Great changes have taken place in China's employment processes since market-oriented socio-economic reforms were introduced. Different from the state-controlled job assignment scheme in a planned economy,Footnote 1 emerging labour markets are witnessing increasing unemployment and intense competition in job search processes in urban China. Chinese job seekers are using diverse methods in their attempts to find work such as attending job fairs, responding to job advertisements, directly contacting employers, and obtaining help from relatives, friends and so on. According to available statistics, the use of market-oriented job search methods increased remarkably in the early years of the 21st century. For example, in 2006 there were 49.51 million registered job vacancies and 47.36 million registered job seekers, 53 per cent of whom successfully secured a job via market channels.Footnote 2 Under such circumstances, can we argue that after more than 30 years of economic reform, market-oriented methods have overtaken the utilization of hierarchical channels or social networks as the dominant method of finding a job in China today?
The existing literature provides a mixed response to this question. According to Yanjie Bian and his colleagues, the ongoing labour market reform has not yet weakened the effect of social networks in job searches; rather, social networks play an active role in transmitting information, building trust and creating binding obligations between job applicants and prospective employers.Footnote 3 Conversely, Douglas Guthrie argues that the importance of guanxi 关系 has waned during China's economic transition, and Amy Hanser further finds that the decline of social networking is coupled with the rise of market forces in employment processes.Footnote 4 My interviews with job seekers in Chinese cities indicate that certain types of job search methods seem to be particularly useful for winning offers in certain types of jobs. For instance, social networks remain important in securing state-sector jobs, highly desired jobs and “soft-skill” jobs.Footnote 5 A review of the existing studies into job search methods reveals a deficiency in that they have not systematically and quantitatively explored to what extent job search methods vary in diverse spaces in the emerging labour market or what underlying institutional logics exist.
This article attempts to address these issues and is organized as follows. I first situate my research in China's transitional society and “hybrid economy.”Footnote 6 Given that market forces penetrate economic sectors to varying degrees and that China has adopted gradualist reform policies,Footnote 7 I then propose an institutional niches approach to examine whether and how job search methods are associated with employment outcomes in different segments of the labour market. I continue by conducting quantitative analysis using the Chinese General Social Survey (CGSS) data collected in 2003 to verify hypotheses. Finally, I summarize my findings and discuss theoretical implications of the institutional niches approach for China's emerging labour markets.
Market Penetration in a Hybrid Economy
Three decades of market economy reform have transformed China's state socialist redistributive economy into a hybrid system characterized by the continuous growth of market institutions and the strong and persistent involvement of government.Footnote 8 Under this hybrid economy, certain segments reflecting the continued legacy of planning have been retained owing to the reform's partial, gradual and uneven nature; however, overall market forces have continued to make deeper inroads.Footnote 9 According to Ivan Szelenyi and Eric Kostello, there have been two types of market penetration (i.e. local markets in the redistributive economy, and a socialist mixed economy that featured the coexistence of the market and the redistributive system) in the Chinese context since the late 1970s.Footnote 10 Each type of market penetration, arguably, has boosted the development of main market institutions such as the commodity market, labour market, and capital market. Compared to the redistributive power of the state, the more the markets dominate the allocation of goods, products and resources, the deeper the market penetration. For instance, in the context of labour markets, the more the allocation and price of labour is determined by the rules of supply and demand rather than by the planned system, the more the market is free and developed. Market methods then play a more effective and important role in securing employment than other non-market methods such as state assignment or using personal connections.
However, market penetration is not a linear and straightforward process in China. As documented, the survival of the authoritarian hierarchy,Footnote 11 the persistence of political power,Footnote 12 and the resilience of guanxi and social networkingFootnote 13 have all interwoven with market forces and brought more complexity to the transition period. Max Boisot and John Child point out that China is not moving towards a market order as experienced by Western countries, but towards a form of economic organization that can be labelled “network capitalism.”Footnote 14 This indicates that the Chinese system of networked transactions is relatively uncodified and based on trust and longstanding personal connections. Xueguang Zhou and colleagues show that social networks are employed as a business strategy to manage uncertainty in economic transactions and to develop inter-firm contractual relationships in China's transitional economy.Footnote 15 David Wank finds that private businesses operate in networks of personal ties that centre on local government, and that through an institutional commodification process, public resources controlled by the state and the bureaucracy have become commodities that serve the interests of both parties in patron–client exchanges, i.e. officials and private entrepreneurs.Footnote 16
These studies, although varying in foci of analysis and detailed interpretations, all point to the fact that although the penetration of market forces is apparently visible, market forces have not yet become dominant in China's hybrid economy, and so further market building and institutionalization are necessary.Footnote 17 The complex nature of market penetration provides the institutional context for my research of job searches in China's labour market.
China's Labour Market: An Institutional Niches Approach
As a transitional and hybrid economy, China has shifted gradually from the planned allocation of labour in state-sector jobs to a more open labour market. Nevertheless, the labour market reform has been halting, uneven and difficult, with much additional reform still required.Footnote 18 Because the hybrid economy is closely entangled with the durability of the political system and the influence of social networks, it is reasonable to suspect that, as far as employment processes are concerned, market-oriented methods may have not worked effectively and evenly in different segments of the labour market. Rather, other mechanisms such as hierarchical methods and social networks maintain their advantages in certain locations and erect barriers to further market penetration, preventing the development of free labour markets.
To capture the institutional variations associated with the effectiveness of job search methods, I adopt an institutional niches approach which helps to specify which locations in labour markets favour particular job search methods owing to the underpinning institutional rules. This approach is derived from a theoretical construct of “niche.”Footnote 19 This construct has developed at the intersection of the field of human ecology and classic structural macrosociology and describes the environmental space confined by parameters on multiple dimensions in which a population of species can sustain itself.Footnote 20 From the early 1980s, scholars have applied the concept of niche to organizational sociology and spurred a research tradition that has greatly advanced our understanding of the relationship between organizations and their environments. A niche helps to explicate how an organization's position in an industry shapes its life chances as well as organizational behaviours within a particular market segment.Footnote 21 That is, whether an organization is located at the market centre, on the periphery or in between affects its ability to access resources and resist competition. Apart from organizational studies, the niche concept is employed in a metaphorical way to describe the unique nature of specific markets. For instance, Franklin Wilson defines ethnic niches as any labour market based on a social collectivity in which members of an ethnic group are concentrated at a higher level than members of other groups.Footnote 22 He disclosed a pattern of niche employment in America, showing that niches in service and blue-collar occupations were primarily occupied by indigenous minority and non-European groups; in contrast, niches in professional, managerial and technical occupations were dominated by European, Middle Eastern and selected Asian groups. In Australia, Val Colic-Peisker and Farida Tilbury report that refugee groups are concentrated in employment niches consisting of low-status and low-paid jobs that locals tend to avoid.Footnote 23
How can the institutional niches approach be applied to an examination of China's emerging labour markets? As we know, the reform of the employment structure in China marked a shift from a rigid state-assigned scheme to a market-oriented mechanism. Its basic tendency was to open up the state sector and encourage the development of a non-state sector that had been artificially depressed. This resulted in the diversification of employment structures and the coexistence of state, collective, private and joint businesses. In particular, following the aggressive restructuring and layoff (xiagang 下岗) programme that marked the end of the “iron rice bowl” (tiefanwan 铁饭碗) of guaranteed employment and benefits for workers in state-owned enterprises in the late 1990s, urban unemployment became a grave concern. Consequently, official encouragement was given to expanding small non-state enterprises and service-sector employment.Footnote 24 Under the segmented labour market that has emerged in cities, urban residents receive preferential treatment in some sectors but compete with migrant workers in others.Footnote 25 Owing to these circumstances, I suspect that the institutional niches approach may provide a suitable framework and contribute to revealing that labour market segmentation is more highly differentiated than indicated by the previous binary divisions (for example, the “core versus periphery”Footnote 26 or the “state versus private” sectorsFootnote 27 ).
Based on the relevant literature and my earlier qualitative study, I assume that two dimensions of analysis are important to revealing a fuller picture of how Chinese emerging labour markets have been shaped by concrete institutional niches.Footnote 28 First, the type of ownership of work organizations (state owned versus non-state owned) is a fundamental factor that draws the line between two utterly different labour markets.Footnote 29 In the state sector, the openness of organizations to the market varies considerably. Some work organizations have a higher degree of marketization. For example, as pointed out by John Burns, China's civil service system is far from homogeneous and has in fact two systems.Footnote 30 At the central or high levels, entry into the civil service is relatively performance-oriented and the system selects “the best and brightest” through competitive mechanisms and rigorous selection processes. To boost recruitment, the government has attempted both to increase its profile as an employer, especially within universities, and to rely more on the internet and information technology for publicizing posts. By contrast, at the local level, civil service positions continue to be filled through non-competitive and rather opaque processes, and potential employees are selected on the basis of different criteria, some of which may be irrelevant to the job, such as personal influence or guanxi, and can even involve relatively high levels of corruption. In addition, Chenggang Xu argues that the fundamental institution in China consists of a regionally decentralized authoritarian (RDA) system, which is characterized by highly centralized political and personnel controls at the national level, and a regionally decentralized administrative and economic system.Footnote 31 This combination of political centralization and economic decentralization deeply influences decision making and policy implementations in the state sector.
Similar to the scenario in government organizations or the civil service system, the degree of marketization of a state-owned enterprise's hiring practices depends on its organizational size or rank in the hierarchical system. For instance, local branches of the Bank of China in every city had to apply for recruitment quotas from the bank's headquarters in Beijing, as revealed by an interviewee in my previous research.Footnote 32 In contrast, in smaller state-owned enterprises both the hiring process and decision making are more flexible and less competitive or standardized. On the whole, however, it cannot be denied that the government and bureaucracy still have tight control over economic resources and personnel policies in the state sector. The non-state sector, on the other hand, is not constrained by this system and has applied market rules to doing business and hiring employees. Compared to Western countries, the type of ownership of work organizations remains imbued with the socialist characteristics of segmented labour markets that are unique to China.
The second dimension looks at how the status of work organizations within the market divides the Chinese labour markets into two distinct types, monopolistic or competitive. The monopolistic type of work organization occupies an advantageous position in markets and extracts excessive monopolistic profits by relying on national preferential policies or technical monopolies. These monopolies enable them to offer high remuneration, stable employment and attractive benefits. Generally, there are three categories of monopolistic organizations: monopolistic institutions (for example, state bureaus and government organizations); policy-monopoly enterprises (for example, banks, insurance companies and telecommunications companies); and technology-monopoly enterprises (for example, water/gas supply companies and electric power companies). This is consistent with Loren Brandt and Thomas Rawski's observation that the important sectors in China's economy are still dominated by state-owned enterprises.Footnote 33 By comparison, market-driven work organizations are at a relative disadvantage and are seen as less desirable in the labour market.
In short, I propose that work organizations in the emerging labour markets in China are differentiated by these two dimensions (i.e. ownership type and status in the market) into four institutional niches: state monopolistic, state competitive, non-state monopolistic, and non-state competitive (see Figure 1). These niches can be understood as four theoretical ideal types. They illustrate that market forces penetrate into the power structure of employment decision making to varying extents. In state monopolistic work organizations (Niche I), market penetration tends to be weaker than in state competitive work organizations (Niche II) and non-state work organizations (Niches III and IV). As shown in Part A of Figure 1, from Niche I to Niche IV the strength of market penetration is assumed to be increasing because of China's gradualist reform and its uneven nature.Footnote 34 In the niches where market forces are comparatively weaker, non-market methods for job searches that include hierarchical arrangements and social networks would have advantages over market methods. This institutional perspective demonstrates finer and essential institutional differences between work organizations, and allows us to go beyond the dichotomous classification of employment sectors implied in the classical theory of labour market segmentation to explore how the reforming institutional context in China influences concrete conditions for job searches.Footnote 35 The series of hypotheses proposed below will guide the empirical study to test the explanatory power of the theoretical framework of “institutional niches.”
Figure 1: Predicted Effects of Job Search Methods in Institutional Niches
Job Search Methods in Institutional Niches: Hypotheses
Taking into account the different institutional conditions governing employment processes in China, the methods used by job seekers can be classified into three types: hierarchical, market, and network.Footnote 36 The hierarchical method refers to job assignments and organized transfers by a state authority or organization. This was the most frequently used method in the pre-reform period because it was legitimized by the state redistributive regime. The market method includes job search activities in which people go through the formal labour market channels to seek out employment. The assessment of eligibility or value and the hiring decision made by the prospective employer are based on merit. Using the market method may entail voluntary geographic mobility, attending job fairs where employers gather to recruit employees, using formal recruitment services, applying directly to employers, responding to media advertisements, undertaking online job searches, and so on. Market methods were mostly unlawful in the pre-reform period but became legal after the introduction of market reforms in the 1980s. Finally, the network method covers any searching activity in which the personal assistance of others – that is, the mobilizing of social ties and connections – is drawn on to secure employment, and the assessment of the job seeker's eligibility or value and the hiring decision can be influenced by power, favour or other particularistic criteria.
Given this, what is the relative effect of these job search methods on gaining entry into different institutional niches in China's emerging labour markets? According to the theory of market transition, human capital is becoming increasingly important while the value of political capital is declining in transitional China.Footnote 37 In terms of this theory, hierarchical methods will be overshadowed and replaced by market forces. In a similar vein, Guthrie finds that in the area of economic decision making, hard-budget constraints and market competition combine to increase rationalization and cause social networks to decline in significance.Footnote 38 Alternatively, the power persistence theme has surfaced in the market transition debate. This theme maintains that the Chinese party-state will retain its political power and social influence and that it is not possible for these forces to be removed instantly by market reforms.Footnote 39 Which one of these competing accounts, then, has stronger explanatory power in understanding Chinese labour market processes, and which is more suitable for investigating employment in certain locations of the labour market? Based on the above view of institutional niches and my qualitative study of job seekers in China, I find it reasonable to argue that in the emerging labour markets, the three job search methods (hierarchical, social networks and market methods) coexist and play different roles across institutional niches.Footnote 40 Thus, both the market transition theory and the power persistence theme may fit by being grounded in different institutional niches.
In Part B of Figure 1, I present my assumptions by considering the institutional environment and recruitment policies in each niche. As illustrated, four types of institutional niches are differentiated: Niche I and Niche II are the state monopolistic niche and the state competitive niche, respectively. While both belong to the state sector, the former contains monopolistic work organizations and the latter comprises competitive work organizations. Niche III and Niche IV apply to the non-state monopolistic niche and the non-state competitive niche, respectively, and while both belong to the non-state sector, they differ in status in terms of market monopoly. In each type of niche, the three kinds of job search methods are expected to exert different effects. For the purpose of formulating hypotheses, I choose Niche IV as the reference category and discuss the relative effects of hierarchical methods and social networks when compared to market methods.
In both niches of the state sector (I and II), hierarchical methods and social networks are the entry mechanisms that are particularly favoured. First, most state organizations maintain strong institutional links with political authorities,Footnote 41 and still use the hierarchical allocation of jobs as an essential means for personnel arrangements, especially at high levels.Footnote 42 Despite economic reforms, the party-state has never loosened its control over personnel.Footnote 43 For instance, Party cadres and government officials are appointed by the government to fill important positions in public institutions (for example, universities, hospitals and unions) and state-owned enterprises. Furthermore, the internal operations of these organizations are subject to routine inspections by the supervising agencies.Footnote 44 In some cases, the local government must provide non-competitive employment opportunities for demobilized soldiers.Footnote 45 Meanwhile, the principal–agent separation problem in the state sector creates an environment in which social networks play a role.Footnote 46 For managers (agents) who control resources and make employment decisions, after taking their own interests and the internal harmony of enterprises into account, they are inclined to favour job applicants with whom they have a connection via a social network.
In addition, when the monopolistic feature of organizations is considered in the state sector (i.e. in Niche I), the effects of hierarchical methods and social networks will be more significant. The reasons are twofold. First, the state monopolistic organizations are of great importance in China's hierarchical administration structure and come under the stricter control of the state with the result that hierarchical job allocation methods continue to have a bearing on maintaining the “internal labour market” for jobs at high levels. Second, state monopolistic organizations provide better paid jobs that are highly desired in the labour markets and, thus, social networking tends to be heavily mobilized to enhance the possibility of securing these jobs. For example, in 2003, the average wage in the farming sector was at the bottom of the spectrum at 6,912 yuan, and in the finance and insurance industry, a type of policy-monopoly industry, it was at the top end, with 23,075 yuan. Therefore, average earnings within the top industry were 3.3 times those in the bottom industry. This huge income differential remains the same after a decade. According to the most recent national statistics, the average wage in the finance industry (108,273 yuan) was 3.8 times the average wage in the farming industry (28,356 yuan).Footnote 47 Apart from the differences in wages, mandated subsidies and other benefits in urban enterprises are increasingly dependent on the profitability and financial resources of employers.Footnote 48 Hence, the excessive inequality that exists in wage levels and various social benefits between monopolistic and competitive industries provides a strong incentive to securing employment within state monopolistic organizations. Hanser observes that market methods such as attending job fairs or seeking jobs on the internet are not very useful for gaining entry to the state sector, although they do seem to have gained weight in some new job-searching avenues in labour markets.Footnote 49 Bearing the above in mind, I predict that in Niche I, hierarchical methods and social networks are both significant in obtaining a job with reference to market methods and would outperform market methods in the strongest way (indicated by “+++”). Hypotheses are formulated as follows:
Hypothesis 1a:
In the state monopolistic niche compared to the non-state competitive niche, hierarchical methods are likely to be much more useful than market methods to secure a job.
Hypothesis 1b:
In the state monopolistic niche compared to the non-state competitive niche, social networks are likely to be much more useful than market methods to secure a job.
In Niche II, although competitive work organizations in the state sector may have a deeper degree of marketization than do state monopolistic work organizations, it is still premature to anticipate a significant use of market methods in job search processes because the ongoing labour reform is characterized by gradualism and experimentation,Footnote 50 with political power and social influences still persisting strongly.Footnote 51 Provided that the market penetration is not strong enough, the relative advantage of using hierarchical methods and social networks would remain even though they may be weaker than the counterparts in Niche I (indicated by “++”). This suggests two hypotheses:
Hypothesis 2a:
In the state competitive niche compared to the non-state competitive niche, hierarchical methods are likely to remain useful to secure a job with reference to market methods.
Hypothesis 2b:
In the state competitive niche compared to the non-state competitive niche, social networks are likely to remain useful to secure a job with reference to market methods.
In contrast to niches in the state sector, Niche III and Niche IV in the non-state sector provide different implications for the job search methods. It is widely observed that work organizations in the non-state sector niches have adopted market-oriented methods to recruit staff by posting job advertisements publicly, participating in job fairs to meet applicants, holding screening interviews and examinations, and so on. As a result, job seekers can rely on market methods to secure a job in these work organizations. Compared to the rapid growth of market methods, hierarchical methods appear to have been minimized or phased out in both niches of the non-state sector. As for social networks, their role may vary according to the status of the work organization in the competitive market. In Niche IV, as competitive jobs in the non-state sector are open to applicants who will be assessed by market standards such as formal qualifications and task-specific skills, there is little opportunity to utilize social networks. In Niche III, however, which hosts a range of monopolistic work organizations in the position to offer well-paid jobs and associated benefits, clientelist ties are prone to be widely cultivated and social networks may still be used to help job applicants “squeeze in” even though the effect can be minimal (indicated by “+”).Footnote 52 Therefore, I propose:
Hypothesis 3a:
In the non-state monopolistic niche compared to the non-state competitive niche, hierarchical methods are likely to be insignificant with reference to market methods to secure a job.
Hypothesis 3b:
In the non-state monopolistic niche compared to the non-state competitive niche, social networks are likely to remain useful to secure a job but the effect is minimal with reference to market methods.
Methods and Variables
The 2003 Chinese General Social Survey (CGSS) data are employed for statistical analysis.Footnote 53 A multi-stage stratified sampling scheme was implemented in this survey to collect a nationally representative sample of 5,894 urban households across 30 provinces, municipalities and autonomous regions. The age criterion for respondents in the sample was between 18 and 69. Respondents were selected randomly from all eligible household members and interviewed using a structured questionnaire. In this paper, the working sample for analysis is generated by two principles: first, respondents had to have a history of employment, and second, respondents had to have gained their current or last job since labour market reform. In this way, respondents who had never been employed or who had obtained their current job before 1980 were excluded. In total, 4,146 were eligible for statistical analyses. The purpose of these principles is to ensure that the institutional niche of a respondent's current job could be identified and also to exclude cases of workers who had secured jobs in the pre-reform period, which are irrelevant to the current study on reforming China. Multinomial logistic regression is used to test the hypotheses concerning the association of job search methods with the entry to institutional niches.
Dependent variable
“Institutional niche” is the dependent variable of the multinomial logistic regression. I construct this variable by considering the ownership of work organizations and the category of industries.Footnote 54 First, the ownership of work organizations is classified into two types: the state sector, which refers to the items of “Party, government, or government agency or office,” “state-owned enterprise,” “state-owned institution” and “collective-owned institution or enterprise” in the questionnaire; and the non-state sector, which includes “individual-operated enterprise,” “private institution or enterprise,” “foreign investment enterprise” and “other types.” Second, based on statistics released by the National Bureau of Statistics in the same year of the survey, “monopolistic industries” covers the “production and supply of electricity, gas and water,” “transport, storage, post and telecommunications,” “banking and insurance,” “real estate trade” and “state and Party organizations” in the questionnaire, while “competitive industries” refers to the remaining industries.Footnote 55 In this way, institutional niches are differentiated into four types: state monopolistic niche (“1”), state competitive niche (“2”), non-state monopolistic niche (“3”) and non-state competitive niche (“4”). The fourth type is taken as the reference category in the multinomial logistic regressions.
As shown in Table 1, 15.8 per cent of 4,146 respondents in the post-reform era worked in the state monopolistic niche, 58.3 per cent in the state competitive niche, 2.1 per cent in the non-state monopolistic niche, and 23.8 per cent in the non-state competitive niche. From the early reform period (1980–1992) to the further reform period (2000–2003), the percentages of the state monopolistic niche and state competitive niche decreased in contrast to an increasing pattern in both the non-state monopolistic niche and the non-state competitive niche. This tendency is consistent with the fact of economic reform in the state sector.
Table 1: Descriptive Statistics of Variables in the Analyses
Source: Author's estimates from Chinese General Social Survey 2003.
Independent variables
Job search methods are independent variables. Respondents were asked to check what methods they used in job search processes for their current or last work organizations. Among 11 items, “replaced parents or relatives,” “assigned by the government or reallocated by the organization” and “intra-organization recruitment” are regarded as using only hierarchical methods. Items such as “directly applied by myself,” “employment services,” “human resources exchange centre,” “self-employed,” “invited by employer,” “recruitment and examinations” and “other” are treated as using only market methods. If a respondent only chose “ask for recommendation from others,” he/she was regarded as using only social networks. If a respondent chose this item together with any item of hierarchical methods or market methods, he/she was coded into the fourth category, i.e. networks jointly used with other methods. However, it should be noted that a respondent who simultaneously selected hierarchical methods and market methods without using social networks was treated as a missing case because, theoretically speaking, hierarchical methods and market methods are two contradictory institutional arrangements. Thus, four categories of job search methods are constructed: only hierarchical methods; only market methods; only social networks; and networks jointly used.Footnote 56 They are mutually exclusive and employed as dummy variables in the analysis. Among these, the category of only market methods is chosen as the reference group. As can be seen from Table 1, only hierarchical methods was the major strategy reported by 52 per cent of respondents, only market methods accounted for 28.8 per cent, only social networks 10.5 per cent, and networks jointly used accounted for 8.7 per cent. Across the three post-reform periods, data based on respondent recall reveal that hierarchical job search methods dropped from 61.5 per cent to 52.9 per cent and then to 42.2 per cent. In contrast, market methods and social networks (including “only used” and “jointly used”) continued to rise.
Control variables
Control variables include individual attributes variables such as gender, age, Party membership and educational level, as well as labour market context variables such as region of residence and size of city. Research has shown that Party membership and education are good indicators of political capital and human capital in China.Footnote 57 In the emerging labour markets, it is reasonable to assume that since job seekers equipped with different packages of capitals would utilize different search methods, these two variables should be controlled for in models. In addition, because of the varying degrees of marketization across regions and cities in China, the region of residence and the size of city are two important factors to be considered as they may play a role in structuring labour market opportunities. Finally, I use a variable of “time of job entry” to divide my working sample into three subgroups, which helps to detect interaction effects between independent variables and the periods of reform. This variable is coded into three categories reflecting major stages in the reform era: (1) 1980–1992, the early reform period featured the emergence of labour markets and the decline of state job assignments; (2) 1993–1999, the fast reform period began following Deng Xiaoping's 邓小平 Southern Tour and the 14th National Congress of the Chinese Communist Party in 1992, and was characterized by tremendous growth in the private sectors and the increasing prevalence of labour markets; and (3) 2000–2003, in the further reform period, markets had become an increasingly dominant indicator in employment processes. Involuntary job mobility, caused by state-owned enterprises' policies to lay off workers, and voluntary job mobility, as a choice for higher education degree holders, were both apparent. The details about the control variables are reported in Table 1.
Findings
Figure 2 reports a significant association between job search methods and institutional niches (p < .001), which serves as a preliminary test on the predicted effect of job search methods. It is shown that hierarchical methods and market methods respectively play a major role in obtaining jobs in the state sector and the non-state sector, regardless of the differences that exist between monopolistic and competitive work organizations. In detail, 66.5 per cent of jobs from the state monopolistic niche and 64.7 per cent from the state competitive niche were obtained through hierarchical methods, whereas 55.2 per cent of jobs from the non-state monopolistic niche and 59.8 per cent from the non-state competitive niche were secured by market methods. In the meantime, hierarchical methods were used markedly less in the non-state sector (with 8.0 per cent in Niche III and 14.3 per cent in Niche IV), while market methods reached 17.1 per cent in Niche I and 18.5 percent in Niche II in the state sector. Thus, the type of ownership of work organizations is a significant factor involved in differentiating between hierarchical and market job search methods. As for social networks (with “only used” and “jointly used” summed), overall there were noticeable effects in obtaining jobs in four types of institutional niches. Comparatively, using social networks was more common in the non-state sector (36.8 per cent in Niche III and 25.9 per cent in Niche IV) than in the state sector (16.4 per cent in Niche I and 16.8 per cent in Niche II). This can possibly be explained by the sharp decline in hierarchical methods coupled with the insufficient growth of market methods in the non-state sector.Footnote 58 This cross-tabulation result shows significant differences in job search methods in institutional niches, but does not reveal the relative efficacy of these methods. For this purpose, a multinomial regression that controls other variables when examining the relationship between job search methods and the entry to institutional niches is needed.
Figure 2: Association between Job Search Methods and Institutional Niches
Table 2 presents the multinomial regressions concerned with the relative effect of job search methods on entering different institutional niches in three periods, and thus tests the hypotheses. Findings reported in the first column of each period provide the tests for Hypotheses 1a and 1b. Hierarchical methods and social networks had significant effects in comparison to market methods in predicting entry into the state monopolistic niche with reference to the non-state competitive niche. In particular, hierarchical methods were found to have a very large effect (odds ratios equal to 7.152, 18.219, and 9.924 in the three periods, respectively) on obtaining jobs in the state monopolistic niche. Take the biggest odds ratio, 18.219, as an example. It means that the odds of a job seeker utilizing hierarchical methods to gain a job in the state monopolistic niche in 1993–1999 would be about 17.2 times greater than for a job seeker who utilized market methods. The odds ratios related to social networks suggest that the use of “only networks” or “networks jointly used” would increase the odds of winning job offers from the state monopolistic niche when compared to market methods. For instance, the odds ratios for the use of “only networks” in the two periods 1993–1999 and 2000–2003 are 6.008 and 2.383. And, the odds ratios for “networks jointly used” are 2.177 and 2.447 in the two periods of 1980–1992 and 1993–1999. Hypotheses 1a and 1b are supported by these results.
Table 2: Multinomial Odds Ratios in Predicting Effects of Job Search Methods on Entering Institutional Niches
Source: Author's estimates from Chinese General Social Survey 2003.
Notes: a The reference category is Niche IV (non-state competitive niche); Niche I = state monopolistic niche, Niche II = state competitive niche, and Niche III = non-state monopolistic niche. b The reference category is only market methods. c Variables of individual attributes and labour market context are included in the models but not presented owing to space constraints. Details are available upon request. !p < .1, *p < .05, **p < .01, ***p < .001.
Similarly, findings reported in the second column of each period provide the tests for Hypotheses 2a and 2b. It is shown that hierarchical methods and social networks were still more useful than market methods for obtaining jobs from the state competitive niche. This pattern has remained consistently significant since the reform started in 1980. Even in 2000–2003, the likelihood of hierarchical methods helping a job seeker to gain entry to the state competitive niche was still about 6.7 times (odds ratio = 7.701) greater than when market methods were utilized. As for social network variables, whether they are “only networks” or “networks jointly used,” when compared to market methods they had an increased likelihood of securing jobs from the state competitive niche in all three periods. Given this, Hypotheses 2a and 2b gain significant statistical support.
Finally, as disclosed in the third column of each period, hierarchical methods and social networks do not assume a more significant effect on entering the non-state monopolistic niche than if market methods were used. Therefore, Hypothesis 3a is supported but Hypothesis 3b cannot be confirmed yet because social networks do not show a minimal, significant advantage over market methods to securing a job in the non-state monopolistic niche.
To sum up, the pattern observed from the above statistical results indicates that institutional niches in China's emerging labour markets favour different job search methods. The work organization's ownership and its status in the market are two major factors that differentiate these niches in which hierarchical, social network and market methods of job search have their own roles. Compared to market methods, hierarchical methods and social networks remain powerful in the state sector, including in both monopolistic and competitive work organizations. By and large, these findings suggest that hierarchical influences and network effects persist in China's labour market; market forces are gradually penetrating employment processes but as yet have not impacted the hard core of the state sector. This implies that the degree of permeability of employment opportunity structures differs according to the institutional niche within the emerging labour markets.
Discussion
The empirical study outlined in this paper supports the analytical framework of institutional niches and discloses a pattern of coexisting job search methods in reforming China. Two key factors, the work organization's type of ownership and its status in the market, are found to segment emerging labour markets into diverse institutional niches. After 30 years of market reforms, the coexistence of job search methods does not seem to be a transitional phenomenon, as non-market job search methods are still powerful in various institutional niches. Theoretically, the institutional niches framework, which derives from an overall argument of market penetration, contributes to the market transition debate by specifying the concrete contexts within which competing arguments can be grounded. That is, the market transition theory gains increasing explanatory power in niches of the non-state sector,Footnote 59 while the power persistence theme retains its strong position in niches within the state sector.Footnote 60 This suggests that we should not expect a sweeping answer to the market transition debate; rather, institutional analysis of concrete contexts counts. In the emerging Chinese labour markets, the expansion of markets is not a sufficient condition for the decline of redistributive power and social networks,Footnote 61 and it proves to be linear and simplistic to claim that surging market forces would replace hierarchical influence and social networks instantly and universally. As revealed in this study, the toughest barrier to market penetration lies with the boundary of state ownership. Within the state sector and regardless of the monopoly of work organizations, hierarchical methods are still the dominant entry mechanism, with the role of market methods confined to the non-state sector.
In addition, the significant role of social networks in the state sector is worthy of more discussion, and I propose two accounts, both of which are consistent with the market penetration argument. First, between the state sector and the non-state sector, the influence of market forces on running and managing a business is still sharply different; the most substantial difference in employment processes can be ascribed to the degree of transparency of recruitment procedures, which is decisive in terms of the effect of social networks on securing jobs. In a state sector undergoing partial reform and which has institutional insufficiencies and employment processes imbued with nepotism and favouritism,Footnote 62 recruitment procedures are much more complicated than those in the non-state sector, which has implemented both marketized and standardized staffing criteria. Moreover, the separation of principal and agents in the state sector has led to a tendency for agents to engage in rent-seeking activities by exchanging their power, influence or favour with contacts during the recruitment processes for their own immediate economic profit or long-term social benefit.Footnote 63 These factors provide soil for cultivating particularism. The consequence of this tendency is that within niches circumscribed by the state ownership, it is possible that recruitment activities are conducted in a non-transparent way to suit the personal interests of a particular social network.
Second, exploiting social networks is a rational choice for job seekers when they compete for a profitable and attractive job that is offered in a monopolistic niche. As mentioned earlier, there is a deep chasm between the monopolistic and the competitive industries in terms of pay and other benefits. Owing to the extremely high profits generated within the monopolistic industries, job applicants will use any means available to secure a job in that sector.Footnote 64 Social networks excel in collecting information about desirable jobs, function as low-cost, flexible and reliable channels in job search processes, and can provide access to spheres of influence through contacts.Footnote 65 As such, networks are likely to outperform market methods that adhere to formal and standardized procedures. Thus, competing applicants for monopolistic jobs tend to utilize social networks as much as possible, reflecting a rational choice when faced with keen competition in such an uncertain, institutionalized context. In China, the practice of using social networks has long been argued to be instrumentally rational.Footnote 66 Provided that market penetration does not reduce the huge gap between the profits to be made in the monopolistic and competitive industries, the profit-driven rational choice of social networking will remain.
In sum, the institutional niches framework is based on institutional characteristics in the emerging labour markets in China, and illustrates the importance of conducting institutional analysis. Because the ongoing labour reform in China is marked by gradualism and experimentation, it should be kept in mind that the coexistence of job search methods is a product of institutional settings that are in transition and subject to the scope of market penetration.Footnote 67 To grasp their dynamic relative efficacy and conditional effects, it is vital to adopt an institutional perspective. If the institutional environment does not change drastically, hierarchical methods and social networks will persist even though some variations might occur as the labour markets develop. If the marketization process of labour markets goes well, it is reasonable to expect that the weight of market methods in employment processes will accrue significantly beyond the current niches. Therefore, in terms of future research, it will be rewarding to discern how the ongoing socio-economic reform brings about changes in institutional niches in labour markets by conducting well-designed longitudinal studies.
Finally, this study has demonstrated that using the conceptual tool of “niches” can shed new light on the various spheres of China's emerging labour market and that further investigations based on new data would be welcomed and promising. This paper contributes quantitative evidence and findings to advancing research in this field but has some limitations owing to the constraints of the empirical data. A key issue is that large-scale, general social surveys (for example, CGSS) are often expensive, time-consuming, and limited by space in the questionnaires. This makes it impractical to follow trends in real time and probe into some aspects of society in sufficient detail. Even though the CGSS2003 dataset presents valuable and detailed information on individuals' social networking activities and job search processes in China, it cannot capture the most recent changes or latest phenomena in the labour market and the broader, fast-changing social context. For instance, over the past decade Chinese society has witnessed an upsurge in graduate unemployment,Footnote 68 the emergence of the “ant tribe” (yizu 蚁族, referring to low-income university graduates who live in poverty in the major cities),Footnote 69 and mounting challenges for the hai gui 海归 (returning high-skilled graduates who have studied overseas) in securing employment.Footnote 70 Each of these groups (i.e. unemployed or disadvantaged university graduates and highly skilled returnees), arguably, can be construed as a type of niche in China's labour market, yet these specific niches are very hard to identify in large-scale social surveys because of their relative small share of China's general population. Consequently, future research would benefit from combining the strengths of various empirical approaches, such as in-depth interviews, ethnographic methods and mixed methods, to produce more relevant and high quality data in order to enhance the understanding of the diversity of niches in China's labour market and how job seekers secure employment in these particular market segments accordingly.
Acknowledgement
The research for this paper has been supported by the Australian Research Council (project number DP130100690).
Biographical note
Xianbi Huang is senior lecturer in sociology at La Trobe University, Australia. Her main research interests include social networks, social stratification, employment processes, social well-being, and contemporary Chinese society. She has published articles in journals such as Social Networks, Sociology, Work, Employment and Society, American Behavioral Scientist, Research in the Sociology of Work, Chinese Journal of Sociology, International Journal of Japanese Sociology, and SAGE Research Methods Cases.
