Precarious labour in waiting: Internships in the Chinese Internet industries

Abstract

Internships are prevalent in new media industries and have become the focus of news reports and popular writings in recent years. This article addresses the gap between intern research and digital labour research. It asks: How are these interns working and living? What are the power dynamics behind interns’ experiences? To what extent can digital labour theories be applied to explore these experiences? Based on empirical research conducted at two Chinese Internet companies, this article shows that interns in Chinese Internet industries experience poor working conditions and difficult living conditions. These are caused by power dynamics within the companies, such as tensions between interns and full-time Internet workers, and power dynamics within Chinese society, such as those between Chinese universities and Internet companies involved in these internships. The article argues that such difficult conditions are caused by Internet companies and the Chinese higher education system, both of which engage in forms of coercion and alienation. Digital labour theories need to take greater account of intern labour and of interns’ experiences of precarious work in the new media industries.

Introduction: Internships and digital labour research

After joining my current university at Shanghai, in early 2018, I built up a new module called Working in New Media. The original idea behind setting up this module was to introduce digital labour theories to undergraduates. I tried to lead students to critically explore work in the new media era by showing different types of digital labour in new media industries. However, in the seminars, the students showed me a type of digital labour more relevant to them: internships in new media industries. In a group activity, I asked them to undertake a survey on working hours and payment in the Internet content production industry. Contrary to expectation, each group focused on internships and produced detailed descriptions of the experiences of interns in well-known Internet companies, many of the students having had internship experiences in these companies themselves.

This situation reminded me of an incident I had observed when I came to Shanghai in 2011 for fieldwork related to my PhD thesis: a well-known Internet company in Shanghai hired a large number of interns to replace full-time employees, in order to reduce labour costs. Owing to pressure from their universities and the need to make themselves attractive to future employers, interns had been obliged to suffer through low-paid, low-security, high-pressure work.

The issue of internships has become a hot topic in Hollywood movies in recent years. Examples include The Devil Wears Prada in 2006, The Internship in 2013, and The Intern in 2015. The topic has also aroused extensive discussion in academia since Ross Perlin’s (2011) publication of How to Earn Nothing and Learn Little in the Brave New Economy. For this book, Perlin conducted a 4-year survey on the basis of which he claimed that internships are increasingly becoming an important part of university students’ lives in the USA and beyond. He focused on the impact of internships on young people’s daily lives, as well as social inequalities caused by internships.

TripleC: Communication, Capitalism & Critique, an important journal in the field of critical digital media studies, also published a special issue on internships in 2015. Topics included a clarification of the concept of internships (Corrigan, 2015; Frenette, 2015; Hope and Figiel, 2015), the relationship between the internship system and the development of cultural and creative industries (Boulton, 2015; Ciccarelli, 2015; Mirrlees, 2015), and the relationship between internships and higher education (Chong, 2015; Einstein, 2015; Smeltzer, 2015), as well as labour movements (Cohen and Peuter, 2015; Webb, 2015). Within the same special issue, Rodino-Colocino and Berberick (2015) highlighted the role higher education institutions play in students’ decisions to take internships, colleges and universities encouraging students to do so by offering them credit-based internship programmes. In all these ways, internships can be seen to blur the line between students and workers.

Such research links internships to discussions of the nature of digital labour and its impacts on participants. Indeed, digital labour research nowadays has two main focuses: the exploitation of professional labour (Gill, 2002; Hesmondhalgh, 2010; Kennedy, 2012; Qiu, 2009; Ross, 2008; Xia, 2014) and the exploitation of audience-labour (Barbrook, 1998; Fuchs, 2015; Hills, 2002; Jenkins, 2006; Van Den Broek, 2010). The former focuses on the exploitation of different forms of professional workers in different media industries, while the latter focuses on the exploitation of audience-labour in terms of time spent and data/information produced when users are online, such as Facebook users.

The discussion on audience-labour originates in Dallas Smythe’s seminal article Communications: Blindspot of Western Marxism (Smythe, 1977) with a focus on the audience commodity. Possibly, the work on free labour and immaterial labour by the Italian Autonomist Marxists, such as Michael Hardt and Antonio Negri (2001, 2004), are further antecedents to the concept. In the mid-2000s, the concept appears in Barbrook’s (1998) discussion of the gift economy and Terranova’s (2004) development of free labour theory. Both formally include audiences in their examination of digital labour.

More recently, researchers such as Graham et al. (2017) have developed the theorisation of audience-labour by focusing on digital labour working on platforms that support transnational workflows. Similarly, these writers explore outsourcing work in digital connective markets. They regard the rise of digital labour as a result of regional concerns towards un- and under-employment as well as rapidly changing connectivity. They explore digital labourers’ complicated experiences working on platforms in economic margins via four interests: bargaining power, economic inclusion, intermediated value chains, and upgrading. They argue that platform digital labour is exploited and solidarity among these workers becomes difficult, due to the dispersed geography (Graham et al., 2017: 153). As platforms are owned and run by unaccountable hands, they are out of the control of national governments, and so workers’ benefits, such as minimum wages, and taxes are able to be avoided (Graham et al., 2017).

Following this work on the nature and impact of digital labour, the costs and benefits on internship in media industries are also now receiving attention. Christian Fuchs, writing from a critical Marxist approach, has been working in recent years on the efforts of non-professional workers and professional digital labourers to organise and resist capitalist exploitation. Fuchs and Sevignani (2013) define digital work as

… a specific form of informational work that makes use of digital media as an instrument of work that is employed together with the human brain to organise human experiences in such a way that symbolic representations, social relations, artefacts, social systems and communities emerge as new qualities. (p. 257)

In particular, they argue that Facebook exercises ‘a social form of coercion that threatens the user with isolation and social disadvantages’ (Fuchs and Sevignani, 2013: 257) in order to exploit his or her everyday activities. They also recognise that ICTs (information and communication technologies) today create a plenitude of exploited labour, ranging from mineral workers in Africa that contribute to producing hardware, industrial workers in China who assemble hardware tools, as well as low-paid software engineers in developing countries (Fuchs and Sevignani, 2013: 263–264).

In comparison, the discussion on professional digital labour centres on the concepts of the creative class (Florida, 2002), knowledge labour (McKercher and Mosco, 2008) and creative labour (Hesmondhalgh, 2010). All these concepts criticise capitalist exploitation of digital labour, although with different definitions. In particular, Hesmondhalgh (2010) argues that the current debates on free labour and exploitation are unconvincing and fail to explore the potential agency of professional labourers. He further questions the extent to which the existing pairing of the concept of free labour with exploitation is coherent – is capital accumulation based on Internet users’ free time and data really the most important concern in digital labour research? Instead, he argues that the internship system in current cultural industries is one the most significant issues in any analysis of free or unpaid labour.

Hesmondhalgh (2010) points out that unpaid workers, such as interns, are a source of technical and economic rent for media companies. This unpaid system not only helps media companies save costs on training and mentoring inexperienced workers, but also depresses wages for professional workers. Furthermore, he argues that unpaid internships increase inequality: young people from poor families are less likely to be able to enter the media industries. Third, he points out that internships are increasingly provided as part of college and university degrees. This certainly encourages young people to carry out such unpaid work. He reminds us from a social development perspective that the increase in unpaid internships is at the expense of time that ‘young people might be spending exploring ideas and broadening their intellectual horizons’ (Hesmondhalgh, 2010: 279). The unpaid internship thus erodes the benefits of education for social development.

Corrigan (2015) claims that some concepts in digital labour research, such as self-exploitation, could be applied to understand interns’ experiences. He applies digital labour theories to internships in creative industries, but retains an appreciation for both the meaningful part of such labour (e.g. work satisfaction and skill acquisition for a future career), and the problematic part (e.g. exploitation and self-exploitation). He takes work experiences and exposure in the industries as a form of coercion to force interns to do unpaid or underpaid work. Meanwhile, he points out interns’ limited power to control meaningful and hands-on experiences that they value for future work functions as an alienation experience, which we must appreciate in order to understand issues of exploitation and self-exploitation. Again then, Corrigan’s work indicates the value of exploring interns’ experiences in media and creative industries using digital labour theories.

Finally, in the context of China, Brown and deCant (2014) claim that industrial interns have become cheap labour deployed in Chinese industries. It is not uncommon to find large numbers of interns performing regular factory workers’ work for subminimum wages across the whole of China. Smith and Chan (2015) argue that internships in China only involve simple work experience irrelevant to the students’ academic specialisation, constituting an unimportant pre-employment experience that is nevertheless necessary to complete one’s education.

The main discussion about internships in China focuses on Foxconn, the world’s biggest maker of electronics and a major supplier to Apple and other companies (Chan and Pun, 2010; Perlin, 2011; Pun and Chan, 2013; Su, 2011). However, little academic attention has been paid to internships in the Chinese Internet content industry, at companies such as Facebook and YouTube, and Chinese equivalents like Weibo and Youku-Tudou, companies that accumulate massive economic and cultural capital by providing online content.

The Chinese Internet content industry has developed exponentially since the end of 2002 when the market was revived from the dot-com crash in 2001. In 2012, the annual market value had reached 385.04 billion RMB (about USD57.37 billion), an increase of 54.1% from 2011 (iResearch Inc., 2013). The capital accumulation enabled by this growth largely relies on the labour efforts of Internet workers, and huge numbers of interns are involved in the production process. The number of Chinese Internet workers increased to 12.3 million by the end of 2009 (iResearch Inc., 2013). Despite these staggering numbers, little academic attention has been paid to these workers, in particular to interns.

As mentioned above, nearly all students in my two undergraduate modules have had internship experiences in the Internet content industry, such as at public WeChat platforms, small apps start-ups, and social media departments in media companies. According to what they have told me about their experiences in the industry, most of the problems they are experiencing in 2018 are similar to those I observed in my fieldwork in 2011. In this article, I therefore critically interpret interns’ experiences in the Internet content industry based on materials I collected for my PhD fieldwork in 2011. In doing so, I will answer the following questions: How are interns in the Chinese Internet content industry working and living? What are the power dynamics behind interns’ experiences? To what extent can digital labour theories be applied to explore these experiences? Before addressing the interns’ subjective experiences, I will explain the methodology adopted in this research.

Methodology

I carried out empirical, at times ethnographic, research in two Chinese Internet companies, in order to study the workers in this article. I will call the first company S. S focuses on online entertainment, such as online gaming and online fiction. I will call the second company X. This company provides social networking services. I used observation and in-depth interviews as my primary methodology.

I conducted in-depth interviews in three periods at X: seven interviews in February 2010 (a very early stage of my research, and I do not directly use this part of the data here, as they are not related to internships); nine interviews in August 2011; and five interviews in December 2011. In addition, I spent 3 months at S conducting participant observation, where I worked as an intern to observe and kept a journal about workers’ daily practices. I also invited one worker at X, who I will call 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 August–December 2011. Finally, thanks to his own habit of keeping work journals, Galeno gave me his work journals from September 2009–December 2011. Through these mixed, qualitative methods, I explored how interns get involved in digital production in the Chinese Internet content industry, which I will discuss in the next section.

My participant observation at S was covert, for a number of reasons. First, 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. Second, this article develops from my PhD project, which not only focuses on the quality of working life but also emphasises the involvement of worker agencies. Using a covert method, I felt that I would be able to witness more ‘genuine’ acts of worker agencies, 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 role played by Internet workers in that society. In order to protect all participants, I anonymised both the companies and participants here. As all my participants have left these two companies and most of them have even left the industry, I believe the data used in this article will not harm their safety in work.

Most personal narratives quoted here indeed were from my observation journal, a summary of daily activities and conversations with all participants during the period August–December 2011. I blurred participants’ information in all personal narratives used in this article to minimise any possible harm to them. However, I received permission from all interviewees to use their data in academic-related publications via information sheets and participant consent forms.

‘Tears of blood’ in the Chinese Internet content industries; tears as well as joy?

Interns in the Internet industries are mostly treated as full-time workers in terms of their working hours. Brown and deCant (2014: 161) illustrate that interns at the Wintek Corporation (a company supplying products to Apple and Nokia) usually work 11 hours per day, 7 days a week. At S, interns are required to clock in and out, just as full-time workers are required to do. Interns in both companies rarely have the opportunity to decide their working hours. Most interns follow the working pattern of full-time workers, and some need to work overtime, again, like the full-time workers do. Comments pertaining to work hours by intern-participants included:

I had interned here [at X] for five months before I was formally employed. All new employees, who are recruited before graduation, are required to do internships, before they formally join the company … My team leader wanted me to do more work, so he asked me to do [an] internship earlier than [the] others … (Galeno, technical worker at X, 24 August 2011, interview)

… I did more work when I was an intern than now. During my internship, we three interns were required to carry out seven full-time workers’ work … (William, technical worker at X, 26 August 2011, interview)

Sometimes, interns seemed to be able to have more flexible working hours than full-time workers, because they needed time to focus on their studies. But this time off work was considered holidays, and holidays needed to be approved by their team leaders who usually encourage interns to focus on their work. For example, Janet, an intern at S, was struggling to write up her MA dissertation, find a formal job, and conduct her intern work near the end of her internship at S. Therefore, she decided to quit the internship. But her application to do so was rejected, as her work could not be suspended because of her imminent departure. It was suggested that she find a new intern to take over her work, if she really wanted to leave. Here, the pressing issue is that interns are forced to take over the company’s responsibility of finding new interns to fill vacancies. This indicates that interns indeed do not have ‘flexible working hours’: interns do not even have the right to quit their internships. Corrigan (2015: 345) suggests that interns’ experiences of menial work and indifferent/unwilling supervisors are not only a result of coercion, but are also alienating experiences. In the last example cited, Janet’s loss of control over her work also addresses two criteria of exploitation: coercion into underpaid work and alienation from her control of work.

Despite its clear importance, interns’ heavy work appears never to be rewarded equally with that of employees. Smith and Chan (2015: 312) point out that interns at Foxconn have the same starting wage of 950 RMB (about USD141) per month as new full-time workers; however, interns are not entitled to a skill subsidy of 400 RMB (about USD60) per month. At both S and X, interns’ pay was quite low compared to that of full-time workers. At S, interns were paid 120 RMB (about USD18) per day, no matter how many additional hours they worked overtime, compared to full-time workers’ 10,000 RMB (about USD1490) monthly salary. At X, the disparity was equally great:

Most interns [at X], who are confirmed to be full-time workers after graduation, are paid 2,000RMB (about US$298) per month … (Carl, technical worker at X, 27 August 2011, interview)

Salaries for interns were less than half that of full-time workers, although interns mainly did the same work as full-time workers. Both companies recruited interns as a cheap and easily manageable alternative to full-time workers. For example, Janet told me that her team needed to recruit some Spanish and Arabic translators for the technical departments. But, after negotiating with some interviewees, the director decided to recruit interns who majored in Spanish and Arabic to carry out the work, because the salaries of full-time translators were too high for their budget. Interns in this case were regarded as an alternative to full-time workers because they were perceived as able to do the same work.

As another example, when I contacted my colleague at S 1 month after I had left, she told me that 30 out of the 40 full-time workers in the HR department had been laid off, including herself. In contrast, all the interns were kept on, replacing these full-time workers. In other words, the company used these cheap interns as replacements for the full-time workers in order to cut back on labour costs. As Perlin (2011) points out, interns become a dream solution for employers to ‘test drive young workers for little or no cost’ (p. 29).

Beyond the disparities in remuneration, interns’ heavy workload also did not guarantee them the same status in companies as full-time workers. Instead, interns were regarded as an inferior group:

Most interns are following our [full-time workers’] steps. They are doing the jobs we assigned to them… They are a separated group from us… They are still students rather than professionals… They are the second-class workers… (Galeno, technical worker at X, 20 December 2011, interview)

According to full-time workers, they were the people who control the production process, whereas the interns were in an inferior position regarding work. This indicates the tension between these two groups. Some examples below will show the struggles interns go through owing to their perceived inferior positions.

Sometimes, interns were required because of their inferior positions to take over the responsibilities of full-time workers. For example, Janet was blamed by her team leader when a new employee could not find anyone in her team to begin his commencement procedures when she was on leave. Janet argued with her leader that it was her holiday and that she was not supposed to be responsible for the work. But she was still blamed because she was the only intern involved in the work. As Janet said, every time there was a mistake, it was the interns’ fault.

Some of my observations made at S also suggested further social differences between interns and full-time workers. For example, I observed that interns, ordinary full-time workers and team leaders all had their own dining areas. Team leaders usually drove together to good restaurants in the downtown area of Shanghai, full-time workers usually went to the canteen in the centre of the Software Park, and interns usually went to cheaper snack bars near underground stations. Most members were conscious of belonging to these different groups – it was rare to find ordinary full-time workers going with team leaders for lunch, for example.

In addition to the difficult working conditions caused by their inferior positions, interns experienced difficulties in gaining full-time positions after finishing their internships. Sometimes, having a full-time position after finishing the internship indicates certain characteristics of Chinese culture, such as guanxi, the Chinese term for personalised networks. Bian (2002: 107–108) claims that guanxi facilitates all aspects of occupations in the Chinese context, from entry into a job to internal mobility at work. He argues that guanxi benefits guanxi users by promoting job opportunities to them, while constraining people who are ‘poorly positioned in the networks of social relationships’ (Bian, 2002: 107). Gold et al. (2002: 3–20) argue that guanxi is an essential component of Chinese culture, which relates to some important issues in Chinese society, such as ‘ganqing (sentiment), renqing (human feelings), mianzi (face), and bao (reciprocity)’ (p. 4). It is widely believed in China that it is necessary to have good guanxi with executives in order to be promoted in a company.

In the Internet content industry, the necessity of guanxi was also born out within certain practices, among them how interns obtained full-time positions at S. Janet, Lily, Cathy and Shelly, four interns at S, all wanted to be recruited as full-time workers after graduation, but only Cathy and Shelly were recruited after four rounds of interviews, even though Janet had received the highest score. Cathy was even rejected in the first round, but was subsequently aided by her team leader who negotiated with executives for her to get the job. This was not only because of Cathy’s good relationship with her team leader, but also because the leader, who had just joined the department, needed someone who would be loyal to her. Shelly was rejected in the third round and was then also helped by her team leader because one of the executives was interested in hiring her – the team leader wanted to please the executive by helping Shelly.

This example shows how guanxi shapes interns’ working life. It shows that workers with strong guanxi have better opportunities than workers with weak guanxi, no matter what skills and motivation the latter may have. Guanxi here refers not only to relationships with executives but also to workplace politics: Cathy and Shelly got the job because of their ability to help the manager ensure his or her personal interests.

Such difficult working experiences thus raise a question: Why do large numbers of university students still pour into the intern market when they have inferior positions there and are required to work as hard as full-time workers? This might relate to the high competition among fresh graduates. As I argued in an earlier article (Xia, 2018), the large number of low-quality university graduates, caused by the problematic education system, has resulted in employment difficulties. Thus, university students are forced to enter the internship market to gain working experience before entering the job market. Below, I analyse the power dynamics that result in students’ difficult experiences in their internships.

Power dynamics behind the internship system: A form of coercion

Qiu (2009) focuses on the recent reform of the education system in the 1980s when the CCP (the Communist Party of China) further extended national compulsory education to ‘peasants and the proletariat working class’ (Qiu, 2009: 135). In particular, in 1999, the CCP issued two policies: ‘The Action Plan for Education Development in the Forthcoming New Century’ (Mianxiang 21 ^st shiji jiaoyu zhenxing jihua) and the ‘Decision of Deepening Educational Reform and Promoting Quality-oriented Education’ (Guanyu shenhua jiaoyu gaige quanmian tuijin suzhi jiaoyu de jueding), aimed at expanding college education. This education reform caused some problems: to some extent, university degrees were devalued because of the large number of students graduating; employment difficulties followed on from the large number of graduates and devaluation of university degrees; no-pay internships became pervasive because of high competition in the job market; and rural students struggled because of increasing tuition fees.

Figure 1 shows an increase in the higher education population from 1996 to 2016. For example, the proportion of acceptances rose from 40% in 1996 to 82.15% in 2016 (see Figure 1), meaning that more than 80% of students who took the college entrance exam were guaranteed places in colleges in 2016. This would not be problematic if quality remained high, but the problem was that the high quantity of university students has not necessarily guaranteed a higher quality of higher education.

Figure 1. University and technical college entrants, China, 1996–2016.

Source: Qinxue Education (2018).

At the time of writing, there is a widely popular understanding of higher education in Chinese society: if education can be conceived as a train to the job market, then there is standing room for junior college graduates, there are seats for undergraduates, hard berths for masters graduates, and soft berths for PhD students. However, when the train arrives at the labour market, all students are cheap labour. According to this understanding, there is no difference between various education degrees, because all graduates, no matter their level of education, only qualify as cheap labour (Sohu News, 2013).

The postgraduate programme is a good case in point to understand the devaluation of higher education degrees. In 2004, the top three universities in China, Peking University, Tsinghua University and Zhejiang University, claimed that the number of postgraduates (master and PhD students) was larger than the number of undergraduates (China Youth Daily, 2012). The expansion of postgraduate programmes, then, required more supervisors to be recruited, because of the one-to-one teaching model. But few supervisors were newly recruited to these three universities to match their largely increased postgraduate numbers. Instead, supervisors were encouraged to supervise more postgraduates than they were able to. The result of the education reform is that students have had more chances to access higher education, but the most important part of the education, the quality, has been reduced.

Employment difficulties resulting from the increased numbers of graduates were reflected in my research, for example, by the experience of Janet. In one of her hundreds of job interviews, 7 out of 10 interviewees had master’s degrees. Janet’s university has been held in high esteem by many employers in Shanghai over the previous few years; however, it became quite hard for its graduates to find jobs in 2011, as the university had expanded its undergraduate and postgraduate programmes and enrolments in 2008 and 2009. Such difficulties in employment were also shared by graduates of other major universities. According to a member of staff at the Employment Guidance Centre at Janet’s university, few of the university’s students were able to secure jobs in 2011, and their average monthly salary had been reduced to 2659 RMB (about USD397) after taxes, even lower than a migrant peasant’s salary.

Nevertheless, despite the employment difficulties reported by university graduates, the official statistics regarding graduate employment appear quite encouraging: the employment rate of university graduates in 2011 was 90.2% (China News, 2012). Indeed, there are two issues with the state’s ‘active policies’ contributing to this encouraging number: universities’ activities regarding beijiuye, a term I explain below, and the alliance between employment rates and graduation rates.

In 2009, a new term, beijiuye, meaning ‘to be worked’ became popular among university students. The term refers to the situation in which ‘a person is given a job without his or her knowledge’ or ‘a person is given a job that may not exist at all’. The term was used by some graduates to criticise their universities who falsified work contracts for students without informing them (ifeng.com, 2010).

Accurately or not, it was believed among some students in my field study that universities were adjusting the numbers of graduating students in response to the employment conditions in each given year, and this was seen as a way to increase the official employment rate. There was a common belief among the interns at S, whom I studied that it was harder to graduate in 2011 than before because of the stricter standards for dissertations and vivas. During my time at S, Janet received notification of updated dissertation standards, indicating that a dissertation with 20 misused punctuation marks would be failed. Shelly, another intern at S, stated that it was quite common for all universities to stop too many fresh graduates from pouring into the job market by making it more difficult for them to graduate.

However, the large number of graduates in 2011 was the result of the temporary solution used to solve the employment difficulties in 2008: postgraduate programmes, which had been 2-year courses before 2008, had been extended to 2½ and 3 years in most universities because of the employment difficulties that year. Lily, an intern at S, accepted a 2-year postgraduate offer in 2008, but on the day she started her course, she was informed that she had to spend 1 further year at university. The expansion of postgraduate programmes and the extension of courses in 2008 led to the pipeline effect of increasing graduate numbers in 2011, which in turn created new employment difficulties. The connection between employment difficulties and the expansion of postgraduate programmes then became a vicious circle. Ironically, each consecutive year after 2011 has been called ‘the most difficult year of employment for university graduates’ (China Education, 2014; ifeng.com, 2013).

Put simply, the expanded requirement for internships ‘to be worked’ and the linking of graduation rates to employment opportunity rates can be seen as means whereby universities helped mask employment difficulties rather than solve them. This placed back on students the responsibility of solving employment access difficulties via undertaking multiple internships. An unstable future awaits them after graduation, even following completion of multiple internships. As Smith and Chan (2015) argue, student interns are a result of ‘the coercive partnership among institutional actors, including the provincial governments, employers and vocational schools’ (p. 306). These authors argue that ‘the organisation of internship and collaboration between institutional actors (firms, local states, vocational schools) ensured that firm interests overruled the interests of education’ (Smith and Chan, 2015: 320).

Su (2011) portrays the close relationship between Chinese technical schools and enterprises using students as cheap labour via internships: it is expected that the state’s combination of learning and working (gong xue jie he) policy will enable technical school students to practice their skills in SOEs (State-Owned Enterprises), which will help them become more experienced. However, technical schools now send their students to enterprises such as Foxconn to work as cheap and flexible labourers, in exchange for firm subsidisation via equipment and funding that will help address institutions’ financial difficulties. The internship has become an important node connecting technical schools, students and enterprises. Smith and Chan (2015: 315–319) further point out that teachers in technical schools play a key role in this process: teachers ensure that entire classes do internships in enterprises and follow factory rules during the internships; they are also responsible for ‘counselling and deflecting students from feeling dejection at their work and employment situation and the risks of resistance and self-harm that could flow from this’ (p. 315).

Arguably, a similar relationship now exists between universities and enterprises, with supervisors in universities play a similar role to that argued by Su in the case of technical school teachers – the commodification of education.

This dynamic can be seen in the experiences of my participants:

Normally, every supervisor has certain vacancies for ‘special postgraduates’, who are good at socialising rather than academic work. These sociable students are necessary for supervisors to negotiate projects with companies or the state. For example, my friend, who is doing an MA at Zhejiang University, told me that one of his classmates was from an unknown university, and had poor academic ability. He was the special student the supervisor chose, because of his ability to socialise. Every time the supervisor needed to negotiate business with companies, he would be the person to help with the socialising. (William, technical worker at X, 19 December 2011, interview)

In this case, the student was chosen and used to help the supervisor’s business with outside firms: such cooperative projects with companies are reportedly quite lucrative. For example, a survey conducted by The Paper (also known as Pengpai in Chinese, the first news app launched in China), received reports from a significant number of masters students in science and engineering who were working in their supervisors’ factories or enterprises as free or low-cost labour, with supervisors benefitting economically from this structure (The Paper, 2016).

In addition, during my field study, I received anecdotal accounts that students majoring in humanities and social sciences were required by their supervisors to write up books for free. For example, Lily, the intern at S who was majoring in communication studies, had been required to write two books for her supervisor to publish (quality was not an issue in the case of compulsory reading texts for the supervisor’s modules, which course participants would be required to purchase). As is common among postgraduates, she received no payment, nor was her name on the publications. Before that, she had also been required to write-up a movie script for her supervisor, which was sold to a film company for 20,000 RMB (about USD2966), although she was paid nothing.

Janet, the intern at S, worked on a design project with her supervisor, which was evaluated at around 5000 RMB (about USD741) by a company, while her work was evaluated by her supervisor at 800 RMB (about USD118), the cost of a new dress. Cathy, another intern at S majoring in communication studies, had been asked by her supervisor to produce some commercial videos for a cooperating company, again without pay. In neither was the student keen to contribute her creativity and labour for free, but both felt obliged to carry out the work because their supervisors could decide whether they would successfully obtain their degree. There were further instances such of value appropriation in my data:

I was recruited by an online game lab which was a joint venture between my university and a Korean online game company. I was not paid as usual … (William, technical worker at X, 26 August 2011, interview)

… I was also forced to conduct projects for my supervisor in my three years’ postgraduate study. To be honest, my supervisor is nicer than others, because sometimes he bought dinner for us as pay for our work. Although we all know that out work helps him earn a lot of money, we have no chance to negotiate with him … I know some PhD students are forced to extend their courses, because their supervisors need them to conduct projects for free … (Lara, non-technical worker at X, 19 December 2011, interview)

Some months ago, we [X] cooperated with the Chinese Academy of Sciences on a search engine project. Some students from the university were sent to the company as interns. But we didn’t pay for them, since their supervisors took charge of it [though we know supervisors do not pay their students]. Personally, I guess, as a common understanding in Chinese universities, these interns are never paid by their supervisors … (Louis, former technical worker at X, 18 December 2011, interview)

In these cases, supervisors and companies cooperated to set up business relationships to make profits. Both supervisors and companies earned money from this for-profit relationship: supervisors received payment from companies and also published papers with data from the projects; and companies obtained economic benefits by profiting from the products of the co-operative projects. The postgraduates, working long hours as interns in the cooperating companies, such as S and X in this research, received no fair reward.

The coercion that the student interns reported corresponds to the discussion of exploitation in the literature on digital labour. For example, Fuchs (2014), citing Marx’s definition of exploitation, points out that coercion can be divided into three forms:

physical violence (such as overseers, security forces, military), structural violence (markets, institutionalised wage labour contracts, legal protection of private property, etc.) and cultural violence (ideologies that present the existing order as the best possible or only possible order and try to defer the causes of societal problems by scapegoating). (Fuchs, 2014: 158)

On this basis, supervisors coercing students to provide cheap or free labour belongs to the category of structural violence. It is the leveraging of supervisors’ power; as the supervisors evaluate students’ academic performance and have a determining influence on students’ research topics, publications and oral defences, they are able to force students to accept unfair experiences and demands. Supervisors are arguably in charge of student interns’ destinies, destinies that students cannot easily escape from, as I claimed above.

This power structure has certainly resulted in many social tragedies in recent years. A postgraduate student at Wuhan University of Technology committed suicide in early 2018, supposedly due to mental abuse by his supervisor. The case addresses concerns about the problematic professor-student power dynamic in the Chinese education system, posing a significant reason as to why we must critically evaluate interns’ experiences in the Internet content industry.

Moreover, as Hesmondhalgh (2010) claims, the internship model closes the door to the media industries for certain groups of people, including young people from poor families. In the context of the Chinese Internet content industry, this inequality between young people from different family backgrounds needs to be understood primarily as an inequality between those from urban and rural areas. Rural students in China pay a higher price for taking on internships: they take on internships as a way to economically support their lives in big cities, even though the pay is very low, as shown above.

In 2011, the education forum on NetEase, one of the main portals in China, edited a special report about the rise of university tuition fees. The report was based on an online survey conducted among its users. The results showed that the new century’s university students suffered from high tuition fees, rapidly growing living expenses and non-guaranteed futures, as compared to university students in the 1980s and 1990s who were guaranteed jobs by the state. In this report, many Internet users from rural families and low-income urban communities who completed BA degrees in the 2000s had difficulties paying the high tuition fees. Families borrowed money from relatives to pay the steep tuition fees, but were unable to pay back the money until the graduates found jobs. This was the experience shared by Janet and her friends who were from rural China.

Olivia, Janet’s friend, was from a rural family in Xi’an, a developing area in China. Before she came to Shanghai, her family borrowed money from all relatives in order to pay the high tuition fees, but she could still not cover the high living expenses in Shanghai. Olivia then spent most of her free time doing internships, regardless of whether they related to her academic specialisation. She did so during her 6½ years of study (4 years of undergraduate education and two and a half years postgraduate) in order to cover her own living costs. As she could not find a job soon after her graduation, owing to the state of the labour market, she did not have any money for subsistence. When Janet told me that Olivia had asked her whether she could share some of the porridge Janet had cooked, because she (Olivia) had no money for dinner, I immediately developed a sense of responsibility to explore the struggles of these rural students in good universities.

If the high tuition fees have frustrated large numbers of rural students, then the employment difficulties in recent years, which I addressed in this section, break these students’ dreams about the future. Suicide seems to be a way for them to escape from the stress caused by companies, universities and the state:

It is understandable that these students choose suicide. Some students are from rural families who are in debt because of high tuition fees. But when they realise how difficult it is to find jobs and even support themselves in big cities, after seven years of hard work, they must feel hopeless. (William, technical worker at X, 19 December 2011, interview)

This makes clear a sense of depression and desperation surrounding university students’ insecure futures. Cedrstrom and Fleming (quoted in King, 2012) argue that workers’ suicide cases are ‘failed escape attempts’. The authors put forth the example of highly paid bankers who dramatically killed themselves – ‘jumping, for instance, from a prestigious restaurant with [a] glass of champagne in hand’ (King, 2012: 455). In contrast, interns do not, and cannot choose to commit suicide in such a ‘romantic’ way. Neither do interns have a way of escaping from the difficult conditions they are suffering under.

This, I would argue, is another form of ‘failed escape attempt’, based on ‘helpless choice’, a choice forced by the state, universities (including supervisors) and companies, who appear to be acting in concert to maximise economic benefits from economically disadvantaged students. These words from Janet’s friend, Olivia, capture the problem:

… I thought I was lucky when I received the offer from the university, since I thought I had found a chance to change my fate. I was quite confident studying, working and settling down here, in the big city, and changing my life via my university degree. But, now, I realise that it is impossible and my dream is broken … (Olivia, a rural student in Shanghai, 14 December 2011, observation journal)

The hopelessness, helplessness and frustration expressed by Janet and her friends is still fresh in my memory when I write up this article 7 years later. These rural students, who already struggled with basic survival, lost even the opportunity to refuse the problematic internship system. Therefore, as scholars, when we critically adopt theoretical frameworks of digital labour to interpret dynamics between capital and labour, is it not our responsibility to include these sacrificing students in digital labour by looking at their experiences of working in the Internet industries? In the end of my module Working in New Media, a student asked me the question, ‘is free labour really free?’. I think this is a question that deserves our attention and one we need to ask ourselves when exploring the issue of internships.

Conclusion: On whom does digital labour research need to focus?

In this article, I answered three questions regarding internships in Chinese Internet content industry pertaining to the working life of interns, the power dynamics behind their working lives, and the extent to which digital labour theories help understand these experiences. By addressing experiences such as underpayment, overwork, inferior positionality in companies, and the difficulty of becoming a full-time worker after graduation, I highlighted a structural coercion caused by education reform and exercised by postgraduates’ supervisors, a coercion based on collusion between universities and companies. All data used in this article were collected in 2011, yet tragedies like postgraduates’ suicide due to overwork and pressure from supervisors have not stopped in the period between then and the publication of this article. Evidence from students in my two current modules suggests that the working lives of interns in new media industries have not improved. I believe it is still necessary to address this problem 8 years after I collected the data reported here.

At the same time, these students cannot easily escape the pressure placed on them by their supervisors to provide unpaid labour, as this would have even more deleterious effects on their future than merely ‘quitting’ their degrees. These experiences suggest that interns suffer alienating losses of control over their work, examples of which include the difficulties of quitting internships, and not controlling working time and workload as a result of inability to refuse work assigned by supervisors. Therefore, it is plausible to argue that interns’ exploitation is a result of collusion between industry and the higher education system.

Hesmondhalgh (2010: 279) suggests a possible solution for the inequality caused by unpaid internships: media companies can build up a common fund to support young people as payment for their work, or they can pay extra taxes in order to support education. It is, however, outside my expertise to suggest possible solutions for the difficult conditions experienced by student interns in the Chinese context. Rather, I suggest that more attention from academia, such as research on digital labour, is needed to address the costs imposed on interns in new media industries in China.