“In Accordance with Local Conditions”: Policy Design and Implementation of Agrarian Change Policies in Rural China

Abstract

An important part of Beijing's strategy to reduce the welfare gap between urban and rural parts of China has been the promotion of urbanisation. Replacing peasant agriculture with commercial operations of scale is an integral part of this endeavour. This article analyses the implementation of policies meant to transform the structure of Chinese agriculture. It argues that the central government is using a set of very flexible policies, project-based implementation and adaption to local conditions to guide and support an existing dynamic of structural transformation in agriculture. Local governments, in turn, appreciate the flexibility, the political predictability, the potential revenue improvements and the cognitive framework inherent in these programmes. The article is primarily based on interviews with leading cadres at the township and county levels in the provinces of Shandong, Sichuan and Guizhou between 2008 and 2010.

Introduction

In the face of massive Chinese policy projects to transform entire sectors of the economy and to change the lives of large sections of society, many observers have been wondering about the capacity of the Chinese state to implement these plans. The most recent case in point is the Chinese government's current urbanisation drive. This article looks at the implementation of an important subsection of these policies: those that promote agrarian change – in other words, the transition from smallholder farming to commercial agriculture. Drawing on insights from policy choice theory and the work of Michael Howlett (2004) in particular, this article aims to contribute to the ongoing debate on local policy implementation from a new angle. While previous studies have focused on the tools available to the state to push forward its agenda, this article will discuss the related but less well-covered question of how the design of a policy in terms of its resource needs, political efficiency and legitimacy may also influence its adoption. The empirical case presented here is a set of policies for agrarian change to be implemented by county and township cadres in the provinces of Sichuan, Shandong and Guizhou.

Since the end of the 1970s, the Chinese countryside has witnessed a continuous stream of reforms that address almost all aspects of rural life. At various stages of these reforms, scholars addressed the capability of the Chinese state to implement these policies (O'Brien and Li 1999; Cai 2004; Li 2006; Kennedy 2009; Göbel 2011; Mei and Pearson 2014). Though they share an interest in the factors that help to increase the prospects for successful implementation, they have arrived at different and sometimes conflicting conclusions.

Kevin O'Brien and Li Lianjiang noticed a “pattern of selective policy implementation” (O'Brien and Li 1999: 167) in rural China in which policies to control and tax villagers were strictly enforced while those meant to protect their interests were more likely to be ignored. They argue that selective policy implementation was a function of the hierarchical structure of the political system, with its dominance of immediate superiors, its obsession with hard and quantifiable targets and the relative autonomy of the local state vis-à-vis the demands of the population. Others have expanded upon this line of thinking. Graeme Smith, for example, suggests that leading local actors carry out policies primarily according to a cost–benefit analysis. If policies appear prominently in the cadre evaluation system, increase the revenue sources of local governments, benefit “individual cadres and their partners in the ‘shadow state’” (Smith 2013: 1028–1029) and can be carried out with existing resources, they are more likely to be implemented (see also Smith 2009: 30). At the core of Smith's argument as well as that of O'Brien and Li is a hierarchical perspective in which any variation from original policy guidelines indicates an implementation gap. For Linda Chelan Li (2006), this is a zero-sum perspective on policy implementation. Her alternative is a “non-dualistic” perspective of implementation in which controversial policy initiatives, such as the Tax-for-Fee reform she uses as an example, become “a central–local ‘joint venture’” (Li 2006: 172). Local opposition to and criticism of central initiatives and the resulting bargaining process allow for the adaptation of these measures to local conditions and do not necessarily indicate an implementation gap.

Christian Göbel (2011) suggests that, while both of these approaches have their merits, they are unable to explain local variation of the same policy in different localities. In his empirical example, he also looked at the Tax-for-Fee reform, some locations became eager “pioneers” of the reform while others turned into staunch “resisters.” To explain this variation, he argued for the recognition of market-like mechanisms such as competition as a third important policy-steering mechanism next to hierarchy and networks (based on Hood 1998). According to Göbel, policy implementation in the case of the Tax-for-Fee reform encompasses all three elements. In particular, however, the addition of the element of competition was an important factor leading to the variety in results. The competition for funding and rewards turned out to be a further motivation for some well-equipped counties, who thus became “pioneers” of this reform. Counties with a large agricultural base which may have also been experiencing financial difficulties since even before the reform were unable to compete with these pioneers. They had to come up with creative solutions to deal with the new obligations introduced by the Tax-for-Fee policy – not all of them legal and some of them even in stark contrast to the very spirit of the reform. In short, all of these studies focused on the different approaches (policy-steering mechanisms) used by central and local government bodies to push forward potentially contentious policies.

Policy-choice theorists look at policy implementation through a different analytical lens. They argue that “different instruments involve varying degrees of effectiveness, efficiency, equity, legitimacy and partisan support, and changes in a particular situation [affect] their appropriateness” (Howlett 2004: 6). The appropriateness of a policy instrument would, in turn, have a decisive impact during the implementation stage. Howlett specifically focuses on three aspects of the design of a policy: its legitimacy, resource needs and perceived efficiency.

Legitimacy refers to the acceptance of a policy by the public and among “those directly involved in policy-making in the issue area or subsystem involved” (Howlett 2004: 6). The inclusion of legitimacy is based on the realisation that “many dynamics in the organisational environment stem not from technological or material imperatives, but rather from cultural norms, symbols, beliefs and rituals” (Such-man 1995: 571). For Suchman,

legitimacy is a generalised perception or assumption that the actions of an entity are desirable, proper or appropriate within the socially constructed system of norms, values, beliefs and definitions. (Suchman 1995: 574)

He further differentiates three aspects of organisational legitimacy relevant for this article: pragmatic, moral and cognitive legitimacy. Pragmatic legitimacy “rests on self-interested calculations of an organisation's most immediate audiences” (Suchman 1995: 578). In other words, an organisation and its actions are legitimate because the intended outcome aligns with the interest of its audience. Moral legitimacy is based on the judgement as to whether the given activity is “the right thing to do” (Suchman 1995: 579). Cognitive legitimacy may come into existence if an organisation and its activities provide “cultural models that furnish plausible explanations for the organisation and its endeavours” (Suchman 1995: 582).

Taking into account resource needs acknowledges that there are huge differences among policies in relation to target size and intended distributional effects. For example, policies that entail a massive transfer from one group to another or that target a large audience are more costly than those that only encourage a change in behaviour among a small group of people. Finally, policies differ not only in their resource needs but also in terms of their perceived likelihood to produce results. The more administrations base their system of promotions and demotions on performance in the “target world” (Hood 2006), the more the perceived ability of a chosen measure to achieve its intended political target becomes an important point of consideration for those charged with its execution. Howlett refers to this as the efficiency of a policy.

The following analysis will first introduce policies for agrarian change in more detail. The fate of these policies will to a large degree be decided by the acceptance or resistance of the grassroots level of the administration during the implementation stage. This level of government is subject to a set of very specific contextual pressures, which may greatly influence their decision-making process – this will be discussed next. The final section of the analysis then dissects the fit of the current agrarian-change policies in terms of legitimacy, resource needs and political efficiency.

The article is based on three rounds of fieldwork carried out mainly in the provinces of Shandong, Sichuan and Guizhou. The fieldwork was part of two joint projects by the China Center for Global Governance and Development in Beijing, Tübingen University and Duisburg-Essen University, funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) and the Bundesministerium für Bildung und Forschung (BMBF, German Federal Ministry of Education and Research).

In Shandong research was conducted in September 2008 and 2009 in the administrative region of Laixi, a county-level city under the administration of the vice-provincial city Qingdao. With a per capita income of CNY 7,159 in 2007, Laixi was the richest of all field sites (Anonymous 1 2008). We conducted semi-structured interviews in Laixi with government and party officials responsible for the implementation of agricultural policies. We also conducted interviews with township and village officials in two towns (, zhen), Xiagezhuang Town and Jiangshan Town. In total, more than 40 interviews took place. In Sichuan, fieldwork was carried out in September of 2008 and 2009 in the administrative region of Anju District, a county-level district under the prefectural city Suining. Anju, one of the latest cases of a county-turned-district, is located halfway between Chengdu and Chongqing. Per capita income in 2007 was CNY 3,415 (Suiningshi Tongjiju 2008: 7). Anju District has 21 towns and townships, in four of which – Yufeng Town, Anju Town, Changli Township and Majia Township – we conducted approximately 40 interviews with township and village officials. In Guizhou, research was conducted in September of 2010 in the counties of Xifeng and Meitan. The per capita annual income in 2009 for peasants in the latter two counties was CNY 4,343 and CNY 5,594, respectively (see Guiyangshi tongjiju 2010: 419). Here, about 20 interviews with county, township and village officials took place. In addition, we conducted several expert interviews with scholars in Chengdu, Qingdao, Jinan, Guiyang and Beijing.

Agrarian Change in China

One of the obstacles in addressing agrarian-change policies is that they are not part of a unified framework but rather consist of numerous individual measures. However, they all tend to promote a shared understanding of the purpose of agriculture and have certain similarities in their design. It is argued here that these similarities are enough to allow their being lumped into one category. The remaining part of this section will introduce in more detail the rationale of these policies and their reception at the local level at the field sites.

Beijing has grown increasingly dissatisfied with the state of Chinese villages over the years, especially in the less developed central and western parts of the country. While urban China experienced steadily growing incomes and improving welfare, rural China did not benefit to the same degree from the Chinese economic boom. The central government has identified the current structure of agriculture as one of the main reasons for these differences in development. In relevant policy documents, the suggested solution to this problem turned out to be the promotion of a highly efficient agriculture (, gaoxiao nongye). This vision of agriculture entails commercial enterprises, scientific/modern production and a separation of land, labour and capital. Smallholder farming, in contrast, is portrayed as being unable to provide the necessary economic growth and profitability to fund the intended massive improvement of rural living conditions. Recently, local governments in many locations have begun to enthusiastically welcome this idea and have even initiated numerous projects on their own.

The current drive to promote agrarian change is only the most recent variation of a much older theme. At least two previous waves of national interest in the promotion of a commercial restructuring of agriculture can be identified. Document No. 11 from 1993, famous for extending the land-usage rights of villagers to 30 years, was also one of the first legal signs that the centre envisioned a change in the structure of agriculture (Zhongfa 1993). This document already justified changes to the Household Responsibility System by pointing to shortcomings in productivity and low levels of investment in agriculture. In other words, it legitimised a move away from the system of egalitarian redistribution of land with concerns over the efficiency of land usage and actively advocated the rearrangement of land into bigger holdings in order to create economies of scale (for a good overview of this change, see Unger 2002).

Another major step toward a structural reform in agriculture was the “Opinion Regarding the Work to Complete in Agriculture and the Countryside in the Year 2001” (Zhonggong Zhongyang and Guowuyuan 2001). This document firmly established the role of the efficiency paradigm as a cornerstone of Chinese agricultural policies. It argues that structural inefficiencies were the major reason for low income levels among the rural population and admits that instead of catching up with urban areas, the rural regions were falling further behind. The 2001 Opinion suggests that the only way out of this mess would be to conduct a thorough structural reform of agriculture that would emphasise the quality and efficiency of agriculture. Its key mantra was that if 900 million peasants did not change their ways of earning a living, peasant prosperity would never take off, and the modernisation of the countryside would therefore be difficult to realise.

The latest spike in the interest to transform agriculture came with the 2008 “Decisions of Central Committee of the CCP Regarding Several Big Issues in Pushing Forward the Reform and Development of the Countryside” (Zhongfa 2008). The 2008 Decisions state that in the future agriculture should prioritise high yields, high quality, high levels of efficiency in the production process, the protection of environmental resources, more security for the producers and the use of scientific methods and modern equipment. In this vision, almost no aspect of agriculture remained untouched. Structural reform was identified as the key to raising rural incomes and narrowing the rural– urban income gap. Since 2006, similar arguments have consistently appeared in each of the annual joint statements of the State Council and the Standing Committee of the CCP known as the “No. 1 documents,” which target rural issues. While the programme to “Build a New Socialist Countryside” – introduced with the 2006 No. 1 Document – addressed many issues beyond agriculture, agricultural upgrading was among its top priorities. Although at times during the interviews literally all of the activities of grassroots-level governments seemed to belong to this framework, the programme is usually seen as an attempt to establish more advanced and specialised economic structures, to improve infrastructure and, moreover, to better rural welfare (Saich 2007; Schubert and Ahlers 2011, 2012). The recently published “Opinions on the All-Around Deepening of the Rural Reforms [and] Speeding-Up of the Push for a Modernization of Agriculture” continues to emphasise a structural reform of agriculture (Zhonggong Zhongyang and Guowuyuan 2014). Importance, however, is also given to sustainable agriculture, improving food quality and making production increasingly “green.” The effects of this potentially new direction have yet to be seen.

This article argues that, in contrast to previous spikes in Beijing's interest in transforming agriculture, this latest initiative may have caught on at the grassroots level. Leading cadres at the township and county levels at all field sites saw success in the transformation of agriculture as an important step to reach the goal of Building a New Socialist Countryside. One leading cadre from Laixi further explained:

Our biggest success is raising the level of agricultural industrialisation (, nongye chanyehua de shuiping), raising the income of the villages, developing dragonhead enterprises. Our biggest problem is that the funds for “Building a Socialist Countryside” are not enough. Because of these investment needs, the financial power on every level of the administration is not sufficient. (Anonymous 2 2008)

Some of the measures to encourage agrarian change have proven particularly popular among leading township and county cadres. These include dragonhead enterprises, the promotion of all activities in the field of land transfer and strong informal financial and administrative support for new enterprises and potential investors.

Dragonhead enterprises appear in the interviews as one of the most important tools of the (local) state to promote commercial agriculture (Anonymous 3 2008; Anonymous 4 2009). These enterprises, initiated by local and provincial governments, have been instrumental in the creation of an integrated and commercial agriculture in China (for more on dragonhead enterprises and their role in the commercialisation, see Lingohr 2007). Examples of such enterprises at the field sites can be found in the pork industry in Suining, where dragonhead enterprises have established long-term supply contracts with farmers, provided food-processing services and manage the selling of the final product (Anonymous 4 2009). A similar practice was visible in Laixi, where dragonheads help to organise the production of beef, milk, pork, poultry, vegetables, fruits and peanuts (Anonymous 5 2009).

Many local cadres have pointed to the link between the management of agricultural land and the current structure of agriculture. According to their understanding, land needs to be concentrated into bigger holdings in modern agriculture. The collective nature of land in the villages has turned the local state in many locations into a major player in arranging and preparing land-transfer solutions and thereby coordinating between villagers, who have land-usage rights, and new actors in agriculture, who have an interest in agricultural land. At the field sites, we encountered several variations of this process, among them land-transfer cooperatives (, tudi liuzhuan hezuo-she), specialised cooperatives (, zhuanye hezuoshe) and land-transfer service centres (, tudi liuzhuan fuwu zhongxin) (see Trappel 2011).

In addition to these methods of direct support for commercial farming, local governments provide substantial informal support. For instance, they provide information about loans and subsidies to key commercial activities in their localities. Even if local governments themselves have no funds to invest, they might use their social and political capital to assist. One township party secretary in Suining, for example, recounted in great detail all the wining and dining with officials and scholars (from an agricultural science faculty of a university in the provincial capital) that went into the preparation of specialised agrarian production in his township.

Why have these local cadres recently started to embrace agrarian-change policies even though most previous central initiatives in this field have proven to be rather short-lived? This article argues that two factors are responsible for the new enthusiasm towards these policies: the difficult political, economic and moral context for these local administrations and the nature of these policies. The following two sections will explore these aspects in more detail.

Conditions for Local Policy Implementation in the New Socialist Countryside

This section explores the position of township and county administrations in rural China as the main bodies for local policy implementation observed at the field sites. Three different aspects of the position of local governments within the Chinese political environment have been selected: financial conditions in regard to the resources available for implementation, the link between political efficiency and careers of cadres and, finally, organisational legitimacy.

The Financial Conditions of Local Governments

The financial difficulties of lower levels of the Chinese rural administration have been widely discussed (Wong 2009; Zhao 2007; Zhou 2010). The proliferation of new government initiatives as part of the New Socialist Countryside (Schubert and Ahlers 2012) and a substantial rise of transfer funding (documented by Lin and Wong 2012) have triggered a debate as to whether there have been substantial changes to the fundamental structural problems discussed in these earlier works. Do township- and county-level administrations still shoulder most costs for public services, social protection, health care and infrastructure investments at the grassroots level – given all the negative financial consequences for them, as described by Christine Wong (2009)? Linda Chelan Li and Wen Wang (2014) have pointed out that the Chinese state has begun to use massive amounts of earmarked funds to close, for example, the gap in education. Indeed, project-based and earmarked funding as two important trends in the countryside may have considerable effects on the structure of local funding in the long-run. Yet the lack of adequate budgets and revenue sources remained an important issue at our field sites and was especially tangible at the township level.

A leading township cadre in Sichuan, for example, noted that his township government was ordered by higher-level administrations to improve (among many other things) the transportation infrastructure – irrespective of the local financial capabilities to fund the construction of such roads. Given the funding limitations, the official argued that they had to prioritise some policies (read: roads) over others (read: almost everything else) (Anonymous 6 2009). This led to the implementation of the New Socialist Countryside in many poorer locations being limited to a few model villages (Anonymous 7 2009) (this is well documented in Ahlers and Schubert 2013; Ahlers 2014). The provision of public services was another major drain on local funds, and even wealthier localities were struggling with these obligations. In Laixi, the richest of our field sites, a cadre complained:

For the two administrational levels of county and township government, the wages for teachers are a great financial burden. We have some townships here that had to use all of their financial income this year to just pay the teachers’ wages. (Anonymous 3 2008)

A township cadre from Anju mentioned that, compared to all other costs, the wages of the cadres and work expenses still account for the largest part of expenses. This is the case despite a strict supervision of all expenditures and very low wages (Anonymous 8 2009).

The need to transfer taxes to the centre was another frequent complaint among county cadres. A cadre from Laixi pointed out that they had collected CNY 2 billion in taxes in 2008. In his understanding, this sum would be enough to solve, for example, all local funding problems in education. However, most of these collected taxes, which represent a 30 per cent increase from the previous year, had to be transferred to the centre. They would end up with only CNY 500 to 600 million in transfer payments from the centre (Anonymous 3 2008).

The financial obligations compare very unfavourably to the available income of most local administrations. One example from the field sites may clarify the great difficulties local governments find themselves in. In one of the townships visited, the monthly available operational funds stood at CNY 59,000, of which 54,000 went directly into minimum social-insurance allowances (, wubaohu,, dibao) (Anonymous 6 2009). The remaining operational costs hovered at CNY 45,000 per month. In sum, this township alone had an average operational loss of CNY 40,000 per month (Anonymous 6 2009). In many locations, township governments did not even have an independent account of their own. All of their financial affairs were managed at the county level (Anonymous 9 2008). In one Anju township, cadres complained that they did not have any independent income at all and were completely dependent on transfer payments from higher levels of the administration (Anonymous 10 2009).

Leading local cadres argued that, as a consequence of the mismatch between numerous policy obligations and insufficient funding, they have incurred enormous debts. A cadre from Anju District estimated the collective debt of all 21 towns and townships to be approximately CNY 200 million (Anonymous 11 2008). The lending activities and growing debt of many township governments have also recently been discussed by Liu Yia-Ling (2012), who shows how normal deficit spending has become for township cadres. Zhou Xueguang notes the great influence that several national infrastructural policies had on the development of the “collective debt” in the countryside (Zhou 2012).

At the field sites, another important recent trend in the finances of rural administrations, in part related to the New Socialist Countryside, was also clearly visible: the rise of project-based funding. Zhe Xiaoye and Chen Yingying provide one of the best introductions to this development (Zhe and Chen 2011). They describe project-based policy implementation as a process in three stages. First, higher levels of the administration prepare a set of projects local administrations can apply and compete for (, fabao). Second, county- and township-level governments combine these different projects into integrated packages (, dabao). These packages basically represent a coordination of several independent projects and funding sources to achieve a shared objective. Third, villages can compete for these packages, whose contents might change again by the time the projects reach the village (, zhuabao).

Almost all of the field sites reported on how they combine different programmes in order to achieve their targets. At one location in Anju, poverty-alleviation funds, infrastructural funds and funds to improve living conditions in regions inhabited by ethnic minorities were all used for the construction of the same road in the end (Anonymous 9 2008). The tight financial situation of many local governments also puts them in direct competition with each other. In Laixi, officials mentioned that the competition with other county-level administrations over transfer payments from Qingdao is part of their normal business. Without participating and succeeding in this competition, they would have considerably fewer funds (Anonymous 1 2008).

In the eyes of local administrators, a particularly frustrating aspect of the current system of transfer payments in rural China was that of matching funds. As a rule, most earmarked or transfer funds have to be matched locally. In other words, higher levels would only match the amount the local administration could invest by itself. These “forced” matching funds (, qiangxing peitao) leave cadres little choice but to withdraw from less important projects in order to get priority projects started (Anonymous 8 2009).

In order to cover the costs for bigger infrastructure projects, townships still also had to ask villagers for contributions (Anonymous 12 2008). While these kinds of contributions are not uncommon in rural China, it is important to highlight how normal it has become for many villages to fund their basic infrastructure in this way. One of the townships in Suining even asked migrant workers who had left for urban employment long ago to fund a portion of local projects (Anonymous 13 2008).

The constant pressure to find new funding sources has promoted the proliferation of questionable practices of generating revenue (Wong 2009: 949). For example, the (often illegal) sale of farmland for construction purposes was said to account for 30 to 70 per cent of all government revenues (Lin 2007: 1832). Although more recent data regarding this topic is hard to come by, the profitability and ease of land sales and the limited dependence of the local state on the opinion of the rural population certainly have contributed to the widespread nature of these practices (Ping 2011). Kennedy also notes that township governments continue to use land-leasing as a means to balance their expenditures and wonders openly about the sustainability of this phenomenon (Kennedy 2013: 1021).

Political Pressure and Career Opportunities

The local state is subject to a very complex system of near-constant supervision and monitoring. Several previous studies have pointed to the importance of this system for the strategic planning of local cadres (Edin 2003; Gao 2010; Heberer and Trappel 2013; Liu, Hou, and Tao 2013; O'Brien and Li 1999; Whiting 2001, 2004; Zhong 2003). While the results of the annual evaluations are not the only factor considered when making decisions on promotions or sanctions, they do carry weight.

At our field sites, local governments were subject to three different evaluation procedures: an evaluation of the execution of political programmes (, mubiao kaohe), a performance audit of leading cadres (, lingdao ganbu kaohe) and the so-called “one-item veto rule” (, yipiao foujue). For the programme evaluation, each year all levels of the administration at the local state had to outline (in communication with their higher-ups) targets in several important policy fields. At the end of the year, their performance was evaluated and graded either excellent (, youxiu), qualified (, hege) or not qualified (, bu hege) (Anonymous 14 2010). While most of this evaluation process was based on an exchange of numerical data among different levels of the administration, additional on-spot investigations were conducted to discourage the transmission of incorrect data. The grading is important, as those units with better evaluation results will receive additional funds the following year. In a similar spirit, the individual performance of leading cadres at the township and county levels was evaluated. According to regulations, only cadres who are given an “excellent” two or more times in a row can be promoted. Finally, all targets that have the status of a one-item veto rule must be achieved, lest the local leadership and all organisations become subject to severe punishment and sanctions. In addition to national targets, such as the one-child policy and provincial ones – for example, in regard to energy consumption (Kostka and Hobbs 2012), many counties had a set of veto targets in relation to local priorities (see Heberer and Trappel 2013). In Guizhou, for example, a county cadre listed the following six veto targets:

• ▄

  the one-child policy,

• ▄

  “Clean and Honest Government” (, dangfeng lianzheng),

• ▄

  [fighting] drugs and substance abuse,

• ▄

  public order and responsive government,

• ▄

  a secure work environment, and

• ▄

  social stability (Anonymous 14 2010).

While drawing a direct connection from the priorities in the evaluations to policy change would omit a lot of other important factors, the evaluations indicate what is prioritised within the political system. They create considerable pressure on the local leadership and most importantly on the county party secretary to meet the goals set by higher levels of the administration. At the same time, they provide these cadres with a framework to get ahead in their careers. Heberer and Trappel (2013) argue that, when facing evaluations, cadres have three options: they can decide to stick to policy solutions they already know to work to some degree; they can opt for new and innovative solutions, which may carry more potential risks but also could be more rewarding; or, they can collude in an attempt to deceive the evaluators.

Legitimacy and the Local State

In addition to financial and political pressure, local administrations and their leadership face considerable moral pressure, which comes from three main sources: First, the centre propagates a carefully designed narrative that links most political corruption to the local state. Especially within Chinese academic literature, we find many accounts that link the local level to corrupt behaviour. Zhao Pan (2010), for example, describes the great harm corruption inflicts at the grassroots level from the perspective of the state. Ultimately, he argues, corrupt behaviour will erode all possible moral authority of the local political leadership. In other words, if local governments are perceived as lacking moral legitimacy, their actions will also be seen as lacking this legitimacy. Li Lianjiang (2004) points to the severity of this problem by showing that the rural population has little trust in the local state and its motivation (as opposed to generally high approval rates for the central state). Christian Göbel (2012) notes the existence of a narrative carefully crafted by the centre that paints the peasants as victims of corruption on the part of local administrators who do not implement beneficial policies for the rural population as they should. Even if cadres are not corrupt, many doubt their work ethic. Cai Yongshun (2004), for example, points to cases where local governments have opted to invest in “face-building” projects, which may help leading cadres improve their image and achieve their career goals but have little, if any, lasting positive effects for the rural population.

Second, the reports of clashes between local administrations and the rural population, especially in the field of land sales, are another source of great moral pressure for local governments (in addition to the political pressure arising from their apparent inability to manage social stability, which falls within the targets of the one-item veto rule). Ray Yep (2013: 278–280) has compiled an impressive list of recent clashes over land deals in rural China. He argues that there may not even be an upward trend in land grabs and that these protests may instead be the sign of a new rights awareness of the villagers. Nevertheless, the news about these clashes harms the reputation of all local cadres.

Third, and somewhat neglected in the discussion so far, many members of local governments themselves, irrespective of their bad reputation among portions of the general public, have a positive image of themselves and their role in the countryside. One township cadre outlined his motivation to become a cadre:

To develop one's native place. One [main motivation] is to see that the native place is good, that family and friends are happy and harmonious. That gives [me] a good feeling. (Anonymous 5 2009)

While a cadre can certainly speak in this way and still be corrupt, we should acknowledge that many of them want to be seen as having a positive impact on local welfare.

“In Accordance with Local Conditions”: Implementation of Agrarian-Change Policies

This section explores how the design of agrarian-change policies may have affected the fate of these policies during the implementation stage. The argument presented here is based on three assumptions: First, leading cadres at the township and county levels in China have some (limited) leeway in selecting and prioritising policies (see Heberer and Schubert 2012). Second, irrespective of the nature of a policy, its implementation will require a certain amount of resources. Therefore, the local leadership will have to decide which policies it wants to work on. Third, based on Howlett's framework, these decisions relate to the expected impact of a policy on local resources, its perceived political efficiency and its legitimacy.

Agrarian Change and Resource Needs

The project-based nature of many policies to promote agrarian change ensures that localities can keep costs low by implementing projects only in promising locations and in accordance with their resources. This is usually described as implementation “in accordance with local conditions” (, yin di zhi yi). Unsurprisingly, local administrations primarily select villages which are likely to succeed in competition with other villages.

Most projects have a modest scope and aim to push forward existing trends in agriculture. The goal of land-transfer service centres, for example, is simply to reduce the high transaction costs of existing land-transfer methods. Providing this kind of service tends to be cheaper and less risky than engaging directly in agriculture. A substantial part of agrarian-change policies aims to create a favourable investment climate for external investors.

Not only are the costs of these projects relatively low (at least for local administrators), they may even provide several new sources of revenue, of which one of the potentially most important is new investors. In theory, given the tax exemption for smallholders, this kind of agricultural restructuring should also have a positive effect on taxation revenue. The tax benefit from promoting agricultural enterprises, however, has not been as clear-cut as one might expect.

The situation in Laixi provides a good example of this development. All sectors of agriculture there reported GDP growth in 2011. Cash crops and livestock industries, for example, grew by 4.1 per cent and 6.2 per cent, respectively. Still, this did not translate into higher taxation revenues. While in 2011 the overall financial revenues of Laixi increased by an astonishing 30.1 per cent to hit CNY 2.96 billion, local taxation revenues during that same year decreased by a similarly impressive 23.2 per cent (down to CNY 1 billion) (Qingdaoshi Laixishi Tongjiju 2012). There are several possible explanations for this, the most obvious, however, being a general tax cut and a change in the way taxes were calculated in 2011 (Jingji Cankao Bao 2012).

The anticipation of a higher tax income through a restructuring of the local economy and in particular through the shift toward operations of scale in agriculture may perhaps be a valid long-term strategy, but short-term regular taxation income of local governments did not increase substantially. Still, there are two other important financial benefits attached to this structural change: more transfer payments from higher levels of the administration, and direct financial support from the newly founded enterprises for local governments and the general public.

The local state and especially the township level may apply for considerable project funding in order to push forward agricultural change. Even if most of these funds are earmarked, they are a welcome means to improve the local GDP and the rate of investment. In Anju District, for example, each new specialised land-transfer cooperative was supported by Suining City, the next-highest level of the administration, with a subsidy of CNY 5,000 to 10,000 (Anonymous 15 2009). In the Laixi region, Qingdao handed out CNY 400 per cow purchased in 2007 and increased that sum in 2008. There are also substantial subsidies in place for other types of livestock farming (Anonymous 3 2008). Furthermore, in Laixi there was considerable monetary support for new greenhouses, ranging from CNY 1,000 to 5,000 per completed greenhouse (Anonymous 1 2008). Still, assessing the amount of financial support from higher echelons of the administration remains difficult, as a substantial part of the funding in agriculture in recent years has been hidden within the New Socialist Countryside programme, and the representatives of the local Bureaus of Finance at all field sites were among the most reticent interview partners. One high-ranking official from Suining City explained the sources of funding for the New Socialist Countryside in the following way: 10 per cent came from the centre, the province contributed 40 per cent, the city/prefecture gave another 10 per cent and the county level contributed the final 40 per cent (Anonymous 16 2008). An article by Lin and Wong (2012) shows the rising level of central subsidies for rural development but also remains relatively vague about the contributions from the subnational level.

To understand the true financial importance of a new and restructured agriculture for local governments, we have to look beyond the immediate and institutionalised financial benefits. In general, having more successful local enterprises (to which the new enterprises in agriculture belong) may bring financial benefits in addition to taxes and transfer payments. In a world where local governments and villagers often lack funds to pay for infrastructure and other public services, it is entrepreneurs who make the difference with their donations. In several locations, these companies or individual entrepreneurs had invested considerable amounts in new infrastructural projects (Anonymous 10 2009). Sometimes these contributions to the efforts of the local state are organised in quite rigid ways and entrepreneurs are systematically asked to donate rather large sums. In Laixi, these contributions provided a substantial part of the funding for a local development initiative called the “Five Changes” (, wuhua). In a campaign-style approach, all local entrepreneurs were asked to select one village to “volunteer” to support financially in its efforts to improve its infrastructure and its living conditions. In 2008 the total investment in this programme was CNY 120 million, of which enterprises contributed 30 million and the peasants themselves added another 22 million (Anonymous 5 2009).

Finally, while there is always a risk of failure, as there is in all new investments, the short-term financial risks for local administrations appear to have been relatively small. Project-based and earmarked funds came from higher levels of the administration. Investments were usually made by entrepreneurs and not the administrations themselves. Other urbanisation projects, such as the resettlement of villagers into new urban communities (, nongcun shequhua), seem to carry much more financial and political risk for local administrations. Moreover, through this system of project-based funding, local cadres are designing projects to align with the priorities of higher levels of the administration, and this may have provided those higher levels with a way to harness the image-building of their subordinates, described in Cai Yongshun (2004), for the common good (or, in other words, their own development priorities).

Political Efficiency and Career Opportunities

According to Howlett, implementing bodies prefer policies that are perceived as being more likely to achieve their targets and try to avoid risky policies or those with unclear targets (this is also one of the key arguments made by O'Brien and Li 1999). One could argue that this is the case even more so in political systems such as China's, which increasingly focus on “governing from a distance” by incorporating complex monitoring and evaluation systems (Kipnis 2008).

Every location we visited had a detailed plan of how to further develop commercial agriculture. All of these plans included binding targets to build up capacity for specific cash crops, fruits, vegetables, livestock industries, food-processing and other aspects of agriculture. In Anju, the following quantitative targets were given high priority in the annual evaluations: GDP growth, investments and attraction of investors, average peasant income, government revenue and fixed investments/assets (Anonymous 8 2009). Investments in modern agriculture and livestock industries are among the top priorities of the annual evaluation process, as one township cadre explained (Anonymous 8 2009). Particularly interesting has been the inclusion of land-transfer cooperatives in the evaluation system in Anju (Anonymous 15 2009). Xifeng also had a very elaborate system of monitoring all variants of land transfer (Xifengxian Tudi Liuzhuan Zhongxin 2009).

In Laixi, every step in building commercial agriculture was rewarded in the evaluation. Township cadres, for example, could gain more points for modernising agriculture (55) than for implementing the Five Changes programme (Laixishi Renmin Zhengfu 2009). Specific measures mentioned by the evaluation system in Laixi included building biogas facilities, constructing greenhouses, installing cooperatives, planting orchards and undertaking many other similar tasks (Anonymous 1 2008).

Finally, calculations of political efficiency are also based on previous experience in implementation. While many administrators have considerable experience directly engaging in or guiding the local economy (, zhidao or , yindao), necessary for agrarian-change policies, they have less experience with providing services(, fuwu). However, attracting new investors (and creating a proper investment climate) has been part of their daily business for many years. In fact, Liu, Hou, and Tao (2013) argue that the importance of attracting new investors to agriculture has become a dominant theme in the evaluations of township governments. While in the year 2000 only 17.55 per cent of their surveyed township administrations mentioned agricultural industrialisation (of which attracting investors is a core element) as a top priority of the annual evaluations, this figure stood at 40.55 per cent in 2007 (Liu, Hou, and Tao 2013: 89).

Agrarian Change and Organisational Legitimacy

In rural practice, Suchman's three categories of organisational legitimacy (pragmatic, moral and cognitive) are closely connected to each other. However, for many leading cadres in our interviews, the cognitive aspect of agrarian-change policies turned out to be the most important. The majority of them subscribed to the centre's portrayal of smallholder farming as being unable to raise rural incomes and substantially improve welfare. The centre's drive to modernise the countryside provides these cadres with a specific role to play: that of agents of change.

Virtually all officials interviewed began with a summary of their activities to create a modern countryside and their efforts to overcome the remnants of traditional agriculture (, chuantong nongye). One township mayor in Anju, while proudly presenting a newly built pig husbandry, explained: “In the past, our mode of production was a few dispersed households raising livestock, no economies of scale and not much of a profit.” He continued:

Now, after founding this company, our peripheral masses (, zhoubian de qunzhong) get to know technology. From this they can learn that livestock industries are not about, again and again, raising just one pig and that's it. They can learn to raise pigs and how to make a profit from their production. (Anonymous 17 2009)

Arguments like this may also be an explanation for the omnipresent usage of the term “high efficiency” (, gaoxiao). Efficiency here stands for more than just the commercial reorganisation of agriculture. It is also a way to dismiss alternative models of agrarian development that are not perceived as fitting within this cognitive framework. “Traditional farming” and “traditions” have become terms used to devalue the work and lifestyle of smallholders. In our interviews, only one situation emerged in which local cadres found welcoming words for traditional farming in relation to the economic future of villages, and this was in the context of rural home-stay tourism (, nongjiale) (Anonymous 17 2009; Anonymous 13 2008).

To the frustration of the implementing local administrations, the villagers do not always seem to agree with this vision of an agrarian modernisation. Many local cadres complained that implementing these changes would require substantial “thought work” (, sixiang gongzuo) to convince the villagers (Anonymous 12 2008). A city-level cadre from Laixi explained:

The stance of peasants is just not to conform [to a modern, specialised production]. For example, in grain cultivation, if there were only one product in the village, the mode of production would be the same for everyone; farming would be easier, too. The various seeds/plants would not be mixed, there would be a unified planting of seeds. In this way, the production would be efficient and labour could be saved, and certain quality standards in production could be guaranteed. In reality, the current situation is not like that; between the two of us, you would use this seed and I, very likely, would not agree [with the choice of seed …]. The more you guide them to cultivate a certain kind of seed, the more they will refuse to do so. This has something to do with traditions of being used to growing one seed passed down. (Anonymous 1 2008)

However, there is one important aspect in which the promotion of agrarian change may have had a surprisingly positive effect on inclusiveness. The rise of project-based funding and the fierce competition attached to this type of policy process requires local governments to prepare the best possible applications for projects. In order to do so, they have to communicate and coordinate with parts of the local population, as noted by Zhe and Chen (2011). This participatory aspect of agrarian-change policies may be a source of moral legitimacy for the local state and its activities if villagers agree that these new methods have made cadres more attentive to their needs.

Conclusion

How Beijing interacts with its local governments to get them started on projects and to what degree it succeeds in so doing has been one of the most interesting and controversial topics among observers of the Chinese political system in recent years. Much of the existing debate has focused on the role of policy-steering instruments for policy implementation. However, what has been less featured in the analysis so far is the link between the design of a policy and its implementation. This article has set out to address this question and uses policies in the field of agrarian change as the empirical basis to illustrate its argument. It argues that one important factor for the success of agrarian-change policies in terms of their local adoption is the great fit of these policies for the difficult political, economic and moral context in which these administrations find themselves.

Most policies promoting agrarian change encountered at the field sites were designed in a project-based manner with a focus on providing support for investors in agriculture. Direct involvement in agricultural production was very limited. This reduced the costs of the implementation for administrators and gave local cadres the flexibility to implement “in accordance with local conditions.” The predictability of the measures included in these packages and the quantitative targets attached to them have turned agrarian-change policies into politically efficient policies, which, when pursued, have a high likelihood of achieving results compatible to the annual evaluations. Finally, the careful placement of these policies within a modernisation discourse provides local cadres with cognitive legitimacy or, in other words, a sense of purpose. Given these features, the current set of agrarian-change policies seems to be a good fit for its main target audience, the township and county administrations in less developed inland regions with strongly agriculture-based economies and ways of life.

There is a certain irony in the fact that perhaps the most disruptive changes to the Chinese countryside in the history of the People's Republic take place virtually unnoticed, without much active political steering and long after the end of radical political experiments in agriculture. It appears that peasant resistance against this latest transformation is more difficult to carry out than it ever was against previous socialist experiments (Zweig 1989).