Contracting Welfare Services to Social Organizations in China: Multiple Logics

Abstract

As part of a broader direction of welfare and governance reforms, China has launched a policy to contract welfare services out to social organizations. Scholars have explored the implementation of the policy in a few socioeconomically advanced cities such as Guangzhou, Shenzhen and Shanghai. In this article, we examine how local governments lacking nongovernmental services suppliers or resources for contracting respond to the policy. We developed a framework of multiple logics to analyse services contracting in a county-level city in eastern China. We found that local officials follow three logics in implementing the policy: to meet the central state's targets, to balance policy outcomes and risks, and to stimulate a more participatory society. This generates a mix of policy behaviour, including entrepreneurialism, welfarism, innovation, risk-sharing and collaboration. We thus argue the interplay of the logics determines the local policy process of services contracting in China.

摘要

作为其福利改革和治理转向的一部分，中国政府制定并实施了购买社会组织服务的政策。学者们探究了该政策在社会经济发达或政策试点地区——如广州、深圳和上海——的实施。本文将目光投注于非政府服务力量发展较晚、政府购买资源相对不足的地区，探究这些地方的基层官员对自上而下的政府购买政策的应对。研究发现基层官员的政策行为体现了三种逻辑，分别为实现中央目标、平衡政策成果与风险、以及催生更具参与性的社会；据此产生的是一系列具有企业家性质、福利主义、创新、风险共担、合作性的政策行为的混合。文章因此提出国家、官员制度和社会逻辑的互动形塑了政府购买社会组织服务的基层实践。

In 2013, the Chinese central government initiated a policy to contract welfare services out to social forces.¹ This policy is part of China's governance reform, which introduces procurement (zhengfu caigou 政府采购) to “improve the efficacy of the use of governmental funds” and “promote the building of a clean government.”² It is also proposed as a strategy of welfare reform that moves towards a marketized and pluralized provision system.³ Despite years of experimentation in pilot cities like Guangzhou, Shenzhen and Shanghai, services contracting is a top-down decision for many localities. Local governments are expected to implement the policy under the general ideologies and principles defined by the central state and, at the same time, according to their own conditions. Contracting has then been extended across China with models varying in their financial arrangements, services areas and the degree of competition and transparency.⁴

A growing number of studies has examined the implementation of services contracting in China, with most attention being paid to the few advanced and/or pilot cities, namely Guangzhou, Shenzhen and Shanghai.⁵ Research has shown that several conditions are essential to services contracting, including a sufficient number of social organizations (SOs) to bid for contracts and officials’ capacity to design, manage and assess the delivery of services.⁶ Scholars have also demonstrated the advantages of pilot cities in implementing the policy. For example, Shenzhen was found to benefit from its position as a special economic zone and an experimental site for governance reforms when designing a contracting system and connecting with social work experts in Hong Kong to deliver services.⁷ In Guangzhou, universities and academics have significantly contributed to the development of SOs and contracting.⁸ Some argue that Guangdong province in general has secured “flexible arrangements for the implementation of central policies as well as greater discretionary control over local investment and financial planning.”⁹ But how do local governments lacking these conditions, such as those with a limited supply of nongovernmental providers, respond to the policy?

We address this question by looking at services contracting in a medium-sized, county-level city in China, where officials face the top-down mandate to outsource welfare services but, at the same time, lack nongovernmental providers or the resources needed for contracting such as access to experts. In the Chinese bureaucracy, county-level jurisdictions are a crucial sphere for policy implementation because they represent the meeting point of top-down policies and the expression of bottom-up opinions.¹⁰ The article engages with the theories on the Chinese policy process, which mainly stem from the economic area. The increasing relevance of welfare as a policy domain begs research to uncover the particularities (if any) of its policy dynamics. Our study found that county-level officials respond to the top-down welfare policy with three logics. We take “logics” to mean institutionally constructed rules and principles that bureaucrats elaborate on in policy implementation and which generate certain patterns of behaviour.¹¹ First, officials at the county-level city have an overriding priority to meet the central state's targets, which include the reform of the public sector, welfare reform and social governance reform. This shapes officials’ policy implementation and can lead to entrepreneurial and welfarist behaviour. Entrepreneurial behaviour is exhibited when officials take risks and invest resources in developing SOs, or even act as SOs themselves. Welfarist behaviour refers to practices that expand welfare provision and regulate SOs to focus on services. Second, as part of the bureaucratic system, officials follow the logic to balance policy outcomes and risks. This can generate innovative behaviour, as officials adjust the top-down instruction to contract services to suit local conditions, or lead to risk-aversion behaviour, as officials adopt strategies to share the risks of implementing welfare policies. Finally, officials elaborate on the principle of stimulating a more participatory society, which leads to a set of practices inviting non-state professional forces to participate in the making and implementation of the policy. Officials juggle these logics, combine them in different ways, and swing between different patterns of policy behaviour according to the risks involved. We thus argue that the interplay of the multiple logics determines the implementation of services contracting in China.

The article begins with the introduction of the policy of services contracting in China. This section sets the scene and raises the often-overlooked question of the implementation of the policy in non-pilot, less socioeconomically developed regions. We then briefly review the key theories of the Chinese policy process, identifying gaps before proposing an analytic framework of multiple logics. We go on to present our case city and the specific conditions it faced when implementing the contracting policy. The fourth section compiles our empirical findings before analysing the officials’ behaviour and explaining their logics. In the conclusion, we pull the findings together and discuss their implications for policy implementation and services contracting in China.

This article draws on a research project examining services contracting to SOs in China. As discussed in later sections, we chose City A to exemplify less advanced, non-pilot regions where officials face constraints in implementing the top-down contracting policy.¹² The research involved an extensive review of Chinese and English literature and an analysis of relevant policies, laws and regulations. The data presented in this article comes from interviews, conducted between 2018 and 2019, with seven government officials from the civil affairs, finance and police bureaus, and with 22 SOs varying in size, services areas, registration status and degree of dependency on governments (see Appendix I). The interviewed SOs were direct-participant agents of local contracting and were sampled from officially published lists of contractors or through snowballing.

The Policy of Services Contracting in China

This section introduces China's policy of contracting out welfare services, including its multiple rationales and the top-down nature in most localities. Since the 1990s, nongovernmental forces have been providing public services in a few Chinese cities like Shanghai, Guangzhou and Shenzhen. Services contracting has expanded from these experimental practices and been legitimized as a national policy. At the central level, it has been proposed in three sets of policy agendas.

First, contracting forms part of the public sector reform. In 2002, China promulgated the Government Procurement Law, which proposed procurement to “improve the efficacy of using governmental funds” and “promote the development of a clean government.”¹³ This suggested a new public management approach as it introduced the rhetoric of competition, efficiency and performance measurement in governance.¹⁴ The reform also downsized public institutions (shiye danwei 事业单位) such as schools and hospitals, which used to be the primary providers of services. In 2011, China published a policy which required public institutions with non-basic welfare functions to be independent of government and to marketize.¹⁵ The 18th Central Committee of the Chinese Communist Party (CCP) declared its intention of “greatly reducing the government's role in the direct allocation of resources and allocating resources according to market rules, market prices and market competition.”¹⁶

The second set of policy reforms concerns welfare. While China's economic reform has produced massive achievements, it also has intensified the commodification of labour, reduced the obligation of state-owned enterprises to offer services and brought challenges to China's socialist welfare regime.¹⁷ To balance economic and social development and increase social harmony, the Hu–Wen administration (2003–2013) introduced a variety of policy programmes such as the abolition of rural taxes and the return of the rural cooperative medical system. In 2011, the 12th Five-Year Plan on Economic and Social Development announced the goal to innovate public goods provision, including contracting.¹⁸ This was further explained in the 12th Five-Year Plan on the Development of Public Service, which aimed to build “a multilevel, channelization of public services delivery system.”¹⁹

The third aspect is social governance reform. In the past few decades, China has transitioned from “social management” (shehui guanli 社会管理) to “social governance” (shehui zhili 社会治理), a move which has required a new framework through which to develop and also regulate societal forces.²⁰ On the one hand, policies have been announced to activate society and support the development of welfare organizations, social workers and volunteers.²¹ In 2013, the Third Plenary Session of the CCP's 18th Central Committee put forward the proposal to “kindle the vigour of social organizations.” This included easing registration requirements for certain types of SOs and letting SOs “undertake public affairs and solve problems in suitable areas.”²² On the other hand, policies have been implemented that extend Party power to SOs,²³ restrict fundraising²⁴ and hinder the registration and activities of international organizations.²⁵ These have resulted in complex and changing state–society relations in China.

As part of the broad policy reforms, the Chinese central government launched the policy of contracting out welfare services.²⁶ This started the formal, contracted collaboration between the state and SOs based on the following rationales. First, services contracting should “[be] connected with the public sector reform, promote the division between administrative and public institutions and between state and society.”²⁷ Second, contracting supports “liberalization of access to the public service market” and selects providers “in a competitive and merit-based manner.”²⁸ Finally, services contracting is a means to grow and regulate SOs. In 2014, the Ministry of Finance published a policy that clarified that contracting also aimed at promoting SOs’ ability to offer services.²⁹ The “Guiding opinions on supporting the development of social organizations through services purchasing” (2016) further suggested steps to achieve this goal, such as relaxing registration requirements, extending contract lengths and offering training. Meanwhile, contracting regulates SOs with techniques such as drawing up lists that exclude certain types of organizations or issues, using criteria like Party-building in selecting contractors, and adopting a results-based evaluation to control SOs’ activities.³⁰

To recap, the contracting policy is informed by complex policy agendas to reform China's public sector, pluralize its welfare provision and reform social governance. Officials are required to implement the policy according to these rationales. Meanwhile, they need to adapt the central design to suit local conditions. As services contracting has been practised across China, studies have emerged that explore the various contracting arrangements and models,³¹ interactions between the state and SOs³² and effects of contracting on SOs and welfare provision.³³ These have added greatly to our knowledge of China's welfare regime and state–society relations. However, most empirical studies focus on advanced or pilot regions, where contracting “occurred without conscious coordination by the central government.”³⁴ This limitation has led to the claim that services contracting in China is “generally less seen as ‘designed’ than ‘emergent’ in an inevitable, quasi-automatic process involving the actions and interests of the multiple public and private stakeholders active in complex modern societies and economies.”³⁵ The reality, however, is that contracting for many localities is a top-down policy, which includes the aforementioned rationales that might be hard to achieve in local conditions. Understanding how the policy is implemented in non-pilot, less advanced regions is thus a significant issue.

Understanding the Chinese Policy Process

The Chinese policy process has changed considerably in the reform era. Scholars have proposed a variety of theories to characterize the patterns, features and rationales of policymaking and implementation in contemporary China.³⁶ Here, we briefly review the three most used theories – namely, fragmented authoritarianism, experimentation under hierarchy and policy innovation. We outline their limitations in analysing policy implementation in the welfare domain. We then introduce multiple logics as our analytical framework through which to understand policy implementation in second-tier localities in China.

Introduced by Kenneth Lieberthal and colleagues, the “fragmented authoritarianism” model sought to explain bureaucratic decision-making processes in post-Mao China.³⁷ By examining the changes in China's economic and bureaucratic reforms since 1978, the model asserts that the central state and its policies have become increasingly malleable to serve the goals and interests of subnational agencies. This leads to the emergence of bargaining and negotiation within the bureaucratic system. Policy outcomes are thus shaped by the incorporation of the interests of the implementation agencies into the policy. Fragmented authoritarianism has been one of the most popular concepts applied to understanding the Chinese policy process. Rich studies have adopted the framework to examine how Chinese bureaucrats deal with top-down policies that contradict their interests. These suggest a range of coping strategies like evasion, distortion, selective implementation and collusion.³⁸

While fragmented authoritarianism focuses mainly on the tensions between the centre and local states, policy experimentation pays attention to the collaboration between different levels of the state. Analysing China's economic policies, Sebastian Heilmann identified an “experimentation under hierarchy” model which encourages local officials to innovate so that central policies can be changed using the results of the experiments.³⁹ The pattern involves a point-to-surface process (youdian daomian 由点到面), including innovation in “experimental points,” the extraction of “model experiences” and the expansion and emulation of results. Experimentation thus “minimized the risks and the cost to central policymakers.”⁴⁰ Scholars have applied this concept to examine experimentation behaviour in areas like housing and the administration system reform.⁴¹ Some claim experimentation has had positive impacts on China's prosperity.⁴²

The theory of “experimentation under hierarchy” acknowledges the authority of central leadership in policy design and the regularized channels to generalize local experiments. In practice, however, the leading role of the central state has been questioned. For example, by studying news in People's Daily, Ciqi Mei and colleagues found that only 37.81 per cent of experimentation was led by the central state.⁴³ Some scholars have then turned their attention to spontaneous policy innovation in which local bureaucrats play a more active role. This body of literature has mainly focused on officials’ incentives to innovate, such as the shortage of administrative resources to solve local problems, desire for promotion and peer pressure.⁴⁴ Some examine bureaucrats’ individual attributes. For example, Reza Hasmath, Jessica Teets and Orion Lewis distinguished three types of personalities: authoritarian, consultative and entrepreneurial. Their study showed how these personalities affect officials’ policy behaviour.⁴⁵

The above theories illustrate and attempt to explain the various types of local bureaucratic behaviour in China: the fragmented authoritarianism model focuses on the incentives and patterns of local bargaining; the experimentation under hierarchy and policy innovation theories examine creative policy practices of local officials, whose targets correspond with the central state's goals to varying degrees. These theories suggest some key elements of the Chinese policy process, such as the tensions between the central and local governments and a performance-based evaluation system. The literature, however, has the following gaps. First, it is assumed that China's political decentralization has allowed local bureaucrats to attain a high degree of autonomy in amending central policies. While this might be true in the early stages of the reform, the situation has started to change. Research has found that Xi Jinping's 习近平 administration displays considerable centralization, which reduces the space afforded to local officials for innovation.⁴⁶ Whether local bureaucrats still have the discretion to bargain with higher authorities or facilitate innovation becomes a question. Second, existing studies often see local Chinese states as a whole unit, and/or focus on the leader while overlooking the tensions within the bureaucratic system. China has a complex governing system, comprising the Communist Party, the administrative governments and their subordinate agencies. The policy process includes different, or even conflicting, rationales.⁴⁷ The third gap is on the role of non-state actors. It is accepted that studies of the Chinese policy process need to consider corporate, collective and individual actors in society.⁴⁸ However, while some empirical studies have discussed the role of social actors in areas like environmental protection and health reform,⁴⁹ they often focus purely on state–societal tensions and fail to connect these with other tensions. Finally, most theories about the Chinese policy process, such as fragmented authoritarianism and experimentation under hierarchy, are based on the economic realm. Their applicability in the welfare domain needs to be tested.

To address these gaps, we develop a framework with multiple logics drawn from Xuegang Zhou and Yun Ai's study about China's institutional change to analyse the implementation of services contracting in China's second-tier localities. In their study of the evolution of China's village election system, Zhou and Ai identify three logics, or “stable institutional arrangements and corresponding mechanisms that lead to certain patterns of behaviour in the field.”⁵⁰ These include the state logic, referring to the general frame defined in policies and institutional arrangements; the bureaucratic logic, meaning officials consider the cost and outcomes of policy implementation and act in ways that benefit their promotion prospects; and finally, the rural logic, which refers to the participation of village people in economic, political and social affairs. The authors then argue that the changes in the election system are determined by the interweaving of the three logics.

Zhou and Ai's framework rightfully reminds us that local policy practices include various aspects of relations and interaction, rather than a single dimension of top-down or state–society relations. We believe services contracting, as a top-designed policy that brings in new actors like SOs, would also create such a complex situation. We thus develop a framework based on the state, bureaucracy and society to examine the implementation of the policy at the local level. Unlike Zhou and Ai's study, which discusses multiple actors and their respective logics, we focus on local officials as the implementors of the policy. We define “logic” as institutionally constructed rules and principles that bureaucrats elaborate on in policy implementation and which generate certain patterns of behaviour. As services contracting is a top-down policy, we first examine the state logic and how local officials acknowledge the goals and rationales defined in central policies and fulfil these in practice, which has been a core theme in studies about the Chinese policy process. Second, as our case exemplifies localities with little space or resources with which to bargain or innovate, we explore the bureaucratic logic to see how officials weigh up the benefits and risks to themselves in implementing the policy. Finally, considering the nature of services contracting, we examine the officials’ logic in relation to society. The Chinese government wants the policy to stimulate a more active and participatory society, but the extent and forms of collaboration vary according to local conditions. We use the case of City A to analyse the three logics in the following sections.

The Case of City A

This section introduces our case city, anonymized as City A, and its conditions for implementing the contracting policy. City A is a county-level city in east China. It has eight towns, seven functional districts and more than 1.2 million residents. The development of the city is unbalanced. On the one hand, City A is industrial and economically advanced. It is one of the 100 wealthiest counties in China, with a GDP of 285 billion yuan in 2019. On the other hand, the city lacks welfare provision and societal forces. In 2011, City A had only 375 registered SOs, of which 177 were people-run non-enterprise organizations (minfei 民非).⁵¹ These were mainly informal self-organized groups and cultural groups, which were “undeveloped, lacking staff with requisite skills, and incapable of services delivery.”⁵² In 2012, City A's government stated in its annual report that “our public services cannot meet people's diverse needs.”⁵³

Since 2013, City A has had to manage a top-down mandate to contract welfare services. Following the announcement of the central policies, the provincial and prefecture governments under whose jurisdiction City A is located published a series of policies that echoed the rationales to reform the public sector, innovate welfare provision and reform social governance. These policies also added details and practical methods, mainly concerning what type of services should be outsourced and the roles and responsibilities of different government departments in the policy process.

In October 2013, the provincial government promulgated its “Guidance on contracting out public services,” which was a rapid response to the 2013 Guiding Opinions. This document reiterated the Guiding Opinions’ goal to “deepen the reform in the social area, transfer government roles … improve the quality and efficiency of public services” but also added practical targets like “define and extend public services” and “regulate the contracting practices.”⁵⁴ In 2015, the provincial-level finance and civil affairs departments published the “Implementation methods of the government purchasing of services from social organizations,” which had two differences compared to the central-level policies. First, the policy defined SOs as welfare services suppliers rather than use the vague concept of “social forces.”⁵⁵ It omitted the requirements for SOs to be eligible for contracting, as listed in the 2014 Management Rules.⁵⁶ Second, the policy clarified the institutional arrangements for contracting: the financial departments were responsible for setting up the system, designing and reviewing the purchasing plans, and were to play a leading role in relevant arrangements and evaluation; the civil affairs departments were responsible for supervising and managing the involvement of SOs; and the evaluation departments were to monitor and assess the process.

The contracting policy was further clarified at the prefecture level. In 2014, the prefecture-level government published “Guidance on the government purchasing of services from society and the implementation methods of the purchasing of services from society (temporary).” This presented a detailed list of welfare services to be contracted out, including 28 areas such as services to the elderly and disabled people, and community services. Services listed in the catalogue “have to be contracted out, we are not allowed to do [offer] them ourselves.”⁵⁷ The document also further clarified the institutional arrangements, defining the roles and responsibilities of seven departments, including the departments of civil affairs and finance and the development and reform committee. The policy also defined practical issues such as the length of contracts. In 2015, the finance department published “Methods of budget management and performance evaluation of services contracting to society,” which introduced specific management and evaluation methods of contracting.

These policies put pressure on City A's officials to contract services although local conditions imposed notable constraints. The first issue was the lack of nongovernmental services suppliers. While higher-level policies required the contracting process to adopt the principles of public competition and fairness in selecting contractors, City A lacked eligible SOs to undertake services, not to mention a sufficient number to compete in the market. Officials described the dilemma as “we have to cajole [the few existing] SOs into doing the jobs … they are already busy.”⁵⁸ Second, officials had little in the way of administrative resources as well as limited knowledge or experience of drafting contracts, setting up assessments and monitoring services delivery, all of which were found to be essential to contracting out services in advanced cities.⁵⁹ Third, unlike advanced cities where academics supported the creation and training of SOs, there were no nearby universities with relevant expertise, meaning City A's officials had fewer options to turn to for help. Finally, compared to pilot cities, officials had less room to manoeuvre and faced higher risks when adapting the central instructions. All of these factors put City A on a different starting block when it came to contracting out welfare services. By choosing City A as a case study and examining how local officials behave under such circumstances, we present the little-explored side of non-pilot cities in services contracting.

Services Contracting in City A

Faced with the top-down mandate on the one hand and local constraints on the other, officials in City A implemented services contracting in two steps, with the introduction of an investment programme first and then the establishment of an intermediate foundation.

First, officials developed a “creative investment for public goods” programme (gongyi chuangtou 公益创投, CIPG hereafter) as the city's main method for contracting.⁶⁰ In 2014, officials in the civil affairs bureau (CAB) invited a third-party organization from Shanghai, which has a strong reputation for growing SOs, to establish a branch in the city (anonymized here as PM). With the help of PM, officials set up a fund that had a separate budget and was approved by City A's finance bureau. The main source of funding came from the city's welfare lotteries. This CIPG fund was opened to local non-profit forces, including registered and recorded SOs, and “grassroots individuals who are interested in public affairs.”⁶¹ This was designed to include as many nongovernmental services suppliers as possible. The CIPG programme used a competitive process to select contractors, which included the following procedures: the CAB defined the general categories of services, such as social work, community services and Party-building. SOs designed their services plans to fit within the categories and submitted their proposals, which were assessed according to the area covered, design, staffing and budget. Once selected, the SOs signed a contract with the CAB, which was for one year and offered various amounts of funding between 10,000 and 200,000 yuan. The services delivery was evaluated by third-party experts.

The CAB's design was soon replicated by the town- and district-level governments in City A, using their own fiscal budgets. By 2016, the CIPG programme had become the main contracting process in all the towns and districts. While the programme had been efficient in developing SOs, some issues emerged, such as the overlap between services purchased by different levels of governments, conflicts arising when SOs were participating in multiple purchasing processes, and a shortage of officials available to manage the process. In response, City A's officials took the second step to establish a “community development” foundation in 2016 to facilitate contracting. This was claimed to be the first county-level community development foundation in China.⁶² It was registered as a civic foundation (minban jijinhui 民办基金会) but was created and fully funded by the CAB. A village Party secretary was nominally appointed as the foundation's legal representative although, in practice, the foundation was managed by PM.

Officials then outsourced contracting-related work to the foundation/PM. In 2016, the CAB required town- or district-level governments to put their budget for CIPG contracting into the foundation, which would then organize the purchasing as a whole. The town/district-level officials could decide what services they wanted to buy, and the foundation would deal with practical issues like publishing information, contacting SOs and collecting their plans, and arranging the contracts. The CAB's officials switched role from being buyer to acting as regulator, fulfilling functions like approving the contracting plans and guiding the development of SOs. The cost was shared between the CAB (30 per cent) and lower-level governments (70 per cent). Via the foundation, officials also rolled out services contracting from the civil affairs system to other government institutions. For example, the organization department of the CCP in City A initiated a scheme to offer 200,000 yuan a year to each community for activities that centred on Party-building, public services and community self-governance – 30 per cent of the funding went into the foundation to be used by SOs.⁶³ The foundation also started to raise money from the private sector, including a local bank and a company.

The above contracting practices were deemed to be successful. From 2014 to May 2018, the CIPG programme spent 28 million yuan in governmental funds and launched 586 projects with the participation of 134 registered and 37 recorded organizations, 1,772 social workers, and more than 17,500 volunteers. The rate of competition for SOs submitting tenders varied between 1:2 and 1:3. The projects offered services to over 46,900 direct users in all the 180 communities and villages in City A.⁶⁴ Officials claimed that contracting improved the quality of services and “eased the tasks of officials.”⁶⁵ The CIPG programme also became the “biggest and most innovative direction SOs could take.”⁶⁶ By 2018, City A had 2,361 registered SOs, seven times more than in 2011, including 1,017 minfei organizations. Overall, from a difficult starting point, City A's officials actively and efficiently implemented the policy to contract out welfare services. We next analyse their behaviour and the multiple logics behind this.

The Multiple Logics of Officials’ Behaviour

Our analyses show that officials in City A followed three logics in implementing the contracting policy: they wanted to meet the central state's targets, to balance policy outcomes and risks, and to stimulate a more participatory society. Their policy behaviour has been framed in a certain way, as shown in Table 1.

Table 1: Services Contracting at the County Level: Logics, Behavioural Patterns and Examples

Meeting the central state's targets

As explained, China's contracting policy derives from complex rationales for reforming the public sector, pluralizing welfare provision and reforming social governance. These highest-level decisions were made without involving the opinions of local officials from non-pilot regions or considering regional conditions. We found City A's officials endeavoured to meet the directives, which generated entrepreneurial and welfarist behaviour in their policy implementation.

Entrepreneurial behaviour

Research shows that Chinese bureaucrats often display entrepreneurial behaviour in promoting economic growth, meaning they take risks and directly engage in business in search of profits.⁶⁷ Our study found similar behaviour in the domain of welfare. Through the CIPG programme, officials in City A invested financial, administrative and social resources in developing local SOs. The programme offered “seed funding,” particularly to small-scale and unregistered organizations, which “solved our [SOs’] problem of survival.”⁶⁸ Officials shared administrative resources with SOs, such as workplaces and access to services users, and connected them to the media so that SOs could attract wider attention. Contracting also provided SOs with legitimacy, which was key to gaining public trust in places like City A.

City A's officials also created SOs to undertake governmental services. As explained by a town-level official, the CAB had a specific task to “help to create nine SOs every year.” The official was assigned the task of creating an SO to serve disabled children in a context where “few [non-governmental] people were familiar with this field.”⁶⁹ This government-led organization (GONGO) then utilized the administrative resources of the government – for example, its power and channels to mobilize volunteers – to offer services. It has an advantage when participating in the CIPG programme, as “rural people don't know about social workers, so they trust us [as] government [officials].”⁷⁰ In such ways, officials acted as SOs themselves in the policy process.

The behaviour of City A's officials mirrors the behaviour of the entrepreneurial state in that officials invested bureau funds or arranged resources for businesses, or they became entrepreneurs themselves. The primary goal here, however, was not personal or institutional profit, as is the case in state entrepreneurialism.⁷¹ Officials claimed the CIPG programme was a necessary strategy to implement the contracting policy, as “there is no other way. In theory, we should have enough SOs and then transfer governmental work to those who are capable, but how can SOs develop without money?”⁷² This entrepreneurial behaviour carries risks, as will be explained in later sections.

Welfarist behaviour

In her studies on China's state–society relations, Jude Howell defined the term “welfarist incorporation” to refer to a pattern of collaboration between the state and civic organizations in the provision of services, rather than the representation of interests.⁷³ This study confirms the trend. We found that City A's officials introduced practices to expand welfare provision and regulate SOs to focus on services. First, in City A, contracting was expected to create new, additional welfare services because “we [the government] cannot offer services to each door.”⁷⁴ For this purpose, officials used the CIPG programme to task SOs with investigating and developing services according to local demand. They valued the creativity and quality of services, rather than the retrenchment of public expenditure. An official explicitly claimed this in an interview: as “unlike public bidding, we do not select services according to a low price; that's not our aim.”⁷⁵

At the same time, officials introduced a set of strategies to regulate SOs to focus on services. The CIPG programme provided paperwork templates, which used rigid and bureaucratic indicators such as a specific number of service users, to help SOs to design, deliver and report their activities. The programme required that all SOs’ plans and activities be reviewed, approved and monitored by community staff. Also, the CAB purchased capacity-building services from PM – described as “nurturing growth” (peibanshi chengzhang 陪伴式成长) – which directed SOs to focus on practical issues rather than strategic work centring on rights and power relations. These arrangements put pressure on SOs to orient all or part of their work towards the state's purposes.

Both the entrepreneurial and welfarist behaviour of City A's officials responded to the central state's targets to reform welfare and social governance. This differs from theories of experimentation under hierarchy or policy innovation, where local officials invent distinctive patterns that may affect the policy directions at the national level. Also, unlike findings on fragmented authoritarianism, officials did not bargain with higher authorities, even when faced with significant constraints in implementing policy. Neither did they use coping strategies like distortion or symbolic compliance.⁷⁶ Instead, City A's officials developed a range of strategies to seed a social sector they could work with and build upon. Their behaviour suggests a more active response to top-down mandates.

Balancing policy outcomes and risks

Similar to many other Chinese policies, the policy language of services contracting at the central level is full of vague statements, providing general guidance and leaving room for local adjustments. In implementing the policy, bureaucrats need to target positive outcomes to obtain promotion through the performance-based ranking schemes. At the same time, there is uncertainty regarding policy outcomes, which may put officials’ careers at risk. Officials in City A thus followed a logic to balance policy outcomes and risks, which generated innovative and risk-sharing behaviour.

Innovative behaviour

Officials in City A used an innovative approach to adjust services contracting to suit local conditions. First, owing to the lack of nongovernmental services suppliers, officials relaxed the requirements for organizations applying to supply services. The CIPG programme allowed unregistered SOs and even individuals to apply for the funds.⁷⁷ This differed from the requirements at higher levels but was effective at encouraging a large number of SOs to participate in welfare provision. Second, officials in the CAB limited the funding for individual projects to below 200,000 yuan, which, at higher levels, was the threshold amount for review by the finance bureau. This was to save time and avoid potential conflicts of ideas between governmental departments. As one official explained: “we couldn't have these many projects if they [the finance bureau] were involved [in reviewing SOs’ plans]. They just don't understand welfare services.”⁷⁸ Third, the CIPG programme required SOs to investigate the communities in which they intended to work and develop service plans to meet people's demands. This was to generate new services and innovate welfare provision in the areas in which local officials had limited knowledge. All these adjustments were made to achieve positive policy outcomes with limited resources.

Risk-sharing behaviour

Officials recognized the risks associated with their adjustment of the policy and so adopted strategies to share the risks with other actors. There were two areas in particular that presented potential risks to their career prospects: investment efficiency and the management of SOs. First, the CAB invested financial, administrative and human resources, including workspace and money, when contracting out work to SOs and it expected high-quality welfare services in return. But as contracting was a new practice in City A, there were no agreed standards on the quality and appropriateness of services. Also, since the CIPG programme was the only funding source for most SOs in City A, officials found it hard to not renew contracts, as this would undermine any achievements in developing the social sector. Officials thus lacked exit strategies to free up resources for the next investment. To reduce these risks, the CAB assigned responsibility for the organization of services contracting to the foundation, which in turn shared the funding responsibilities with lower-level governments (70 per cent). Via the foundation, the CAB also invited other government institutions, such as the organization department of the CCP, to participate in the contracting of welfare services. This reduced the risk of incurring financial shortages and also expanded the scale of contracting out, which was potentially a positive policy outcome.

The management of SOs was also an area of potential risk. Officials acted entrepreneurially to develop SOs to undertake welfare services, but “they may lose their job if [SOs] did something wrong.”⁷⁹ For example, in 2014, when the CIPG programme just started, City A's financial bureau conducted a review of how governmental funds were being used and discovered a few CIPG projects using “non-standard invoices.” This put the CAB's officials under pressure, particularly against the background of the strict fiscal accountability demanded under Xi's anti-corruption campaign. Also, by inviting nongovernmental forces into public affairs, the CAB officials were questioned by other departments, such as the education department and the police, about the possibility of SOs threatening social stability. To counter such risks, the CAB designed protocols for community staff to manage SOs. Since 2015, SOs applying for CIPG funding have been required to conduct field research, design their plans according to identified needs, and contact the community or village in which they planned to work for approval. A formal, stamped letter from the local leader is required in the application. The CAB also insist that community staff attend, assist and monitor SOs in the services delivery process. Such strategies enable officials between different departments to spread the risks of managing SOs and to minimize issues and activities that might be considered as politically sensitive.

The above behaviour has been shaped by the logic that officials aim to achieve positive policy outcomes and, at the same time, minimize any potential risk to their careers. City A's officials introduced innovative measures in minor aspects of policy implementation, such as relaxing the registration requirements and keeping funding below 200,000 yuan. This differs from large-scale innovation that aims to produce distinctive models (as researched in policy innovation theories) or even to change central designs (discussed in the experimentation under hierarchy theory). This “practical” innovation, balanced with risk-sharing behaviour, reflects the dynamics between different parts of the bureaucratic system.

Stimulating a more participatory society

The contracting policy per se encourages the participation of non-state actors in public affairs. Previous studies demonstrating how this institutionalized collaboration affects China's state–society interactions have produced controversial findings.⁸⁰ On the one hand, scholars argue contracting has offered SOs chances to improve their standardization, services and fundraising capacity as well as their administrative and media advocacy,⁸¹ which contributes to the transformation of authority, the enhancement of public trust and even collaborative governance.⁸² On the other hand, contracting has been found to extend governments’ control over SOs by requiring duties beyond the contracts, employing fragmented, workload-oriented evaluation and distributing resources to SOs that most fit the state's goals.⁸³ SOs thus become the “foot soldiers,” rather than equal partners, of local governments.⁸⁴

Unlike the much-researched cities such as Guangzhou and Shenzhen, City A lacked a mature social sector for the state to collaborate with in providing welfare services. Local officials thus had to stimulate an active and participatory society along with the implementation of the policy. The most important strategy they adopted was to invite PM, the non-state professional agent, to participate in the making and implementation of local policy. PM played multiple important roles in the policy process. First was as rule maker. Officials asked PM to draft the content and procedures for contracting, such as categories of services to purchase and assessment methods. Although the drafts needed to be approved by the CAB, the officials chose an approach that included non-state actors in policymaking. One of the interviewed officials even claimed that they “respect the third party very much and do not know its criteria [in selecting services suppliers].”⁸⁵ Second, officials appointed PM as the manager of the foundation and the CIPG programme. This meant that their own administrative tasks, such as publishing information, organizing contracting events and collecting paperwork, were transferred to PM. It also gave PM the power to pre-review SOs’ proposals before formal submissions and even to expel SOs from the tender process if they failed or refused to revise plans according to PM's advice. Third, officials authorized PM as the regulator of City A's SOs. The CAB had been funding PM to offer long-term training and consultancy to local SOs. As stated in the section on welfarist behaviour, the training was to meet the state's targets to develop services-oriented SOs; however, it also allowed PM to make adjustments to the local social sector.

What we can see here is the invitation and full engagement of non-state actors in policymaking, which has been underestimated in previous studies about China's services contracting. Officials in City A chose PM as a key partner and awarded this non-state organization power and authority throughout the policy process. The director of PM described the dynamic: “It's wrong to say we have no power, or we have absolute power. Most of the time it's co-work.”⁸⁶ That is, City A enacted deeper participation and collaboration, compared to other presented studies in which SOs mainly acted as government foot soldiers.⁸⁷ The invitation to participate in policymaking, however, was made exclusively to PM, an “outsider” third-party agent with limited connections with the local social sector; local SOs were only included in the provision of welfare affairs. An interviewed SO expressed their discontent about this, as “they [PM's staff] are young college graduates, who know little about this city or our specific areas [of services].”⁸⁸ In other words, officials were selective in stimulating a participatory society. This suggests the interplay of the society logic with the bureaucratic logic, as officials need to balance policy outcomes and risks.

Conclusion

This article set out to examine how local officials respond to the top-down policy of services contracting against a background of significant constraints. We developed a framework of multiple logics to analyse how China's policy of services contracting is implemented at the county level in a less developed region, an area that has been underexplored in the literature. Our research found that officials reacted to the policy following three logics: to meet the central state's targets, to balance policy outcomes and risks and to stimulate a participatory society. These logics led to entrepreneurial, welfarist, innovative, risk-sharing and collaborative behaviour in the implementation of the policy.

We argue that the implementation of services contracting at the county level is determined by the interweaving of the three logics, rather than simply the top-down pressure to implement policies or the altered relationship between China's state and society. Local officials actively respond to top-down mandates even when facing significant constraints, which contrasts with previous findings that local bureaucrats bargain with higher authorities or adopt strategies to cope with central policies.⁸⁹ While the state logic determines the key approach taken in policy implementation, officials also consider their benefits and employ strategies to balance policy outcomes and risks. The two logics and the behaviour they generate are accompanied by the logic to expect a more participatory society. This echoes China's social governance reform and has been observed in welfare policies and practices in other Chinese cities. In our case, however, the society logic generated a warm invitation and the deep engagement of a non-state professional agent in the whole policy process, including the making of local policy. The three logics of the state, bureaucracy and society also overlap in determining the practices of services contracting. For example, the decision to invite PM to participate in local contracting suggests both the logic to stimulate a participatory society and the bureaucratic logic to reduce policy implementation risks. The officials’ collaboration with PM to offer training to local SOs, combined with their entrepreneurial behaviour to provide resources and opportunities, echoes the goal of the central state to develop a services-focused social sector. That is, local officials juggle the three logics, combine them in different ways, and sway between the various patterns of policy behaviour according to the risks involved.

Our study presents novel evidence to studies of services contracting in China. Most existing studies in the field focus on former-pilot or advanced cities, where contracting occurred before the initiation of the national policy and/or local governments had resources and capabilities for policy innovation.⁹⁰ While these studies offer many insights, particularly on China's new state–society relationship, it needs to be noted that the few advanced cities do not represent the whole picture of contracting in China. Similar to our case city, many localities have a different starting point and face multiple and complex challenges in implementing the policy. Examining how officials in these places behave and the multiple logics driving their behaviour contributes to understanding the variation in services contracting in China and its impact on China's state–society relations.

The article also confirms the complexity of policy implementation in China, especially in the domain of social welfare. Theories about the Chinese policy process have mainly stemmed from the economic area. They reveal how officials bargain with higher authorities for potential benefits, or how officials that have power, resources or opportunities experiment or innovate with central policies. Our research shows how officials in a non-pilot city, without a special position, resources or strong economic incentives, behave when implementing a national welfare policy. Our study, as qualitative research of a rather overlooked sub-field, also has its limitations. First, it did not examine the variation of contracting behaviour among City A's towns and districts. Second, our analysis covered three services areas, while there may be variations in contracting behaviour across services sectors. We do not claim our arguments to be representative or generalizable, but we believe Chinese local officials elsewhere and in other policy domains may also need to juggle and swing between different logics in their policy implementation. We hope our findings will pave the way for further research to examine the complexity of the Chinese policy process in other sectors.