easily stops them. Such rough intervention from the state not only influences workers’ practices and creativity, but also increases their work intensity, as the Tian’anmen Square Protests case indicated. Therefore, as an important part of Chinese economy, Chinese Internet industries reflect certain characteristics of the economic model, such as the close relationships between capitalists and the state; the industries also show some problems of the ongoing Chinese economic reform, such as the bad working conditions (excessive working hours without equal pay, which I discuss later) caused by the close relationships between the state and the market. Before I illustrate the working conditions, I explain the methodology I adopted in this research.

3. Methodology

I carried out empirical, at times ethnographic, research in two Chinese Internet companies, in order to study the workers in this paper. The first will be called Grand; this company focuses on online entertainment, such as online gaming and online fiction. The second one will be called Campus; this provides social networking services. I used observation and in-depth interviews as my primary methodology.

I conducted in-depth interviews in three periods in Campus: Seven interviews in February 2010; nine interviews in August 2011; and five interviews in December 2011. I also spent three months in Grand conducting participant observation, where I worked as an intern to observe and keep a journal about workers’ daily practices. I also invited one worker at Campus, who will be called Galeno, who participated in the interviews in two periods in 2011, to conduct self-observation, by keeping a journal about his working life during the period of August to December 2011. And finally, due to his own habit of keeping work journals, he gave me his work journals from September 2009 to December 2011. I had hoped that more workers would agree to engage with this process, but in the event, they did not. Through these mixed, qualitative methods I explored how Internet workers get involved in the digital production of the Chinese Internet industries, which I discuss in the next section, and in the final section, I outline how I recognised a sort of worker agency as a hope in the industries.

My participant observation in Grand was covert for a number of reasons. Firstly, because Chinese companies tend to reject requests for access to do academic research, unless the research could bring them commercial benefits. Such rejection would certainly have been the case for my research into workers’ practices. Secondly, this paper develops from my PhD project, which not only focuses on the quality of working life, but also emphasises acts of worker agency. By using a covert method, I felt that I would be able to witness more “genuine” acts of worker agency, which was an important part of my PhD project.

However, covert research necessarily brings with it ethical concerns. I felt that I was deceiving “participants” as I simultaneously built personal friendships and gathered their stories. Participants told me their personal stories because they saw me as a friend; friendship therefore helped me to gather data. This then presents me with a dilemma regarding sharing the stories that participants confided in me. This dilemma and feelings of deception remain, yet I choose to write about the research in this public domain, because I feel that it contributes to understanding Chinese society and the roles played by Internet workers in that society. Indeed, this is why I chose to pursue the research through what might be seen as an ethically problematic means. I hope that if my participants read this paper and recognize themselves in it, they understand my motivations for carrying out covert research and the benefits it may bring, and that they do not feel deceived by me.

4. Joy and Tears in the Internet Sweatshop

4.1. Working Hours and Wages

Most of my interviewees, technical workers in both companies, state that overtime work is a common phenomenon in the Internet industries. For example, Tim, a new technical worker who joined Grand at the same time as me, stated that he had worked overnight for several days since he joined the department half a month ago. The following are some quotations from other workers who also experienced this issue.

In Campus, the standard working time is 10am to 9pm […] This [long working hours] is quite common in the industries nowadays, where 10 hours are the average working hours […] I have a friend working in another Internet company, where he usually finishes his work at 11pm or 12am […] As far as I know, workers in the 3G department now are still staying in a hotel to work day and night for a new program […] (Louis, former technical worker in the Open Platform Department at Campus, 28^th August 2011, interview)

Non-technical workers, who deal with administrative-related work, in both companies, also share this overtime experience. For example, Katy, a new HR worker, who joined Grand at the same time as me, worked overtime for nine days, which usually meant she finished work at 11pm or 12pm, during the first ten days she joined the company. She said that sometimes she even considered sleeping over in the office, because her home was far away from the company. If she was “lucky” to finish work at 11:30pm, she could claim reimbursement for the taxi fee, would arrive home after midnight, and go to sleep by 1am. But she would need to get up at 5am, as she needed to change three times on the underground to arrive at the office. Thus, the overtime seriously reduced her sleeping time and made her very tired.

There are several reasons for Internet workers to choose overtime work, such as rewards they receive from the work—to satisfy themselves when having certain achievements in the work. But one of the important issues that forces Internet workers to work overtime is the high competition given by managers.

Ross (2005) points out that workers in the Chinese outsourcing high-tech industries work overtime because of high competition in the industries: They need to work harder to help avoid the risk of other workers taking their places. Some of my participants show the anxiety caused by high levels of competition.

My leader always reminds me to keep myself in a high competitive status. For example, to think about who will be fired if the company needs to lay off staff, if I were the person, it means I need to work more. (Galeno, technical worker in the Product Administration Department at Campus, 24^th August 2011, interview)

Managers and companies play a key role in pushing these workers to work overtime, because they seek high profits with low labour costs. For example, Galeno shows how his manager keeps him in a sense of crisis: He may be laid off if he does not work hard. It is also common to promote “geek culture” in all Internet companies, which encourages workers, especially technical workers, to show their love of the Internet work via working day and night.

Such long working hours causes a striking issue in terms of Internet workers’ well-being in recent years: Karoshi, a Japanese term meaning death from overtime work. There were some cases in the Internet industries during the five months that I spent there, where some workers died from exhaustion, which was usually in relation to overtime and high pressure.  For example, one of these examples was in Tencent, one of the main portals in China with the famous instant messenger system QQ, where an online editor died because of overtime work. The news was circulated on microblog sites because the editor talked about his overtime situation on his microblog (Sina News 2010). Some of his posts showed that he even worked until 8am in the morning.

Another karoshi case was in Baidu, the main search engine company in mainland China, where a member of staff in the online game department died from overwork on 14^th November 2011 (Tencent Technology 2011). According to this news, the average age of death from overtime in the Internet industries is just under 38 (Tencent Technology 2011). Sam, the senior manager in HR department in Grand, says that it is not unusual to find cases of karoshi in the Internet industries. According to him, there were some cases of karoshi in Grand before I joined, but the PR department covered them up.

All these stories show the serious overtime situation in the industries. The most significant issue here is whether the overtime is reasonably rewarded. According to my participants, indeed, little of this overtime work is rewarded. In Campus and Grand, full-time workers’ overtime work during weekdays is not paid. Instead, workers working late during weekdays can benefit from a free dinner or the money to buy a dinner, which is usually offered with certain restrictions. For example, in Campus, workers can only benefit from a free dinner after working for 12 hours. In Grand, workers only receive £1.80 for a dinner when they work after 8:30pm, and can be reimbursed for taxi fees after 11:30pm. Working during weekends is paid at double time, but in both companies, weekend overtime work needs to be approved by department leaders, who usually encourage workers to finish their work during weekdays. In other words, full-time workers are forced to work overtime without reasonable rewards. Put the unrewarded overtime work aside, it is important to ask how much Internet workers earn for their hard work. In both companies, technical workers are paid £13,000 per year, and non-technical workers receive £4,500 per year, compared to SOE (State-Owned Enterprise) workers’ yearly salary of over £10,000. SOE workers also receive various bonuses from the state and some of their living costs are covered by the state. For example, according to my interviewees, workers in a state-owned flight company only need to pay £0.1 per day for their meals, which are good quality (most food is organic). In comparison, Internet workers, as workers in private enterprises, live only with this fixed salary, without extra bonuses and cheap meals.

My salary is £13,000 per year before tax [£1,000 per month x 13 months] with the endless overtime, while my friend who works in a SOE, earns around £20,000 per year after tax [£800 per month x 16 months, and a bonus of £200 per quarter x 4 quarters] without any overtime (William, technical worker in the 3G Department at Campus, 19^th December 2011, interview).

According to William, SOE workers are paid much higher than Internet workers, even technical Internet workers with high skills. The difference might be assumed to relate to their different work intensity, but the reality is that Internet workers are paid less, and have much higher work intensity. In contrast, SOE workers receive higher pay without working any overtime; neither do they have to devote themselves to their work during work time, as Internet workers do. In other words, Internet workers are paid unequally, compared to SOE workers.

This is what Wright (2010) points out in his research in Chinese society: SOE workers receive benefits from the party-state that “have been unavailable to other poor individuals” (3). Such inequality between SOE workers and Internet workers can also be found in their different benefits: SOE workers and executives in Internet companies enjoy specially delivered organic food as a benefit, whereas most Internet workers do not.

There is another issue relating to Internet workers’ benefits: Sharing companies’ stocks. Internet work has been fetishised in recent years because the workers are likely to receive companies’ stocks as their bonus. For example, Xinhua News reported that the listing of Baidu—the dominant search engine company in China—in the USA NASDAQ stock market in 2005 enabled 400 workers in the company to become millionaires because they were given some stocks before the listing (Xinhua News 2005). It sounds as if every Internet worker is able to become a millionaire once he or she receives stocks. However, large numbers of Internet workers are struggling in difficult working conditions with the dream of being the next hero, but only a small number of individuals achieve the dream.

I had some stocks before I left [Campus]. They were just 3,000 shares. It was £0.4 per share when I received them, which was evaluated as £1 per share when I left [Campus]. It seemed that I could receive £3,000 when I left. But, indeed, the company had a rule, which meant we could only sell a quarter of our shares every year. So this year I can only sell a quarter of my shares, which is £750. But because income from stock sharing is windfall in the income tax law, I need to pay 47% of this amount of money as tax, which means I only receive £397.50 after tax. This is even less than my monthly salary […] (Louis, former technical worker in the Open Platform Department at Campus, 28^th August 2011, interview)

According to Louis, only some experienced workers who had joined Campus when it was founded could receive certain stocks. These stocks are not worth as much as workers in other industries imagine them to be and few of the more recently employed workers can receive stocks.

Most of us were so disappointed when the company went public. All the managers celebrated it, but it’s none of our business. We do not get any benefit from it […] Even to the workers who received stocks, such sharing does not guarantee them anything. It only becomes a way for the company to stop [workers] job-hopping: once you want to leave the company, HR would suggest you stay for one more year, in order to sell another a quarter of your shares. This does not only happen in our company, but also happens in other Internet companies. It [stock sharing] becomes a way for companies to bargain with us [workers] […] (Galeno, technical worker in the Product Administration Department at Campus, 20^th December 2011, interview)

Here, it is evident that stock sharing does not authorise these workers any managerial power; instead, it becomes a way for the company to control workers. With such unequal pay, some Internet workers experience difficulty of housing in cities where they work. For example, most Internet workers are struggling to buy a house in the big cities with their low salary.

I never think about settling down here, in Beijing, because it’s too expensive to be here, especially buying a house […] Now I rent a house with my colleague, each of us pays £200 per month for a small and old room. It is still not cheap for us […] (Wynn, technical worker in the Open Platform Department at Campus, 24^th August 2011, interview)

In these big cities, it seems that only SOE workers, civil servants, government officials and the rich can afford to buy a house. As Shelly, an intern in Grand, told me, civil servants in Shanghai are usually offered discounted houses by their work units (danwei), for which they only need to pay less than half of the market price, because these houses are built by their work units with free land provided by the local government. Here, the work unit again plays an important role in workers’ benefits, as it did in Mao’s era. But the difference is that the work unit in contemporary Chinese society only protects and guarantees certain classes benefits, such as the higher middle class—civil servants and SOE workers.

Admittedly, there are some jobs that pay more than others, which cannot be criticised as a result or form of exploitation. But, in this study, Internet workers receive low pay, in spite of their long working hours and high work intensity, because they are controlled by the higher classes, such as the capitalists in this study—Internet companies, via work contracts (Internet workers cannot escape these experiences of unrewarded overtime work, because it is so common in the Internet industries). This shows that Internet workers’ labour efforts and skills are appropriated in the capitalist production process via long working hours without equal pay.

As I pointed out earlier, some technical workers receive certain stocks as part of their bonus, but such stock sharing does not guarantee them good pay, nor does it authorise the workers any managerial power. With such limited benefits, Internet workers still need to sell their skills to survive. In other words, Internet workers are forced to sell their labour efforts and skills in the Chinese capitalist market, which has been subsumed in the global value chain as the second largest economic system, with an unreasonable price. I argue that such unjust experience is caused by a sort of exploitation between classes: The bureaucratic capitalist class—the state and Internet companies in this study—dominates the lower-middle class and appropriates the labour of this class, because the former own means of production and the political power to allocate these resources. I explain the way I discuss classes in section five. This exploitation between classes results in an injustice in relation to Internet workers’ pay and work time: Internet workers are not rewarded for their high work intensity, and in contrast, government officials and capitalists appropriate this part of labour efforts. Therefore, the structural concept, exploitation, is the force behind Internet workers’ unequal and unjust experiences in terms of work intensity and pay. I clarify the way I use this concept in the theoretical part—sections five and six.

4.2. Job-Hopping or Lay-Off?

Some theorists argue that job-hopping becomes a common phenomenon in Chinese high-tech industries (the hardware market). For example, Ross (2005) points out that the average turnover rate in the Chinese high-tech industries is about 20%, which is quite high in his understanding. This high turnover rate was reflected in my fieldwork: Six out of the seven interviewees, whom I contacted in 2010, had left Campus by the time I contacted them again in 2011.

As some of my interviewees say, two years is quite a long period to stay in the same company in the Internet industries.

Most workers in my department, almost 20 people, left in just one and half years […] Every time we gathered for team building, I saw new faces […] (Galeno, technical worker in the Product Administration Department at Campus, 24^th August 2011, interview)

This obviously indicates a high turnover rate in the Internet industries. Some theorists point out certain problems caused by this high turnover rate. For example, Ross (2007) points out that such frequent turnover, results in high costs for training, because workers usually leave after benefiting from training, and then new workers have to be trained to replace them. Ross (2007) also claims that job-hopping becomes a bargaining tool for experienced workers to negotiate with companies for good pay.

But, indeed, such bargaining experience only applies to certain experienced technical workers, who have the high skills to bargain with companies. In contrast, most workers change their jobs because they struggle to earn good money and want an easier working life.

Now everything’s getting expensive, I need to find another position with better pay, so that I can survive in this city […] But autonomy in work is the more important reason for me to job-hop […] (Walter, technical worker in the Advertisement Department at Campus, 25^th August 2011, interview)

Here, Walter points out that good pay is an important reason for him to change jobs, though autonomy is another important reason for him to consider changing jobs. As in most countries, working and living in big cities is not easy. Some workers, most of whom are in middle class locations, choose the new lifestyle of a “weekly couple” (zhoumo fuqi) or “monthly couple” (yuemo fuqi), which refers to the family gathering that takes place once a week or once a month, because of the high living expenses in big cities. For example, Sam is a senior manager in Grand’s Shanghai office, and his family are based in Najing, another city two hours away from Shanghai by train. The high cost of living in Shanghai stops his family from moving there with him, and the terrible traffic jams, means it takes Sam more than four hours to go back home, and stops him from gathering with his family every day. Then, the only choice for Sam is to be a “weekly couple” with his wife by going back home every weekend. This new lifestyle obviously influences workers’ life quality, as it separates workers from their families. This goes against the traditional Chinese family life, which is based on living in groups. It is the high cost of living in big cities and their unequal pay that push workers to choose this modern lifestyle and decrease their quality of life.

Put simply, some experienced technical workers do benefit from the frequent job-hopping in the industries, because they have high skills to bargain with companies for good pay. In contrast, many workers choose to change their jobs because they suffer from bad pay and high living costs in big cities. These workers pursue a good job, with good pay and a convenient lifestyle, by frequently job-hopping. Moreover, there is another side to the coin: in some cases, companies lay off workers.

I had a colleague, who had been at Campus for four years, but he still received the same salary as me, and I had just joined the company… It was one way the company forced us to leave […] Finally, he left as most new employees were paid higher than him… If I were my colleague, I would also choose to resign, as salary is a way to evaluate a person’s talent and to show respect to that person […] It doesn’t matter how much you are paid, but it matters how much more you are paid than others, especially people whom you think are less talented than you […] (Galeno, technical worker in the Product Administration Department at Campus, 24^th August 2011, interview)

In this case, workers are forced to leave by the company in an indirect way, because the company needs “fresh blood” to bring creative ideas. Workers, indeed, are laid off in an indirect way.

Three months after I finished my internship in Grand, I contacted my friends there, and was surprised to find that more than half of the full-time workers in the HR department had been laid off, because the company wanted to cut its labour costs. According to the work contracts, these workers were informed of their redundancy one month before they left and were paid one more month’s salary as compensation. These workers did not have the time to find new jobs in one month, but they had to accept the situation. In the Chinese context, work unions do not work well. The Chinese unions usually stand with companies, by persuading workers to accept unfair working conditions, rather than protecting workers.

In both examples, workers were laid off in an indirect or direct way, without any protection from the unions. The lay-off issue addresses an urgent problem in the Internet industries: Internet workers lack protection from the work unions, such as job security. Admittedly, some workers benefit from frequent job-hopping. But, more workers change their jobs because of the bad working conditions they experience. And some workers are even forced to leave without protection, because companies want to save labour costs or employ “fresh blood”.

4.3. Life After Retirement

As I have discussed in the last two sections, Internet workers experience serious overtime work, unreasonable rewards for their hard work and unstable working status. But this is not the complete picture of their working life. They also experience insecurity in their working life, such as the insecurity of life after retirement.

Similar to some countries, in contemporary Chinese society, certain types of retired workers are protected by the pension system. For example, Internet workers are required to pay 8% of their salaries into a pension fund, to which the companies give an amount equal to 20% of workers’ salaries. The workers will receive a certain amount of this money every year after retirement. Compared to the pension system in which Internet workers—employees in private enterprises—participate, SOE workers and civil servants benefit from another pension system, in which the work units pay into the pension fund, but they will be given more money than private enterprises’ employees after retirement.

After the pension reform in China, workers now need to work longer and pay more into their pension funds than they have done before. Workers deposit part of their salaries in a pension fund, in order to guarantee their quality of life after retirement. However, in China, this amount of money is shrinking, due to the failure of investment in pension funds in recent years. It is reported that the growth rate of pension funds is much slower than the growth rate of the CPI (Consumer Price Index) (Chinese Economics 2013). In other words, the pension that workers will receive after retirement will no longer cover their living costs.

Such inequality surrounding pension systems points to injustice between Internet workers and SOE workers: Internet workers experience difficult working conditions, such as long working hours without reasonable rewards, while needing to pay into their pension fund. In contrast, SOE workers have good working conditions and benefit from a state-secured pension. In this sense, the higher middle class, such as SOE workers, seems to have better working conditions than the lower-middle class, such as the Internet workers, because of the support they receive from the state and companies—the bureaucratic capitalist class, which I clarify later.

In the contemporary Chinese context, there is another issue links such inequality to exploitation: pension corruption. Workers are worrying about their lives after retirement, even though they pay into a pension fund every month, because government officials might embezzle their pension before their retirement. For example, there were several scandals regarding pension corruption from 1993 to 2010: around £8.9 million of pension money was embezzled in Guangzhou in 1993 (Youth Times 2012); £0.86 million pension money was embezzled in Taiyuan in 2003  (Youth Times 2012); and £32 million pension money was embezzled in Shanghai in 2006 (Youth Times 2012).

These news reports make workers feel unsafe and insecure about their work, especially about life after retirement. As Monica says:

I’m not sure whether I could get back my pension after retirement, because it is possible that some officials have already embezzled it before I retire (Monica, HR worker at Grand, observation journal).

This pension corruption indicates that government officials appropriate workers’ labour efforts, shown as the money they pay into their pension fund. To take a more structural perspective, this is resulted by an appropriation of labour of a low class by a high class: The low class’s labour is “robbed” by the high class via acts of corruption, due to the strong political power to allocate social resources and control the lower classes owned by the bureaucratic capitalist class. This makes “the poor becomes poorer, while the rich becomes richer”. In other words, the bureaucratic capitalist class exploitation of labour of the lower-middle class results in insecure experiences in Internet workers’ working life.

Some of my interviewees even state that they could not imagine their lives after the age of 40 in the Internet industries:

We [Internet workers] sometimes made jokes that we might die before [our] 40s […] To be honest, I cannot imagine my life after [my] 40s. Maybe I will leave the industries […] (Galeno, technical worker in the Product Administration Department at Campus, 24^th August 2011, interview).

Galeno’s recognition of high risks in the Chinese Internet industries echoes Gill’s (2002) findings in her study of new media industries: Work in new media industries is characterised by issues of insecurity, low pay, and long working hours. These unsafe and insecure working conditions, such as lay-offs, karoshi, and unguaranteed pension fund, make Internet work unstable and precarious, though there are other things that make it precarious, such as long working hours and low pay. These precarious and risky working-life experiences of workers in the Internet industries, answers my question addressed in the beginning of this paper, specifically how these workers are involved in the global value chain. I now turn to explain why workers still stay in the industries, despite the bad working conditions.

4.4. Why Don’t Internet Workers Leave the Sector? Autonomy?

As I pointed out earlier, Internet work has been fetishised in recent years because some workers have become millionaires by receiving companies’ stocks. Meanwhile, the Internet work is fetishised also because of offering a high level of autonomy to the workers. It is worth to ask, whether autonomy is the force of keeping Internet workers in the industries, despite the difficult working conditions.

Hesmondhalgh and Baker (2010, 39–44) clarify autonomy in the cultural work with two concepts: Workplace autonomy and creative autonomy. Workplace autonomy refers to the degree of workers’ self-determination within certain work situations, such as how they balance their work and life in their daily practices. Creative autonomy refers to the freedom in terms of practicing creativity. Here, I focus on workers’ self-determination in their work practices and creativity, under the pressures that are exerted by the state and firms.

It is easy to understand that workers’ freedom of self-determining their project is affected by companies. In some cases, Internet workers are forced by companies to conduct some projects that they are not good at. For example, during my observation, all workers in some technical departments in Grand, no matter what areas they were specialised in, were encouraged to learn developing Android—an open source operating system, which was the next key project for the company. It seemed that workers were “encouraged” to pick this option, but the company issued a new rule to process the program: Department leaders would have deductions from their salaries, if 30% of the employees in their departments could not pass the Android test. Put simply, the workers were forced to learn new technologies and conduct new programs without any consideration of their capabilities and interests. Two workers in the operations department told me that they needed to conduct the Android project at the same time as learning the skill, which was quite stressful to them. Therefore, the workers involved in the Android project did not have any freedom to decide which projects to carry out.

However, in both Internet companies, there are also workers who consider themselves to have a considerable degree of freedom to self-determine practices and creativity, such as workers in the online novel department in Grand.

The main job of workers in the online novel department is online editing. Their work includes managing online writers’ writing and emotions. For example, when online writers face difficulties writing novels, these online editors are responsible for cheering them up, providing ideas, searching writing materials, and even suggesting structures for their stories. These editors are allowed a high level of freedom in determining these practices, such as deciding when and how to manage the writers and novels.

Such authorisation of their high level of professional autonomy covers some disadvantages within the work: The workers need to work day and night without receiving equal rewards. In order to maintain good relationships with the online writers, who usually start their writing at night because they are only part-time writers, these editors need to work until late at night. Surprisingly, although they carry out such day and night overtime, these editors receive just £300 a month without any overtime pay compared to other full-time workers’ £1,000 monthly pay.

Thus, it might be claimed that some Internet workers, such as workers in the online novel department, stay in the industries, because of the high level of autonomy given by companies in their daily practices and creativity, though there are other difficult working conditions along with such autonomy. It then becomes necessary to ask, why Internet workers suffer such ambivalent working conditions—excessive working hours with unequal pay, lay-off without protection from unions, insecure work status, and certain autonomy. Below, I adopt Eric Olin Wright’s concept of exploitation to answer this question.

5. The Class Analysis Approach

Erik Olin Wright (2009) illustrates three approaches relevant to class analysis in order to clarify his approach towards exploitation: Stratification research, which defines classes with “the attributes and material life conditions of individuals” (102); the Weberian approach, which centres on “the ways in which social positions afford some people control over economic resources while excluding others” (102); and the Marxist approach, which regards classes as “being structured by mechanisms of domination and exploitation, in which economic positions accord some people power over the lives and activities of others” (102). 

The stratification approach focuses on class background, which consists of attributes such as sex, race, religion, age, education, and material life conditions, which refers to adequate income, dire poverty, and pleasant suburban houses. This approach identifies the middle class as people who “have enough education and money to participate fully in the vaguely defined ‘mainstream’ way of life (which might include particular consumption patterns, for example)” (Wright 2009, 103). But, as Wright acknowledges, this approach lacks serious consideration of the injustice and inequalities between different social positions, such as why some jobs are better than others. 

The Weberian approach focuses on the unequal “opportunity hoarding” (104), which highlights the restricted access to certain positions. For example, high levels of education are restricted to the upper classes, because of the high tuition fees. Good education then further benefits the upper classes, as it usually relates to good jobs. In other words, unequal locations within market relations are causally connected to unjust opportunity hoarding among different social positions. According to Wright (2009), this approach is usually adopted by sociologists to analyse American society, where the middle class is defined by “mechanisms of exclusion over the acquisition of education and skills” (106). In Wright’s discussion, the Weberian approach has a critical difference to the stratification approach, as it indicates that, “the economic advantages gained from being in a privileged class position are causally connected to the disadvantages of those excluded from such positions” (106). Put simply, the upper class’s economic advantages are causally related to the lower class’s disadvantages. Breen (2005) has discussed Weberian understandings of the mechanisms sustaining the privileges of advantaged classes in terms of the concept of “life chances”. Life chances are chances that “individuals have of gaining access to scarce and valued outcomes” (43). He then claims three aspects of the distribution of power in society, which is widely adopted by Weberian approach, as factors that influence the distribution of life chances: Classes, status groups, and parties. All these dimensions overlap, while none of them can be reduced to others. The status groups imply “some level of identity in the sense of some recognised ‘positive or negative social estimation of honour’” (Weber 1978, 932; Wright 2002, 834). In other words, members of a status group are conscious of being members of the group. Wright (2002) points out that Weber distinguishes status from class by highlighting the different mechanisms through which they shape inequalities of the material conditions of people’s lives: Status affects people’s well-being with ‘the monopolisation of ideal and material goods or opportunities’ (835); by contrast, class influences people’s well-being via economic assets that people bring to market exchanges. Weber distinguishes status groups from classes by highlighting members’ consciousness, and regards classes as objective places.

The Weberian approach usually relates inequalities between different classes to different life chances: Workers are given different chances to access to scarce and valued outcomes. Both Weberian and Marxist approaches agree that occupants of different classes enjoy different life chances, though they provide different schemata for understanding class. The Weberian approach might understand these inequalities and injustices in terms of salary and benefits shown above, as a form of exploitation, because they show inequalities in people’s daily experiences. But the Marxist approach, especially the neo-Marxist approach adopted by this research, understands exploitation and domination as the mechanism linking different class locations to variations in life chances. According to Wright, the Marxist approach focuses on mechanisms of exploitation and domination. Unequal opportunity hoarding does not only relate to restricted access to certain positions and resources, but also depends on the ability of the exploiting/dominating group controlling the labour of the exploited/dominated group. Put simply, this approach highlights “an ongoing relationship between not only the conditions but also the activities of the advantaged and disadvantaged” (108). The traditional Marxist approach distinguishes three class locations based on ownership of means of production: The capitalist class, the petty bourgeoisie, and the working class.

Some theorists in the 1980s argue against this Marxist approach to class locations, stating that the working class has diminished or even disappeared in capitalism, especially the manual working class, which only occupies a small part of the workforce, and that white-collar workers are already in the position of the middle class. However, Callinicos (1983, 193–195) claims that many people still occupy the position of the working class, even though they are not engaged in manual labour in Marx’s industrialist capitalism. For example, because of the industrialisation of office work, clerical workers are doing similar work to manual workers and suffering from a similar working condition to manual workers, with the massive introduction of new technologies. Therefore, they are in the same position as the working class, as manual labourers, because they are “compelled to sell their labour-power in order to live” (193), even if they do non-manual work. As a result, Callinicos argues that the change in class location and class relations since Marx’s period is “a shift in the structure of the working class, not its abolition” (195). In later work, Callinicos (2004) introduces Wright’s work on class to explain the fragmentation of class structure in contemporary capitalism. As Wright (1985) argues, class locations in modern capitalism are contradictory, as some positions share properties of both labour and capital. For example, managers perform some functions of capital by directing others’ work, but still sell their labour-power in order to live.

In the book Classes, Wright (1985) explicitly introduces his framework of contradictory class locations and fragmented class structure. He argues that Marxist criteria for class are an approximate framework for class structure in capitalism, rather than an elaborated classification. He then develops a much more complex typology of class in capitalism, where he divides typology into two parts: Owners of means of production and non-owners. Among these non-owners, their locations are divided by organisation and skill/credential assets. The class locations of wage labourers in a capitalist society are classified into expert managers, non-managerial experts, and non-skilled managers, etc. Wright (1996) further modifies this typology of class locations in his later work by specifying three dimensions that clarify class relations: Property, authority, and expertise/skill, which is where questions of symbol making and manipulation come in. The property dimension consists of employers, the petty bourgeoisie, and employees; The authority dimension is divided into managers, supervisors, and non-managerial employees; and the expertise/skill dimension contains professionals, skilled employees, and non-skilled employees (704).

Wright (2009) aims to move beyond the traditional Marxist approach to class analysis by developing a detailed typology of class locations. He identifies certain key aspects that constitute the new class structure of his model: The mechanism of exploitation and domination in the traditional Marxist approach, the mechanisms that sustain the privileges of advantaged classes in the Weberian approach, and the individuals’ class locations in the stratification approach. He argues that a completely elaborated class analysis needs to combine the ‘macro-model of conflict and transformation with the macro-micro, multi-level model of class processes and individual lives’ (111) (see Figure 3). Put in another way, Wright argues that individuals’ lives depend not only on the micro-model of attributes and material life conditions, but also on the macro-model of social conflicts and transformations where their lives take place.

Figure 3: Combined class analysis: macro and micro processes (Wright 2009, p.111)

Wright’s work (2009) then suggests a necessity to analyse class locations by locating individuals’ lived experiences, such as “class background”, in the context of social conflicts and transformations. It is no longer the problem of individuals who fill these positions, but rather, it is important to recognise the mechanisms shaping the privilege of certain class positions. As Wright points out, the middle-class problem is not who is excluded from the position, but is the fact that “there are mechanisms of exclusion that sustain the privileges of those in middle-class positions” (109). Likewise, I adopt a neo-Marxist class analysis approach, which combines both the macro model of transformation and the macro-micro model of individual lives. It is not my interest to just identify the scope of the Chinese middle class, by clarifying who is excluded from the position; rather, my aim here is to recognise the important and unique positions of Internet workers in the general Chinese social structure, and to clarify the mechanisms that sustain and change their unique positions (probably privileged positions) in the Chinese context, which result in inequalities in their working lives, as shown in section four.

6. Why Bad Working Conditions? Exploitation?

6.1. The Lower Middle Class: Chinese Internet Workers

Based on Wright’s model of class structure, here, I explain Internet workers’ position in the Chinese class structure, and clarify how certain mechanisms shaping their working lives. So (2003) points out that a new dominant class emerged in China during the process of privatisation of SOEs in 1992: Cadres set up their own businesses, which at times cooperated with foreign capitalists, by usurping resources from SOEs where they had executive positions. Capitalists also joined the existing structure using bribery to access to the market and gain resources. The new partnerships between cadres and capitalists enabled the new private sectors to “save on the additional costs of pension schemes, health and welfare insurance, environmental protection facilities” (368), which ultimately led to the deterioration of working conditions in private enterprises.

Such discussion highlights that the bureaucratic capitalist class accumulate capital via appropriating lower classes’ efforts, which ultimately leads to the deterioration of working conditions. The class typology in the contemporary Chinese context is complex,

but my focus in this paper is the low-class location of Internet workers, due to which the labour of this class is appropriated by another higher class.[1]

In the Western context of the UK cultural industries, Hesmondhalgh and Baker (2010, 68–69) place creative workers in these industries principally in a middle-class class location, although they acknowledge that there are various classes involved in cultural production (such as working-class cleaners). Based on Wright’s schema, which is also a fundamental framework for this study, they claim that most creative workers in the cultural industries occupy lower-authority, higher-skilled positions, such as skilled workers with little managerial power.

However, in the Chinese context, I argue that creative workers’ location in the middle class, such as Internet workers’ location, is not only decided by their skills and managerial powers, but also decided by their family backgrounds, work units, education level, and political authority (see Figure 4).

Figure 4: Typology of Chinese middle class

According to Figure 4, education level, family background, political authority, skills, and work units generally decide individuals’ positions in the middle class: Workers in SOEs and civil servants have higher positions than private enterprise workers; workers who are highly educated have higher positions than workers who are not; workers who have family members as cadres have privileged positions; party members are more likely to be guaranteed stable work and lives than others; and high-skilled workers have more possibility to have well-paid jobs than others. It is hard to quantitatively evaluate the influence of these five factors on individuals’ locations in the middle class, such as whether individuals who are not party members but have high skills have higher positions than individuals who are party members but work in private enterprise, and it is not my aim to do so here. Instead, I highlight these issues to give a variation within classes to analyse Chinese Internet workers’ social class.

As I stated in section two, most Internet workers in China are working in private enterprises, which are excluded from certain advantageous resources that are only available for SOEs. Due to the inequality between SOE workers and workers in private enterprises, which I illustrated earlier, it is possible to say that large numbers of Internet workers do not have family members in the bureaucratic capitalist class, who have priority to send their offspring to beneficial jobs, such as SOE work and civil servant positions.

Most of these workers still conduct intellectual work, which is defined as middle-career (Zhou 2008, 114-117), though based in private enterprises, which are inferior to SOEs, due to their limited access to certain advantageous resources. As I pointed out earlier, in the existing media reports and academic research, there is no survey conducted amongst Internet workers to report their education background and income. Instead, according to a sample survey conducted amongst IT workers (workers in the hardware market) in some big cities (such as Shanghai, Beijing, Wuhan, and Dalian) in 2010, 97.13% workers were educated at college level (Li 2010, 128). This figure enables us to deduce that a large number of Internet workers are also educated at colleges. Indeed, according to my qualitative research, all the participants and interviewees are educated at college level, which might help us recognise that generally most Chinese Internet workers are highly educated.

Likewise, as little research investigates Internet workers’ income, it is hard to give an authoritative figure about Internet workers’ income. Rather, according to an annual report about salary information in various industries, which was conducted by a professional HR service company, PXC, in 2013, the increase in salary rate in the Internet industries was 16.2%, which was the highest among all industries (excluding SOEs and civil servants’ positions) (GRlib 2013). Meanwhile, according to Guangzhou Daily, a local newspaper, annual salaries of fresh graduates who find jobs in the top 5 Internet companies are between £10,000 and £15,000, which is a middle-level salary for most jobs (Gangzhou Daily 2013). This indicates that Internet workers have a high-level salary among jobs in private enterprises.

Though CCP intends to control big private enterprises via subsuming employees there into its party system, as its managerial slogan in the cultural industries indicates: “Control the big, let go the small” (O’Conner and Gu 2012, 4), it is still hard to find large numbers of party members in the Internet industries. For example, according to one of CCP’s official magazines Oriental Outlook, only 9 Internet companies in Beijing had organised Party Committees until 2011, and there were only 2,680 party members in all Internet companies in Beijing, who were mostly in Baidu and Sina. Most of these members joined the Party after 2010 (Oriental Outlook 2013). In other words, large numbers of Internet workers are non-party members.

Unquestionably, as Hesmondhalgh and Baker (2010) claim, just as most cultural workers are highly skilled, either with professional skills or technological skills, Internet workers also have high professional or technological skills. Therefore, the picture here becomes clear: Most Internet workers are based in private enterprises, which indicates that they do not have family members in the bureaucratic capitalist class; most workers are college-educated with low political authorities, as they are non-Party members; and they earn high salaries among people in the middle-class location, as most of them are highly skilled. As I stated earlier, it is hard to evaluate Internet workers’ location in the middle class with any sophistication with these figures. However, this generally shows that the Internet workers here occupy an inferior position to SOE workers and civil servants, but this does not indicate these workers have an inferior location in the Chinese social structure—the Internet workers still occupy more privileged locations than those in working-class locations. In other words, most Internet workers occupy the lower position in the middle class.

6.2. The force behind the unequal working-life experiences: exploitation

After clarifying Internet workers’ lower-middle class position, it is necessary to explain how certain mechanisms shaping these workers’ working-life experiences, due to their class position. Following Marx’s work, many theorists regard injustice as central to understanding Marxist ideas of exploitation (Callinicos 2000, Roemer 1982, Cohen 1985 and Wright 1996).

Roemer (1982) pays particular attention to exploitation in existing socialism (Roemer 1982, Chapter Eight). He claims that exploitation still exists in socialism, and that socialist exploitation, based on the inequality in ownership of skills, is socially necessary at a certain stage. The historical task of socialism is to eliminate capitalist exploitation, rather than socialist exploitation. However, Roemer, as a follower of Marx who devotes considerable efforts to explain Marx’s work, still shows his ethical concern about socialist exploitation, by asking “if a form of exploitation is socially necessary, what should one’s attitude toward it be? Should its existence be endorsed?” (240). Roemer answers these questions with reference to the level of “social consciousness—how the people involved think” (248). He puts it thus: If the exploited fight against injustice, even though the revolution or rebellion is doomed to fail, the social necessity of the exploitation then should be questioned. Because of this, it would be morally wrong to accept a form of exploitation that seems to be socially necessary, without criticising it.

Cohen (1995) argues that Roemer correctly states that exploitation is not based on natural injustice, but that he is incorrect to indicate that an unequal product flow is unjust “only if it reflects an unjust initial asset distribution” (204). The work of exploitation needs to focus on the unjust exploitative allocation. This is because Roemer’s work directs our interests to unjust asset distribution, which in Cohen’s understanding is caused by the unjust product flow. As an alternative, Cohen states that it is necessary to focus on the “injustice of an exploitative allocation” (207) rather than the “injustice of the initial distribution” (207), as the former generates the primary injustice that drives the latter to be unjust.

Wright (1985) criticises Roemer’s work on exploitation because of the elimination of class relations in his analysis of injustice. For example, Wright states that Roemer fails to point out that “real transfers from one actor to another” (74) create unjust inequalities. According to Wright, Roemer fails to introduce the notion of dominance in his game-theory approach to exploitation. As an alternative, Wright defines exploitation as a process that contains both “economic oppression” (1985, 74) and the “appropriation of the fruit of the labour of one class by another” (74). As a criterion of exploitation, the benefits of the exploiter must depend on the work of the exploited.

Wright (1976, 28–29) claims that exploitation needs to be discussed within varied modes of production, as different forms of exploitation correspond to different modes of production. For example, workers in industrial capitalism are exploited in a way that is distinguished from workers in the earliest capitalism: on the one hand, they cannot control the labour process as producers in cottage industries did, because they are gathered in factories; on the other hand, the labour force is deskilled and the production process is fragmented, because of the introduction of new technologies in factories. Meanwhile, capital is not a commodity in existing socialism as it is traded in capitalism. Burawoy and Wright (2002, 478-480) distinguish existing socialism from Marxist socialism by using the example of Soviet communism, which is entitled “state socialism”. They claim that state socialism refers to a central planned system: a class of “planners” take charge of the “redistribution of surplus”, which is extracted from a class of “direct producers”. This extraction is legitimised in the name of “the superior knowledge of the planner about the needs of the people” (479).

Roemer (1982) divides modes of production into four categories, based on the different forms of exploitation: Feudal exploitation, which is based on injustice generated by unequal distribution of labour power assets, in which lords and serfs are the main classes; capitalist exploitation, which is based on injustice generated by unequal distribution of alienable assets, in which relations between bourgeoisie and proletariat are the main class relations; status exploitation, which exists in the existing socialism, a historical stage between capitalism and socialism; and socialist exploitation, which is based on injustice generated by unequal distribution of inalienable assets, in which experts and workers are the main classes. In status exploitation, exploiters control labour power and property because of their high status in the social structure. This is different from the injustice generated by either means of production or skills.

Wright (1985) agrees with Roemer that skill-based exploitation would exist in a Marxist socialist society, and it could only be eliminated in Marxist communism. But Wright indicates that Roemer’s concept of status exploitation is problematic in two ways: First, it is not necessarily related to production at all and second, it is hard to distinguish it from feudal exploitation. As an alternative, Wright (1985) points out a post-capitalist mode of production that exists between the stages of capitalism and socialism, statism, which is based on organization asset. In this mode of production, bureaucrats and managers occupy the class location of the exploiter.

This is agreed by Callinicos (1983), who interprets that existing socialism is “bureaucratic state capitalism” (183), as “a state bureaucracy, which competes with its Western counterparts” (183) exploits the working class. In the context of the Soviet Union, socialism, or “bureaucratic state capitalism”, did not self-emancipate the working class, as it claimed. The followers of the Soviet Union, such as China, reproduce this mode of bureaucratic state capitalism in their societies. Callinicos (2004) further explains his arguments in his later work. He states that the existing socialist societies are “state bureaucratic socialist, combin[ing] the statist and socialist modes of production” (223). This includes multiple occurrences of exploitation based on the unequal ownership of varied resources: “Skills, organisational assets, means of production, labour-power” (225).

Some theorists who work on modern Chinese society have acknowledged this exploiter class, which allies bureaucrats and capitalists. I have demonstrated such research in the last section, in order to clarify the social class of Internet workers. Here, I continue my argument from the last section, based on Wright’s and Callinicos’ work, recognising the social mode of production in modern China as bureaucratic state capitalism or bureaucratic state socialism. The bureaucratic capitalist class, which I defined in section five, occupies the location of exploiter class, with ownership of the means of production, organisational assets, and political authority. This class accumulates huge wealth by controlling labour power and the skills of the middle class and the working class. This activity of appropriation then generates inequality and injustice between the bureaucratic capitalist class, the middle class, and the working class.

In the Chinese context, the working class sells labour power in order to survive, as their livelihoods are not guaranteed by society. The bureaucratic capitalist class owns the means of production, such as factories/firms, raw materials, and telecommunication, and has the political authority to allocate these means of production. For example, executives in the party-controlled enterprises (the SOEs in contemporary Chinese society) and government departments own the main raw materials and economic resources, such as oil and telecommunication. Officials in the bureaucratic capitalist class with certain political power (similar to Wright’s state power) and capitalists with certain economic power allocate these raw materials and economic resources. But what does the middle class own, and what are the relationships between the middle class and these other two classes?

Internet workers, as members of Chinese lower-middle class, own certain means of production, such as professional and technical skills, but this ownership is only helpful when they place it in the capitalists’ service. According to Wright (1997, 19), this is the general problem of the middle class, who sell their labour power as they lack the means of production, while they do not regard themselves as the working class. Under the movement towards globalisation, precarious and uncertain work and life status are shared by workers in different social contexts, both in socialism and capitalism. Workers in Western societies share the severe economic pressure faced by Chinese Internet workers, which I showed earlier, as it becomes a general problem of the middle class, to use Wright’s terms.

In the Chinese context, the bureaucratic capitalist class dominates the working class and the middle class, because of the ownership of the means of production, and the power to allocate these resources. It is the bureaucratic capitalist class, where officials and capitalists gain benefits from corruption and bribery, which appropriate the labour of other classes. Due to this, the wealth of the bureaucratic capitalist class is based on the labour efforts of the working class, who contribute labour power, and the middle class, who contribute skills.

Due to the inferior position in the middle class, Internet workers lack the power to allocate the resource they own, such as skills; instead, their labour efforts are appropriated by the bureaucratic capitalist class via long working hours, unequal pay, insecure work status, and unguaranteed pension fund. Such structural activity of appropriation, exploitation in Wright’s explanation, thus, becomes the force behind Internet workers’ poor working-life experiences. Here, exploitation becomes a mechanism to explain why Internet workers suffer bad working conditions, it is worth asking: Why do Internet workers still work in the industries, despite inequalities resulted by the exploitation?

7. There is still hope: agency of Internet workers

As I stated in last sections, Internet workers experience inequalities and injustice, such as unequal pays with long working hours, insecure work status, and unguaranteed life after retiring, due to the structural exploitation. But, this does not indicate that Internet workers accept such difficult working conditions, due to certain level of autonomy given by companies, without any acts of agency. Rather, Internet workers use various acts of negotiation and resistance to improve the quality of working life.

For example, some workers regard intervention from the state as influencing their practices and creativity in an unacceptable way:

We [workers] definitely don’t like the rule. We prefer to stand with users, who could bring us money. But, as the state could easily stop our service, we still need to follow the rules in certain ways […] (Galeno, technical worker in the Product Administration Department at Campus, 20^th December 2011, interview)

Here, Galeno shows the necessity of balancing the state’s requirement and Internet users’ needs for free space in his daily practices; moreover, he states that some workers feel unsatisfied with the state’s intervention. Thus, these workers, who are unsatisfied with such intervention, apply their professional knowledge and skills to acts of negotiation and resistance, in order to gain more autonomy.

For example, In an Election Meeting of Candidates of the National People’s Congress in Shanghai, which took place during my fieldwork, representatives of workers in Grand expressed discontent about current working conditions, and made a case for more work-related benefits. They asked the local government to build a new kindergarten near the company in order to benefit the workers with children. They also raised the issue of overtime work in the industries. Moreover, these representatives questioned whether their voices could be heard by the departments responsible for bringing in changes via the Congress system. Since I left Grand two months after this event, I do not know the result of such bargaining. But the voices of the workers in this event indicate Internet workers’ special forms of response towards state control and intervention. It is rare to hear the voices of other workers in the Chinese context, such as SOE workers.

This form of negotiation may not be so unusual in other geographical contexts, such as the UK, but such direct questioning of authority can hardly be found amongst workers in other industries in China. Internet workers’ direct expression of discontent and their questioning of authority mark a fundamental shift in attitudes towards worker agency in the authoritarian Chinese context.

Here, I acknowledge workers’ subjectivity and agency in such events. As Burawoy and Wright (2002, 474–475) claim, the exploited classes tend to resist the appropriation of their labour efforts. The exploited class–the lower-middle class–has the potential to resist the exploiter class–the bureaucratic capitalist class–via acts which attempt to eliminate inequalities and injustice in their working life. This then creates potential for changing working conditions in the Internet industries in the future. I regard this as a sign of hope in contemporary Chinese Internet industries.

Notes

¹Also present in the Chinese social hierarchy in the post-1992 period were various other classes, such as the petty bourgeoisie and small employers.

References

Andreas, Joel. 2010. A Shanghai Model? On Capitalism with Chinese Characteristics. New Left Review. 63–85.

Barbrook, Richard. 2005. The High-Tech Gift Economy. First Monday. Special issue 3: Internet banking, e-money, and Internet gift economies.

Breen, Richard. 2005. Foundations of a Neo-Weberian Class Analysis. In Approaches to Class Analysis, edited by Wright Eric Olin, 31–50. Cambridge: Cambridge University Press.

Burawoy, Michael and Eric Olin Wright. 2002. Sociological Marxism. In Handbook of Sociological Theory, edited by Turner Jonathan, 459–486. New York: Plenum Books.

Callinicos, Alex. 1983. The Revolutionary Ideas of Karl Marx. London: Bookmarks.

Callinicos, Alex. 2000. Equality. Cambridge, Oxford, and Malden: Polity Press and Blackwell Publisher.

Callinicos, Alex. 2004. Making History: Agency, Structure, and Change in Social Theory. Leiden: Koninklijke Brill NV.

Cohen, Gerald Allan. 1995. Self-Ownership, Freedom, and Equality. Cambridge: Cambridge University Press.

Dickson, Bruce. 2003. Red Capitalists in China: The Party, Private Entrepreneurs, and Prospects for Political Change. Cambridge: Cambridge University Press.

Dickson, Bruce 2011. Who Consents to the ‘Beijing Consensus’: Crony Communism in China. In In search of China's Development Model: Beyond the Beijing Consensus, edited by Hsu Philip, Wu Yu-Shan and Zhao Suisheng, 24–36. London: Routledge.

Dyer-Witheford, Nick. 2010. Digital Labour, Species-Becoming and the Global Worker. Ephemera 10 (3/4): 484–503.

Gill, Rosalind. 2002. Cool, Creative and Egalitarian? Exploring Gender in Project-Based New Media Work in Europe. Information, Communication and Society 5 (1): 70–89.

Harvey, David. 2008. The right to the city. New Left Review Sep-Oct: 23-40.

Hesmondhalgh, David and Sarah Baker. 2010. Creative Labour: Media Work in Three Cultural Industries. Abingdon and New York: Routledge.

Hills, Matt. 2002. Fan Cultures: Sussex Studies in Culture and Communication. London: Routledge.

Huang, Yasheng. 2008. Capitalism with Chinese Characteristics: Entrepreneur and the State. Cambridge, New York: Cambridge University Press

Kennedy, Helen. 2012. Net Work: Ethics and Values in Web Design. Basingstoke: Palgrave Macmillan

Manzerolle, Vincent. 2010. Mobilizing The Audience Commodity: Digital Labour In A Wireless World. Ephemera 10 (3/4): 455–469.

Mok, Ka Ho and Yat Wai Lo. 2007. The Impacts of Neo-Liberalism on China's Higher Education. Journal for Critical Education Policy Studies 5 (1): 316–348.

Noon, Mike and Paul Blyton. 2007. The Realities of Work: Experiencing Work and Employment in Contemporary Society. Houndmills, Basingstoke: Palgrave Macmillan.

Ornebring, Henrik. 2008. The Consumer as Producer-of What? User-Generated Tabloid Content in the Sun (UK) and Aftonbladet (Sweden). Journalism Studies 9 (5): 771–785.

Roemer, John E. 1982. A General Theory of Exploitation and Class. Cambridge, Massachusetts, and London: Harvard University Press.

Ross, Andrew. 2005. Fast Boat to China: High-Tech Outsourcing And The Consequences Of Free Trade: Lessons From Shanghai. New York: Vintage Books.

Ross, Andrew. 2007. Nice Work if You Can Get It: The Mercurial Career of Creative Industries Policy. In My Creativity Reader: A Critique of Creative Industries, edited by Lovink Geert and Rossiter Ned. Amsterdam: Institute of Network Cultures.

So, Alvin. 2003. The Changing Pattern Of Classes And Class Conflict In China. Journal of Contemporary Asia 33 (3): 363-376.

Terranova, Tiziana. 2004. Network Culture: Politics For The Information Age. London: Pluto Press.

Wang, Hui. 2003. China’s New Order: Society, Politics, and Economy in Transition. Cambridge, Massachusetts, and London: Harvard University Press.

Wang, Jing. 2008. Brand New China. London: Harvard University Press.

Wright, Eric Olin. 1976. Class Boundaries In Advanced Capitalist Societies. New Left Review 98: 3-41.

Wright, Eric Olin. 1985. Classes. London: Verso.

Wright, Eric Olin. 1996. The Continuing Relevance Of Class Analysis—Comments. Theory and Society 25: 693–716.

Wright, Eric Olin. 1997. Class Counts: Comparative Studies in Class Analysis. Cambridge: Cambridge University Press.

Wright, Eric Olin. 2002. The Shadow Of Exploitation In Weber’s Class Analysis. American Sociological Review 67: 832–853.

Wright, Eric Olin. 2006. Compass Points: Towards A Socialist Alternative. New Left Review 41: 93–124.

Wright, Eric Olin. 2009. Understanding Class. New Left Review 60: 101–116.

Wright, Teresa. 2010. Accepting Authoritarianism: State-Society Relations In China’s Reform Era. Stanford, California: Stanford University Press.

Zhao, Yuezhi. 2003. Transnational Capital, The Chinese State, and China's Communication Industries in a Fractured Society. The Public 4: 53-74.

Zhao, Yuezhi. 2007. After Mobile Phones, What? Re-Embedding The Social In China's 'Digital Revolution'. International Journal of Communication 1: 92–120.

Zhou, Xiaohong. 2008. Chinese middle class: reality or illusion. In: Christophe, J. & Van der veer, P. ed. Patterns of Middle Class Consumption in India and China. Los Angeles: Sage

Websites

Chinese Economics. 2013. Shouyilv Nan Di Tongzhang, Yanglaojin Ruhe Baozhi Zengzhi. Accessed July 29, 2013. http://finance.chinanews.com/cj/2013/07-09/5017318.shtml.

GhRlib. 2013. Nian Hulianwang Hangye Xinchou Zengzhang Yuce. Accessed April 14, 2014. http://www.yixieshi.com/zhichang/1547.

Guangzhou Daily. 2013. Jinnian Yingjiesheng Qixin Gao, Neidi Wuda IT Qiye Nianxin 10wanyuan Qibu. Accessed April 14, 2014. http://finance.ifeng.com/a/20131209/11234834_0.shtml.

iResearch Inc. 2013. iResearch China Internet Economy Research Report 2011–2012. Accessed June 11, 2013. http://www.iresearch.com.cn.

Liaoning Research Institute of Industry and Information Sciences. 2013. Zhongguo IT Keji Renyuan Keyan Chengguo Zhuanhua Fenxi. Accessed June 10, 2013. http://www.lniis.gov.cn/dwdt/ShowArticle.asp?ArticleID=6921.

Oriental Outlook. 2013. Hulianwang Qiye Dangjian Diaocha. Accessed April 14, 2014. http://finance.sina.com.cn/roll/20130225/170114639663.shtml.

Sina News. 2010. Tengxun 37sui Zhuanti Zhubian Yu Shihong Cusi. Accessed January11, 2014. http://eladies.sina.com.cn/qg/2010/0521/1631993336.shtml.

Sohu IT. 2013. Zhongguo Hulianwang Yifenzhongnei Hui Fasheng Shenme? Accessed January 14, 2014. http://it.sohu.com/20130322/n369907395.shtml.

Tencent Technology. 2011. IT Chanye Cheng Wu Da Jibing Zhong Zaiqu: Guolaosi Pingjun 37.9 Sui. Accessed July 19, 2012. http://tech.qq.com/a/20111124/000177.htm.

The Boston Consulting Group. 2012. Clicks Grow Like BRICS: G-20 Internet Economy to Expand at 10 Percent a Year Through 2016. Accessed June 16, 2014. http://www.bcg.com/media/PressReleaseDetails.aspx?id=tcm:12-100468.

Xinhua News. 2005. Li Yanhong Zai Meiguo Chuangzao Shenhua: Baidu 400 Lao Yuangong Yi Ye Bao Fu. Accessed September 23, 2013. http://news.xinhuanet.com/fortune/2005-08/08/content_3323159.htm

Xinhua News. 2013. Xinhua Insight: Internet Industry Optimism To Buoy China's Economy? Accessed October 10, 2013. http://news.xinhuanet.com/english/indepth/2013-08/17/c_132638709.htm.

Youth Times. 2012. Wei He Shuo 'Yanchi Tuixiu' Bu Ke Xing. Accessed October 29, 2012. http://www.qnsb.com/fzepaper/site1/qnsb/html/2012-06/11/content_373983.htm.

About the Author

Bingqing Xia
Bingqing Xia currently is a PhD candidate in the Institute of Communications Studies at University of Leeds, UK. Her research focuses on labour and the quality of working life in the Chinese Internet industries. Bingqing's background is in history and new media studies. She has presented her work at a range of international, annual and biannual conferences such as the Association of Internet Researchers, the International Association for Media and Communication Research, and the Crossroads Cultural Studies conference.