Research Questions Xi Jinping became

Xi Jinping became the supreme leader of the Chinese Communist Party at the 18th Party

Congress in 2012. The following year, at the Third Plenary Session of the 18th Central

Committee, he initiated a comprehensive reform blueprint. In this plenary session, the Central

Comprehensively Deepening Reforms Commission was established. This reform was

intended to be even more far-reaching than the previous ones, concerning the economy,

democracy and legal systems, culture, social institutions, and the CCP’s discipline inspection

system.

In general, this plenary session announced many reform policy initiatives that tended

toward economic liberalization. For the first time, it emphasized that the market should play a

“decisive role” in the allocation of resources. Thus, it was said to be necessary to develop an

open and competitive market; improve the decisive role of the market in determining prices;

reform the fiscal and taxation system; establish a unified land market across urban and rural

areas; and promote urban and rural integration. Excessive governmental intervention into the

market was to be constrained. It also stressed improvement of the property right protection

system. For the state-owned enterprises, this session proposed to promote the separation of

government and business; continue to implement corporate autonomy and self-management;

and finally establish a modern enterprise system. Most importantly, the plenary session also

accentuated the role of the non-public sectors in promoting growth, innovation, employment

and fiscal health; and proposed to support the development of the non-public economy by

upholding the equal rights, opportunities and rules, and abolishing unreasonable regulations.

These new policies encouraged non-public enterprises to participate in SOE reform and

develop the mixed ownership system.

The ambitions of the CCP’s new leadership to push forward a comprehensive market-

oriented reform could not easily be more salient than in this plan. The reform intentions of

the CCP are evidently not a passing fancy but a serious, long-term commitment. For decades,

since the policy of “reform and opening” was adopted in 1978, the Communist Party has been

successful in pushing China’s transition from a planned, totalitarian economic system under

Mao to a market system (Fu 2000; Zhou 2008). Given that the communist political

institutions in the former Soviet Union and some East European countries were so rigid and

hostile to innovation (Frye and Shleifer 1997), China’s political and economic elites would

have been expected to defend their vested interests by blocking market reforms. But the

CCP’s top leadership overcame such obstacles to achieve remarkable success in economic

reform and development. What puzzles scholars most is that, while this market transition has

fundamentally changed almost every aspect of China’s economy and society, the CCP’s

authoritarian system survives with its basic features intact (Nathan 2003).

Why did China succeed in economic reform? Why could the private economy break

through the barriers of communist ideology and the command system to play an increasingly

important role in China’s economic development? How could foreign investment bypass

institutional barriers and take root in China's economic system? What are the outcomes of

Chinese political institutionalization? And what are the roles of political institutionalization in

promoting economic reform performance?

One common view, which emphasizes China’s success in developing formal, market-

friendly institutions, is as follows. Since the early 1980s, the Chinese Communist Party’s top

leadership has engaged in a 30-year-long ideological dispute over whether to prioritize

conservatism or reform. These ideological struggles are about whether the planned economy

or the market economy is more in line with the ideological fundamentals of communism and

about which ideology is more conducive to the continued rule of the CCP. In this view, the

reformists within the CCP gradually prevailed over the conservatives. The neo-liberal market

economy was grafted onto the one-party dictatorship, and China’s economy was gradually

shifted from an closed and internally recycling system to an export-oriented market system.

Under the conditions of an export-oriented development strategy, institutional evolution

brought about the success of China’s economic reform (Ang 2016). China has developed a set

of formal institutions that take the form of laws and regulations that protect the development

of private economy.

This gradual change is not only economic, but also political in nature (Shirk 1993).

Shortly after the Cultural Revolution, Deng Xiaoping proposed the reform and opening up

policy to transform the communist regime and make it more pragmatic (Yang 1997). In the

1980s, the central government allowed regional autonomy. The provincial units could deal

with local affairs related to economic development, thus giving local bureaucracies incentives

to benefit from reform from its early stages. Meanwhile, the CCP established a personnel

management system to guide its bureaucracies in accordance with the established reform

policy. Through this formal arrangement, the interaction between economic decentralization

and political dominance leads to the success of reform and opening up. The inefficient state

economy is gradually phased out. Private property rights are protected to promote rapid

growth of private economy. Also, China is deeply involved in the tide of economic

globalization. Through these methods, economic change and political reform work together

to establish market-oriented, rule-of-law institutions.

However, this view has at least two weaknesses. First, it may exaggerate the intensity of

ideological confrontation within the party center. Ideological conflicts on the developmental

path of China did exist shortly after Mao’s death. Even though the small circle of Cultural

Revolution supporters was defeated politically, the remaining conservative revolutionaries

still favored a Stalin-style planned economy and opposed radical market reforms. In this

context, the top factions that supported pragmatic economic policies won the competition

against the conservative revolutionaries by offering policy benefits. Their cronies in the

lower-level bureaucracy implemented some reform policies and successfully promoted local

growth. This helped their superiors at the party center to appeal to the elite selectorate by

demonstrating improving regional economic performance (Cai, Park, and Zhao 2008). After

the top leadership was consolidated and unified by the reformists, the party center gradually

ratified the policy decision to move toward reform and opening up (Shirk 1993). The 1989

Tiananmen Square protests and the corresponding mass killing further destroyed the CCP’s

political legitimacy. After that, the CCP’s top leaders agreed that only economic reform and

improved economic performance could legitimize the CCP’s authoritarianism. Since the early

1990s, therefore, there has been no ideological conflict over the fundamental reform path

among the CCP’s top leaders. The political struggles, if any, are mainly focusing on power

conflicts rather than ideological confrontations.

Second, it may also overstate the achievements of China’s institutional evolution. In fact,

a defining feature of the socialist economy in China during the market transition is the lack of

fully legitimated and well-defined private property rights (Nee 1989). Although the CCP has

accepted a more open economic system, the communist hierarchy is frozen in time. The

hierarchical system shapes the process of Chinese market transition and consistently adjusts

its economic policies in response to the forces unleashed by the changing international

political economy (Lardy 2012). The stagnation of political reform blocks China from

establishing rule-of-law institutions. The CCP’s dictatorship and lack of checks and balances

may discredit and undermine its market-oriented reforms.

Under these conditions, market forces are still constrained by political interventions.

China has to continue its economic development using structurally weak rule-of-law

institutions. Personal connections between state agencies and firms may provide crucial

informal linkages. Entrepreneurs have to mobilize these personal networks or even bribe

cadres to gain access to economic resources and opportunities. Therefore, many of the basic

features of the reform process may be explained by the informal structure of the CCP’s

political system. Informal institutions here refer to implicit norms and customs, and most

importantly, to the patronage network within the CCP’s bureaucratic system that fulfills the

changing needs of the politicians to sustain their rule (Tsou and Nathan 1976; Nathan and

Tsai 1995; Tsou 1995). I argue that the CCP’s informal politics can help to improve the

market-oriented reform performance. In the absence of strong rule-of-law institutions to

sustain the market economy, the CCP’s organizational resources enable the top leaders to

manipulate their political factions, thus inducing the economic behavior of non-public

businesses and investors to follow the party’s economic development goals. It seems that the

CCP’s informal politics may not result primarily in inefficiency or chaos. Instead, it may be a

driving force for informal cooperation between the party officials and private entrepreneurs,

which makes well-connected provinces especially attractive destinations for private

investment.

By arguing this, I do not intend to deny the importance of formal institutions. I contend

that market reforms in China are not shaped by formal institutions alone. Instead, it is the

formal and informal institutions that jointly drive China’s economic reform. Markets are not

perfect and have various limitations and weaknesses. Thus, they require institutions that

support property rights, regulation, macroeconomic stabilization, social insurance, and

conflict management (Rodrik, 2009). These formal institutions also reduce the uncertainty

and transaction costs of economic activities, induce entrepreneurship, and encourage

technological innovation, all of which contribute to economic liberalization and development.

However, the rigid authoritarian regime may prevent the evolution of formal institutions from

meeting the requirements of economic reform and development. In this situation, informal

institutions encourage the local officials to follow the policy preferences of the CCP’s top

leadership. Also, informal institutions strengthen the collusion of political and economic

elites. Given that the officials affiliated with strong factions may have more political

resources to preserve their power, they could do more to ensure the private investors’

prospects of rewards. In this way, informal institutions can help the private economy to

circumvent the obstacles of the old system and obtain future benefits.

In the following sections, I discuss how past researchers elaborate the relationship

between China’s economic decentralization and institutional development, and how the

CCP’s formal personnel management system produces reform incentives. These studies

reveal the great importance of institutional evolution in China’s reform and opening up. In

addition, I summarize some recent studies and evidence on the role of informal politics in

Chinese economic reform. They explain how informal institutions, usually seen as the CCP’s

patronage network, complement the weaknesses of China’s legal system and bolster market

prosperity.

1.2 Economic Decentralization and Institutional Evolution

Naughton (2008) argues that the starting point of China’s economic reform was the

command economy equilibrium, within which the communist leaders distributed physical

goods and promotion opportunities in exchange for their followers’ allegiance and political

support. Economically, the protected state-owned enterprises delivered enormous profits to

the government through the financial system. In turn, the party allowed consistent

governmental protection of SOEs and delivered job and promotion opportunities to their

clients in SOEs through personnel management. Governmental officials and urban workers

were all fully incorporated into this hierarchical system. China’s economic transformation

would be expected to be very costly in the absence of market-like flexibility that drives

resource allocation toward Pareto Optimality. Under these conditions, decentralization within

the system may be helpful in breaking the ice (Qian and Xu 1993).

China’s command economy to some extent, if not fully, differed from the planned

economy of the classical Stalinist system (Jowitt 1992). The Stalinist system was based on a

series of hierarchical organizations along ministerial lines, within which there was almost no

substantial participation of regional governments in economic plans. In contrast, China’s old

system entailed meaningful participation of local governments in the planning process. This

might be because Mao Zedong tended to promote local officials’ political radicalism with

regard to the achievement of rapid industrialization (Yang 1998; Kung and Chen 2011). Thus,

it was the radical movements during Mao era that broke the Stalin-style centralized structure

to form a decentralized system. Maskin, Qian, and Xu (2000) regard the decentralized

structure as an M-form system, which is organized along regional lines rather than along

industrial and ministerial lines. Since the late 1950s, this M-form economic system replaced

the power of specialized ministries with regional power, establishing an almost self-contained

and self-sufficient system in function at the county level and above. Under the M-form

system, the subnational governments, including the provinces, municipalities, and counties,

are empowered with local economic affairs, controlling a substantial amount of resources,

such as land, energy, raw materials, firms and budgets, and holding the responsibility for

initiating, coordinating and regulating reform process, providing public services, as well as

making laws and enforcing policies within their jurisdictions (Qian, Roland, and Xu 2006,

2007).

Since 1958, local governments had gradually acquired the power of economic plan and

enterprise management, distribution of goods and materials, approvals of infrastructure

construction, investment and bank credit, taxation and finance, as well as labor management,

which were also strengthened during the Cultural Revolution (Wu, 2005). Under the authority

of the central government, the local government in China gained the economic power far

beyond the local governments of the Soviet Union. Without market-oriented reforms to

encourage enterprises and individuals, Chinese local governments became the main forces in

promoting economic development, which was very different from the planned economic

system in the Soviet Union. In such a system, China’s four-level administrations below the

central government were also given wide discretion in implementing new policies to promote

economic transition (Chung 2016). Thus, economic decentralization continued to be

influential after Mao’s death, which created favorable conditions for the subsequent economic

reform (Landry 2008). The combination of centralized policy-making and delegated

implementation was better than the Soviet hyper-centralized system in facilitating local

reform experimentation.

Although disputes on market reform policies remained intense between the conservative

factions and the reformists in the central government, Chinese local governments could take

steps toward market reform without political instructions or pressure from the top. China’s

crucial early agricultural reform—which moved peasants from a collectivized system to a

household responsibility system—was carried out under the leadership of local governments.

After that, the economic reform of the lower-level units triggered political reform among the

ruling groups. A heated debate over the right and wrong of rural reform was launched among

the central leaders. The conservatives asserted that it shook the foundation of the socialist

system under the CCP’s leadership, while the more pragmatic reformers proposed that any

policy conducive for economic development should be accepted. The debate over the

tendency of reform policy eventually evolved into a fierce political struggle. Historians

dispute on the roles of Hua Guofeng and Deng Xiaoping in this political struggle (Ruan,

1994). But there is no doubt that the reformists successfully demonstrated brilliant economic

performance in the provinces they controlled. Their achievements were proved to be more

urgently needed by a poor and backward China. Deng Xiaoping used the power of the

reformists to defeat Hua Guofeng. One of his major achievements was the establishment of

the so-called collective leadership of the CCP in the 1980s. In this system, there were checks

and balances among various factions within the party but nobody could challenge Deng’s

personal power. Ironically, it also institutionalized the political process of China and

promoted power-sharing among various elite groups (Shirk 1993). The Chinese Communist

Party is seemingly no longer a Leninist party dominated by a charismatic leader, but a power-

sharing system for China’s ruling elites.

Such political changes opened the door to further reform. After that, the CCP carried out

comprehensive reforms of China’s economic system. Private firms were allowed to enter the

economic sectors previously monopolized by the state. Although reducing the supply of

tradition forms of patronage, emerging technocrats on balance benefited from reform. Reform

expanded obtainable asset values and created new income streams. Reform also increased the

space available for rent-seeking activities. Consequently, the ruling elites became less loyal to

the old system. Rigid communist institutions were gradually abandoned and replaced by

market-oriented institutions (Naughton 2008). Highly educated technocrats were substituted

for fanatical revolutionaries. These technocrats followed the rules and principles of the

market economy in designing industrial policies. Agricultural collectivization policies in rural

areas had been abolished. The autonomy of managing land was given back to the peasant.

The restrictions on labor mobility were gradually abrogated. A large number of rural people

were allowed to enter the city to provide cheap labor for the process of industrialization,

which forms China’s comparative advantage in international trade (Cai, Park, and Zhao

2008).

Price controls were completely liberalized. Privatization reform also occurred in urban areas.

Inefficient state-owned sectors were transformed. Economic resources were reallocated to

improve efficiency. Investments in low-efficiency regions and industries in old system were

transferred to more efficient areas and businesses, which enabled the rapid development of

the coastal provinces (Chan, Henderson, and Tsui 2008). The renewed protection of private

property rights boosted investors’ confidence in future profits and led to a boom in the private

economy (Lardy 2014). Therefore, market-oriented policies and institutions promoted the

development of China’s economy. In turn, the development of China’s economy in

legitimized the CCP’s authoritarian regime.

In this context, most of the past literature views China’s economic reform as a process of

institutional evolution from a communist system to a market-oriented economy. Ang (2016)

argues that institutional evolution helped China to escape from the poverty trap. Because of

the decentralization and innovation incentives, bottom-up improvisation among ground-level

officials drove China’s economic transformation. Through gradual reform and development,

local officials actively overcome the old institutional weaknesses, such as rampant corruption,

weak property rights protection and under-regulation. Thus, economic growth could precede

good governance. The development of the market economy then generated problems that

required the local officials to improve formal institutions, which in turn fostered further

economic development. This implies that China’s institutional evolution is based on a flexible

and adaptive governing system that empowers the local bureaucracies to solve evolving

problems. Thus, China could harness existing resources to initiate reform without full-fledged

market-friendly institutions.

Apart from its effects on domestic reform and growth, China’s newly established

institutions also attracted large-scale foreign investment. Based on theories regarding effects

of institutions on economic behaviour (North 1990), Fu (2000) claims that China benefitted

from the evolution of pro-competition, predictable, and transparent institutions. These

institutions are favorable for protection of property rights and reduction of transaction costs.

This is especially true for the growth of foreign direct investment, because formal regulatory

frameworks are far better than informal factors in facilitating large-scale foreign investment

projects that are capital-intensive, complex, interpersonal, intertemporal, and geographically

distant from the source. Therefore, China has come a long way from a command system with

bureaucratic interference to an open market system. Its sustainable economic development

relies on the steady improvement and solidification of a market-oriented rule of law

institutional framework.

Overall, institutional theorists of economic development stress the crucial roles of

Weberian-style bureaucracies, market-oriented economic systems and an effective legal

framework that can protect private property. The more complete the shift to market

coordination, the less likely that economic transactions will be embedded in political

networks, and the more likely economic power will be located in market institutions.

Advocates of this standard explanation claim that, based on decentralization, China gradually

established a market system through institutional innovation. China’s market reform has

resulted in bypassing hierarchies to establish horizontal market relationships. The new system

could support tax reform, high accumulation and investment, free flow of domestic labor, an

export-oriented economy, as well as a restructuring of state-owned enterprises, all of which

are conducive to sustain rapid growth.

However, the role of formal institutions in China’s economic reform and development

remain controversial in some recent literature. For instance, Cai and Treisman (2006) are

skeptical about the effectiveness of decentralization on regional reform experimentation

because the initiators of some crucial reform were not the empowered local officials.

Empirical evidence on the causal influence of economic decentralization on hardened SOE

budget constraints is weak. The performance incentives of fiscal decentralization are offset by

the sharp decline of available tax revenue. And more seriously, without substantial political

reform, economic decentralization cannot enable local governments to monitor the power

center to limit predation and convince investors that reforms will last. Instead, the central

government can change the rules at any time to shrink the pool available for sharing, which

undermines the credibility of economic incentives. Based on this reasoning, Cai and Treisman

(2006) propose that it is the central political struggles among rival factions that stimulate the

lower-level reform competition. Although being retarded by the conservative factions, the

reformist leaders tended to support and promote the local officials who achieved brilliant

economic performance to endorse their reform preferences and consolidate their central

power. While the local officials connected with these factional leaders have sought to impress

their superiors at the power center through spontaneous and sometimes illegal initiatives to

break the old system. These initiatives might be sufficient to break up the old system to

promote growth-promoting market reforms. Thus, Cai and Treisman (2006) remind us that

the driving forces of China’s economic reform may be present in the factional politics of the

CCP’s personnel management system.

1.3 The Rules of the CCP’s Personnel Management

Leninism advocates establishing a totalitarian state with a pyramid structure, within

which the vanguard party possesses absolute power over people’s thoughts and actions.

Following Leninist principle, the CCP established a party-state system in the year 1949. The

CCP tried to promote national industrialization at the expense of the agricultural sector. This

was an important strategy for China to develop from a backward agricultural country to an

advanced industrial country, which would also prove the superiority of the Leninist system.

Soon after that, however, the efforts made to rapidly industrialize this country were

interrupted by political fanaticism fueled by Mao Zedong. The Cultural Revolution almost

destroyed China’s national economy and broke the communist bureaucracy’s control over the

state. A large group of radical revolutionaries who had absolute allegiance to Mao replaced

the orthodox Marxist bureaucracy at the power center. But this did not weaken the party’s

leadership.

After Mao Zedong’s death, the party’s conservative cadres regained power. Radical mass

movements and factional struggles were no longer the main aspects of political life. More

pragmatic policies were carried out to liberalize the economic and even the political system.

During the Deng era, a separation of powers was implemented to adjust the relationship

between the party and the state. Although the party was seemingly weakened shortly after

Cultural Revolution, Deng Xiaoping was not willing to accept the prospect of losing power.

He formulated the Four Cardinal Principles to accentuate the party’s sacrosanct position in

Chinese politics: China still adhered to the CCP’s leadership, Marxist ideology, the socialist

system, and the people’s democracy and dictatorship. After that, economic reform continued,

while political liberalization movements were suppressed.

Nevertheless, Mao Zedong’s political heritage was seemingly replaced by a modern

bureaucratic system. The party power was withdrew from some economic domains, but

remained dominant in this bureaucratic system. In this context, China’s bureaucracy has also

become the basic unit for political analysis. Huang (1996) argues that the party center

distributes political and economic resources by appointing different kinds of provincial

officials. When a province is assigned a leader who has a close relationship with the

localities, the region’s privileges would be better defended. On the other hand, if a leader who

has a close relationship with the party center is appointed, the local units would be more of

the agent faithfully implementing the mandate of the central government. Through this

bureaucratic mechanism, the party center influences the central-local conflicts in various

policy areas.

Because the reform-era CCP legitimizes its authoritarian regime by fostering economic

growth, the bureaucratic selection system is designed based on meritocratic principles (Wang

2008). The CCP selects its cadres by examining their political integrity and ability. Education,

expertise, work experiences, political loyalty, and usually most importantly, past economic

outcome performance, are scrutinized. The candidates must show such qualifications to gain

access to higher positions. Advocates of the CCP’s meritocracy speak highly of this system,

because it can select qualified officials to increase developmental efficiency and achieve

economic goals (Bell 2016). Under this system, the bureaucracy has more formal legal

legitimacy. It formalizes China’s political process and enhanced the credible commitment of

institutions as a whole. And political loyalty is no longer the primary criterion of selecting

cadres (Zhou 2013). Along these lines, some empirical studies show that provincial leaders

who have good economic performance are more likely to be promoted to senior positions (Li

and Zhou 2005; Chen, Li, and Zhou 2005).

Accordingly, Zhou (2008) argues that a promotion tournament plays a crucial role in

achieving China’s economic miracle. Although keeping a very tight grip on political power,

the party center can delegate economic power to lower levels without worrying about

regional centrifugal tendencies. The superiors make development plans and evaluate the

economic performance of their inferiors. At the same time, local officials have discretion

about local economic affairs. They take responsibility to implement developmental policies

and achieve economic goals. The top leaders work closely with the bureaucrats to implement

policies that favor rapid growth (Lieberthal and Oksenberg 1988). In this promotion

tournament, the local GDP growth rate becomes a major criterion for evaluating local

officials’ career performance. Local officials seeking career promotion spare no effort to

increase the growth rate in their jurisdictions. This highly institutionalized cadre evaluation

system encourages competition to increase the efficiency of public goods production

(Naughton 1995; Wu 2005; Xu 2011).

However, some more recent literature questions this reasoning. Although economic

development and improved living standards of ordinary people in authoritarian countries are

important for the survival of dictators (Gallagher 2002; Doner, Richie, and Slater 2005; Wang

2016), the top leadership is as much preoccupied with winning the political struggle as with

achieving economic performance (Tullock 1987; Svolik 2009). This is because the

distribution of power may have a direct effect on consolidating authoritarian rule (Nathan

1973; MacFarquhar and Schoenhals 2006; Svolik 2012; Albertus 2015). In China, the

peaceful periods serve renewed conflicts while the periods of conflicts motive the subsequent

power dominance (Dittmer and Wu 1995). That is, the balance of power at the party center

results in political struggle among different factions, which may in turn improve the power

center’s control over the provincial bureaucracy. Political struggle may result in cross-

factional power-sharing with effects similar to formal checks and balances (Li 2012), and

fending off the centrifugal tendencies deriving from prosperous provinces (Sheng 2007).

Political struggle may also help to purge ambitious officials who have better economic

performance but are not loyal to the CCP’s top leadership. In this way, the party ensures that

power is limited to a political elite coalition loyal to the current leaders. Therefore, it is

argued that the priority of the party’s personnel management is not so much selecting gifted

cadres to promote economic growth, as preserving the bureaucracies’ political loyalty to the

party’s incumbent leadership (Guo 2007; Shih, Adolph, and Liu 2012).

Theoretically, promotion tournaments constrain the discretion of the authoritarian leaders

in selecting cadres. Merit-based recruitment seems to be a set of fair rules that induce local

officials to serve economic development goals. Yet, Tao et al. (2010) argue that such rules go

against the nature of authoritarianism. If superior leaders select cadres according to some

clearly defined criteria, they cannot promote their cronies and cultivate a ruling elite group

that is loyal to its patrons. They also point out that there are no open sources showing that the

government has established and effectively implemented quantifiable criteria for cadre

selection. Instead, public information with regard to the criteria for assessing cadres are rather

vague. In addition, exaggeration and concealment coexist. Chinese local officials commonly

exaggerate local GDP data to show better performance. On the other hand, local governments

in some poor areas deliberately curtail GDP data to attract and siphon off additional financial

subsidies. Advocates of promotion tournament theories fail to explain how to correct for the

impact of widely existed data falsification among local officials.

Landry, Lü, and Duan (2018) conjecture that the party employs a multilevel appointment

strategy. Specifically, the power center controls personnel management by appointing and

dismissing the main local leaders in the provincial party standing committee, while the

provincial leadership has a political voice in determining the sub-provincial leaders. The

career promotion of the sub-provincial level officials is mainly based on their working

performance, while factional politics may become more important when selecting provincial-

level leaders. For Chinese authoritarianism, this strategy can synthesize the advantages of

meritocracy and cronyism. Economic technocrats are selected to hold local governmental

posts with direct influence on regional economic development, while trusted cadres are more

likely to enter the power center. This can achieve economic development incentives and

simultaneously ensure that the party’s power is in the hands of loyal political elites.

Therefore, the influences of informal politics among the high-level officials of the Chinese

Communist Party should not be ignored. Given the decisive role of the central government in

advancing economic reform policies, the impacts of informal politics of economic reform

needs further discussion.

1.4 Informal Politics and Economic Reform

It is far from clear that increasing economic performance in exchange for promotion

opportunities is the only significant rule for the functioning of the CCP’s bureaucracy. It

seems that the CCP still depends on informal politics during reform era. Opper, Nee, and

Brehm (2015) show that officials who have close relationships with superiors are more likely

to be promoted. Jia, Kudamatsu, and Seim (2015) prove that personal connections with

China’s top politicians in the central government accentuate the effects of a province’s

economic growth on its leader’s promotion chances. In other words, patronage networks help

the provincial leaders’ hard work get recognized and rewarded. This might be because

patronage networks solve the commitment problem, ensuring that the officials newly

promoted are less likely to become potential competitor political threats for incumbents (Pye

1992; Egorov and Sonin 2011).

Therefore, incentives are likely to be especially effective for well-connected followers.

Officials who have brighter vocational prospects are more likely to comply with their bosses’

wishes. Strong evidence is provided in some past studies showing that officials within

patronage network coteries are inspired to cooperate with their superiors and follow their

policy preferences (Kung and Chen 2011) and to work hard to increase economic

performance (Jiang 2018). Therefore, Wang (2000) argues that clientelism complements the

weak Chinese meritocracy in generating reform incentives. The CCP’s patronage network

provides trusted followers with incentives to comply with the center’s policy preferences and

boost market-oriented reform. At the same time, these additional incentives are constrained to

officials within this network.

Wang (2001) further argues that officials’ personal connections with the top gives private

entrepreneurs greater confidence that that their investments are safe and may yield higher

returns. In China, the subtle relationship between power and economic interests makes

entrepreneurs concerned with the politics behind investment programs. In the absence of

adequate rule of law and property rights protection, private sector actors prefer to seek extra

political protection. Chan, Henderson, and Chan, Henderson, and Tsui (2008) claim that

Chinese private firms obtain special deals to either break the formal rules or gain favorable

access to scarce resources. This is mainly because, on one hand, Chinese local officials have

enormous administrative capacity to provide a “helping hand” to favored firms, while on the

other hand, these local officials can benefit from collusion between the government and

private sectors. The cronies of the top leaders have better chances to be promoted to higher

levels of power, thus gaining greater ability to allocate economic resources to reward

connected investors. Hence, seeking a reliable political backer is sometimes more rewarding

than cultivating a market. Entrepreneurs have an incentive to invest in setting up intimate

politics-commerce collusion (Wei and Yang 2020).

One interesting example is the recent explosive growth of private investment in Guizhou

province. As a remote province located in the hinterland of southwest China, Guizhou is one

of China’s most impoverished regions. It is located on a high, hilly plateau, with

comparatively poor climate conditions and few natural resources. Due to geographical

limitations, infrastructural facilities in Guizhou are backward. In such a region, it is not easy

to develop economic critical mass by increasing agricultural output, carrying out large-scale

manufacturing production, or establishing a large service industry. Thus, economic growth

there has long been slow and lagged behind most other provinces. It is not an ideal place for

private investment. Although growing fast during Jiang Zemin era, Guizhou failed to keep

pace with the upward economic cycle and expansion of private investment during the first

term of Hu Jintao. From 2003 to 2007, the average value of the national private investment

growth rate peaked at 39.7%. Guizhou province only reached 23.6%. In the downward cycle

of Hu Jintao’s second term, the reform performance of Guizhou did not rebound. It slightly

decreased to 23.3%, which was still much lower than the average value of 25.5% for the

whole country and ranked at the bottom end.

Yet Guizhou’s economic situation has changed dramatically since the arrival of its new

leader, Chen Min’er. Chen spent most of his career working in Zhejiang province. He used to

serve as the Propaganda Chief of the CCP’s Zhejiang Committee from 2002 to 2007. During

this period, he worked under Zhejiang party secretary Xi Jinping. This experience gave him a

precious opportunity to establish a close patronage connection with Xi Jinping. From then on,

Chen has been portrayed by foreign media as a confidante of Xi Jinping and a rising political

star. Chen Min’er became a vice governor of Zhejiang in 2007. In 2012, the year when Xi

became the CCP’s General Secretary, Chen was transferred to the southwestern interior

province Guizhou as the governor. He became the party secretary in 2015. Chen Min’er was

fortunate to be one of the youngest provincial party secretaries in China. Under the rigid age-

based promotion norms in the CCP’s personnel management system, taking a provincial

leadership position at a relatively young age is a great political advantage for one’s future

career. The advantages of Chen’s young age and his shared work experience with the CCP’s

top leader, Xi Jinping, jointly signaled that Chen was a candidate for next-generation supreme

leadership of the CCP.

After becoming the provincial leader in Guizhou, Chen had been devoted to advancing

Xi’s policies. His most important contributions were made to promote the Guizhou’s

economic growth to lift the province out of grinding poverty. Chen enthusiastically led an

initiative to set up a program of regional development and invested heavily in making

Guizhou a center of innovation for recently emergent industries. The brightest fields for new

investment, such as electronic information technology and big data industries, high-tech

equipment manufacturing, bioengineering and drug research, and modern mountain

agriculture and tourism, were introduced in Guizhou to drive economic growth. As one of the

poorest provinces in China, the relative advantage of Guizhou to develop new industries is its

low electricity price for supporting large high-tech factories and big data computation

centers. International high-tech giants, such as Apple, Tencent and Huawei, have shifted their

equipment factories to Guizhou. Electronic platform construction, big data innovation and

cyber transaction technology were located there, making this backward province into a

gigantic computer laboratory for the whole country.

Chen’s performance in promoting private investment of Guizhou is remarkable.

According to the data from China’s National Bureau of Statistics, private investment grew

very fast in Guizhou from 2013 to 2017, rising to 27.4% and ranking the first in the whole

country. In contrast, the average value of the whole nation is only 16.2%. The extent to which

the success of Guizhou in attracting private investment to advance big data industries relied

on Chen Min’er’s personal connection with Xi Jinping is still unknown. But it is suggestive

that Chen did not depend only on private investment. Actually, the development of Guizhou

also relied on governmental investment. In recent years the central government had poured in

billions of yuan in fix-asset investment. Urban and rural infrastructure construction was

accelerated. Highway, railway, and water-carriage facilities were upgraded. Removing this

bottleneck on long-term economic and social development has improved the efficiency of

economic resources allocation and increased regional competitiveness. In Guizhou, both the

proportion of private investment in total fixed-asset investment and the difference between

the growth rates of private and state-owned investment are not larger than the national data. It

suggests that, alongside fostering private sectors, the development of Guizhou has also relied

heavily on governmental investment.

Therefore, both theory and anecdotes indicate that the effects of personal connections

with the top on provincial economic reform performance are positive. These effects might be

more significant in the growth rate of private investment than in the increase of reform policy

environment indicators. By appointing their confidants to the important provincial leadership

positions, the CCP’s top leaders are able to better implement their reform intentions. This also

shows the significance of informal relationships in promoting China’s economic reform.

When the party’s supreme leaders encounter resistance to changing status quo policy

preferences, patronage networks might be the most reliable instruments to overcome these

difficulties.

1.5 Plan of the Dissertation

Politics and economics are closely interrelated. This is especially true in China, since it is

a highly politicized authoritarian regime ruled by a Leninist party and integrated with a quasi-

market economy. Studies of contemporary China cannot deliberately separate economic

issues from political changes. For an authoritarian regime that has based its legitimacy on

economic performance, promoting efficient growth of the domestic private sector and

deepening engagement with the outside world are among the most important political

strategies. Thus, it is important to figure out how political factors affect China’s reform

performance in fostering private and foreign investment.

Several empirical studies are conducted to discuss the CCP’s personnel management and

test its effects on China’s private and foreign investment distribution. The empirical part of

the dissertation is organized into three chapters. The second chapter analyzes how the party’s

top leaders control provincial personnel through factions. I identify a number of specific

variables that are regarded as important factors affecting provincial leaders’ career prospects

and prove that provincial leaders’ personal connections with the incumbent general secretary

matter the most. This reveals the highly informal nature of China’s political development,

where changes have been far less fundamental than those in the economic realm. The third

chapter analyzes the relationship between factional politics and the growth rate of domestic

non-public investment. The purpose is to find out how the provincial leaders’ factional

coalitions affect the overall situation of provincial investment. The fourth chapter further

illustrates the impacts of such factions on FDI. Both of the two chapters on investment show

that, although informal politics may contribute to sub-optimal distribution of economic

resources, it may also compensate for the weaknesses of Chinese formal legal system in

promoting the non-public economy. The fifth chapter draws conclusions and discusses some

larger issues.

Chapter 2: Selection of China’s Top Leadership Cadre: The Roles of

Supreme Leaders, Factional Networks, and Candidate Attributes

Abstract. This article provides empirical evidence to show how the general secretaries of the

Chinese Communist Party dominated provincial personnel through their factional ties. Based

on panel data from 1993 to 2017, this study finds that the provincial leaders’ personal

connections with the incumbent party head significantly increased their promotion chances.

The positive effect of the incumbent party heads on promotion did not depend on provincial

leaders’ economic performance and seniority. This study further uncovers that working

experiences in the prefectural leading positions strongly increased the likelihood of

promotion. However, connection with other important top leaders did not have similar

effects. These findings challenge the traditional wisdom on the collective leadership and

indicate the dominance of the CCP’s heads for provincial personnel arrangements.

2.1 Introduction

Recently, a growing body of literature examines authoritarianism using institutional

perspectives (Geddes 1999; Magaloni 2008; Boix and Svolik 2013; Reuter and Robertson

2015). Formal political institutions help mitigate many of the worst features of

authoritarianism. Perhaps most importantly, they enable regular interaction between dictators

and their allies, thereby facilitating power-sharing. In this way, institutions prevent conflicts

among ruling coalitions from escalating into ruinous confrontations, thus enhancing

authoritarian stability (Svolik 2012). In China, however, this mechanism may not function

effectively. The absolute power concentration at the top, and the absence of overt ideological

differentiation, might preempt or mitigate serious political conflicts at the center. Under these

conditions, institutions may be no more than a tool of the top leadership to regulate lower

levels of government and induce them to follow the center.

What are the roles of the supreme leaders of the Chinese Communist Party in these

established institutions? Do the CCP’s formal or informal institutions have a significant

independent influence on the behavior of these supreme leaders? Or do these institutions

become the dependent tool of each new supreme leader? China’s governmental structure is

mainly based on geographical principles. This self-contained structure allows provincial

leaders to impose substantive direction regarding regional economic policies (Qian and Xu

1993; Qian, Roland and Xu 2006, 2007). The CCP’s provincial personnel management power

provides more political resources to determine policy orientation than does central personnel

power. Therefore, the CCP’s provincial-level personnel management is an excellent

dependent variable with which to evaluate these broader questions.

Scholars who study the determinants of the CCP’s personnel management disagree

about the importance of meritocracy and clientelism. Pro-meritocracy scholars believe that

the CCP selects its cadres based on their performance (Zhou 2008; Bell 2016).

Institutionalization of this meritocracy occurs via a promotion tournament that encourage

local bureaucracies to align with the top (Nathan 2003; Miller 2008). However, pro-

clientelism scholars depict China’s bureaucracy as being a web of political networks (Keller

2016). They see the CCP’s collective leadership as the basis of underlying informal patron-

client networks that sustain the factional balance of power at the center (Shih, Shan, and Liu

2010). Under these conditions, the CCP’s leaders strive to appoint their trusted followers to

important positions.

These studies enlarge our knowledge of the CCP’s personnel management. However, the

influence of the CCP’s incumbent supreme leaders on high officials’ career promotions seems

to have been underestimated. Therefore, This article discusses the power of the CCP’s

supreme leaders in selecting the CCP’s top leadership cadre. I argue that the CCP’s personnel

management institutions may be a tool of the CCP’s supreme leaders to extend their factional

power. Although institutions seem to constrain the career mobility of province-level leading

cadres, the party heads in power might be able to overcome these constraints and appoint

sycophants and loyalists from their own factions. To test these arguments, this study draws on

a set of panel data to test the effects of performance-based and seniority-based institutional

constraints on provincial leaders’ career mobility from 1993 to 2017. Apart from the greater

availability of data sources, I focus on this historical period mainly because it incorporated

the major institutionalization of China’s personnel management, which has arguably

generated both political stability and economic prosperity. Thus, focusing on this period

enlarges our understanding of China’s bureaucratic politics. Through quantitative analyses of

career mobility, I find that personal connections between the CCP’s supreme leaders and the

provincial party secretaries and governors significantly increase the probability of provincial

leaders’ promotion. However, connection with other incumbent top leaders imposes opposite

effects. It may largely dim their political prospect.

This study advances the existing literature on the China’s leadership promotion methods

and outcomes in the following ways. First, I summarize multiple quantifiable factors related

to career mobility, and compare their different effects in a provincial-level panel. I also

discuss the interactions among them. Second, I distinguish the influence of the CCP’s

supreme leaders from that of other top leaders, which helps to distinguish the uniquely

important role of party heads in China’s political system. Based on this, the current study

attempts to answer the broader question concerning the extent to which the Chinese

Communist Party’s supreme leaders are able to impose discretionary control over the whole

political system. The Leninist party cultivates its leaders and shapes their political behavior in

political practice. The party’s supreme leaders, in turn, enjoy a unique position in this system,

which empowers them to change the status quo of the party’s institutions (McAdams 2017).

Therefore, it appears that the party’s heads are by far the most powerful actors in Chinese

politics.

The roadmap of this article is as follows. In the next two sections, I briefly review the

prior literature on the determinants of career promotion at the higher levels of the CCP party-

state, and develop my own theories. The fourth section discusses the data and methods. The

fifth section presents baseline results and robustness checks. Moreover, I further test the

effects of the personal connection with other Politburo Standing Committee members affect

career promotion. The sixth section discusses how my results relate to prior findings. The

conclusions are presented in the final section.

2.2 Institutionalization of Authoritarianism and Meritocratic Bureaucracy

Since the 1990s, the CCP’s ruling elites have gradually come to a consensus over

sustaining long-term prosperity as a strategy to preserve the leadership of the CCP (Cai and

Treisman 2006). For this purpose, the party center tried to institutionalize personnel

management, decentralize power and transfer economic profits to generate reform incentives

(Tsou 1984；Solinger 1987; Wong 1987). Contrary to what many expected, the ensuing

expansion of local autonomy, conflicts of interest, and corruption among cadres has not

undermined reform performance as in other transition economies (Burns 1989; Huang 1996;

Frye and Shleifer 1997; Bo 2004; Landry 2008). Most scholars attribute this to the CCP’s

formal meritocratic institutions (Huang 2008; Bell 2016).

First, meritocracy provides a solution to the inconsistency of policy preferences among

levels of governments. The Chinese Communist Party stands at the pinnacle of central

autocracy to master the organizational resources (Yan 1995; Burns 1994; Huang 1996).

Economic reform has not substantially attenuated party control over the bureaucratic system.

Instead, the party has built a performance-based elite promotion system (Naughton 1995; Xu

2011). It can reward those officials achieving better economic performance by appointing

them to important posts, which increases bureaucratic competition among different regions to

increase economic growth (Chen, Li and Zhou 2005; Li and Zhou 2005). Under these

conditions, bureaucracies often stay in line with the party’s overall economic purposes.

Therefore, this “veritable bureaucratic revolution” (Li 1998, 395) induces local bureaucrats to

be devoted to growth (Lieberthal and Oksenberg 1988).

Second, this meritocratic bureaucracy places importance on professional qualifications

to select officials with the potential to advance growth. A nominee should show his or her

future performance capabilities to most of the top leaders (Xu 2011). In this case, quantitative

evidence revealed that the candidates’ past political experiences matter (Zhang 2010; Du,

Zeng and Wu 2012). This is mainly because these job records convincingly show the

candidates’ political training in policy-making, regional development, social control, crisis

management, and other important policy areas. In addition, a wider variety of career

experiences may reduce the likelihood of collusion with particular regions and increase

political loyalty to the party center (Zhou 2016; Zhou et al. 2018; Xiang 2019).

Age is another crucial factor in examining the qualifications of provincial leaders. To

allow younger but more qualified technical officials to replace older revolutionaries, the de

facto lifetime tenure of leading cadres has been abolished since the early the 1980s (Li 2004).

Before the strict retirement age arrives (usually 65 years old for provincial-level leaders),

younger officials have more time to accumulate the leadership experiences required; they also

have more time to secure higher positions. In contrast, provincial officials are less likely to be

promoted to a higher position when approaching retirement age. Because of this, holding

provincial positions at relatively young ages has gradually become a political advantage for

future leaders.

Meritocratic scholars believe that the CCP cultivated a pro-market bureaucracy through

established institutions. However, Tao et al. (2010) were skeptical of this viewpoint, claiming

that any unambiguous formal institutional configuration would constrain the behavior of

authoritarian rulers, which runs counter to their discretion and interests. For autocratic rulers,

supervising the subordinates’ political behavior to ensure loyalty is actually more important

than improving performance (Landry, Lu and Duan 2018). Thus, China’s dictators may place

a higher priority on maintaining short-term power by buying the support of winning

coalitions than on improving economic performance (Shih, Adolph and Liu 2012; Jiang

2018). Therefore, opinions that performance concerns have trumped factions in determining

the career promotion of China’s provincial leaders may be misunderstandings. Along these

lines, more recent studies have stressed the role of personal favoritism or clientelism in the

CCP’s personnel management (Jia, Kudamatsu and Seim 2015; Opper, Nee, and Brehm

2015). The presence of collective leadership has been widely accepted by some Chinese

politics observers (Nathan 2003; Hu 2013; Li 2016). This collective leadership system

empowers the Standing Committee of the Central Political Bureau of the Chinese Communist

Party, usually known as the Politburo Standing Committee, to make the final decisions on

provincial leaders’ career mobility (Zhang 2009). Thus, a close relationship with incumbent

PSC members is expected to be a crucial factor in such mobility.

2.3 Institutional Constraints on the CCP’s Supreme Leaders

Previous researchers have revealed multiple criteria of political selection, proving that

both performance and factionalism are relevant for elite mobility (Walder 1995; Zhang 2009;

Choi 2012; Jia, Kudamatsu and Seim 2015). However, they might underestimate the

capacities of the CCP’s heads to change the status quo of the bureaucracy via factional ties.

Goldstein (1991) argued that, in the Chinese Communist Party’s established hierarchy, a

commitment to the unified image and organizational vitality of the Leninist vanguard party

sustains a well-functioning political system, which endows the CCP’s supreme leaders with

ultimate authority. Tsou (1995) claimed that the political struggle of the CCP is basically “a

game to win all.” He thus stood in sharp contrast with Nathan’s (1973) popular view that

balance-of-power politics can be sustained on a regular basis within the party center. As the

individuals who hold the real power of the CCP, the personal authority of the supreme leaders

is at the apex of the party–state’s political order. Successive supreme leaders endeavor to

concentrate power tightly in their hands, which may even affect the collective leadership of

the PSC. In this view, the role of formal institutions in constraining the CCP’s top leaders

should not be exaggerated. Neither the formal institutions of the personnel evaluation system

nor the so-called collective leadership at the apex can effectively constrain the political

behavior of the CCP’s supreme leaders. The party heads are able to develop their own

factions to change China’s bureaucracy (Keller 2016; Meyer, Shih and Lee 2016).

Therefore, from my perspective, the CCP’s incumbent supreme leaders seem to be able

to bend the rules of meritocracy. In addition, they may have a strong desire to do so. First, the

supreme leaders are most anxious about the political loyalty of their subordinates. Under the

CCP’s pyramid hierarchy, the party’s supreme leaders stand at the peak of the political

pecking order (Yan 1995). Any political threats to their power are most likely to come from

below. The expansion of autonomy and the centrifugal tendency of local governments may

pose substantial threats to the power of the center, and may trigger a sense of insecurity in the

party’s supreme leaders. Therefore, they tend to adjust personnel selection rules to fulfill their

demands for personal loyalty. Second, the party’s supreme leaders do not rely completely on

economic performance to consolidate power. Although prosperity legitimizes the CCP’s

authoritarian regime, it does not necessarily strengthen the power of the supreme leaders.

Instead, different factions within the party may challenge their authority by building the

prestige of their circles through regional economic achievements (Cai and Treisman 2006). If

the performance-based personnel selection rules were strictly followed, hostile factions

would have expanded their strength through excellent performance, and would thus be able to

challenge the power of the CCP’s supreme leaders. From this understanding of personnel

management in China, I derived the following hypothesis (H1):

H1: Factional connections with the CCP’s incumbent supreme leader make provincial

leaders more likely to be promoted.

Under the CCP’s institutionalized personnel management rules, economic performance,

past working experiences, and age should be the crucial factors affecting provincial leaders’

career promotions. However, any established rules in the party seem to run counter to the

dictators’ ambition of pursuing power, because they may impose a constraint on the political

behavior of the dictators. Strong leaders should be able to break constraining rules when they

have a personal political interest in doing so. Thus, the interactive hypotheses (H2–4) specify:

H2: The effect of factional connections on career promotion do not depend on the economic

performance of provincial leaders.

H3: The effect of factional connections on career promotion do not depend on the past

working experiences of provincial leaders.

H4: The effect of factional connections on career promotion do not depend on the age of

provincial leaders.

2.4 Data and Methods

2.4.1 Panel Data

To test the theories and hypotheses above, panel data on provincial leadership was

gathered. It takes the province-year as the unit of analysis, incorporating almost all of the

provincial party secretaries and governors from 1993 to 2017. I replaced leaders whose tenure

was less than half a year with their successors because their short tenure was unlikely to

generate any meaningful policy influences. I excluded leaders who died during their tenure. I

also excluded those transferred to the People’s Congress or to the Consultative Conference

after the age of 63, because their exit from power circles seemed to be due to age rather than

performance or any other political concerns.

Biographical information was mainly from Jiang’s (2018) Chinese Political Elite

Database. I double-checked the data by using other independent data sources from the

websites China Vitae, Wikipedia, and Baidu Baike. Data on provincial economies were

obtained from China’s National Bureau of Statistics website and the annually published

China Statistical Yearbooks.

2.4.2 Baseline Model

This study incorporates a binary variable accounting for promotion or non-promotion as

its independent variable. In this case, a logit model is valid for estimating the effects of

factionalism on provincial leaders’ career mobility. The probability of promotion is given by

Promotionp, t+1 = α0 + α1Factionp, t + α2Χp, t + α3Factionp, tΧp, t + α4Ζp, t + γ p, t + η p + ε p,t

where p denotes the province and t indicates the year. Promotionp, t+1 is the career mobility

status of one provincial leader. Factionp, t is the main independent variable indicating the

personal connection of the provincial leaders to the incumbent party head. Χp, t is a vector of

the performance, experience, and age characteristics of provincial leaders. I incorporated

Χp, t and its interaction with Factionp, t to test whether there is any differentiation of the

effects of these promotion determinants between connected and non-connected leaders. For

technical reasons, the continuous variables in Χp, t are centered to avoid collinearity caused

by interaction terms. Ζp, t is a vector of control variables. The year dummies γ p, t and the

province dummies η p capture common shocks and time-invariant heterogeneity among

different observations. Both of them are allowed to differ between party secretaries and

governors. The standard errors are clustered at the province level to avoid serial correlation.

For convenience of explanation, I will report the odds ratio of coefficients in the final

models.

2.4.3 Dependent Variables

As the main dependent variable of this study, Promotion indicates whether a party

secretary or governor of province p in year t was appointed to a higher position during the

period between July of year t and June of year t+1, which was to be matched with their

economic performance in year t. Given the ambiguous rules of the CCP’s bureaucratic

hierarchy, scholars have held varying opinions on criteria for judging provincial officials’

upward mobility (Geng, Ling and Pang 2014; Zhong, Geng and Chen 2016). I regard a

provincial leader as being promoted if he or she assumed a national-level office, including

becoming a member of the Politburo, a vice president of the PRC, a vice-premier, or a state

councilor in the central government. Provincial governors becoming provincial party

secretaries or heads of a central party department or a ministry could also be seen as

promotions. For comparability to some previous studies (Jia, Kudamatsu and Seim 2015), I

also regard occupying the secretary of the Secretariat of the CCP’s Central Committee, the

vice chairmanship of the National People’s Congress or the Chinese People’s Political

Consultative Conference, and the head of the Supreme People’s Court as a promotion. In

robustness check, exclusion of these factors generates very similar results.

2.4.4 Independent Variables

In this study, the most important independent variable shows the factional connections of

provincial party secretaries and governors. I designated the variable Personal Connection as

the measure of the relationship between provincial leaders and the CCP’s incumbent heads

(general secretary of the Central Committee). In this regard, past researchers often regarded

shared working experiences as evidence of interpersonal ties (Keller 2016). Working in the

same place during the same period increases the probability of knowing each other and

establishing mutual trust. However, intimate factional ties are also built on hierarchical

relationships within a regional or departmental bureaucracy. In particular, incumbent supreme

leaders might be expected to promote their past subordinates. Therefore, I coded the variable

Personal Connection as 1 if a provincial party secretary or governor used to work with one of

the subsequent supreme leaders as a subordinate within a regional or departmental

bureaucracy at the same time; otherwise, it was coded as 0. This patronage connection is only

coded as existing when the supreme leader is in office, and not before or after he is in

office.

To encourage local bureaucrats to follow the party center, the CCP’s top leaders may

evaluate the performance of their inferiors via important economic indicators (Zhang 2009).

Accordingly, I measured a provincial leader’s Economic Performance using the average

provincial GDP growth rate. It was measured by the mean value of the provincial GDP

growth rate during one leader’s tenure. The average provincial GDP growth rate reflects the

provincial leaders’ contributions to regional prosperity, which would be expected to be a

criterion for calculating the performance of provincial cadres.

To capture seniority, I constructed several dummy variables, including SOE,

INTERPRO, REMOTE, CENT, CITY, and CYL. SOE measures work experiences at state-

owned enterprises. INTERPRO measures work experiences in other provinces. To ensure the

significance of these political experiences, all these positions need to be at or above the

departmental level. REMOTE measures work experiences in the local government of

Xinjiang and/or Tibet. CENT measures past experiences of working in branches of the central

government. CITY measures past experiences of assuming prefectural-level party secretary

and/or city mayor posts. CYL measures whether the officials used to work at or above a

departmental level branch of the Communist Youth League (Wu 2006). I simplified my

model by creating a count variable Experiences, which ranges from 0 to 6. It measured the

number of these different work experiences that provincial leaders have. In addition, I

included the age of the provincial leaders as an independent variable. All of these variables

show the seniority of provincial leaders within the party.

2.4.5 Control Variables

Career promotion may be influenced by other confounding factors. In the baseline

models, several control variables were introduced to isolate such potential effects. First, I

followed previous studies by including Provincial GDP to control for provincial welfare

effects (Li and Zhou 2005; Opper, Nee, and Brehm 2015), which allows for the possibility

that taking charge of economically developed regions may provide crucial professional

experiences for promotion. Second, I included a dummy variable, Graduate, to show when

officials have a graduate degree or above. Finally, I incorporated a continuous variable,

Tenure, to show the number of years a provincial leader has already served in the current

post. I assume that longer office terms reduce the promotion chances of a provincial leader,

because it tends to strengthen officials’ local favoritism and decrease their linkage with the

center. Moreover, longer tenures might imply poor performance at current positions.

Table 1 Descriptive Statistics

(1)

(2)

(3)

(4)

All Leaders

(last tenure

year)

Party Secretary

(last tenure year)

Governor

(last tenure year)

All Samples

Promotion

42%

22%

61%

12%

Personal Connection

10%

12%

10%

Economic Performance

111.10

111.12

111.08

111.13

(2.33)

(2.29)

(2.36)

(2.68)

SOE

30%

28%

31%

29%

INTERPRO

40%

39%

41%

37%

REMOTE

10%

CENT

34%

0.37%

32%

34%

CITY

66%

70%

63%

65%

CYL

21%

22%

Experiences

2.00

2.04

1.96

(1.07)

(1.13)

(1.00)

(1.04)

Age

60.04

61.20

59.02

58.72

(4.19)

(4.16)

(3.95)

(4.08)

Tenure

4.37

4.60

4.17

3.19

(2.29)

(2.38)

(2.20)

(2.15)

Graduate

0.43

0.40

0.46

GDP (in trillion RMB)

1.02

1.00

1.03

1.01

(1.30)

(1.20)

(1.39)

(1.32)

Observations 358 168 190 1526

Notes: Reported in each cell are the sample means for continuous variables and the percentage for dummy variables. Reported

in parentheses is the standard deviation for continuous variables. SOE means work experiences at state-owned enterprise.

INTERPRO = work experiences in other provinces; REMOTE = work experiences in Xinjiang and/or Tibet; CENT = past

experiences of working in the central government; CITY = past experiences of assuming prefectural-level party secretary

and/or city mayor posts; CYL = past experiences of working in the Communist Youth League.

The descriptive statistics are shown in Table 1. It shows that about 42% provincial

leaders are promoted at the end of their tenure in office. But governors have much more

chances than party secretaries in career promotion, 61% and 22% respectively. About 10%

observations have personal connection with the CCP’s incumbent top leader. And over 60%

of them used to work at the prefectural leading positions. On average, each observation has at

least two working experience items.

2.5 Empirical Results

2.5.1 Baseline Models

The baseline results are presented in Table 2. A personal connection with the incumbent

supreme leader has significantly positive effects on promotion in several separate models,

including isolated regression (Model 1), regression with separate interaction terms of

economic performance (Model 2), experiences (Model 3) and age (Model 4). Even with the

inclusion of the complete set of variables and fixed effects, the effects of the personal

connection variable are significant at the five per cent level (Model 5). The odds ratio

suggests that, all else being equal, the likelihood of provincial leaders’ career promotion is

increased by approximately 105.7 per cent when there is a personal connection with the

incumbent supreme leader.

In addition to factionalism, past working experiences and age also had significantly

positive effects on provincial leaders’ career promotion. Models 3 and 5 jointly show that

richer work experiences increase the probability of career promotion, and the results are still

significant at the ten per cent level, even when all of the control variables are included.

Specifically, the odds ratio shows that the likelihood of career promotion is increased by 60.6

per cent for each additional item of past working experience. The effects of interaction

between personal connections and past working experiences are not significant once I

incorporated control variables in Model 5, which suggests that the decision of the supreme

leaders to promote provincial officials may not depend on their seniority. However, the

effects of age on promotion are more complicated. Models 4 and 5 demonstrates that the

likelihood of career promotion does not increase with age. Nevertheless, Models 4 and 5

show that the coefficient on the interaction term of personal connection and age is negative

and statistically significant at the one per cent level, implying that increased age might

decrease the likelihood of the career promotion of the connected officials.

Table 2 Baseline Results

(1)

(2)

(3)

(4)

(5)

Promotion

Odds

Ratio

Personal Connection

0.879***

0.886***

0.666**

0.629*

0.721**

2.057**

(0.313)

(0.337)

(0.276)

(0.326)

(0.365)

Economic Performance

0.034

0.064

1.066

(0.049)

(0.064)

Personal Connection*

0.127

0.042

1.043

Economic Performance

(0.089)

(0.094)

Experiences

0.360***

0.474***

1.606***

(0.120)

(0.165)

Personal Connection*

0.575**

0.359

1.432

Experiences

(0.265)

(0.291)

Age

0.029

-0.004

0.996

(0.022)

(0.030)

Personal Connection*

-0.261***

-0.246***

0.782***

Age

(0.072)

(0.092)

Tenure

0.336***

(0.072)

1.399***

Graduate

0.223

(0.259)

1.250

GDP

-0.184

(0.176)

0.832

Fixed Effects

Yes

N of Province

pseudo-R^2

0.406

0.408

0.422

0.413

0.459

Log likelihood

-324.10

-322.84

-315.38

-320.33

-295.36

Obs. 1526 1526 1526 1526 1526

Notes: Independent variables are lagged one year. Robust standard errors clustered at the province level are reported in

parentheses.

*** p<0.01, ** p<0.05, * p<0.1

In sum, first, both personal connection and seniority impose significantly positive effects

on career promotion. Second, neither performance-based nor seniority-based criteria

effectively influence the decision-making of the CCP’s supreme leaders on bureaucracy

mobility. Third, only personnel management rules about age limits constrain the CCP’s

supreme leaders’ efforts to promote their cronies into higher positions. Therefore, these

stepwise logit regressions strongly support the hypothesis (H1) of a positive association

between factionalism and career promotion chances. H3 is weakly accepted. H4 is strongly

rejected. In contrast, the evidence does not support the traditional wisdom on meritocracy.

Neither economic performance nor the interaction terms between personal connection and

economic performance significantly increase the chances of promotion. Therefore, H2 is

strongly accepted.

Table 3 Robustness Checks

(1)

(2)

(3)

Modification of

Promotion

Modification of

Personal Connection

Modification of

Economic Performance

Personal Connection

0.650*

0.655**

0.977**

(0.360)

(0.330)

(0.453)

Economic

Performance

0.071

0.086

0.057

(0.066)

(0.064)

(0.059)

Experiences

0.480***

0.512***

0.477***

(0.164)

(0.155)

(0.165)

Age

-0.009

-0.019

-0.002

(0.035)

(0.029)

(0.030)

Personal Connection*

0.027

-0.107

-0.149

Economic

Performance

(0.094)

(0.153)

(0.095)

Personal Connection*

0.260

-0.024

0.358

Experiences

(0.270)

(0.203)

(0.288)

Personal Connection*

-0.298**

-0.190*

-0.198*

Age

(0.116)

(0.101)

(0.103)

Tenure

0.310***

0.343***

0.340***

(0.069)

(0.074)

Graduate

0.430

0.240

0.232

(0.268)

(0.241)

(0.258)

GDP

-0.162

-0.204

-0.208

(0.175)

(0.176)

Fixed Effects

Yes

N of Province

pseudo-R^2

0.459

0.451

0.460

Log likelihood

-278.12

-299.65

-294.84

Obs. 1526 1526 1526

Notes: Independent variables are lagged one year. Robust standard errors clustered at the province level are reported in

parentheses.

*** p<0.01, ** p<0.05, * p<0.1

2.5.2 Robustness Checks

For robustness checks, I conducted a series of modifications on the baseline models.

First, I modified the coding strategies of Promotion, as mentioned before. Second, I changed

the coding strategies for Personal Connection. I followed some new literature (Keller 2016;

Meyer, Shih and Lee 2016) to take overlapping time and rotation of provincial unit into

consideration. In this case, a leader’s personal connection was coded as 1 if he or she used to

be a provincial subordinate of the subsequent supreme leader for at least one year, and while

working together, he or she was transferred into or out of their patron’s work unit. Third, I

took the difference of average GDP growth rate relative to adjoining provinces to measure

provincial leaders’ performance. This reflects the superiority of a given provincial leader in

regional economic competition. This variable was calculated by subtracting the mean value

of economic performance of the adjoining provinces from the economic performance of the

provincial leader’s province. Table 3 shows that they yield similar results to the baseline

models. In addition, I regressed the five indicators of past working experiences separately.

Table 4 shows that working as a party secretary and/or mayor of a prefecture-level city

increased the likelihood of promotion by 174.2 per cent when all the variables are included.

Past working experiences in state-owned enterprises, central government and Communist

Youth League also impose some positive effects on their likelihood of promotion.

Table 4 Working Experiences

(1)

(2)

(3)

(4)

(5)

Promotion

Odds Ratio

SOE

0.561*

0.618**

0.638**

0.611*

0.603*

1.827*

(0.290)

(0.296)

(0.303)

(0.327)

(0.333)

INTERPRO

0.085

0.020

0.023

0.199

0.158

1.171

(0.255)

(0.263)

(0.266)

(0.340)

(0.353)

REMOTE

0.687

0.780

0.743

0.594

0.692

1.998

(0.465)

(0.479)

(0.509)

(0.590)

CENT

0.307

0.360

0.394

0.553*

0.570*

1.769*

(0.263)

(0.273)

(0.319)

CITY

0.740***

0.822***

0.835***

0.989***

1.009***

2.742***

(0.231)

(0.254)

(0.260)

(0.281)

(0.290)

CYL

0.689***

0.634**

0.685**

0.574*

0.508

1.662

(0.265)

(0.271)

(0.297)

(0.348)

(0.370)

Personal Connection

0.894***

0.849***

0.844***

1.087***

1.095***

2.988***

(0.312)

(0.308)

(0.316)

(0.338)

(0.344)

Economic Performance

0.088

0.091*

0.101

0.083

1.086

(0.054)

(0.070)

(0.069)

Age

0.025

-0.032

-0.031

0.969

(0.026)

(0.029)

Tenure

0.350***

0.353***

1.423***

(0.075)

Graduate

0.178

(0.304)

1.195

GDP

-0.219

(0.167)

0.803

Fixed Effects

YES

N of Province

pseudo-R^2

0.423

0.425

0.426

0.455

0.456

Log likelihood

-314.55

-313.46

-313.11

-297.55

-296.58

Obs. 1526 1526 1526 1526 1526

Notes: Independent variables are one-year lagged. Robust standard errors clustered at the province level are reported in

parentheses.

*** p<0.01, ** p<0.05, * p<0.

2.5.3 Patronage Ties with PSC Members

Some quantitative studies have indicated significantly positive effects of personal ties

with the incumbent PSC members on provincial leaders’ career promotion (Jia, Kudamatsu

and Seim 2015; Opper, Nee, and Brehm 2015). However, I posited that this was caused by

incidental correlation, because the party’s supreme leaders were among the PSC members.

One way to separate the confounding effects of the supreme leaders from other PSC members

is to regress on the personal connection with them. I coded one provincial leader as having

Personal Connection with other PSC if he or she used to work with at least one incumbent

PSC member, except for the CCP’s supreme leader. For comparison, I also measured the

provincial officials’ Personal Connection with all PSC.

I regressed these variables separately using logit regression models, and incorporated the

interaction term of Personal Connection and Personal Connection with other PSC to test

their complementary effects. I also included the control variables from the periodicity

regressions. Table 5 displays the results. Model 1 suggests that personal ties with any one

PSC member increase the likelihood of promotion, but the result is not significant. Model 2

generates similar results to my baseline models. Models 3 shows that personal ties with any

other PSC members do not significantly improve the promotion chances of provincial

leaders. More interestingly, Models 4 and 5 suggest that the probability of promotion

becomes much higher after including the connection with the top, the connection with other

PSC leaders, and their interaction term. Model 5 shows that the promotion chance of those

provincial leaders exclusively connected with the incumbent general secretary is over 8 times

larger than others. However, their promotion chances declined dramatically if they also have

personal connection with other PSC leaders.

2.6 Discussion

This article, in general, shows the importance of patronage networks in affecting elite

selection of the CCP at provincial level. However, there is little evidence to support either the

effects of economic performance alone, or the complementarity of economic performance

and political connection. Given the widely accepted notion that merit-based recruitment has

played a crucial role in China’s economic and political reform after the Mao era (Chen, Li

and Zhou 2005; Li and Zhou 2005; Zhang 2009; Choi 2012, 2016), these results may seem to

run counter to common sense. Nevertheless, not all the studies on the cadres assigned to

provincial leading posts strongly support the performance hypotheses (Opper, Nee, and

Brehm 2015). One plausible explanation for this distinction is that the CCP employs a

multilevel appointment strategy in which sub-provincial level officials are promoted

according to their performance (Guo 2007; Landry 2008; Yao and Zhang 2015), while

political conformity may become more important for provincial-level leaders (Li 2014). This

is because lower-level officials are essential to regional economic development, but have no

substantial influence at the center of power. Promoting competent officials is crucial to

regional growth, while the political consequences of promoting potentially disloyal officials

are acceptable. Conversely, higher-level cadres directly influence the selection of central

leaders. A national official with a high degree of competence and a low degree of loyalty may

increase political risks for supreme leaders. Thus, performance-based recruitment at higher

levels is replaced by political concerns (Landry, Lu and Duan 2018).

Table 5 Personal Connection with PSC Members

(1)

(2)

(3)

(4)

(5)

Promotion

Odds

Ratio

Personal Connection with all PSC

0.398

(0.275)

Personal Connection

0.879***

2.303**

2.236***

9.352***

(0.313)

(0.965)

(0.841)

Personal Connection with other PSC

0.100

0.151

0.096

1.100

(0.282)

(0.309)

(0.349)

Personal Connection*

-2.001*

-1.965**

0.140**

Personal Connection with other PSC

(1.059)

(0.999)

Controls

Yes

Fixed Effects

YES

N of Province

pseudo-R^2

0.403

0.406

0.401

0.411

0.426

Log likelihood

-325.57

-324.10

-326.64

-321.26

-313.19

Obs. 1526 1526 1526 1526 1526

Notes: Independent variables are one-year lagged. Robust standard errors clustered at the province level are reported in

parentheses.

*** p<0.01, ** p<0.05, * p<0.1

More importantly, unlike some previous works (Jia, Kudamatsu and Seim 2015; Opper,

Nee, and Brehm 2015), this study accentuates the significance of personal ties with the CCP’s

supreme leaders in career promotion of China’s provincial elites. This article shows that, first,

the exclusive patronage ties with incumbent incumbent general secretary of the CCP, to the

largest extend, increases the probability of the provincial leaders’ career promotion; second,

the personal ties with any other PSC members alone can not help them to be promoted; and

third, personal connection with both incumbent general secretary and any other PSC

members only slightly increases the likelihood of promotion. In other words, the empirical

evidence does not support the complementarity between connection with supreme leaders and

connection with other top leaders. The effects of the two are offsetting. This is probably

because connection with other PSC members undermined the perceived loyalty of the

provincial officials to the incumbent supreme leaders. This finding is very important. It

further reveals not only concerns of the CCP’s top leaders on personal loyalty but also the

weaknesses of the collective leadership rules at the top.

2.7 Conclusion

Based on panel data from 1993 to 2017, this study proves that personal connections with

the CCP’s incumbent supreme leaders significantly increased the promotion chances of

provincial leaders. These positive effects did not depend on economic performance or past

work experience. Age limits and retirement rules still mattered for provincial leaders’ career

promotion. Past working experiences of being a prefectural party secretary and/or city mayor

also increased provincial leaders’ promotion chances. In addition, this study does not support

the argument that economic performance is an important determinant for the provincial

leaders’ political future. The study also challenges our understanding of the CCP’s collective

leadership. The effects of factional ties with other incumbent PSC members on career

promotion were weak. Therefore, this study strongly supports the existence of the power of

the incumbent supreme leader in selecting the CCP’s high-level leadership. The CCP’s

established institutions seem to be a dependent tool for each new supreme leader.

Given the crucial role of provincial leaders in implementing central policies, controlling

the provincial personnel is pivotal to advancing the policy agenda. The party’s supreme

leaders are able to use career incentives to induce their factional coalitions to follow their

policy preferences. In this way, they consolidate their personal control over the party as a

whole, which in turn allows them to change the status quo policies of the party-state. For

instance, Jiang’s dominance was reflected in his efforts to allow capitalists to join the party.

In this way, he transformed the class base of the CCP (Gore 2015). Another example is Xi

Jinping’s power of control. Although Xi’s seniority as a national leader was that of a

lightweight, he successfully purged the opposing factions and assigned his cronies to

important posts to control the CCP’s bureaucracy before the 19th Party Congress (Li 2016).

This was the political outcome of the anti-corruption movement. Based on this makeover, he

could move on to abolish the institutionalized tenure restriction of the national leader by

amending the constitution. In 2020, the appointment of his retired cronies to crucial

departments supervising Hong Kong seemingly signals a break in enforcing age limit rules

for assigning higher officials. Xi Jinping’s decisive determination and power once again

demonstrates the unchallenged authority available to the CCP’s supreme leaders.

Some may be interested in the rise and fall of Hu Jintao’s power as well as in his

association with the CYL. It is unclear whether the CYL is Hu Jintao’s personal faction. Past

observers often regarded the CYL as Hu’s power base (Wu 2006). Nevertheless, the evidence

showing Hu Jintao’s efforts to affect the CCP’s provincial personnel through this faction is

very weak. The effects of CYL on provincial leaders’ career promotion disappear after

including some control variables. From this standpoint, it seems that the CYL is not an

effective faction in the CCP’s bureaucracy. From another perspective, however, Xi Jinping’s

antipathy toward and vigilance against the CYL suggests that the CYL used to be an

independent faction influencing Chinese politics and possibly even challenging the power of

the supreme leaders. Was the CYL a subordinate faction under Hu Jintao’s personal control or

an independent elite group? Or should we regard it as only a valuable type of work

experience in the CCP’s bureaucracy? So far, these questions remain unresolved and require

further study.

Furthermore, This study does not tell us whether the past and incoming supreme leaders

did not have significant effects on the promotion of connected provincial leaders. But some

previous studies have shown the impacts of already designated and retired general secretaries

of the party on the composition of the party center. For instance, Meyer, Shih, and Lee (2016)

suggest that Jiang Zemin helped his factions come into the 16th Central Committee, and Xi

Jinping exerted a strong influence on the composition of the 18th Central Committee.

However, my study was able to find no strong evidence of Jiang Zemin’s or Hu Jintao’s

impact on provincial leaders’ promotion when they were not in power. It seems, rather, that

the Party Congress is a crucial component of a political mechanism in generating the next

selectorate, within which two successive generations of party leaders and their factional

coalitions compete with each other directly. The Party Congress meeting finally confirms the

legitimacy of these political outcomes (Wu 2018). After that, the newly selected supreme

leaders are able to exclude other leaders from intervening in the selection of officials to more

important provincial leading posts. Therefore, I supposed that the previous and future leaders

might influence choices of new supreme leaders, but not subsequent choices of provincial

leaders. Further studies should be done to testify such theories and assumptions.

Chapter 3: Informal Political Coalitions and Private Investment in China

Abstract. This paper attempts to estimate the effects of informal political coalitions on

China’s private investment. Theoretically, the party-state clients of China’s supreme leaders

are expected to have stronger incentives to foster economic growth. One way of doing so is to

encourage private investment by reducing its political risks. Analysis of provincial-level

panel data from 1993 to 2017 shows that personal connections—based on shared experience

in the same work unit—between provincial leaders and the Chinese Communist Party’s

incumbent supreme leader significantly increase the growth rate of private investment. This

suggests that informal institutional relations may assist the development of China’s private

economy by partially compensating for the weaknesses of formal rule-of-law institutions.

3.1 Introduction

Fixed asset investment has long played an important role in promoting China’s economic

prosperity. Compared with state-owned investment, private investment is more conducive to

efficient allocation of economic resources. Due to the importance attached to more market-

oriented economic development since the time of Deng Xiaoping’s leadership, the Chinese

government should have strong incentives to promote private investment. Overall,

conventional wisdom ascribes increasing private investment in China to the evolution of

Weberian-style formal institutions (Fu 2000). These institutions maintain a balance between

decentralization and centralization, through which the high-ups encourage their subordinates

to fulfill the party’s developmental goals (Miller 2008；Zhou 2018).

Nevertheless, informal institutions are of growing interest to China specialists. More

recent studies prove that informal politics, particularly as seen in the factionalism of the

Chinese Communist Party, help to explain a wide range of political and economic outcomes

in China (Wang 2000; Shih 2004; Bai, Hsieh, and Song 2020; Jiang and Zhang 2020). This

implies that factional ties may assist the CCP’s local bureaucracies in their efforts to compete

for economic resources and improve development performance. What are the effects of

informal political coalitions on the development of private investment in China? Do personal

connections with the Chinese Communist Party’s supreme leaders bolster provincial private

investment? This paper hypothesizes that provincial leaders having personal connections with

the CCP’s incumbent general secretary should perform better in promoting the growth rate of

overall private investment at the provincial level. Given weak institutions of private property

protection, this patronage network may encourage provincial leading figures to facilitate

private investment. It may also attract private investors by reducing the investors’ political

risks. Thus, informal institutions may not always be synonymous with inefficiency (Jiang

2018; Bai, Hsieh, and Song 2020). Instead, they may serve as a path for cooperation between

provincial officials and private entrepreneurs under the CCP’s system, which makes

connected provinces more attractive destinations for private investors.

In order to test such effects, I design a quantitative analysis on a panel of China’s

province-level administrations using time-series, cross-sectional data across 31 provinces

from 1993 to 2017. The empirical evidence is consistent with my assumption that provincial

connections with the CCP’s top leaders deliver better private investment performance. During

the sample period, the growth rate of fixed-asset investment by private firms in the provinces

having a close relationship with the top is about 15.0% higher than in other provinces. This

result persists across several robustness checks. Empirical evidence also indicate that

governors have more significant effects on private investment growth rate than provincial

party secretaries. This suggests that provincial governments exert more influence than party

agencies on regional economic policies.

The remainder of the paper is organized as follows. The next section theorizes the effects

of factional coalitions on private investment. The third section provides a research design.

The fourth section summarizes the empirical results; and the final section offers conclusions.

3.2 Personal Connection and Provincial Private Investment

Long-term, rapid economic development is typically based on formal institutions that provide

rule of law and transparent, fair market competition (North 1982). Based on this argument,

past researchers attribute China’s remarkable growth of private investment to its institutional

evolution since Deng’s reform (Ang 2016). Some assert that China’s self-sufficient and

internally diversified economy facilitates regional autonomy and reform experiments (Qian

and Xu 1993). Tax-sharing programs and state-owned enterprises reforms harden budget

constraints and compel local governments to promote growth by encouraging private

investment (Cull et al. 2017; Qian and Roland 2018). Additionally, other observers attach

great importance to promotion tournaments (Zhou 2004; Li and Zhou 2005). Specifically, the

leaders reward those who deliver economic prosperity by providing career promotion, which

in turn promotes regional competition and gives local officials enduring incentives to

promote growth. Under such pressure, local administrations may foster private sectors as a

long-term strategy of increasing economic efficiency.

However, these formal institutions seem not to be as reliable as scholars suppose. This is

partly because China has a very strong and effective authoritarian regime. First, central

government can change the rules at any time to shrink the pool available for sharing, which

may weaken the institutional incentives of economic decentralization (Cai and Treisman

2006). Next, the CCP’s leaders are inclined to appoint local officials based on personal

favoritism, as illustrated by some recent empirical studies on the homophilic features of

political career mobility paths, which leads to dysfunction of promotion tournament (Opper,

Nee, and Brehm 2015). Thus, Chinese authoritarianism may undermine the credibility of

formal institutions.

Under these conditions, property rights protection remains weak. The formal institutions

underlying economic reform may not provide adequate incentives for private investment.

However, such less-than-ideal institutions do not necessarily result in low private investment.

China’s bureaucracy has distinctive ways of regulating markets, which do not function

exactly like traditional market-oriented, rule-of-law institutions. Instead, they provide private

investors with additional inducements through factional coalitions.

Central leaders’ consensus on marketization requires economic growth as a strategy to

legitimize the regime. The development of the private economy tends to improve the

efficiency of the national economy as a whole. Arguably, then, the party’s top leadership

seeks to implement policies favorable to private sector growth. For provincial government

officials, economic decentralization enables them to have discretion in how the private-

sector-oriented policies are enforced in their jurisdiction. They can regulate the use of natural

resources, land tenure, labor welfare, investment projects, bank credit, and even environment

protection (Huang 1996; Zhou 2008), which are important for the development of the private

economy. Therefore, the provincial party secretaries and governors, as the top leaders of

provincial governments, may have more tools than the party center to boost economic growth

in their regions.

In this case, local officials face two choices. One is to seek career promotion by

following the economic development priorities coming down from above. To this end, local

governments should rationally allocate resources at hand to induce private investment. The

other one is to feather their own nests by engaging in corruption and plundering the private

economy. In this case, local officials may capitalize on the power at hand for rent-seeking.

Because the party values political loyalty when selecting senior cadres, the officials having

factional ties with the party’s top leaders are more likely to be trusted and promoted.

Therefore, the connected provincial leaders hold stronger incentives to bolster development

through inducing private investment. They may help the private investors navigate the

complexities of the bureaucratic regulatory system to gain higher returns on investment. By

contrast, local officials without connections to the top have weaker promotion prospects and a

correspondingly greater incentive to discourage private investment by engaging in corrupt

regulatory activities.

Due to frequent transfer and promotion, the provincial leading figures often have short

tenures in office of provincial secretaries and governors. Because of this, some may argue

that the provincial officials do not have sufficient time to help private sector actors realize a

higher return on long-term investments. Yet a few years spent with connected officials are

better than nothing. Such officials can help to reduce the political risks of private investment

in their jurisdictions. Moreover, those few years spent cooperating with the provincial leaders

may provide time and an advantageous position to develop other local connections, among

which the city-level leaders may provide more direct and near-term benefits (Landry 2008).

How exactly might private investors benefit from ties with well-connected local officials?

First, political backing by connected officials decreases the short-term uncertainty of private

investment. Empirical evidence shows that officials having factional ties with the party’s top

leaders are less likely to be purged in anti-corruption campaigns (Jiang and Xu 2015). Because

of this, private investment can be better protected through sustained leadership stability and

more limited, predictable corruption (Zhu and Zhang 2017a). More importantly, with the help

of the connected officials, the private investors may escape from judicial sanctions if they are

involved in such illegal business activities as rent-seeking or commercial bribery. Although

immoral, these informal channels can help the development of private economies.

Second, political backing by connected officials sustains private investors’ long-term

confidence. The entrepreneurs who tend to curry favour with promising officials are often

rewarded in business. For example, Li Shufu, a rising private entrepreneur, seemed to

establish close relationships with the local bureaucracies of Zhejiang province in the 2000s.

Li Shufu made a fortune after 2010, when the past leading figure of Zhejiang province, Xi

Jinping, became the CCP’s general secretary. Therefore, it is the prospect of future rewards

that motivates present confidence and activities. Local bureaucracies having a close

relationship with the top are more likely to be promoted in the next few years or at least to

prolong their political influence (Jia, Kudamatsu, and Seim 2015). If these connected officials

are promoted in the future, for example, by entering the Politburo Standing Committee or

becoming the CCP’s next generation supreme leader, they may enable private investors to

gain even greater commercial benefits. In this way, well-connected bureaucracies have a

better prospect of continuing the reciprocal exchange of benefits in the future, which makes

connected private entrepreneurs believe that their investments are safer and may yield greater

long-run returns.

Additionally, it is worth noting that not all the provincial leaders connected with the

power center can reduce the political risks of private investors. Elite competition among CCP

factions may stimulate intensive anti-corruption campaigns, which often occur during the

transfer of power from one provincial party leader to the next (Zhu and Zhang 2017b). When

outgoing leaders and their successors belong to competing factions, the successors tend to

purge the subordinates of their predecessors. It hinders predictability of provincial leadership

and may undermine the associated private entrepreneurs (An et al. 2016). Therefore, general

cooperation is sometimes risky for private entrepreneurs, while cooperation with the faction

of the party’s incumbent supreme leader may offset such risks. This is because the party’s

supreme leaders rarely if ever become the target of anti-corruption campaigns. They may also

preserve their political influence on personnel arrangements after retiring, which may help to

protect their old subordinates from political purges.

Based on these considerations, I contend that provinces led by those with connections to

the supreme leader are more attractive to private investors. In addition to relying on market

competition, China’s private investors are always eager to seek and exploit the more reliable

patronage networks between local officials and the top. These local factions are often related

with the CCP’s incumbent supreme leader, the general secretary. This reasoning supports the

following hypothesis:

H: Provincial private investment is higher where provincial leaders have personal

connections with the CCP’s incumbent supreme leader.

3.3 Research Design

3.3.1 Provincial Private Investment

Provincial private investment is the dependent variable. The Chinese government has a

dominance on almost all investment projects. Through national economic plans,

governmental scrutiny and approval are required for any fixed-asset investment above a fairly

low threshold (Haggard and Huang 2008). Nevertheless, private investors seem to be more

sensitive to the factions of provincial leadership than others. In addition to their benefit-

seeking motivations and flexibility in decision-making, private sectors are disadvantaged in

ideology and politics and thus have an incentive to seek political backing. In contrast, state-

owned enterprises already have built-in patrons (Chen et al. 2011). It is costly for SOEs to

curry favor with new influential leaders by shifting their investments, because it may offend

their old patrons and thus undermine the credibility of the established system of mutual

benefits. Therefore, the inflow of private investment may be affected by investors’

assessments of changes in the local political environment, whereas SOEs may be more

constrained and derive less benefit from seeking new patrons. Based on these arguments, I

take the annual growth rate of fixed asset investment by de jure private and individual firms

as the proxy for the provincial private investment growth rate. I expect that the provincial

leaders’ connections with the top may increase the private investment growth rate.

3.3.2 Measuring Factional Ties with the CCP’s Supreme Leaders

Factional ties are the main independent variable. There are two commonly used

approaches to trace personal ties among high-level officials. The explanatory approach uses

open qualitative information to identify such ties, while the structural approach uses

biographical data and quantitative analyses to code factions (Keller 2016). Through

explanatory analyses, we find that, first, Jiang Zemin and Xi Jinping developed most of their

factional networks within prefecture-level or even county-level bureaucracies when they were

in office as provincial party secretary and governor in Shanghai, Zhejiang and Fujian; second,

Hu Jintao cultivated his faction when he served as the leading figure of the Communist Youth

League. It seems quite likely that domestic private investors would be aware of such personal

connections—more so, probably, than would foreign investors. In addition, Meyer, Shih, and

Lee (2016) derived a series of structural approaches to measure factional ties. Combining the

two approaches, I define a provincial leader P as a factional member of the CCP’s incumbent

general secretary G if and only if all of the following conditions are met:

1. G served as the head of a ministerial or province-level unit, including minister,

provincial party secretary or governor.

2. While G was working in the unit, P was transferred into G’s work unit or out of G’s

work unit, except when P was transferred out to the Party Congress or the Chinese People’s

Political Consultative Conference.

3. Before or after such transfer, P and G have shared working experiences in that unit for

over one year within four administrative steps of one another.

Past research on provincial leadership ties tends to focus on the effects of provincial

party secretaries because they are the most powerful leading figures in the provincial-level

unit (Sheng 2007). But party secretaries are often focused on personnel management, while

governors may have more direct responsibility for economic policy (An et al. 2016). Thus,

the governors should be taken into consideration for their potential influence on private

investment growth. In order to combine the two, I coded the provincial independent variable

value as 1 if the factional variable value of either the party secretary or the governor was 1,

and 0 otherwise.

3.3.3 Control Variables

To isolate the confounding effects of provincial economic performance from the key

factional effects, I introduced a vector of economic control variables capturing provincial

economic conditions. They are GDP (annual provincial GDP value), GDP growth (annual

provincial GDP growth rate value), Per Capital GDP (annual provincial GDP per capita),

Manufacturing/GDP, and Service/GDP (provincial manufacturing and service industries

added values as a percentage of provincial GDP). To isolate the confounding effects of

certain key characteristics on leader competence and policy preferences, I include the Age,

college Education, and Tenure length of the provincial leaders. I also control for CYL, career

experiences in the Communist Youth League, and EXP, career experiences in some

prosperous provinces (Beijing, Tianjin, Shanghai, Jiangsu, Zhejiang, Fujian, and

Guangdong).

3.3.4 Data Sources

The data for this study come from several sources. The economic statistics were from the

National Bureau of Statistics of China website and the annually published China Statistical

Yearbooks. I changed the current economic statistics to the constant value of the year 1997

because the data of Chongqing begin in that year. The biographical information was manually

collected from Jiang’s (2018) Chinese political elite database.

3.3.5 Baseline Specifications

The hypothesis above presumes that the effects of personal connections to supreme

leaders on private investment do not change with regions and time. In this case, a two-way

fixed-effects regression makes it possible to control for unobservable heterogeneity across

provinces and years. The model equation takes the following form:

Private Investment Growthp, t+1 = α0 + α1Personal Connectionp, t + α2Ζp, t + γt + ηp + εp, t

where Private Investment Growthp, t+1 indicates the growth rate of private investment in

province p at time t+1. I set the dependent variable at t+1 to mitigate time lag and reverse

effects. Personal Connectionp, t reflects affiliation of the provincial leaders with the currently

ruling supreme leader at time t. Ζ is the vector of controls, including provincial economic

attributes and provincial leaders’ demographic characteristics. Time-related terms γt are

included to capture omitted country-wide economic and policy shocks. Province dummies ηp

capture time-invariant heterogeneity among different provinces. To avoid correlation of the

residuals with the province-level clusters, I applied clustered standard errors to estimate the

significance of the coefficients.

Table 6 Basline Models

(1)

(2)

(3)

(4)

Dependent Variable

Private

Investment

Growth (t+1)

Private Investment

Growth (t+1)

Private Investment

Growth (t+1)

Private

Investment

Growth (t+1)

Personal Connection

0.167***

0.156***

0.156**

0.150***

(0.061)

(0.046)

(0.061)

(0.047)

Constant

0.079

-0.480*

0.181

-0.326

(0.054)

(0.258)

(0.280)

(0.314)

Economic Attributes

YES

Demographic Characteristics

YES

Province Fixed-effects

YES

Year Fixed-effects

YES

Number of Provinces

Number of Observations

680

R-squared 0.579 0.587 0.578 0.583

Notes: Independent variables are one year lagged. Robust standard errors clustered at province level are reported in parentheses.

*** p<0.01, ** p<0.05, * p<0.1

3.4 Empirical Results

Table 6 presents the effects of personal ties on the province-level private investment

growth rate from 1993 to 2017. Model 1 show the most parsimonious results without control

variables. Model 2 and Model 3 add the economic attributes and demographic characteristics

respectively. Model 4 includes all of the variables in my baseline specification. They provide

remarkably similar results. Personal connections with the CCP’s incumbent supreme leader,

whether held by the province party secretaries or the province governors, produce positive

and statistically significant effects on the growth of private investment. Model 4 estimates

that, all else equal, the growth rate of fixed-asset investment by private firms in the provinces

having a close relationship with the top is about 15.0% higher than in other provinces. This is

a considerable performance premium. The private investment for each province in this study

is 171.7 billion yuan (24.2 billion US dollars) on average. Every year, a personal connection

with the top provides an additional 25.8 billion yuan (3.6 billion US dollars) of private

investment.

As robustness checks, I use several procedures to modify my models. First, I apply

alternative coding strategies for age and tenure controls by designating age dummies for over

65 years old (less than or equal to 65 years old being the reference group) and tenure length

dummies for one year, one to three years, and three to five years (more than five years being

the reference group). Second, I use alternative time-related trend terms to replace the year

fixed-effects. They are Hu Era from 2002 to 2012 and Xi Era from 2013 to 2017. Third, I

adopt the Arellano-Bond Generalized Methods of Moments procedure to estimate the

coefficients (Arellano and Bond 1991). Fourth, I follow Meyer, Shih and Lee’s (2016) broad

tie measurement to assess whether a provincial leader is connected with the CCP’s general

secretary. Here, two officials sharing a native place and institutions of higher education are

also coded as factional allies. In Table 7, Models 1-4 exhibit all these empirical results, which

are consistent with the main results in Table 6. I also regress the effects of provincial party

secretaries and governors in Model 5. It shows that, while the factions of both party

secretaries and governors have some positive effects on private investment, the effects of

governors are much more significant than those of party secretaries. Overall, my hypothesis

is strongly supported.

Table 7 Robustness Checks

(1)

(2)

(3)

(4)

(5)

Alternative

Age and

Tenure

Control

Alternative

Time-related

Trend Control

GMM

Regression

Broader

Personal

Connection

Separate

Personal

Connection

Dependent Variable

Private

Investment

Growth (t+1)

Private

Investment

Growth (t+1)

Private

Investment

Growth (t+1)

Private

Investment

Growth (t+1)

Private

Investment

Growth (t+1)

Personal Connection

0.148***

0.137***

0.170***

(0.046)

(0.030)

(0.066)

Personal Connection (Broad)

0.056**

(0.027)

Personal Connection (Sec)

0.066

(0.061)

Personal Connection (Gov)

0.166**

(0.076)

Hu Era (2003<=year<=2012)

0.006

(0.039)

Xi Era (2013<=year<=2017)

-0.421***

(0.061)

Constant

-0.492

-0.788**

-0.313

-0.230

(0.299)

(0.347)

(0.310)

(0.308)

Economic Attributes

YES

Demographic Characteristics

YES

Province Fixed-effects

YES

Year Fixed-effects

YES

Number of Provinces

Number of Observations

680

558

680

R-squared 0.584 0.256 0.577 0.584

Notes: Independent variables are lagged one year. Robust standard errors clustered at the province level are reported in

parentheses.

*** p<0.01, ** p<0.05, * p<0.1

3.5 Conclusions

This article discusses the relationship between the CCP’s patronage networks and the

growth rate of provincial private investment. Based on a quantitative research design, I find

that the provincial leaders’ personal connections with the CCP’s incumbent general secretary

significantly increase the growth rate of private investment. This might happen through

several channels, all of which can be outlined as areas for further research. First, due to career

promotion incentives, the cronies of the supreme leaders may have stronger incentives to

promote economic development by fostering the private economy. Second, cooperation with

the connected provincial leaders may foster private investment by reducing private sector

political risks. If these causal mechanisms do exist, future research should investigate in

detail the specific favorable treatments and privileges that the connected provincial leaders

provide for private investors. Case studies may offer details regarding these open or

clandestine transactions, while data analyses may show the extent to which these additional

benefits, along with the reduced political risks, stimulate private investment. In addition,

firm-level analyses may help to decipher how the factional ties interact with other individual

characteristics to affect private sector investment behavior.

China’s market-oriented reform is not a matter of expediency, but a long-term strategy

for the survival of the CCP. Due to the elites’ consensus on marketization, the CCP’s

economic reform intentions have been coherent since at least the mid-1990s. However,

because the party is not willing to fundamentally transform the authoritarian regime, its

efforts to establish market-oriented, rule-of-law institutions are greatly constrained. Thus, the

institutions underlying the development of the private economy are still weak. Nevertheless,

informal institutions seem to provide a sub-optimal alternative for privatization reform

(Dickson 2008).

Under these conditions, the implications of the informal institutions for the broader

system are of growing interest to scholars. More recent studies disagree on whether political

connections affect corporate investment efficiency and profitability of private firms. Some of

them find that political connections help to protect private firm equity values. This is because

seeking new and strong political connections may enhance private sectors’ relatively weak

position, given China’s weak legal infrastructure and discretionary, preferential treatments

and policies (Wang et al. 2018), as compared to SOEs (Chen et al. 2011). However, the

uncertainty of political turnover may reduce investment and profitability of private firms (An

et al. 2016). The results of this study offer some additional support for the hypothesis that, in

reform-era China, informal institutions may provide a partial, sub-optimal substitute for

traditional rule-of-law institutions.

3.6 Appendix

This appendix provides complementary information that is not shown in my research

note. The information includes descriptive statistics on the measures of personal connections

and the demographic characteristics of provincial leaders, models addressing potential

endogeneity problems, and additional analyses on provincial reform performance. I also test

the association between informal politics and prospects for future GDP growth. Finally, a

firm-level analysis is provided to strengthen the empirical evidence.

3.6.1. Descriptive Statistics

I manually collected demographic characteristics of provincial leaders from the Chinese

political elite database. As this dataset ends in 2015, I supplement missing demographic

information for 2016 and 2017 through other websites, including China Vitae, Wikipedia, and

Baidu Baike. I build a panel of data on provincial leadership. It takes the province-year as the

unit of analysis, incorporating the provincial party secretaries and governors from 1993 to

2017. The panel is composed of 771 valid observations, including 166 provincial party

secretaries and 212 governors. The observations of Chongqing before 1997 are excluded,

because it was not until 1997 that Chongqing became a provincial unit. The descriptive

statistics are in Table 8. Panel A reports the sample means. It shows that about 9% of all

observations have at least one provincial leader who has working ties with the CCP’s

incumbent general secretary. About 26% of observations have at least one provincial leader

who has broad ties with the CCP’s incumbent general secretary. Panel B reports sample

distributions by year. Panel C reports sample distributions by province.

Table 8 Descriptive Statistics

Panel A: Overall statistics

(1) (2) (3)

Secretary or Governor

Secretary

Governor

Measures of personal connections

Working Ties

0.09

0.06

0.05

Birth Province

0.15

0.10

0.06

Alumni

0.03

0.02

Broad Ties

0.26

0.16

0.13

Demographic Characteristics

Age

59.70

57.87

Tenure Length

3.35

3.09

College Education

0.83

0.88

Served in CLY

0.20

0.23

Served in Prosperous Provinces

0.30

0.37

Observations

771

Panel B: connection variables by year

Secretary or Governor

Secretary

Governor

Year

working ties

broad ties

working ties

broad ties

working ties

broad ties

1993

0.03

0.27

0.03

0.20

0.03

0.13

1994

0.03

0.27

0.03

0.20

0.03

0.13

1995

0.03

0.30

0.03

0.17

0.03

0.20

1996

0.03

0.30

0.03

0.17

0.03

0.20

1997

0.03

0.26

0.03

0.19

0.03

0.13

1998

0.03

0.29

0.03

0.19

0.03

0.16

1999

0.03

0.29

0.03

0.16

0.03

0.19

2000

0.03

0.39

0.03

0.19

0.03

0.23

2001

0.06

0.42

0.06

0.26

0.03

0.19

2002

0.06

0.35

0.06

0.19

0.03

0.19

2003

0.13

0.29

0.10

0.19

0.03

0.10

2004

0.13

0.29

0.10

0.19

0.03

0.10

2005

0.13

0.29

0.10

0.19

0.03

0.10

2006

0.13

0.32

0.10

0.23

0.03

0.10

2007

0.10

0.26

0.06

0.23

0.03

2008

0.10

0.23

0.00

0.13

0.10

2009

0.10

0.23

0.00

0.13

0.10

2010

0.06

0.19

0.03

0.13

0.03

0.06

2011

0.06

0.19

0.03

0.13

0.03

0.06

2012

0.06

0.19

0.06

0.16

0.00

0.03

2013

0.10

0.13

0.03

0.10

0.13

2014

0.10

0.13

0.03

0.10

0.13

2015

0.13

0.16

0.10

0.13

2016

0.19

0.26

0.13

0.19

2017 0.23 0.32

0.13

0.10

0.19

Panel C: Connection variables by province

Secretary or Governor

Secretary

Governor

province working ties broad ties

working ties broad ties

Beijing

0.04

0.20

0.04

0.00

0.16

Tianjin

0.16

0.40

0.08

0.16

0.40

Hebei

0.08

0.44

0.00

0.36

0.08

Shanxi

0.04

0.20

0.00

0.16

0.04

Inner

Mongoria

0.12

0.40

0.12

0.40

0.00

Liaoning

0.12

0.24

0.12

0.24

0.00

Jilin

0.20

0.88

0.12

0.48

0.08

0.40

Heilongjiang

0.00

0.04

0.00

0.04

Shanghai

0.44

0.48

0.40

0.44

Jiangsu

0.28

0.68

0.28

0.56

0.00

0.40

Zhejiang

0.16

0.48

0.16

0.32

0.16

0.32

Anhui

0.16

0.28

0.00

0.16

0.28

Fujian

0.08

0.36

0.00

0.16

0.08

0.20

Jiangxi

0.12

0.08

0.04

Shandong

0.00

Henan

0.08

0.00

Hubei

0.00

0.28

0.00

0.04

0.00

0.24

Hunan

0.00

0.20

0.00

0.20

0.00

Guangdong

0.00

0.20

0.00

0.20

0.00

Guangxi

0.00

0.08

0.00

0.08

0.00

Hainan

0.12

0.04

0.08

Chongqing

0.00

0.29

0.00

0.10

0.00

0.19

Sichuan

0.16

0.48

0.16

0.48

0.00

Guizhou

0.00

Yunnan

0.00

0.16

0.00

0.16

Tibet

0.28

0.36

0.00

0.08

0.28

Shaanxi

0.00

0.20

0.00

0.20

Gansu

0.04

0.00

Qinghai

0.00

0.16

0.00

0.16

Ningxia

0.00

0.36

0.00

0.36

0.00

Xinjiang

0.00

3.6.2. Endogeneity Issues

The CCP’s supreme leaders have the arbitrary power to appoint their cronies to provinces

with distinctive economic features. Therefore, the provinces exhibiting the trends of fast

growth in private investment may be systematically assigned a crony from the top, which

would create endogeneity. In order to investigate whether this problem exists, I designate a

series of dummies showing statuses ranging from two years prior to forming a connection to

two years after losing such connection. Figure 1 shows the dynamic effects of personal

connections on private investment. There is a rise of private investment precisely after

appointment of leaders connected to the top. Then it deteriorates back to the bottom two years

after these connected officials leave. It is evident that the growth rate of private investment is

significantly stimulated only when the province is connected with the top. As a result,

endogeneity does not appear to be a severe problem.

Figure 1 Dynamic Effects of Personal Connection

3.6.3 Additional Analyses on Reform Performance Incentives

This article assumes that informal political connections may encourage provincial leaders

to promote growth by fostering private investment. One way to achieve this is to reduce the

political risks of private investors in anti-corruption campaigns. In this section, I provide

additional evidence on the effects of informal politics on reform performance, which proves

that the provincial leaders who have personal connection with the CCP’s incumbent general

secretary perform better in supporting the private economy.

Arguably, the inflow of private investment indicates the investors’ assessment of their

policy environment. Under weak institutions, private investors have a low status in the

political pecking order of Chinese firms. They are inclined to seek political protection (Huang

2003). One way of doing this is to cooperate with the state-owned sectors through mixed or

even ambiguous property ownership. Haggard and Huang (2008) claim that regulatory and

other policy barriers lead to the rise of mixed property firms. Namely, private sectors register

their firms as red-hat firms to access economic resources that are exclusively reserved for

state-controlled sectors. Therefore, the ratio of private investment to state-owned investment

should be lower when the policy environment is not good for private sectors. In contrast, if

some provincial leaders improve the policy environment by liberalizing business entry and

reducing biases against the private sector, this may increase private investment. The public

sector, by contrast, may be less sensitive to the provincial leaders’ factions. Also, public

sector investment may decline because of privatization policies. I thus assume that the ratio

of private investment to public investment should be higher when the policy environment is

good for the private sector.

To verify this argument, I create two indicators for the changes of policy environment.

One is PFA/SFA Growth, the annual growth rate of the ratio of private investment to state-

owned investment. The other one is PFA/SCFA Growth, the annual growth rate of the ratio of

private investment to state-owned and collective investment. I control for the economic

attributes of each province and the demographic characteristics of the provincial leaders.

From Table 9, we can see that, regardless of which control variables are included, the

estimated effects of connected provincial leaders on policy environment are positive and

statistically significant at 1%. This implies that provincial leaders who have personal

connections with the CCP’s incumbent general secretary may perform better than other

leaders in improving the policy environment for the private economy.

Table 9 Reform Performance Incentives

(1)

(2)

(3)

(4)

Dependent Variable

PFA/SFA Growth

(t+1)

PFA/SFAGrowth

(t+1)

PFA/SCFA Growth

(t+1)

PFA/SCFA

Growth

(t+1)

Personal Connection

0.211***

0.185***

0.205***

0.179***

(0.073)

(0.064)

(0.071)

(0.065)

Constant

0.129

-0.254

0.136

0.048

(0.083)

(0.455)

(0.082)

(0.459)

Economic Attributes

YES

Demographic Characteristics

YES

Province Fixed-effects

YES

Year Fixed-effects

YES

Number of Provinces

Number of Observations

680

679

R-squared

0.169

0.267

0.263

Notes: Independent variables are one year lagged. Robust standard errors clustered at province level are reported in parentheses. Economic attributes

include GDP, GDP growth, Per Capital GDP, Manufacturing/GDP, and Service/GDP. Demographic characteristics include the provincial party

secretaries’ and governors’ age, college education, tenure length, career experiences in the Communist Youth League and in some prosperous provinces.

PFA = private fixed asset investment; SFA = state-owned fixed asset investment; SCFA= state-owned and collective fixed asset investment ***

p<0.01, ** p<0.05, * p<0.1

3.6.4 Personal Connection, Private Investment and Prospects for Future GDP Growth

This article argues that the provincial leaders’ personal connections may assist China’s

private economy by partially compensating for the weaknesses of formal rule-of-law

institutions. Moreover, the CCP’s factionalism does not necessarily result in increased overall

inefficiency. This may be because the CCP makes efforts to select new leaders from the cadres

having past experience working in some developed coastal provinces, which enables these

prosperous areas to establish closer relations with the party center. This is conducive for the

party to control economic resources and makes the economic resources tilt toward these

coastal provinces, which in turn may be one of the driving forces to reform the planned

distribution of economic resources to some inefficient northeast and western inland provinces.

Namely, the party’s factions may be actively reallocating economic resources to some highly

efficient eastern coastal areas. In this sense, the CCP’s informal politics seem not to conflict

with efficiency of resource allocation. Additionally, the private sector may be more prone to

adapt to the party’s factional politics than the state-owned sector, which helps the connected

provinces to attract private investment while avoiding crowding-out effects of state investment

on private investment. These connected provinces thus become the areas with a stronger

private economy.

Nevertheless, empirical evidence with regard to these arguments is still rare. In this case,

I design a two-stage regression to test these points. First, I regress the private investment

growth at the year t+1 on personal connection at the year t. Column (1) in Table 3 shows that,

after including the control variables of provincial GDP, GDP per capita,

Manufacturing/GDP and Service/GDP, personal connection can significantly promote private

investment. Second, I regress the provincial GDP growth at the year t+1 on private

investment at the year t. Column (2) in Table 3 shows that private investment growth can

significantly promote economic growth. Third, I regress the provincial GDP growth at the

year t+1 on personal connection of the provincial leaders at the year t. Column (3) in Table 10

shows that personal connection has no expected effects on GDP growth in the next year. This

indicates that the CCP’s factionalism may promote growth through fostering private

investment. But factionalism seems to have no direct effects on growth.

3.6.5 Firm-Level Evidence

This study shows the effects of provincial leaders’ personal connections on overall

private investment at the provincial level. Thus it uses the growth rate of provincial level

private fixed-asset investment as the dependent variable. In this section, I strengthen the

argument by using a firm-level analysis. It examines how personal connections of provincial

leaders affect corporate investment for private enterprises. In this analysis, both provincial-

level factors and firm-level characteristics can be controlled for. Table 10 Connections,

Investment and Prospects for Future GDP Growth

(1)

(2)

(3)

Dependent Variable

PFA Growth (t+1)

GDP Growth (t+1)

Personal Connection

0.153***

-0.067

(0.047)

(0.408)

Private Investment Growth

0.784*

(0.430)

GDP

0.014*

-0.456***

-0.603***

(0.007)

(0.164)

(0.172)

GDP per capita

-0.046

-5.285***

-4.440***

(0.145)

(1.649)

(1.573)

Manufacturing/GDP

0.701*

-2.444

-0.518

(0.352)

(3.567)

(3.031)

Service/GDP

0.906

-7.309

-3.989

(0.648)

(6.518)

(6.004)

Constant

-0.331

14.918***

14.286***

(0.267)

(2.967)

(2.374)

Province FE

YES

Year FE

YES

N of Province

Obs.

680

649

740

Adjusted R^2

0.584

0.648

0.614

Notes: Independent variables are one year lagged. Robust standard errors clustered at province level are reported in parentheses. PFA

= private fixed asset investment

*** p<0.01, ** p<0.05, * p<0.1

I obtained firm-level accounting data from the Chinese Industrial Enterprises Database.

This contains data on all Chinese industrial firms with sales of over RMB 5 million from

1998 to 2013. I only included private firms that met the following criteria: (1) it is not

reported as bankrupt, merged, nor closed; (2) total assets, total fixed assets and depreciation

expenses are greater than zero; (3) total debt and total sales are greater than zero; and (4) it

was established after 1900. The dependent variable is Corporate Investment at the year t+1,

which is calculated as the firm’s net fixed assets at the year t+1, minus its net fixed assets at

the year t, plus depreciation at the year t+1, divided by its total assets at the year t. The

independent variable is the provincial leaders’ Personal Connection with the CCP’s

incumbent general secretary. I control for firm-level corporate investment, industrial output,

sales growth rate, gross profits and corporate age at the year t. I also control for provincial

GDP, provincial GDP per capita, provincial GDP Growth and the ratio of provincial total

fixed asset investment to provincial GDP. Age and tenure of the provincial leaders, as well as

firm and year fixed effects, are also included in the regressions. I regress all of the enterprises

and the enterprises with total sales greater than RMB 30 million separately.

Table 11 Firm-Level Evidence

(1) All

sample

(2)

Total sales greater than RMB30

million

Dependent Variable

Corporate Investment (t+1)

Personal Connection

0.129***

0.098***

(0.023)

(0.022)

Corporate Investment

-0.005

0.001

(0.019)

(0.029)

Industrial Output

0.000

-0.000

(0.000)

Sales Growth Rate

-0.000***

(0.000)

Gross Profits

-0.000**

-0.000

(0.000)

Corporate Age

0.001

-0.000

(0.001)

GDP

0.000***

(0.000)

GDP per capita

0.000**

0.000

(0.000)

GDP Growth

0.035***

0.024**

(0.012)

(0.010)

Total Fixed Asset Investment/GDP

0.074

0.132*

(0.075)

(0.072)

Constant

-4.407***

-3.252***

(1.477)

(1.176)

Age and Tenure of the Provincial

Leaders

YES

Firm Fixed Effects

YES

Year Fixed Effects

YES

Obs.

288743

190869

Adjusted R^2 0.002 0.003

Notes: Independent variables are one year lagged. Robust standard errors clustered at province level are reported in parentheses.

*** p<0.01, ** p<0.05, * p<0.1

Table 11 shows that the estimated coefficients of Personal Connection are significantly

positive at the 1% level, which suggests that the provincial leaders’ connection with the top

may positively affect corporate investment of private enterprises. The estimated coefficient of

Personal Connection for all the enterprises in this database is -0.129, shown in column (1).

The estimated coefficient of Personal Connection for the enterprises with total sales greater

than RMB 30 million is 0.098, shown in column (2). These findings provide further evidence

that investment of private enterprises may be driven by the CCP’s informal politics.

Chapter 4: Factional Politics and Foreign Direct Investment in China

Abstract. This article uses quantitative analyses to discuss the effects of the Chinese

Communist Party’s factional politics on the regional distribution of foreign direct investment

inflows from 1993 to 2017. Empirical evidence shows that provincial leaders’ personal

connections with the CCP’s incumbent general secretary had positive and statistically

significant effects on the annual growth rate of provincial foreign direct investment inflows.

These effects were more salient in inland provinces and during Xi era. These findings

challenge the conventional wisdom about the evolution of institutions under the leadership of

the Chinese Communist Party and demonstrate the importance of informal politics in

promoting China’s economic reform and prosperity.

4.1 Introduction

China is known to have benefited from reformist policies, which have opened the country

to foreign investment since the 1970s. Arguably, the combination of regional economic

advantages and, more importantly, advanced formal institutions help governments attract

foreign investment (Henisz and Delios 2001; Li and Resnick 2003; Li and Filer 2007;

Pajunen 2008; Pajunen 2008). Legal and other regulations conducive to growth have been

regularly enacted, despite the nation’s weak institutional foundations (Ang 2016). These

newly established formal institutions could reduce the uncertainty surrounding economic

activities in China by standardizing transactions and maintaining uniformity of treatment,

encouraging foreign investors to take an optimistic view of China’s long-term prospects for

growth and profitability (Gao 2017). Some scholars thus have argued that China's long-term

competitiveness in attracting foreign investment can be explained by studying the gradual

improvement of its institutional framework (Fu 2000).

Despite improvements, however, other scholars point out that institutional barriers still

hinder China’s long-term development (Xia 2018). Even after political and economic reforms,

formal institutions remain very weak. Inefficient state-owned enterprises often monopolize

China’s economic opportunities. The government frequently intervenes in economic

transactions, ignoring established rules and principles. Collusive corruption is rampant and

widespread. All of this breaks down political accountability and undermines social trust.

Recent developments in China's domestic politics, such as the anti-corruption movement, may

further erode the code of civility and the institutional environment (Pei 2006).

In this situation, the fact that China lacks checks and balances, transparency, and the rule

of law should have dimmed its hopes of achieving economic prosperity and discouraged

foreigners from investing in China (Acemoglu and Robinson 2012). However, China’s

seemingly immature institutions still induce international enterprises to make large, long-term

investments. China has maintained rapid growth for many decades and continued to attract

international business. This raises the following questions. What are the roles of the weak

institutional foundations in attracting foreign direct investment inflows in China? Do

informal institutions to some extent compensate for the weaknesses of China’s formal legal

system in promoting foreign investment? Specifically, does the Chinese Communist Party’s

factional politics at the top affect the provincial allocation of foreign direct investment?

A recent study has argued that many private firms rely on special deals to circumvent the

obstacles presented by formal rules, thereby obtaining scarce resources and illegal benefits

from a political monopoly (Hallward-Driemeier and Pritchett 2015). In particular, Chinese

informal institutions, usually involving a patronage network within the bureaucracy, seem to

compensate for the deficiencies of formal institutions by providing special deals (Bai et al.

2020). Based on the arguments above, this study contends that factional politics within the

Chinese Communist Party may influence the extent to which foreign corporations engage with

China’s economy, generating regional discrepancies in foreign direct investment inflows. CCP

factions informally monopolize control over China's economic resources and influence the

commercial prospects of foreign investors. The pre-reform planning system augmented the

Chinese Communist Party's resource-mobilization capability. Since the 1970s, market-

oriented reforms have helped to promote economic liberalization while maintaining

China's authoritarian regime. These intact political institutions influence the way in which the

economy functions and create incentives for local bureaucracies to shape the economy’s spatial

configuration (Chan et al. 2008). Beyond their political rank in formal institutions, local cadres

who show allegiance to the top can form political factions to maintain influence over economic

resources. The factions attract foreign investors via economic resources they control because

foreign companies naturally prefer to invest in provinces run by trusted leaders.

The present study creates a provincial-year panel and designs regression models to

verify the arguments above. After controlling for various factors that past researchers

considered influential in the regional distribution of FDI, this study shows that, from 1993 to

2017, personal connections between provincial leaders and the current CCP general secretary

significantly increased FDI inflows to particular provinces. Factional connections have had

more influence over FDI inflows in inland, as opposed to coastal, provinces. The variation

over time indicates that factional politics became important in attracting foreign investment

during the Hu Jintao era. By demonstrating that factional coalitions can increase the growth

rate of provincial FDI inflows in China, this research stresses the role of informal institutions

in stimulating FDI inflows (Boisot and Child 1996; Peng 2002; Seyoum 2011; Jr et al. 2013;

Zhao et al. 2013; Munasib and Tian 2015). This paper also suggests that China's rapid

economic growth over the past three decades may not have fundamentally changed weak

institutions. Under these conditions, foreign investors may prefer to continue to seek informal

political protections through factional politics.

The remainder of the article proceeds as follows. The next two sections summarize past

research on regional allocation of FDI inflows and then offer new theory about how informal

politics have determined Chinese foreign direct investment. The fourth section covers

methodology, data, variables, model building, and descriptive statistics. The fifth section

presents the empirical evidence including the baseline results and robustness checks. The

sixth section further discusses the geography and time differences in the effects of informal

institutions. The final section presents conclusions.

4.2 Neoliberal Theories of FDI

There is a vast literature on the factors that determine the regional distribution of FDI

inflows (Faeth 2009). The Heckscher-Ohlin-Samuelson model of neoclassical trade theory

views FDI as a component of international capital trade, linked to the surplus in capital

returns between capital-abundant countries, such as the U.S., and labor-abundant countries,

such as China (Aliber 1970). To obtain these extra benefits, foreign firms must overcome

multiple difficulties, including a different culture, an unfamiliar commercial environment,

legal uncertainty, and even political risk (He 2002). They use several strategies to address

these problems. First, multinational enterprises with market power may capitalize on their

monopolistic advantages, such as product differentiation, managerial expertise, and new

technologies, to balance out the risks associated with competing with local firms in China

(Dunning 2015). Second, multinational firms are essentially products of market

internalization across national borders. Through hierarchical institutions with cohesive force,

they can avoid market deficiencies caused by ambiguous rules, opportunistic behavior, and

increased transaction costs (Buckley and Casson 2002). Third, the geographical and sectoral

diversification of some large countries may enable multinational enterprises to reduce their

investment risks. Foreign investors can circumvent the sectors where state-owned enterprises

enjoy preferential treatment and invest in market-oriented, labor-intensive industries (Lin and

Kwan 2011).

While these theories parsimoniously account for the broad profile of FDI inflows across

the world, more recent research has turned to explain why and how foreign investment flows

into certain newly emerging economies, such as China. For instance, Sun, Tong, and Yu

(2002) argue that location theory helps to explain the determinants of foreign investment

across China. They have identified eight important factors that affect China's FDI distribution:

market size and demand, concentration of economic activities, labor quality, labor costs,

scientific-research development, economic openness, political stability, and financial

competition. Their findings are supported by other quantitative research (Wei et al. 1999; Hon

et al. 2005). However, some of these economic explanations may be effects, rather than

causes, of foreign investment. For example, the inflows of foreign investment help to diffuse

advanced technologies, promoting the economic development of host countries (Hermes and

Lensink 2003). Financial deregulation has been shown to accelerate FDI technology diffusion,

which, in turn, helps to expand market size, develop scientific research, and elevate labor

quality (He et al. 2012). By contrast, economic concentration, openness, and political stability

are often functions of political institutions and industrial policies.

Thus, private investment behavior may not be shaped by market forces alone (Mudambi

and Navarra 2002; Fan et al. 2009). Economists have highlighted the role of institutions in

sustaining long-term growth (North 1990; Rutherford 1995), and scholars now discuss the

nexus between investments and non-market institutions. Institutions are rules and procedures

enforced through official channels. In this sense, institutions encompass formal rules, laws,

and regulations that are pro-competition, predictable, and transparent. They are designed to

protect property rights and reduce transaction costs, thus promoting FDI in specific locations.

According to Fu (2000), the case of FDI in China converges with this argument. The bloom

of foreign investment across China has been based on gradual reform and the solidification of

market-oriented institutions (Gao 2017). Domestic legal institutions have been established to

check sovereign discretion, improve dispute resolution, protect intellectual-property rights,

broaden market access, promise the free repatriation of profits, offer preferential taxation, and

generate other fiscal incentives. This process has boosted the confidence of foreign investors,

making them believe that their interests were sufficiently guaranteed in the long run.

4.3 Informal Politics and China’s FDI Inflows

Overall, transnational corporations view China as an attractive place to locate their

businesses, because of its large internal market, abundant low-cost workforce, improved

physical infrastructure, and easy access to capital goods (Yi et al. 2004; Liu et al. 2012). In

turn, such corporations contribute funds, advanced technologies (via capital goods),

managerial experience, and professional training for employees, all of which help to make

China one of the largest emerging economies in the world (Buckley et al. 2002; Branstetter

and Lardy 2008).

In this case, conventional theories provide holistic explanations of the surge in foreign

direct investment in China (Sethi et al. 2011). They claim that regions with greater economic

advantages and better institutions should have attracted greater FDI inflows. However, many

of them often ignore informal institutions. In comparison to formal institutions, informal

institutions are defined as unwritten rules, usually based on shared culture and values, created

and enforced outside officially sanctioned channels (Helmke and Levitsky 2006). According

to Wang (2001), such informal institutions generally consist of intimate interpersonal

connections among economic elites and influential local bureaucrats. Arguably, such social

ties can reduce information costs and alleviate commitment problems by increasing mutual

trust between foreign investors and their hosts (Lu et al. 2018). For example, scholars have

observed that investment from Hong Kong and Taiwan makes up a large proportion of the

total foreign investment in China; it is also geographically concentrated in Guangdong and

Fujian. This shows that linguistic and cultural connections among these coastal regions can

foster mutual financial flows (Wei 1995; Naughton 1996).

Moreover, the scope of such informal factors, affecting the regional distribution of FDI

in China, may go beyond linguistic and cultural ties, ultimately involving the core

characteristics of the Chinese Communist Party’s economic and political rules. Despite

decades of reforms, China's authoritarian regime has not changed fundamentally. Although

the reforms have allowed more space for local economic autonomy (Huang 1996; Landry

2008), the party center still monopolizes the authoritative distribution of political resources.

After the 1980s, the CCP tried to replace informal rules with consolidated regulatory rules

through a bureaucratic reshuffle (Su and Yang 2000; Huang 2008; Miller 2008). However, the

effectiveness of these reforms should not be overstated. Within the hierarchical bureaucracy

of the CCP, the informal distribution of power did not change fundamentally. Outside

established formal political institutions, lower-level officials form political factions,

exchanging their personal allegiance for various resources (Nathan and Tsai 1995). In other

words, the power and economic resources held by lower officials depend not only on their

formal political rank within the party and government, but also on the degree to which they

are trusted by their superiors (Dittmer 1978).

This system has produced at least two political outcomes. First, trusted officials, who

often have personal connections with people at the top, are strongly encouraged to follow the

economic goals of higher-level officials and to implement policies that favor growth (Jiang

2018). In this way, the political connections of local officials influence their policy

preferences (Tung 2014; Jiang and Zeng 2019). Second, a lower cadre with informal ties to a

powerful leader may obtain preferential treatment through network-based policy coordination

(Jiang and Zhang 2020). The preferential treatment may reduce the political risks of

investment projects and improve investors' expectations of earnings. Investors can hedge

their risks by colluding with local leaders who have informal connections with CCP top

leaders. Thus, interpersonal connections with local officials can compensate for the

institutional risks associated with a weak legal system (Wang 2001). Merchants can hedge

their risks by colluding with local leaders who have informal connections with CCP top

leaders. Foreign investors may thus prefer to invest in provinces that have closer political

connections with power center. This strategy better ensures good relationships with both local

government and the center of power.

This study therefore posits that the political networks of provincial leaders generate

privileges that appeal to foreign investors. First, provincial leaders with good connections

have a substantial advantage when it comes to obtaining scare resources controlled by central

government; various kinds of resources tilt in their direction (Shih 2004). Most importantly,

their political backing seems to give them more discretionary power to use these resources to

attract foreign investors, possibly by circumventing institutional barriers to investment. For

example, connected officials can help foreign investors circumvent performance

requirements, including export ratios and compulsory technology transfer requirements (Fu

2000). Under fiscal decentralization, they can capitalize on special channels outside

established rules to help foreign investors avoid land-conflict-related policy risks (Wu and

Heerink 2016). Second, provincial leaders linked to party heads are less likely to be defeated

in political struggles and more likely to be promoted in the next five or ten years (Jiang and

Xu 2015; Opper et al. 2015). Their ability to protect investors’ property rights thus seems

stable enough to guarantee long-term returns. This prospect of preserving long-term personal

power provides more certainty for investors for the foreseeable future. For all of these

reasons, provincial leaders’ personal links to top officials make their provinces more

attractive to foreign investors. On this basis, the following hypothesis is proposed:

H: Provincial leaders' personal connections with the incumbent supreme leader should

increase the growth rate of FDI inflows.

4.4 Methodology

4.4.1 Panel Data

To test the theories and hypotheses above, this study has gathered panel data, including

factors related to the regional distribution of FDI inflows into China, between 1993 and 2017.

This time frame is ideal because it covers the main period of economic reform and

development in China. The time-series, cross-sectional data cover 30 provinces, each of

which provides diverse incentives for foreign investors and receives vastly different levels of

foreign investment. In a country with obvious geographical differences and vast market size,

these data provide a good sample for analyzing FDI distribution. Tibet has been excluded

from the research sample because insufficient data were available. Chongqing did not

become a provincial-level administrative zone until 1997. For this reason, the data from 1993

to 1996 includes 29 provinces only.

To examine the hypothesis above, provincial leaders are the appropriate units of analysis,

because they provide the main pathways through which the CCP’s supreme leadership work

with their high-level bureaucracies to promote FDI inflows. Party secretaries and governors

are the most powerful leaders in provincial governments. Party secretaries are in charge of

political issues, including general strategies, ideological control and personnel management.

Governors often take responsibility for economic-development policies and promoting

growth. In this sense, both types of provincial leader should have influence on FDI

distribution. In the study panel, each provincial observation was matched with one party

secretary and one governor; there were 164 party secretaries and 206 governors in total.

Provincial leaders who had held office for less than a year were excluded, because they were

unlikely to be sufficiently influential.

4.4.2 FDI Inflow Growth

The main dependent variable I use in this study is the annual growth rate of provincial

FDI inflows. Specifically, foreign investment inflow is measured using actual foreign direct

investment, which includes the actual capital and loans provided by foreign investors. As this

study concerns the provincial distribution of realized FDI, annual FDI inflows are collected

from provincial statistical yearbooks from 1993 to 2017. FDI is measured in millions of U.S.

dollars. The annual reference exchange rate is used to calculate its RMB values.

4.4.3 Patronage Ties

In this case, shared working experiences serve as an affective clientelist bond, since

working in the same organization increases the probability of knowing each other and

establishing steady patronage networks. Intimate factional ties are likely to be built on

hierarchical relationships within a regional or departmental bureaucracy. Apart from that,

previous studies also claim that factional ties may appear between officials who share

birthplaces or between officials who graduate from the same universities (Shih et al. 2012;

Meyer et al. 2016). The biography of each provincial leader was therefore analyzed to

identify any personal relationships with the CCP’s top leaders. The Personal Connection was

coded as 1 if such relationships occur between a provincial party secretary or governor and a

CCP supreme leader—and otherwise 0. When a top leader leaves office, the patronage

connection disappears. This study posits that both party secretaries and governors influence

provincial reform policies. Thus, the value of the independent variable Personal Connection

was coded as 1 if the factional variable value of either a party secretary or a governor was 1,

and 0 otherwise.

4.4.4 Control Variables

Demographic Characteristics. It is assumed that some personal features of the provincial

leaders may affect their personal competence and policy preferences. For example, due to

promotion incentives, younger cadres, who are in the early stages of their careers, may be

more active in promoting investment policies. The officials with work experiences in the

Communist Youth League or in prosperous provinces may be more liberal-minded. To control

for the effects of these key demographic characteristics, I include the provincial leaders’ age,

tenure length, and career experiences in the CYL and in some prosperous provinces (Beijing,

Tianjin, Shanghai, Jiangsu, Zhejiang, Fujian, and Guangdong). Biographical data on

provincial leaders came from Jiang’s (2018) Chinese Political Elite Database, and was

double-checked using independent data sources.

Previous Performance. The previous economic performance of provinces is included to

account for the possibility that provincial leaders with informal connections to the top may be

systematically assigned to the provinces that perform better in terms of key economic

indicators, including GDP, GDP per capita, and GDP growth. All of these variables are

measured in the final year of the party secretaries’ and governors’ predecessors’ terms.

Economic Controls. Past researchers have shown that several economic and institutional

factors may influence the regional distribution of foreign investment inflows. Based on

available data, the factors included below were market size, labor costs, economic

decentralization, institutional strength, openness policy, and basic infrastructure. The given

province’s current GDP and GDP per capita capture the market-size effect. The average wage

of employees in urban units in each province has been used to capture labor costs. Economic

decentralization delegates power to local governments, which facilitates regional reform

experiments and more liberalized policies designed to attract foreign investors. I measure this

via the ratio of provincial fiscal revenue to GDP. I measure the quality of formal institutions

by using the ratio of private investment to state-owned investment. This is because greater

relative private investment may indicate a better environment for non-public economies,

including deregulation of investment projects, easy access to bank loans, rule of law and

property rights protection (Huang 2003). The number of special economic zones is used as a

proxy to measure provincial economic openness policy. Finally, highways per square

kilometer is used to capture the quality of basic infrastructure. Original data for the economic

variables are collected mainly from the NBS website, supplemented by China’s annual

statistical yearbooks. Current economic statistics were changed to constant values in 1997

prices, when the Chongqing data begin. The list of SEZs come from China’s National

Development and Reform Commission.

4.4.5 Descriptive Statistics

Table 12 displays the basic summary statistics of the main quantities of interest. Panel A

shows that the average growth rate of FDI inflows is about 16.50% and that about 30%

provinces are assigned at least one top official having a personal connection with the top.

Panel B shows that, on average, the provincial FDI growth rate was about 15.08% during

Jiang era, 23.90% during Hu era, and 4.38% during Xi’s first term. Thus, the Hu Jintao era

saw the greatest increase in inflows of foreign investment, which is consistent with the

proliferation of factional politics at that time. Panel B also shows that, albeit on the secondary

status, the inland provinces have absorbed an increasing share of foreign investment inflows.

4.4.6 Model Building

Based on the panel data, a multiple regression analysis was carried out. The two-way

fixed effects regression captured both province and year effects, testing the determinants of

FDI inflows into different regions. Many previous studies lagged location determinants by

one year to allow for the delayed effects of economic factors and to avoid the endogeneity

problem (Liu et al. 2012). The present study has assumed that international investment

decisions are handled more cautiously than other investments. Foreign investors may need

more time to collect information about the frequently changing patronage networks among

provincial bureaucracies and to accommodate any changes they encounter. Thus, the location

determinants in this model were lagged by two years to allow more time to elapse. This made

it possible to test whether foreign investment growth was influenced by informal connections.

The regression model specified for this study is as follows:

FDI Inflow Growthp, t+2 = α0 + α1Personal Connectionp, t + α2Ζp, t + α3Χp, t+1 + γt + ηp + εp, t

In this model, FDI and the personal connections of provincial leaders are the most

important variables. FDI Inflow Growthp, t+2 is the growth rate of foreign investment inflow

into province p at time t+2. Personal Connectionp, t indicates the political affiliation of

province p with the CCP supreme leader at time t. Ζ is the vector of demographic

characteristics and previous performance. Χ is the vector of economic and political factors

that are associated with FDI inflows, including GDP, GDP per capita, labour cost, economic

decentralization, institutional strength, openness policy and basic infrastructure. All of the

economic variables are standardized for comparison. The year dummies γt and province

dummies ηp were included to capture any unobservable shocks during different time periods

or among different regions. As a statistical strategy to avoid correlation between the residuals

with the province-level clusters, clustered standard errors were applied to estimate the

significance of the coefficients.

Table 12 Descriptive Statistics

Panel A: Overall statistics

Observation

(1)

Mean

Std. Dev.

Min

Max

(2)

(3)

(4)

(5)

Measures of Foreign Investment

Provincial FDI Inflows (100 Million RMB)

746

296.06

421.51

0.69

3825.35

Provincial FDI Growth Rate (%)

Measure of Political Faction

717

16.50

0.48

-80.92

427.93

Personal Connection

Demographic Characteristics

746

0.30

0.46

Age (Sec)

746

59.80

4.04

Tenure Length (Sec)

746

3.35

2.30

Working in CYL (Sec)

746

0.20

0.40

Working in Prosperous Provinces (Sec)

746

0.31

0.46

Age (Gov)

746

57.83

3.98

Tenure Length (Gov)

746

3.08

2.03

Working in CYL (Gov)

746

0.22

0.42

Working in Prosperous Provinces (Gov)

746

0.39

0.49

Economic Controls

GDP (100 M RMB)

746

10119.75

12825.44

157.7838

95383.77

GDP per capita (RMB)

746

24027.32

24751.26

1875.23

168527.5

Labour Costs

741

26334.96

23788.99

2747.73

143758.9

Revenue/GDP (%)

746

8.50

0.03

0.26

Private/State-Owned Investment (%)

717

70.07

0.69

0.02

3.98

Total Number of SEZs

746

38.68

38.29

174

Highways Per Square Kilometer

746

0.50

0.40

0.01

1.71

Panel B: Main variables by year

Year

Broad Ties

FDI Growth Rate (%)

FDI Inflows

Total (100 M RMB)

Distribution by region (%)

Inland

Coast

(1)

(2)

(3)

(4)

(5)

1993

0.31

2216.32

18.03

81.97

1994

0.28

67.57

3558.30

16.82

83.18

1995

0.31

-1.52

3331.19

16.78

83.22

1996

0.31

0.94

3553.27

16.12

83.88

1997

0.33

18.50

3861.46

19.05

80.95

1998

0.33

6.09

3946.76

18.31

81.69

1999

0.33

-3.12

3943.11

17.77

82.23

2000

0.40

7.87

4047.55

18.57

81.43

2001

0.40

7.75

4530.96

19.02

80.98

2002

0.37

31.62

5557.44

19.80

80.20

2003

0.23

24.46

6949.35

20.17

79.83

2004

0.23

13.51

6681.14

22.93

77.07

2005

0.23

23.71

7347.90

21.89

78.11

2006

0.30

26.56

8676.99

24.52

75.48

2007

0.23

31.54

9881.24

27.62

72.38

2008

0.23

7.66

10002.31

30.47

69.53

2009

0.23

5.34

10638.11

33.11

66.89

2010

0.20

14.25

11739.84

36.22

63.78

2011

0.20

138.94

25532.81

39.99

60.01

2012

0.23

-47.01

13646.59

39.53

60.47

2013

0.27

2.92

13949.66

42.18

57.82

2014

0.27

1.94

14227.60

42.63

57.37

2015

0.33

9.45

13862.70

37.62

62.38

2016

0.43

3.26

13921.13

39.42

60.58

2017

0.43

4.35

15259.97

38.30

61.70

Sum 0.30

16.50

220863.70

32.29

67.71

Table 13 Baseline Models

Dependent Variable

FDI Growth (t+2)

(1)

(2) (3) (4) (5) (6)

Personal Connection

0.084*

0.116**

0.112**

0.115**

0.118**

(0.048)

(0.053)

(0.054)

(0.052)

(0.053)

GDP (t+1)

-0.029

-0.034

-0.028

-0.033

(0.032)

(0.030)

(0.031)

GDP Per Capita (t+1)

-0.013

-0.014

-0.018

(0.053)

(0.055)

(0.057)

(0.058)

Labour Cost (t+1)

-0.029

-0.028

(0.022)

(0.026)

Economic Decentralization (t+1)

-0.015

-0.026

(0.047)

Institutional Strength (t+1)

-0.001

-0.010

(0.025)

Openness Policy (t+1)

-0.020

-0.026

(0.044)

(0.043)

Basic Infrastructure (t+1)

0.032

(0.044)

Constant

-0.038

0.692

0.643

0.615

0.536

0.511

(0.041)

(0.443)

(0.469)

(0.467)

(0.552)

(0.543)

Demographic Characteristics

Yes

Previous Performance

Yes

Province FE

Yes

Year FE

Yes

N of Province

Obs.

686

681

652

Adjusted R^2

0.424

0.427

0.426

0.429

0.431

0.430

Notes: Robust standard errors clustered at province level are shown in parentheses. ***

p<0.01, ** p<0.05, * p<0.1

4.5 Empirical Evidence

4.5.1 Baseline Results

Table 13 presents the baseline results. Overall, the empirical evidence shows that,

regardless of variation in which control variables are included, provincial leaders' personal

connections with the CCP top leader had positive and statistically significant effects on the

growth of FDI inflows into their regions. In particular, the final model in Column (6) shows

that, after including fixed effects and other controls, FDI inflow growth rates in well-

connected provinces are about 11.8 percentage points higher than those in other provinces, at

a significance level of five per cent. In this sample, the average annual FDI inflow in a

province is about 29.6 billion yuan (about 4.60 billion U.S. dollars). The well-connected

provinces receive an additional 3.49 billion yuan (about 0.54 billion U.S. dollars) in annual

FDI inflows. Personal connections have not only a statistically significant effect but also a

substantively significant economic effect on FDI distribution. Thus, the hypothesis is strongly

supported.

4.5.2 Sensitivity Analyses

As a robustness check, an Arellano-Bond Generalized Methods of Moments estimator is

adopted to account for the effects of the lagged variables. Table 14 shows that the positive

effects of personal connection remain statistically significant when adopting GMM

regressions.

Table 14 Sensitivity Analyses

Dependent Variable

FDI Growth (t+2)

(1)

(2)

(3)

Personal Connection

0.144*

0.186**

0.187**

(0.082)

(0.091)

(0.090)

Demographic Characteristics

Yes

Previous Performance

Yes

Economic Controls

Yes

Province FE

Yes

Year FE

Yes

N of Province

Obs. 627 627 561

Notes: Robust standard errors clustered at province level are shown in parentheses. ***

p<0.01, ** p<0.05, * p<0.1

4.5.3 Endogeneity Issues

One possible concern about the above results is that there may be an endogeneity

problem. The CCP’s leadership may assign their trusted followers to the provinces that are

more favorable for high FDI growth. These well-connected officials thus may appear to have

better performance in attracting foreign investment and promoting economic growth simply

because they are assigned to more desirable provinces. In order to test whether such

endogeneity issues exist, I apply three strategies. First, I conduct an event study to test the

parallel trend assumption. I make several dummies from 2 years prior to the occurrence of

connection to 4 years after the connection. The dynamic effects of connection on FDI inflows

are presented visually in Figure 2. It shows that the FDI performance premium slightly

increases only after a connected provincial leader is appointed. The performance premium

reaches its peak 2 years after connection and begins to decrease 3 years after connection.

Figure 2 Event Analysis

Second, I conduct a series of placebo tests. I change the periodicity of the dependent

variable FDI from t-3, t-2, t-1, t to t+1 and finally t+2. Table 15 displays the results. The

coefficient of personal connection is only significantly positive when the dependent variable

is FDI t+2, and not in prior years. This also shows that two years is the appropriate lag in the

regression.

Third, I directly test whether selective appointment is present. Several connection

indicators at t+1 are regressed by a series of socioeconomic variables. They are the three-year

moving average of FDI growth, GDP growth, revenue growth, and the growth of the total

number of SEZs. I also include other controls in the final model. Table 16 shows that

connection status does not appear to be significantly associated with a province’s prior

growth rates of FDI, GDP, fiscal revenue and the number of SEZs. All three tests provide

reassuring evidence that strategic appointment is not what drives the observed premium of

the FDI growth rate.

Table 15 Placebo Tests

Dependent Variable

FDI

Growth

(t-3)

FDI

Growth

(t-2)

FDI

Growth

(t-1)

FDI

Growth

(t)

FDI

Growth

(t+1)

FDI

Growth

(t+2)

(1)

(2)

(3)

(4)

(5)

(6)

Personal Connection

0.020

0.002

0.031

0.008

0.032

0.118**

(0.037)

(0.027)

(0.035)

(0.033)

(0.030)

(0.053)

Constant

2.473**

1.618**

-0.131

1.803**

1.121*

0.511

(0.936)

(0.642)

(0.596)

(0.729)

(0.563)

(0.543)

Demographic Characteristics

Yes

Previous Performance

Yes

Economic Controls

Yes

Province FE

Yes

Year FE

Yes

N of Province

Obs.

593

623

624

653

682

652

Adjusted R^2 0.448 0.441 0.443 0.432 0.418 0.430

Notes: Robust standard errors clustered at province level are shown in parentheses. ***

p<0.01, ** p<0.05, * p<0.1

4.6 Regional and Periodic Divergence

4.6.1 Inland and Coast Provinces

By showing the effects of factional connections on provincial FDI inflows, this study

highlights that the CCP’s informal politics may partially compensate for the weaknesses of

formal institutions. Also, The relative importance of potential FDI-inflow determinants may

vary across different regions in China (Liu et al. 2012). There are reasons to expect some

geographic discrepancies in the effects of CCP factional politics on foreign investment.

Relative to the interior regions, China's coastal provinces have inherent advantages in

attracting foreign investment. These advantages are reflected not just in highly developed

economies, larger market size, better education, more skilled workers, and a relatively mature

rule of law, but also in locational advantages and closer cultural ties with Hong Kong and

Taiwan. Such factors make coastal developed provinces more attractive to foreign capital. By

contrast, inland provinces have corresponding natural disadvantages when it comes to

competing for foreign investment. They may therefore be more likely to resort more to

informal political ties and adopt temporary preferential policies to attract foreign investors

(Ang 2016).

Table 16 Prior FDI Growth Trends and Connection

Dependent Variable

FDI Growth (t+2)

(1)

(2)

(3)

FDI Growth (3-year avg)

0.041

0.034

0.016

(0.136)

(0.172)

(0.164)

GDP Growth (3-year avg)

0.043

0.056

(0.035)

(0.037)

Revenue Growth (3-year avg)

0.221

0.226

(0.573)

(0.595)

SEZ Growth (3-year avg)

0.058

0.093

(0.075)

(0.095)

Constant

0.563***

-4.533

-5.858

(0.086)

(3.854)

(4.182)

Economic Controls

Yes

Year FE

Yes

Province FE

Yes

N of Province

Obs.

628

349

348

R-squared 0.146 0.149 0.159

Notes: Robust standard errors clustered at province level are shown in parentheses.

*** p<0.01, ** p<0.05, * p<0.1

To test these arguments, the following regional model is applied:

FDI Inflow Growthp, t+2 = α0 + α1PCp, t*Region + α2Ζp, t + α3Χp, t+1 + γt + ηp + εp, t where

Region indicates the coastal and inland provinces. The interaction terms, of personal

connections with region, were used to test whether the effects of provincial leaders’ patronage

networks vary significantly with regional differences.

As Table 17 shows, personal connections were more effective in inland provinces than

they were in coastal provinces. In particular, Model 3 suggests that, on average, a patronage

connection with the CCP’s top leader increases the FDI growth rate by 17.2 per cent in inland

provinces; this result is statistically significant at 5per cent. By contrast, the patronage

connections have almost no effect in coastal provinces.

Table 17 Regional Models

Dependent Variable

FDI Growth (t+2)

(1)

(2)

(3)

Personal Connection*

Inland Provinces

0.118*

(0.069)

0.159**

(0.072)

0.172**

(0.071)

Personal Connection*

Coast Provinces

0.033

(0.040)

0.043

(0.041)

0.019

(0.043)

Constant

-0.015

0.619

0.400

(0.048)

(0.453)

(0.572)

Demographic Characteristics

Yes

Previous Performance

Yes

Economic Controls

Yes

Year FE

Yes

Province FE

Yes

N of Province

Obs.

686

652

R-squared 0.424 0.428 0.433

Notes: Robust standard errors clustered at province level are shown in parentheses. ***

p<0.01, ** p<0.05, * p<0.1

4.6.2 Periodicity of Informal Politics

Past studies have suggested that China's economic institutions evolved over time (Fu

2000). According to this logic, interpersonal connections should have become less important

as formal institutions developed. During the Jiang Zemin era, rampant corruption may have

forced foreign investors to seek informal political protection. Although corruption remained a

serious problem after Hu Jintao came to power, China's membership of the WTO

strengthened formal institutions designed to attract foreign investment. As a result, it can be

argued that the role of informal institutions in attracting foreign investment would be

expected to decline. After 2012, Xi Jinping's anti-corruption movement moved a step further,

strengthening the established rules and weakening the power of informal connections. All of

these factors would have helped Xi-era China attract even more foreign investment without

recourse to personal connections.

To examine this line of argument, consider the following regional model:

FDI Inflow Growthp, t+2 = α0 + α1PCp, t*Periodicity + α2Ζp, t + α3Χp, t+1 + γt + ηp + εp, t

where Periodicity indicates the Jiang, Hu, and Xi eras. The interaction terms of personal

connection with these three periods of supreme rule are included to test whether the effects of

provincial leaders’ patronage networks depend on the tenures of CCP top leaders.

As Model 3 in Table 18 shows, after including control variables, the personal

connections of provincial leaders are effective in influencing FDI inflows only during the Hu

era. Specifically, personal connections increase provincial FDI inflows by 23.5 per cent

during the Hu era. These results are statistically significant at 5 per cent. This factor is not

important in determining FDI inflows during the Jiang or Xi eras. Thus the argument of

China’s evolution of institutions may be challenged. The development of China’s market-

oriented rule-of-law institution may be nonlinear but fluctuant. China has not yet established

a real market economy. In this case, informal politics can make up for institutional

weaknesses in promoting economic development. Table 18 Periodic Models

Dependent Variable

FDI Growth (t+2)

All Samples

(1)

All Samples

(2)

All Samples

(3)

Inland

Samples

(4)

Coast

Samples

(5)

Personal Connection*Jiang Era

-0.010

0.014

0.002

0.077

-0.040

(0.053)

(0.048)

(0.046)

(0.071)

(0.079)

Personal Connection*Hu Era

0.179**

0.226**

0.235**

0.311***

0.069

(0.080)

(0.092)

(0.095)

(0.097)

(0.042)

Personal Connection*Xi Era

0.098

0.115

0.113

0.207

0.067

(0.089)

(0.111)

(0.114)

(0.132)

(0.090)

Constant

0.011

0.443

0.196

1.144

0.374

(0.046)

(0.399)

(0.500)

(1.655)

(0.326)

Demographic Characteristics

Yes

Previous Performance

Yes

Economic Controls

Yes

Year FE

Yes

Province FE

Yes

N of Province

Obs.

686

652

412

240

R-squared 0.427 0.431 0.435 0.421 0.619

Notes: Robust standard errors clustered at province level are shown in parentheses.

*** p<0.01, ** p<0.05, * p<0.1

Do the results differ between coastal and inland provinces? One way of answering this

question is to combine the regional and periodicity models; the periodicity models are

regressed using sub-samples. One sample contains data from coastal provinces, while the

other contains data from inland provinces. Models 4 and 5 in Table 18 present the results. In

inland provinces, personal connections have a positive effect on the growth rate of provincial

FDI inflows during the Hu eras. In coastal provinces, personal connections have almost no

effect on the growth rate of provincial FDI inflows during the whole period.

4.7 Conclusion

Previous studies have generally argued that greater prevalence of informal rules

fundamentally undermine the conditions for long-term development. Based on this argument,

one must assume that the massive influx of FDI into China cannot rely on informal

institutions forever. However, the present study challenges this traditional wisdom by

showing that China's informal rules may indeed drive foreign investment and lead to regional

discrepancies in FDI inflows. Empirical evidence proves that provincial leaders' personal

connections with the CCP general secretary have both positive and statistically significant

effects on the growth rate of provincial FDI inflows. These effects were more salient in inland

provinces and during the Hu era. Given the increasing volume of FDI inflows that go to

inland provinces, factional ties indeed have great effects on the Chinese foreign investment

environment. The periodic effects of factional politics also coincide with the peak of FDI

growth. The influence of factional politics on the regional distribution of FDI was mostly

significant during Hu era, which had the fastest growth rate of FDI inflows.

From the perspective of institutional evolution, China’s economic reforms and

development have a weak institutional base, which appeared to prompt foreign investors,

during the Jiang era, to seek informal political protection. One might therefore assume that

the development of an institutional framework would decrease the effects of informal rules

during the Hu and Xi eras. However, here the empirical findings of this research present a

different picture. They show that personal connections had more effects on provincial FDI

inflows during the Hu era than the other eras. One possible explanation involves the policy

changes adopted by the CCP's leadership. Jiang Zemin adapted macroeconomic policies that

were conducive to centralization and greater efficiency (Lardy 2014). Apart from

strengthened national banking system, reformed taxation, controlled inflation, and reduced

SOE inefficiencies and bloat, Jiang utilized Shanghai Pudong New Area to offer preferential

treatments to foreign investors, including the liberalized laws and policies for attracting and

regulating foreign participation in the country’s economy (Fu 2000). All of these, arguably,

improved the overall institutional environment for foreign investors. In comparison to Jiang

Zemin and Xi Jinping, Hu Jintao was a weak political leader. Under this interpretation, within

the constraints of the CCP’s factional balance, Hu lacked the political influence to promote

additional liberalizing policies. Stagnation of SOE reforms, proliferation of governmental

intervention, and pervasive corruption weakened the investment environment (Hsueh 2011).

Thus, foreign investors relied more on local factions to obtain preferential treatment. Xi's

anti-corruption campaign strengthened the rule of the party and consolidated his personal

influence. It may also have deterred corruption and weakened the influence of factional

politics (Fabre 2017). Xi also initiate a comprehensive reform plan in 2013 to support

market-oriented development (Rosen 2014). Foreign investors may rely less on factions to

protect their commercial interests. I understand that this explanation is easily challenged.

Jiang’s liberalization policies were indeed associated with rampant corruption, while Xi’s

campaigns were far from establishing real rule-of-law institutions (Chen 2017). These issues

must be left for future research.

Future research can also examine the scope conditions of the theory. I conjecture that the

influence of factional ties on FDI is likely to extend downward to lower-level leaders. As

many city leaders are active in courting FDI, it is important to conduct research on

prefecture-level units. City-level leaders who have personal connections with the incumbent

provincial leaders may perform better in attracting FDI, especially when the provincial

leaders belong to a strong faction at the party center. This is mainly because factional leaders

may encourage their followers by providing better opportunities for career promotion. But

this hypothesis remains to be tested.

In past decades, China’s authoritarianism has successfully developed a globalized

economy. However, China has not established institutions that promote the long-term

development of a market economy. In sharp contrast to the viewpoints of classical economists,

this paper argues that the political foundations of China’s economic reforms are still very

weak. The Chinese Communist Party has preserved a strong and effective form of

authoritarianism. This regime has been able to capitalize on its massive economic resources to

deal effectively with international capitalists. It has traded China’s huge market and cheap

labor for the funds and technology needed to develop its economy. Essentially, the rules of this

commercial transaction are part of China’s informal politics. As the economic prosperity

driven by this transaction does not necessarily lead to institutional evolution, the

sustainability of this development model remains questionable. If the Western world develops

a general suspicion of and aversion to the CCP regime, China may need to reform its

institutions to retain its favorable position in a globalized world. In other words, China may

need to establish a market economy that conforms to Western standards, incorporating respect

for the rule of law, property-rights protection, and labor security, to create a more stable basis

for developing its long-term economic prosperity.

Chapter 5: Conclusion

The term “institution” refers to a platform providing rules influencing the incentives of

agents and helping to coordinate their activities. Institution constrains the behavior of

political and economic actors when they maximize their interests (Hall and Taylor 1996).

Well-established formal institutional arrangements can reduce the uncertainty of economic

activities, reduce the costs of transactions, induce entrepreneurship and private investment,

and thus increase economic efficiency and promote long-term development. Thus,

institutionalists often ascribe national economic growth to development-oriented formal

institutions. Accordingly, they believe that China’s success in reform and opening up policies

depends on formal institutional evolution that results in rule of law, high efficiency, and

transparency. In sharp contrast with these thoughts, the CCP continues to monopolize

political power and govern the state through its patronage network. In order to maintain

growth, the party should have pushed forward substantial formal institutional reforms. But

such efforts are obstructed by its lag in political reform. Although constantly adapting and

innovating to ensure success in a growing market, the CCP has never transformed its

authoritarian ruling system, which blocks China from fully building market-oriented rule-of-

law institutions. In this dissertation, several quantitative analyses are conducted to prove that

informal politics may still dominate China’s provincial bureaucracy and to some extent

compensate for the weaknesses of formal institutions in attracting non-public economic

investment.

By analyzing panel data from 1993 to 2017, this study testifies to a controversial causal

link running from patronage networks with the top to increased career promotion chances.

Specifically, the provincial leaders’ personal connections with the CCP’s incumbent general

secretary, mainly measured using shared working experiences, significantly increase their

opportunities for postings to higher bureaucratic positions. But connections with other

important top leaders do not have similar effects. Except for the general secretary,

membership in the Politburo Standing Committee members’ patronage networks appears to

reduce the likelihood of promotion. Moreover, meritocracy plays a less important role than

previously thought, at least for positions at the provincial level or above. Empirical evidence

implies that the positive effects of provincial leaders’ previous economic performance and

seniority may be exaggerated in previous studies.

These findings indicate the dominance of the CCP’s supreme leaders in making

provincial personnel arrangements. Based on career promotion incentives, connected

provincial leaders may actively promote reform policies favorable for the private sector as a

strategy to increase economic efficiency. They also increase the confidence of private

investors with regard to long-term profits. This study further shows that provincial leaders’

patronage linkages with the top significantly promote the private and foreign investment

within their jurisdictions. Empirical evidence also implies that policy inclinations in support

of market-oriented reform are consistent among the major factions. There are no salient

ideological divergences in market reform orientations among the political leaders. It is worth

noting that this situation differs from that of the 1980s, when political factions within the

party center fell into the conflicts over economic reform goals. At that time, the conservative

factions attempted to block market-oriented reform polices, whereas their political rivals

utilized reform policies to improve the efficiency of regional economic growth and thus reap

advantages in political struggles (Cai and Treisman 2006). Since then, reform consensus has

dominated the CCP’s top leadership. Factional cooperation facilitates the implementation of

reform policies. In turn, these political elites share the economic benefits of reform through

factional cooperation. Therefore, unless and until there are fundamental changes among the

CCP’s top leadership cadres, it seems that market-oriented reform has become an irreversible

trend in China.

This study tells us that the reform-era Chinese economic system combines market-

friendly development with the dominant position of government in a unique way. The

centrally managed state-owned enterprises, entrepreneurial local governments, and powerful

planning institutions have jointly influenced the state-market nexus (Zheng and Huang 2018).

But in contrast with the East Asia development model, which also incorporates a strong role

for government, China’s state apparatus uses political factions to cooperate with

entrepreneurs in an informal way. Even though China’s rule of law institutions are not full-

fledged, the CCP’s top leaders are still able to exploit their personal factional ties to fulfill

their market-oriented economic goals. Rampant corruption caused by informal institutions

and factional politics have never undermined rapid growth of China’s national economy. This

differentiates China from other new emerging countries, such as Russia, where the stagnation

of political reform contributes strongly to the stagnation of economic growth. This might be

because the Chinese government has a stronger developmental ideology and can deliver its

developmental priorities from above to the bottom mainly through informal factional

connections.

Apart from showing the uniqueness of China’s model, this study also enlarges our

understanding of the general roles of institutions in sustaining authoritarian regimes. Some

scholars argue that whether a authoritarian regime is strong enough seemingly depends on the

extent to which it can monopolize political resources (Reuter 2017). When dictators

monopolize power, they tend to exclude the political participation of other forces. If the

authoritarian leaders are relatively weak in resources to defect from bargaining with the

ruling elite groups and compel them to relinquish their own autonomy, they have to share

their political power in exchange of the assistance of the elites, or even politicize social

resources to buy the support of the public in authoritarian elections (Greene 2010). However,

institutional arrangements can provide more stable support for the survival of

authoritarianism in the long run (Boix and Svolik 2013). Apparently, a formidable party

system may be more durable than other kinds of authoritarianism (Geddes 1999; Magaloni

2008; Reuter and Robertson 2015). This is because strong authoritarian parties are effective

vehicles to facilitate cooperation and prevent factionalism. Svolik (2012) argues that, for

these parties, three functions are important. They are hierarchical assignment of service and

benefits, political control over appointments, and selective recruitment and repression. They

effectively exploit their members’ opportunism and career aspirations to create a stake in the

perpetuation of the regime among the most productive and ideologically agreeable segments

of the population.

In China’s case, the main purposes of authoritarian institutions are to simultaneously

maintain economic development and political stability. The dictatorship is legitimized via its

market-oriented reform and brilliant economic performance (Ang 2016). The Chinese

Communist Party needs to avoid the inefficiency caused by the rigid political system on the

one hand and prevent economic development from threatening its dictatorship on the other.

Therefore, at least during the early stages of reform and opening up, the CCP had to give up

its one-sided pursuit of the lower cadres’ political loyalty and encourage professional

bureaucrats to develop the economy. Under these conditions of transitioning to market-

oriented reform, the party elites developed the exchange of concrete rewards and

punishments through chains of actual acquaintance, rather than relying on abstract,

impersonal principles. Thus, it is the functioning of power-sharing mechanisms, rather than

ideological commitments, that inspires the loyalty of inferior officials to their leaders (Kung

and Chen 2011). But the party leaders still fear that allowing a group of technocrats who have

economic talents and uncertain political disloyalty to hold power may threaten the party’s

dominant position. Therefore, the party does not select senior officials according to the

meritocratic principles used to select grassroots cadres. The party leaders tend to select their

trusted acquaintances to enter the party leadership. These promoted cadres formed a ruling

elite coalition with their leaders.

China’s informal institutions manifest both flexibility and authority. But this multilevel

appointment strategy may also result in rampant corruption. Once grassroots cadres realize

the ceiling effect of career promotion, they may give up their pursuit of local economic

performance and participate in corruption for personal gains. The incentives of career

advancement are likely to be less effective in influencing marginalized or disconnected

cadres, because these cadres are outside the factional networks of the top party leaders and

less likely to be promoted even with outstanding working performance. Therefore, Chinese

experts should not be surprised by the CCP’s selective anti-corruption campaigns and should

not necessarily view them as a strategy to suppress rivals in political struggles. An alternative

explanation seems to be that the officials without high-level factional connections are more

motivated to engage in corruption. Not coincidentally, they also lack factional protection

when facing anti-corruption investigations.

Given the detrimental effects of corruption on economic growth, the Chinese Communist

Party should seek to completely eradicate the roots of corruption. Formal institutions should

be established to ensure checks and balances of power, maintain judicial independence,

tolerate press freedom and supervision by public opinion, and solidify private property

protections. DeLisle (2008) argues that, essentially, China indeed has rule-by-law institutions,

in the sense of governmental tools to effectively and consistently regulate economic, political

and social behaviour of its citizens. In China, economic laws are implemented to cultivate the

private sector; and perhaps even more importantly, to attract foreign investors who can

provide markets, capital and techniques, yet require a legal system with familiar contents.

Relying on these formal institutions, China’s authoritarian regime successfully established a

framework for market-oriented reform and economic growth. However, the Chinese legal

system differs fundamentally from conventional rule-of-law institutions. It merely allows

institutions with hierarchical constraints on the bottom, yet falls short of any substantial

restrictions on the rulers themselves. Thus, corruption cannot be eradicated in China.

Students also viewed