Regulating labour dispatch in China: A cat-and-mouse game

Abstract

This article suggests cautious optimism toward the prevailing Polanyian countermovement discourse by providing a timely and comprehensive examination of the enforcement of the labour dispatch regulation in China. Since the enactment of the regulation, some enterprises have narrowed the remuneration gap between agency workers and formal employees, while others have retained a large gap in overtime pay, bonuses, and welfare benefits between these two groups of workers. The regulation has reduced the number of agency workers, but has invoked the abuse of the more precarious ‘outsourced’ workers as well. The regulation has had little effect on limiting the use of agency labour to temporary, auxiliary, or substitute positions, raising the requirements of engaging in the labour dispatch business, or stabilizing the employment of agency workers. This article contributes to the extant literature on regulatory enforcement by examining the effects of non-standard employment regulation, highlighting the variance of labour law compliance among enterprises with different types of ownership, and demonstrating how China’s ongoing transformation from a planned to a market economy since the 1980s and from high growth to a new normal since 2010 has fundamentally constrained the full implementation of its labour protection regulation.

Keywords

labour market regulation agency workers labour dispatch new normal outsourced workers precarity

Karl Polanyi’s notion of countermovement has gained worldwide currency in recent years. Data on public opinion about the re-embedding of markets in 20 countries during the 1990s show that a countermovement in public opinion was emerging at that time.¹ Data on social assistance programmes in 84 developing countries demonstrate that state-driven social protection policies have been rising and have spread throughout the past two decades in the Global South.² In China, Wang Shaoguang claims that a flourishing countermovement has emerged since the late 1990s when the government began to re-embed the market into social and ethical relations through redistributive social policies.³ In particular, the consecutive enactment of the Labour Contract Law (劳动合同法), the Employment Promotion Law (就业促进法), and the Labour Dispute Mediation and Arbitration Law (劳动争议调解仲裁法) in 2007 has been interpreted as a counter-movement in the sphere of industrial relations.⁴ Clearly, this countermovement discourse is based on the assumption that compliance in the case of social protection policies has reached satisfactory levels. Given the enforcement crisis that labour laws are facing, as discussed in the article, the extent to which this countermovement has been effective is questionable.

This article contributes to solving this countermovement puzzle by providing a timely and comprehensive examination of the implementation of agency labour regulation in China. Agency labour is a new form of precarious labour that prevails in present-day China. According to the Interim Provisions on Labour Dispatch (hereafter Interim Provisions), enforced on 1 March 2014, the two-year transitional period for enterprises in China to lower the proportion of agency workers in the entire workforce to a maximum of 10 per cent ended before 1 March 2016. The Interim Provisions are the latest and most rigorous piece of regulation that tightens the use of agency labour in China following the 2007 Labour Contract Law and its 2012 amendment. The research questions of this article are: how has China’s labour dispatch regulation been enforced; what are its effects; and what are the constraints of its implementation?

The rise of agency labour and related regulation in China

According to the International Labour Organisation, agency workers are those workers who are ‘deployed and paid by a private employment agency, but the work is performed for the user firm’.⁵ As a new form of precarious labour, compared to formal employees at the accepting unit, agency workers tend to have irregular working hours, irregular income, and less protection from arbitrary dismissals.⁶ The International Confederation of Private Employment Services estimated that, in 2013, there were over 40 million agency workers worldwide (excluding China).⁷ In China, agency labour has expanded since the late 1990s.⁸

The 2007 Labour Contract Law requires that labour dispatch agencies hold a minimum of RMB 0.5 million as registered capital and that they sign two-year or longer full-time labour contracts with dispatched workers, and that user enterprises put agency labour ‘generally’ in temporary, auxiliary, or substitute positions. Moreover, agency workers are entitled to equal pay for equal work as that received by regular employees in the accepting unit. The effects of these requirements, however, have largely been offset by the effects of Labour Contract Law’s requirements pertaining to regular employment in terms of open-ended labour contracts, social insurance, and severance fee. In response to this, many employers have proactively transformed their formal employees into agency workers, and also used agency workers when new hands were needed. In Shanghai, the number of agency workers increased from 0.68 million in 2008 to 1.33 million in 2010.⁹

As a result, according to the All-China Federation of Trade Unions, in 2011, agency workers totalled five million in state organs and public institutions, and 37 million in enterprises, that is, 13.1 per cent of the workforce of enterprises. Agency labour accounted for 16.2 per cent of the workforce in state-owned enterprises, 14 per cent in foreign enterprises, 4.5 per cent in collective enterprises, and 2.7 per cent in private enterprises; over half of the agency workers in China were migrant workers.¹⁰

To counter the rapid growth of agency labour after the enactment of the Labour Contract Law, an amendment to the Labour Contract Law was introduced in 2012. It increased the minimum registered capital of labour dispatch agencies to RMB 2 million. It required labour agencies to obtain labour dispatch licences before providing related services. It further detailed the provision of equal pay for equal work by requiring the unit which accepts agency labour to adopt a unified compensation scheme for its own employees and agency workers. The amendment also changed the word ‘generally’ to ‘only’ and defined three kinds of positions open for agency labour: temporary positions that last for less than six months; auxiliary positions that provide supplementary services to the primary business; and substitute positions which are ad hoc vacancies created by the temporary absence of regular employees. The 2014 Interim Provisions, as mentioned, imposed a 10 per cent cap on enterprises using agency labour (state organ users and public institution users are excluded).

A review of the enforcement of labour protection legislation

In an era of growing working-class precariousness, labour protection legislation is more desired than ever. Nonetheless, how to enforce labour protection regulation is equally important as how to design it. Unfortunately, while there are ongoing design efforts, a crisis of enforcement has taken place worldwide. In the UK, most work-related safety violations are either undetected or filtered out of the official resolution channels.¹¹ In Canada, few of the many firms which repeatedly violate the occupational health and safety regulation get punished.¹² The lack of enforcement is even more pronounced in developing and transitional countries.¹³ In addition, the compliance level varies by region,¹⁴ size of firms,¹⁵ regulatory style (punitive or persuasive),¹⁶ and so on.

What factors contribute to the failure to comply to labour laws? Guy Davidov argues that the difficulties that hinder labour law compliance lie in the following facts: employers have incentives for non-compliance; employees encounter barriers to enforcing compliance; and the state finds effective enforcement costly and complicated. These difficulties have been exacerbated in recent years by intensified global competition, the proliferation of migrant labour and non-standard employment arrangements, and so on.¹⁷ Moreover, political scientists find that political legacies, local interest groups and their elected representatives play a key role in shaping the compliance with labour codes.¹⁸ In developing countries in particular, the dominant argument is that labour inspectorates lack resources and independence for the enforcement of laws.¹⁹

While there is a plethora of studies exploring the enforcement and effects of the regulations on employment standards and occupational health and safety,²⁰ few studies investigate the enforcement and effects of non-standard employment regulations. This is because non-standard employment is usually regarded as a result of either the over-regulation of regular employment²¹ or the deregulation of the labour market.²² One study on the effects of non-standard employment protection acts in South Korea found that these acts, while reducing the proportion of non-standard employment to a certain extent, encouraged employers to take advantage of more precarious forms of workers.²³

In China, the chief culprit of non-compliance with labour laws is the lack of independence of labour inspectorates. Ching Kwan Lee refers to this as ‘authoritarian precarization’ – the Chinese government’s arbitrary and selective enforcement of labour laws is responsible for the sheer extent of informal employment in China. The opportunistic implementation of the Labour Contract Law is due to the changing economic, political, and social priorities of the state.²⁴ Under the current central–local relationship, local governments in China are responsible for both promoting local economic growth and implementing labour laws.²⁵ In practice, it is common that local governments help investors to circumvent the formal regulation of labour use and sidestep environmental protection requirements.²⁶ In addition to this, China’s labour inspection teams are often underfunded and understaffed.²⁷ Weak unions, the bureaucratic ‘command and control’ style of inspection and dispute resolution, and negligence on the part of the private sector also restrain the effectiveness of Chinese labour laws.²⁸

In terms of the effects of agency labour regulation, a questionnaire investigation conducted in Guangzhou in 2011 and 2012 showed that agency workers in China were still discriminated against in terms of wages, social security, and benefits, and were vulnerable to occupational hazards.²⁹ Drawing on 59 court cases adjudicated in 2012 involving labour dispatch disputes from the database of China’s Supreme People’s Court, Philip Huang found that agency workers had difficulties in claiming overtime pay and equal pay for equal work as compared to the regular employees of the accepting unit because of the murkiness of the concept of ‘equal work’.³⁰ A questionnaire survey of approximately 5000 enterprises in 20 cities conducted jointly by the Ministry of Human Resources and Social Security and the Beijing Foreign Enterprise Human Resources Service Corporation in mid-2013 indicated that, among those enterprises that used to have 10 per cent or more of agency workers in their workforce, 28.6 per cent of these hired the majority of agency workers directly; in 40.2 per cent of the cases, the majority of agency workers were made outsourced workers; and 2.2 per cent of the enterprises moved the agency-labour-intensive business out of China.³¹ A national survey conducted by the All-China Federation of Trade Unions between May and October 2014 found that outsourcing the peripheral business had been the main solution to reduce the use of agency labour nationwide.³² Using a similar approach to the one discussed above, Huang found that, between 2013 and 2016, the regulation of dispatch agencies was tightened with respect to obligations for social insurance, while managerial abuses pertaining to overtime and unilateral termination of agency workers continued.³³

While these studies shed light on some of the effects of agency labour regulation on agency workers, the whole picture of the regulation enforcement remains unclear. We do not know how the regulation has been carried out at the local level on a day-to-day basis, what the current working conditions are of outsourced workers who were previously employed by the same company as agency workers, what the specific constraints of this regulation are, and so on. Moreover, previous studies tend to regard enterprises as homogeneous with regard to their compliance with the labour dispatch regulation, an assumption which is dubious given the different proportions of agency workers in enterprises with different kinds of ownerships. This article fills these gaps by offering a systematic examination of how and how well the labour dispatch regulation has been implemented in enterprises with different kinds of ownerships.

Methodology

This article is based on multi-sited data collected between July and December 2016. The envisaged differences in labour law compliance at state enterprises, foreign-invested enterprises, and private enterprises, the relatively geographic dispersion of these three kinds of enterprises, and envisaged differences in labour law compliance in different regions necessitate a multi-sited approach at the research design stage. Fieldwork sites are shown in Figure 1.

Figure 1.

Multi-sited fieldwork undertaken in the study.

Although a study of several cities cannot claim statistical representativeness for all of China, the cities involved in this research are arguably the most theoretically significant cases to understand the regulation of labour dispatch in China. Beijing–Tianjin, Shanghai–Suzhou, and Guangzhou–Shenzhen–Dongguan represent the top three production hubs and thereby also the hubs of agency labour in China: the Pan-Bohai Region, the Yangtze River Delta, and the Pearl River Delta. Beijing, Tianjin, Shanghai, and Guangzhou are traditional industrial cities with large numbers of all three types of enterprises. Suzhou, Shenzhen, and Dongguan are newly industrialized cities attracting much foreign capital. Jinan and Changsha are second-tier cities.

The data examined in this article represent the collective efforts of the Labour Dispatch Research Group, which consisted of the author, an administrative staff from the Ministry of Human Resources and Social Security, and several labour NGO workers based in different cities of China. Thanks to professional connections between the Ministry of Human Resources and Social Security staff and local governments in the above-mentioned cities, as well as the author’s personal connections, the coverage of agents involved in labour dispatch in China was comprehensive. Most of the data was collected via semi-structured individual interviews with agency workers in state, foreign and private enterprises in order to gather information about their working conditions after the promulgation of the Labour Contract Law, its amendment, and the Interim Provisions, and via semi-structured group interviews with: (1) local labour officials to gain a general understanding of how they had implemented the regulation in their individual jurisdictions, their concerns in this process, and the enterprises’ reactions to the regulation; (2) human resources (HR) managers of state, foreign and private enterprises to obtain a clear picture of why they used agency labour, the working conditions of the agency workers at their enterprise (including pay, position, and social insurance), their reactions to the labour dispatch regulation, and the impact of the regulation on their enterprise and workers; and (3) managers of labour dispatch agencies to learn about their reactions and those of their clients to the regulation.

Individual interviews were conducted with 85 agency workers, including 44 working in seven state enterprises, 36 in 19 foreign enterprises, and five in three private enterprises. Most interviewees were reached in a stop-and-interview manner or through snowball sampling. Group interviews involved provincial labour officials in Guangdong and Shandong, their municipal counterparts in Suzhou and Shanghai, HR managers of 11 state enterprises, seven foreign enterprises, and two private enterprises, and managers of 19 labour dispatch agencies.

Why agency labour?

To some extent the reason why agency labour is prevalent in China points to how the regulation is enforced. Based on fieldwork interviews, this section explains why different enterprises in China use agency labour.

State enterprises began to use agency labour initially due to the tight wage quota and the unique accounting principle authorized by the State-owned Assets Supervision and Administration Commission (hereafter State-owned Assets Commission) and its predecessor. To boost the efficiency of state enterprises, from 1985 onwards, the annual general wage quota of individual state enterprises has been linked to its annual financial performance. According to the accounting principle of state enterprises, only the wages of formal workers enter the wage quota; wages of temporary workers are categorized as operational fees, independent of wage quotas and free from the direct control of the State-owned Assets Commission and its predecessor. Meanwhile, the remuneration of formal workers in many state enterprises remains significantly higher than the societal average. Under such circumstances, cheap temps have been massively deployed in state enterprises as rank-and-file staff to lower labour costs and raise profitability.³⁴

Foreign enterprises in China initially embraced agency labour due to path dependency. At the dawn of the reform and opening, foreign enterprises in China recruited Chinese workers via labour dispatch agencies because they themselves were not allowed to recruit workers and they were happy to have a strong local partner who could manage HR for them. Although they were later allowed to hire workers directly and they began to understand China better, some continued using workers via labour dispatch agencies. In addition, some foreign enterprises in China use agency workers because their headcount is controlled by their headquarters.³⁵ For small private enterprises, it is sometimes cheaper to use the HR management services provided by labour dispatch agencies instead of recruiting their own HR staff.³⁶ Electronics contract manufacturers are giant users of agency labour due to their labour-intensive production mode and fluctuating orders. In 2015, ASUS, a contract manufacturer for Apple, accounted for over half of agency labour in Suzhou New District.³⁷

All these enterprises expanded their use of agency labour to circumvent the Labour Contract Law after 2008. This law required employers de jure to pay the wuxian yijin (五险一金), that is, five kinds of social insurance – for healthcare, retirement, occupational disease, unemployment, and maternity – and the housing fund. When added up, employers’ compulsory contribution of wuxian yijin every year can amount to 30–40 per cent of the total wages of employees. Hiring through a third party, the employer de jure, allows enterprises to obscure and shirk some responsibilities as employers de facto, including paying the wuxian yijin. By returning agency workers to labour dispatch agencies when necessary, enterprises are also in a better position to avoid open-ended contracts and severance fees, which is of vital importance to enterprises handling fluctuating orders. Besides, many enterprises use agency labour to bypass the maximum probationary term of six months required by the Labour Contract Law. To cut labour costs, they prolong such terms to years.³⁸

Labour dispatch is also a common response to the shortage of unskilled workers. Due to the draining of China’s labour pool, employers have increasingly favoured the option of outsourcing recruitment to labour dispatch agencies over hiring staff themselves. In Suzhou, one labour official estimated that about 70 per cent of new hires in local factories were obtained via labour dispatch agencies in 2015.³⁹ This number is even higher for enterprises with a seasonal production rhythm. They use labour dispatch agencies to rapidly triple or quadruple their workforce to ensure that orders can be met in time. Rather than taking one-time commissions for recruitment and leaving workers to sign labour contracts with enterprises, labour dispatch agencies often leverage labour shortages and sign contracts with workers in order to earn monthly commissions for their labour dispatch services.⁴⁰

How and how well is the regulation implemented?

This section outlines how the labour dispatch regulation operates in China and its effects. The Ministry of Human Resources and Social Security has proposed three guiding principles concerning the enforcement of the regulation: the regulation cannot result in unemployment; disturb the operation of enterprises; or threaten social stability.⁴¹ After the initial campaign-style initiatives at the local level, the daily enforcement of the regulation adopted a ‘no report from below, no action from above’ approach. Aggrieved agency workers could turn to the local labour inspectorate and the local labour disputes arbitration committee for help, and might then file a lawsuit or petition when necessary.⁴² As our fieldwork progressed, differences in ownership appeared to be the overarching explanation for variations in legal compliance of different enterprises. This article largely omits the differences in labour law compliance between different regions.

Equal pay for equal work

Unequal pay for equal work provokes fierce criticisms of labour dispatch in China. The remuneration gap between formal employees and agency workers is the largest in state-owned enterprises where the wages of formal employees are beyond market regulation.⁴³ According to a manager of a key state-owned enterprise, equal pay for equal work, once fully realized, could amount to RMB 26 billion (out of the RMB 30 billion profit) in his industry in 2012. Thus, state-owned enterprises have become strong opponents of the equal pay for equal work provision, and have successfully forced the proponents, including the Ministry of Human Resources and Social Security and the All-China Federation of Trade Unions, to make a concession: equal pay for equal work should not include social insurance and welfare benefits.⁴⁴

Local labour officials were sluggish in enforcing equal pay for equal work because employers loathed this provision. They also found it hard to implement because of the ambiguity of the concept of equal work and equal pay. As a result, formalistic compliance, rather than substantive compliance, was pursued through a creative implementation of the regulation. The related provision is phrased as:

Agency workers shall have the right to receive the same pay as that received by employees of the accepting unit for the same work. The accepting unit shall adopt the same compensation policy for agency workers and formal employees in comparable positions according to the principle of equal pay for equal work.

When informed of related violations, the local labour inspectorate would ask the enterprise to provide the documentation of its compensation policy to see whether it included agency workers. If it did not, then it was required to include them. If it did but the agency workers were underpaid, the inspectorate would ask the enterprise whether the compensation policy had been approved by the congress of employees’ representatives or its equivalents. If it had been approved, the current compensation policy had to be respected even though it discriminated against agency workers. If not, the inspectorate would ask the enterprise to submit the compensation policy to the congress or equivalents for discussion. However, even labour officials themselves doubted whether the congress of employees’ representatives or equivalents could represent the real interest of workers, to say nothing of the fact that it normally involved only formal employees.⁴⁵

Litigation is of hardly any help to agency workers. The Supreme Court of China requires that all judicial documents (with few exceptions) issued by local courts should be available online with effect from mid-2013.⁴⁶ Ninety-two judicial documents, which were issued by courts in Shanghai, were available online in January 2017. These pertained to claims from plaintiffs, all agency workers, who asked for equal pay for equal work. No plaintiffs succeeded in court either because judges insisted that these agency workers could not prove that they were underpaid, or that they were not in comparable positions to formal employees in the accepting unit, or because employers had the right to tailor compensation according to each worker’s diploma, job tenure, work experience, morale, and so on.

To avoid legal risks, some employers took steps to narrow the remuneration gap between agency workers and formal employees. For example, Beijing Post and Shandong Airways have launched several rounds of wage increases specifically for agency workers in recent years. Some employers, particularly labour-intensive foreign enterprises, however, obscured equal pay by adopting different compensation schemes for agency workers and formal employees. At Foxconn (Shenzhen), agency workers were paid solely according to hourly rates, while formal workers were paid base wages, overtime wages, and bonuses. To earn the same level of wage, agency workers had to work longer hours than formal employees in comparable positions. After the promulgation of the regulation, some employers adopted a partially unified payment system for the two groups of workers, while retaining the large gap in bonuses between them. In a state-owned steel company in Changsha, the normal festive bonuses were RMB 200 for agency workers and RMB 20,000–50,000 for formal employees when business was thriving.

In terms of welfare benefits, formal employees of state enterprises could enjoy the wuxian yijin, and some extra insurance and benefits, while agency workers were covered by only three kinds of social insurance (for healthcare, occupational disease, and retirement) and had no access to any extra insurance and benefits, which could amount to a significant percentage of their total wages in monetary terms. In most foreign and private enterprises, wuxian yijin and overtime pay remained the privilege of formal employees. Labour agencies would buy agency workers cheap commercial insurance against occupational disease (approximately RMB 20 per month per capita) instead.⁴⁷ To ensure that workers did not do equal work, some employers avoided placing agency workers and formal employees in similar positions. In the foreign enterprises of Tianjin Xiqing industrial park, it was common for agency workers to be placed in rank-and-file positions while formal employees were in management.⁴⁸ Some employers, however, just changed job titles to avoid equal work. For example, agency workers were ‘drivers’ while formal employees were ‘senior drivers’ although they might have similar driving skills and work experience.⁴⁹

The 10 per cent cap

The number of agency workers in China actually did decrease after the promulgation of the Interim Provisions. In Shanghai, the number of agency workers dropped from a peak of 1.33 million to 0.88 million at the end of 2015.⁵⁰ In Guangdong, the numbers fell from a peak of 1.63 million to 0.85 million in mid-2016. According to a labour official in Guangzhou Labour Bureau, of the 100 randomly selected enterprises in Guangzhou in 2016, 76 per cent had reduced the number of agency workers by at least 20 per cent after the enactment of the Interim Provisions.⁵¹ The majority of labour dispatch agencies investigated also reported a significant shrinkage of their labour dispatch business. However, it is too early to conclude that the regulation has successfully prevented the abuse of agency labour in China.

Employers mainly use two methods to deal with the 10 per cent cap: temp-to-formal conversion and dispatch-to-outsourcing conversion.⁵² Temp-to-formal conversion means that agency workers are made formal employees of the accepting enterprise. It is what agency workers expect and also what the regulation aims to achieve, since agency workers are better off after a temp-to-formal conversion. Dispatch-to-outsourcing conversion means that agency workers are sent out under the guise of ‘outsourced’ workers. The standard practice of labour dispatch, outsourcing and the common practice of dispatch-to-outsourcing conversion are specified in the following.

In standard labour dispatch, the accepting enterprise signs a labour dispatch contract with the labour dispatch agency; the labour dispatch agency signs labour contracts with agency workers and dispatches them to the accepting enterprise which puts them to work; the agency receives a monthly payment package from the accepting unit, including commission for itself, wages for dispatched workers, and insurance fees for local insurance bureaus; and only the commission is taxed. In standard outsourcing, the issuing party signs an outsourcing contract with the contractor and contracts out certain tasks to the latter; the contractor signs labour contracts with its employees and gets them to accomplish those tasks on its premises using its machines, as is the arrangement between Apple and its contract manufacturer Foxconn;⁵³ the contractor receives a revenue based on the quantity and quality of the task finished, and the whole revenue is taxed. Although there is no legislation clarifying the main difference between labour dispatch and outsourcing in China, in practice, the interviewees concurred that the difference lay in who managed the workers and what type of revenue was taxed.⁵⁴

By contrast, the common practice of dispatch-to-outsourcing conversion triggered by the 10 per cent cap is as follows: the enterprise re-signs a business outsourcing contract to replace the previous labour dispatch contract, usually with the same labour dispatch agency, or with companies jointly owned by the enterprise and the labour dispatch agency. The latter pays taxes for such business as contractors, rather than as labour dispatch agencies. Given that the regulation does not present clear definitions of agency workers or outsourced workers, by changing the documentation and the taxation arrangement, user enterprises can use agency workers in the guise of outsourced workers. Such workers often continue working at the facilities of the accepting unit and under its management. That is why such practice is called ‘actual labour dispatch, disguised as labour outsourcing’. Employers may seek to outsource some business to contractors, as in the case of Apple with Foxconn. Unfortunately, they simply cannot find enough mature contractors since agency labour is widely and massively deployed in all kinds of industries in China, which is at the lower end of the global commodity chain. Besides labour, production facilities, processing techniques, and management are all crucial for standard contractors, thus creating high-entry barriers for asset-light labour dispatch agencies. With formal workers placed in management and the majority of informal workers working in rank-and-file positions, coupled with the need to impose quality control, the issuing enterprises therefore have to be deeply involved in the management of this variant of outsourcing, inevitably resulting in actual labour dispatch, disguised as labour outsourcing. Such practice has been so prevalent that labour officials in Shanghai said that the majority of business outsourcing derived from labour dispatch fell into this category.

The lawmakers of the Interim Provisions were aware of the prevalence of such creative compliance. In the consultation draft of the Interim Provisions, Article 2 stipulates:

Labour dispatch refers to the employment practice in which the labour dispatch agency dispatches its employees to the accepting unit who manages these workers. Outsourcing business to contractors but retaining the management of workers of contractors is one form of labour dispatch.

According to this provision, the practice of actual labour dispatch, disguised as labour outsourcing, will be regarded as one form of labour dispatch, which cannot be used by enterprises to reach the 10 per cent cap. However, Article 2 in the consultation draft version was deleted in the enforced version. Local labour officials concurred that the 10 per cent cap was so stringent that it had forced employers to act creatively. They were aware of the prevalence of this kind of outsourcing but imposed no penalties on employers. They also complained that the 10 per cent cap lacked social support. Employers loathed it while agency workers cared more about income than employment status.

Interviews with local labour officials show that in Suzhou New District in 2016, 70–80 per cent of agency workers were in the guise of outsourced workers; in Guangzhou, after the enactment of the Interim Provisions, approximately 33 per cent of enterprises which deployed agency workers formalized all agency workers, while 54 per cent introduced outsourcing as a guise. At the enterprise level, as shown in Table 1, state enterprises were highly responsive to the 10 per cent cap. A substantial percentage of agency workers became formal workers, while a considerable number of agency workers and new recruits were signed on in the guise of outsourced workers (layoffs were rare). Among foreign enterprises, those using agency labour due to the headcount constraint imposed by their headquarters tended to prioritize compliance with the Chinese law and turn the majority of agency workers into formal employees. Those placing agency workers in labour-intensive business such as retailing and manufacturing, however, tended to ignore the cap or used agency workers in the guise of outsourced workers. Labour dispatch remained their best solution for coping with fluctuating tasks.⁵⁵ Agency workers continued dominating the shop floors of many electronics contract manufacturers in Tianjin, Changsha, and Shenzhen.⁵⁶

Table 1.

A sample of state enterprises’ response to the 10 per cent cap.

No.	Location	Industry	AW peak %	AW 2016 %	OW 2016 %
1	Guangdong	Communication	-	0	47.6
2	Guangdong	Energy	80	6.5	14.8
3	Guangdong	Logistics	50	19	17.8
4	Shanghai	Communication	-	2.6	39.5
5	Shanghai	Logistics	61	36	-
6	Shanghai	Energy	-	1.7	5.1
7	Suzhou	Energy	-	0	86.8
8	Shandong	Communication	70	7	-
9	Shandong	Manufacturing	37	14.9	8.3
10	Shandong	Transporting	-	9.9	0
11	Shandong	Logistics	-	24.7	9.3

Notes: AW = agency workers; OW = ‘outsourced’ workers; - = not known.

Source: Compiled from interviews with HR managers of related enterprises.

The cap has various effects. As shown in Table 2, both temp-to-formal conversion and dispatch-to-outsourcing conversion have significantly raised labour costs. For temp-to-formal conversion, employers have to close the remuneration gap between newly made formal workers and previous formal workers in comparable positions in terms of wages, basic social insurance, extra insurance, the housing fund, and bonuses, and pay extra fees associated with formal employees, such as trade union fees. The enterprise will also have to increase the remuneration of these new formal workers according to length of tenure and pay them pensions after they retire as it does for formal workers. In this way, the increase in labour costs due to temp-to-formal conversion is relatively low in the first year of the conversion but will grow annually. Dispatch-to-outsourcing conversion is also more expensive than labour dispatch. Only the commission paid by the accepting unit is taxed in labour dispatch, while the whole revenue is taxed in outsourcing. Moreover, apparent contractors need money to buy extra business insurance to hedge the risk of more employer responsibility de jure. In comparison, agency workers’ welfare increases when they are formalized, and decreases when these workers are sent out in the guise of outsourced workers. These apparently outsourced workers are more remote from the accepting unit in terms of employment relationship and cannot be made formal workers (unlike agency workers, who can). Some enterprises even lowered the pay of these apparently outsourced workers to offset the cost increase due to dispatch-to-outsourcing conversion. While Chinese labour laws have special provisions to counter the abuse of agency workers, they do not have special provisions against the abuse of such outsourced workers who actually fall under the category of labour dispatch. The implementation of the 10 per cent cap caused many apparently outsourced workers to leave the enterprise. The annual turnover rate of agency workers in a state-owned enterprise in Guangdong used to be about 6–7 per cent, but the company lost some 20 per cent of its workers who worked in the guise of outsourced workers between January and September 2016.⁵⁷

Table 2.

A sample of enterprise labour costs increase due to the cap.

No.	Location	Ownership	Increase in labour costs in the first year
1	Guangdong	State	Not known due to TFC At least 8% due to DOC
2	Guangdong	State	At least 8% due to TFC RMB 30 million due to DOC
3	Shanghai	Foreign	12% due to TFC 17% due to DOC
4	Shanghai	State	20% due to TFC 25% due to DOC
5	Shandong	State	At least 12% due to TFC Not known due to DOC
6	Shandong	State	RMB 10,000–20,000 per capita due to TFC Not known due to DOC

Notes: The base for the percentage increase is previous labour costs of related agency labour. TFC: temp-to-formal conversion; DOC: dispatch-to-outsourcing conversion.

Source: Compiled from interviews with HR managers of related enterprises.

Other provisions

Another major provision to counter the abuse of agency labour is that agency workers can only be placed in temporary, auxiliary, or substitute positions. Although the 2013 amendment to the Labour Contract Law defines these three kinds of positions, it is still difficult to judge whether the placement of agency workers is legitimate in many cases. For example, it cannot be easily ascertained whether the position of HR managers in a toy factory is auxiliary according to the amendment which defines auxiliary staff as those providing supplementary services to the primary business – more legal details are needed to differentiate primary from supporting. In practice, as in the case of the implementation of equal pay for equal work, local labour bureaus trusted the congress of employees’ representatives or equivalents in enterprises to determine the list of temporary, auxiliary, and substitute positions. Thus, procedural justice again triumphed over resultant justice.⁵⁸ Agency workers were still employed to work at the assembly lines of contract manufacturers, call centres of communication service providers, and other similar positions.⁵⁹

The effort to raise the entry threshold of labour dispatch agencies was also in vain. An increase in the minimum registered capital from RMB 0.5 million to RMB 2 million can hardly prevent the proliferation of small, asset-light agencies. In all the cities where we had interviews with local labour officials, we were told that labour dispatch agencies could easily buy an asset verification report stating assets worth at least RMB 2 million from professional asset firms at a cost between RMB 5000 and RMB 20,000. Local labour bureaus had no capacity to authenticate these reports, and they were exposed to the risk of administrative litigation if they refused to issue licences for labour dispatch to labour agencies with such reports in hand, even if they knew that those agencies were family-run and operated in the owners’ home. The entry threshold was so low that the labour dispatch market remained crowded with tiny, unscrupulous agencies which docked wages and insurance fees daily.⁶⁰

The provision that aims to stabilize the employment of agency workers fails to regulate those employers that use short-term agency labour to cope with fluctuating orders. The lawmakers intended to use the requirement of two-year or longer labour contracts with agency workers to force labour dispatch agencies to channel agency workers to enterprises requiring such workers at different peak times. None of the agencies that we interviewed, however, liked this idea because it was too risky. Once a worker is returned, the agency is required by the law to pay the local minimum wage before the worker is re-dispatched. For agencies earning only meagre commissions from enterprises, this responsibility is too heavy to shoulder. In practice, the most popular form of agency labour in electronics manufacturing in Dongguan and Shenzhen was hourly paid workers. They were on three- to six-month contracts with agencies, and were paid by the hour. Even if agency workers were on two-year contracts with the agency but were returned by the accepting unit halfway, agencies could still circumvent the minimum wage responsibility by, for example, forcing them to resign by threatening to dispatch them to remote places.⁶¹

Contextualizing the limited success of the regulation: Old legacies and the new normal

China’s ongoing transformation and the new normal (slower growth and economic restructuring) give us clues to understand the limited success of the regulation.

On the one hand, the legacy of the planned economy is still felt, even if market-oriented reform has been ongoing for over 30 years. One labour official in Shanghai put it bluntly: ‘The remuneration gap between agency workers and formal employees is the largest in state enterprises not because agency workers there are too poorly paid, but because formal employees are too well paid’.⁶² Equal pay for equal work cannot be achieved in state enterprises with the compensation of formal workers protected by the state and the compensation of agency workers determined by the labour market. Moreover, as in the case of the practice developed in the planned economy era, the wage quota of state enterprises in present-day China is still authorized by the government. Although the Interim Provisions require state enterprises to comply with the 10 per cent gap, the State-owned Assets Commission insisted that the wage quota of individual state enterprises could only be adjusted according to financial performance, not according to the number of employees, thus forcing state enterprises to resort to making agency workers outsourced workers in order to lower their percentage of agency workers.⁶³

The entrenched rural–urban dualism consolidated in the planned economy era also hampers the enforcement of the regulation by locking migrant agency workers – who comprise the majority of agency workers in China – in a state of continuous migration between cities and their hometown villages. Our interviews with migrant agency workers indicate that working in big cities to make money and settling down in small cities or hometown villages remained the dominant trajectory of this group. Emboldened by the shortage of blue-collar workers in recent years, the migrant workers at prime ages that we interviewed said that they changed jobs frequently and they had no worries of un-employment. Under such circumstances, the regulation alone can hardly stabilize their employment. Meanwhile, although the gap in social insurance between agency workers and formal employees is salient, particularly in foreign and private enterprises, migrant workers are discouraged from contributing to or from demanding that employers contribute to the pension pool of the cities where they work. In the eyes of our agency worker informants, such contributions did not only lower their present income but also might not be translated into real benefits because of their mobility and the obstacles in transferring social insurance entitlements between different regions of China.⁶⁴

On the other hand, the fact that China is experiencing a new normal has diverted the priority of the government away from labour protection toward economic revitalization and employment promotion. In 2014, Chinese President Xi Jinping said that China had to adapt to a new normal characterized by slower growth, the upgrading of industry, and an innovation-driven economy.⁶⁵ Notably, China’s GDP growth rate has dropped from around 10 per cent (2002–11) to 7-odd per cent (2012–14), and further to 6-odd per cent since 2015.⁶⁶ One of the Chinese government’s countermeasures is to lower the operating costs of enterprises, particularly labour costs. To do so, enterprises’ contributions to social insurance and to the housing fund have been lowered and increases to the minimum wages have been reduced. For example, according to labour officials in Guangdong, the provincial government lowered the average premium rate of occupational injury insurance which enterprises contributed to from 0.63 to 0.43 per cent in 2015; lowered the premium rate of unemployment insurance which enterprises contributed to from 1.5 to 0.8 per cent in 2016.⁶⁷ Labour dispatch had already been widely used by employers to avoid contributions to social insurance and the housing fund.

Under such circumstances, the Ministry of Human Resources and Social Security suggested that local labour bureaus softened the 10 per cent cap and focused more on equal pay for equal work, which excludes social insurance and welfare benefits. Moreover, in 2016, the Ministry of Human Resources and Social Security suspended the initiative to issue any guidelines that distinguished outsourcing from labour dispatch despite the fact that considerable preparatory work had been carried out in 2015.⁶⁸ Labour dispatch and its variations, such as actual labour dispatch disguised as labour outsourcing, had, therefore, been tacitly incorporated into the central government’s general plan to cut the operating costs of the real economy. A labour bureau official in Shanghai said incisively: ‘It’s good news when enterprises are still playing cat-and-mouse games with you. At least, they are still doing business in China. Real troubles come when they stop playing and close down.’⁶⁹ Meanwhile, with the slowdown in economic growth, the Ministry of Human Resources and Social Security has shifted the focus from the quality to the quantity of employment. Flexible employment was promoted and valued as a means to buffer the impact of an economic slowdown on employment.⁷⁰

Conclusion

The labour dispatch regulation in China is like a cat-and-mouse game involving rounds of pursuit, near captures, and repeated escapes. In the first round, the central government introduced the Labour Contract Law to enhance the protection of agency workers, only to prompt employers to employ more agency labour in their workforce. In the second round, an amendment to the Labour Contract Law and the Interim Provisions was enacted with a more detailed requirement of equal pay for equal work and a stringent 10 per cent cap, only to trigger more countermoves from local governments, employers, and labour dispatch agencies to hollow out the regulation.

This article has made a timely and comprehensive examination of how and how well the labour dispatch regulation in China has been implemented. To buffer the impact of the regulation on enterprises and to circumvent the ambiguities of some provisions, local labour inspectorates prioritized procedural justice over resultant justice in implementing the provisions concerning equal pay for equal work and temporary, auxiliary, or substitute positions. Such practice created loopholes for enterprises to escape. Aligned with local governments, local judicial systems leveraged ambiguities in the regulation and the complexities of reality to go against agency workers’ appeals for equal pay for equal work. As a result, although the remuneration gap between agency workers and formal employees had narrowed down in some enterprises, it remained large in terms of overtime pay, bonuses, and welfare benefits in the other enterprises. Even after the new regulation, agency workers were still massively employed in rank-and-file positions crucial to enterprises. Local governments also turned a blind eye to enterprises’ massive transformation of agency workers into more precarious apparently outsourced workers. The labour dispatch market was still crowded with asset-light and unscrupulous agencies. Many agency workers were still in unstable employment. This situation suggests cautious optimism toward the countermovement discourse.

Second, this article argued that enterprises with different kinds of ownership comply with the labour dispatch regulation differently. Although the research design took into consideration the varying levels of regulation compliance of different regions, the major differences in compliance, however, have unfolded along with the ownership of enterprises. Agency workers in most state enterprises had access to basic social insurance although they were likely to be denied access to extra welfare benefits and bonuses. Agency workers in private and foreign enterprises, however, tended to be denied access to basic social insurance, overtime pay, and bonuses, which remained the privilege of the formal employees. State enterprises had simultaneously relied on temp-to-formal conversion and dispatch-to-outsourcing conversion to approach the 10 per cent cap. Foreign enterprises using labour dispatch because of restrained headcounts tended to formalize the majority of agency workers. Foreign enterprises using labour dispatch to lower labour costs, however, tended to resort to using apparently outsourced workers.

Third, this article explained why the labour dispatch regulation in China has had limited success so far. Apart from the ambiguities of key provisions, the enforcement of the regulation has unfolded in an era in which China is undergoing transformations from a planned to a market economy, and from rapid and unsustainable development to slow and more balanced growth. The implementation of the regulation has thus been subjected to the gravity of old legacies and the new normal. The sheer inequality between formal and agency workers as a legacy of state enterprises inherited from the planned economy has invoked the regulation but has also restrained its full implementation. With the entrenched rural–urban dualism, the regulation alone can hardly stabilize the employment of migrant agency workers. The central government has kept the upper hand in coordinating its sometimes conflicting policy goals. In the face of intensified social grievances, it introduced the labour protection regulation to create the impression that social justice is emphasized. It then, however, fine-tunes the implementation of the regulation according to the needs of economic performance.

Footnotes

Notes

References

七成企业派遣用工比例低于10%: 人力资源流程外包需求量将增加 (Agency workers accounted for less than 10% of employment in 70% of investigated enterprises: The need for HR outsourcing will increase) (2013) 29 October. http://learning.sohu.com/20131029/n389159663.shtml, accessed 1 July 2018.

All-China Federation of Trade Unions General Office and Legal Work Department 中华全国总工会办公厅及法律工作部 (2015) 《劳务派遣暂行规定》贯彻落实情况报告 (A report on the implementation of the ‘Interim Provisions on Labour Dispatch’). 26 February. http://ghpf.acftu.org/art/2015/2/26/art_106_84312.html, accessed 24 June 2017.

All-China Federation of Trade Unions Project on the Issue of Dispatch Work全总劳务派遣问题课题组 (2012) 当前我国劳务派遣用工现状调查 (Investigation of China’s present dispatch labour use). 中国劳动 (China labour), no. 5: 23–5.

Almeida

Rita

Carneiro

Pedro

(2009) Enforcement of labor regulation and firm size. Journal of Comparative Economics 37(1): 28–46.

Brown

Richard

(1994) Theory and practice of regulatory enforcement: Occupational health and safety regulation in British Columbia. Law & Policy 16(1): 63–91.

Cairns

Daniel S. S.

(2015) New formalities for casual labor: Addressing unintended consequences of China’s Labor Contract Law. Washington International Law Journal 24(1): 219–52.

Chan

Jenny

Pun

Ngai

Selden

Mark

(2013) The politics of global production: Apple, Foxconn and China’s new working class. New Technology, Work and Employment 28(2): 100–15.

Cooney

Sean

(2006) Making Chinese labor law work: The prospects for regulatory innovation in the People’s Republic of China. Fordham International Law Journal 30(4): 1050–97.

Cooney

Sean

(2007) China’s labour law, compliance and flaws in implementing institutions. Journal of Industrial Relations 49(5): 673–86.

10.

Davidov

Guy

(2010) The enforcement crisis in labour law and the fallacy of voluntarist solutions. International Journal of Comparative Labour Law and Industrial Relations 26(1): 61–82.

11.

Gallagher

Mary

et al ., (2015) China’s 2008 Labor Contract Law: Implementation and implications for China’s workers. Human Relations 68(2): 197–235.

12.

Gimpelson

Vladimir

Kapelyushnikov

Rostislav

Lukyanova

Anna

(2010) Employment protection legislation in Russia: Regional enforcement and labor market outcomes. Comparative Economic Studies 52(4): 611–36.

13.

Harris

Kevan

Scully

Ben

(2015) A hidden counter-movement? Precarity, politics, and social protection before and beyond the neoliberal era. Theory and Society 44(5): 415–44.

14.

Huang

Philip C. C.

(2011) The theoretical and practical implications of China’s development experience: The role of informal economic practices. Modern China 37(1): 3–43.

15.

Huang

Philip C. C.

(2017) Dispatch work in China: A study from case records, part I. Modern China 43(3): 247–87.

16.

Huang

Philip C. C.

(2017) Dispatch work in China: A study from case records, part II. Modern China 43(4): 355–96.

17.

International Confederation of Private Employment Services (CIETT) (2015) Economic Report 2015 Edition (Based on Data of 2013/2014). http://www.wecglobal.org/fileadmin/templates/ciett/docs/Stats/Economic_report_2015/CIETT_ER2015.pdf, accessed 24 June 2017.

18.

International Labour Office (2006) Labour Inspection. Geneva: International Labour Office.

19.

International Labour Office (2016) Non-standard Employment Around the World: Understanding Challenges, Shaping Prospects. International Labour Organization. http://www.ilo.org/wcmsp5/groups/public/—dgreports/—dcomm/—publ/documents/publication/wcms_534326.pdf, accessed 5 October 2017.

20.

Lee

Ching Kwan

(2007) Against the Law: Labor Protests in China’s Rustbelt and Sunbelt. Berkeley: University of California Press.

21.

Lee

Ching Kwan

(2016) Precarization or empowerment? Reflections on recent labor unrest in China. The Journal of Asian Studies 75(2): 317–33.

22.

Lee

Joohee

(2017) More protection, still gendered: The effects of non-standard employment protection acts on South Korean women workers. Journal of Contemporary Asia 47(1): 46–65.

23.

Levien

Michael

Paret

Marcel

(2012) A second double movement? Polanyi and shifting global opinions on neoliberalism. International Sociology 27(6): 724–44.

24.

Meng

Jie 孟捷

Yile 李怡乐

(2013) 改革以来劳动力商品化和雇佣关系的发展 – 波兰尼和马克思的视角 (The commodification of labour and the development of employment relations since reform – From the perspectives of Marx and Polanyi). 开放时代 (Open times), no. 5: 74–106.

25.

Michie

Jonathan

Sheehan

Maura

(2003) Labour market deregulation, ‘flexibility’ and innovation. Cambridge Journal of Economics 27(1): 123–43.

26.

人社部: 劳务派遣同工同酬不包括福利和社保 (Ministry of Human Resources and Social Security: Equal pay for equal work for agency workers excludes welfare benefits and social insurance) (2013) 2 September. http://www.scio.gov.cn/zhzc/8/4/document/1345141/1345141.htm, accessed 24 June 2017.

27.

Portes

Alejandro

Castells

Manuel

Benton

Lauren A

. (eds) (1989) The Informal Economy: Studies in Advanced and Less Developed Countries. Baltimore, MD and London: Johns Hopkins University Press.

28.

Ronconi

Lucas

(2012) Globalization, domestic institutions, and enforcement of labor law: Evidence from Latin America. Industrial Relations: A Journal of Economy and Society 51(1): 89–105.

29.

Rutkowski

Jan J.

Scarpetta

Stefano

(2005) Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union. Washington, DC: World Bank.

30.

Juliana

(2014) Exploring the plight of dispatch workers in China and how to improve their conditions: A preliminary study. Working USA: The Journal of Labor and Society 17(4): 531–52.

31.

The World Bank (2017) China GDP growth (annual %): World Bank national accounts data, and OECD National Accounts data files. http://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=CN, accessed 24 June 2017.

32.

Tombs

Steve

Whyte

David

(2008) Crisis of enforcement: The decriminalisation of death and injury at work. Centre for Crime and Justice Studies Briefing 6, June. https://www.crimeandjustice.org.uk/sites/crimeandjustice.org.uk/files/crisisenforcementweb.pdf, accessed 17 June 2018.

33.

Vosko

Leah F.

(2000) Temporary Work: The Gendered Rise of a Precarious Employment Relationship. Toronto: University of Toronto Press.

34.

Vosko

Leah F.

et al ., (2011) New approaches to enforcement and compliance with labour regulatory standards: The case of Ontario, Canada. Comparative Research in Law & Political Economy Research Paper no. 31/2011. http://digitalcommons.osgoode.yorku.ca/cgi/viewcontent.cgi?article=1069&context=clpe, accessed 11 June 2018.

35.

Wang

Haiyan 王海燕

(2011) 本市劳务派遣用工规模趋升 (Agency labour is on the rise in Shanghai). 解放日报 (Jiefang daily), 10 September, section 3.

36.

Wang

Shaoguang 王绍光

(2008) 大转型: 1980年代以来中国的双向运动 (The great transformation: Two-way movement in China since the 1980s). 中国社会科学 (Social sciences in China), no. 1: 129–48.

37.

Xi’s ‘new normal’ theory (2014) 10 November. http://www.china.org.cn/world/2014-11/10/content_34012582.htm, accessed 24 June 2017.

38.

Feng

(2009) The emergence of temporary staffing agencies in China. Comparative Labor Law & Policy Journal 30(2): 431–61.

39.

Zhang

(2015) Inside China’s Automobile Factories: The Politics of Labor and Worker Resistance. Cambridge: Cambridge University Press.