The allocation mechanism of authority and accountability in Chinese government hierarchies: A case study of the “Earmarked Project System”

Based on basic organizational theories, this paper revisits studies about intergovernmental relationships in China and proposes a systematic explanation framework from the perspective of the allocation mechanism of authority and accountability. The institutional feature of government organization is unfolded from two dimensions, authority and accountability. In China, the authority and accountability of different hierarchies are strongly related but not clearly defined. Theories such as “Federalism, Chinese Style” are the basis for different conditions. The trade-off between control and incentive, specification, and coordination should be given more consideration with optimal analyses. As the main accountability practicing mechanism, Cadre management from upper-layer hierarchies greatly affects the allocation mechanism between different hierarchies. However, the tournament theory about promotion lacks vision, empirical bases, and explanation power. This paper studies the project system from an organizational perspective and indicates that the subcontracting form differs from regional decentralization, which accentuates interdepartmental competition and expansion.

The institutional structure of the state deeply influence individual behavior (North 1981; Kornai 1992; Evans 1995; McGuire and Olson 1996). One of the most intensively studied topics is the inner structure of governmental organizations. Theories relate these intergovernmental relationships to the overall behavior and performance of governments. Our approach follows Max Weber’s concept of hierarchy. In organizational economics, markets and hierarchies are the two polar modes. They contrast to each other in autonomy (Simon 1951), property rights (Alchian and Demsetz 1972; Grossman and Hart 1986), and incentive intensity (Williamson 1991). Scholars have pointed out that hierarchy has three main characteristics: a low level of autonomy, employer-owned property, and low-powered incentives; these three features are complementary (Williamson 1991; Holmstrom and Milgrom 1994). Moreover, Williamson (1991) locates hybrid modes that differ from both markets and hierarchies, being more elastic than the former but more legalistic than the latter. However, the modes mentioned above are based on the analysis of economic firms and the relationships between these firms. When talking about the intergovernmental relationships of China, the focus is mainly on how the definition and operation of intergovernmental authority—exercising power with legitimacy—changed through the years, and how this change engendered the phenomenon of rapid growth while combining with social and environmental problems since China’s reforms. Economists and sociologists have attempted to answer these questions by discussing the boundaries of different organizations.

From the perspective of authority, although local governments have limited autonomy in a unitary government like China, it is fundamentally different from federalization governments. In reality, upper-tier governments have strong incentives and strong control over lower-tier governments among hierarchies. However, neither Federalism, Chinese Style nor the administrative subcontracting model has captured this feature. Theories should be consistently revised to fit the reality, and the complicity of reality is a chance to test and update theories. Therefore, selectively applying materials to mechanically fit theories is like cutting the foot to fit the shoe (Prendergast 1999).

On the one hand, from the perspective of authority, Montinola et al. (1995) point out that fiscal federalism empower local governments to resist decisions from the central government, providing the case that in 1989, Guangdong successfully resisted central tax reform, but they ignore the change of personnel and tax-share reform (Cai and Treisman 2006). In addition, previous literature mainly focused on relationships among blocks (kuaikuai), while there has been little treatment of the problems of branches (tiaotiao) beyond sociological studies. Since the mid-1990s, the mechanism of authority and accountability between Chinese government hierarchies has changed dramatically, with the increased authority of branches and changes in relationship among blocks. Unfortunately, theoretical and empirical research on the current intergovernmental relationship is limited.

From the perspective of accountability, there is no independent third party to monitor and supervise the exercise of authority in the governmental organization. How can governments guarantee that their power is being exercised legally and effectively? In developing countries, the problem of accountability is as significant as the problems of authority and influences the efficiency of governance (Bardhan 2002). There are two basic views on the problem of governmental accountability. First, from the view of supervision and monitoring from the upper level to the lower level of government, this research focuses on the efficiency of these systems and takes it as a factor of “state capacity” (Huang 1995; Edin 2003; Wang and Hu 1993). Some scholars use the promotion-tournament theory to explain it. In reality, however, personnel promotion is one of the most shrouded realms in Chinese government, and this obscurity limiting the credibility and availability of the data increases the difficulty of measurement, which caused wide divergence in the conclusions of previous research works (Li and Zhou 2005; Tao et al. 2010; Shih et al. 2012; Yao and Zhang 2013; Yang and Zheng 2013). Second, research on limitations of accountability reveal the phenomenon of soft budget constraints, clientelism, soft-risk constraints, adverse soft constraints, property rights of relationship, and conspiracy (Kornai 1992; Walder 1986; Liu 2005; Zhou 2012, 2008; Shi 2015). Although existing analyses in sociology have provided several models, the problem of government organization is formidably complex and new modes of analysis are still needed.

Along these discussions of authority and accountability, we have three main objectives in this paper. First, we explore different types of authority and accountability in Chinese hierarchies and aim to demonstrate the features of intergovernmental relationships. Revealing political imaginations in governmental studies, we can determine the same mechanism in different concepts and analyses. For example, in the classical federalism model, government’s main role is to provide public goods (Tiebout 1956; Oates 1999), while in fiscal federalism, the main role of government is preserving the market (Weingast 1995). Although they are both in favor of decentralization, the analyses had different logical approaches. Additionally, some research assumes that the main role of government is tax extraction (Levi 1988), which fosters analyses of the political risk (Cao 2011; Zhou 2014). What then is the institutional environment and internal operation logic in the Chinese government? How do these affect the mechanism of authority and accountability in China?

The second objective of this paper is to survey formal theoretical literature on intergovernmental relationships and advance the discussion with a new framework of analysis. On the one hand, the subcontracting system in Chinese government hierarchies provides the opportunity for competition among contractors, through which the upper-tier government controls the lower-tier governments. On the other hand, different from federal government and unitary governments, the structure of intergovernmental authority and accountability is not transparent and symmetrical, which invites the coexistence of a contractual system and a hierarchy system with different management approaches of control, coordination, and incentive forms.

Finally, to illustrate our arguments, we examine the earmarked-project system. Different from the hierarchy system, the earmarked-project system is a newly developed organizational mechanism. Shi (2015) points out three key features of the project system: temporality, goal-oriented operation, and flexibility. This paper explores how the earmarked-project system operates in China, and what kind of intergovernmental relationship it generates and aims to analyze the phenomenon of the bureaucratized-project system.

Government organizations differ from corporations in their multiplicity of goals, lack of performance comparison, heterogeneity of citizens’ tastes, and dispersed ownership (Tirole 1994). In simple-task organizations, the goal of incentives is to balance the intensity of incentives and the degree of risk, while in organizations with multidimensionality of goals, the design of the incentive system is concerned with arranging different tasks and managing workers’ attention (Holmstrom and Milgrom 1991). Thus, entirely applying the theory of firm to explain government organization may be greatly problematic.

Researches have addressed the different goals of government, namely maximizing social welfare (Musgrave 1959), maximizing department budget (Niskanen 1971), maximizing revenue (Brennan and Buchanan 1980; Levi 1988), and maximizing monopoly rent (Acemoglu 2002). To date, there are basically two views of government goals. First, governments have their own interests (Evans et al. 1985), which may be distinct from their principals—the people they govern. Three models of governments, namely the invisible-hand model, the helping-hand model, and the grabbing-hand model, are demonstrated in different legal environment and regulatory structures (Frye and Shleifer 1997). Second, the government is an agency of the people, at least in name, while the regulator of their owners (Wilson 1989). The government extracts from citizens economically, physically, and mentally, and in return, they receive the products of personal security, profit safety, social equality, public order, and infrastructures (North 1981). However, the social arrangements for distribution of income and productivity of public goods are determined by the institutional environment and space for citizens in collective actions, a representation system, and the power of exit and voice (Hirschman 1970; McGuire and Olson 1996). The characteristics of government organizations lie in two dimensions: authority and accountability. The former includes governmental extraction and production, while the latter includes institutional capacity and a structure by which citizens can supervise their governments directly and indirectly.

Extraction and production: authority of the government

Extraction by the government

The depth of government extraction can be divided into three scales. In the small scale, the government exercises its discretion in control of legislation and justice, and its extraction capacity is limited by regulations and laws. In the medium scale, although the budgetary system is established through non-budgetary revenues and extra-budgetary revenues,^{Footnote 1} the government can still collect implicit taxes, issue currency, deploy loans, and deposits beyond the legal framework. As a result, substantially independent bureaucrats pursue their own agendas and extract rent (Shleifer and Vishny 1993; Frye and Shleifer 1997). In the large scale, governments extract not only by taxation but also by revenues and human capital. In this model, the government is a super company, while citizens scarcely own residual claims and residual control rights to invest in human resource and economics. Consequently, this type of government impedes the investments in private sectors and reduces the extraction capacity of the government in long term (Hayek 1945) (see Table 1).

Production of government

On the issues of governmental production, the larger the production scale the more controversial the issue. An extreme model is the small government model in which Adam Smith observed the government acts as a “night watchman” (Smith 2003/1776). The so-called Washington consensus and neoliberalism were deeply influenced by these arguments of small government. Another extreme model is called developmental government, which follows the success of “East Asia’s Miracle”. Under this framework, the government has embedded autonomy, which means that bureaucratic autonomy and societal embeddedness are joined together (Evans 1995). Small governments leave allocative decisions to the market, and the market provides positive incentives that reward economic success while government regulation plays a limited role. Also, the market provides negative incentives in terms of punishing economic failure. In this model, the government does not impose soft budget constraints on corporations (Weingast 1995; Kornai et al. 2003). In contrast, in the large government framework, the government is intimately involved in promoting private economic activities. It selectively supports some kinds of organizations and pursues industrial policies and control over the market. In some cases, the government becomes the headquarters of the market. Scholars have explored how the enhanced financial incentives of local governments strengthen economic performance while avoiding rent seeking and corruption (Oi 1992; Walder 1995).

Monitoring and accountability of the government

Focusing on how governments exercise authority effectively, Bardhan (2002) leads us to identify who monitors, and how elected officials carry out their responsibilities. Strikes, appeals, and so on are negative instruments of governmental accountability, while elections and commentary can be deployed as positive instruments.

The mechanisms of accountability include first, direct participation of and voting by the public, and second, indirect deployment of institutional devices both formally and informally. In reality, we see a combination and coexistence of these two mechanisms. The problem of asymmetric information exists throughout government hierarchies, among government officials and policy makers, politicians, and bureaucrats. Thus, monitoring by government agencies alone is not costless. This asymmetric information problem can be solved by direct participation of the people and their attention to administrative procedures. However, as high-level hierarchies impose sanctions, the government in a unitary authority structure may suffer from the political costs of wrongdoing although it can provide the benefits of administrative efficiency. Additionally, the problem in the case of China is the absence of a public monitoring system. This absence will trigger aggrieved citizens to apply methods near the boundary of authorized regulations, such as collective actions and leapfrog petitions (O’ Brien 1996). These conflicts may serve as an impediment to social development and bring about unorganized threats to the state due to the lack of institutional capacity to settle conflicts within an authorized framework (Zhou 1993).

In regard to theories of the modern state, the relationship between the people and the government is described as a principal-agency model. In China, governmental organizations are organized as a unitary form in a simple three-tier hierarchy, with the central government at the top, local governments in the middle, and enterprises or citizens at the bottom. The central government issues contracts to local governments, then local governments implement the contracts. The problem is there is no effective monitoring system by which the public can supervise the behavior of local governments. Thus, on the one hand, the central government empowers local governments with limited autonomy and appropriate incentives to utilize local advantages to fulfill the goals set by the upper-tier government. However, it may cause incentive distorting because of abuse of power at the central level (Baker et al. 2002). On the other hand, besides the authority of assigning tasks to local governments, the central government is also responsible for monitoring and supervising the lower-level governments. To sum up, these are two dimensions of state capacity. As is usually, the case when a subject draws advocates from sharply different standpoints, different scholars have reached different conclusions regarding state capacity (Wang and Hu 1993; Wong 1992; Edin 2003).

The fundamental political dilemma of decentralization and centralization is decentralization may balance the authorities of different bureaus in the government and is effective in hardening budget constraints and uniting workers to commit to organizations’ long-term goals as a whole. On the other hand, the government consists of a large number of substantially independent bureaucrats pursuing their own agendas, including taking bribes. In extreme cases, the government becomes so disorganized that it loses its ability to ensure laws and provide basic legal protections (Frye and Shleifer 1997). Why does decentralization galvanize the “tragedy of the commons” (Harding 1968) in some cases instead of establishing balance of authority? To answer this question, we need to clarify the allocation mechanism of authority and accountability in governmental organizations.

The central-local framework

In China, the intergovernmental mechanism is neither federalism nor extreme centralism as in the Soviet Union. Our analysis suggests two main facts. First, in China, the relationship between the central and local governments neither fits market-preserving federalism in which central governments have the primary regulatory responsibility for the economy, the lower governments face a hard budget constraint, and the common market is ensured by the central government (Weingast 1995), nor a clear division structure of accountability and authority between the central government and local governments (Montinola et al. 1995). In reality, local governments have only limited authority (the right to execute), while the central government maintains veto authority (effective control over decisions) (Aghion and Tirole 1997). Fundamentally speaking, local governments are self-interested and possess limited autonomy.

Second, historically and ideologically, local governments in China gained more autonomy compared with those under the extremely centralized government structure of the Soviet Union. China is closer to the M-form hierarchy, while the Soviet Union to U-form hierarchy. From the perspective of incentives, under China’s hierarchical structure competition among local governments provides higher-powered incentives for regional economic development and hardened budget constraints (Maskin et al. 2000; Zhou 2007). From the perspective of coordination, functional matching between hierarchies are vital in an M-form hierarchy, and the flexibility of this structure benefits the innovative experiments that otherwise have little chance to survive in the market. Yet, this structure has been less efficient in utilizing scale economies (Qian et al. 2006). The role of “branches” in China was less significant in the 1980s yet thrived after the tax-share reform in the 1990s.

Features of the allocation of authority and accountability in Chinese government hierarchies

What is the allocation mechanism of intergovernmental authority and accountability in China if neither federalism nor extreme centralism can illustrate the reality in China? Some scholars use the theories of Federalism, Chinese Style and administrative subcontracts to answer this question. However, focusing on the relationship between “blocks” (kuaikuai), these ideal types do not capture the fundamental characteristics of the relationships among the hierarchies in Chinese governments (Qian and Xu 1993; Zhou 2014). Zhou and Lian (2012) point out the interaction and evolution of these two models in terms of residual control but do not analyze the complicated allocation structure of authority and accountability in Chinese hierarchies.

First, from the aspect of clarity, for federalist states, there is a clear boundary of authority and accountability between the central government and local governments with local citizens’ voting and monitoring rights (Tiebout 1956; Xu 2011). In mature centralism states, a clear clarification of accountability is ascribed to a clear clarification of authority among different hierarchies. However, in China, the structure of authority is not lucid, institutionally or executively (Lou 2013; Li 2010). This obscurity stems from the organizational structure of the Chinese government. It is more than a technical problem and, as a consequence, hard to rectify. “The Organic Law of the People’s Republic of China on the People’s Congresses and People’s Governments at All Local Levels” stipulates that local governments should execute resolutions and decisions of the People’s Congress and the Standing Committee of the People’s Congress at the local level, as well as decisions and orders from the upper-tier government. This stipulation generates a dual role for local governments. The central government can participate in the decision-making progress of local governments indirectly through appointments, contracts, assessment through “blocks”, and direct orders and guides from “branches”. The clarification of authority and accountability for each block and branch is not clear, causing disorder in the hierarchy system. This has two results. First, the central government puts pressure on and shifts responsibilities to local governments, which will inevitably bring about pressure mechanisms, adverse soft budget constraints, and soft risk constraints. Second, low-tier governments will seek extra opportunities and ask for protection from the upper-tier government and may even conspire or selectively execute decisions to avoid punishment (Kornai 1992; Walder 1986; Rong 1998; O'Brien and Li 1999; Zhou 2005a; 2008; Liu 2005) (see Table 2).

Second, from the aspect of multiplicity, a multitask organization should theoretically balance extraction and production institutionally. However, policies of ideology, movement of iron and steel before the Chinese reform, and policies of taking economic growth as the central task during the 1980s serve as supporting examples of the unbalanced strategy that Hirschman (1998) describes. China’s M-form hierarchy is effective in reducing coordination costs of the central government (Qian and Xu 1993; Qian et al. 2006), but it is harder for the central government to supervise and coordinate through bureaucratic structure and administrative procedures (Wong 1992). On the other hand, to reduce these negative effects, the Chinese government relies more on the branches (tiaotiao). However, the lack of public participation and satisfactory institutional design of the branches causes competition in interests and organization expansion (Shi 2015). This leads to territorialism, departmentalism, and fragmentation of authoritarianism, and even worse, frustration of the national market (Shen and Dai 1990).

Different types of contracts in the extraction and production of the government

Contracts for economic growth

Since the economic reform, China has generated significant growth by adopting policies to “loosen and stimulate the economy” (fangkaigaohuo) and changed to a fiscal revenue sharing system from the centralized fiscal system. Although schemes vary across provinces, provincial governments enter into contracts with the central government on the total amount or share of taxes and profits revenue to be remitted for the next several years, and the provincial governments will keep the rest. By the late 1980s, the provinces were keeping almost all of the marginal revenue growth (Qian and Xu 1993). Furthermore, the extra-budgetary revenues of local governments had ballooned. In 1992, local extra-budgetary revenues went up to 110 percentage of local budgetary revenue (Montinola et al. 1995). Thus, although a nominal tax rate was set by the central government, local governments controlled the real tax rate through policies of tax reduction and tax exemption for local enterprises. In addition, regional credit was under the control of local governments before the banking system recentralized in 1998 (Oi 1992; Wong 1992; Walder 1995).^{Footnote 2}

In terms of organizational economic theories, scholars use theories of second-generation fiscal federalism when discussing China’s intergovernmental relationships. Similar to the function of horizontal decentralization and property rights protection, in second-generation fiscal federalism, local governments have become almost residual claimants to local revenue and have incentives to maximize local revenues by protecting local enterprises, even when confronted by the upper-tier government regarding their over-extraction (Montinola et al. 1995; Qian and Weingast 1997). In regard to hierarchical structure, the lower the level of the government the greater the possibility that local fiscal revenue is highly dependent on the profits of local enterprises, and the smaller the possibility that the local government will invest in state-owned enterprises and collective enterprises that are less profitable in comparison. Moreover, with reduction of the centrally controlled capital flows, opportunity costs of paternalism for local governments have increased because of regional competition. This horizontal competition hardened budget constraints (Qian and Roland 1998). Although the budget constraints have not completely hardened, it has suppressed the “grabbing hand” of the government (Jin et al. 2005).

Nevertheless, the theory is problematic in the following two aspects. First, it does not provide a comprehensive analysis of the relationship between local governments and business enterprises and thus lacks a sound basis to deal with the problem of the relationship between local governments and the market. Scholars have demonstrated how local governments utilize the positive incentive effect of the market by clarifying property rights and have explored the issue of how local governments developed the private sector and promoted privatization reform (Jin et al. 2005). However, according to this fiscal incentive framework, local governments and enterprises become a community of shared interests in government-oriented reform, and their relationship becomes more opaque, injurious to the market’s negative incentives and to market positive incentives in long-term.^{Footnote 3} Second, although researchers acknowledge the outgrowth of Federalism, Chinese Style, they have not pointed out how these internal negative results undermine the efficiency of fiscal federalism or explored how these problems can be solved. They merely emphasize the five preliminary requirements of market-preserving federalism without capturing the substantial features of the Chinese government structure.

Contracts for extraction

Levi (1988) states that the level of government extraction capacity is restrained by three conditions: the capacity of the government to control social resources and relative bargaining power; the capacity of the government to formulate, coordinate, and execute extraction contacts; and finally, the governmental expectations for the future, in other words, the withdrawal rate. In short, governments are expected to extract resources from their subjects without annoying them (Tirole 1994). Kiser (1994) explores two types of tax structures—a hierarchical form employing salaried state officials and a market form (tax farming). These frameworks of extraction have triggered controversy.

Contracts for public goods provision

As Hayek (1945) stresses controlling scale economics and externality, decentralization of authority has the benefits of more efficient use of dispersed local knowledge possessed by the local government. Moreover, assuming unrestricted migration and abundant regions, regional competition functions as market competition to provide more satisfactory public services (Tiebout 1956). In contrast, in centralized hierarchies collecting, clarifying, and disseminating information about public good preference is costly, and along the information, channel toward the central government potential information leakage will be counterproductive to governmental decision-making progress, causing loss of control (Williamson 1967). Qian’s study (1994) indicates a greater loss of control in a massive and complicated hierarchy system. However, this assumption of regional competition does not apply to the character of the Chinese hierarchy structure. Admittedly, as some scholars have demonstrated, fiscal competition may hinder the process of public goods in quantity and quality, compared with fiscal centralization (Qian and Roland 1998). They also point out that earmarked funds transferred from the central to local governments can play a favorable role.

Local fiscal expenditure planning is a major function of local governments, yet the decisions on public good provision are selectively “up-side-down” (Gao 2008). On the one hand, China has generated significant growth, even compared with developed countries, in its communication system, industrial estate, and urban infrastructure, which were closely related to economic performance (Zhang et al. 2007). On the other hand, selective lack of development in more fundamental areas like public health, education, and rural infrastructure was detrimental to China’s reform. This situation results from local governments’ rational selection to achieve economic growth in a multicontract system and called forth central institutional reform in the provision of public goods.

Contracts for social equality and regional balance planning

The regional gap and the urban-rural gap have widened since the economic reform due to the intentionally unbalanced policies of the regional contractual system (Wong 1991). Moreover, although the contractual system is easier to set up strong incentive, it is inferior to the centralization system in solving the externality. In addition, centralization can prevent conspiracy among local contractors (Tirole 1986).

Changes and adjustments

Although the contractual system is result oriented, the restriction of contractor’s behavior is limited. While the hierarchy system is process oriented, governments supervise and monitor the entire process. In the 1980s, the rate of central expenditure and revenue to total expenditure and revenue of China was rather low. Some scholars point out that it has reached the baseline of decentralization (Wang and Hu 1993). Since 1994 and especially since 2004, the Chinese government has redesigned policies not only in extraction institutions (separation of state and local taxation) and public goods provision but also in economic growth mode, social order, and public security in which the hierarchy system has played a more significant role. The emergence of the new mobilization form of “leading group” (lingdao xiaozu) and “project headquarters” (zhihuibu), along with the new organizational form of the “province-administrated county” and the “fiscal-administrated county”, has brought about an alteration in the intergovernmental authority and accountability mechanism.

Furthermore, the mechanism of authority and accountability varies among regions and government organizations. Enjoying preferential policies, some regions are assigned special economic or political functions in the reform while others not (Yang 1990). For instance, in economic terms, the southeastern coastal regions, which enjoyed superior factor endowments, are empowered with more institutional flexibility by the central government compared to the interior. Although regional policies may be unconstitutional in some other countries, they are a crucial part of the Chinese institutional structure and influence the contractual systems and authority-accountability mechanism in different regions.

The efficiency of authority lies in accountability. Two modes monitor and assess the output. First, the process-oriented mechanism is implemented when monitoring is observable or cooperation is valued over competition. This accountability mode is based on individual behavior in the decision-making progress, yet the costs of monitoring remain high. The second method is the outcome-oriented mechanism, which can be divided into absolute-performance and relative-performance evaluations. Given that the latter can alter incentive levels in different situations, the relative performance evaluation gains more flexibility. However, the agents share risks while enjoying profits and risk aversion hastens incentive distorting, which must be balanced (Lazear and Rosen 1981).

The Chinese government system combines these two mechanisms. By comparing relative performance of contractors, the central government controls personnel incentives and material incentives due to economic and social development policies that mainly depend on the contractual system. Meanwhile, competition among lower-tier governments lends credibility to the upper-tier government. The upper-tier government thus remains highly authoritative (Bendor et al. 1987). In addition, governance by procedural rules and regulations became increasingly significant in the Chinese government system. As the result of the abovementioned features of Chinese governance, one way to solve this dichotomy of progress and output of policy monitoring is by factions (Dittmer and Wu 1995). Faction members share group solidarity in ideology and general ideas. Factions can be used to promote different sets of goals. The most prominent areas are security, material interests, and ideological and policy commitments. The consensus in factions provides stability, compatibility, and flexibility, and it partly solves the authorization problem of the contractual system (Whyte 1973).^{Footnote 4} This phenomenon challenges the theories of developmental government and rank-order tournaments.

Promotion: economic-oriented tournaments?

In multitask organizations, the incentives are “low-powered” (Holmstrom and Milgrom 1991). “High-powered” relative performance incentives are suitable for lower-level employees with simple tasks rather than higher-level employees who have multitasks. Many research works focus on political elites in the Chinese government; although reasonable, this focus leaves many problems unaddressed.

Maskin et al. (2000) has illustrated that the economic status of one province has a positive correlation with province’s status in the Central Committee, and regional competition determines the status. Some scholars then applied this theory to individual officials. Based on the theories of tournaments and yardstick competition (Lazear and Rosen 1981; Shleifer 1985), they argue that GDP-growth-order tournaments existed in the promotion of local officials (Zhou 2004, 2007; Li and Zhou 2005; Chen et al. 2005). The premises of promotion tournaments in China are central control of the personnel system, preventing conspiracy, measurable indexes for tournaments, controlling power over local economic by local government leaders and individual performance assessment, effective promotion commitment, and analogy among regions under deep M-form.^{Footnote 5} Scholars carried out statistical research on these topics (Li and Zhou 2005; Zhou 2005b; Chen et al. 2005; Du et al. 2012).

In contrast, Tao et al. (2010) argues that Li and Zhou (2005) have not set standards for measurement and finds that local economic growth has no relevance to the promotion of officials. Zhou (2004) argues that political tournaments will bring about local protectionism and barricade regional cooperation but tests of this argument have not brought statistically significant results (Chen et al. 2005). In addition, some scholars consider economic scale as a more fundamental element in promotion incentives (Bo 1996; Yang and Zheng 2013; Yao and Zhang 2013).

In this literature on promotion incentives in the Chinese government, the problem of endogeneity has not been addressed seriously or formally analyzed. What is significant in the promotion of government officials if they are sent to a more economically developed region as a result rather than a reason? Some researchers turned to non-economic factors like factional ties. Shih et al. (2012) explores the advancement of Central Committee members in China but fails to find evidence that exceptional economic growth contributes to high ranking in the party hierarchy. Their findings suggest that factional ties with various top leaders boost the chance of climbing into the upper echelons of the CCP.

Implications

These discussions about political tournaments have not systematically analyzed the mechanism of promotion and incentive in Chinese government hierarchies with respect to authority and accountability. First, from the perspective of designing an incentive system, it is more than a system of indexes. Diversity of authority and accountability will cause incentive distorting even in a “big government” like China, if only by some so-called objective indexes and ranks. Rational decision makers see this problem. Introducing subjective performance measures and explicit incentive contracts can mitigate incentive distortions caused by imperfect objective measures. In some circumstances, objective and subjective measures are complements (Baker et al. 1994). From the management perspective, it is suspicious that a candidate who can fill the current position would also be eligible for a higher position. This is the reason why members of organizations spend considerable time, effort, and ingenuity to influence decision makers (Rosen 1982; Milgrom and Roberts 1988). Furthermore, the position for officials of each level in the government hierarchy system after promotion may not be economically oriented. Finally, loyalty is a fundamental aspect in institutional system designing for any organization; it is essential in the incentive system.

Second, analyzing the nature of authority can also help us explore this issue. Supporters of the tournament model point out that the central government holds centralized control of personnel promotion. Now that the central government has this centralized authority, why would it restrain itself by some fixed indexes? Blau (1963) presents the relationship between authority and indexes: on the one hand, decision makers use numbers as a way to offer incentives and a way to increase their credibility and avoid disputes. On the other hand, these upper-tier officials are not manipulated by numbers. In addition, the tournament model addresses the idea that promotion is an effective commitment because it will not occupy extra resources, and the system needs to eventually find someone who is qualified for the position. However, these arguments lack awareness of the uncertainty that exists in the process of government promotion. The perspective of political power is essential in theories of government organization (March 1962; Rajan and Zingales 1998); otherwise, the phenomenon that Montinola presents cannot be explained: because of fiscal incentives, many local officials in developed regions do not pursue advancement because they have more autonomy and possibility of monetary profit in their original position (Montinola et al. 1995). One explanation of this phenomenon is the diversity of individuals, indicating that officials in bureaucracy have different goals (Downs 1967). Another explanation is power; Zhou (2016) uses the concept of stratified mobility to explore this problem.

Previous studies of the project system mainly focused on the hierarchical operation (She and Chen 2011; Zhou 2012a, 2012b; Qu 2012; Huang et al. 2014). Scholars seldom approach the problem by applying organizational theories (Shi 2015). To illustrate our arguments, we reconsider the project system from the organizational perspective.

The earmarked project system in a multiple governmental authority and accountability system

After the tax-sharing reform, central fiscal revenue was transferred down to local governments in three forms: tax rebates, general-purpose transfer payments, and special purpose transfer payments.^{Footnote 6} The three are designed for different goals and correspondingly have different intergovernmental authority-accountability features. First, the form of tax rebates is designed to encourage local tax extraction, and the main goal of tax rebates is to balance resistance to the economic reform, especially from developed regions in eastern China. Thus, tax rebates have an unfair effect. In the early years of the reform, tax rebates increased to almost 80% of government expenditures, while it remained at the rate of 50% before 2000, and the majority went to developed regions. Second, the general-purpose transfer payment was designed to balance expenditure gaps among regions, and it contributed the political effect of achieving social equality and maintaining social stability (Treisman 1999; Wang 2002). The majority of these funds flowed to the western or frontier areas, minority regions, rural areas, and impoverished areas. Third, the form of special transfer payments is unique in its purpose, special distribution mechanism, and practical results. According to economic analyses, special transfer payments are a modification for externality through subsidies to regional marginal revenue and altering relative prices to solve the financial shortage of public goods, especially goods that have strong externality. However, scholars have pointed out that the function of special transfer payments works by political logic (Dixit and Londregan 1998). The profit of special transfer payments belongs to specific regions while the national fiscal system bears the costs; thus, local government may over-apply for special transfer payments. The distributors of special transfer payments make decisions politically rather than economically, so there are a considerable number of inefficient projects (Weingast et al. 1981).

In China, the ratio of central fiscal expenditures to government expenditures was rather low, fluctuating at the rate of 20%, which was one of the lowest in the world (Li 2010; Lou 2013). In 2014, the national general public budget expenditure was around 15.18 trillion RMB. The central-level general public budget expenditure was about 2.26 trillion RMB, accounting for 14.87%, and tax rebates and transfer payments to local government was approximately 5.16 trillion RMB, accounting for 33.99%, which accounted for 48.86% of the annual central level budget expenditure as a whole (see Table 3). According to these numbers, China is not unique. In other words, the particularity of Chinese fiscal expenditure system lies in its fiscal structure in which the rate of indirect expenditure from the central government is rather high. This led to ambiguity in intergovernmental authority-accountability relationships.

As illustrated in the fiscal expenditure categories, national defense expenditures and science and technology development expenditures were the responsibility of the central governments. Except for expenditures in public security, the expenditures in other categories went downward, flowing to local governments through earmarked transfer payments. The expenditures in health care, social security, and employment were transferred down to local governments, while their counterparts in other countries like the USA are direct expenditures of the central government (see Fig. 1).

Expenditure structure of central fiscal budget of China (2014)

Full size image

The rate of tax rebates to annual central expenditures declined over the years, and it was the highest among the three types of central expenditures until 2005 (see Table 4 and Fig. 2). Tsui (2005) observes that in those years, fiscal inequality among regions was aggravated by the tax rebates and transfer payments. Then, from 2008 to 2010, special transfer payments occupied the highest proportion, reaching 43.6% in 2010. After this period, general transfer payments occupied the greatest proportion, and in 2014, the rate went up to 53.4%. We can thus almost conclude that intergovernmental relationships have changed since 2011 because the superficiality of these numbers cast doubt on this conclusion. The central government earmarked 262.22 million RMB of central general expenditures in 2011. In the two types of central transfer payments, up to 61.7% of the funds were earmarked, which means local governments had no autonomy to redesign the use of these funds. In 2012, the number went up to 64.42%. In 2014, about 918 billion RMB (about 19%) in general transfer payments were earmarked (Liu 2014). Additionally, some basic expenditures of the central government budget were also earmarked as project funds.

The structrue of central expenditure to local level (2003–2014)

Full size image

Organizational features of the earmarked project system

In China, general transfer payments and special transfer payments are two different contractual types. The former flows down through blocks, which are based on a standard structure of government hierarchies, while the latter is transferred down through branches that are decentralized and flexible. There are two characteristics of the contractual system: first, the compulsory homologous structures (Zhu and Zhang 2005) in different levels of hierarchies made it a compensation of central direct expenditures; second, in the official definition, special transfer payments are used for projects commissioned by the central government and central-local collective projects, and they share the features of the contractual system and regional competition.

First, from the view of horizontal relationships, the earmarked project system increases competition among central ministries. The balance between government organizations is essential in hardening budget constraints and increasing governmental decision quality. However, in China, the absence of an efficient supervision mechanism leads to decentralized and unorganized expansion of these departments. Every department aims to have more projects and expand their authority in projects. The National Audit Office of PRC addresses this phenomenon as “fragmentation and departmentalization” (Liu 2014). As an example, projects in agriculture, forestry, and water resources are under the control of the National Development and Reform Commission, Ministry of Finance, Ministry of Land and Resources, Ministry of Science and Technology, Ministry of Agriculture, Ministry of Water Resources, State Forestry Administration, Office of State Flood Control and Drought Relief Headquarters, China Meteorological Administration, The State Council Leading Group Office of Poverty Alleviation and Development, State Administration of Grain, Federation of Supply and Marketing Cooperatives, Ministry of Education, National Health and Family Planning Commission, Ministry of Culture, and Ministry of Civil Affairs at the central level. These ministries create special departments unrelated to their original functions for these projects. Thus, ministries have no incentive to monitor and supervise themselves by their own authority. To sum up, the Chinese government is neither an M-form organization because of centralized authority among branches nor a U-form organization because of the disordered division, even antidivision, among central ministries.

Second, from the view of vertical relationships of government organizations, the earmarked project system causes interdependency in the multilevel hierarchies. Different from central-local relationships in other countries (Dixit and Londregan 1998), governments at the provincial and county level rely on the upper-tier government to expand project funds, while the upper-tier government relies on the local governments to strive for more fiscal budget from projects and gain more authority in these projects. In addition, the indistinct authority and accountability relationship among projects gives rise to a symbiotic relationship between upper and lower hierarchies. Conspiracy between upper and lower governments is one result. Therefore, in the monitoring and supervision of projects, even the problems of misuse of funds and low quality of products will not be punished by upper-tier governments. This unsubstantial supervision softens budget constraints. Furthermore, to achieve better performance, central and local governments prefer to highlight certain projects. Despite the lack of perfect competition, the preference for particular projects galvanizes long-term factions between hierarchies, establishing an informal relationship within formal hierarchies. This leads to the “Matthew Effect” (She and Chen 2011). This relationship among hierarchies is a way to soften budget constraints rather than the result of it.

Government organization is an interdisciplinary topic, and many theories contribute to the discussion, especially in economics and sociology. These years have witnessed an increased interaction among different subjects. For instance, economists have become more valued in the research in sociology, political science, and psychology. Sociological studies should focus on rising new concepts rather than answering the questions that previous theories have not explained.

The studies on the “campaign government” of China epitomized these phenomena. Scholars give different definitions and explanations on the basis of different literature (Feng 2011; Zhou 2012; Zhou 2014). However, they do not logically point out the preconditions of their mechanism and thus have lost connection with basic theories. Why did the Chinese government operate more by means of campaigns even compared with socialist counterparts, such as the traditional Soviet Union? Why are national campaigns in China designed to achieve one goal rather than several goals? Why is this campaign mechanism more flat than other mechanisms? We can try to answer these questions with three hypotheses. First, the prevailing theories on campaigns indicate the weakness of the traditional hierarchy system. Compared with the centralized U-form hierarchy of the Soviet Union, China’s M-form hierarchy remains decentralized with a lower level of specialization and professionalization. These features of government structure may give rise to national campaigns. Second, organizations have different goals and tasks, and they go into different orders in different levels of hierarchies. A national campaign mandatorily unifies the orders nationwide. Thus, on the basis of the first hypothesis, governments under pressure to catch up to developed countries tend to deploy this national campaign. The movements of the iron and steel industry to catch up with Britain and America help illustrate this phenomenon. Third, in certain conditions, the hierarchy chain and range of control are substitutable (Qian 1994; Hart and Moore 2005). Thus, campaign mechanism shortens hierarchy chains with direct communication from upper to lower governments, and central control is widened. This is particularly applicable for some urgent tasks and projects.

The performance and legality of national organization is based on the capacity of the government to institutionalize change in society. Kornai’s (1992) concepts such as soft budget constraints, investment hunger, and mandatory increase remain convincing and applicable with slight changes. This is a warning for our social scientists: how will the theories remain popular in these years of intense change? Thus, to illustrate the mechanism of realities, clarifying the boundaries is more significant than updating concepts.

1. 1.

   Extra-budgetary revenues and non-budgetary revenues are not in the category of budget revenue in China. Although the Chinese government issued “Decisions on Extra-budgetary Revenues Regulations” in 1996, clauses in these regulations were not detailed or specified. In addition, there are no formal regulations for non-budgetary revenue, making it difficult to supervise and control.

2. 2.

   As far as it is concerned, the criticism by Cai and Treisman (2006) are weakened. They argue that although the marginal fiscal retention rate of government has increased, the proportion of fiscal revenue to national income has declined. By using simple multiplication, they came to the conclusion that the rate at which the fiscal revenue of local government can control to the whole local fiscal revenue has dwindled.

3. 3.

   Some scholars even view a township or village as a corporation, and the government of the township as the board of directors and the management team of the corporation. Under fiscal incentives, local governments support the reproduction of enterprises instead of corruption with revenue they gain from these enterprises (Oi 1992). However, the effect of this selective paternalism is unclear and differs by region.

4. 4.

   Vertical protection is pivotal in the inner structure of factions (Walder 1986). This protection system from the upper-tier in China is contrary to rule-based governance.

5. 5.

   It is difficult to fulfill these premises. The term “tournaments” was originally used in sport events, with fewer participants and more clarity of the indexes. Admittedly, some indexes before the economic reforms such as steel production and grain output fit the premises (Zhou 2009). In addition, China implicated regional development policy with regions differing in the policy priority, and the upper-tier government has the controlling power to orientate the regional function, index of land use, and credit policies.

6. 6.

   The names and ranges of central expenditure have changed over the years see Li (2010) for a discussion of this issue.

The research for this paper was conducted with the generous support of Liu Shiding, Zhou Xueguang, Zhang Yonghong, Chen Nabo, Liu Yuzhao, Feng Meng, Huang Xiaochun, Di Jinhua, and Tian Xianhong. Also, the referees and editors are gratefully acknowledged.

Authors’ contributions

PS, the main author, was director of the whole research plan and wrote the manuscript. MN, the second author, was the research assistant of PS and helped collect data. Both authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests, since the data in this paper is open to the public.

Affiliations

1. Zhejiang University, Hangzhou, People’s Republic of China
   • Puyuan Shi
2. Peking University, Beijing, People’s Republic of China
   • Mengwei Ni

Corresponding author

Correspondence to Puyuan Shi.

Reprints and Permissions