Towards a hybrid model? A systematic review of human resource management research on Chinese state-owned enterprises (1993–2017)

Abstract

Over the past two decades, there has been a proliferation of research on human resource management (HRM) in Chinese state-owned enterprises (SOEs) as a result of the wide-reaching state-sector reform of the late 1980s. This article aims to provide a systematic review of literature on this topic and capture the nature of HRM in Chinese SOEs, both in research and practice. The article draws on 178 studies from 43 English academic journals over a period of 25 years (1993–2017). In analysing this literature and by taking stock of theoretical frameworks, research methods, themes and analysis of academic articles in this area, we have gained a number of insights. The study has found that the research methods used have shifted from qualitative and interpretive methodology toward quantitative and sophisticated modelling. A further insight is that there has been a relatively heavy reliance on institutional theory in the earlier studies reviewed, and since then a switch towards organizational behaviour perspectives. The level of analysis has moved from macro to micro level and thematic foci have become more diverse and complex. We highlight a number of avenues, theoretical and empirical, for future studies in this field.

Keywords:

• China
• state-owned enterprise
• institutional theories
• cultural theories
• employee behavior
• strategic HRM

Introduction

Since the mid-1980s, significant changes have taken place in the Chinese state-owned enterprises (SOEs). Employment in the state sector has shrunk. In 1978, it made up 78% of the total urban employment, whereas by the end of 2016, it was 15% of the same. This change has largely resulted as a consequence of the country’s opening up of its economy since the late 1970s (see Table 1). Research on human resource management (HRM) in SOEs has proliferated, capturing and embodying some of these transformational processes and outcomes in Chinese SOEs. While HRM researchers have provided several reviews that have sketched the emergence and development of HRM research and practice in the Chinese context (e.g. Cooke, 2009; Zhao & Du, 2012; Zheng & Lamond, 2009; Zhu, Thomason, & De Cieri, 2008), no review has been conducted that provides a systematic review of HRM in SOEs. This is an important research gap, given that the state sector still provides jobs to over 414 million workers as of 2016 (see Table 1), and that the business environment of SOEs differs markedly from that of the private sector (e.g. Huang, 2008). SOEs continue to play a major role in the Chinese economy, as they dominate many vital industries, such as energy, financial services and telecommunications (Fu & Kamenou, 2011). More broadly, researchers have called for more attention to ‘how and to what extent HR practices in China are changing in the face of significant changes in economic and political policies’ (Liang, Marler, & Cui, 2012, p. 55).

Table 1. Employment in state-owned enterprises in China (1978–2016) (figure in 1000 person).

Year	Total of urban employment	Total employment of state-owned units and % of total of urban employment
1978	95,140	74,510 (78%)
1985	128,080	89,900 (70%)
1990	170,410	103,460 (61%)
1995	190,400	112,610 (59%)
2000	231,510	81,020 (35%)
2003	262,300	68,760 (26%)
2006	296,300	64,300 (22%)
2009	333,220	64,200 (19%)
2013	382,400	63,650 (17%)
2016	414,280	61,700 (15%)

Source: compiled from the China Statistical Yearbook 2017 (National Statistical Bureau of China, 2017).

Studies of HRM in Chinese SOEs extend the scope of research subjects. Extant HRM studies often classify the subject into types (e.g. public sector and government organizations) (Guo, Brown, Ashcraft, Yoshioka, & Dong, 2011). Little attention has been paid to hybrid organizations that mix for-profit and non-profit characteristics. SOEs have been regarded as hybrid organizations in that they pursue not only financial organizational goals, but also public and non-profit goals (Bruton, Peng, Ahlstrom, Stan, & Xu, 2015). Hence, an in-depth review of SOEs will extend the scope of HRM research.

Finally, examining HRM in the Chinese SOE context may contribute to HRM theory in general. Most HRM theories dichotomize HRM into either commitment-focused (employee well-being and service quality) or control-focused models (cost reduction and exploitation of employees) (Nishii, Lepak, & Schneider, 2008). However, hybrid models that combine both commitment- and control-focused practices remain less well explored. The contemporary HRM system in SOEs is characterized as a hybrid model: a fusion of some market-driven elements of western HR practices with the vestiges of the socialist-oriented ‘iron rice bowl’ system (Warner, 2014). We therefore believe that exploring the HR practices in SOEs can balance these two competing HRM models (market-driven HR practices and socialist-oriented personnel management system) and have the potential to build a novel HRM theoretical framework.

As part of state institutions and a key stakeholder in the economy, SOEs are more likely to be affected by changes in economic and political policies than their private counterparts, as is evident in SOE reforms of the last three decades. Since 1993, the Chinese government proposed a set of policies to transform SOEs in a gradual fashion. For instance, the ‘Modern Enterprise System’ was launched in 1994; The State-owned Assets Supervision and Administration Commission was established in 2003 (Jing & McDermott, 2013); ‘Guiding Opinions of the Central Committee of the Communist Party of China and the State Council on Deepening SOE Reform’ were proposed in 2015 (Leutert, 2016). All of these institutional arrangements were designed to boost the performance of SOEs and motivate them to be more competitive in the global market.

Over the last two decades, researchers have revealed a set of marketization reforms and their penetrating effects on Chinese SOEs, including strategic HR practices, employment relations and organizational productivity (e.g. Child, 1995; Cooke, 2002; Ding, Goodall, & Warner, 2000). More recent studies have also shown signs of a more strategic approach to HRM in SOEs (Ngo, Lau, & Foley, 2008; Zhu, Cooper, De Cieri, & Dowling, 2005). Questions that arise are: what aspects of HRM in SOEs have been researched? What journey have SOEs travelled in their ‘modernization’ as demanded by the state? Has the current research adequately documented this ‘journey’? In the process of capturing the evolution of HRM in SOEs, how has the trajectory of research, in terms of theoretical focus, methodological orientation and level of analysis, for example, evolved? In other words, what trajectories have HRM researchers travelled in their study of the transition of SOEs in China from a state-planned economy to a market-oriented economy environment?

To shed light on the above issues, this study provides a systematic review of research on HRM in Chinese SOEs published in 43 English journals over a 25-year-period (1993–2017). The key questions examined were:

1. What are the key characteristics of HRM systems in Chinese SOEs? How have they evolved in the past three decades?

2. What have been the theoretical orientations of HRM research in Chinese SOEs?

3. What research methods have been used?

4. What thematic foci have been examined? How have the thematic foci changed over the last three decades?

5. What are the general trends, main research theories and levels of analysis?

6. What future research opportunities exist to address some of the gaps identified in this review?

The remainder of this article is structured as follows. First, we outline how the sample is collected and the methods used to analyze the sample. Second, we summarize the key features of HRM in Chinese SOEs and how they have evolved in the past three decades. Third, we identify types of methods in existing studies that have been adopted to examine HR practices in Chinese SOEs. Fourth, we discuss the theoretical underpinnings of this body of HRM research. Fifth, we uncover and provide comment on a variety of thematic foci in existing studies. Sixth, we discuss the implications of studying HRM in Chinese SOEs for general HRM theory building, and suggest directions for future research. Finally, we point out the limitations of our study.

Methods of data collection and analysis

Data collection for the review

For this review study, we focus on academic English journal articles. First, we searched the university library databases including EBSCO, Elsevier, Emeralds, JSTOR Arts and Sciences, Springer, ProQuest, Taylor and Francis, Thomson Reuters’ Web of Knowledge, Wiley Online Library. The key search words used to find articles included: ‘China, HRM’; ‘China, HR practices (including staffing, compensation, training)’; ‘China, employment relations’; ‘China, personnel management’, among other similar terms. The search period was from 1980 until December 2017. A total of 2660 articles were found. We used Mendeley to select articles that included the keyword ‘state-owned’ and identified 148 that focused on SOEs.¹ Second, we then searched the same databases using the key words: ‘China, state-owned enterprise’; and ‘China, state-owned company’. We manually screened the title, abstract and keywords and downloaded those articles that had an explicit focus on HRM, personnel management and employment relations. A total of 109 articles were selected. Third, we used the citation pearl-growing method to search for more articles from the reference lists of the articles which we selected through the first two steps. As a result, a total of 178 articles were included in our review.²

Data analysis

Given that the purpose of this review was to take stock of the theoretical and methodological trends of research on Chinese SOEs’ HRM practices, we adopted a qualitative coding method to analyse the selected articles. Each selected article was entered into the data coding template to record the author, journal, year of publication, empirical or review study, research method (mixed, quantitative and qualitative), level of analysis (macro, meso and micro), theories used, thematic focus and key findings. As shown in Table 2, the vast majority of studies were published in the late 1990s. This coincided with the radical restructuring of the SOEs which led to the displacement of millions of SOE workers (e.g. Hassard, Sheehan, Zhou, Terpstra-Tong, & Morris, 2007).

Table 2. Number of articles on HRM in SOEs published from 1993 to 2017 (N = 178)^a.

Publication Year	No. of articles found	Publication Year	No. of articles found	Publication Year	No. of articles found
1993	1	2002	10	2011	14
1994	2	2003	8	2012	15
1995	2	2004	7	2013	7
1996	1	2005	4	2014	5
1997	3	2006	14	2015	8
1998	1	2007	5	2016	13
1999	7	2008	11	2017	6
2000	6	2009	12
2001	3	2010	13

^a 1993 was when the first article was found.

Publication outlets

As Table 3 shows, the 178 articles were published in 43 journals in the business and management field (broadly defined), of which 157 were empirical studies (88%) and 21 review articles (12%). Generic management journals were not the main outlets (25/178, 14%). The proportion of articles published in top-ranking journals remains limited, indicating that the quality of the studies in terms of theoretical contributions and empirical rigor may need to be further improved. Only a handful of articles (9/178, 5%) were published in strategic management focused and organizational behaviour (OB) focused journals (8/178, 5%).

Table 3 List of academic journals included in this systematic review (N = 178).

Types of journals	Journal titles	Total number of articles
General business and management journals	Journal of Business Ethics (6); Management and organization review (5); Organization Studies (3); Journal of Management Studies (2); Accounting, organizations and Society (2); Academy of Management Journal (1); Academy of Management Perspectives (1); British Journal of Management (1); Business & Society (1); International Journal of Contemporary Hospitality Management (1); Journal of Business Economics and Management (1); Journal of Organizational Change Management (1)	25
Strategic management and international business journals	Journal of International Business Studies (4); Journal of World Business (2); Strategic Management Journal (2); Journal of International Management (1)	9
Organizational behaviour journals	Journal of Organizational Behavior (3); Journal of Managerial Psychology (2); Leadership & Organizational development Journal (1); Journal of Occupational and Organizational Psychology (1); Organizational Behavior and Human Decision Processes (1)	8
HRM and industrial relations journals	International Journal of Human Resource Management (60); Human Resource Management (7); Employee Relations (4); Human Resource Management Review (3); Personnel Review (3); British Journal of Industrial Relations (2); Human Resource Development Quarterly (2); Human Resource Management Journal (2); Human Relations (2); Industrial Relations (2); International Journal of Manpower (2); Work, Employment and Society (2); Journal of Industrial Relations (1); Industrial and Labor Relations Review (1); International Labour Review (1)	94
Specialized regional studies journals	Asia Pacific Business Review (14); Asia Pacific Journal of Management (8); Asia Pacific Journal of Human Resources (10); Chinese Management Studies (5); The China Quarterly (3); Journal of Contemporary China (1); Modern China (1)	42
Review articles: 21 Empirical articles: 157

The majority of the studies found their home in HRM journals, of which The International Journal of Human Resource Management (IJHRM) has been the main outlet (60/178, 34%). The majority of these studies are based upon comparative analysis, demonstrating the disparate HR practices between SOEs, private enterprises and multinational corporations (MNCs). These comparative studies provide evidence as to ‘whether the development of unique ownership models in China has been a vehicle to level the HRM playing field’ (Zhu et al., 2008, p. 148). A substantial proportion of the articles (42/178, 24%) were also found to have their home in journals specializing in regional studies. These studies have tended to focus their attention on the broad Chinese context and aspects believed to have shaped the unique characteristics of HR practices in SOEs (Warner & Zhu, 2002).

Key features and the evolvement of HRM in Chinese SOEs

HRM in Chinese SOEs has gone through a series of transformation in the last three decades. As the privatization and marketization of SOEs deepens, the HRM model in SOEs has moved along the continuum from the socialist-oriented personnel management system towards the market-driven HRM model.

Under the state-planned economy (1949–1978), the dominant personnel management system in Chinese SOEs is described as a socialist ‘iron rice bowl’ system, including life-time employment, egalitarian compensation, centrally planned job allocation and management promotion, high job security and cradle-to-grave welfare and benefits (Warner, 1993; Zhu & Dowling, 2000). These practices are based on the combination of Soviet regulations and the Chinese traditional values of egalitarianism and obedience (Chen & Wilson, 2003).

Following the open-door policy in 1979, the Chinese government has launched a set of institutional policies to accelerate the marketization and privatization of SOEs. During the process of institutional transformation, the traditional ‘iron rice bowl’ system was abolished and market-oriented HR practices were adopted. More specifically: (1) permanent employment was largely replaced by the contract employment system; (2) performance appraisal was based on meritocratic rather than morality-focused criteria; (3) the egalitarian compensation system became more performance-oriented; (4) the centrally planned job allocation was shifted to market-based recruitment and selection practices; (5) training and development became more systematic and widely used (Akhtar, Ding, & Ge, 2008).

In the new millennium, the HRM system in Chinese SOEs is becoming a hybrid model: a fusion of some market-driven elements of western HR practices with the socialist-oriented vestiges of the traditional ‘iron rice bowl’ system (Warner, 2009; Zhu, Zhang, & Shen, 2012). The majority of SOEs have adopted some elements of western HR practices and abolished certain traditional people management practices. However, the socialist-oriented ‘iron rice bowl’ regime has not been fully phased out, and its lingering effect on the contemporary HRM system is still observable (Chen, Su, & Zeng, 2016). For example, social benefits and welfares are provided to a greater extent in SOEs than in non-SOEs, which may be a remnant of the cradle-to-grave welfare system (Zheng, Morrison, & O’Neill, 2006). Egalitarianism in pay still prevails in SOEs, and therefore managers may find it difficult to fully implement the western practices of pay-for-performance (Wang, Nicholson, & Zhu, 2009). In addition, recruitment and promotion practices are not completely market-oriented: the central and local governments still possess the administrative power to influence the appointment of senior executives and core staff in SOEs (Hassard, Morris, Sheehan, & Xiao, 2006).

Taken as a whole, HR practices in Chinese SOEs consist of a mix of the market-based HRM model in the western countries and elements of the socialist ‘iron rice bowl’ regime, which may differ from the non-SOE and non-Chinese contexts.

Theoretical underpinning of studies of HRM in Chinese SOEs

The bulk of empirical research on HRM in Chinese SOEs is descriptive, with limited theory building/extension, in that 22% (35/157) of the empirical articles lack clear theoretical underpinnings. This finding is in line with Cooke’s (2009, p. 20) observation that ‘a relatively large proportion of the empirical studies, particularly those published in the late 1990s and early 2000s in the non-top-ranking journals, tend to be descriptive and not sufficiently theoretically informed.’

Since the mid-2000s, cross-fertilization between HRM and other fields such as strategy, economics, organizational psychology and sociology, has been more frequent. This has enabled HRM scholars to bring together theories from other disciplines to provide an interdisciplinary picture of what constitutes ‘real’ HR practices (Zhu et al., 2008) in Chinese SOEs. As indicated in Table 4, studies published since the mid-2000s have mobilized a range of theories, such as institutional theory, strategic HRM and OB theories (also see Figure 1).

Figure 1. Theoretical paradigms of HRM research on Chinese SOEs.

Table 4. Major theories used in empirical articles (N = 157).

	Publication period
Theories^a	1993–1997	1998–2002	2003–2007	2008–2012	2013–2017	Total
Organizational behavior	0	2	7	17	16	42
Cultural theories	0	3	8	9	7	27
Institutional theories	1	8	4	4	8	25
Strategic HRM	0	2	6	10	4	22
Social relationship	1	1	1	6	4	13
Gender	0	1	0	3	2	6
Resource-based view	0	0	2	2	2	6
Sociological theories	0	1	3	0	0	4

^a A few empirical studies may adopt two or three theoretical frameworks. Instead of listing all the theories, this table only includes the main theoretical foundations appearing in 157 empirical studies.

Institutional theory: its previous dominance and current fading trend

Institutional theory has initially been a dominant theory in the research of HRM in Chinese SOEs. This is not surprising, given that the earlier studies were preoccupied with the reform of SOEs, where the state was the driver of radical reform as well as the employer of a significant proportion of the urban workforce (see Table 1). Subsets of institutional theories have been mobilized to present a moderate picture of the motives, processes and outcomes of SOE reform and its consequences for HRM (see Table 5).

Table 5. A summary of theoretical foci, main assumptions and examples of studies.

Theories^a	Theoretical focus^a	Main theoretical assumption	Examples of studies
Institutional theories	Path dependence	The evolutions of HR practices are to some extent in historically situated institutional contexts	Chen et al. (2016); Zhu et al. (2012)
	Institutional isomorphism	A set of HR practices in SOEs are becoming similar with those in private and foreign invested enterprises in part because they are under similar institutional constraints	Gamble (2003); Whelan and Muthuri (2017)
	Regulative institution	A wide variety of government policies, legal systems (e.g., contract responsibility system and modern enterprise system) and coercive rules shape HR dimensions related to SOEs	He and Fang (2016); Morris et al. (2001)
	Normative institution	HR activities in SOEs are under pressure from informal constraints, such as moral responsibilities, ethical systems and work norms	Björkman (2002); Cooke et al. (2015)
	Institutional logic	SOEs would adjust their HR practices to conform to multiple or incompatible institutional logics, such as legitimacy-oriented and market-oriented logics	Huang (2016); Wang et al. (2016)
Cultural theories	Hofstede’s (1980) model on cultural dimensions	China is characterized in Hofstede’s (1980) cultural dimensions as ‘high power distance’. As such, there could exist low trust among (horizontal) coworkers and a high level of standardization between hierarchically differentiated (vertical) positions	Kim et al. (2016); Kwon (2012)
	Confucian culture	Confucian cultural values have considerably facilitated employees’ loyalty, respect for authority and seniority, and workplace harmony in SOEs	Huang and Gamble (2011); Snell and Tseng (2003)
	Corporate culture	Every SOE is equipped with its unique organizational culture, which acts as a kind of social control that guides employee behaviours and business strategies	Chen, Wakabayashi, and Takeuchi (2004); Liu (2003)
Strategic HRM theories	Strategic HRM	Human resources are valuable resources for corporate success. HRM systems in SOEs should be in line with their business strategies. In doing so organizational performance could be improved	Ding and Akhtar (2001); Wei et al. (2011)
	High-performance work systems	There are a bundle of best work practices (e.g., performance appraisal, sophisticated training, and information sharing) which can give rise to firm performance	Zhang and Li, (2009); Zhang and Jia (2010)
	High-involvement work systems	High-involvement work systems comprise several flexible practices, including employee participation, information sharing, autonomous problem solving and work design, which may lead in some ways to better firm performance (Boxall & Macky, 2009)	Ouyang et al. (2016)
Social relationship theories	Guanxi	Guanxi not only symbolizes interpersonal relationships but also represents semi–family ethics (i.e., responsibility for coworkers and supervisors as well as community identity) rooted in the workplace	Chen, Friedman, Yu, and Sun, (2011); Gu and Nolan (2017)
	Social network	A mix of instrumental and expressive networks in the workplaces have an impact upon employee behaviour and task performance. Moreover, the business–to–government and business–to–business networks may also shape the evolution of HR practices in SOEs	Han and Han, (2009); Ma (2012)
	Social capital	Companies’ HR decisions are somewhat based on social capital criteria. To be more specific, companies are more likely to hire employees with rich social capital in order to grapple with the market competition	Zhang, Liu, Loi, Lau, and Ngo, (2010); Zhang and Lin (2016)
OB theories	Organizational identity	Organizational identity is closely linked to employees’ attitudes (i.e., intentional turnover and work satisfaction) and behavioural outcomes (i.e. task performance)	Begley et al. (2006); Xing and Liu (2016)
	Motivation theory	Employees have a low work motivation in traditional work units, like danwei, whereas the modern HRM system is able to motivate employees	Ebrahimi (1999); Zhang et al. (2016)
	Organizational commitment	Individuals who are committed to their companies tend to have an emotional connection with their companies, which could have an influence upon job satisfaction, individual and organizational performance	Chen, Wang, Huang, and Rodgers (2012); Wang (2004)
	Organizational citizenship behaviour	Organizational citizenship behaviour is investigated as a mediating factor of the association between high-performance work systems and firm performance	Sun, Aryee, and Law (2007); Wong, Ngo, and Wong (2006)
	Leadership theory	Different types of leadership styles (e.g., servant and transformational leadership) have a direct effect on the relationships between employees and managers and may indirectly influence employee behaviour	Huang, Shi, Zhang, and Cheung, (2006); Zhang et al. (2011)
	Expectancy theory	Expectancy theory claims that a suitable role and expectancy of that role may enhance job motivation, leading to high performance both at the individual and organizational level (Guest, 1997)	Wang, Yi, Lawler, and Zhang (2011); Wu, Chen, and Leung (2011)
Gender theories	Gender	The career experiences (e.g., recruitment, pay, and promotion) of men and women are different in Chinese SOEs due to gender discrimination	Leung (2002); Xiu and Gunderson (2013)
Resource- based view	Resource-based view	SOEs adopt certain HR practices based on their existing resource endowments, including both observable assets and intangible capabilities	Zhou, Li, Zhou, & Su (2008); Zhu et al. (2005)
Marxist theories	Class theory	The marketization of SOEs has initialized employees’ class consciousness, enabling labor struggle to trigger positive outcomes	Chen (2006)
	Socioeconomic framework	Workers’ motivation to engage in education and training is associated with their position vested in the socio-economic systems as well as their life course	Price & Fang (2002); Xiao & Tsang (2004)
Critical sociological theories	Labor process theory	The transformation from the state-planned to the market economy has brought about significant changes in diverse components of the labor process	Chiu (2006)

^a Not all the theories and theoretical foci are included in this table due to space limitation. We only listed the most used theories explicitly adopted in previous empirical studies.

In general, current studies assume a ‘top-down’ mode of institutional forces, presenting SOEs as passive recipients of pressures from a competitive business environment together with Chinese broader economic reforms. In other words, it has been argued that institutional forces can exert considerable control in a unilateral way upon compliant SOEs. Heeding Scott’s (2005) call for supplementing and amending this one-way and determinist institutional framework, researchers need to explore further the presence of ‘bottom-up’ or interactive models by viewing SOEs as active players with creative capacity for responding to diverse contexts (c.f., Jing & McDermott, 2013). In doing so, researchers might be able to better recognize the agency role of SOEs (along with their HR staff) and their capabilities of altering, or at least shaping, governmental regulations and professional standards which guide, rather than determine, their HR dimensions.

Growing popularity of OB theories and strategic HRM

There has been a rapid growth of interest in adopting an OB theory in examining HRM of SOEs, as evidenced by 79% of the studies (33/42) based on OB theories having been published since 2008 (see Table 4). The unit of analysis is moving away from the macro institutional level of key stakeholders toward the micro level of individual employees. Current research draws its theoretical foundations extensively from organizational psychology and is attracted to individuals within SOEs as the focus of study. Divergent OB theories have been revealed, such as identity theory (e.g. Begley, Lee, & Hui, 2006), motivation theory (e.g. Zhang, Deng, Zhang, & Hu, 2016) and leadership theory (Zhang, Cone, Everett, & Elkin, 2011). It has been argued that organizational psychologists regard employees as ‘cognitive dopes’, constrained by rudimentary rules of traits and behaviour and without capability to interpret the heterogeneity and complexity of realities (Wood, Roberts, & Whelan, 2011). Consequently, OB scholars are likely to construct intricate models that abstract employees from the real workplace. Participants are then evaluated via these models through sophisticated research design composed of a number of hypotheses about micro affection and motivations and with advanced techniques for data analysis (Godard, 2014; Zickar, 2004). This is detrimental to developing a holistic understanding of what goes on in the SOEs, because the strong and growing preference for psychologizing HRM research bears little connection with the actual realities of work and employment in workplaces that have unique and evolving institutional and cultural characteristics.

In addition, various strategic HRM theoretical lenses, such as strategic HRM theory (e.g. Ding & Akhtar, 2001), high-performance work systems (e.g. Zhang & Li, 2009) and high-involvement work systems (e.g. Ouyang, Liu, & Zhang, 2016) have been used in the research on HRM in Chinese SOEs. This body of strategic HRM literature has primarily adopted a content-based theoretical assumption which highlights the inherent power of the content of the HRM system to interpret performance (Wei & Lau, 2008; Zhang & Morris, 2014).

However, they neglected the unique psychological process through which employees make sense of the HRM system in SOEs and contribute to firm performance. We argue that the psychological processes in the Chinese SOE context can differ from those in non-Chinese and non-SOE contexts, due to the impact of vestiges of the ‘iron rice bowl’ system. For example, employees with long organizational tenure in traditional SOEs may have become used to the socialist-oriented egalitarian pay system, because they view themselves as the ‘Master(s) of the country’ (Zhu et al., 2012). In this regard, they could perceive the performance-based compensation in SOEs as an unfair HR practice that disadvantages them as well as breaking the harmonious employment relationship. In contrast, employees with experience in foreign-invested companies may prefer the western compensation system. They expect that the performance pay practice is motivated by HRM philosophies of empowering employees and encouraging them to undertake tasks innovatively. Accordingly, future research should embrace a process-based approach (Bowen & Ostroff, 2004; Nishii et al., 2008) to investigate HR attribution in Chinese SOEs to reveal more fully the links between HRM and performance in a specific context.

On balance, the growing preference for OB theories, and to a lesser extent strategic HRM, in recent studies demonstrates that most articles on the HRM of SOEs in China are not engaged in an ‘inside out’ approach (Tsui, 2006) to their theoretical contributions. Instead, they mainly apply well-recognized western theories in the Chinese context, and examine the applicability of these theoretical foundations. As such, these studies make limited contributions to developing a comprehensive understanding of the (unique) features of HR practices in the Chinese SOE context.

Growing indigenous theories: cultural and guanxi theories

Over the past decade, there has been a modest, yet steady, increase of ‘homegrown’ theories, such as cultural and guanxi theories. This is consistent with the call for a contextual focus in HRM studies (Cooke, 2018).

Through an analysis of traditional Chinese values, such as respect for hierarchy, high collectivism and emphasis on harmony (Wang & Fulop, 2007), scholarship aims to construct a fine-grained conceptual HRM model with Chinese characteristics. For example, Zhu, Zhang, and Shen (2012) built a paternalistic and transactional HRM model centered on harmony and loyalty; Zhu, Warner, and Rowley (2007) constructed a hybrid people-management system. Their efforts have been considered as ‘inside out’ approaches, in that Chinese traditional cultural elements are presented in frameworks that can inspire US and European-based HRM systems.

Along with the development of cultural theories, one of the most widely known and useful concepts in Chinese societies—guanxi—has gained increasing attention (Gu & Nolan, 2017). The meaning of Chinese guanxi is more complex than the English equivalents of ‘relationship’ and ‘network’ (Vickers & Fox, 2010), in the sense that guanxi symbolizes not only interpersonal relations but also semi-family ethics (e.g. responsibility for coworkers and community identity) embedded in the workplace ( Luo, Cheng, & Zhang, 2016; see also Chen, Chen, & Huang, 2013 for a comprehensive review). Researchers have been theorizing guanxi patterns in indigenous HRM studies to complement or even substitute for the well-accepted HRM models that are dominant in the western world (e.g. Law & Jones, 2009). In light of the fact that the SOE context is featured as a familistic hierarchical pattern (Zhu et al., 2012), guanxi theory will contribute to future research on the process through which dyadic and sentimental ties play out in the SOE workplace.

Methods used in extant studies

As seen in Table 6, the majority of empirical studies (92/157, 59%) are quantitative, with first-hand questionnaire surveys as a primary method for data collection. It is worth noting that the vast majority of these studies were published after 2000. These quantitative studies adopted various kinds of advanced models for analysis, including a multinomial logistic model (e.g. Xiao & Tsang, 2004), structural equation modelling (e.g. Wang, Yi, Lawler, & Zhang, 2011) and panel logistic regression (e.g. Hass, Tarsalewska, & Zhan, 2016).

Table 6. Empirical articles and methodology (N = 157).

	Methodology
Publication year	Qualitative methods	Quantitative methods	Mixed methods	Total
1993–1997	6	2	1	9
1998–2002	13	10	3	26
2003–2007	9	24	2	35
2008–2012	13	35	2	50
2013–2017	12	21	4	37
Total	53	92	12	157

A couple of factors may explain HRM scholars’ heavy reliance on sophisticated models. First, most US-based journals have a preference for advanced quantitative methods that make research findings more scientifically convincing (Kaufman, 2015, Meyer, 2014). An increasingly demanding requirement of Chinese universities for their staff and students to publish in leading English journals means that resources naturally channel to positivist studies. In fact, many universities only focus on quantitative methods in their postgraduate research training, and qualitative studies may be more difficult to get published. Second, the growing popularity of strategic HRM and OB theories may also contribute to a dominance of quantitative research. These statistical models are based upon either organizational-level analysis (e.g. examining firm performance effects of strategic HR practices) or individual-level analysis (e.g. investigating the impact of HRM systems on employees).

Qualitative studies made up 34% of the empirical articles (53/157), while mixed methods studies accounted for less than 8% (12/157). As noted above, there has been a definitive shift in the choice of methods, and the surge of quantitative studies is at the expense of qualitative studies, a situation further affected by the reduction in general of studies of HRM of SOEs in recent years. Compared with quantitative research, which has a preoccupation with demonstrating statistical links between HR dimensions and firm performance or other outcomes, qualitative research methods are more suitable for documenting rich and deep-level information (Busse, 2017; Zhu & Dowling, 2000). Early studies usually chose small-scale SOEs as cases and used descriptive and general analysis to examine what was going on within and outside the SOEs (Lewis, 2003).

Studies of SOEs were mainly conducted in developed cities and coastal provinces (e.g. Beijing, Shanghai, Guangdong, Jiangsu), which are at the forefront of Chinese economic reforms and open-up initiatives, whilst less developed inland regions remain under-researched (c.f., Jing & McDermott, 2013). Moreover, large SOEs in the manufacturing sector took centre stage in previous research, while less attention has been paid to small- and medium-sized SOEs (c.f., Danford & Zhao, 2012) and those operating in the high-tech (c.f., Derong & Faure, 1995) and service sectors.

In a nutshell, the methodological approaches for studies on HRM in Chinese SOEs have been moving from qualitative interviews toward sophisticated quantitative methods. Nonetheless, the complex and interrelated mechanisms of HRM evolution within SOEs cannot be completely captured by advanced statistical models. As such, we need more longitudinal and in-depth qualitative studies to obtain first-hand information of day-to-day HRM challenges faced by both employers and employees (Tsui, 2006).

Thematic foci and level of analysis of HRM in Chinese SOEs

We utilized the ‘antecedents-phenomenon-consequences’ logic which has been widely adopted in other systematic reviews ( c.f., Narayanan, Zane, & Kemmerer, 2011; Pisani, Kourulla, Kolk, & Meijer, 2017) to organize our findings. Within the three broad categories, we recorded the subject matter examined in these articles and clustered them into sub-categories (see Table 7). This categorization method is able to pinpoint the intellectual structure and evolving trends of research on HRM in Chinese SOEs.

Table 7. An analysis of levels of context/antecedents phenomenon consequence of HRM studies in Chinese SOEs (N = 157).

Themes	Main theories	Main methodology	1993–1997	1998–2002	2003–2007	2008–2012	2013–2017	Entire period
Levels of context/antecedents
National-level context
Institutional transformations	Institutional theory	Qualitative	3	7	5	7	4	26
Traditional culture	Cultural theory	Qualitative	0	1	2	6	6	15
Industrial-level context
Downsizing	Institutional theory	Qualitative	0	7	2	0	1	10
Inter–organizational networks	Social relationship theory	Qualitative	1	1	2	0	2	6
Organizational-level context
Leadership styles	Social relationship and OB theories	Qualitative and Quantitative	0	0	1	4	4	9
Personal relations	Social relationship and OB theories	Quantitative	0	1	1	1	0	3
Organizational culture	Cultural theory	Qualitative and Quantitative	0	0	3	1	0	4
Business Strategy	Institutional theory and Resource–based view	Qualitative and Quantitative	0	0	1	1	2	4
Phenomenon (firm-level)
Key dimensions of HR practices
Compensation and benefits	No clear theories	Qualitative	3	2	3	2	1	11
Recruitment and selection	Social relationship theory	Qualitative	2	3	1	2	1	9
Training and development	No clear theories	Qualitative	3	2	2	0	0	7
Conflict management	No clear theories	Qualitative	1	2	2	1	0	6
Performance management	No clear theories	Qualitative	1	1	0	1	0	3
Labor contract	No clear theories	Qualitative	2	1	0	0	0	3
Career management	No clear theories	Qualitative	1	0	0	1	0	2
Sub total^a			5	6	4	5	2	22
Outcomes
Organizational outcomes
Firm performance	Strategic HRM	Quantitative	0	2	6	9	4	21
Individual outcomes
Intangible outcomes
Employee Perception	OB theories	Quantitative	0	0	2	4	3	9
Organizational commitment	OB theories	Quantitative	0	1	2	3	2	8
Organizational citizenship behavior	OB theories	Quantitative	0	0	3	2	1	6
Employee Satisfaction	OB theories	Quantitative	0	0	1	2	2	5
Employee stress	OB theories	Quantitative	0	0	1	2	1	4
Ethical behavior	OB theories	Quantitative	0	0	0	1	2	3
Employee motivation	OB theories	Quantitative	0	0	0	1	2	3
Organizational identification	OB theories	Quantitative	0	0	2	0	0	2
Tangible outcomes
Job security	OB theories	Quantitative	0	0	1	0	1	2
Financial output	Gender and sociological theories	Quantitative	0	0	0	0	2	2
Individual creativity	OB theories	Quantitative	0	0	0	1	1	2
Individual performance	OB theories	Quantitative	0	0	0	0	1	1
Sub total^a			0	1	8	16	12	37
Total			9	26	35	50	37	157

^a A sizeable proportion of empirical articles included in ‘Key dimensions of HR practices’ and ‘Individual outcomes’ categories examined various sub-categories simultaneously. Hence, the aggregated number listed in the sub-categories would exceed the sub-total number shown in the categories.

It should be noted that quite a few articles are multi-themed. In reviewing these articles, we have tried our best to group them into what we believed to be the most appropriate category, although some degree of subjectivity exists. A detailed analysis of these studies has enabled us to develop an integrative model of existing HRM research on Chinese SOEs (see Figure 2). It also assists in considering the implications for future research.

Figure 2. An integrative framework of research of HRM in Chinese SOEs. *HPWS: high-performance work systems; HIWS: high-involvement work systems; HCWS: high-commitment work systems.

Levels of contexts/antecedents

Articles included in the ‘contexts/antecedents’ category focus on three broad themes. These themes are at three levels of analysis: national, industrial and organizational. Various theoretical frameworks have been constructed to explain the multilevel drivers triggering the development of HR practices in Chinese SOEs. For example, institutional and cultural theories have been used to frame national contexts. Studies exploring industrial factors fostering SOEs’ HR activities have primarily adopted institutional and guanxi theories. Additionally, guanxi, OB and cultural theories have been favoured more in studies that focus on the organizational climate.

Macro-level contexts

Institutional transformations

A number of articles in this category are not theoretically informed, especially those published in the 1990s. In recent years, a majority of studies have adopted institutional theories to focus on the impact of institutional transformation upon the hybrid HRM systems within SOEs.

First, past studies examined how the institutional arrangements created the conditions for the abolition of the traditional ‘iron rice bowl’ regime (Ding et al., 2000). For example, Hassard, Morris, Sheehan and Yuxin (2010) systematically examined the 30-year transformation path (from the late 1970s to the early 2000s) of SOEs through in-depth interviews with executives of ten large SOEs. They argued that the initial ‘Contract Responsibility System’ launched in the late 1980s and the subsequent ‘Modern Enterprise System’ were two step-by-step governmental policies that jointly empowered SOEs to abolish the traditional personnel management system and build a modern HRM model.

Second, other researchers focused on a plethora of institutional constraints that would slow down SOEs’ pace of fully adopting a western HRM system (e.g. Warner, Goodall, & Ding, 1999; Zhu, Warner, & Feng, 2011). For example, the State-Owned Asset Supervision and Administration Commission may influence the recruitment and promotion of crucial supervisors in SOEs (O’Connor, Deng, & Luo, 2006). In this regard, the recruitment and selection system in SOEs is not completely market driven. Furthermore, the trade union in SOEs is controlled by the Chinese Communist Party, and therefore does not possess autonomy in conflict management, which could differ from their western counterparts (Liu, 2010).

Traditional culture

Articles included in the ‘traditional culture’ group drew on cultural elements rooted in the traditional Chinese value of social order and customs (Huang & Gamble, 2011). Although the traditional ‘iron rice bowl’ system has been abolished, the heritage of socialist management is still visible, owing to the last lock-in effect of several cultural factors—including paternalistic values (e.g. Zhu et al., 2012), norms of respect for hierarchy (e.g. Huang, 2008) and collectivist customs (e.g. Liu, 2003). As Zhu et al. (2012) stated, the legacy of paternalism might shape the nature of current HRM systems, characterized as moral motivation, benevolent management and familial employment relationships.

From 2008 to 2017, the amount of research in traditional culture has caught up with studies on institutional transformations, which was the paramount thematic focus in the 1990s. This thematic shift reflects a general theoretical trend from institutional theory to various cultural theories, indicating that HRM researchers are branching out with diverse theories to conceptualize the hybrid form of HRM in Chinese SOEs. The expansion of the range of thematic foci and theoretical lenses require scholars to rely on multi-level, multi-contextual and multi-disciplinary studies. For example, Tang and colleagues (2015) have contributed to the developing concept of inclusion management, identifying several unique Chinese characteristics of the enactment of inclusion management at both the individual and organizational level.

Industrial-level contexts

Articles in this category have reflected scholarly interest in industrial-level contexts and their impact on the development of HRM in SOEs. Based on the subject matter investigated, two sub-themes have been identified: downsizing related to SOEs in particular industries and inter-organizational networks.

Downsizing

We identified ten articles centering on the downsizing over time of the industrial-level workforce in SOEs, which aimed to break the ‘iron rice bowl’ regime and establish a modern HRM system. Most of the articles are descriptive in nature and lack theoretical frameworks. The literature can be divided into two broad categories. Firstly, scholarship has investigated the strategies SOEs in the steel industry used to accomplish the task of downsizing, which practically ended state-controlled appointments and the permanent employment system (Hassard et al., 2006; Morris, Sheehan, & Hassard, 2001). These strategies aimed to soften the impact of large-scale unemployment in the steel industry, because most manufacturing workers only possessed skills specific to the steel industry, had not been exposed to the external labor market all their life, and were therefore rather vulnerable (Hassard, Morris, & Sheehan, 2004). Secondly, a range of problems resulting from industrial restructuring have been assessed. In terms of emotional issues, downsizing of SOEs has brought about growing psychological insecurity among laid-off workers and those who remained (Price & Fang, 2002). In terms of financial problems, forcing out surplus employees in the automobile industry, for example, could have a negative impact on employees’ financial income and benefits (Chen, 2003).

The majority of these studies were published in the late 1990s and early 2000s after the then Premier Zhu Rongji initiated the workforce restructuring strategy (e.g. Cooke, 2000; Hassard et al., 2007). Nearly a decade on, only one article has recast light on this topic (Jing & McDermott, 2013). Furthermore, previous literature placed an emphasis on workforce reduction in the manufacturing industry, with insufficient attention paid to the service industry or the tertiary sector.

Inter-organizational networks

Inter-organizational ties can be categorized into two distinct forms: business-government and business-to-business ties, which have divergent influences on HRM in SOEs. First, business-government guanxi ties have been viewed as a political personnel control affecting the decision-making processes in SOEs (Liang, Ren, & Sun, 2015). As such, some political aspects of the older socialist-oriented HR practices have not been abolished in a real sense. For example, payroll expenditure is still controlled by the political institution (Chen et al., 2016). The governmental agency, rather than the SOE, is in the driver’s seat when it comes to conflict management and negotiations (Derong & Faure, 1995).

Second, business-to-business networks, especially the interactions between SOEs and western MNCs, have been investigated as factors conducive to the generation and progression of western HR functions in Chinese SOEs (Björkman, 2002). For instance, Wang, Bruning, and Peng (2007) indicated that the overall relationships between SOEs and foreign-invested enterprises in China motivated SOEs to realize the potential of internally consistent HR policies and practices. These have been widely introduced in foreign partners, and therefore dramatically changed centrally planned HR policies and practices prevailing in Chinese SOEs.

Organizational-level contexts

The majority of articles in this category were published after 2003, indicating that there is a growing interest in organizational-level contexts over recent years.

Research on ‘leadership styles’ constitutes the largest part of the literature on organizational-level contexts. Most of these leadership styles, including aesthetic leadership (Zhang et al., 2011), participative leadership (Huang, Shi, Zhang, & Cheung, 2006), and ethical leadership (Liu, Kwan, Fu, & Mao, 2013), are not specific to Chinese SOEs. Researchers mainly applied these western terminologies in the SOE context to test their generalizability, but they seldom provided an in-depth analysis of the key features of SOEs. Zhang, Cao, and Tjosvold’s (2011) study is an exception. They argued that transformational leadership, defined as leaders encouraging subordinates to transcend self-interest and prioritize group interest, has unique implications for SOEs that seek to develop a cooperative, as opposed to competitive, strategy for conflict management. This cooperative approach is consistent with the SOE context, where collective interest may be preferable to individual interest.

Another cluster of studies has explored organizational culture and its implications for HRM. This body of research bears similarities with the theme of ‘traditional culture’ in that both explore the cultural types of Chinese characteristics. However, organizational culture studies examine the values and beliefs embedded at an organizational level rather than at national level. For example, Wei, Liu, Zhang and Chiu (2008) assess the differences in the corporate culture of different ownership structures. They maintain that, even after two decades of transformations, SOEs still have a higher level of hierarchical organizational culture and are less likely to realign their HRM system to meet market demand than their foreign counterparts.

Finally, the last cluster of articles identified in this category focuses on whether (or how) business strategies relate to HRM. According to these studies, the relatively high level of political involvement and administrative intervention may prioritize managerial attention to state interests and impede them from being independent in business operations (Cordeiro, He, Conyon, & Shaw, 2013). In this sense, SOEs can be faced with a set of HR-related challenges (i.e. job insecurity, labour disputes and adoption of western HR strategies) throughout their international expansion (Cooke et al., 2015) and marketing planning processes (Zhou, Li, Zhou, & Su, 2008), for example, when developing joint ventures with foreign firms.

Firm-level phenomenon: key aspects of HRM studied

This thematic category contains firm-level phenomenon studies primarily targeting the functional HR practices in Chinese SOEs, ranging from compensation, recruitment and training, to performance management. Having segmented the literature, we discern four general themes. First, functional HRM topics were popular two decades ago, featured in five out of the nine articles published from 1993 to 1997 and 6 out of the 26 articles published from 1998 to 2002. These studies focused on the end of the ‘iron rice-bowl’ system which epitomized traditional management practices before the economic reforms in 1979. Key findings include that: (1) life-time employment was replaced by the fixed-term contract system; (2) the egalitarian pay system was shifted towards the performance-based compensation system; (3) the seniority- and morality-based performance appraisal practice was changed to merit-orientation; (4) the centrally planned job allocation was transformed to market-oriented recruitment and selection practices and (5) ‘cradle-to-grave’ welfare was phased out (Warner, 1996; Zhu & Dowling, 2002).

Second, there is a paucity of empirical research on the unique HR dimensions of Chinese SOEs in recent years (5 out of the 50 articles published from 2008 to 2012 and 2 out of the 37 articles from 2013 to 2017). Among these limited studies, the hybrid form, involving a fusion of market-driven HR practices with remnants of the socialist ‘iron rice bowl’ system, is the main focus. The adoption of western HRM systems in SOEs is somewhat fragmented and partial. For instance, Huang (2016) observed that SOEs are motivated to partially implement the market-oriented responsible pay system, because they are subject to both administrative intervention and market competition. In a similar vein, Kong, Cheung, and Zhang (2010) found that western career management systems are not comprehensively deployed in SOEs. Managers tend to pay attention to the pre-job and on-the-job training activities, rather than to the formal training programmes commonly practised in western companies.

Third, there was a dearth of theoretical and methodological frameworks in articles published in the early stage of research, while recent articles are becoming more methodology-oriented and theory-focused. For example, Zhang and Lin (2016) utilized a three-level multilevel mixed effects model to demonstrate that foreign-invested enterprises are likely to set higher social capital criteria in the recruitment process than are SOEs. Moreover, when it comes to recruitment and selection, guanxi theory has been favoured by authors as the main theoretical framework (e.g. Han & Han, 2009; Huang, 2008). It has been argued (Gu & Nolan, 2017) that the ongoing SOE restructuring associated with economic transformation has created considerable uncertainty as to how to hire the best talent for a given position. As a consequence, HR managers have relied on their social networks to look for suitable job applicants (e.g. Bian, 1994; Huang, 2008; Zhang & Lin, 2016). This indicates that recruitment in SOEs is not entirely dependent on the market, which may differ significantly from the non-Chinese and non-SOE context.

Fourth, the majority of articles published in early phases of research covered a bundle of HR dimensions simultaneously (e.g. Lewis, 2003; Zhu & Dowling, 2002), and some dimensions were more underscored than others, such as compensation, recruitment and training. In contrast, recent studies have made great efforts to bring the single under-researched HR function into analysis, such as training programs (Kong et al., 2010) and benefit schemes (Lin, Kelly, & Trenberth, 2011). Whilst it is vital to investigate individual HR practices or HR subsystems in contemporary SOEs, a systems perspective of HRM that examines the overall set of HR activities needs more attention. This is because a configuration or aggregation of HR practices may provide a more comprehensive insight into the key features of HRM systems in modern SOEs.

Outcomes

Based on the level of analysis, articles included in the ‘Outcomes’ category were grouped into two broad themes: firm-level and individual-level outcomes of HR practices. Strategic HRM and OB theories have logically been the main theoretical frameworks used to shed light on the firm and micro-level consequences respectively.

Strategic HRM and firm performance

Studies in this category have discussed the organizational-level effects of HR practices. While strategic HRM studies also touch on a range of HR practices that should be considered as HRM, we classified them into an ‘Outcomes’ category. This is because strategic HRM research pays special attention to firm performances derived from the pattern of planned HR activities (Wright & McMahan, 1992).

Strategic HRM studies did not appear until the year of 1999, when Ebrahimi (1999) published an article on the diffusion of strategic HRM among Chinese firms with various ownership forms. Since then, this field of research has grown rapidly. Studies categorized in this theme primarily use sophisticated statistical models to compare the HRM-performance linkages among types of ownership. As a key firm characteristic, ownership type is usually included in the quantitative models as the independent, control or moderator variable. First, a handful of studies considered ownership type as an independent variable, explaining the association between ownership type and the adoption level of strategic HRM (dependent variable) (Law, Tse, & Zhou, 2003; Ouyang et al., 2016). On the one hand, the levels of implementation of strategic HR practices were lower in SOEs than in their foreign counterparts. This was because SOEs had to face more institutional scrutiny, and enjoy less autonomy or flexibility in adopting western-based high-performance work systems (Law et al., 2003; Zheng et al., 2006). On the other hand, compared with the private enterprises, SOEs were more likely to adopt high-involvement work systems. In recent years, the government agency is advocating harmonious employment relations (taking employees as the foundation). In light of the fact that SOEs are subject to administrative control and rely on governmental resources for development, they are motivated to respond to the governmental call by introducing a series of HR practices aiming to enhance employee involvement and wellbeing (Ouyang et al., 2016).

Second, ownership type has been absorbed into the quantitative models as a control variable because of its predicted effect on organizational performance (Zhang & Jia, 2010; Zhu et al., 2008). Akhtar et al. (2008), for instance, controlled for the impact of ownership type in their model and observed that the adoption of strategic HRM improved product performance. They also elucidated the effect of ownership type (control variable) on performance: compared with SOEs, foreign enterprises might generate better product performance, as managers in foreign enterprises were more likely to configure their HR practices to fit in with their desired competitive strategy.

Third, ownership type may act as a moderator on the HRM-performance relationship (e.g. Wang et al., 2011; Zhang & Morris, 2014). For example, Wei, Liu, and Herndon (2011) suggested that the relationship between strategic HRM and product innovation would be moderated by corporate culture. Given the fact that SOEs are less market-oriented and have a less developmental corporate culture, the relationship would be stronger for SOEs than for their counterparts in the private sector.

Despite the tremendous effort scholars have made, there remain some downsides in approaches to the measurement of HRM-performance relationship in SOEs. Perhaps one of the most important drawbacks is that researchers have underplayed the unique features of organizational performance in SOEs.

Extant studies focus mainly on financial performance (e.g. Akhtar et al., 2008; Qiao, Khilji, & Wang, 2009), while ignoring non-financial or public service performance (e.g. social responsibility, public image and good will). The absence of non-financial performance measures, we argue, is flawed, because SOEs, as hybrid organizations, not only pursue financial organizational goals but also public and non-profit goals. In some cases, SOEs might not regard profit maximization as their top priority (Bruton et al., 2015). Hence, we cannot assume that SOEs will implement their HR strategies in the same way as their foreign and private counterparts who focus largely on profit or market value. Unfortunately, we only found one study taking into consideration the unique organizational performance in SOEs (public image and good will) (Zhang & Morris, 2014).

Strategic HRM research has experienced a downturn since the mid-2000s—only four articles have been published since 2013. This substantial decline may indicate that scholars no longer view ownership type as an important variable in explaining HRM-performance linkage, or it may be that researchers have turned to the private sector for strategic HRM research.

Individual outcomes

The other subgroup of ‘outcomes’ studies have investigated individual-level outcomes resulting from the multilevel drivers of HRM evolvements in Chinese SOEs. As indicated in Table 7, individual consequences studies, primarily using OB theories as their theoretical framework, have been burgeoning in recent years. As also discussed earlier, this may reflect the general trend of HRM research over the last two decades (Cooke, 2018; Godard, 2014).

Individual outcomes are both intangible and tangible, having been logically epitomized in a range of subcategories. Among these, sub-themes on employee perception, organizational commitment and organizational citizenship behaviour, are predominant. There are several characteristics of this body of research. First, the majority of studies (24 out of the 37 articles in this theme) did not have SOE as a key focus. Instead, they examined individual outcomes within a wide range of ownership forms for comparison (e.g. Jiang, Wang, & Zhao, 2012; Wong, Wong, Ngo, & Lui, 2005). Unfortunately, there was no consensus amongst previous literature on whether employees in different ownership organizations had different levels of identification or satisfaction with HR practices. For instance, based on the data from a large auto manufacturing SOE and a joint venture, Yu and Egri (2005) revealed that employees in the SOE had a lower level of satisfaction with their HR practices than those in the joint venture, because the joint venture had adopted high-performance work systems to a greater extent than the SOE. In contrast, Huang and Gamble (2015) recently studied 22 state- and foreign-owned retail firms. They found that there was not a significant relationship between male employees’ satisfaction and ownership types, even though HR practices and employment relations varied between SOEs and foreign enterprises. These mixed findings can perhaps stem from certain cross-contextual variables (e.g. location, industry and time).

Second, a few studies (13 out of the 37 articles in this category) chose SOE as their single context to provide an in-depth analysis of employees’ special needs and perceptions of HRM in SOEs (Begley et al., 2006; Fu, 2014). Tsai and Wang’s (2013) study on employee attitudes towards performance appraisal accuracy is a good example. Western performance appraisal practices have been increasingly adopted in Chinese SOEs, but the tradition of subjective and non-job-related assessment criteria is by no means uncommon. The performance appraisal system was viewed as accurate by employees when SOEs broke away from the traditional management style and established impartial, objective and task-related evaluation criteria.

Third, among 37 articles related to individual outcomes, only seven (19%) touched on tangible outcomes. This means that existing studies focus extensively on intangible or psychological outcomes while marginalizing tangible or material outcomes like job insecurity (Wong et al., 2005) and financial rewards (Gunderson, Lee, & Wang, 2016). The lacuna of tangible/material individual outcomes is not surprising, given the growing dominance of OB theories that regard employees as psychologized objects and instruments, subject to personal traits and cognitive capabilities (Wood et al., 2011). Nevertheless, this presents an important research gap that may be unique to China as a less developed country with a paternalistic culture. This is even more so when considering SOEs, in that they have a long and heavy legacy of extensive workplace welfare and benefits provision to supplement low wages. Like other less developed Asian countries, workplace benefits play an important role in smoothing employment relations and in boosting workers’ total income. This is in part due to the paternalistic culture, and in part because the majority of the people are still not well off and are eager to improve their material life. Indeed, material incentives have proved to be more effective than they have been given credit for in recruiting and retaining talent (e.g. MichaelPage, 2016; Nankervis, Cooke, Chatterjee, & Warner, 2013).

Discussion and future research avenues

The hybrid nature of HRM in Chinese SOEs and its implications for general HRM research

The current HRM system in SOEs can be seen as a hybrid form, combining some aspects of western HR practices with the legacy of traditional management systems (Warner, 2009). We believe that the use of hybrid model in the Chinese SOE context has twofold implications for general HRM research.

First, most HRM studies view organizations dichotomously. That is, organizations are labelled as either for-profit or non-profit organizations (e.g. public sector and government organizations) (Guo et al., 2011). Hybrid organizations which blend the elements of for-profit organizations with some aspects of non-profit organizations have not attracted much attention. Due to the absence of hybrid organizations, the extant theoretical frameworks and empirical findings may not be generalizable. In order to fill this gap, we need to bring Chinese SOEs into the analysis. As business firms, SOEs may be charged with achieving profit or market value goals on the one hand. In this regard, it is unsurprising that a large number of SOEs adopted several market-driven HR practices (e.g. performance-oriented pay, market-based recruitment and contract employment system) to enhance firm performance. On the other hand, as public sector organizations, SOEs could highlight state/public interests and pursue public and non-profit goals (Bruton et al., 2015). They may be accountable to the state or public at the expense of organizational efficiency. In other words, the way in which SOEs operate differs considerably from non-SOEs. Therefore, we maintain that studying SOEs, or hybrid organizations more broadly, contributes to existing HRM research by extending the scope of research subjects.

Second, current research tends to categorize HR practices in into either commitment-focused (employee well-being and service quality) or control-focused models (cost reduction and exploiting employees) (Nishii et al., 2008). The possibility that two contrasting models may co-exist in a single HRM system remains less explored, with some exceptions (e.g. Cooke, 2004). Examining HR practices in Chinese SOEs could fill this research gap. As discussed before, SOEs are likely to adopt market-driven HR practices to respond to the competitive market on the one hand. These strategic HR practices bear some similarities with the high-performance work systems in the western context, with a primary concern for reducing costs and improving profits. On the other hand, the vestiges of the socialist-or administrative control are visible (e.g. comprehensive benefits and welfare and egalitarianism in pay), reflecting the conformance to the socialist ideology in China. When implementing HR practices in SOEs, managers are likely to show sufficient concern for the interests of multiple stakeholders (e.g. the governmental agencies, the public, ordinary workforce, managers, private and foreign enterprises), rather than focusing merely on the SOEs themselves. The processes through which SOEs balance the two competing HRM models (market-driven HR practices and socialist-oriented personnel management system) in a single system could add value to the current HRM theory, which is largely based on one specific HRM model.

Although past studies have to some extent revealed the key features of HRM in SOEs, a number of issues remain unexplored. In the next section, we identify further research avenues for future studies (Table 8). Many of these research areas have implications for both academics and practitioners.

Table 8. A summary of current research gaps and directions for future research of HRM in SOEs.

Subject matters	Current research gaps	Directions for future research
Theoretical matters	Lack of homegrown theories	Carrying out the ‘inside out’ approach: building indigenous HRM theory with Chinese unique characteristics, such as the hybrid HRM model
	Absence of mutual connections between theories at different levels	Redirecting special attention to the heterogeneity and complexity of macro-level institutional settings and combining theoretical foundations at different levels
Methodological matters	Problems related to performance measure	Placing a greater emphasis on subsequent performance rather than past performance Conducting more research on non-financial and public performance (e.g., public service performance and public image)
	Limited data collection methods	Acquiring ‘big data’ on a large number of SOEs across the country Requiring more unstructured interviews and in–depth ethnographic studies
	Lack of narrative methods	Utilizing narrative techniques to investigate the lives and experience of employees in SOEs
Thematic matters	Remiss about recent institutional reforms in SOEs	Attending to the up-to-date national reforms, such as the reforms in SOEs’ compensation systems, and exploring their short-term and long-term influences on HR practices
	Insufficient attention paid to functional HR practices	Focusing on the recent impact of ICT on single HR functions (e.g., training, performance management, and talent management) in SOEs
	Dearth of studies on single SOE issues, such as the evolution of pay practices	Considering SOEs together with their special HR practices as a single subject matter to bring forth a more thorough explanation regarding any new and particular developments in SOE-related HR activities

Theoretical matters

While articles published in the 1990s rarely articulated their theoretical underpinnings, this gap has been gradually filled in recent studies. Current research primarily adopts multiple theories to explore the HR practices within Chinese SOEs. Despite their considerable contributions to constructing theoretical paradigms, existing studies share a couple of limitations: (1) the large absence of real homegrown theories; and (2) a dearth of mutual connections between theories at different levels of organizational analysis.

First, the majority of researchers are employing the ‘outside in’ approach by borrowing western-oriented theories (strategic HRM and OB theories in particular) to examine Chinese HRM phenomenon. This approach assumes that Chinese SOEs can serve as an illustration as well as confirmation for the explanatory validity of western HRM theories. The over-emphasis on western-oriented theoretical frameworks, however, affords little insight into the unique characteristics of HRM in SOEs, since patterns of economic development, regional cultures and political systems of China bear little resemblance to western contexts (Warner & Rowley, 2010). Indeed, Chinese scholars are searching for, and debating if there is, a Chinese theory of management, or just a management theory of China (Barkema, Chen, George, Luo, & Tsui, 2015; Barney & Zhang, 2009). Some researchers adopted an ‘inside out’ approach (Warner, 2009; Zhu et al., 2007). This approach, however, may need to be further refined to take into consideration multiple variations (Zhu et al., 2012). For example, in terms of individual-level variations, would core and peripheral employees have similar perceptions of the hybrid HR practices? In terms of organizational-level variations, to what extent would HR practices vary across firm sizes and industries, even amongst SOEs? In terms of institutional-level variations, would the hybrid HRM model be able to capture the nature of HR practices in different local institutional contexts? Future research may pay more attention to these variations in order to examine the generalizability of the hybrid HRM model and extend HRM theories.

Second, recent literature has a micro-level orientation of OB theories and is deficient in interactions between theoretical foundations at different levels. In articles in the first 10-year period in this review (1993–2002), institutional theory focusing on complex and dynamic political, business and societal environments has occupied a central stage. Studies in the ‘contexts/antecedents’ category primarily mobilize institutional and cultural theories to demonstrate the impact of contexts on the hybrid HRM systems in SOEs. In the second decade (2003–2012), strategic HRM, highlighting organizational-level practices and their positive relationships with organizational performance, have become a popular focus. Studies included in the ‘firm-level outcomes’ category typically used strategic HRM theories to investigate the organizational-level effects of HR practices. In recent years (2013–2017), OB theories, rooted in individual-level traits and behavioural styles, are the predominant theoretical frameworks. The majority of studies in the ‘individual outcomes’ category relied on OB theories to explore the individual outcomes of the development of HRM systems within SOEs. In general, these three theoretical paradigms are connected with distinct thematic foci based on distinct analytical levels and are seldom combined to provide a multilevel analysis. To make things worse, the recent bias towards micro-level OB theories inevitably results in a reduction of articles at a more macro, organizational level of analysis. Considering that the genealogy of HRM research in China has a rhizomatous relationship with traditional institutionalism, the lack of attention to the heterogeneity and complexity of institutional settings may be detrimental. Thus, future research should combine various theoretical paradigms and analytical units to develop a more nuanced understanding of HRM inside SOEs. More importantly, instead of applying existing western-developed theories to the research, future studies may examine new evidence from SOEs and China more broadly. They may also investigate how this new empirical evidence may challenge existing theories and lead to the development of new theoretical insights.

Methodological matters

Several methodological considerations warrant further discussion here. First, existing studies of HRM in Chinese SOEs have produced multiple statistical models with sophisticated hypotheses and advanced techniques for data analysis, but there are several drawbacks in relation to organizational performance measures. For instance, many studies used a post-predictive measurement design (Wei & Lau, 2008), examining the functions of HR practices after a performance period. This is inconsistent with the underlying principle of examining causal relationships, as these HR practices actually reflect past performance instead of subsequent performance (Wright, Gardner, Moynihan, & Allen, 2005). Hence, future research should be designed to measure SOEs’ subsequent performance after the implementation of strategic HR practices. Moreover, current studies mainly pay attention to financial performance rather than public service performance. Future studies may be designed to incorporate a wide variety of non-financial performance measures to capture the hybrid performance of the SOEs in relation to their HR practices.

Second, first-hand surveys and small-scale semi-structured interviews (usually much fewer than 60 interviews) have been the main methods adopted in previous research. The scarcity of large-scale secondary databases presents a challenge to developing a fuller understanding of the hybrid HRM model in SOEs. Future studies could adopt more innovative approaches to data collection. For instance, ‘big data’ on thousands of SOEs across China is required to make a systematic comparison of SOEs located in different regions. This task is made possible with growing digitization and the interest of organizations in China poised to exploit the opportunities of big data. Moreover, unstructured interviews and in-depth ethnographic studies (e.g. participant observation) may be beneficial in revealing the deep-level mechanisms between broad institutional environments, organizational-level HR strategies and practices and micro-level individual perceptions and behaviours.

Third, earlier narrative accounts of HRM in SOEs have provided insights into the lives and experience of employees. However, the shift towards quantitative studies in recent years means that these insights have been largely lost. In fact, the last 5-year period (2013–2017) of our analysis has seen a substantial reduction of research interest in SOEs specifically. This is perhaps because SOEs have been absorbed into the quantitative studies of various ownership forms. It may also be a result of the significant and continuing reduction of the proportion of SOE employment in the economy as a whole.

Thematic matters

Although previous studies have shed light on multiple thematic foci, ranging from contexts, HR dimensions, to outcomes, several important research themes remain under-researched.

First, the subjects of institutional reforms, industrial downsizing and HR practices in SOEs have been well documented in the 1990s. Most previous research, likewise, was indicative of the future of the HRM transformation trajectory. Nevertheless, their key predictions were contingent upon the institutional changes happening nearly two decades ago, which may be out-of-date given the more dynamic and competitive business environments in the current global age. Since the mid-2000s, the central government has proposed a set of policies to further the reforms of HRM systems in SOEs. For example, according to the State-Owned Assets Supervision and Administration Commission established in 2006, the central government still maintains a supreme position and guides the operation of SOEs in pillar industries. In addition, in 2018, the State Council of China carried out another round of reform of the compensation system in SOEs. These reforms not only influence millions of employees working in the SOEs, but can also have an effect on the economic development of China at a broad level. Nevertheless, these phenomena have yet to be studied. Thus, we call for more empirical studies to examine the recent institutional reforms and investigate their impacts on the current hybrid HRM form in SOEs.

Second, insufficient attention has been paid to functional HR practices in Chinese SOEs in recent years. To the best of our knowledge, core functional aspects of HRM, including training, talent management, performance management and remuneration have not been explored in depth in the past few years, although they often form part of the HPWS examined. In addition, as the application of digital technology in the HRM has developed rapidly in recent years, how have SOEs made use of digital technology and HR analytics to inform their HRM decisions? Similarly, with the increased global competition, how have SOEs implemented talent management practices to maintain knowledgeable and skilled workforce?

Third, quite a few articles published in the early stages treated SOEs as a single subject matter with the purpose of offering a more in-depth picture of the unique HR practices of SOEs ( c.f., Child, 1995; Morris et al., 2001). Conversely, in the last decade, most researchers have empirically examined HR practices in disparate Chinese ownership enterprises as a group, and focused on the similarities and differences between them (c.f., Kim, Yoon, Cho, Li, & Choi, 2016). Although comparative studies of HRM across different ownership forms are informative, the drawback is that these comparative studies are often at the expense of an in-depth understanding of what goes on in SOEs.

For instance, at an organizational level, are workforce reductions still pending and have they achieved the desired outcomes (facilitating efficiency and profits)? Has productivity of the SOEs improved? And, if so, is it related to improved HR practices and human capital? Have SOEs become more market oriented or competitive following successive waves of reforms, and with appointment of a younger and more highly educated cohort of managers? How is the emerging digital economy and digitalization in management impacting SOEs? Are they in a better or worse position to exploit digital opportunities than their private counterparts? What are the skill and career implications for the SOE workforce? At an individual level, what impact have radical reforms of SOEs had on the remaining workforce? Has the ‘state-owned’ mentality changed as a result of the reforms? Have they become more market oriented, and if so, what impact does this have on their psychological contract? Do SOE employees now share similar perceptions with their counterparts in the private sector with regard to why they work and how they experience their working lives?

Limitations

There are several limitations in this review article. First, the key words we used may not have captured all relevant articles, although we did go through the references in the sample to identify articles not captured in our initial search. Second, our university library database may not subscribe to all the English academic journals related to the field. Nevertheless, we are confident that our sample incorporated the majority of studies of HRM in SOEs. Third, our review did not include articles published in Chinese academic journals. This means that our review has not included findings or perspectives of the majority of indigenous scholars on the topic. Fourth, the thematic categories we identified are indicative, in that some studies address multiple themes. We grouped themes into those we believe to be the most appropriate categories. Fifth, we were not able to analyze in depth all the theories or themes in the sample, due to space constraints. This limitation is excusable, as the purpose of this review was to provide a general picture of the transformation of research on HRM in Chinese SOEs, rather than projecting a precise picture in detail.

Conclusion

Drawing on 178 articles from 43 English journals over a period of 25 years (1993–2017), this article systematically reviewed the present state of research on HRM in Chinese SOEs. We revealed that SOEs have developed a hybrid model: a fusion of some market-driven elements of western HR practices with the vestiges of the socialist-oriented ‘iron rice bowl’ system. In addition, we found that theoretical paradigms, methodological choices and levels of analysis have shifted in the last three decades as researchers’ preferences have changed, and as Chinese SOEs have reformed and transformed and different issues of priority have surfaced. We have argued that despite the sharp decline of employment in the state sector, research on HRM in SOEs in China has continuing significance, not least because of the continuing size and functions of the sector, but also because of the relevance to, and implications for, other, slower developing countries that are undergoing similar economic transformations and public sector restructuring processes. Finally, we maintain that studying HRM in Chinese SOEs can contribute to the general HRM research by extending the scope of research subjects and highlighting novel HRM models (e.g. the hybrid HRM model).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

Additional information

Funding

We gratefully acknowledge funding support from the National Natural Science Foundation of China (grant No: NSFC 71832003).

Notes

1 Many of the articles, especially those quantitative studies, only touched upon SOEs without really studying them.

2 A full reference list of the 178 papers utilized in the data analysis can be provided by the authors on request.