Introduction
From the late 1970s, China gradually transitioned from a centrally planned state socialist economy to an export-focused, modern socialist market economy ‘with Chinese characteristics’ (Reference HuangHuang, 2008). This included introducing open labour markets and gradually dismantling the Maoist-era ‘iron rice bowl’ under which employees, particularly in state-owned enterprises (SOEs), had enjoyed lifetime employment, ‘a cradle-to-grave welfare system and relatively equal wages’ (Reference Warner, Lee, Sheldon, Kim and LiWarner and Lee, 2011: 22). Whereas previously, there had been very limited external labour mobility, employees now had much greater freedom to shift jobs through contracting for employment, and employers were now free to hire and fire. Nonetheless, state interventions continued, including through the hukou locality residential permit, which greatly discriminated against rural workers migrating to China’s rapidly growing cities for better-paid work. As well, despite dramatic restructuring of the SOE sector that devastated some of their heavy industries and their workforces, SOEs have continued to provide advantageous employment conditions (Reference Warner, Lee, Sheldon, Kim and LiWarner and Lee, 2011).
The transition from planned economy to markets created deep structural workforce inequalities (Reference Cooke, Sheldon, Kim and LiCooke, 2011) within an overall boom in economic growth and in living standards. Perceptions of systemic unfairness in treatment weakened the legitimacy of the new labour market regime, increasingly also weakening the ‘social harmony’ desired by China’s central government. Introduction of the 2008 Employment Contract Law (henceforth ECL or ‘new law’) sought to address these concerns. The ECL’s main focus is on uplifting labour rights at the individual level (Reference Hendrischke, Sheldon, Kim and LiHendrischke, 2011). One of the main issues the ECL addresses is the widespread employment (and hence income) insecurity at the heart of many other forms of labour market unfairness. Crucial here was the new law’s promotion of continuing contracts of employment in place of fixed-term ones. The ECL’s provisions regarding continuing contracts took effect on 1 January 2008. The new law stimulated controversy over whether mandatory written employment contracts, in particular non-fixed term (‘continuing’) ones, would reduce employers’ flexibility in hiring and firing and increase labour costs (Reference Gallagher, Giles and ParkGallagher et al., 2016; Reference Hendrischke, Sheldon, Kim and LiHendrischke, 2011). If this were to transpire, some have argued, it might have negative impacts on China’s economic growth (Reference Zhao and ZhangZhao and Zhang, 2010). Indeed, an OECD (Organisation for Economic Co-operation and Development) Employment Outlook (2013) argues that excessive restrictions imposed on companies may hinder job creation.
Opponents of the ECL – mainly employers, business lobbyists, employers’ associations, and neo-classical economists – argued that employers would be more cautious when recruiting through continuing contracts. This reflects employer perceptions that continuing contracts accord workers excessive protection on issues such as termination and severance, while compromising employers’ own managerial flexibility. In these situations, employers facing uncertainty arising from such a new environment (changes in institutions, economy and policies) would seek more ‘strategic ways’ to manage employment relations (Reference Hendrischke, Sheldon, Kim and LiHendrischke, 2011; Reference Sheldon and ThornthwaiteSheldon and Thornthwaite, 1999). On the other hand, supporters of the law, including unions and labour-oriented scholars, beyond arguing on the basis of fairness and social harmony, believed that continuing contracts enhance workers’ sense of belonging and commitment to the enterprise and improve employment relations. They dismissed the claim that there is any direct contradiction between continuing contracts and enterprise employment flexibility, and warned that marginalising workers’ rights for labour cost minimisation might turn economic issues into social problems (Reference WangWang, 2015).
The ECL was China’s first labour legislation since the 1995 Labour Law, itself the Chinese Communist Party’s (CCP) first labour law intended to protect workers’ rights. The 1995 law’s overarching aim was to ensure social harmony and economic development in a socialist market economy with ‘Chinese characteristics’. In this, continuing contracts were to provide job security in the new system based on voluntary employment contracts, and a smoother transition from the old to the new employment system. As well, it represented some ‘compensation’ for older workers who had contributed under the planned economy (Reference YanYan, 2015). The Labour Law represented a fundamental shift, focusing on individual employment within labour markets and on transition from high labour stability to high mobility (Reference Li and ZhaoLi and Zhao, 2012). While well intentioned, the Labour Law coincided with China’s first round of drastic reforms of its SOEs. That produced millions of layoffs and dramatic labour oversupply, particularly in some regions. Employment relations were unstable and, this, in addition to enforcement problems with the 1995 law, left many vulnerable workers exposed to exploitation (Reference PicklesPickles, 2015).
In September 2005, the Labour Law Enforcement Inspection of the Standing Committee of the National People’s Congress revealed that more than 60% of enterprises and workers had signed short-term employment contracts. Most signed every year, and some even signed a few times each year. These contracts accorded employers greater employment flexibility and minimised compensation available to workers in termination cases (Reference Yang and YiYang and Yi, 2007). They also resulted in the widespread exploitation of workers (Reference Wang, Appelbaum and DegiuliWang et al., 2009).
A core purpose of the 2008 ECL was therefore to promote harmonious and stable employment relations (Reference Zhao and ZhangZhao and Zhang, 2010). Improving on the 1995 law, the ECL obliges the employer to enter into a written, continuing contract in any of the following circumstances: (1) the worker has worked for an uninterrupted period of 10 years for that employer, (2) the worker has had continuous employment on two previous fixed-term contracts with that employer or (3) where the employer fails to sign a written contract after 1 full year of employment (Article 14). Under the new law, if an employer fails to enter into a written contract after more than 1 month but before 1 year after the employment starting date, the employer shall pay the worker double their monthly salary for each month the worker is employed (Article 82). In sum, the new law has expanded the scope and application of continuing contracts as well as increasing non-compliance penalties (Reference Gallagher, Giles and ParkGallagher et al., 2016).
This suggests our research questions: has the new law increased the likelihood of employers and employees signing continuing employment contracts? Has it redressed some of the main structural workforce inequalities over employment security by giving advantage to previously disadvantaged categories of workers? How have employers responded to this inhibition on their prerogatives over hiring and firing? How might organisational theory explain diverse employer choices?
To conduct the study, we draw on data from the 2007 and 2012 All-China Federation of Trade Unions (ACFTU) surveys, which cover approximately 80,000 individuals. In addressing the questions above, this article contributes to contemporary policy-related literatures on employment relations in China. Furthermore, it contributes to debates on employer flexibility and/or employee security, including in transition societies (e.g. Reference Judzik, Khan and SpagnoloJudzik et al., 2016; Reference SilSil, 2017). Researchers and policymakers need to know more about the impacts of significant new labour legislation providing additional protections to workers, and how employers might respond, particularly for much-debated legislation like the ECL (Reference Hendrischke, Sheldon, Kim and LiHendrischke, 2011). For example, Reference Katz and DarbishireKatz and Darbishire (2000), in studying 20 years of data from seven industrial countries, found similar changes in employment patterns across countries that increased diversity of employment methods and restricted employment flexibility through legislative means.
The following section provides very brief literature reviews of China’s labour market and employment versus security debates. We then turn to briefly introducing the role organisational neo-institutional theory plays in the article and our hypotheses. Subsequent sections present, respectively, the research design and methods, findings, discussion and conclusion.
Literature review
China’s employment contract system
In European countries, continuing employment is the norm, with fixed-term labour contracts, often defined as ‘temporary’, mostly the exception (Reference GuestGuest, 2004). When fixed-term contracts expire, employers are not subjected to the dismissal protection system. Contract renewals or extensions are strictly limited, and employers can only terminate these contracts for serious employee misconduct or force majeure (Reference CountourisCountouris, 2016; Reference SchÖ mann, Rogowski and KruppeSchomann et al., 1998). Under continuing contracts, employees can only be dismissed with legitimate reasons, restricting employers’ prerogatives (Reference Banker, Byzalov and ChenBanker et al., 2013; Reference PissaridesPissarides, 1999; Reference QianQian, 2011).
In contrast, the Chinese labour law neither specifies a fixed-term contract’s scope, nor restricts the maximum or minimum periods; the duration is entirely determined by the employer and employee, but mostly the former. The same conditions for relinquishment, economic compensation and financial compensation apply to fixed-term and continuing contracts. This has caused many obstacles in implementing continuing contracts (Reference Wu and LiWu and Li, 2013). In 2012, an ACFTU survey showed that 23.46% of managers and 15.59% of ordinary employees had signed continuing contracts. China’s Ministry of Human Resources and Social Security (Reference HuangHuang, 2016) found that only 17.9% of 3000 surveyed workers had signed a continuing contract. Thus, despite the new law, fixed-term contracts remain the main type of labour contract.
Continuing contracts: Security versus flexibility?
Some scholars argue that continuing contracts reduce an enterprise’s employment flexibility, increasing the cost of workforce adjustment. For example, Reference Gallagher, Giles and ParkGallagher et al.’s (2016) study found that 34.5% of managers felt that the ECL made it harder for enterprises to hire and dismiss workers. Reference YanYan (2015) claims that continuing contracts increase dismissal costs and recruitment thresholds, perhaps leading to employment system rigidities, with the risk of carrying China back to the ‘iron rice bowl’ era, in turn, discouraging hiring. Reference VennVenn (2009) believes that China’s continuing contracts impose employee protections well in excess of the OECD average. Moreover, when the intensity of law enforcement and effectiveness in implementation of the new law were positively correlated, Reference LiuLiu (2016) found that it reduced enterprises’ levels of employment.
Other research (e.g. Reference Wang, Song and ChengWang et al., 2016) suggests that implementation of the ECL has increased the proportion of migrant workers signing continuing contracts, thereby also narrowing the gap with urban workers. Reference Li and FreemanLi and Freeman (2015) investigated employment contracts signed in 2006, 2008 and 2009 by migrant workers in nine Pearl River Delta localities in South China. Here, the proportion of migrant workers with continuing contracts rose slightly after implementation of the new law: from 15.19% in 2006 to 17.35% in 2008, while it was 17.28% in 2009. Having analysed rural–urban migration data from 15 cities across 9 provinces, Reference MengMeng (2017) found that less than 2% of migrant workers had signed continuing contracts in 2005, but after the new law, this had reached 15%–20%. On the other hand, Reference He and LuoHe and Luo (2011) found that to avoid the ECL’s rigid requirements and the higher costs associated with continuing employment, enterprises often adopt discriminatory employment strategies by signing part-time contracts with migrant workers with negotiating capacities.
Furthermore, some studies indicate wide variations in the willingness of workers themselves to sign continuing contracts. This reflects the fact that while the ECL mainly seeks to bind employers, employees on continuing contracts must serve sufficient notice before leaving. Furthermore, continuing contracts often contain punitive clauses impeding employees joining a competitor company within a certain period after resigning. Thus, frontline workers are more likely to sign continuing contracts for their job security, and because their generic skills make them less mobile (Reference DongDong, 2016). On the other hand, competent employees with professional skills and management experience have strong competitive advantages and negotiation power (Reference GengGeng, 2014). ‘Top talents’ and skilled professional employees are less likely to desire being restricted by a continuing contract.
The ECL’s continuing contract requirements also differently affect diverse enterprises, by ownership and scale. In addition to their economic purpose, SOEs also play societal and public service roles, including promoting social harmony through secure employment. As such, SOEs are more likely to sign continuing contracts with their employees. Conversely, small and micro private enterprises, often having restricted finances and in highly competitive markets, face financial pressures to remain nimble and competitive and offer shorter fixed-term contracts (Reference ShengSheng, 2017).
While existing research provides solid foundations, it still displays shortcomings. First, there is a lack of comprehensive research on the practical effects of the ECL. Most studies are qualitative, and where they are quantitative, their scope lacks nationwide survey data. One exception, Reference Wang, Song and ChengWang et al. (2016), does use such data but addresses different research questions, investigating the ECL’s effects on employers entering signed employment contracts rather than verbal ones. Second, previous studies focus primarily on differences in rates of signing of continuing contracts among workers with different household registrations. Third, previous studies are mainly descriptive, making little attempt to interpret their phenomena using theory.
This research makes several important contributions in addressing these shortcomings. First, we quantitatively analyse a huge survey sample of 80,000 workers. Second, this study has controlled the factors affecting the likelihood of signing continuing contracts by individuals, enterprises and regions. It extends the field by examining other characteristics such as employee’s length of service and rank, and ownership type of employer. To do this, we construct a logistic regression model, verifying the main effect of the ECL and the moderating effects of corporate ownership and employee characteristics. This helps produce more accurate findings. Third, in China’s labour market, an employer’s decision of whether to sign continuing contracts depends greatly on the organisational characteristics and behaviours. Previous studies pay little attention to investigating this aspect. Using organisational neo-institutional theory (institutional theory), we explore the reasons why employers may accede to or avoid the ECL’s implementation regarding continuing contracts, and how such avoidance may transpire.
Theory and hypotheses
Impact of the ECL on signing continuing contracts
Institutional theory provides an important perspective for understanding, analysing and predicting organisational behaviour in organisational research. It argues that laws, rules, beliefs and other systems in the external environment constrain organisation behaviour. The institutional environment requires an organisation to conform to legitimating mechanisms and adopt practices that are taken for granted in their institutional environment, regardless of whether these practices affect the efficiency of its operations (Reference ScottScott, 2008). While, as a field, institutional theorising is now highly developed and increasingly complex, it has three core components (also known as three ‘pillars’): regulative, normative and culture-cognitive, and the ECL has enhanced the legitimating mechanisms of China’s institutional environment through all three elements, as we explain below.
Regulatory element
The new law expanded the scope of continuing labour contracts signed. It also increased penalties against enterprises flouting the law. If an employer does not sign continuing contracts, workers can directly sue it for double wages. Employers will tend to comply with the new law when they deem legally derived environmental pressures too high to neglect. These encourage them to sign continuing contracts even if it restricts their flexibility (Reference Li and ZhuLi and Zhu, 2017).
Normative element
Protecting labour rights is one critical aspect of corporate social responsibility (CSR). The ECL guides enterprises to guarantee employment security for workers, thereby helping align corporate behaviours with social norms and values. Therefore, constrained by ethics, enterprises must fulfil their responsibilities by signing continuing contracts with employees.
Culture-cognitive element
In responding to changes in the ‘rules of the game’ regarding continuing contracts, enterprises will promote common values, beliefs and behavioural orientations. They do this through particular cultural cognitions, either attaching great importance to following those regulations, or avoiding them.
For an example of avoidance, Huawei management was concerned that under the new law, old employees would become a ‘burden’ very difficult to get rid of. So, the company started a ‘7,000 people resignation campaign’, encouraging employees with close to 8 years’ service to resign in return for compensation. Thereafter, those workers would receive fixed-term contracts as new employees. As a prominent privately owned and strategically important Chinese multinational company, Huawei’s decision to ‘circumvent’ the new law was not made lightly. Senior management understood the risk and yet proceeded, apparently perceiving as too heavy the potential burden of continuing contracts (among other restrictive conditions; Reference Wu and PengWu and Peng, 2007).
Another challenge results from cognitive biases. Workers’ perceptions of the meaning of continuing contracts have deviated from the original legislative intentions, increasing employers’ practical difficulties with dismissals. In a survey of over 3000 workers, 29.18% simply understood continuing contracts as akin to the ‘iron rice bowl’ (Reference HuangHuang, 2016). In sum, the ECL imposed legislative constraints on the signing of continuing contracts through regulative, normative and culture-cognitive pressures. Based on the discussion above, we propose Hypothesis 1:
Hypothesis 1
The ECL significantly increases the likelihood of workers signing continuing contracts.
The moderating effect of enterprise ownership
Institutional theory suggests that different types of organisations in different operating environments tend to develop similar structures and behaviours due to pressures from their institutional environment (Reference Meyer and RowanMeyer and Rowan, 1977). By choosing similar organisational structures and behaviours, they gain societal recognition in order to obtain legitimate resources (Reference ScottScott, 2008). Reference DiMaggio and PowellDiMaggio and Powell (1983) analyse organisational mechanisms generating convergent behaviour (‘isomorphism’) under the influence of institutional environments. They suggest three mechanisms as discussed below.
Coercive convergence
Regulatory institutions coerce organisations to adopt certain structures or behavioural patterns. Under the influence of coercive mechanisms, organisations generate tendencies towards convergent behaviours. Indeed, observance of legality is conducive to enterprises acquiring key resources and enhancing their economic interests and competitiveness.
Mimetic convergence
Organisations will interpret and imitate the behaviours and practices of successful organisations in the field, thereby generating organisational convergence. This is a key to narrowing the gap between enterprises enjoying advantaged status and others (Reference Dacin, Oliver and RoyDacin, et al., 2007).
Normative convergence
The existence of shared cognitive patterns across society will also encourage organisational convergences. In China, prior to the new law, foreign-owned companies had the lowest percentage of continuing contracts and SOEs the highest. After the ECL’s implementation, the percentage of continuing contracts in enterprises under all types of ownership increased by between 3.1% and 7.1% (based on 2007 and 2012 ACFTU data). According to Reference Cheng and YangCheng and Yang’s (2010) survey of human resources (HR) managers in 600 Chinese enterprises, for continuing contracts and their ratio to fixed-term contracts after implementation, the gap among different types of enterprise ownership narrowed.
The new law has had less impact on SOEs as they already had higher rates of continuing contracts. Another reason may be that as SOEs can attract higher quality job candidates, it is in their interest to retain them through continuing contracts. Conversely, the new law has significantly shifted private firms’ employment contract policies: 66.7% of private enterprises extended the contract period of their workers after ECL implementation. It is important to note that enterprises of different ownership types also differ widely in terms of development status, characteristics and interests. Accordingly, their response and strategies will differ due to these factors when responding to changes in the new law. Thus, SOEs are usually influenced by the will and interests of the government. Therefore, they exhibit stable labour relations (Reference DingDing, 2010). On the other hand, non-SOEs are likely to conform through behaviours similar to SOEs under strict enforcement and supervision of the new law. Accordingly, our Hypothesis 2 is as follows:
Hypothesis 2
The moderating effect of enterprise ownership is significant, and the new law significantly increases the likelihood of non-SOEs signing continuing contracts.
Moderating effects of hukou identity, professional title and length of service
Before the ECL, the length of a worker’s service was an important factor affecting signing of continuing contracts. The 1995 Labour Law did not specify financial penalties for employers for not signing continuing contracts. This lowered the transaction costs of non-compliance, encouraging opportunistic behaviours among enterprises in regard to continuing contracts. Companies were more likely to sign continuing contracts with employees with high human capital. This focus on those who helped maximise profits the most also reduced each company’s employment transaction costs.
Many studies suggest that China’s vulnerable labour market groups, notably migrant workers and low-level frontline workers, tend to have low human capital, are easily substitutable and experience high inter-job mobility. In highly competitive labour markets with excess low-skilled labour supply, they are in inherently weak positions. Prior to the ECL, they were more likely to suffer discrimination over access to continuing contracts (Reference LuoLuo, 2008).
The ECL’s mandatory stipulation regarding continuing contracts represents top-down institutional change aimed at this labour market failure. One focus of the ECL was to protect those workers’ rights and interests, thereby narrowing the employment security gaps between disadvantaged (low-value) and advantaged (high-value) workers (Reference Feng, Wang and SunFeng, 2009). A second aim was for the law to apply fairly and equally to all workers, regardless of hukou identity and professional title (Reference WangWang, 2013).
Reference PengPeng (2003) argues that, in emerging markets, those organisations facing dramatic changes in their institutional environments, when making strategic choices, would first consider those institutional factors affecting their overall environment. Indeed, implementation of the new law restricts employers’ flexibility in selectively signing continuing contracts. Therefore, in the signing of continuing contracts, the ECL reduces the likelihood of firm-level discrimination based on hukou identity, professional title and length of service. Accordingly, we propose the following hypotheses:
Hypothesis 3
The moderating effect of workers’ hukou identity is significant, and the new law significantly improves the likelihood of signing continuing contracts among migrant workers.
Hypothesis 4
The moderating effect of workers’ level of professional title is significant, and the new law significantly improves the likelihood of signing continuing contracts among low-level workers.
Hypothesis 5
The moderating effect of workers’ length of service is significant, and the new law significantly improves the likelihood of signing continuing contracts among workers with shorter lengths of service.
Research design and methods
Data
The study used a dataset of two surveys conducted by the ACFTU. One took place just before the ECL (the sixth survey of Chinese employees in 2007) and the other was undertaken 4 years after the ECL’s implementation (the seventh survey of Chinese employees in 2012). This dataset surveys current employees; it does not contain data on those who were not employed or dismissed. In 2007, using random sampling techniques, the ACFTU statistical office and State Statistical Bureau jointly conducted a national employee questionnaire investigation of a total of 42,000 employees in 15 provinces (regions, municipalities): Beijing, Shanxi, Heilongjiang, Shanghai, Jiangsu, Zhejiang, Anhui, Henan, Hubei, Guangdong, Sichuan, Yunnan, Shannxi, Gansu and Xinjiang.
For the 2012 survey, the ACFTU statistical office organised general unions in every province (region, municipality), to conduct a national employee questionnaire sampling investigation of a total of 45,000 employees in 15 provinces (regions, municipalities): Beijing, Inner Mongolia, Liaoning, Jiangsu, Zhejiang, Anhui, Fujian, Shandong, Henan, Hubei, Guangdong, Chongqing, Sichuan, Shanxi and Gansu. For the survey, they selected 40 offices, respectively, in each provincial city and, prefecture-level city, plus 10 offices in each county (city). In a municipality, they selected 40 offices in an urban district and 10 offices in a suburban district (county). In all, 10 employees were selected in each office. The valid response rates were 100% and 99.6%, respectively. Descriptive statistics of sample characteristics are shown in Table 1.
Table 1. Descriptive statistics of sample characteristics.
Measurement
Dependent variable
Whether to sign a continuing labour contract
In both 2007 and 2012, all respondents were asked the term of the employment contract they had signed with their employers, with answer options of ‘less than 1 year’, ‘1 to 3 years’, ‘above 3 years’, ‘until completion of specified project/task’ and ‘continuing’. For the requirements of this project, we divided the answers into two categories: to sign a continuing labour contract, or not. ‘The term until completion of a specified project/task’ was taken as a deficiency. Thereafter, we created a dichotomous variable: 0 = signed a fixed-term contract; 1 = signed a continuing contract. We used the likelihood of signing a continuing contract to indicate employers’ employment flexibility. Specifically, the larger the likelihood of signing a continuing contract, the lower the flexibility.
Predictor variables
Enforcement of the ECL
The ECL took effect in 2008. This study used a dichotomous variable to represent the period before and after the new law: 0 = 2007 and 1 = 2012.
Hukou identity
This study divided household registration status into urban employees and migrant employees. Urban employees included local and non-local non-agricultural household registers. Migrant employees included local and non-local agricultural household registers. We then created a dichotomous variable: 0 = migrant employee and 1 = urban employee.
Type of enterprise ownership
The type of enterprise ownership included state-owned, collectively owned, privately owned and foreign-invested enterprises. We divided them into state-owned and non state-owned enterprise, with dichotomous variable: 0 = non-SOE and 1 = SOE.
Level of professional title
Both survey waves asked a question about, ‘the highest professional title you own currently’. Options included ‘no title’, ‘junior title’, ‘middle title’ and ‘senior title’. The professional title was an ordinal variable.
Tenure
The variable of seniority was a continuous variable written in by employees.
Control variables
Besides the predictor variables above, other factors, like individual characteristics of employees, enterprise characteristics and regional factors, may also influence whether employees sign continuing contracts. We therefore selected gender (0 = female; 1 = male), age and education (0 = below university; 1 = beyond university), whether the enterprise had a trade union (0 = has a trade union; 1 = no trade union) and industry type (1 = primary industry; 2 = secondary industry; 3 = tertiary industry) as control variables.
Analysis
As the dependent variable was dichotomous, we tested our hypotheses using logistic regression analysis. The logistic regression coefficient represented the prediction effect of the logarithmic odds ratio of the independent variable to the dependent variable. The positive regression coefficient represented the positive effect of the independent variable on the dependent variable, and the negative regression coefficient represented the negative effect of the independent variable on the dependent variable.
In the first step, we defined whether employees had signed a continuing contract as the dependent variable, ECL enforcement as the independent variable, and put all the control variables into the logistic regression to test whether the new law had had an impact on the likelihood of an employee signing a continuing contract. Based on this first step, the interaction items of moderator variables and independent variable were entered in step 2 to test whether the corresponding moderating effects were significant.
Findings
First, multiple collinearity analyses showed that the variance inflation factor (VIF) values of all variables were less than 2 (Table 2). This indicated that there was no multiple collinearity issue, allowing for subsequent correlation and regression analysis. Table 2 contains the correlation analysis between research variables.
Table 2. Correlation analysis.
VIF: variance inflation factor; ECL: Employment Contract Law; SOE: state-owned enterprise.
* Represents significant correlation at level of 0.05 (both sides).
** Represents significant correlation at level of 0.01 (both sides).
As Table 3 shows, model 1 indicates that after the ECL’s implementation, the likelihood of signing a continuing contract increased significantly (b = .292, p < 0.01). This would indicate that the enforcement of the new law had a negative influence on employers’ flexibility. Therefore, Hypothesis 1 was supported.
Table 3. Results of logistic regression analysis.
SOE: state-owned enterprise; ECL: Employment Contract Law.
** p < 0.01.
* p < 0.05.
In model 2 in Table 3, the interaction item (enforcement of the law × type of enterprise ownership) was significant (b = −.208, p < 0.01). Furthermore, Figure 1 indicates that the ECL increased the likelihood of SOEs signing continuing contracts, but it also significantly increased the likelihood of non-SOEs signing continuing contracts, thus supporting Hypothesis 2. This outcome has narrowed the employment security gap between these two ownership types. It also means that enterprise ownership has a moderating effect on the relationship between enforcement of the new law and likely employer flexibility (Figure 2).
Figure 1. Effect of interaction between ECL and enterprise ownership type.
Figure 2. Effect of interaction between ECL and hukou identity.
According to the results in models 3 and 6 in Table 3, Hypothesis 3 was also supported. The interaction item (enforcement of the law×household registration) was significant (b = −.663, p < 0.01). This means that employees’ household registration moderated the relationship between enforcement of the new law and continuing employment. In this, the new law has a stronger influence on the likelihood of migrant (rather than urban) employees signing continuing contracts, reducing this important employment security gap between the two groups.
The result for Table 3’s model 4 indicates that the interaction item (enforcement of the law × professional title) was significant (b = −.089, p < 0.01). This means that employees’ level of professional title has a moderating effect on relationships between enforcement of the new law and the likelihood of continuing contracts. Here, the new law significantly improves the likelihood of low-level workers signing continuing contracts. Thus, Hypothesis 4 was supported. Figure 3 shows the ECL slightly increasing the likelihood of workers, across professional levels, signing continuing contracts. Indeed, the likelihoods for all levels of workers converge after the ECL’s implementation, offering similar levels of employment security for workers across employment hierarchies.
Figure 3. Effect of interaction between ECL and level of professional type.
Hypothesis 5 was also supported according to the results in models 5 and 6. The interaction item (enforcement of the law × seniority) was significant (b = −.015, p < 0.01). Here, seniority (length of service) moderated the relationship between the new law and the likelihood of signing continuing contracts. Furthermore, the new law significantly improved the likelihood, of signing continuing contracts, among workers with shorter length of service. As Figure 4 shows, it increases this likelihood for all workers, regardless of the length of service, but narrows the gap between workers with short and long lengths of service. It thus supports our Hypothesis 5.
Figure 4. Effect of interaction between ECL and length of service.
Discussion
The new law has increased labour market equity
This study showed that the new law significantly increased the likelihood of signing continuing contracts between 2008 and 2012. The law particularly addresses entrenched forms of structural inequality across China’s workforce: hukou identity, employment in SOEs versus private firms, level of professional title and length of service. In those 4 years, the ECL reduced differences in the likelihood of signing continuing contracts across all these categories. As a result, negative implications of short-term contracts affected fewer workers, with particular improvements for more vulnerable workers. Our research reveals that the new law increased the proportions of private sector workers, migrant workers, those with low levels of professional title and those with shorter length of service signing continuing contracts.
These promising results appear to reflect the power of institutional pressures inherent in the ECL’s enforcement. The legislative mechanisms providing this pressure include provisions to allow workers to unilaterally press to conclude continuing contracts, which restrict the number of times fixed-term contracts can be renewed, and which stipulate double compensation to workers when enterprises refuse to award continuing contracts under certain circumstances. These punitive measures pushed enterprises to sign continuing contracts with qualified workers. In sum, the new law changed the environment of Chinese labour relations. Firms’ employment focus shifted from ‘giving priority to efficiency with due consideration to fairness’ to ‘balancing efficiency and fairness’. In other words, an across-the-board increase in signing of continuing contracts, rendered more ‘fairness’ through improved job security. From employers’ perspectives, this may have come at the expense of their flexibility. We can only assume that this process has continued with recovery from the global financial crisis and continued central government emphasis on social harmony, including through further legislation (see below).
Divergence in employers’ strategic choices
The new law has brought positive impacts in terms of labour market fairness. Gaps in signing continuing contracts among workers in different household registrations, rankings, working ages and employed in different types of enterprise ownership still exist. This phenomenon can be explained by two factors. First, in the process of generating legal pressure on enterprises, the system is subjected to the inertia resistance of different enterprises’ ‘efficiency logics’. In a modern society, organisations are embedded in increasingly diversified institutional logic conflicts that are not only influenced by the logics of the government and law, but also face the pressure of market logics (Reference Lounsbury and BoxenbaumLounsbury and Boxenbaum, 2013).
The new law strengthens firms’ focus on a ‘legitimacy’ logic, but it may also conflict with their long-term ‘efficiency’ logics in two ways. First, continuing contracts directly or indirectly reduce employment flexibility, potentially increasing short-term labour costs and employment risks. In these ways, they may hinder an organisation’s cost minimisation and/or profit maximisation goals. Second, each employer’s judgements and perceptions of the law’s effect may ultimately reflect on whether it effectively influences managerial decisions more generally. Organisational responses may thus depend heavily on the perceived effectiveness of the law’s implementation and, in the Chinese context, of governmental resolve regarding enforcement.
When strong laws are enacted and strictly enforced, employers are more likely to perceive they have no option but to comply. When this perception is widely shared, it reduces the likelihood of inter-employer competition based on regulatory avoidance, making compliance easier for more firms. Here legitimacy logic trumps market logic. Conversely, market logic is likely to prevail when laws are weak and/or inadequately enforced (Reference GodardGodard, 2002). Then, employers view non-compliance as a viable option. Even those who might otherwise comply may choose avoidance (and market logic) once they see that choice is both prevalent and unpunished. That is why the Huawei example, discussed earlier, was important. Given the diversity and unevenness of labour law enforcement regimes across China’s provinces (Reference Hendrischke, Sheldon, Kim and LiHendrischke, 2011), we would also expect to see great variation in outcomes.
The type of enterprise ownership remains important in affecting the possibility of signing continuing contracts. Here, the strategic choices of SOEs, foreign-owned enterprises and small private enterprises differ widely. Even so, even SOEs, with their strong government links and reliance on the legitimacy logic, have found methods to avoid compliance in order to reduce costs. In SOEs, continuing contracts are mainly intended for permanent employees. To mitigate the new law’s restrictive requirements, many SOEs converted long-service, low-level frontline workers into dispatched workers, creating a surge in the number of China’s dispatched workers. According to Reference HuangHuang (2016), there were very high proportions of dispatched workers in a number of prominent SOEs, including China Mobile (62%), China Post (52.6%) and Sinopec (40%).
To protect vulnerable workers, a July 2013 amendment to the new law stipulated that enterprises can only use dispatched workers in temporary, auxiliary and alternative positions, and dispatched labour should not exceed 10% of the total workforce. To legally avoid this new requirement, many SOEs converted dispatched labour into outsourced workers. Due to their need to meet the legitimacy criteria in China, foreign-owned enterprises tend to be more compliant with the new law. For their part, small private enterprises usually last only between 3 and 6 years because they operate in highly competitive markets and have limited resources. As such, even if they do sign continuing contracts, for many of these employees, this does not bring long-term employment security.
Furthermore, individual characteristics such as workers’ household registration (hukou), value (professional title) and working age are also important factors affecting the possibility of signing continuing contracts. Demand exceeds supply for high-value and urban workers, who usually have higher human capital. It is in the employer’s vested interests to sign continuing contracts with them. Conversely, low-level frontline and rural workers create less value for the enterprise and are easily replaceable. As such, employers are less willing to sign continuing contracts with them. Compared with workers with many years of service, the employment security of workers with shorter service is more significantly influenced by the new law. In terms of working age, there is also evidence that continuing employees are usually older than fixed-term ones (Reference Brown and SessionsBrown and Sessions, 2003).
Conclusion
Our study illustrates how the new law has clearly achieved important successes in protecting China’s most vulnerable workers. Yet, divergence in strategic choices among different types of employers in dealing with the provisions of the new law remains. This means that the work of Chinese legislators is an ongoing process, as was evident through the 2013 legislative amendment.
Crucially, the new law provides an impetus that fundamentally shifts some of the power imbalance from employers to employees. This is a significant and important achievement from two perspectives. First, if a labour market with underdeveloped institutional protection is left to free market forces, it will create opportunities for power imbalances and exploitation of weak members of the society. This is unsustainable in the long run. Employers get things done through workers. Exploited workers will not be happy workers. Unhappy workers are not likely to contribute to sustainable productivity. They can also be disruptive by participating in strikes, as is evident in the increasing labour unrest across China in recent years (Reference Elfstrom and KuruvillaElfstrom and Kuruvilla, 2014). Ultimately, it is not in the interests of any of the parties to have widespread labour unrest.
As the Chinese economy progresses and matures, sustainable development of the labour market is crucial to the economy’s sustainable development. Second, there have been widespread misconceptions among employers that pressure for continuing contracts brings the (disguised) return of the ‘iron rice bowl’ of the planned economy era, undermining their ‘efficiency’ logic. Here, the enforcement of the new law plays a crucial role to instil a sense of moral responsibility among employers and ‘forces’ them to see things from the workers’ perspective. This is important during a period in which the Chinese government is actively promoting social harmony − a central theme which will continue to feature prominently in Chinese society in the foreseeable future. The new law has demonstrated the Chinese government’s resolve in creating a more equitable and sustainable labour market that will in turn, contribute to social harmony and sustainable economic development.
From a firm-level perspective, one implication is that employers and their managers need to learn that cost-minimising labour strategies have their limits and that other strategies are available to satisfy both legitimacy and market logics facing firms. For example, effective HR management systems may fairly manage workers on continuing contracts and deliver high performance outcomes.
This study, while having important policy-related findings and conclusions, also suffers limitations. Our data analysis is mainly based at the individual (employee) level, rather than (also) at the enterprise level. This reflects the lack of enterprise information coding in the dataset. Future research may consider collecting data and undertaking analysis at the enterprise level.
Acknowledgements
We are grateful to the ACFTU for the data supplied. We would like to thank the journal’s two anonymous referees for their helpful feedback and journal editor, Peter Sheldon, for his guidance and assistance in revising the article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
