In 1979, the United Nations (1979) proclaimed, “…States parties shall ensure to women appropriate services in connection with pregnancy, confinement and the post-natal period….” In most OECD countries, “maternity leave,” “paternity leave,” and “parental leave” are concepts that encompass economic compensation and the provision of healthcare services to women in the prenatal, natal and postnatal periods.¹ In China, the most popular semantic reference is “maternity insurance,” which was initially introduced by the Chinese government in 1951 (Zhuang, 2009).

China is one of only a few countries with a specific insurance scheme to cover the social risks resulting from childbirth.² Yet few studies have explored maternity insurance in China systematically, even though the number of women in the country reached 667 million in 2014 (National Bureau of Statistics, 2015). In international academic journals, few articles explicitly focus on maternity insurance, and some only mention maternity leave under the umbrella of overarching topics without elaborating on the subject in detail (Chiu, Luk, & Tang, 2002; Granrose, 2007; Darimont, 2009). In Chinese literature, several studies have addressed this topic but from a more or less descriptive standpoint; few have elaborated it from a theoretical and analytical angle. Therefore, maternity insurance in China represents a “forgotten corner” compared to other forms of insurance such as old-age insurance and health insurance (Hong, 2003).³ Whereas the issue of educational and occupational inequality of both genders fuels debates, attracting both academic and public interest, the issue of gender-related social protection remains under- researched.

Against this backdrop, this article seeks to fill this research gap by examining the conceptual design and local implementation of Chinese maternity insurance. Unlike conventional policy- and problem-oriented studies within Chinese academia, this article attempts to trace the inherent trend and development of maternity insurance in China from an analytical perspective. Based on a gender perspective and the concept of “social globalization” (Deacon, 1997, 2005), we explore maternity insurance from a global, national, and local perspective, and seek to answer two corresponding research questions. First, we look at maternity insurance from the angle of the globalization and internationalization of social policy, asking to what extent Chinese legal systems regarding maternity insurance converge with global treaties and conventions operating at an independent level.⁴ The convergence to or divergence from a global framework of maternity insurance will be explored by comparing Chinese legal documents with international treaties introduced by the UN and the International Labour Organization (ILO). Next, we examine local implementation; accordingly, the second research question concerns how far and to what extent the sub-national practices of maternity insurance are substantially enforced or are disconnected from global standards and national laws, with a focus on the discrepancy between the law on the books and the law in action.

Based on in-depth documentary analysis of international treaties and Chinese legal documents, we find that while maternity insurance has been set up by the Chinese government, this pioneering practice exhibits two main limitations. First, it lags behind in terms of meeting international standards. For example, parental leave and paternity leave are not included in the Chinese legal system, and more importantly, only urban employees are covered by maternity insurance, excluding the majority of the female population, including non-working urban women, informal female workers (e.g. migrant workers in cities), and rural women. Moreover, in its implementation, maternity insurance falls short of substantially benefitting local targeted groups. For instance, it is common to find that employees of formal sectors cannot receive the benefits of maternity insurance, mainly because non-compliance with gender-related maternity laws is widespread across China. Following this introduction, Section 1 introduces the analytical framework employed in this research. Section 2 explains the international treaties on maternity protection introduced by the UN and the ILO. Section 3 details the Chinese legal documents regarding maternity insurance and compares the Chinese regulations with international standards. Section 4 examines local implementation in China. The conclusion discusses the findings and their implications.

With regard to “maternity protection,” the United Nations (UN) and the International Labour Organization (ILO) are the two leading international organizations. The primary institutional measures that these global actors apply in order to influence nation states or member states, are termed “soft” methods. International “soft power” is significantly different from “hard power” since the former is mainly based on international treaties, conventions, and declarations with a symbolic implication for nation states, while hard power is connected with direct intervention in national social reforms through direct financing by international organizations such as the World Bank (Liu, 2015). Member states of international organizations (IOs) are subject to the concrete international obligation of fulfilling the proclaimed targets if they have signed and ratified these conventions. However, even after the ratification of international treaties, the UN and ILO treaty bodies lack the “hard” or “coercive” means to affect and steer national social policy directly if members fail to comply (Leisering, 2005; Leisering & Liu, 2010).

Nevertheless, the importance of international treaties should not be underestimated even if efficient and coercive methods of enforcement are lacking. In fact, soft power might have greater clout than hard power if these ideas are internalized by actors as socially accepted rules and values (Koh, 1997), if these ideas are globally disseminated and enjoy high “legitimacy” among actors (Risse & Sikkink, 1999) or if to openly deny and violate these global treaties will generate “international isolation” and a “naming and shaming” effect (Meernik, Aloisi, Sowell, & Nichols, 2012).

The international framework regarding maternity protection includes several core conventions introduced by the UN and the ILO (see Table 1). The UN General Assembly introduced the “International Covenant on Economic, Social and Cultural Rights” (ICESCR) in 1966 and the “Convention on the Elimination of all Forms of Discrimination Against Women” (CEDAW) in 1979. The ILO introduced the “Maternity Protection Convention, 1919 (No. 3)” (C003), the “Social Security (Minimum Standards) Convention, 1952 (No. 102)” (C102), the “Maternity Protection Convention (Revised), 1952 (No. 103)” (C103), the “Maternity Protection Recommendation, 1952 (No. 95)” (R95), the “Maternity Protection Convention, 2000 (No. 183)” (C183) and the “Maternity Protection Recommendation, 2000 (No. 191)” (R191).

The UN treaties have set up the principal standards via normative proclamation, which are mirrored in the semantic rhetoric of human rights, women’s rights, and gender equality. In this vein, maternity protection has been proclaimed a basic citizenship right for all nation states, which has provided global validity. For instance, Article 9 of the ICESCR mandates a state’s obligation to ensure the rights of everyone to social insurance without detailing each policy field (United Nations, 1966). Additionally, in the General Comment to Article 9 introduced by the Economic and Social Council of the United Nations (2007), “maternity protection” has been explicitly addressed:

Art. 12(g) Maternity: 19. Article 10 of the Covenant expressly provides that ‘working mothers should be accorded paid leave or leave with adequate social security benefits’. 15 Paid maternity leave should be granted to all women, including those involved in atypical work, and benefits should be provided for an adequate period. 16 Appropriate medical benefits should be provided for women and children, including perinatal, childbirth and postnatal care and care in hospital where necessary.

Compared to the principles declared and proclaimed by the UN treaties, the ILO plays a more dominant role in the process of standardization and formalization of maternity protection worldwide through introducing, revising, and adjusting the conventions. Through an examination of the long-term development of ILO conventions on maternity protection, we not only explore its historical consistency but also chart the evolution of standards concerning benefits, services, eligibility requirements and target groups. Through analyzing the core conventions and recommendations introduced by the ILO, we find that from the first convention on maternity protection in 1919 to the recent convention in 2000, the main catalogues and principles remain consistent. All the conventions and recommendations state that maternity protection should provide a wide range of benefits and services – including cash benefits, medical services, and health care – to women in the perinatal, confinement, and postnatal period, and should also offer protection to pregnant women from employment dismissal. These conventions and recommendations propose that maternity protection should be funded through public funds or compulsory social insurance funds, which are financed by taxes or by the contributions of employers. In some countries, such as Sweden, the financial resources for maternity protection rest on a mixed-finance model, including both contributions and tax revenues (Carlson, 2013). Nevertheless, the ILO standards have been periodically adjusted and improved with regard to the target groups, leave period, benefit levels, and the integration of paternity leave.

Firstly, as for the target groups of the maternity protection, in the ILO convention (1919), it targeted employees in public, industrial, and commercial undertakings (Art. 3 of C003). In 1952, this “narrow” coverage was expanded by ILO (1952a) to include women employed in “non-industrial” and “agricultural occupations, including women wage earners working at home” (Art. 1.1 and 1.3 of C103). The target group of agricultural occupations has been specifically defined as “all occupations carried on in agricultural undertakings, including plantations and large-scale industrialized agricultural undertakings” (Art. 1.4 of C103).

Secondly, as for the leave period, it has been prolonged over time. In 1919, the leave period was about 6 weeks after confinement (Art. 3a of C003),⁵ which has been extended to “at least twelve weeks” in 1952 (Art. 3.2 of C103). In the same year, the ILO (1952b) recommendation (R95) suggested, “[W]here necessary to the health of the woman and wherever practicable, the maternity leave provided for in Article 3, paragraph 2, of the Maternity Protection Convention (Revised), 1952, should be extended to a total period of 14 weeks” (Art.1.1 of R95). In 2000, the leave period was extended by ILO (2000a) to “not less than 14 weeks” (Art. 4.1 of C183), and the ILO (2000b) recommendation (R191) suggested that signatories shall endeavor to extend this period to at least18 weeks (Art. 1.1 of R191).

Thirdly, as for the level of cash benefits for employed women, in 1952, the convention 103 read, “[W]here cash benefits provided under compulsory social insurance are based on previous earnings, they shall be at a rate of not less than two-thirds of the woman’s previous earnings taken into account for the purpose of computing benefits” (Art. 4.6 of C103). However, in the same year, the recommendation was revised to suggest that these cash benefits should be set at a higher rate, in the amount of 100 percent of a woman’s previous earnings (Art. 2 of R95).

Lastly, the most significant transition of the ILO conventions regarding maternity protection lies in the widening of the scope of protection from solely maternity leave to parental leave comprising both maternity and paternity leave. The claimants of maternity protection also include employed fathers (ILO, 2000b), not only in some exceptional cases where fathers “should be entitled to take leave of a duration equal to the unexpired portion of the postnatal maternity leave” (Art. 10.1 of R191), but also in more conventional cases, where the fathers can claim a leave when the mother cannot look after the child (Art. 10.2 of R191). The granting of paternity leave fulfills the social function of significantly unburdening women as sole caregivers.

The trajectory of international maternity protection has undergone constant transition, shaped by the extension of its degree of coverage, the prolongation of the leave period, the rise in benefits and the integration of parental and paternity leave into the overall maternity protection scheme. The revisions of international conventions have been aimed at correcting the disadvantaged position of women resulting from giving birth, changing the uneven gender power structure, and empowering individual women in a monetized market economy. The fundamental change in core beliefs relates to the recognition of paternity leave, which can be seen as a realistic response to a post-industrialized society shaped by semantics over gender equality. The basic assumption of paternity leave is that the model of the male breadwinner in industrialized societies has been dispelled; women can also earn a high income and act as the main breadwinner, while husbands can also take on the task of child rearing. This mutual exchange of traditional gender roles signifies the progressive development of the international maternity protection regime and the rising status of women in contemporary global society.

In ancient times, feudal China was basically a patriarchal society that reflected the dominant role of men and the suppression of women (Hamilton, 1990). Under this male-hegemonic rule, men were located at the top of the pyramid of political power, while the women remained largely powerless and subordinate (ibid). In thousands of years of feudal history, there has never been an institutional welfare policy protecting women’s fundamental rights and improving their wellbeing. The establishment of maternity insurance is a product of modern Chinese history. After the founding of the People’s Republic of China (PRC), the State Council enacted a Labor Insurance Regulation (LIR) in 1951, which was seen as an embryonic form of institutional and state-organized social protection for women in modern China. As a member of the socialist block, China emulated the Soviet model of labor protection and included comprehensive social protection arrangements such as old age pensions, work safety protection, health insurance, and maternity insurance in the 1951 regulation (Dixon, 1981). A labor insurance model rooted in the prevalence of state-owned enterprises emerged, providing social benefits to employees directly (ibid). According to this regulation, female employees were entitled to claim 56 leave days after birth, during which continued remuneration would be ensured, and medical treatments would be reimbursed by the state-owned enterprise. The LIR was the first institutional maternity program in modern Chinese history. The only notable limitation was that this scheme was principally enterprise- based labor insurance, not a social insurance arrangement financed by a social-pooling fund.

Since China joined the UN in October 1971, the behavior of the Chinese state has been increasingly internationalized and socialized in the global community (Johnston, 2007). Shifting from a country isolated from the international organizational framework to a full member state integrated into the international community, China has been participating in multiple international governmental and non-governmental organizations and has signed international treaties and conventions (Kent, 2002). Regarding the protection of women’s rights, the Chinese government signed the “Convention on the Elimination of all Forms of Discrimination Against Women” in 1980. In 1997, the Chinese government signed the “International Covenant on Economic, Social and Cultural Rights,” which was approved by the People’s Congress in 2001. Through the signing of these core international documents regarding social rights and women’s rights, China has committed itself to fulfill its international obligation to ensure maternity protection.

Within the Chinese legal system there are several legal documents that address the issue of maternity insurance in China, including a series of laws, decrees, regulations, and measures – for example, the “Law on the Protection of Rights and Interests of Women” (LPRIW), the “Law on Maternal and Infant Health” (LMIH), the “Law on Population and Family Planning” (LPFP), the “Regulation on the Placement of Surplus Staff and Workers of State-owned Enterprises” (RPSSWSOE), the “Notice on Several Matters concerning Maternity Benefits of Female Employees” (NSMMBFE), the “Special Regulation on Labour Protection of Female Employees” (SRLPFE) and the “Trial Measures on Maternity Insurance for Enterprise Staff and Workers” (TMMIESW).

The Chinese legal system, like the international gender regime, emphasizes the protection of women’s rights and pursues the goal of gender equality. For instance, Art. 48 of the Chinese Constitution (2004) stipulates, “Women in the People’s Republic of China enjoy equal rights with men in all spheres of life, in political, economic, cultural, social and family life.” The “Law of the People’s Republic of China on Safeguard of Women’s Rights and Interests” (LSWRL) was introduced in 1992, and it is regarded a milestone in maternity protection in China. Art. 28 of the LSWRL dictates that the state shall develop social insurance, social assistance, social welfare, and medical services to protect women’s social entitlements, and Art. 29 explicitly states that the state shall promote a maternity insurance system and shall develop further social protection systems regarding childbirth, and local governments and relevant local authorities shall deliver maternity assistance for impoverished women. In 1995, the “Labour Law of the People’s Republic of China” (LL) was introduced and Art. 61 to 63 address the rights of employed women to maternity protection, maternity insurance and health services pertinent to childbirth. In 2008, the “Labour Contract Law of the People’s Republic of China” (LCL) was put into practice, Art. 6 of which notes the entitlements of employees to social insurance and maternity insurance. Recently, the “Social Insurance Law of the People’s Republic of China” (SIL) was promulgated in 2010, which establishes a legal foundation for all five branches of social insurance in China, including maternity insurance, old-age insurance, health insurance, unemployment insurance and work-injury insurance. Its Chapter VI has set up a legal framework for the entitlements of employed women to maternity cash benefits and maternity medical expenses (Art. 54, 55) and it regulates the employer’s obligation to pay contributions for maternity insurance (Art. 53).

Until now, China has behaved in a passive rather than an active manner considering ILO treaties on maternity protection; it has not signed any of them. However, this passive behavior towards ILO conventions does not reduce the importance of ILO treaties to China. On the contrary, ILO treaties have a significant impact on providing a reference framework and a global standard for the development of maternity insurance in China, as ILO treaties are tentatively studied, cited, and disseminated by Chinese scholars and policy makers (Hong, 2003; Tang, 2009; Jiang & Wang, 2015). ILO treaties play a preeminent role in appeals for reforms of and improvements to maternity protection, and in legitimizing domestic discourses and debates on maternity insurance (Shao & Xie, 2004). In order to have a clear understanding of the similarity and differences between Chinese regulations and ILO treaties, we compare them along five dimensions: target groups, leave period, benefit levels, the inclusion of paternity leave, and finance source (see Table 2).

First, with regard to the target groups for maternity insurance, these vary from one legal document to another. The Labour Law stipulates that the employees who have set up a labor relationship with employers be covered by the maternity insurance system (Art. 2). According to the TMMIESW, the target group is employees in urban enterprises (Art. 2). The SRLPFE, on the other hand, stipulates that employees in enterprises, public institutions, state-owned institutions, and government agencies should be covered by maternity insurance (Art. 2). Therefore, the SRLPFE has wider coverage than the other two legal documents. Despite this slight difference among them, all three legal documents target “employees” in urban regions. The target groups in the Chinese legal documents are different from multiple ILO conventions on maternity protection, which have extended the target groups to include non-industrial and agricultural sectors, including the rural population (e.g. Art. 1.1, 1.3, 1.4 of C103). According to the Chinese legal documents, the agricultural population and employees of informal sectors are totally excluded from the maternity insurance system. Instead of making a legal claim, they normally turn to social networks for assistance (Sun, 2012).

Second, the period of maternity leave in China was initially 90 days and was extended to 98 days in 2012 according to the most recent maternity document (i.e. Art. 7 of the SRLPFE). It complies with the requirements of the ILO treaties R95 and C183 fully (14 weeks and “not less than 14 weeks”) yet it is slightly lower than R191 (18 weeks).

Third, as for the benefit level of maternity insurance, the SRLPFE and the TMMIESW, the two core Chinese legal documents, contain indicators that differ from those used in ILO treaties. In China, maternity allowances shall be paid on the basis of the average earnings of all staff members of the enterprises in the previous year. The ILO treaties specify that the calculation of maternity cash benefits should be based on the previous earnings of individual workers instead of the average earnings of all staff. Another difference is that the Chinese maternity insurance grants cash benefits for maternity leave with one hundred percent of previous wage earnings to staff in enterprises, while the ILO treaties only demand the contracting party to grant two thirds to one hundred percent of previous wage earnings of the employee (Art. 4.6 of C103; Art. 2 of R95).⁶

Fourth, Chinese legal documents still largely lag behind ILO treaties with regard to paternity and parental leave. So far, Chinese legal documents have not taken parental leave into consideration, which means the beneficiaries covered by maternity insurance are only women, and fathers are not entitled to claim a leave period.

Lastly, it worth noting that all core legal documents in China mandate that the cash benefits, medical services, and health care in the maternity insurance system be financed through a special “maternity insurance fund,” which shall be organized via a social insurance program (Art. 8 of SRLPFE; Art. 29 of LPRIW; Cha. 6 of SIL). This conforms to the regulations of ILO treaties stipulating that maternity protection shall be organized through either public funds or compulsory social insurance system(s) (Art. 4.4 of C103; Art. 6.8 of C183).

In sum, the Chinese maternity regime has adjusted itself to international norms and standards to a certain degree (e.g. regarding the leave period and benefit level). However, due to its limited definition and scope, it remains a “social insurance” scheme that limits the target group to only urban employees. In contrast, the international maternity regime has expanded its beneficiaries to include the entire population and has evolved into universal “maternity protection.” The most significant difference between China and the international community lies in the non-recognition of parental leave by the Chinese maternity regime, which still adheres to a more conventional view, assuming that only women are entitled to maternity leave, not their husbands. The basis of this ideational foundation is still linked to traditional sex roles and a male breadwinner model, which is consistent neither with the reality of contemporary Western societies nor with that of contemporary Chinese society.

Institutional disconnection refers to the gap between paper law and factual development. It suggests that at times it is easier to proclaim global values than actually put them into practice, since the enforcement of paper laws is dependent upon many prerequisites such as the availability of institutional capacity, political-administrative infrastructure, financial resources and the political will to implement a law by national and sub-national actors (Leisering, 2005). We explore the disconnection of maternity insurance in China from both international standards and national regulations. Firstly, the large number of excluded women and the absence of parental leave are the main features of the disconnection from ILO treaties. Secondly, poor implementation and undesirable outcomes reveal disconnection from national regulations.

Compared to ILO treaties, maternity insurance in China exhibits social exclusiveness. In 2014, less than 15 percent of the total population in China or less than 25 percent of the urban population was covered by the maternity insurance.⁷ To date, China has applied the “limited” definition of maternity insurance introduced during the early industrial era, which only covers the industrial population and urban employees in social and public institutions, while neglecting the enormous demands for maternity protection by other groups, such as female workers in informal sectors (i.e. migrant workers), unemployed women, rural women and the female workers who engage in employment activities with marginal and low-wage income and thus have a discontinuous employment biography (Shao & Xie, 2004; Tang, 2009). According to the legal exegesis and explication of the series of laws on maternity insurance provided by the China Law Counsel Center (2014), unmarried women are in fact entitled to claim maternity leave. However, any person who breaks the “Law on Population Planning” will lose legal entitlements to social insurance and maternity leave. Therefore, the maternity insurance system in China features an enormous coverage gap and fosters social exclusiveness. Additionally, a significant difference between the national regulation on maternity protection and ILO treaties is the fact that Chinese legal documents fail to include parental leave and paternity leave, which constrains the choices of female employees after childbirth. The paternal obligation and contribution to child-care is ignored by the law, even though fathers play an increasingly important role in childcare.

The complexity of maternity insurance is shaped not only by the multiplicity of relevant legal documents at the national level, but also by legislative federalism, as nearly each province and autonomous region in China has additionally enacted local regulations on maternity insurance. For example, in 1997, Fujian province was first to introduce a “Regulation of Maternity Insurance for Staff in Enterprises in Fujian Province” (following TMMIESW). Since then, more than 30 local regulations have been enacted at the sub-national level.

Tang (2009) categorizes local maternity insurance coverage into three types – narrow, medium and broad. The narrow type covers mainly employees in state-owned enterprises; it has been put into practice in Shandong, Fujian, Sichuan, Inner Mongolia, Beijing, Jiangxi, Yunnan, etc. The medium type covers employees in state-owned enterprises and private enterprises, which exists in Zhejiang, Chongqing, Jiangsu, Shaanxi, etc. The broad coverage extends to employees in state-owned enterprises, private enterprises and public institutions; it is found in Henan, Guangdong, Shanghai, Hunan, Jilin, Ningxia, etc.

When examining the disconnection of maternity insurance regulations from actual practices, we find that the outcome is undesirable compared to other branches of social insurance. According to the annual report on the development of social insurance in China (2014) released by the Ministry of Human Resources and Social Security (MOHRSS, 2015), 170.39 million urban employees, accounting for 40.5 percent of total urban employees, were covered by maternity insurance in 2014. The number of insured employees covered by unemployment insurance, work-injury insurance, health insurance, and old-age insurance are respectively 170.43 million, 206.39 million, 282.96 million, and 341.24 million (see Figure 1). Compared to the other four branches of social insurance in China, maternity insurance has the lowest number of participants and coverage rate. The coverage rate of old-age insurance for urban employees is almost double that of maternity insurance. In 2014, around 6.13 million urban employees made use of maternity benefits under the maternity insurance program, making up a mere 0.2 percent of the total urban population.

Figure 1. 
The Number of Urban Employees Participating in Five Branches of Social Insurance

Source: authors’ compilation based on the annual report of social insurance in China (MOHRSS, 2015)

The low coverage and take-up rates in the maternity insurance arrangement are mainly due to individual enterprises’ unlawful behavior. For example, many employers do not pay their maternity insurance contributions to local maternity insurance funds. In practice, different types of enterprises have different participation rates. Local government officials, at the level of cities and counties, usually adhere to a model of economic growth and they lack the fiscal incentives to fully implement social insurance arrangements such as maternity insurance. Generally speaking, most state-owned enterprises pay the contributions for their staff members’ maternity insurance as required. However, among private enterprises, joint ventures and foreign enterprises, the participation rate for maternity insurance varies; many private companies try to minimize their legal obligation to protect their workers from social risks (Pan, 2003; Feng, 2011). A large number of companies with private ownership commonly ignore legally binding regulations in maternity laws and neglect to protect female workers against dismissal due to pregnancy and childbirth. Studies reveal that many female employees are dismissed as soon as their employers are informed of their pregnancy, while some are dismissed promptly after childbirth (Li, 2004; Cheng, 2005).

Furthermore, the extremely low take-up rate of maternity insurance is also linked to structural factors and employment status in the labor market. Despite the achievements of promoting women’s rights in China, scholars have uncovered a countertrend impeding further gender equality. One study shows that the number of female employees in the formal sector, particularly in state-owned enterprises, dropped by 14.77 million between 1995 and 2000 (Jin, 2006). In contrast, the number of female laborers working in the informal sector increased (Jiang, 2003; Gu, 2006; Chen, 2015). In all five branches of social insurance, the coverage rate for female employees lags behind that of male employees (National Bureau of Statistics of China, 2004, 2012). This development is closely linked to the uneven distribution of job opportunities and positions between men and women, since informal sector employees are largely excluded from access to social insurance benefits. Thus, the “informalization” of female labor can partially explain the low coverage rates of the maternity insurance scheme.

As maternity protection is an unheeded issue in both gender studies and social policy studies, this article has sought to shed light on maternity protection by taking China as a case study. With the largest female population in the world, China is one of a few countries that have set up a specific maternity insurance policy. Additionally, several Chinese laws and other legal documents have explicitly addressed the issue of maternity insurance, indicating the political will of Chinese government when it comes to women’s rights. However, through in-depth documentary analysis, we find that, firstly, although Chinese laws and regulations have applied the overall framework of UN and ILO treaties and conventions on maternity protection, the former is not fully convergent with the global standards set by the ILO with regard to some core technical aspects. Secondly, the local practice of maternity insurance in China reflects the process of disconnection from both international standards and domestic laws, exemplified by the low participation rate due to poor implementation.

This study has examined and compared international treaties and Chinese national legal documents on maternity insurance. On the one hand, since the reform and opening-up policy in 1979, the Chinese government has introduced several laws and regulations on social rights and women’s rights in line with core international conventions introduced by the UN human rights bodies, which make it seem as if China has responded to the call for maternity protection proposed by the UN. However, on the other hand, we find that China has not signed any ILO treaties regarding maternity protection, which indicates China behaves in a more passive and restrained way towards fully applying the ILO treaties and conventions detailing specific regulations and global standards on maternity protection. The inconsistency between this embrace of the guideline-oriented international treaties (such as the UN’s CEDAW and ICESCR) and concrete international obligations (in the ILO treaties) reveals the ambivalent attitudes of the Chinese maternity regime towards the international maternity regime.

Through a comparison of ILO treaties and national laws in China on maternity protection, we find that as for the leave period, benefit levels, and finance source, the Chinese regulations on maternity protection are close to international standards. There are two major differences between the Chinese maternity insurance scheme and international maternity treaties. The first is that parental leave and paternity leave are not included in the Chinese legal system, which stipulates maternity leave can only be claimed by mothers, not fathers. Adherence to a sole “maternity” leave makes the maternity insurance system in China different from parental leave schemes in most OECD countries. In this aspect, China adheres to an outdated welfare model that hinges on fixed sex roles and gender division. Clearly, China is not keeping pace with current developments and the new semantics of gender equality in post-industrialized societies, as, for example, Scandinavian nations do.

The second is the fact that Chinese maternity insurance targets merely urban employees, while ILO-standards have extended the insured groups to include the agricultural population and employees in the informal sector. In other words, in China, the non-working urban female population, informal female workers (e.g. migrant workers in cities), and the rural female population are excluded from the maternity insurance system even though they make up the majority of the female population. This social exclusiveness undermines the legality and legitimacy of this system, and contradicts the principle of social equality and universalism.

Although China has set up an independent maternity insurance scheme, the implementation of maternity insurance in China is rather poor. Compared to the four other social insurance branches (i.e. old-age insurance, health insurance, unemployment insurance, and work-injury insurance), maternity insurance has the lowest coverage ratio. Taking the example of old-age protection, the Chinese government has recently extended the conventional social insurance system to include the whole population by introducing a universal and non-contributory social pension (Liu & Sun, 2015). In 2014, less than 15 percent of the national population was covered by maternity insurance. Even for those who qualify for insurance, it is common to find that they cannot receive benefits, mostly because of employers’ non-compliance, which is widespread across China (Sun & Liu, 2014, 2015). In local practice, many employers minimize their legal obligation and do not pay maternity insurance contributions for their employees. Owing to the lack of penalties and enforcement, non-compliance with these laws rarely exposes employers to legal punishment, which in turn encourages employers to act arbitrarily at the sub-national level, making female workers’ rights more or less a matter of luck.

Through the case of maternity protection, we argue that unlike the common perception that women in China have been to a certain degree emancipated in the Sino-communist revolution that intended to promote women’s liberation, China remains a backward gender-related welfare regime with a rudimentary maternity insurance system characterized by enormous gaps. By using major indicators of modern social insurance arrangements such as the coverage ratio and take-up of benefits, we have come to the conclusion that the maternity insurance scheme in China targets only the small part of the female population that is primarily confined to the formal sector in the urban regions. The global trend of the inclusion of paternity leave into maternity protection is still a marginalized issue, and has been neglected in China. In fact, China is still far from a gender-equal society since proportionally more women are employed in informal sectors than male workers and are thus exposed to high social risks and excluded from receiving social insurance benefits. Therefore, although some improvements have been achieved, China remains a patriarchal society shaped by the highly uneven distribution of social resources between men and women (e.g. opportunities, jobs, and positions, as well as uneven access and entitlements to social benefits). The social welfare culture in Chinese society is still conservative and hinges on a socially constructed “naturalness” concerning sex roles and gender division.