Brunel Business School, Brunel University, the United Kingdom
The Business School, University of Nottingham, the United Kingdom
Department of Economics, University of California, Davis, the United States
Forthcoming in World Development, 2006
# Corresponding Author: Nottingham University Business School,
Jubilee Campus Wollaton Road
Nottingham NG8 1BB,
Tel: +44 (0)115 84 66361
Fax: +44 (0)115 84 66667
Email Address: Pei.Sun@nottingham.ac.uk
The paper examines how the privatization of Chinese state-owned enterprises (SOEs) can be successfully triggered and completed. By identifying the motives of local government leaders and the constraints facing them during transition, we conclude that: first, whether local governments are motivated to privatize their SOEs depends on if the ownership transfer is expected to stimulate sufficiently high growth of local tax revenues without sacrificing bureaucrats private control benefits. Second, a specific privatization program can succeed only if it manages to satisfy the managerial cooperation constraint, the workers compensation constraint, and the bank debt-servicing constraint. The motives-cum-constraints political economy approach offers an important explanation for the pace and scope of ongoing Chinese-style privatization.
Keywords: Privatization, Enterprise Reform, Political Economy, Local Government, China
The authors greatly appreciate the access to the Year 2004 Enterprise Survey data permitted by the Institute of Enterprise Research, the Development Research Center of State Council, People’s Republic of China. We are also indebted to Stephen Green, Xiaoxuan Liu and Mike Wright for their valuable comments, and to the anonymous reviewers of World Development, whose constructive advice has helped to improve the quality of the paper significantly. The usual disclaimer applies.
Unlike the big-bang mass privatization approach adopted by the Eastern European and Former Soviet Union (EEFSU) countries, the Chinese government until the mid 1990s was still trying to improve the SOE performance by establishing market-oriented incentives1 while maintaining its ownership and control over a great majority of industrial enterprises. Since then, China has entered a new phase of enterprise reform: the government has explicitly pursued a ‘2-R’ strategy – retain government control of large enterprises that operate in the strategic sectors and retreat from small and medium-sized enterprises that operate in highly competitive markets (e.g. Green and Liu, 2005).
With regard to the restructuring of large SOEs, corporatization and stock flotation are the key measures used to privatize a fraction of government cash flow rights, in return for funds to the extent that the government is still able to maintain ultimate corporate control (Liu and Sun, 2005b). Meanwhile, hundreds of thousands of small and medium SOEs at local level have been granted privatization permission during the last decade. Moreover, a large amount of evidence shows that the major players behind the rise of privatization are local governments, especially those at municipal and county levels (Tenev and Zhang, 2002; Garnaut, Song, Tenev and Yao 2005; OECD, 2005).2 Then natural questions arise on what motivates the government to relinquish its control of industrial firms, what constraints the smooth progress of privatizations, and what economic and socio-political factors affect the dynamics of government-initiated privatization programs. Unfortunately, in comparison with a vast amount of literature devoted to assessing the extent of success of privatization by examining the profitability and operational efficiency of privatized enterprises, in-depth analyses of the factors that initially trigger privatization and further constrain its progress are relatively inadequate both in general theoretical formulations and in empirical studies.
Amongst a limited number of existing works relevant to the research questions just posed, Yarrow (1999) makes a general conjecture that the worldwide spread of privatization during the 1980s and 1990s can largely be attributed to the escalating government expenditure relative to GDP growth, together with the increasing cost of government finance. So governments privatize SOEs in order to ease their fiscal pressures. This ‘fiscal pressure’ view seems to be a parsimonious answer, but it fails to capture the deep political economy nature of the privatization decision-making process. For instance, Biais and Perotti (2002) provide a model showing how privatization can be employed as a strategic policy by incumbent politicians to maximize their probability of winning the re-election.3 Although not directly applicable to the Chinese context, the ‘political-benefit’ view is illuminating in reminding us that political and institutional factors do have a direct bearing on government officials’ privatization decision.
Moving to the developing country context, Bienen and Waterbury (1989) not only mention the significance of fiscal austerity and international donor duress in triggering privatization, but argue that ‘to assess the constraints on and possibilities for privatization, one must have a clear picture of … the gains and losses that will be sustained by the constituent elements of dominant political coalitions’ (p.618). ‘… even in nonelectoral systems, leaders must find enough support or create new bases of support to sustain privatization policies’ (p.629). Such insights are echoed and further elaborated by the World Bank (1995) in a chapter entitled ‘The Politics of SOE Reform’. On the basis of a comprehensive study of the reform experience in the developing world up to the early 1990s, the policy research report generalizes some essential conditions of a successful privatization program, notably the political desirability from the leaders’ viewpoint and the political feasibility ensured by government leaders in withstanding opposition from potential losers.
This paper closely follows the ‘political-benefit’ view suggested by the aforementioned literature in general; and the discussion of motives and constraints involved in China’s privatization in the last decade serves to update and enrich our understanding of the critical issues concerning political desirability and feasibility in particular. In fact, the World Bank (1995) did touch upon the case of transition China when discussing the two conditions. But obviously its assessment was grounded on the extremely limited, if any, formal private ownership presence in the Chinese industrial sector in the late 1980s and early 1990s. The transformation of the industrial landscape since the mid 1990s, however, prompts us to rethink how the incentive structure within the authoritarian regime and the change of wider institutional environments can interact to significantly alter the desirability and feasibility of privatization in transition China.4 In the paper we aim to answer the ‘why-privatize’ question by identifying the unique institutional settings that have shaped the incentive structure of the privatization-friendly Chinese local government leaders. Based on an analytical framework that integrates the extant China-based studies, we maintain that local government leaders will be keen to privatize their SOEs only if they can assure themselves of both higher growth of fiscal revenues and the retention of their private benefits from the privatized firms.
Moreover, the willingness of government officials to privatize SOEs alone does not necessarily translate into the success of privatization programs in practice. Rather, they are subject to a series of entangled economic and socio-political constraints that are always of country-specific nature, since privatization itself means a complete reshuffling of the extant interest structure concerning not only governments, but also managers, workers, and creditors. In the paper we highlight some essential constraints imposed by the key stakeholders in privatization, namely the managerial cooperation constraint, the workers compensation constraint, and the bank debt-servicing constraints. In particular, for a detailed privatization plan to proceed smoothly, government leaders must be able to enlist the cooperation of the SOE managers, compensate the workers to an extent they deem sufficient, and guarantee the payment of existing bank loans after the restructuring.
The rest of the paper is organized as follows: section two provides an aggregate picture of the latest ownership structure in the Chinese industrial sector. Section three is devoted to an in-depth analysis of the evolution of institutional environments during the 1990s China, which shapes the incentive structure of local leaders in making their privatization decision. Based on the recently introduced integrative framework, section four illustrates the conditions that need to be satisfied to make local leaders committed to privatization. Section five moves on to identify the key constraints that need to be overcome if the desire for ownership transfer is to be turned into reality. The sixth section presents further statistical evidence in support of the motives and constraints propositions. Section seven concludes.
THE OWNERSHIP TRANSFORMATION OF THE CHINESE INDUSTRIAL SECTOR: LATEST DEVELOPMENTS
More than twenty-five years of economic reform has completely transformed China’s industrial landscape to the extent that, at the beginning of the century, according to the official statistics shown in Table 1, the private sector already accounted for more than 40% of national industrial output and employment, in comparison with the SOE sector that now only accounts for less than one quarter in the two aggregate indicators. This is in stark contrast to the composition in 1980, when the state, collective and private sectors respectively contributed 76%, 23.5% and 0.5% of the national industrial output.
It is worth noting, however, that the official data acts, at best, as a crude estimate, since the mixed ownership structure has become commonplace in corporate China, and is present in virtually all enterprises however they are classified in Table 1. In particular, the collective sector shown in Table 1 is a highly ambiguous ownership category, as a large number of listed and limited liability companies, or even the collective firms, are ultimately controlled by either government units or the private ones (e.g. Liu & Sun, 2005a, b). An early attempt to remedy the problem was undertaken by us through the use of the China Statistical Bureau 2001 national survey data. By combining the survey data and the official statistics, we find that, in terms of ultimate ownership, the state, the collective, and the private in 2001 respectively accounted for 31%, 12% and 36% of China’s industrial output.5 Subsequent estimation efforts in this respect, despite some arbitrary assumptions that have to be made, have managed to illustrate a clear trend of ownership evolution from the late 1990s to the early 2000s (Garnaut et al, 2005; OECD, 2005). For example, Table 2 suggests a gradual but steady rise of the private sector during 1998-2003: by the end of 2003, domestic private firms had gained an equal weight to that of the state-controlled ones in their contribution to China’s GDP. If we focus on the industrial/non-farm business sector, Table 3 reports more rapid growth of the private sector and decline of state ownership during the same period, with the collective segment nearly halving.
It goes without saying that the change of ownership structure at the macro level is accompanied by significant divestment and transformation of the SOEs across industries and regions.6 While a complete record of SOE ownership restructuring is not available in China, the latest national-scale survey7 reveals both the pace and the main forces driving the restructuring, details of which are shown in Table 4. It can readily be seen from the Table that the pace of ownership restructuring was gradual but gained momentum at the start of the century. Notwithstanding the potential over-sampling of non-restructured SOEs in our survey,8 more than 42% of firms have been fully or partially privatized during the whole period under investigation.9 And within the restructured firms, more than three quarters undertook the transformation during 2001-2003. Regarding the choice of full versus partial privatization, unfortunately only 997 firms reported the information. It is found that only 31.4% of the firms have undergone a full privatization so far, which suggests that a limited amount of control transfer has taken place.
Moreover, a breakdown of the sample SOEs based on central and local government affiliation indicates that it is local governments that have been keen to promote the ownership restructuring programs over the past few years. While less than one quarter of the centrally controlled SOEs in our sample experienced ownership restructuring, more than one half of their local counterparts were already privatized by the mid 2004. The difference in the choice of full versus partial privatization is also significant in the sub-sample just mentioned: nearly 40% of the restructured local SOEs can be classified as full privatization, in contrast to 10% in the transformed central SOEs.
Overall, the preceding data presentation implies two stylized facts on the pace and scope of China’s ownership restructuring over the last decade. First, the progress of the transformation is gradual but steady, with a mixture of full and partial privatization arrangements. Second, SOEs controlled by local governments take the lead in the transition process both in terms of scale measured by the percentage of firms having been restructured, and of depth indicated by the proportion of firms that have undergone full privatization.
CHANGING INCENTIVE STRUCTURE OF LOCAL GOVERNMENT LEADERS: AN INSTITUTIONAL ENVIRONMENT PERSPECTIVE
Given the latest developments of the ongoing privatization in China, in the section we examine the multifaceted institutional forces responsible for the emergence of the pro-privatization attitude on the part of local government leaders. While several pieces of work have already touched upon the causes of privatization in China, they either omit some of the key institutional factors driving privatization (Qian & Roland, 1998; Cao, Qian, & Weingast, 1999; Li, Li, & Zhang, 2000), or amount to little more than a collection of loosely knitted hypotheses (Garnault et al, 2005; Guo & Yao, 2005).
Specifically, Qian & Roland (1998), Cao et al (1999), and Li et al (2000) address the question by means of the following reasoning: first, the fiscal and monetary recentralization in the 1990s contributed to the significant hardening of budget constraints on local governments, which means that they now have to assume the primary responsibility for local economies, in correspondence with their control of independent revenue sources granted by the 1980s decentralization. Second, the intensification of cross-regional competition in product markets is the driving force behind local governments privatizing their enterprises to supply more incentives to managers and to improve the enterprises’ competitiveness, which in turn increase their fiscal revenue. These arguments do touch upon the fundamental contributing forces that trigger the Chinese-style privatization, but as will be shown below, the combination of fiscal reform and intensified market competition itself is still not adequate to nurture the appropriate conducive institutional environment for privatization. More crucially, none of them explicitly analyze the cadre management system during the reform era and the private benefits the local bureaucrats can capture from the state and privatized firms, which can play a critical part in their decision-making.
Commissioned by the IFC/World Bank, Garnaut et al (2005) offer a more up-to-date review of the latest ownership restructuring process in China. With respect to the causes of privatization, they have identified the main players during the institutional change and grouped the incentives of privatization into several hypotheses. In their companion paper, Guo & Yao (2005) use firm-level panel data to test the validity of the hypotheses.10 Although the acceptance of some hypotheses whilst rejecting the others is informative to those who are interested in China’s privatization, their analysis lacks a general theoretical framework that can integrate these hypotheses in a systematic fashion, not to mention the ad hoc classification of the hypotheses themselves.
Building on the existing literature on the complex privatization process in China, the paper constructs an integrative and coherent theoretical framework which focuses on the dynamic interaction between the incentive structure of local government leaders and the changing institutional environment. On the one hand, the explicit specification of the payoff structure of local leaders enhances the analytical clarity of existing analyses. Previous studies on the evolution of the institutional environment, on the other hand, are systematically organized in our framework to highlight their far-reaching impacts upon the leaders’ decision-making through their own payoff structure. The rest of the section is then devoted to a detailed discussion of the institutional environment relevant to the Chinese-style privatization.
(a) Decentralized state asset administration: Prerequisite of Chinese-style privatization
Unlike the highly centralized EEFSU state asset management system in which a great majority of SOEs are organized along the central line ministries, a substantial number of Chinese SOEs, most of which are small and medium ones, are assigned to the hands of local governments at provincial, municipal and county levels even in the pre-reform era. Furthermore, the income rights, i.e. the tax and profit collection, are closely tied to the specific tier of government control delegated (e.g. Qian & Xu, 1993). Clearly, such an institutional legacy is a precondition for local government leaders at various levels to be the key players in the Chinese-style privatization game.
Another ramification of the institutional arrangement is that local bureaucrats are relatively more immune from fossilized ideological constraints, so when the central government endorses or acquiesces to some reform measures, local leaders will actively promote them provided that it is in their own interests to do so. The ‘reform-from-below’ story fits the Chinese privatization experience quite well: The Zhucheng county in the Shandong province and the Shunde county in the Guangdong province became the pioneers of privatizing their small SOEs as early as 1992 (Cao et al., 1999) when the central government turned a blind eye to their behavior that was then deemed very controversial.
(b) Cadre evaluation system in reform China: The political environment
Being free from excessive political ideology itself, however, is not enough to trigger privatization; local bureaucrats need to discern net political and/or economic gains from doing so. The fiscal federalism and competition story mentioned above is of course essential but still incomplete. We believe that one of the critical missing variables in their otherwise perfect story is the political centralization via the communist party’s evolving cadre control mechanism, or in the comparative politics scholars’ terminology, the nomenklatura.11 This variable enters our integrated analytical framework when we consider the local leaders’ objective functions explicitly. It seems reasonable to assume that local leaders are mainly concerned with their political career path and the private benefits they could derive from their control of local SOEs.
As far as the first part is concerned, political centralization matters because the political future of top regional government officials is in the firm grip of the centre (Burns, 1994; Chan, 2004; Edin, 2003). Then the most important issue for the local leaders who are intent on career advancement is to know the cadre evaluation criteria. While party loyalty and personal connections are essential in the evaluation system, there is ample evidence suggesting that the economic growth record during a leader’s tenure has become a critical factor affecting his or her superior’s promotion decision. Using turnover data of top provincial leaders during 1979-2002, Li & Zhou (2005) and Chen, Li, & Zhou (2005) find a significantly positive correlation between the likelihood of promotion and the average GDP growth rates during their tenure in office, after controlling for other key factors such as the leaders’ connections with the central government. With respect to leading cadres in the lower echelons of the hierarchical regime, Whiting (2001, pp. 101-110) offers detailed case studies to support her claim that not only the incomes of township and village leaders but also their opportunities for career advancement are closely related to their performance in rural industry.
This is not to say that nowadays local leaders in the Chinese nomenklatura have a perfectly stable set of expectations regarding the promotion criteria. Of course there can be a long list of contingencies, or ‘soft targets’, that the center may use to make its promotion decision. The point here, however, is that all else being equal, local economic growth records tend to be a key ‘hard target’ by which the cadres’ performance is evaluated. 12, 13 Aside from the large weight of economic performance assigned by the system, it is worth noting that the centre also specifies some ‘priority targets with veto power’ (Edin 2003, p.39) for local leading cadres. According to Edin, ‘veto power implies that if township leaders fail to attain these targets, this would cancel out all other work performance, however successful, in the comprehensive evaluation at the end of the year’. In this regard, social stability is commonly believed to be a paramount target with veto power. By way of illustration, the occurrence of large-scale demonstration in a region will definitely ruin the political career of its leaders, no matter how much local economy growth they have helped to achieve. Therefore, the Chinese cadre evaluation system during the reform era rewards the promotion of local growth on the condition that social stability is not in serious jeopardy. This observation has a significant bearing on our subsequent discussion of the constraints of the Chinese-style privatization.
(c) Hardened budget constraints on local governments: The economic environment
Given the local developmental imperatives stimulated by the cadre evaluation mechanism, the centralization of the state-controlled banking system, together with fiscal reform and intensified market competition, shapes the route local leaders have to take in order to achieve superior economic performance. Simply put, the room for manipulating GDP growth without bona fide improvement of the performance of local firms, albeit existent, has been significantly confined relative to that before the mid 1990s.
(i) Banking reform
Before the centralization of the national banking system, local governments used to influence easily local bank credit decisions with the aim of re-capitalizing loss-making SOEs and financing large investment projects that could boost short-term GDP figures. In official terms, local banks in China are vertically controlled by their superiors and headquarters, but local leaders have various formal and informal channels available to exert their administrative leverage over credit allocations (Huang, 1996; Lardy, 1998). For example, local officials have a say, at least of a reference nature, in the appointment of the head of local banks in the nomenklatura.14 On the part of bank managers, their remuneration was once independent of performance, and they could also be lured to capture relationship-specific benefits with local bureaucrats through their own lending decisions (Brandt, Li, & Roberts, 2005).
The second half of the 1990s witnessed considerable centralization and commercialization of the banking system: while separating the commercial from the policy lending by establishing new policy banks to take over the function from the four largest state-owned ones, the centre promulgated a commercial banking law in the hope of insulating local branches from local government interference in lending decisions (Lardy 1998, pp.180-181). It is reported that the central government leaders openly ordered provincial governors and mayors not to command local bank presidents (Lardy 1998, p.207). The initiative of improving local bank autonomy has been further bolstered by the institutional reform that replaced the central bank’s provincial branches with newly-established regional/supra-provincial ones, so that provincial leaders have been less able to influence the flow of bank loans to their own jurisdiction.15 Equally important is the introduction of performance-based evaluation criteria for bank managers. Brandt et al. (2005) report that nowadays bank managers’ bonuses are tied, among others, to their performance in risk control measured by the reduction of nonperforming loans.16 In short, the ongoing banking reform both in terms of the relation between branch managers and local governments and in the incentive structure of bank managers has arguably made it more and more difficult for the local leaders to localize the benefits and socialize the risks of their investment projects. Additionally, the reform has had important ramifications on the constraint side of the ownership restructuring process, which will be detailed later on.
(ii) Fiscal reform
Coupled with the ongoing banking reform is the largely completed fiscal reform, which has institutionalized the division of tax revenues between the central and local governments.17 Before that, the Chinese government adopted a fiscal contracting system – a bottom-up revenue-sharing arrangement that required localities to remit only a portion of revenues they collected while retaining the remainder. When the provinces pooled all the tax items, they entered an intense bargaining process with the centre to decide an ad hoc sharing ratio or a lump-sum amount year by year. From 1994 onwards, the tax-assignment system (fenshui zhi) has been introduced to standardize the demarcation of central and local revenue sources by explicitly stipulating the rate and base of central, local and shared taxes. Furthermore, tax administration has been recentralized to the extent that the centre established its own revenue collection body – the national tax service – to collect central and shared taxes,18 leaving the local tax service in charge of local tax collection only.
(iii) National market integration
In sum, the profound institutional changes in the fiscal and banking systems throughout the 1990s have substantially hardened the budget constraint of local governments, so that fast economic growth, even in the short term, seems impossible without a high-performing industrial sector. On the other hand, it is difficult nowadays for the profitability of the local state-owned industry to be propped up by local government via the erection of trade barriers.19 In addition to the evidence of growing market competition in the 1980s provided by Liu and Garino (2001), a study of 32 two-digit industries across Chinese provinces during 1985-1997 (Bai, Du, Tao, & Tong, 2004) reveals a significant rise in regional specialization since the end of the 1980s, reflecting a gradual decline of inter-regional trade barriers. Given that their result is based on provincial-level data, it is safe to argue that the degree of market integration would be even higher at municipal and county levels, due to considerably less political leverage that enables the grass-roots level leaders to effectively practice local protectionism. Therefore, the combination of hardened budget constraints and a leveled market playing field will leave local government leaders, sooner or later, interested in improving enterprise performance through privatization.
Figure 1 illustrates the convolutions of our analytical framework, which integrates all the key background variables that are crucial to triggering the Chinese-style privatization.
WHAT MOTIVATES THE CHINESE-STYLE PRIVATIZATION? THE NECESSARY CONDITIONS
Based on our discussion of the evolving institutional environments, in this section we demonstrate the conditions that are necessary to motivate local government leaders in launching privatization programs. Suppose that before privatization the payoff of local government leaders drives from three sources: local tax revenues including both turnover and corporate income tax; dividends or profit remittances arising from the ownership claims of the state firms; and the private control benefits that can be captured from the local SOEs, such as perquisites either in pecuniary form or otherwise.
It is self-evident that whether firms are state-owned or not, local governments can always benefit from sales and income tax revenues within their own jurisdiction. In addition, business and firm growth will exert a significantly positive effect on local tax revenues. Thus it would be rational for an open-minded government to consider any form of ownership for an enterprise as long as it can bring more revenues into the government’s coffer and create more employment opportunities in the locality. On the other hand, when the domestic market becomes more and more integrated across regions, profit margins will be considerably eroded by increased competition, making corporate income tax and dividends/remittances a much less significant part of the local revenues. This implies that sales taxes levied on products sold will become the foremost concern for a revenue-seeking government. In short, the argument can be stated as follows:
Proposition 1: In a competitive market a revenue-seeking local government can be motivated to allow its firms to be privatized if it expects that privatized firms can induce higher sales and faster growth than would be the case if they remained under state ownership. Another variable that enters the calculation of local leaders is the control benefits they can still expect to capture after the firm is privatized. The range of these benefits can span from pecuniary favors to political interests. Indeed, one of the most important cases for privatization in economic literature is that it is more costly for bureaucrats to capture political/private benefits from a privatized firm than from an SOE (Boycko, Shleifer & Vishny, 1996). So if local leaders are afraid of the loss of private control benefits, privatization will have less chance to proceed. Nevertheless, there are reasons to believe that numerous mechanisms are at work to ensure that the control benefits are sustained. In particular, a collusive or symbiotic relationship between government officials and the new owners can be developed in such a way that the latter pays the former generously in exchange for rent-generating and other advantages. The cost of such a practice may not be as prohibitively high as suggested by Boycko et al (1996), whose study is couched in a developed democratic regime with a sound legal and institutional environment.20 Instead, the prevalence of government capture has already been empirically documented in the EEFSU transition economies (Hellman, Jones & Kaufmann 2003).
While it is technically difficult to measure the exact amount of private benefits due to the political infeasibility of accessing the related data in China, plenty of anecdotal evidence illustrates the enduring influence of local governments in privatized firms and the emerging patron-client relations between local authorities and the private sector. By virtue of a field study in Xihu (a suburban district of Hangzhou, the provincial capital of Zhejiang) during 1992-1996, Sargeson & Zhang (1999) find that local government officials as the organizers of the property-rights reform set ‘the rules of the game’ in a way to ensure that they could continue to interfere in the restructured enterprises, extract revenues for their arbitrary use, and even translate their political influence into individual shareholdings or other assets. As for the fledging private firms, Wank (1995, 1999), based on his nineteen months of fieldwork from 1988 to 1995 in Xiamen, a port city in Southeast China, portrays pervading networks of clientelist exchange between the non-state sector and the local bureaucracy. And he argues that the strategy of the entrepreneurs to cultivate symbiotic clientelism, in the form of bribes and offers of company stocks and board membership, is a rational response to a deficient legal and regulatory environment filled with both rent-seeking opportunities and government predation risks.
One of the most sensational cases comes from the Nanhai Huaguang Co. Ltd, once ‘a model private business’ acclaimed officially by Nanhai city government, Guangdong province. According to media coverage (e.g. Caijing, July 5, 2004) and central government audit, from 1999 to 2002 FENG Mingchang, the company boss, colluded closely with the cadres of the city Fiscal Bureau to take a total of RMB 7.4 billion loans from local state banks by fraud. As we have discussed before, nowadays city government officials can no longer order bank managers to lend money to a certain firm; hard collaterals are required by the banks for credit application. However, it is revealed that various city government agencies were engaged in producing a large number of credentials for fake collaterals (e.g. non-existent or other parties’ land, factories, etc.). In return, more than RMB 2 billion bank loans were transferred to the overseas saving accounts of government officials and Feng’s family as well as to the city government and Huaguang’s overseas subsidiaries. It is no wonder, therefore, why local residents called Huaguang the ‘No. 2 Fiscal Bureau’. Perhaps more revealing than the situations mentioned above is the following fact: Huaguang was initially a collective firm established by a township government in Nanhai in 1990. Despite its privatization in the mid 1990s, there has been a deep involvement of government officials in the company. First, the CFO of the privatized Huaguang is an official of the city Fiscal Bureau whose rank is equivalent to vice-head. And it is said that he is the main coordinator of the financial fraud. Second, the son of the city Fiscal Bureau head owns 40% shares of Huaguang’s parent company in Hong Kong, with Feng and his wife holding the remaining 60%. Finally, the prop-up of Huaguang went beyond the bureau level, as the ex Party secretary of Nanhai city, widely believed a firm supporter of Huaguang, was forced in September 2004 to leave his current more senior post due to the scandal (Caijing, October 4, 2004).21 In view of the previous studies and the most recent case we present, there is reason to believe that local officials can perceive the workability of a collusive relationship with the private sector. And consequently we have the following refined proposition:
Proposition 2: In a competitive market a revenue-seeking local government will be motivated to allow its firms to be privatized if it expects that privatized firms can induce higher sales and faster growth than would be the case if the firm remained under state ownership and if the benefits officials derive from these firms can be sustained. Starting from the formulation of local bureaucrats’ payoff structure, the preceding analysis has identified the necessary conditions for triggering the Chinese-style privatization. Since the local leaders’ payoff is effectively a weighted average between their career prospects and private benefits, privatization is more likely to occur if local governments can trade state ownership for revenue growth and trade policy favors for private benefits. Furthermore, it can easily be seen that, no matter how different the weights assigned to the two variables by the local leaders, there are sufficient incentives for them to make the privatization decision.
STAKEHOLDERS IN LOCAL SOEs: OVERCOMING THE CONSTRAINTS OF CHINESE-STYLE PRIVATIZATION
Privatization in China, like that in the rest of the world, is not free from constraints. Since privatization entails a fundamental redistribution of interests of all the stakeholders involved in a local SOE, including managers, workers and banks, the presence of an ideology-neutral, growth-minded local government is not sufficient to guarantee the success of a particular enterprise restructuring program. Here, by explicitly examining three key players in the complicated process, we identify the crucial constraints that need to be overcome for a successful privatization.
(a) Managerial cooperation constraint
In fear of losing control benefits, firm managers are always fiercely opposing external private investors to acquire local SOEs, unless the outsiders can either assure them of their current management positions after the takeover or offer a large amount of side-payment to buy them out. It is well-known in China that, without the cooperation of incumbent managers, any enterprise restructuring program is in great danger of entering into a deadlock.22 For instance, pricing an SOE or its assets is a complex business, which involves calculating or estimating the preset value of future profit stream. As potential buyers, firm managers have strong incentives to reduce the sale price. Compared to incumbent managers, the government is at an obvious disadvantage in getting the true information concerned. Also, using expected post-privatization profit stream to value the firm may lead to a price higher than the reservation one of the management, which makes the privatization less likely to happen. Keeping this in mind, we can understand why a simple pricing method is in principle applied in China – the net asset pricing. That is, the sale price is in general set to equal the difference between the book value of a firm’s total assets and the sum of its depreciation and total debts outstanding. However, such a practice leaves room for the managers to manipulate the book value upon expectation of the ownership restructuring.
It is expected that the management will participate in privatization if the sale price is lower than the management’s reservation price, which is close to the true economic value of the firm. Obviously the difference between the two price levels is the rent the manager hopes to seek during the privatization. And the local government officials seem to accept the use of discount as a purchase incentive to ensure managerial participation. In practice, it has been found that even the net asset pricing rule is not strictly abided by, as the sale price sometimes is even lower than the book value of net assets. For instance, Chen (2005) reports the pricing practice of a large TVE in the South of Jiangsu province in the early 2000s: the company director paid RMB 3.84 million to acquire the stakes held by township and village governments at the book value of RMB 7.3 million.
The recent development of privatization in China indicates that acquisition of SOEs by external investors is evident as well. But it is wrong to believe that the outsiders’ move can proceed smoothly without winning the essential support from the incumbent management. The 2003 takeover of Nanjing Steel Co. Ltd by the Shanghai-based Fosun Group, an upstart private conglomerate, perfectly illustrates the point.23 Before 2003, Nanjing Steel was a publicly listed company with its ultimate controlling shareholders being Jiangsu provincial government and Nanjing municipal government. In the first half of 2003, a series of secretive negotiations among the Fosun Group, the local bureaucrats and the company management resulted in an off-market block share transfer, making the Fosun Group the new controlling shareholder of the company. The unusual part of the story is that, unlike typical outsider privatizations, there was no change of senior managers in the company at all after the takeover. Our interview with a Fosun Group manager directly involved in the transaction helps answer the puzzle: ‘The key is to get the support from the company management. They have played a big part in persuading the government to sell the profitable firm to us. So that’s the deal.’
In summary, the foregoing analysis suggests the essential role of company management in the privatization process. Whether in the form of significant asset under-pricing or complicated behind-the-scenes contracts containing side payments, obtaining managerial cooperation is a must for both local governments and new owners.
(b) Workers compensation constraint
During ownership transformation, SOE workers need to be compensated as a result of the loss of welfare benefits and lifetime job security they were formerly entitled to by state employment status. In regulatory sense, central government requires that ownership restructuring programs be approved by the employee general meetings. And the Supreme Court rules that the settlement of labor claims is the first priority when a firm undergoes ownership restructuring or bankruptcy (Garnaut et al., 2005).
Although employees are not as powerful as managers in local SOEs, they do have considerable bargaining powers because of their potential influence on social stability. In other words, if their interests are seriously harmed in the restructuring process, they can voice their disgruntlement by staging demonstrations. Based on our previous discussion of the cadre evaluation system, the incidence of social unrest in a region in question, as a veto power in the system, could exert a devastating impact on the local leaders’ career prospects.24 Consequently, local governments have to ensure that the sale price is higher than some bottom line that can cover the labor compensations. Otherwise the local government will become a financial loser in privatization, as it has to reallocate its funds from other sources to subsidize the ownership change.
The importance of the constraint is evidenced by our 2001 field study of SOE reform in Sichuan province. The Mianyang No. 1 Textile Company, a subsidiary of the Sichuan Textile Group that is controlled by the provincial government, started its restructuring program in August 2000. The restructuring plan intended to use part of the net asset of the company to compensate the employees. The rate of compensation is left to the firm itself to decide, though central government gives a recommended standard. However, it transpired that the value of the net assets is too small to compensate the amount of compensation claimed by the workers, so the proposed plan was eventually rejected by the general employee meeting.
Nationwide evidence can be found in the information released from the 2004 enterprise survey mentioned in Section Two. When firm managers are asked to point out the major constraints to be overcome during ownership transformation, Table 5 shows that the lack of funds to cover restructuring costs is the firms’ foremost concern. Since the labor resettlement cost constitutes the bulk of the whole restructuring bill, the questionnaire result lends support to our argument. More direct evidence is also available in the survey: a sub-sample of firms answered the question on whether they offer financial compensations to workers during ownership restructuring. Amid the 242 fully privatized firms, 70% of them answer ‘Yes’ to the question. Interestingly, when it comes to the 597 partially privatized firms25 under survey, only 14% give a positive answer. Such a contrast is a precise illustration of the role of workers compensation constraint in completing the Chinese-style privatization.
(c) Bank debt-servicing constraint
The intensive ongoing banking reform elaborated upon in Section Three dictates a much tougher stance of the local banks during ownership restructuring. Given the generally high leverage ratios in the local SOE sector, the banks are nowadays quite vigilant regarding the firms’ restructuring moves because they have been given a strict target on controlling their own non-performing loans. And the banks’ concerns are quite legitimate if we realize that debt evasion can be an additional motive the local governments have in mind when making the privatization decision (Garnaut et al., 2005, p.35). Accordingly, banks need to be convinced by a restructuring proposal that the privatized firm will have the sufficient capacity to service its outstanding debts. At the very least, the cash flows available to the firm in the next period, consisting of expected profits and net privatization proceeds, should be large enough to cover the interest payment arising from the outstanding loans. Without satisfaction of this inequality, the banks tend to veto any proposed ownership change.
Regarding empirical evidence, Table 5 has actually indicated significant presence of the bank constraint, since debt payment problem, from the perspective of firm managers, was the second largest obstacle to successful restructuring. After all, if banks were not tough in monitoring the debt situation during ownership transfer, they would not consider it a major problem. Anecdotal evidence tends to support the presence of bank influence as well: prior to privatization, Shandong Sanwei Silks Co. Ltd owed the local branch of Industrial and Commercial Bank around RMB 20 million. Meanwhile it was the guarantor for its parent SOE, Shandong Silk Group, which secured RMB 20 million of loans from the Agricultural Bank. The privatization of Sanwei Silks was complicated by the controversies related to the settling of debts between the two firms and the two banks. Although the problem was resolved in the end with the coordination of local government, the company director admitted afterwards that either of the banks could have blocked the restructuring if the arrangement of outstanding debt obligations were not to their satisfaction.26 It is worth noting that Garnuat et al. (2005, pp. 79-81) provide a similar account of the bank constraint on ownership restructuring.
Putting all the above constraints together, we have the following proposition:
Proposition 3: The success of privatization in China depends largely on the meeting of the managerial cooperation constraint, the workers compensation constraint (sale price more than the compensation required to facilitate worker resettlement), and the bank debt-servicing constraint (expected cash flows greater than the interest payment on outstanding debts). This constraints-based perspective reveals that each stakeholder group of the local SOEs acts to defend their own interests in the restructuring process that usually involves considerable negotiation and compromise. The chance of overcoming the three critical constraints in turn depends on the motivation of local officials to engage in ownership reform, and their ability to broker the negotiations.
FUTHER STATISTICAL EVIDENCE
Along with the development of the paper’s arguments, we have provided a large amount of corroborating evidence through our case studies and review of existing studies. In this section, we present further statistical evidence on the privatization motives and constraints that can be duly quantified. Specifically, our theory of Chinese-style privatization posits that ownership change is mainly driven by the expectation to generate greater tax revenues for the local economy through the boost of firm sales, but the desire of local governments is subject to both the workers compensation constraint and the bank debt-serving constraint, providing the support of the incumbent management. So the propositions can be refined to a statistical hypothesis that the probability of privatization is positively correlated to the expected sales growth but negatively associated with the amount of required labor compensation and debt burdens.
To test the hypothesis, we use a sub-set of the aforementioned 2004 Enterprise Survey data to examine the determinants of privatization likelihood in China via Probit model estimations. These are a sub-set of ‘privatizing’ firms that have begun ownership restructuring through the introduction of private/employee stakes. However within the sample some firms still have government agencies as the controlling shareholders, while government relinquishes its control in the others. This unique feature of the dataset offers us a rare opportunity to explore the underlying factors affecting whether a firm can be fully privatized in a given period.27 The dependent variable, defined as ownership switch, has a code 1 if a firm’s largest shareholder has been changed from government agencies to private investors and a code 0 otherwise. With respect to the explanatory variables, lagged sales revenues in logarithm form are chosen to act as the proxy of growth prospects in the next period perceived by state controlling shareholders. Lagged profit margin is also included in the regression to see if it has a definitive impact on the privatization decision of the government given the increased market competition. On the constraints side, we obtained a smaller sample of firms whose managers provided their own estimates of labor compensation costs if full privatization is to proceed. Since we also have the employment data for each firm, we think the lagged compensation cost per worker is a good measure of the workers compensation constraint. The bank debt-servicing constraint is easily measured by the lagged debt-asset ratio for each firm, since higher debt leverage will cause more concern for the banks about the risk of loan default post privatization. Finally, industry and year dummies are employed to control for the potential sector-specific and macroeconomic policy effects. The regression results are shown in Table 6.
Clearly, the results lend substantial support to our hypothesis: a firm’s expected sales growth, equivalent to greater fiscal contribution to the local economy, significantly increases the likelihood of its subsequent privatization, and the finding is robust across the four samples. The effect of profitability on subsequent privatization decision is less conclusive. This is consistent with our previous argument that with the competition intensified, firm-level profits become much less important than sales in affecting government privatization decision, though lower profitability in general reduces the government’s incentive to maintain its corporate control. Such a tendency is clearly reflected in sub-sample 2 and 3.
The test of the workers compensation constraint is confined to the last three samples due to data availability. It is found that the constraint is quite salient in firms affiliated with provincial or central governments, while it is not the case for those with municipal governments. We believe the difference lies in the distinct sizes of employment in the two types of SOEs: in our sample, the manning-level per firm in the latter group was around 1,000 – a sharp contrast to the level of around 6,000 in the former one. Thus the impact of the former group on social stability can be very significant, which makes both the management and government more cautious in relation to their ownership transformation. Besides, the bank debt-servicing constraint is strongly significant especially in the whole firm sample, suggesting that the probability of control switch is adversely affected by historical debt burdens.
After offering a bird’s-eye view of the latest ownership transformation in the Chinese industrial sector, the paper explores the broader and deeper institutional dynamics that have shaped the local bureaucrats’ privatization mentality. A wide array of institutional factors, including the decentralized state asset system, banking and fiscal reform, market competition, and the cadre evaluation mechanism, have been organically synthesized to offer a coherent analytical framework. In particular, the institutional analysis goes a long way towards constructing the explicit payoff function of local leaders, which is critical to identifying the necessary conditions under which privatization will be triggered.
The paper presents a well-grounded theory that identifies the necessary conditions in motivating local leaders to privatize their SOEs and the key constraints imposed by other economic agents for a specific restructuring plan to succeed. Incentives for sale have been created by the increasingly hardened budget constraints faced by local officials, intensified market competition, the likely improved post-privatization performance, and the magnitude of the private benefits from which local leaders can expect to derive after privatization. With regard to constraints, any successful privatization program needs to simultaneously satisfy the management participation constraint, the workers compensation constraint, and the bank debt-servicing constraint. And all our propositions on the motives and constraints are supported by a combination of extant literature, case studies and statistical evidence.
Overall, we argue in the paper that the high-powered local bureaucrats’ incentives to trade ownership for growth while maintaining their private benefits, and the stringent constraints laid out above, have characterized the privatization in China since the second half of the 1990s. The motives-cum-constraints political economy approach offers an important explanation for the two stylized facts during the ongoing privatization in transition China summarized in Section Two: incremental but steady in progress and partial in scale.
Finally, it is worth mentioning that, in this paper, generally, we shy away from normative comments on the privatization process, though we have documented the value-destroying collusion between local bureaucrats and private enterprises. Future research would naturally reside in the assessment of the industrial sector’s post-privatization performance, which is being profoundly influenced by the particular path of institutional evolution.
Among the most widely used was the ‘contract responsibility system’ as a special form of performance contracts between government and firm management. For a detailed analysis of the reasons why this institutional arrangement failed to work well, see Shirley & Xu (2001).
In the survey of China’s business sector, OECD (2005) reports that ‘Eighty-seven per cent of the decline in the number of state held industrial firms in the period 1998 to 2003 came from the prefecture and county levels’ (p.97). It also quotes information provided by the State Assets Supervision and Administration Commission (SASAC) that ‘85% of small and medium local government state enterprises, across all industries, had been restructured by 2003’ (p.120).
Consider the slogan of ‘popular capitalism’ used by politicians in the U.K. in the 1980s and those in the East European countries in the early 1990s to build key electoral support.
The argument advocated here for the study of the urban SOE sector correlates strongly to Oi (1999) and Whiting (2001), who have adopted a similar approach in examining the institutional evolution of China’s rural industry.
Although the nature of the survey data left us unable to identify the ultimate ownership of around 21% of the national industrial output, it is to our knowledge the first attempt to trace the exact ownership composition of the Chinese industrial sector. The technical details of the estimation are not presented here due to space limitations but available upon request.
OECD (2005, p.82) reports that two thirds of the increase in the private share can be attributed to the exit of government ownership, while the remaining one third reflects the decline in the number and output of collectives. It needs to be mentioned that a large part of the collective sector, i.e. the township and village enterprises (TVEs) previously controlled by grass-roots level governments, have been under extensive privatization since the late 1990s (Oi, 1999; Whiting, 2001; Li and Rozelle, 2003; Chen, 2005).
The survey project was sponsored by the World Bank and implemented by the Institute of Enterprise Research at the Development Research Center of State Council, China from August to December 2004. It generated a sample of 2,696 firms across sixteen provinces and province-level cities with regard to information on ownership transformation (Liu & Liu, 2005).
Almost all government-led surveys have potential bias towards firms that have a close relation with government agencies and thus have a higher response rate.
A firm is called restructured if it has undergone full or partial privatization. The former means that the largest shareholder has become a private one; the latter suggests that while the government still remains the largest shareholder, significant private stakes have been introduced in the firm.
They include the efficiency hypothesis, the market liberalization hypothesis, the soft-budget hypothesis, the financial liability hypothesis, and the constraint hypothesis.
Nomenklatura, originating from the Soviet model, refers to control by the apparatus of the communist party over leadership positions of all types in the society.
Recently a questionnaire conducted with around 240 local officials shows the following interesting result: when being asked which factors were most important to determine their prospect of promotion, they ranked local economic performance the second most important one, only next to guanxi, i.e. connections or networks (The 21st Century Economic Herald, September 13, 2004).
That is why Blanchard and Shleifer (2001), when comparing the transition experiences between Russia and China, argue that fiscal federalism itself does not necessarily imply developmental local governments; it is the varying degrees of political centralization that make a difference between China’s rise and Russia’s fall.
Huang (1996, p.43, 81) and Lardy (1998, p.90-91) document the prevailing political influence of local governments at provincial level.
For evidence on the loss of influence on the part of township governments to channel bank loans to their TVEs, see Park & Shen (2002).
Guo and Yao (2005, p.217) mention an extreme case of the ‘life-time responsibility’ rule, which ‘may lead to the dismissal of a loan officer if a single loan passed by him is ever defaulted’.
There is a vast amount of literature on central-local fiscal relations. See Zhang (1999) for an overview of the topic.
The shared taxes would then be split by the standardized ratio between central and sub-national governments after collection.
Colorful anecdotes about local protectionism and market segmentation abounded in the reform period especially before the mid 1990s; and Young (2000) finds converging output structure and diverging price levels across Chinese provinces in most of the reform period, which lead him to conclude that the economic reform has ‘resulted in a fragmented internal market with fiefdoms controlled by local officials’. Nevertheless, using provincial-level input-output tables in 1987 and 1992, Naughton (2003) identifies significant growth of inter-provincial trade, the speed of which is even higher than that of provincial GDP growth and foreign trade during the same time span. His result implies that national economic integration was increasing even in the late 1980s and early 1990s.
Trading regulatory and other goods for private gains from private firms may even be preferred by local officials to doing the same with state firms, since the formers’ finances are not so closely monitored.
It is admitted that the case sketched here is a rather extreme one and we do not deny the crucial helping hand offered by local governments in many private businesses’ takeoff (e.g. Unger & Chan, 1999; Tsai, 2002). It seems safe to argue that the cozy government-enterprise relation is indeed a two-edged sword for business growth in the transition context.
As a matter of fact, Insider control is a typical phenomenon in transition economies; for a general discussion, see Aoki (1995).
The case is based on a field study conducted by one of the authors in May 2003.
For an insightful study of labor unrest in contemporary China and its relation to local governments, see Cai (2002).
Recall that ‘partially privatized’ means the government agencies are still the largest shareholders in spite of the significant introduction of private stakes.
The case is based on the interviews of the company director by one of the authors in July 2003.
We obtained the permission to use a dataset from the Institute of Enterprise Research, State Council Development Research Center, China. For detailed information about the survey data see Liu & Liu (2005) or ask the corresponding author.
Aoki, M. (1995). Controlling Insider Control: Issues of Corporate Governance in Transition Economies, in M. Aoki & H.-K. Kim (Eds.), Corporate Governance in Transition Economies: Insider Control and the Role of Banks, Washington D.C.: The World Bank.
Bai, C.-E., Du, Y., Tao, Z. & Sarah Y. T. (2004). Local Protectionism and Regional Specialization: Evidence from China’s Industries, Journal of International Economics, 63 (2), 397-417.
Biais, B. & Perotti, E. (2002). Machiavellian Privatization, American Economic Review, 92 (1), 240-258.
Bienen, H. and Waterbury, J. (1989). The Political Economy of Privatization in Developing Countries, World Development, 17(5), 617-632.
Blanchard, O. & Shleifer, A. (2001). Federalism With and Without Political Centralization: China Versus Russia, IMF Staff Papers, 48, 171-179.
Boycko, M., Shleifer A., & Vishny, R. W. (1996). ‘A Theory of Privatisation’, Economic Journal, 106 (435), 309-319.
Brandt, L., Li, H., & Roberts, J. (2005). Banks and Enterprise Privatization in China, Journal of Law, Economics, & Organization, 21 (2), 524-546.
Burns, J. P. (1994). Strengthening Central CCP Control of Leadership Selection: The 1990 Nomenklatura, China Quarterly, 138, 458-491.
Cai, Y. (2002). The Resistance of Chinese Laid-off Workers in the Reform Period, China Quarterly, 170, 327-344.
Li, H. & Rozelle, S. (2003). Privatizing Rural China: Insider Privatization, Innovative Contracts and the Performance of Township Enterprises, China Quarterly, 176, 981-1005.
Li, H. & Zhou, L. (2005). Political Turnover and Economic Performance: The Incentive Role of Personnel Control in China, Journal of Public Economics, 89(9-10), 1743-1762.
Li, S., Li, S., & Zhang, W. (2000). The Road to Capitalism: Competition and Institutional Change in China, Journal of Comparative Economics,28 (2), 269-292.
Liu, G. S. & Garino, G. (2001). Privatisation or Competition? A Lesson Learnt from the Chinese Enterprise Reform, Economics of Planning, 34 (1-2), 37-51.
Liu, G. S. & Liu, X. (2005). The 2004 Enterprise Survey Project of Ownership Transformation and Enterprise Reorganization in China, Unpublished Report, the Institute of Enterprise Research, the State Council Development Research Center, China, and the World Bank.
Liu, G. S. & Sun, P. (2005a). The Class of Shareholdings and Its Impacts on Corporate Performance: A Case of State Shareholding Composition in Chinese Public Corporations, Corporate Governance: An International Review, 13 (1), 46-59.
Liu, G.. S. & Sun, P. (2005b). Ownership and Control of Chinese Public Corporations: A State-dominated Corporate Governance System, in K. Keasey, S. Thompson, & M. Wright (Eds.), Corporate Governance Accountability, Enterprise and International Comparisons, Chichester: John Wiley and Sons.
Naughton, B. (2003). How Much Can Regional Integration Do to Unify China’s Markets?, in N, Hope, D. Yang, & M.-Y. Li (Eds.), How Far Across the River? Chinese Policy Reform at the Millennium, Stanford: Stanford University Press.
Organisation for Economic Cooperation and Development (OECD) (2005). OECD Economic Surveys: China, Paris: OECD Publishing.
Oi, J. C. (1999). Rural China Takes Off: Institutional Foundations of Economic Reform, Berkeley and Los Angeles: University of California Press.
Park, A. & Shen, M. (2002). Joint Liability Lending and the Rise and Fall of China’s Township and Village Enterprises, Journal of Development Economics, 71 (2), 497-531.
Qian, Y. & Roland, G. (1998). Federalism and the Soft Budget Constraint, American Economic Review, 88 (5), 1143-1162.
Qian, Y. & Xu, C. (1993). Why China’s Economic Reforms Differ: The M-form Hierarchy and Entry/Expansion of the Non-state Sector, Economics of Transition, 1 (2), 135-170
Sargeson, S. & Zhang, J. (1999). Reassessing the Role of the Local State: A Case Study of Local Government Intervention in Property Rights Reform in a Hangzhou District, China Journal, 42, 77-99.
Shirley, M. & Xu, C. (2001). Empirical Effects of Performance Contracts: Evidence from China, Journal of Law, Economics, & Organization, 17 (1), 168-200.
Tenev, S. & Zhang, C. (2002). Corporate Governance and Enterprise Reform in China: Building the Institutions of Modern Markets, Washington D.C.: The World Bank and the International Finance Corporation (IFC).
The 21st Century Economic Herald. “5+1” litui dangzheng ganbu renshi zhidu gaige (“5+1” promote the reform of the Party and administrative cadre management system), September 13, 2004
Tsai, K. S. (2002). Back-Alley Banking: Private Entrepreneurs in China, Ithaca & London: Cornell University Press.
Unger, J. & Chan, A. (1999). Inheritors of the Boom: Private Enterprise and the Role of Local Government in a Rural China Township, China Journal, 42, 45-74.
Wank, D. L. (1995). Bureaucratic Patronage and Private Business: Changing Networks of Power in Urban China, in A. G. Walder (Ed.), The Waning of the Communist State:Economic Origins of Political Decline in China and Hungary, Berkeley: University of California Press.
Wank, D. L. (1999). Producing Property Rights: Strategies, Networks, and Efficiency in Urban China’s Nonstate Firms, in J. C. Oi & A. G. Walder (Eds.), Property Rights and Economic Reform in China, Stanford: Stanford University Press.
Whiting, S. H. (2001). Power and Wealth in Rural China: The Political Economy of Institutional Change, Cambridge: Cambridge University Press.
World Bank (1995). Bureaucrats in Business: The Economics and Politics of Government Ownership, New York: Oxford University Press.
Yarrow, G. (1999). A Theory of Privatization, or Why Bureaucrats Are Still in Business, World Development, 27 (1), 157-168.
Young, A. (2000). The Razor’s Edge: Distortions and Incremental Reform in the People’s Republic of China, Quarterly Journal of Economics, 114 (4), 1091-1135.
Zhang, L.-Y. (1999). Chinese Central-provincial Fiscal Relationships, Budgetary Decline and the Impact of the 1994 Fiscal Reform: An Evaluation, China Quarterly, 157, 113-141.