Studying the Chinese Policy Process in the Era of ‘Top-Level Design’: the Contribution of ‘Political Steering’ Theory

The recent Chinese discourse on the Chinese policy process and the tensions between centralization and de-centralization in central-local relations has been driven by decisions of the 19th Party Congress held in October 2017 concerning the centralization of the institutional apparatus of China’s political system: Further pushing the rationale of ‘top-level design’, party and government organs are systematically integrated and party committees within government institutions strengthened. This development is politically sensitive. Chinese scholars are very careful to reflect on the centralization of the policy process and the institutional apparatus of China’s political system which has been brought on track since the 19th Party Congress. Most articles published in China remain at the descriptive level and often are apologetic, in the sense that they either legitimize the steps taken by the Xi Jinping government to streamline the policy process and institutional apparatus, or ask for careful implementation of additional reforms to secure administrative efficiency and political stability. See e.g. Lee [46]; Ni and Wang [64]; Zhang [90]; Zhu [95].

The concept of ‘fragmented authoritarianism’ was originally developed and empirically applied in Lampton [44], Lieberthal and Oksenberg [49], and Lieberthal and Lampton [50]. For recent studies see Brødsgaard [10, 11].

Zhou [93] and Zhou et al. [94] have neatly complemented ‘fragmented authoritarianism’ with a more micro-level perspective focusing on collusion of local government officials and a ‘muddling through’ approach - strategies of intermediate party state bureaucracies in Chinese policy implementation. It is the institutional environment which induces local governments to collude for making the best use of the discretion they have been given to implement upper level policies. Moreover, government officials have to come to terms with different ‘bureaucratic logics’ of the tizhi – meeting targets, coalition-building and incentive provision – which make them reactive, urge them to constantly adjust policies and let them focus on short-term solutions. These institutional logics add to the problem of competing bureaucracies and overlapping competencies by enforcing hard targets which can result in ‘goal-displacement’ or distorted policy implementation. However, collusion (as much as ‘muddling through’) ‘may unwittingly act as a corrective and countervailing force to the centralization of decision-making’ and its negative aspects for local policy adjustment as well ([93]: 36).

Heilmann noted that ‘the entire policy process must be conceptualized as an oscillating multilevel interaction rather than as a dichotomized process of centralization vs. decentralization’ ([33]: 12). He also stressed that ‘societal input is still ad hoc and selective, and civil society organizations do not play a significant or even institutionalized role’ ([33]: 20). For a historical contextualization of the ‘point-to-surface’ approach in the Chinese policy process see [32, 35]. For a negative assessment of policy reform at local level, bespeaking ‘symbolic reforms’ see Zeng [91].

As Chen notes, ‘the strict enforcement of the System of Requesting Instructions and Submitting Reports, as well as Rule by Law, have led to local cadres prioritizing self-protection by exclusively and straightforwardly implementing the tasks and policies assigned by the goals of the central government, rather than adapting broad guidelines to local conditions for improved local governance’ ([13]: 669).

Yu and Huang have presented another typology to grasp recent changes in Chinese policy-making. They distinguish, on a continuum, between ‘autonomous exploration’ (zizhu tansuo), ‘autonomous exploration’ as the focus of ‘design and experimentation’ (sheji shiyan), ‘design and experimentation’, and ‘requesting authorization’ (qingshi shouquan). The last type, qingshi shouquan, is the policy-making mode of the Xi Jinping era, by which ‘local governments, for the purpose of solving local problems which are of a general character, come up with a practical plan that demonstrates their willingness to push for reform, report it formally to higher government levels and ask for approval for an experimentation site; after having gained higher level consent and authorization, implementation begins’ ([87]: 94). For the authors, this mode is the most appropriate way to rebalance the tension between hierarchical control and local discretion to the benefit of policy reforms that are relevant for the whole country.

Despite the tightening up of the political system under Xi, this still seems to be the case. For example, the National People’s Congress petition of several law professors against stipulations in recently amended provincial regulations on population and birth planning proved successful; see “Quanguo renda changweihui shencha difang jisheng tiaoli: Jianyi xiugai ‘chaosheng jiu citui’” [National People’s Congress Standing Committee Reviews Local Birth Planning Regulations: Proposes to Revise “Excess-Birth Means Dismissal”]; Pengpai (The Paper), 27/10/2017.

For a condensed version of the main theoretical advances, see [55, 58, 70, 73]. For an overview in historical perspective, see [79]. As often in the case of continental European theory-building, ‘political steering theory’ was widely ignored by U.S. scholars at the time, most likely because it suggested the possibility of systematic public policy planning.

Whether this conceptual shift has been accompanied by parallel transformations in actual practices of governing is debatable, see [8, 52].

This tension is particularly related to the semantic indeterminacy and conceptual inconsistencies inherent to the term ‘governance’, as Offe [67] has convincingly shown.

Mayntz and her colleagues have also discussed wider definitions of ‘governance’ as generic terms for all kinds of coordinated social interaction (soziale Handlungskoordination), sometimes with explicit normative connotations that accentuate claims for ‘good governance’ in the sense of transparency, participation, accountability and legitimate public policy objectives. Mayntz noted in this context that governance can refer either to a structure that regulates action, or to the process of such regulation ([58]: 46). Clearly, the ‘governance’ concept vacillates between these two understandings of the term to the very day.

The distinction of steering and rowing has also entered the literature on (multi-level) governance, see [4, 47].

This original ‘steering optimism’ was later tempered by experience and theoretical objections, and it was system theory à la Luhmann in particular that challenged ‘political steering’ theory by claiming that the steering of autonomous social systems is impossible (see [51, 59, 71]).

Governance without any state involvement (such as network governance) is, however, highly unlikely in a case such as China where even incipient forms of societal self-regulation fall under the ‘shadow of hierarchy’.

A different typology is provided by Levi-Faur [47] who regards steering as one mode of governance and envisions state steering capacities developing in more horizontal ways under contemporary circumstances.

This line of inquiry has to be understood as research on political steering, not research to guide and help political steering, as in the early ‘optimistic’ phase of the steering paradigm, see [79]: 80.

The idea of political steering ‘in the shadow of hierarchy’ has been introduced by Scharpf [72] and refers to the state’s looming power to step in and regulate with direct intervention if delegated social self-regulation, cooperation in networks or corporatist arrangements are ineffective. For an analysis specific to the Chinese context see the work of Heilmann [32‐34].

As Grande [24] remarks, in cases where a steering center still exists, one should not speak of ‘governance’.

With this typology we follow Scharpf who distinguishes between aggregate, collective and corporate political actors in his actor-centered institutionalism [73].

We use ‘meta-steering’ as analogous to ‘metagovernance’ introduced by Jessop [40]. ‘Meta-steering’ refers to the coordination of different structures and practices of steering across all administrative tiers in the Chinese political system. Therefore, ‘meta-steering’ is the mode by which all steering is steered by the political center.

We use ‘governability’ in a strictly analytical sense, i.e., to refer to those actors in specific social fields who are targeted by specific policies and who may either comply with or attempt to shirk them. This is also the term most commonly used in the English-language literature, cf. Jessop [40]; Bevir and Rhodes [8].

Whereas ‘coping’ or ‘shirking’ are typical avoidance strategies set against upper level steering, ‘counter-steering’ is more. We define it as goal-oriented and purposeful action on the part of steering objects (lower-level bureaucracies or non-state actors, even the populace) to bring about social and economic conditions that contradict the original objectives of policies passed down to them. ‘Counter-steering’ has been prominently conceptualized by Jürgen Habermas as strategic action by societal actors to maintain systemic complexity (see [26]), pinpointing the positive significance that counter-steering can have for overall system stability.

According to Fligstein and McAdam, strategic action fields ‘are the fundamental units of collective action in society. A strategic action field is a constructed meso-level social order in which actors (who can be individual, collective or corporate) are attuned to and interact with one another on the basis of shared (though not necessarily consensual) understandings about the purposes of the field, relationships to others in the field (invoking the question who has power and why), and the rules governing legitimate action in the field’ ([20]: 9).

Our approach is hence informed by actor-centered institutionalism à la Scharpf as an analytical framework for interaction-oriented policy research (interaktionsorientierte Policy-Forschung) ([73]: 84), which focuses on strategic action by complex actors – state and non-state – to explain the policy process and assess its outcomes. However, we have no intention of formalizing these constellations and interaction patterns through game theoretical analysis, but instead trace them through qualitative research methods (see below). This is more in keeping with the sociological roots of ‘strategic action field’ theory. Compared with most political science literature, the latter places greater emphasis on perceptions and collective identities rather than mere material self-interest as factors underlying political action (but see [7]).

In fact, every policy field is a ‘strategic action field’.

For instance, policy implementation might also be located at the municipal level, and townships and villages may play a significant role for policy implementation as well.

For an abridged version in English see http://​www.​china.​org.​cn/​china/​third_​plenary_​session/​2013-11/​16/​content_​30620736_​4.​htm. The principal objective of this document, which lists 60 reform measures covering 15 policy fields, as declared by China’s government, is to effect a shift to a new model of growth and development.

As a matter of fact, periods of tightening up (shou), in which the hierarchical control over subordinate state bureaucracies was strengthened, and liberalization/loosening up (fang) could be observed since the early reform days [6, 17, 78].

The other three LSGs-turned-commissions are the ones for cybersecurity and informatization, finance and economy, as well as foreign affairs.

It replaces the State Ministry of Supervision and the National Bureau of Corruption Prevention. Also see [81].

This is reminiscent of Sun Yatsen’s constitutional thinking who proposed ‘five branches’ (executive, legislative, judiciary, control and examination) in his ‘Three Principles of the People’. Such a structure was adopted in the 1946 constitution proposed by the Nationalist Party (KMT). The historical inspiration for Sun was the censorate of imperial China.

CPC News [14].

Alpermann and Zhan [3].

For instance, recent studies on increasing autonomy on the part of business associations [86, 88] suggest that the party state’s ‘shadow of hierarchy’ has been forced to retract. We would instead claim that the increased leeway of business associations to negotiate with the state on private sector reform, including more access to investment in SOEs, does actually not bespeak the state’s waning control of business associations. It is instead evidence of a gradually changing mode from direct ‘command and control’ to more horizontal steering, which may be called ‘steering by deliberation in the shadow of hierarchy’ (see also [76]).

Greenhalgh and Winckler [25].

The distinction between ‘hard’ and ‘soft’ steering must be properly defined. ‘Soft steering’ may include discursive practices of steering which have a long tradition in Chinese governance. For a theoretical foundation of ‘soft steering’ see Göhler et al. [23].

This refers, for instance, to the new campaign mode subjecting every high-ranking cadre to continuous ideological training or intensified supervision of party and government bureaus by the Party’s Discipline and Inspection Commission.

Steering blockades may stem from resistance on the part of collective and individual social actors against local policy implementation or from their dissatisfaction with certain shirking strategies of local governments like bureaucratic inertia or formalism. At the same time, those blockades can be further intensified by collaborative efforts of local party state and societal actors (corporate or collective) to counter upper level demands.

Here, we differ from Heilmann [37].

Streeck [79].