Prior to 1988 water was effectively an open access resource in China. Since then the Chinese government has attempted to use market-based instruments to ease water demand and supply tensions. The instruments chosen, licensing of water use rights and levying charges for possessing these rights, were “market-based” in the sense that by asserting state property over water and fixing a price for it, the central government aimed to bring about adjustments to the demand for water. The intention was to allocate water resources efficiently and cut back on water use. However, in reality, there are indications that the instruments are not sufficiently effective to induce changes in water use behaviour. Areas where water is not abundant continue to be overdrawn. Under increasing signs of water shortages and depletions, we need to ask why these market-based instruments have not worked at the local level. The answers are crucial for an analysis of water management in China.

Current studies on applying market-based instruments in China mainly explain their ineffectiveness by market failure, that is, the market's inability to assess the precise scarcity value of water. Because of this, some argue, abstraction charges are often set too low and provide insufficient economic incentives for users to save water.Footnote ¹ Other scholars suggest establishing a water rights transfer market so that direct valuations of water could be avoided and its “price” would simply vary according to market conditions.Footnote ² These policy discussions reflect the intrinsic limitations of market-based instruments in setting correct levels of charges. While they have considerable success in some countries, why these instruments fail at the local level in China remains unclear.

This article takes a distinctly top-down institutionalist approach, focusing more on the role of government in establishing and maintaining a robust institutional framework for effective market-based instruments than on examining the economic rationale of such instruments. In particular, it suggests that in China these instruments are modified and filtered by institutionalized practices. The way that an instrument is designed and the structure of water management systems in China have caused local water sectors to act differently from original regulatory expectations, leading to local governmental failures. This has affected the implementation of water abstraction policies with several results, including a tendency to set charges at low levels, a lack of necessary monitoring and sanctions, few incentives to collect charges diligently, and failures in providing accessible information for the public. Such governmental failures have turned water abstraction policies into administrative instruments that have very little to do with the “market,” with water users apparently price “insensitive.”

This article starts with a discussion of the need to supplement the rational choice assumption of environmental economics with an institutional dimension. Brief reviews of China's development of water abstraction policies and the current situation of water shortages follow. Four institutional arrangements that potentially lead to local governmental failure are proposed. Then the focus turns to an analysis of the influences of these institutional barriers on local policy design and implementation. The conclusion outlines some of the implications for China's future water resource management.

Rationale for Water Abstraction Charges and some Institutional Considerations

Levies, such as water abstraction charges (WACs), are considered a cost-effective way to correct allocative distortions in the market system.Footnote ³ Based on economic incentives, abstraction charges seek to integrate the interests of the general public with the interests of private abstractors and achieve optimality. The basic idea of WACs is for governments to set a charge for water that a user directly withdraws from surface (rivers and lakes) or ground (aquifers) sources. As profit-maximizers, water users will respond by reducing abstraction to a point where their marginal abatement costs equal charge rates. The ideal charges are often known as Pigouvian taxes.Footnote ⁴ In the case of abstraction charges, Pigouvian taxes should be levied at a level that is equal in value to negative externalities of abstraction, including opportunity costs, scarcity costs and ecological costs. The benefit of using charges is that they allow the desired amount of reduction to be realized at the lowest overall cost to society, as those who can achieve reductions most cheaply make the greatest reductions.Footnote ⁵ In theory, when properly designed and implemented, charges encourage users to adopt cheaper and better technologies, provided a sufficiently low-cost method is available and adopted.Footnote ⁶

This discussion of WACs assumes that the required institutional arrangements exist. To interpret the effectiveness of WACs correctly, however, we need to supplement the rational choice assumption of environmental economics with an institutional dimension. Admittedly, some environmental instruments, such as implementing specific technology standards, can work well with few institutional requirements. However, when policy instruments involve administered prices or created markets, the role of governments and institutions in establishing and maintaining a robust framework for effective market-based instruments becomes more significant. Without the necessary institutional arrangements, the effectiveness of WACs will be diminished for at least two reasons. First, institutional arrangements determine how a policy is designed and interacts with other policies. In particular, they play an important role in the selection of abstraction targets, which is closely related to effective policy design, because identifying an appropriate abstraction charge has proven to be extremely hard.Footnote ⁷ As a result, it is suggested that governments initially establish an arbitrary target and then adjust the levels of abstraction charges to achieve the desired targets.Footnote ⁸ Usually the selection of an appropriate abstraction target involves a process of political bargaining and decision-making to balance economic incentives and a sustainable allocation of water resources.

Secondly, institutional arrangements decide the characteristics of a policy. Once economic charges are fixed to achieve certain targets, the ability and intent of local officials to enforce the charges are affected by the amount of available resources – economic, human and political. Chung has proposed classifying policies according to scope, type and nature.Footnote ⁹ These three factors highlight the level of local discretion in implementation and how resources are pooled together to affect the outcomes of water abstraction policies.

In short, WACs can be useful tools to correct market failure. Without the right institutional setting, however, a WAC can also be an administrative instrument that has very little to do with the “market.” Thus, an institutionalist approach provides deeper insights, explaining why the results of abstraction policies differ across regions, as outlined in the following sections.

The Development of China's Water Abstraction Policies

In China, water abstraction policies (WAPs) focus on water abstraction licences and WACs. They were first introduced by several local governments in the late 1970s to curb groundwater depletion. Despite strong political will to enforce them, early implementation experiences showed that their success was determined to a large extent by local institutional capacity, including inter-organizational co-operation and the ability to monitor water user compliance and to execute sanctions. WAPs were more effective in cities than in geographically larger areas such as provinces. This is best illustrated by contrasting their implementation in Shanghai city and Shanxi province. The administrative area of the latter was 25 times larger than the former, and water users were dispersed across the province, so that a much stronger institutional capacity (which Shanxi province did not have) was required to achieve the same rate of compliance. Subsequently, although both regions implemented WAPs from the early 1980s, groundwater use in Shanghai remained stable during the 1980s while Shanxi province's groundwater abstraction increased rapidly.Footnote ¹⁰

In 1988, WAPs were introduced on a national scale in China's first water law (1988 Water Law, articles 32 and 34). Water was of national concern as overexploitation became serious and local government discontent rose over conflicts in inter-provincial water allocation. Despite the previously mixed results from local implementation, water abstraction charges were chosen probably because they were easier to fit into the institutions at that time, compared to other market-based instruments such as taxes or permit trading schemes, and thus faced less resistance. The 1988 Water Law stated that all individuals or companies withdrawing water directly from surface or ground sources over a certain limit should apply for a licence and pay the corresponding charges. The main targets were industries and water supply companies, and most irrigation was exempted from paying abstraction charges. One benefit of the 1988 Water Law (see article 3) was that it recognized the need to manage water as a state-owned resource, subject to management by China's governments, rather than as an open access resource.

Although provincial governments are required to implement WAPs with local characteristics, not all have given them a high priority.Footnote ¹¹ Besides constraints on institutional capacity in terms of inter-organizational co-operation and monitoring abilities, the main obstacles have involved bureaucratic friction and lack of clarity in water allocation plans. Even though the Ministry of Water Resources (MWR) and its water resource bureaus (WRBs) were given responsibility to implement WAPs, authority was not fully removed from other ministries. In particular, existing water-related sectors still had authority to “co-ordinate” water conservancy sectors in decision-making. For instance, at least two other bureaus were involved in urban groundwater management along with WRBs. Conflicts arose in decision-making because each bureau had its own interests and priorities. In addition, political infighting occurred in inter-ministerial conflicts over exempting water withdrawal activities from WAPs. The inconsistency of rules towards hydroelectric and thermal power plants was a good example of this, highlighting the conflict of interests between the MWR and the Ministry of Electricity and Industry.Footnote ¹² Another explanation of why many provincial governments delayed in implementing WAPs was the lack of clear targets in the water allocation plans to resolve cross-boundary water conflicts. As provincial governments worried that other provinces would “free-ride” on their water management efforts, they chose to wait and see what might actually work.

In 2002, two events indicated the urgency felt by the central government to tackle water shortages. First, it began a very controversial south–north water transfer project, which aimed to divert water to northern China where the rainfall is low and the rivers were running dry. Secondly, the State Council promulgated a new water law. The 2002 Water Law reaffirmed the importance of WAPs and granted MWR and local WRBs more managerial power to implement them. For example, without having to consult or co-ordinate with other sectors, water conservancy sectors now had full authority to process and approve abstraction licences. Today all provinces, autonomous regions and municipalities in China have enacted implementing regulations on WAPs.Footnote ¹³

The National Context

China's demand for water resources has continued to increase in recent decades.Footnote ¹⁴ Industrial demand for water nearly tripled between 1980 and 2007. Ironically, from 2002 when China promulgated the new Water Law, industrial water use began to increase sharply because of industrial booms in several provinces (see Figure 1). Moreover, as a result of population growth and urbanization, total household water use has gone up steadily. In contrast to the increase in demand, China's water supply has been relatively unstable. Annual rainfall is distributed unevenly across regions and seasons. In general, areas north of the Yangtze River are more arid than areas in the south. Also, in some years, precipitation can be eight times higher than in other years. The situation has deteriorated with global climate change, which often manifests itself in a lack of rainwater. Serious droughts in 2009 and the beginning of 2010, for example, led provinces to mine groundwater for daily water use and irrigation. According to official statistics, China's estimated annual water deficit is around 20 to 40 billion cubic metres, which is equivalent to Guangdong province's annual water usage in 2007, and has caused losses in industrial production in the vicinity of 230 billion yuan per year.Footnote ¹⁵

Note:The Ministry of Water Resources derives its water usage statistics from hydraulic project supply data. Realistic water consumption estimates would need to include the quantities of water withdrawn directly from various sources.Source:Shuilibu (Ministry of Water Resources), Zhongguo shuiziyuan gongbao 1997–2008 (China Water Resource Bulletin 1997–2008) (Beijing: Zhongguo shuili shuidian chubanshe). (colour online)

Figure 1: Changes in China's Industrial and Household Water Use

Two signs of overdraft have appeared. First, more areas have seen cases of increased pumping costs, ground subsidence, soil salinity and saltwater intrusion caused by groundwater overexploitation. Drawing groundwater in rural areas is becoming more difficult and costly.Footnote ¹⁶ In a recent survey of 73 villages from five provinces in China, over 60 per cent of the interviewees believed that groundwater depletion had become a severe problem that threatened their livelihoods because village people generally rely on better-quality well water for drinking.Footnote ¹⁷ In addition, continuing overextraction threatens groundwater basins because eventually the gravel and sand in the water-bearing layer compacts so much that it can no longer hold water. The second sign of overdraft is the drying-up of several rivers and disappearance of lakes. For example, the Hai River (Huaihe 淮河), the Liao River (Liaohe 辽河) and the Yellow River (Huanghe 黄河), three of the main rivers in northern China, all suffered serious cut-offs in flows in 2002, the main stream of the Liao for as long as 158 days.Footnote ¹⁸ At that time Nansi Lake (Nansihu 南四湖), the biggest freshwater lake in northern China, was close to being completely dry.

Even officials from MWR have admitted that it is an urgent priority to re-examine the application of China's water abstraction policies.Footnote ¹⁹ As the tension grows, the central government has to revert to planned engineering interventions. As Table 1 shows, investments in water resource projects, which mainly include building reservoirs, water transfer projects and agricultural irrigation projects, reached 46.78 billion yuan in 2008 to become the largest part of the government investments in water conservation projects.

Table 1: China's Investments in Water Conservation Projects (percentage by kind of project)

Source:

Shuilibu (Ministry of Water Resources) (ed.), Shuili tongji gongbao 2001, 2005, 2008 (Statistic Bulletin on China Water Activities) (Beijing: Zhongguo shuili shuidian chubanshe, 2001, 2005, 2008).

The Real Issue: Governmental Failures

Since the 2002 Water Law, the central government has strengthened institutions concerning abstraction licences and abstraction charges policies. At the same time, however, its hesitancy towards new market-based instruments has been reflected in policy design and relevant institutional arrangements. Four characteristics, in particular, reduce local government incentives to implement WAPs.

First, although it is clear that China's water resources are state-owned in principle, the concept of state ownership is not well expounded in law. On the one hand, the state is represented by local governments, which plan for groundwater and regional surface exploitation. On the other, it is represented by the central government, which determines basin-level water distribution and plans major water transfer projects that override local water decisions. This ambiguity in the definition of ownership rights can lead to insecurity and unwillingness to make efficient long-term decisions about investment and harvesting from water resources.Footnote ²⁰ Economic reforms and progress in decentralization have exposed further weakness in this legal structure, creating disputes between state institutions.

Secondly, China's water management system lacks clearly defined abstraction targets. As discussed, an appropriate target is a pre-requisite for setting economic charges and an effective policy design. Systems of total abstraction quantity control (zong liang kong zhi 总量控制) were rapidly put into place at both the national and local level after the central government decided abstraction targets were necessary (1988 Water Law, article 47). Initially, seven river basin management committees (RBMCs) were required to formulate water allocation plans for their basins, in which provincial targets were determined; then provincial governments fixed abstraction targets for prefectures in their provincial water plans. One problem is that this centralized system requires perfect co-operation among all levels of governments and ministries and the availability of reliable hydrological and monitoring data. If any actor fails the plans do not materialize. In fact, although the MWR mandated, as long ago as 1988, that a variety of water plans be promulgated, most were not completed at the local level until recently.

Furthermore, the current system of total abstraction quantity control requires the central government to identify a clear and balanced abstraction target right from the beginning. However, targets have often been fixed in the basin water allocation plans without consultation with local interests. Typically the resolution of conflicts is postponed to the local implementation stage, but either the plans fail to be written or excessive abstraction occurs. Take the targets fixed in the 1987 Yellow River Water Allocation Plan (Huanghe shuiliang fenpei fangan 黄河水量分配方案) as an example. Rather than making the targets a proportion of the available flow, the plan fixed annual water distribution in provincial targets for over ten years (Table 2). The problems are obvious: this did little to respond to the Yellow River's inflow variability between seasons and between years; and fixed targets of abstraction cannot match the dynamic needs of local socio-economic systems. Therefore, it is unrealistic for local authorities to conform to the basin plans and establish water plans based on them, leading local WRBs to issue water abstraction licences without attention to constraining water consumption. Studies show that for provinces like Shandong and Inner Mongolia, the annual water withdrawn constantly exceeded the given allocations.Footnote ²¹ Not surprisingly, such inflexible and partial targets contributed to more frequent and longer periods of drying up in the lower reaches of the Yellow River.Footnote ²²

Table 2: Yellow River Allocation Plan, 1987 Annual Water Distribution

Source: Guanyu Huanghe Kegong shuiliang fenpei fang' an baogao de tongzhi 1987 (Circular of the State Council Concerning Yellow River Allocation Plan, 1987)

A third feature of China's current water management institutions is the lack of an effective institutional process within the RBMCs. The main issues in river basins revolve around the fact that, although abstractions are local activities, their effects involve local water shortages and externality costs in other jurisdictions.Footnote ²³ Decentralized local governments have little control over the amount of water available within their boundaries, as water availability depends on the aggregate level of abstraction in other jurisdictions. Therefore, to achieve optimality, a basin-based institutional process is needed through which local governments can collaborate on issues, including collecting and researching hydrological data, forecasting meteorological events, evaluating the impacts of abstraction, proposing overall planning of water allocation and fixing standards for abstraction charges.

Currently in China, however, WAPs do not have a legal basis for an effective institutional process, which is evident in several processes. First, the RBMCs' administrative rights are much fewer than those prescribed by legislation. China's current water management combines jurisdictional management with river-basin management (2002 Water Law, article 12). In other words, RBMCs have to share authority and responsibilities with provincial governments on issues regarding river basins. In fact, most authority is still retained by local governments. A majority of the abstraction approvals are decided by local WRBs without consultation with RBMCs. In addition, fundamental responsibilities, such as monitoring and penalizing negative abstraction activities, are not given to RBMCs. When facing overexploitation, for example, RBMCs' actions are restricted to “persuading” or addressing notices of warning.Footnote ²⁴ Second, RBMCs can only approve or reject plans of abstraction from the main stream of the river, with no authority over abstraction from tributaries or ground sources.Footnote ²⁵ This poses obstacles to integrated river-basin management because it runs counter to the principle that main streams, tributaries and groundwater in a river basin are closely interrelated. Third, RBMCs lack a process for sharing and discussing critical information on abstraction regularly with provincial governments. Local authorities do not disclose scientific information, such as data on existing water resources and the annual amount of abstraction licences approved by local authorities, unless RBMCs make special requests. The lack of regular hydrological and abstraction information updates means that the amounts of water available and consumed are not clear and establishing a basin water plan becomes unrealistic. Fourth, current legislation does not enable RBMCs to establish a commission of stakeholders from provincial, prefectural and county levels, or to have consultative mechanisms to bring in community voices.Footnote ²⁶ This greatly reduces the benefits of public supervision and participation in river basin management.

Last, but not least, the dual-responsibility system in water administration remains a potential constraint to water management in China. The dual-responsibility system refers to the provincial/prefectural (county-level) WRBs, which are led by both superior-level WRBs and people's governments at the same level. In other words, while receiving technical assistance and infrastructure investments from upper-level WRBs, local WRBs also depend on people's governments for more funding to build and maintain water projects.Footnote ²⁷ Especially in recent years, dependence on local revenue has increased rapidly due to the central government's worsening finances.Footnote ²⁸ As Table 3 shows, local revenue has overtaken central revenue, becoming the biggest source of investments in fixed assets for water conservancy. In 2007, local investments accounted for over half of the total.Footnote ²⁹ In fact, local governments tend to have far more authority over WRBs than do the higher-level WRBs, because the former play a decisive role in appointments and promotion of senior officials at WRBs. Some qualitative research has shown that support from local governments plays a pivotal role in the enforcement of environmental regulations in China.Footnote ³⁰

Table 3: Investments in Fixed Water Conservation Assets, 2005–2007 (percentage by source)

Source:

Shuilibu (Ministry of Water Resources) (ed.), Shuili tongji gongbao 2005, 2006, 2007 (Statistical Bulletin on China Water Activities 2005, 2006, 2007) (Beijing: Zhongguo shuili shuidian chubanshe, 2006, 2007, 2008).

The main problem with the dual-responsibility system is that it separates the WRBs' responsibilities from their sectoral interests, making them less independent and meaning that they do not have water resources protection as their sole purpose. As WRBs are directly led by local governments, their initiatives to protect water resources greatly depend on the attitude of the local government. Since the 1978 economic reform, however, local governments often have more incentive to prioritize economic growth above anything else. This is because central government evaluates the performance of local governments according to economic outputFootnote ³¹; and also, as the decentralization of power is not fully guaranteed or stipulated by law, local protectionism emerges and local authorities tend to maximize their short-term benefits.Footnote ³² Most importantly, fiscal developments have supported local governments' increased pursuit of their own interests by asserting their economic identity (as provinces have been cut off from central support and become economically self-reliant).Footnote ³³ Therefore, when abstraction policies contradict local governments' pervasive “pro-growth” priorities, WRBs are required to “take into account the overall economic situation.”Footnote ³⁴ It is not unusual for local governments to interfere with the collection of charges or even use exemption as a preferential policy to promote trade and attract investments.Footnote ³⁵ If local WRBs insist upon exercising their responsibilities to implement WAPs this can affect their own sectoral interests. Thus, this dual-responsibility system leaves WRBs in a fairly passive position. Lack of government support encourages officials to seek refuge in bureaucracy, manifested in a strict adherence to the book or in the generation of meaningless paperwork in the course of regulatory enforcement.Footnote ³⁶

The implications of the features discussed above are significant. In addition to the 2002 Water Law granting more autonomy to local authorities, the institutional settings have given local WRBs increased independence and discretion to deal with water abstraction activities from local and short-term perspectives. Even though the central government enjoys administrative authority over lower-level governments, it faces the problem of monitoring the compliance of multiple jurisdictions that have very different local conditions. Amid the pervasive “pro-growth” priorities at the local level, more discretion could increase the possibility of governmental failure in forming and maintaining institutions to carry out WAPs effectively. In particular, governmental failure has already manifested in several ways at the local level, through: setting low levels of water abstraction charges; a lack of necessary monitoring and sanctions; low incentives to collect charges diligently; and failure to provide accessible information for the public. The following sections focus on each of these aspects.

Setting Low Water Abstraction Charges

Because the central government did not provide clear targets for abstraction, it is impossible for local governments to design a levy system and base WACs on specific abstraction targets. Instead, provincial governments are given the autonomy to design levy systems and set WAC standards with local features. It is common practice to assess charges based on actual abstraction volumes and a unit charge rate determined by the types of users and sources of abstraction. In general, levies for abstractions are calculated as follows:

where, for type of user i that withdraws water from source j, L _ij is the total of levies, V _ij is the actual volume of water withdrawn, R _ij is the unit charge rate and P _ij is the volume of water permitted on abstraction licences. If the actual abstraction exceeds that permitted, a progressively higher charge rate E _ij is levied on excess volumes. Both R and E are set by provincial governments. Once fixed they usually remain unchanged for many years.

Much more detailed guidelines on designing levy systems and setting WACs should be given to the provincial governments. For example, factors like seasons and water loss should influence the actual payable abstraction charges. At present, the only charge-setting principle given by central government is that both scarcity and ability to pay should be considered while fixing the abstraction charges (2006 Regulation on Management of Water Abstraction Licences and Water Resource Charge, article 29). Although the principle sounds equitable, it has limited applicability in most provinces where undeveloped economies coincide with water shortages. Figure 2 illustrates the level of per capita water availability and GDP per capita by province in 2008. Provinces shown in the bottom left suffer from both a low GDP per capita and low water availability per capita. On one hand the lack of ability to pay requires provincial governments to fix a low level of abstraction charges, while on the other the scarcity of water requires them to increase the level of charges to curb depletions. In this context, the central government's excessively general principle not only provides little guidance for provincial governments in setting charges, but also creates the risk of governmental failure in favour of local protectionism.

Source:China Statistics Yearbook 2009 and China Water Resources Bulletin 2008.

Figure 2: Water Availability Per Capita and GDP Per Capita in 2008

It is therefore not surprising to see nearly all studies on China's WAPs indicate that the majority of provincial governments have set WAC standards so low that they have lost their function as an economic lever.Footnote ³⁷ For example, one of the most arid provinces in northern China, Ningxia province, levied only 0.07 yuan per m³ for taking surface water for industrial use in 2006. Similarly, the groundwater over-exploited Shanxi province levied 0.7 yuan per m³ for taking groundwater privately, only half the price of water supplied by public water companies.Footnote ³⁸ The low level of abstraction charges has not relieved water overexploitation pressures and provides little economic incentive for big industries to change behaviour or promote technological innovations. Instead, the bureaucratic application process for abstraction licences risks reducing the competitiveness of small privately owned enterprises and increasing the possibility of rent-seeking.

Lack of Monitoring and Sanctions

To ensure water users comply with regulations, accurate monitoring of water abstraction as well as proper sanctions exercises are required.Footnote ³⁹ Under the dual-responsibility system, however, China's upper-level WRBs face huge principal-agent problems, because local WRBs are under substantial pressures from local governments to ensure that enforcement will not contradict development strategies. Even an official from the MWR has admitted that it is hard for superior-level WRBs to ensure enforcement of monitoring or sanctions at the local level, especially when enforcement can only be achieved through administrative orders and few resources are provided.Footnote ⁴⁰

In particular, monitoring and sanctioning activities by local WRBs have been hindered by limited personnel and financial resources. Since administration reform, there has been a continuing reduction in China's local WRB personnel. Table 4 shows that while the numbers of staff in the central water conservancy authority increased, WRB staff at the local level were reduced by one-fifth between 2001 and 2008. One consequence is that there are insufficient personnel to undertake managerial responsibilities of WAPs at the street level. An officer from Shihezi city (Shihezi shi 石河子市) WRB in Xinjiang province, for example, complained that there was only one staff member in charge of approving abstraction licences and another two in charge of collecting charges for the entire region.Footnote ⁴¹ Increased management duties can challenge street-level officials, especially because most have backgrounds in engineering and hydrology. With little time, they are burdened with devising strategies to carry out on-site inspections and increase fee collection, while they also have to prevent and prosecute illegal abstractions.

Table 4: Staff Establishment in Central and Local Water Conservancy Authorities (thousand people)

Source:

Shuilibu, Statistical Bulletin on China's Water Activities 2001, 2005, 2008.

WRBs do not have sufficient financial resources available to them. As mentioned, China's decentralization has meant that local WRBs are less dependent on central investment. Instead, they are encouraged to seek new financial channels, such as local government funding, or foreign or domestic loans. This lack of financial resources has led to inadequate investments in staff training and monitoring equipment for industrial users.Footnote ⁴² In Liaoning province, for example, fewer than 50 per cent of the users have properly installed and operating water meters.Footnote ⁴³ As a result, most charges are based on estimates of production, which do not take water wastage into account. In addition, decreases in central investment have removed funding allocation as a tool whereby the MWR can encourage lower-level WRBs to comply with central policies.

Few Incentives to Collect Charges Diligently

China's current institutions do not adequately emphasize supervision of governmental authorities. Under the dual-responsibility system, another principal–agent problem faced by the central government is the method of collecting charges. In the 2002 Water Law, the central government stated that revenues from abstraction charges should be delivered directly to the treasuries administered by local financial bureaus. Local WRBs receive funds separately from these financial bureaus to cover administrative costs. This move to separate revenues from expenditures has reduced incentives for local WRBs to collect charges. In an interview, one local WRB staffer said: “While we have to contribute more resources to increase fee collection, collecting more charges or less in fact makes no difference to our welfare.”Footnote ⁴⁴ Consequently, insufficient charge collection is common, particularly in rural and remote areas. In some areas, actual revenues are less than one-fifth of the due abstraction charges.Footnote ⁴⁵ Kaifeng city (Kaifeng shi 开封市) in Henan province, for example, was reported to have collected a total of 138.5 million yuan in abstraction charges in 2007. However, according to published data, it should have collected at least 167.8 million from water supply companies alone.Footnote ⁴⁶ Furthermore, as the local WRBs' ability to get administrative funds depends more on the bargaining power of officials or their relationships with local governments, enforcement of WAPs becomes less independent.

Inadequate Public Access to Information

Public participation based on adequate information can increase the possibility of achieving efficient outcomes, because local users tend to identify issues relating to local natural resources clearly and have relatively accurate information on local carrying capacity.Footnote ⁴⁷ In addition, effective participation leads to an adjustment of existing power relationships and reduces the possibility of rent-seeking behaviour.Footnote ⁴⁸ Research finds that, when there is a deficiency of organized social forces, officials may adopt a more “flexible” approach to enforcement.Footnote ⁴⁹ As to tackling water shortages, accessible information is particularly important because, unlike other environmental problems that are immediately observable, water shortages are hard for the general public to assess.

Currently, China's legislation specifies neither the kind of information that needs to be published nor the frequency of its publication. Indeed, most of the information is kept as internal resources by local authorities. Accessible information is incomplete, brief or uses inconsistent statistical standards. In addition, very little information about enforcement has been disclosed. For example, the number of licences granted and rate of charge collection are often not made public. With this lack of information and lack of public participation, accountability in the implementation process is almost non-existent. In addition, local levy systems are often designed differently and their standards vary. When there is no system of accessible information, it is difficult for researchers to evaluate the impacts of WAPs quantitatively. Consequently little advice can be given to improve implementation.