Nonstandard Employment and Job Satisfaction across Time in China: Evidence from the Chinese General Social Survey (2006

Abstract

This study examines the association of nonstandard employment with job satisfaction over time in China. An analysis is carried out using the Chinese General Social Survey (CGSS), a large cross-sectional survey that collected data from over 3000 workers across different industries, ownership types and regions in China in 2006, 2008 and 2012. The empirical results show that in 2006, nonstandard employment workers, on average, were less satisfied with their jobs than their counterparts in standard employment, all else being equal. However, these differences in job satisfaction became very small and insignificant in 2008 and 2012. The results from the propensity score matching exercise provide a similar conclusion. These findings suggest that improvements in regulations and employment relations in China have increased job satisfaction for nonstandard employment workers.

Keywords

China job satisfaction Labour Contract Law nonstandard employment

Introduction

Nonstandard employment, also known as informal employment, contingent work, alternative work arrangement or atypical employment, has become prevalent within the global economy over the last several decades due to an increase in global competition in the international market. Similarly, since China shifted from a centrally planned system towards a market-oriented system in the late 1970s, nonstandard employment, such as self-employment, temporary work and dispatch agency work (i.e. similar to temporary-help-agency work in Western countries), has increased dramatically in the urban labour market, becoming a major contributor to the country’s rapid economic development. For example, Wang (2013) provides statistics showing that the number of workers in nonstandard employment increased from 29.8 million in 1990 to 226 million in 2011. Zhou (2012) reports that approximately 60% of workers in the Chinese labour force were engaged in nonstandard employment in the mid-2000s. Owing to these growing numbers, the use of nonstandard employment in the labour market has attracted much attention from both academic researchers and policymakers seeking to identify its consequences.

Although the growth of nonstandard employment arrangements has benefited Chinese businesses and the nation’s economy by helping to generate employment (i.e. lower unemployment), reduce labour costs and enhance labour market flexibility, much evidence shows that this type of employment also causes many problems for workers. Their work is characterized by low pay, high intensity, poor conditions and lack of employment protection and bargaining power (Kalleberg and Hewison, 2013). According to legal requirements in many countries (e.g. the Equal Pay Act in the US and UK), employers should distribute equal pay and benefits to all workers for equal work. In reality, salaries, benefits and other work welfares are often unequally distributed. In particular, most nonstandard employment workers still receive lower pay and are outside the coverage of the welfare system (Kalleberg and Hewison, 2013). In addition, they also tend to lack a sense of security and more frequently intend to quit their job than do those who enjoy standard employment (Zeytinoglu et al., 2013). Such perceptions and experiences, therefore, could cause differences in work attitudes and behaviours between nonstandard employment workers and their standard employment counterparts.

Most of the existing empirical research on nonstandard employment in China focuses on the disparities in economic or monetary outcomes, such as wages, between nonstandard employment workers and their counterparts in standard employment (e.g. Chen and Hamori, 2013; Xue et al., 2014; Zuo, 2013). Other previous studies are based on case studies, individual sectors or single regions. However, scholars have carried out relatively few empirical studies of non-economic or attitudinal outcomes of nonstandard employment, including job satisfaction. Job satisfaction is shown to be one of the most important attitudinal factors associated with improved individual and organizational performance (Judge et al., 2001). Therefore, this study intends to fill this gap by examining the relationship between job satisfaction and nonstandard employment over time in China. In particular, the main research questions of this study include: Are workers in nonstandard employment more likely to report lower job satisfaction than those who are employed in standard employment? And where such differences exist do they persist across time? This study focuses on the 2006 to 2012 transitional period of labour market and employment relations in China, when the Chinese government implemented several new labour laws and regulations, including in 2008 the Labour Contract Law, which aims to improve the working conditions and benefits of both standard and nonstandard employment workers.

This study utilizes the Chinese General Social Survey (CGSS), a large cross-sectional survey that in 2006, 2008 and 2012 collected data from over 3000 workers across different industries, ownership types and regions in China. This study’s sample size is considerably larger than previous studies in the Chinese setting. In a key contribution to the literature, this study sheds light on nonstandard employment in China and its consequences by empirically testing the association of nonstandard employment with job satisfaction across time. This study also makes a theoretical contribution to the literature by providing a contextualized understanding of nonstandard employment and labour-related laws in China, particularly during the period of policy and regulation reforms between the mid-2000s and the early 2010s. The next section reviews the relevant literature on nonstandard employment and its consequences. Thereafter, the dataset and methodology, key empirical results and conclusions are presented.

Literature review

Definition of nonstandard employment in China

Broadly speaking, standard employment refers to a regular or full-time job under contract with a single employer, performed at the employer’s place of business and under the employer’s direction (Kalleberg, 2000). In addition, a standard employment worker is provided with fringe benefits and promotional opportunities. In contrast, nonstandard employment generally refers to an unstable or precarious job that is not regulated or protected by legal frameworks (Kalleberg, 2000; Kalleberg and Hewison, 2013). In addition, many types of nonstandard employment, such as temporary work and contractors (including both employees of contract companies and independent contractors), are not located in the organization’s core business.

In China, scholars have yet to agree on a definition of nonstandard employment (Cooke, 2011a). For example, Wu and Cai’s (2006) definition of nonstandard employment in China includes self-employed workers, workers without a labour contract, migrant workers, contributing family workers, temporary workers, dispatch agency workers and public service workers. Cai and Wang (2004) define nonstandard employment as those who are not registered with the government or do not participate in social insurance. The definition of nonstandard employment in this study closely resembles that of Xue et al. (2014), who follow International Labour Organization (ILO) (2003) guidelines for measuring nonstandard employment. In particular, two main sector types in the labour market are identified: the standard (formal) sector and the nonstandard (informal) sector. The standard sector refers to governmental, institutional and social organizations, as well as state-owned enterprises, collective-owned enterprises, joint venture, foreign-owned enterprises and private-owned enterprises. The nonstandard sector includes individual businesses (i.e. businesses with a small number of workers and businesses whose workers are not registered with the administration office) and households (households that produce goods for their own use and those that employ paid house work).¹ Accordingly, in this study, nonstandard employment in urban China includes the following four types of workers: (1) contributing family workers, freelancers, temporary workers, part-time workers and dispatch agency workers; (2) self-employed workers; (3) employers in the nonstandard sector; and (4) workers who have not signed a labour contract with their employer either in the standard or the nonstandard sector.

Nonstandard employment and labour-related legislation in China

Under China’s centrally planned economy initiated in 1949, all enterprises were owned or under the control of the government. Therefore, most urban workers were employed by state-owned enterprises with a guarantee of lifetime job security, pension, healthcare and housing (the so-called ‘iron rice bowl’ system). During this period, in other words, almost all workers were employed in standard employment (Swider, 2015). However, following the period of economic reform, particularly after the mid-1990s, employment relations and the labour market in China underwent several fundamental changes, including a growing number of private-owned, joint venture and foreign-owned enterprises; a widespread downsizing and shutdown of many state-owned enterprises, creating a large body of laid-off workers; and a large influx of rural migrants who moved to find jobs in urban areas (Park and Cai, 2011). Both rural-to-urban migrant workers and workers laid-off from state-owned enterprises were forced to seek jobs in nonstandard employment. Likewise, many private enterprises have evolved away from the standard and long-term employment arrangements that were so common during the centrally planned economy. For example, dispatch agency work (i.e. agency companies that send their employees to work in client firms) has become very common for enterprises that seek labour flexibility in China (Wang et al., 2016). On the one hand, nonstandard employment has benefitted the Chinese economy: it helped to create jobs that absorbed surplus workers, including rural migrant workers in particular, and it increased labour market flexibility. On the other hand, evidence shows that nonstandard employment workers in China are less well-treated, work in poorer conditions and receive lower (often delayed) pay and benefits than those in standard employment (Cooke, 2011a). Moreover, the legal system, which relies largely on a 1994 Labour Law that applies mainly to those in standard employment relationships, offers little support. The law fails to address the growing tensions that are emerging from the privatized economy and the growing informalization of employment. In fact, it is not clear whether certain laws and regulations should apply to nonstandard employment workers. As a result, many employers take advantage of these loopholes (Cooke, 2008), including the non-provision of employment contracts, the underpayment of wages and non-contributions to social insurance (Park and Cai, 2011; Zuo, 2013). Therefore, workers engaged in nonstandard employment might have lower levels of attitude, particularly job satisfaction, than those employed in standard employment.

Responding to growing concerns about the lack of clear regulations that protect workers’ well-being, the government took action in the late 2000s to improve the situation. In 2008, for example, the government introduced a series of new laws and regulations, including Labour Contract Law, Employment Promotion Law, and Labour Disputes Mediation and Arbitration Law, that aimed to create more equitable labour rights and standards for different employment types and provide better protection to those in nonstandard employment (Cooke, 2011b; Remington and Cui, 2015). The main objective of these laws and regulations in the area of nonstandard employment is to improve employment conditions, offer better employment protection and provide more comprehensive guidance than the 1994 Labour Law. For example, the Labour Contract Law was adopted at the 28th Session of the Standing Committee of the Tenth National People’s Congress of the People’s Republic of China on 29 June 2007, and officially came into force on 1 January 2008. This law is intended to specify the rights and obligations of labour contract parties and give workers more protection. It explicitly presents a probation period, a minimum wage and maximum hours of work for those with nonstandard employment. It also expands the eligibility for social insurance to all kinds of workers, including those in nonstandard employment (Park and Cai, 2011). For example, Freeman and Li (2013) show that the percentage of workers with social insurance coverage increased sharply from 40% in 2006 to 60% in 2008. The trend implies that the new labour law had an immediate impact, improving employment conditions and, consequently, potentially enhancing workers’ attitudes.

Previous studies of the consequences of nonstandard employment

Previous studies of the association between nonstandard employment and workers’ job satisfaction find mixed results.² Indeed, many studies carried out in various countries find a negative relationship between nonstandard employment and job satisfaction. For example, Booth et al. (2002) find that seasonal temporary workers in Britain are less satisfied with their jobs than permanent and fixed-term workers. Similarly, Zeytinoglu et al. (2013) and Aletraris (2010), who analyse Australian data, find that due to the instability and uncertainty of temporary employment, temporary workers are less satisfied with their jobs than permanent workers. Possible explanations of this lower job satisfaction are as follows. First, nonstandard employment is often temporary, and, therefore, it does not offer job security. Also it generally lags behind standard employment in social and legal protections. As a result, this type of employment could lead to insecurity and anxiety among workers, which would lower overall job satisfaction. Second, employers may treat nonstandard workers differently than regular workers, making nonstandard workers feel marginalized. Third, employers are less likely to invest in nonstandard workers because the latter establish a relatively short employment relationship, which reduces opportunities for skill development and career progression. Fourth, working conditions for nonstandard workers are relatively poorer than those in standard jobs. Other studies, in contrast, find a positive relationship between some categories of nonstandard employment and job satisfaction. For example, Wooden and Warren (2004), who examine Australian data, find that workers do not necessarily regard nonstandard employment as undesirable. In fact, their results show that workers on a fixed-term contract are more satisfied with their jobs than other types of workers, including permanent workers. A possible explanation for this relationship is that although nonstandard employment is generally characterized by poor working conditions, low wages and little social and legal protection, some workers may prefer the greater freedom, autonomy and flexibility to balance work and family that these positions provide. For these individuals, nonstandard employment jobs are preferable. In other words, individuals who work in nonstandard employment jobs do not always experience lower job satisfaction. Some voluntarily choose this type of employment because it offers advantages over standard employment. Still other studies find no difference between these two employment groups with regards to job satisfaction (e.g. Bardasi and Francesconi, 2004; Guest et al., 2006). Summarizing the empirical and theoretical viewpoints just described, the association of nonstandard employment with job satisfaction is not clear-cut because it is affected by individual preferences, expectations and financial constraints (Bardasi and Francesconi, 2004).

In China, research on nonstandard employment and how it differs from more traditional employment is gaining increasing recognition among practitioners and academic scholars. Whether workers in nonstandard employment are satisfied with their jobs becomes important when labour policies are designed that affect employment relations. Yet very little research examines whether the experience of workers in nonstandard jobs has any association with individual attitudes, such as job satisfaction. Most empirical studies of nonstandard employment in China focus on economic outcomes, such as wages, benefits and employment. Those that do examine the determinants of job satisfaction in the Chinese labour market focus only on rural migrants and laid-off state sector workers (Cheng, 2014). To address this gap in the literature, this study examines the extent to which nonstandard employment is associated with job satisfaction (a key aspect of workers’ well-being (Bardasi and Francesconi, 2004)) within the Chinese labour market, particularly during the period of policy and regulation reforms that lasted from the mid-2000s to the early 2010s. We hypothesize that nonstandard employment workers’ job satisfaction might be lower than that of workers in standard employment, particularly during the period prior to the 2008 labour-related law reforms. However, the level of job satisfaction among nonstandard employment workers should have significantly improved over time, in part because of the enactment of new laws and regulations, such as the 2008 Labour Contract Law and other related laws.

Empirical strategy and data set

Data and descriptive analysis

The CGSS is a nationwide, repeated and cross-sectional survey conducted by Renmin University of China. It contains detailed information about the lives and employment of a nationally representative Chinese sample across 30 provinces.³ The CGSS was formally launched in 2003 and has included eight waves: 2003, 2005, 2006, 2008, 2010, 2011, 2012 and 2013. This study uses the CGSS in 2006, 2008 and 2012 to examine the association of nonstandard employment with job satisfaction during the transitional period of labour laws and regulations. These three waves contain complete information about employment status, individual and job characteristics as well as job satisfaction. This study’s sample is restricted to urban residents (i.e. workers with urban residential permits (hukou) plus rural-to-urban migrant workers) between the ages of 18 and 65. Rural residents are excluded because the rural–urban dualism is the most basic characteristic of the Chinese economy and significant differences distinguish the two groups. For example, judging from the CGSS (2006–2013), the employment rate of rural residents is about 19%, whereas that of urban residents is 52%. Moreover, the vast majority of rural residents are engaged in agriculture jobs; thus, it seems unreasonable to pool these two groups and describe their attitudes in a single model.

To check the representativeness of the working samples in the CGSS, the distribution of several variables in the CGSS is compared with that of the Chinese Census data. As shown in Table 1 and Figure 1, the employment ratio of individuals with urban hukou calculated from each wave of CGSS and the Census are stable about 50% across years. Additionally, there are no significant differences in the male ratio, proportion of Han ethnicity⁴ or age distribution of the working population with urban hukou between the CGSS and the Census data. This evidence indicates that the sample used in this study can nationally represent individuals who have a job, although the subsamples of each wave are not large. After deleting observations with missing information on key variables, the sample sizes in 2006, 2008 and 2012 are 2222, 1878 and 1382, respectively.

Table 1.

Comparison of employment characteristics in the CGSS and Census data.

	Census 2000	Census 2005	CGSS 2006	CGSS 2008	Census 2010	CGSS 2012
Employment rate	55.58%	51.48%	49.77%	52.69%	52.18%	51.55%
Male ratio	58.98%	58.39%	55.64%	58.64%	57.84%	58.91%
Proportion of Han	93.56%	93.83%	95.70%	94.93%	95.26%	93.89%

Sources: CGSS and Census data.

Based on the information available from the CGSS, the nonstandard employment variable was constructed using the definition discussed in the previous section. Then, a dichotomous variable of nonstandard employment was created, taking the value ‘one’ if a worker is specified as a nonstandard worker and ‘zero’ otherwise. Job satisfaction, representing workers’ feelings about their current jobs, is measured in the survey by the question: ‘Overall, how satisfied or dissatisfied are you with your current job?’. The following options were offered: (1) very dissatisfied; (2) somewhat dissatisfied; (3) somewhat satisfied; and (4) very satisfied. The first two categories are classified as dissatisfied with the job, while the last two categories are satisfied.

Figure 2 provides a first look at nonstandard employment in China based on the CGSS. In particular, it presents the incidence of nonstandard employment in China from 2006 to 2013. The proportion of nonstandard employment in 2006 was about 57%; it then plunged to around 46% in 2008. After that, it stayed quite stable at around 49%. One possible explanation for the decline in nonstandard employment during this period is that the Labour Contract Law and other related laws in 2008 contributed to a reduction in the number of people working in nonstandard employment. Figure 2 also illustrates the prevalence of nonstandard employment by hukou, which refers to the registered place of residence, including rural and urban hukou. Note that under the centrally planned system workers holding rural hukou were not allowed to move to urban areas to find jobs. Since the hukou reform in the late 1970s, rural workers have been able to move to urban cities and hold jobs (as so-called ‘rural migrant workers’). However, they are not eligible to receive urban welfare benefits.⁵ This group has grown in size from the mid-1980s to the present. Figure 2 also shows that rural migrant workers were more likely to engage in nonstandard employment than urban hukou workers.

Figure 3 illustrates the incidence of four types of nonstandard employment from 2006 to 2013. In 2006, Types 1 and 4 were 23% and 21%, respectively, of nonstandard workers, while around 9% and 4% of nonstandard workers were employed in Type 2 and Type 3 employment. A sharp decrease of Types 1 and 4 occurred in 2008. Type 2 employment jumped from 9% to 12% in 2008, fell sharply to around 8% in 2012, and then increased to 10% in 2013. At the same time, Type 3 slowly increased from around 4% in 2006 to approximately 6% in 2013.

Table 2 shows the summary statistics of the main variables used in this study. Approximately 65.7% of workers were satisfied with their job in 2006; that number dropped to 59.7% in 2008 and increased to 68.7% in 2012. Nonstandard workers comprised approximately 57.2% of the workforce in 2006, before a decline to 45.7% and 41.6% in 2008 and 2012, respectively. Of those in nonstandard employment, about 20% were rural migrant workers in 2006 and 2008; that number jumped dramatically to 36% in 2012.

Table 2.

Descriptive statistics.

Variables	2006			2008			2012
Variables	Obs	Mean	SD	Obs	Mean	SD	Obs	Mean	SD
Dependent variable
Job satisfaction (dummy)	2222	0.657	0.475	1878	0.597	0.491	1382	0.687	0.464
4 types of job satisfaction:
1. Very dissatisfied with job (dummy)	2222	0.057	0.231	1878	0.017	0.129	1382	0.222	0.416
2. Somewhat dissatisfied with job (dummy)	2222	0.287	0.452	1878	0.386	0.487	1382	0.090	0.287
3. Somewhat satisfied with job (dummy)	2222	0.607	0.489	1878	0.563	0.496	1382	0.501	0.500
4. Very satisfied with job (dummy)	2222	0.050	0.217	1878	0.034	0.181	1382	0.186	0.389
Independent variables
Nonstandard employment (dummy)	2222	0.572	0.495	1878	0.457	0.498	1382	0.416	0.493
Nonstandard employment type:
Type 1	2222	0.204	0.403	1878	0.165	0.371	1382	0.127	0.333
Type 2	2222	0.098	0.298	1878	0.117	0.322	1382	0.068	0.252
Type 3	2222	0.049	0.216	1878	0.037	0.189	1382	0.073	0.260
Type 4	2222	0.221	0.415	1878	0.138	0.345	1382	0.148	0.355
Male (dummy)	2222	0.563	0.496	1878	0.571	0.495	1382	0.591	0.492
Age	2222	36.401	9.745	1878	36.982	10.018	1382	39.596	10.472
Rural hukou (dummy)	2222	0.204	0.403	1878	0.193	0.395	1382	0.315	0.465
Han ethnicity	2222	0.949	0.220	1878	0.943	0.232	1382	0.948	0.222
Education status:
1. Illiteracy (dummy)	2222	0.013	0.114	1878	0.013	0.112	1382	0.013	0.113
2. Primary school (dummy)	2222	0.068	0.251	1878	0.093	0.291	1382	0.068	0.252
3. Junior high school (dummy)	2222	0.425	0.494	1878	0.370	0.483	1382	0.342	0.475
4. Senior high school (dummy)	2222	0.242	0.428	1878	0.231	0.422	1382	0.204	0.403
5. Above (dummy)	2222	0.253	0.435	1878	0.293	0.455	1382	0.373	0.484
Married (dummy)	2222	0.769	0.421	1878	0.794	0.404	1382	0.816	0.387
Working hours	2222	50.683	14.822	1878	50.090	15.051	1382	49.916	16.995
Last year real income	2222	37.056	40.241	1878	40.886	42.149	1382	62.217	64.010
Firm size:
1. Small	2222	0.328	0.469	1878	0.278	0.448	1382	0.286	0.452
2. Medium	2222	0.318	0.466	1878	0.320	0.467	1382	0.365	0.482
3. Large	2222	0.354	0.478	1878	0.402	0.490	1382	0.349	0.477
Ownership type:
1. State-owned (dummy)	2222	0.390	0.488	1878	0.403	0.491	1382	0.339	0.473
2. Collective-owned (dummy)	2222	0.095	0.294	1878	0.058	0.233	1382	0.055	0.228
3. Private-owned (dummy)	2222	0.495	0.500	1878	0.499	0.500	1382	0.571	0.495
4. Foreign, Hong Kong, Macao, Taiwan (dummy)	2222	0.019	0.138	1878	0.040	0.197	1382	0.035	0.185

Next, the characteristics of standard employment and nonstandard employment are compared with a t-test. As shown in Table 3, job satisfaction in 2006 and 2008 was significantly higher for standard employment (75.4% in 2006, 64.1% in 2008) than for nonstandard employment (58.4% in 2006, 54.4% in 2008). However, these gaps converged over time until by 2012 there was no significant difference. In addition, the proportion of male workers, those holding rural hukou and Han workers engaging in nonstandard employment increased during the same period. Moreover, nonstandard workers tended to work longer, work in smaller working units and earn less income in the previous year than standard workers.

Table 3.

Differences in characteristics of standard employment and nonstandard employment (sample mean).

Variables	2006			2008			2012
Variables	Standard	Nonstandard	Diff	Standard	Nonstandard	Diff	Standard	Nonstandard	Diff
Dependent variable: Job satisfaction (dummy)	0.754	0.584	0.170***	0.641	0.544	0.097***	0.704	0.664	0.039
Dissatisfied with job (job satisfaction = 0):
1. Very dissatisfied with job (dummy)	0.024	0.081	−0.057***	0.011	0.024	−0.014**	0.211	0.238	−0.028
2. Somewhat dissatisfied with job (dummy)	0.222	0.335	−0.113***	0.348	0.431	−0.083***	0.086	0.097	−0.012
Satisfied with job (job satisfaction = 1):
3. Somewhat satisfied with job (dummy)	0.702	0.536	0.166***	0.605	0.513	0.092***	0.512	0.487	0.025
4. Very satisfied with job (dummy)	0.052	0.048	0.004	0.036	0.031	0.005	0.192	0.177	0.015

Control variables:
Male (dummy)	0.595	0.539	0.056***	0.593	0.545	0.048**	0.599	0.581	0.018
Age	36.893	36.034	0.859**	36.582	37.457	−0.875*	39.602	39.588	0.014
Rural hukou (dummy)	0.083	0.294	−0.211***	0.090	0.316	−0.226***	0.192	0.489	−0.297***
Han ethnicity (dummy)	0.961	0.940	0.021**	0.941	0.945	−0.004	0.936	0.965	−0.030**
Education status
1. Illiteracy (dummy)	0.002	0.021	−0.019***	0.004	0.023	−0.019***	0.011	0.016	−0.004
2. Primary school (dummy)	0.021	0.102	−0.081***	0.031	0.167	−0.135***	0.022	0.132	−0.110***
3. Junior high school (dummy)	0.334	0.493	−0.159***	0.288	0.467	−0.179***	0.240	0.485	−0.245***
4. Senior high school (dummy)	0.225	0.254	−0.029	0.239	0.221	0.018	0.207	0.200	0.007
5. Above (dummy)	0.418	0.130	0.288***	0.437	0.121	0.316***	0.519	0.167	0.352***
Married (dummy)	0.785	0.757	0.028	0.779	0.812	−0.033*	0.810	0.824	−0.014
Working hours	45.862	54.284	−8.422***	45.064	56.065	−11.002***	45.682	55.859	−10.178***
Last year real income	40.420	34.543	5.877***	48.459	31.884	16.574***	71.870	48.669	23.202***
Firm size:
1. Small	0.029	0.550	−0.521***	0.028	0.575	−0.546***	0.033	0.640	−0.607***
2. Medium	0.412	0.248	0.163***	0.385	0.242	0.143***	0.447	0.250	0.197***
3. Large	0.559	0.201	0.358***	0.586	0.183	0.403***	0.519	0.110	0.410***
Ownership type:
1. State-owned (dummy)	0.695	0.162	0.533***	0.622	0.143	0.478***	0.533	0.066	0.467***
2. Collective-owned (dummy)	0.135	0.066	0.069***	0.073	0.040	0.033***	0.077	0.024	0.052***
3. Private-owned (dummy)	0.133	0.767	−0.634***	0.240	0.807	−0.566***	0.338	0.897	−0.559***
4. Foreign, Hong Kong, Macao, Taiwan (dummy)	0.038	0.006	0.032***	0.066	0.010	0.055***	0.052	0.012	0.040***

Notes: *, ** and *** indicate the significance levels of 10%, 5% and 1%, respectively.

These comparisons of mean differences, however, do not account for the effect of other characteristics and may confound the association of nonstandard employment with job satisfaction when the influence of other characteristics is considered. To investigate the association of nonstandard employment with job satisfaction, the multivariate approaches are essential.

Empirical strategy

To determine the association of nonstandard employment with job satisfaction, the following multivariate regression was estimated across three waves:

Y_{i} = α_{0} + α_{1} Employment_{status}_{i} + α_{2} X_{i} + θ_{0} + λ_{p} + ε_{i}

(1)

where Y_i is the self-reported job satisfaction of person i. The key explanatory variable is Employment_status_i, which is the employment status for person i (standard or nonstandard employment) at the time of the survey. X_i is a vector of individual demographics and job characteristics, including gender, age, type of hukou, ethnicity, education level, marital status, working hours per week, income of last year, firm size measured by the number of employees in the working unit, the ownership type and occupation (θ₀) and province (λ_p) fixed effects. ε_iis an error term capturing unobserved individual characteristics. Additionally, since the proportion of different types of nonstandard employment exhibits various trends from 2006 to 2012, the key explanatory variable in baseline model (Employment_status_i) is replaced with four dummy variables corresponding to the subcategories of nonstandard employment and re-estimate Equation (1).

In addition, a propensity score matching method was adopted for a robustness check because, unlike the ordinary least square method (OLS), it does not require the researcher to define the exact functional form (Black, 2015). Instead, it balances the covariates non-parametrically during the matching process. Specifically, three matching methods were applied. The first is the nearest-neighbour matching method, which matches a standard employment sample with the closest propensity value as a control group for the nonstandard employment sample. Since this method ignores the information of unmatched observations in the standard employment sample, an additional two matching methods were estimated based on a nonparametric regression: the kernel matching and the regression-adjusted local linear matching. Both methods take into account all standard employment samples and weight them in their closeness to nonstandard employment samples, which are estimated by nonparametric regressions.

Empirical results

Baseline model

This section reports the main regression results. Table 4 shows the baseline results from the linear probability specification for 2006, 2008 and 2012 separately. In particular, columns (1) to (3) show that in 2006 nonstandard employment workers, all else being equal, were 9.9% less likely to be satisfied with the job than standard employment workers, while in 2008 and 2012 no significant differences existed in the level of job satisfaction between standard and nonstandard employment workers.⁶ In addition, and as mentioned earlier, the proportion of different types of nonstandard employment exhibited various trends from 2006 to 2012. A natural question is whether the job satisfaction of these four types also diverged over time. To address this issue, columns (4) to (6) present the results in the baseline model with four nonstandard employment dummy variables. In particular, column (4) shows that in 2006 the nonstandard employment workers who reported lower satisfaction with their job were mainly from Types 1 and 4, while there was no significant difference in job satisfaction between Type 2 and Type 3 workers, on the one hand, and standard employment workers, on the other hand. However, the coefficients of Types 1 and 4 became insignificant in 2008 and 2012 (columns (5) and (6)), which is in line with the baseline regression. In contrast, Type 2 and Type 3 nonstandard employment workers did not induce extra dissatisfaction across the period. That is to say, the baseline results were dominated by workers employed by others without formal employment relationships or labour contracts.⁷ Note that the non-significant estimates of Type 2 and Type 3 nonstandard employment could come from the low proportion of these two types in the sample. Therefore, these two groups were combined and the regression was re-estimated. The result remained the same, confirming the insignificant coefficient of these two types.

Table 4.

Baseline regression results (linear probability model).

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	2006	2008	2012	2006	2008	2012
Nonstandard employment	−0.099***	−0.036	0.042
Nonstandard employment	(0.028)	(0.032)	(0.038)
Employment status: 0. standard employment
1. Type 1				−0.132***	−0.050	0.051
				(0.041)	(0.042)	(0.053)
2. Type 2				−0.024	0.031	0.032
				(0.055)	(0.064)	(0.072)
3. Type 3				−0.030	0.099	−0.029
				(0.060)	(0.077)	(0.072)
4. Type 4				−0.091***	−0.038	0.045
				(0.030)	(0.039)	(0.043)
Rural hukou (dummy)	0.037	−0.012	0.034	0.035	−0.015	0.034
	(0.030)	(0.032)	(0.031)	(0.030)	(0.032)	(0.031)
Male (dummy)	0.015	−0.051**	−0.005	0.011	−0.054**	−0.005
	(0.021)	(0.025)	(0.029)	(0.021)	(0.025)	(0.029)
Age	−0.027***	−0.014*	−0.007	−0.027***	−0.014*	−0.007
	(0.008)	(0.008)	(0.009)	(0.008)	(0.008)	(0.010)
Age²	0.000***	0.000*	0.000	0.000***	0.000*	0.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Han ethnicity (dummy)	−0.039	−0.052	−0.038	−0.045	−0.050	−0.035
	(0.051)	(0.049)	(0.058)	(0.051)	(0.049)	(0.058)
Married (dummy)	0.069**	−0.037	0.113***	0.067**	−0.038	0.113***
	(0.027)	(0.033)	(0.040)	(0.027)	(0.033)	(0.040)
Firm size: 1. Small
2. Medium	−0.048	0.046	−0.006	−0.044	0.072	−0.019
	(0.035)	(0.039)	(0.041)	(0.041)	(0.048)	(0.049)
3. Large	−0.050	−0.020	0.023	−0.047	0.006	0.010
	(0.038)	(0.042)	(0.045)	(0.044)	(0.049)	(0.051)

Observations	2222	1878	1382	2222	1878	1382
R ²	0.149	0.099	0.115	0.153	0.102	0.116
Province fixed effect	Y	Y	Y	Y	Y	Y
Occupation fixed effect	Y	Y	Y	Y	Y	Y

Notes: Constants are included but not reported. The robust standard errors in parentheses are estimated to cover the possibility that the model’s errors are heteroskedastic, which will cause the estimated coefficients to be inconsistent. Not reported independent variables in this table are working hours, last year income, education level and ownership type. *, ** and *** indicate the significance levels of 10%, 5% and 1%, respectively.

To test whether the association of nonstandard employment with job satisfaction in 2008 and 2012 is significantly different from that in 2006, the data were pooled from the three survey years and then nonstandard employment was interacted in the regression with each survey year dummy (with 2006 as the reference). As shown in Table 5, in 2006, nonstandard employment workers, all else being equal, were more likely to report low job satisfaction than standard employment workers. However, the difference in job satisfaction between standard and nonstandard employment declined significantly in 2008 and almost reversed in 2012 when the coefficients of the interaction of nonstandard employment and the year dummies became statistically and significantly positive. This result confirms the proposed hypotheses.⁸

Table 5.

Linear probability model on cross-sectional data with interaction.

Variables	(1)	(2)	(3)
	All	All	All

Nonstandard employment	−0.143***	−0.102***	−0.112***
	(0.018)	(0.019)	(0.024)
Nonstandard employment * year 2006	–	–	–

Nonstandard employment * year 2008	0.050*	0.077***	0.083***
	(0.028)	(0.028)	(0.031)
Nonstandard employment * year 2012	0.086***	0.115***	0.134***
	(0.028)	(0.029)	(0.033)
Rural hukou (dummy)		0.015	0.023
		(0.016)	(0.018)
Male (dummy)		−0.024*	−0.015
		(0.012)	(0.014)
Age		−0.019***	−0.019***
		(0.004)	(0.005)
Age²		0.000***	0.000***
		(0.000)	(0.000)
Han ethnicity (dummy)		−0.045*	−0.044
		(0.027)	(0.029)
Married (dummy)		0.040**	0.041**
		(0.017)	(0.019)
Firm size: 1. Small
2. Medium			−0.019
			(0.021)
3. Large			−0.038
			(0.023)
Observations	6766	6310	5482
R ²	0.041	0.080	0.088
Year dummy	Y	Y	Y
Province fixed effect	Y	Y	Y
Occupation fixed effect	Y	Y	Y

Robustness check

One explanation of the apparent insignificance of nonstandard employment in 2008 and 2012 could be that the association of nonstandard employment with job satisfaction was mediated through individual and job characteristics (i.e. control variables in the regression) and the extent to which shifts in these characteristics across survey years might account for the findings. As shown in Table 6, the Oaxaca-Blinder decomposition (Oaxaca, 1973) was conducted across three waves of data to investigate the influence of the control variables on the relationship between nonstandard employment and job satisfaction. Overall, the results indicated that the differences in the individual and job characteristics between standard and nonstandard employment accounted for 49% of the job satisfaction differential in 2006, and it dramatically increased to 93% and 192% in 2008 and 2012, respectively. This implies that the change in the differences in job satisfaction between standard and nonstandard employment across years can be attributed in part to shifts in the observed individual and job characteristics of these two types of employment.

Table 6.

Oaxaca-Blinder decomposition of job satisfaction between standard and nonstandard employment workers.

	2006	2008	2012
Standard employment	0.754	0.651	0.704
Nonstandard employment	0.584	0.551	0.664
Differences	0.17	0.10	0.039
Explained	0.083	0.093	0.076
Unexplained	0.086	0.007	−0.036
Explained/difference	49.19%	93.36%	192.02%
Unexplained/difference	50.81%	6.64%	−92.02%

Notes: Authors’ calculation from the estimation results shown in Table 4. The complete Oaxaca-Blinder decomposition result is available upon request.

Finally, propensity score matching methods were used to examine the treatment effect of nonstandard employment as a robustness check. Table 7 presents the average treatment effect on the treated (ATT). Columns (1) to (3) report the results from the nearest-neighbour matching, local linear regression matching and kernel matching, respectively. Overall, the results were quite robust to different matching methods. In particular, the influence of nonstandard employment was significantly negative in 2006 but, as in the baseline results, it became insignificant in 2008 and 2012.

Table 7.

Propensity score matching estimates of treatment effect on nonstandard employment.

Year	(1)	(2)	(3)
Year	Nearest-neighbour matching	Local linear regression matching	Kernel matching
2006	−0.181**	−0.126***	−0.187***
	(0.071)	(0.048)	(0.063)
2008	−0.080	−0.068	−0.067
	(0.074)	(0.076)	(0.068)
2012	0.099	0.102	0.072
	(0.092)	(0.086)	(0.069)

Notes: A 0.06 fixed bandwidth and the Epanechnikov kernel are used for kernel and local linear regression matching. Bootstrapped standard errors for ATT are reported in parenthesis. They are based on 200 replications of the data. *, ** and *** indicate the significance level of 10%, 5% and 1%, respectively.

Conclusion and discussion

This study examines the relationship between nonstandard employment and job satisfaction in China from 2006 to 2012. The findings from this study broaden the understanding of nonstandard employment in China, a country currently undergoing a major transition, and contribute to the generalizability of the international literature on nonstandard employment. The empirical analyses point to two conclusions.

First, the empirical results show that, in 2006, nonstandard employment workers were more likely to report low job satisfaction than their standard employment counterparts. This result is consistent with the previous literature on Western settings, which indicates a negative association between nonstandard employment and job satisfaction (e.g. Aletraris, 2010; Zeytinoglu et al., 2013). To explain these findings scholars have hypothesized that nonstandard employment jobs generally offer less favourable conditions and job security than standard employment jobs and that the degree of willingness to accept these jobs is relatively limited. According to De Witte and Näswall (2003), a worker with an insecure job arrangement will experience lower subjective well-being because of higher job stress from more aggravating work conditions or deprivation due to a lack of identity with the employer. In China, evidence shows that nonstandard employment workers, especially before new labour-related laws were issued, were less well-treated, worked in poorer conditions and received lower (and often delayed) pay and benefits than those with standard employment (Cooke, 2011a). In addition, some nonstandard employment workers – and particularly rural migrant workers – were not fully eligible to enrol for social insurance programmes due to the lack of a local urban hukou or written labour contracts. Consequently, this might give nonstandard workers more economic constraints and negatively influence their job expectations, which, in turn, could lower their job satisfaction. According to previous studies, when workers voluntarily select nonstandard employment, such employment will not have negative attitudinal consequences (Aletraris, 2010). Yet, in China, most workers, and particularly rural migrant workers, perhaps went into nonstandard employment not voluntarily but because legislation was not enforced and the hukou system was restricted (Park and Cai, 2011). This highlights the disadvantage experienced by workers who undertook nonstandard employment during that period. The findings also show that those who worked as contributing family workers, freelancers, temporary workers, part-time workers and dispatch agency workers, and those who do not sign labour contracts with employers, are more likely to be dissatisfied with their jobs than standard employment workers. In other words, to better understand the characteristics and consequences for workers of nonstandard employment it is important to break this general category down into more specific categories.

Second, in 2008 and 2012, these differences in job satisfaction became very small and statistically insignificant, indicating that nonstandard workers’ job satisfaction improved over time. Perhaps labour law reforms, such as the Labour Contract Law and two related 2008 laws, improved working conditions and strengthened protections and security for nonstandard employment workers – an outcome that would have increased their job satisfaction. Given that economic outcomes and job satisfaction are positively related (Judge et al., 2010), this finding is consistent with empirical evidence that the average wages and benefits of nonstandard employment workers increased after passage of the 2008 labour law reforms (Chen and Hamori, 2013; Xue et al., 2014; Zuo, 2013). Similarly, the higher rate of social insurance coverage for nonstandard employment workers (Freeman and Li, 2013) probably helped to improve their attitudes. After passage of the Labour Contract Law and related law reforms, many nonstandard workers moved into standard forms of employment, and, consequently, the proportion of nonstandard employment declined over time, as shown in Figures 2 and 3. It is expected that a nonstandard worker who did not move quickly into standard employment will transition to permanent and standard employment with his or her employer in the near future. This occurrence might lead analysts to see nonstandard employment as a transitory stage and a motivation for nonstandard workers to achieve a standard employment position (De Cuyper et al., 2011).

Although this study’s findings significantly advance the empirical examination of nonstandard employment and workers’ attitudes and well-being in China, some limitations need to be acknowledged. First, to reduce the omitted variable bias and to conduct a propensity score-matching exercise, this study controls for several individual and job characteristics, but other unobserved characteristics could influence the relationship between nonstandard employment and job satisfaction. Along these lines, longitudinal data could lead to the development of more confident conclusions about causal relations, which are difficult to draw from cross-sectional designs. To provide a comprehensive picture of nonstandard employment, additional studies of the impact of new laws on such employment across industries and regions are needed. Furthermore, owing to a lack of data in the CGSS, this study does not completely take into account the effect of local governments and trade unions on nonstandard employment and workers’ attitudes. In recent years, the representation of nonstandard workers by Chinese unions has improved. Therefore, future studies should examine how workers’ representatives improve the outcomes of nonstandard workers.

Footnotes

Acknowledgements

The authors would like to thank Hao Zhang, Paul Wolfson, Janroj Yilmaz Keles and three anonymous referees for offering helpful comments and suggestions. We also thank all seminar participants at the 18th International Labour and Employment Relations Association (ILERA) World Congress and at the school of Labor and Human Resources at Renmin University of China. All remaining errors are ours.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kritkorn Nawakitphaitoon

Notes

Kritkorn Nawakitphaitoon is currently an assistant professor in labour relations in the School of Human Resource Development at the National Institute of Development Administration, Thailand. Formerly, he was a faculty member at the Renmin University of China’s School of Labour and Human Resources. His research interests include employee voice, high performance work system, nonstandard employment and occupational human capital.

Can Tang is currently an assistant professor in the School of Business at Shanghai University of International Business and Economics, China. Her research interests include child development and the structure of the labour force.

References

Aletraris

(2010) How satisfied are they and why? A study of job satisfaction, job rewards, gender and temporary agency workers in Australia. Human Relations 63: 1129–1155.

Bardasi

Francesconi

(2004) The impact of atypical employment on individual well-being: evidence from a panel of British workers. Social Science and Medicine 58: 1671–1688.

Bian

(2012) The Chinese General Social Survey (2003–8): sample design and evaluation. Chinese Sociological Review 45: 70–97.

Black

(2015) Matching as a regression estimator. IZA World of Labor 186: 1–10.

Booth

Francesconi

Frank

(2002) Temporary jobs: stepping stones or dead ends? The Economic Journal 112: 189–213.

Cai

(2011) Hukou system reform and unification of rural–urban social welfare. China & World Economy 19: 33–48.

Cai

Wang

(2004) Informal employment and labour market development. Economic Perspective 2: 24–28.

Chen

Hamori

(2013) Formal and informal employment and income differentials in urban China. Journal of International Development 25: 987–1004.

Cheng

(2014) The effects of employee involvement and participation on subjective wellbeing: evidence from urban China. Social Indicators Research 118: 457–483.

10.

Cooke

(2008) The changing dynamics of employment relations in China: an evaluation of the rising level of labour disputes. Journal of Industrial Relations 50: 111–138.

11.

Cooke

(2011a) Labour market regulations and informal employment in China: to what extent are workers protected? Journal of Chinese Human Resources Management 2: 100–116.

12.

Cooke

(2011b) The enactment of three new labour laws in China: unintended consequences and the emergence of ‘new’ actors in employment relations. In: Lee

McCann

(eds) Regulating for Decent Work: New Directions in Labour Market Regulation. Geneva: International Labour Organization; Basingstoke: Palgrave Macmillan, 180–205.

13.

De Cuyper

De Witte

Van Emmerik

(2011) Temporary employment: costs and benefits for (the careers of) employees and organizations. Career Development International 16: 104–113.

14.

De Witte

Näswall

(2003) Objective versus subjective job insecurity: consequences of temporary work for job satisfaction and organizational commitment in four European countries. Economic and Industrial Democracy 24: 149–188.

15.

Freeman

(2013) Has China’s new labour contract law worked? Vox CEPR’s Policy Portal. Available at: https://voxeu.org/article/has-china-s-new-labour-contract-law-worked (accessed 27 January 2018).

16.

Guest

Oakley

Clinton

, et al. (2006) Free or precarious? A comparison of the attitudes of workers in flexible and traditional employment contracts. Human Resource Management Review 16: 107–124.

17.

International Labour Organization (2003) Guidelines Concerning a Statistical Definition of Informal Employment. Geneva: International Labour Office. Available at: http://www.ilo.org/global/statistics-and-databases/standards-and-guidelines/guidelines-adopted-by-international-conferences-of-labour-statisticians/WCMS_087622/lang—en/index.htm (accessed 27 March 2018).

18.

Judge

Piccolo

Podsakoff

, et al. (2010) The relationship between pay and job satisfaction: a meta-analysis of the literature. Journal of Vocational Behavior 77: 157–167.

19.

Judge

Thoresen

Bono

, et al. (2001) The job satisfaction–job performance relationship: a qualitative and quantitative review. Psychological Bulletin 127: 376–407.

20.

Kalleberg

(2000) Nonstandard employment relations: part-time, temporary, and contract work. Annual Review of Sociology 26: 341–365.

21.

Kalleberg

Hewison

(2013) Precarious work and the challenge for Asia. American Behaviorial Scientist 53: 271–288.

22.

Oaxaca

(1973) Male-female wage differentials in urban labor markets. International Economic Review 14: 693–709.

23.

Park

Cai

(2011) The informalization of the Chinese labor market. In: Kuruville

Lee

Gallagher

(eds) From Iron Rice Bowl to Informalization. Ithaca, NY: Cornell University Press, 17–35.

24.

Remington

Cui

(2015) The impact of the 2008 Labor Contract Law on labor disputes in China. Journal of East Asian Studies 15: 271–299.

25.

Song

(2014) What should economists know about the current Chinese hukou system? China Economic Review 6: 200–212.

26.

Swider

(2015) Building China: precarious employment among migrant construction workers. Work, Employment and Society 29: 41–59.

27.

Wang

Cooke

Lin

(2016) Informal employment in China: recent development and human resource implications. Asia Pacific Journal of Human Resources 54: 292–311.

28.

Wang

(2013) An analysis of the development of informal employment in China. Macroeconomic Management 9: 63–64.

29.

Wilkin

(2013) I can’t get no job satisfaction: meta-analysis comparing permanent and contingent workers. Journal of Organizational Behavior 34: 47–64.

30.

Wooden

Warren

(2004) Non-standard employment and job satisfaction: evidence from the HILDA survey. The Journal of Industrial Relations 46: 275–297.

31.

Cai

(2006) Informal employment in urban China: size and characteristics. China Labour Economics 2: 67–84.

32.

Xue

Gao

Guo

(2014) Informal employment and its effect on the income distribution in urban China. China Economic Review 31: 84–93.

33.

Zeytinoglu

Yilmaz

Keser

, et al. (2013) Job satisfaction, flexible employment and job security among Turkish service sector workers. Economic and Industrial Democracy 34: 123–144.

34.

Zhou

(2012) The state of precarious work in China. American Behavioral Scientist 57: 354–372.

35.

Zuo

(2013) Formal and informal employment in China: probability of employment and determinants of monthly wages. Australian Economic Review 46: 405–423.

Nonstandard Employment and Job Satisfaction across Time in China: Evidence from the Chinese General Social Survey (2006–2012)

Abstract

Keywords

Introduction

Literature review

Definition of nonstandard employment in China

Nonstandard employment and labour-related legislation in China

Previous studies of the consequences of nonstandard employment

Empirical strategy and data set

Data and descriptive analysis

Empirical strategy

Empirical results

Baseline model

Robustness check

Conclusion and discussion

Footnotes

Acknowledgements

Funding

ORCID iD

Notes

References