Based on whether the ‘One River, One Policy’ is implemented or not, this paper used the questionnaire data of village (community) river chiefs in nine districts and counties under the jurisdiction of Jinhua City, Zhejiang Province, and used the Difference-In-Differences Model to carry out an empirical analysis. The findings are threefold. Firstly, the ‘One River, One Policy’ implemented by primary river chiefs can improve the water quality of the river reaches in their charge. Secondly, annual river patrol times and public support can improve the water quality of the river reaches in charge of primary river chiefs. Thirdly, the River Chief System (RCS) can promote the treatment of river pollution under ‘Responsibility Contracting’, but for primary river chiefs, there is still a suspicion of whitewash pollution control behavior, so whether to establish a long-term mechanism of the RCS is still uncertain. The purpose of this paper is to provide empirical support and decision-making reference for constructing the positive incentive mechanism of the behavior of primary river chiefs in the reconstruction of the RCS, which has important practical significance for effectively improving the RCS and comprehensively promoting the Chinese experiences of river pollution control.

  • ‘One River, One Policy’ implemented by primary river chiefs can improve the water quality of the river reaches in their charge.

  • Annual river patrol times and public support can improve the water quality of the river reaches in charge of primary river chiefs.

  • The River Chief System (RCS) can promote the treatment of river pollution under ‘Responsibility Contracting’, but for primary river chiefs, there is still a suspicion of whitewash pollution control behavior, so whether to establish a long-term mechanism of the RCS is still uncertain.

The River Chief System (RCS) refers to the appointment of a major Party (The Communist Party of China, or the CPC for short) and government officials at local levels as river chiefs, who are responsible for organizing and leading the management and protection of corresponding rivers and lakes. The work content of the RCS mainly includes six aspects: (1) strengthening water resources protection and fully implementing the strictest water resources management system; (2) strengthening the management and protection of the shorelines of rivers and lakes, strictly managing and controlling water ecological space such as water areas and shorelines, and prohibiting encroachment on river channels and reclamation of lakes; (3) strengthening the prevention and control of water pollution, coordinating the control of water pollution and onshore pollution, identifying pollution sources entering rivers and lakes, and optimizing the layout of sewage discharge outlets into rivers; (4) improving water environment, ensuring the safety of drinking water sources, stepping up efforts to control black and smelly bodies of water, and ensuring that rivers and lakes are clean and beautiful, with clear water and green banks; (5) strengthening the restoration of the water ecosystem, delimiting the scope of river and lake management in accordance with law, and strengthening the systematic treatment of mountains, rivers, forests, farmland, and lakes; and (6) strengthening law enforcement and supervision and cracking down on illegal activities related to rivers and lakes. The implementation of ‘One River, One Policy’ is the core measure in which the RSC plays a fundamental role. It means that local governments at the provincial, municipal, county (district), town (township), and village (community) levels in China formulate and implement specific pollution control policies for rivers and lakes within their jurisdiction.

The RCS is a system with Chinese characteristics to strengthen government functions, which is the full embodiment of environmental control in river pollution control. Its essence is to coordinate the system, strengthen authority, and supervise responsibilities for the existing river management (Manhong, 2018). After the General Office of The CPC Central Committee and the General Office of The State Council issued Opinions on Comprehensively Promoting River Chief System in 2016, the RCS has shifted from a bottom-up independent exploration to a top-down unified will. This system combines watershed distribution and administrative division and concentrates decentralized authority of river management in the hands of the main Party and government leaders at all levels. It realizes top-down target and task operation through the pressure-type leader–responsibility interaction mechanism of leadership–control and control–leadership and, to a certain extent, solves the imbalance problem caused by the overlapping of functions and powers among water administration departments. On the other hand, under the principle of territorial management and the operation mechanism of signing responsibility letters at each level, the main responsibility and great pressure of river pollution control are transferred to primary river chiefs from top to bottom. In the whole system framework, including the meeting system, inspection system, information reporting system, work supervision system, and assessment and accountability system, village (community) river chiefs are often too tired to deal with, which is easy to induce formalism, thus weakening the effect of the RCS on river pollution control. Some studies have found that although the RCS has achieved the initial effect of water pollution control, it has not significantly reduced the deep pollutants in the water, which may reveal the whitewash pollution control behavior of local governments that treats the symptoms rather than the root causes (Kunrong & Gang, 2018).

In theory, many scholars generally recognize the good effect of enhancing river pollution control performance by strengthening vertical mechanisms (Jianguo & Ye, 2017; Ye, 2017). The success of the RCS lies in solving the dilemma of multiple water control in the river and lake management and protection and mobilizing the enthusiasm of local governments in controlling water (Yunyan, 2019). However, scholars have also consistently pointed out the ability dilemma, responsibility dilemma, and organizational logic dilemma faced by excessive reliance on authority (Min, 2015; Jianguo & Ye, 2017; Ye, 2017; Yunyan, 2019). There are some problems in the RCS, such as the lack of rule of law and policy innovation, capacity dilemma and action dilemma faced by policy implementors and agencies, lack of recognition and support from implementors and the public due to poor interinstitutional and public participation, and the lack of support from formal and informal institutional factors (Jiajun, 2019). In addition, the conflict and ambiguity of the RCS lead to the inevitable deviation from policy implementation. The imbalance of power and responsibility structure and the organizational foundation with relatively weak overall capabilities further increase the deviation of policy implementation. The ‘high task, high pressure’ task environment co-shaped by institutional pressure under the pressure-type system, political pressure to fight the battle against pollution, institutionalized pressure to overemphasize work legacy, etc., further increases the policy implementation bias (Xinfu & Jianguo, 2022). Especially from the perspective of new institutional economics, the RSC itself has an inevitable fundamental defect, that is, the information asymmetry under the principal agent (Shuming & Mengmeng, 2011).

In recent years, some scholars have begun to use empirical means to investigate the effect of the RCS on river pollution control and reached inconsistent conclusions (Lin et al., 2022; Wanhua et al., 2022; Yiwen et al., 2022). However, these studies have not conducted in-depth analysis from the micro level, especially from the perspective of primary river chiefs' behavior under ‘Responsibility Contracting’. In view of this, based on whether the ‘One River, One Policy’ is implemented or not, we use the questionnaire data from village (community) river chiefs and apply the Difference-In-Differences (DID) Model to empirically examine the effect of the RCS on river pollution control under ‘Responsibility Contracting’, which is of great importance to summarize China's experiences and lessons on water environmental pollution control. The purpose of our study is to provide empirical support and decision-making reference for constructing the positive incentive mechanism of the behavior of primary river chiefs in the reconstruction of the RCS, which has important practical significance for effectively improving the RCS and comprehensively promoting the Chinese experiences of river pollution control.

Data source

We conducted a sample survey among village (community) river chiefs in nine districts and counties under the jurisdiction of Jinhua City, Zhejiang Province from January to February 2023. A total of 683 questionnaires were sent out and 675 were recovered. After screening and evaluation, 667 were confirmed to be valid, accounting for 98.8% of the recovered questionnaires. The regional distribution of the samples is provided as follows: 163 in Wucheng District (24.44%), 42 in Jindong District (6.30%), 38 in Yiwu City (5.70%), 61 in Dongyang City (9.15%), 84 in Yongkang City (12.59%), 156 in Lanxi City (23.39%), 40 in Pujiang County (6.00%), 63 in Wuyi County (9.45%), and 20 in Pan'an County (3.00%).

Model setup

Water quality is the target of river pollution control, ‘One River, One Policy’ is the key measure to achieve river pollution control through the RCS, and village (community) river chiefs, as primary river chiefs, are the important subjects to implement ‘One River, One Policy’. In order to evaluate the effect of the RCS on river pollution control in China, we divide the study area into the implementing area and the non-implementing area and establish a basic DID Model as below:
formula
(1)
where Q is the water quality, T is the time dummy variable of whether the sample data come from the period after the implementation of ‘One River, One Policy’, P is whether to implement ‘One River, One Policy’, α0, α1, α2, and δ are the parameters to be estimated, θ is the county fixed effect, μ is the random disturbance term. According to Equation (1), the expressions of water quality in the implementing area and the non-implementing area can be obtained. For the implementing area, P = 1, then:
formula
(2)
Before and after the implementation of ‘One River, One Policy’, the expected water quality in the implementing area is:
formula
(3)
Therefore, before and after the implementation of ‘One River, One Policy’, the average change value of water quality in the implementing area is:
formula
(4)
For the non-implementing area, P = 0, then:
formula
(5)
Before and after the implementation of ‘One River, One Policy’, the expected water quality in the non-implementing area is:
formula
(6)
Therefore, before and after the implementation of ‘One River, One Policy’, the average change value of water quality in the non-implementing area is:
formula
(7)
Excluding the system differences between the implementing area and the non-implementing area, if ‘One River, One Policy’ is not implemented in the implementing area, the average change value of water quality in the implementing area should be the same as that in the non-implementation area, with α1. But, in fact, before and after the implementation of ‘One River, One Policy’, the average change value of water quality in the implementing area is α1 + δ, so the net impact of the implementation of ‘One River, One Policy’ in the implementing area is:
formula
(8)
In the empirical analysis, considering the influence of control variables on water quality, Equation (1) is extended as follows:
formula
(9)
where i represents village (community) river chiefs as primary river chiefs, j represents the period; Xk is the set of control variables, and λk is the corresponding set of parameters to be estimated.

Variable definition

Explained variable

In order to investigate the effect of the RCS on river pollution control in China, we set the water quality as the explained variable and take a multi-classification virtual value for it. According to China's Environmental Quality Standards for Surface Water, water quality in China is divided into five classes according to the environmental functions and protection objectives of surface water areas: Class I is mainly applicable to source water and national nature reserves; Class II is mainly applicable to primary protection areas of surface water sources for centralized drinking water, habitats of rare aquatic organisms, spawning grounds for fish and shrimp, feeding grounds for young fish, etc.; Class III is mainly applicable to secondary protection areas of surface water sources for centralized drinking water, winter grounds for fish and shrimp, migration channels, aquaculture areas and other fishery waters, and swimming areas; Class IV is mainly applicable to general industrial water areas and recreational water areas where human body is not in direct contact; Class V is mainly applicable to agricultural water areas and water areas with general landscape requirements. In addition, water quality whose pollution level exceeds Class V is classified as Class VI. The water quality here is rated as 0, 1, 2, 3, 4, and 5, from bad to good. The definition of the explained variable is shown in Table 1.

Table 1

Definition of variables.

VariablesDefinition
Water quality (QWater quality of the river reach in charge of the surveyed village (community) river chief (classified value): Class VI = 0, Class V = 1, Class IV = 2, Class III = 3, Class II = 4, Class I = 5. 
Time dummy variable (TWhether the sample data are from the period after the implementation of ‘One River, One Policy’ (classified value): no = 0, yes = 1. 
Whether to implement ‘One River, One Policy’ (PWhether the surveyed village (community) river chief has formulated and implemented specific river pollution control policies for the river reach in his/her charge (classified value): no = 0, yes = 1. 
Annual river patrol times (X1Number of times in a year that the surveyed village (community) river chief patrols the river reach in his/her charge (continuous value). 
Communicating with the public (X2Frequency of communication between the surveyed village (community) river chief and the public on river pollution control (classified value): no communication = 1, little communication = 2, frequent communication = 3, more communication = 4, lots of communication = 5. 
Training for work (X3Whether the surveyed village (community) river chief has participated in the relevant training of river pollution control (classified value): no = 0, yes = 1. 
Public support (X4Extent to which the public support the work of the surveyed village (community) river chief in controlling river pollution (classified value): very little support = 1, not very support = 2, general support = 3, relatively support = 4, very support = 5 
VariablesDefinition
Water quality (QWater quality of the river reach in charge of the surveyed village (community) river chief (classified value): Class VI = 0, Class V = 1, Class IV = 2, Class III = 3, Class II = 4, Class I = 5. 
Time dummy variable (TWhether the sample data are from the period after the implementation of ‘One River, One Policy’ (classified value): no = 0, yes = 1. 
Whether to implement ‘One River, One Policy’ (PWhether the surveyed village (community) river chief has formulated and implemented specific river pollution control policies for the river reach in his/her charge (classified value): no = 0, yes = 1. 
Annual river patrol times (X1Number of times in a year that the surveyed village (community) river chief patrols the river reach in his/her charge (continuous value). 
Communicating with the public (X2Frequency of communication between the surveyed village (community) river chief and the public on river pollution control (classified value): no communication = 1, little communication = 2, frequent communication = 3, more communication = 4, lots of communication = 5. 
Training for work (X3Whether the surveyed village (community) river chief has participated in the relevant training of river pollution control (classified value): no = 0, yes = 1. 
Public support (X4Extent to which the public support the work of the surveyed village (community) river chief in controlling river pollution (classified value): very little support = 1, not very support = 2, general support = 3, relatively support = 4, very support = 5 

Explanatory variable

The explanatory variable is the cross-multiplication term between the time dummy variable and whether to implement ‘One River, One Policy’. The time dummy variable is taken as a binary dummy value, with a value of 1 assigned to the sample data from the period after the implementation of ‘One River, One Policy’, and 0 assigned to the sample data not from the period after the implementation of ‘One River, One Policy’. Whether to implement ‘One River, One Policy’ is also taken as a binary dummy value, with a value of 1 assigned to having formulated and implemented specific river pollution control policies for the river reaches in charge of the surveyed village (community) river chiefs and 0 assigned to not having formulated and implemented specific river pollution control policies for the river reaches in charge of the surveyed village (community) river chiefs. The definition of the explanatory variable is shown in Table 1.

Control variables

According to the work contents of the RCS, we select four factors, including annual river patrol times, communicating with the public, training for work, and public support, as the control variables to investigate the effect of China's RCS on river pollution control. The annual river patrol times are taken as an actual value, that is, the number of times in a year that the surveyed village (community) river chiefs patrol the river reaches in their charge. Communicating with the public is taken as a multi-classification virtual value. According to the frequency of communication between the surveyed village (community) river chiefs and the public on river pollution control, values of 1, 2, 3, 4, and 5 are successively assigned to those who have no communication, those who have little communication, those who have frequent communication, those who have more communication, and those who have lots of communication. The training for work is taken as a binary dummy value, with a value of 1 assigned to having participated in the relevant training of river pollution control and 0 assigned to not having participated in the relevant training of river pollution control. Public support is taken as a multi-classification virtual value. According to the extent to which the public support the work of the surveyed village (community) river chiefs in controlling river pollution, values of 1, 2, 3, 4, and 5 are successively assigned to those who have very little support, those who have not much support, those who have general support, those who have relatively strong support, and those who have very strong support. These four variables are generally considered to have a positive effect on river pollution control, that is, the more times the annual river patrol, the more communication with the public, participation in training for work, and the more support from the public, the better the water quality of the river reaches in charge of village (community) river chiefs is likely to be. In addition, the county fixed effect is controlled. The definitions of control variables are shown in Table 1.

Descriptive statistics

Table 2 lists the statistical results of sample characteristics for all variables. There are 336 village (community) river chiefs (50.37%) whose sample data are from the period after the implementation of ‘One River, One Policy’. Among them, there are 0 (0%) village (community) river chiefs for Class VI water quality, 2 (0.30%) village (community) river chiefs for Class V water quality, 54 (8.10%) village (community) river chiefs for Class IV water quality, 117 (17.54%) village (community) river chiefs for Class III water quality, 133 (19.94%) village (community) river chiefs for Class II water quality, and 30 (4.50%) village (community) river chiefs for Class I water quality. There are 331 village (community) river chiefs (49.63%) whose sample data are not from the period after the implementation of ‘One River, One Policy’. Among them, there are 118 (17.69%) village (community) river chiefs for Class VI water quality, 108 (16.19%) village (community) river chiefs for Class V water quality, 102 (15.29%) village (community) river chiefs for Class IV water quality, 3 (0.45%) village (community) river chiefs for Class III water quality, 0 (0%) village (community) river chiefs for Class II water quality, and 0 (0%) village (community) river chiefs for Class I water quality. There are 333 village (community) river chiefs (49.93%) who have formulated and implemented specific river pollution control policies for the river reach in their charge. Among them, there are 10 (1.50%) village (community) river chiefs for Class VI water quality, 21 (3.15%) village (community) river chiefs for Class V water quality, 36 (5.40%) village (community) river chiefs for Class IV water quality, 103 (15.44%) village (community) river chiefs for Class III water quality, 133 (19.94%) village (community) river chiefs for Class II water quality, and 30 (4.50%) village (community) river chiefs for Class I water quality. There are 334 village (community) river chiefs (50.07%) who have not formulated and implemented specific river pollution control policies for the river reach in their charge. Among them, there are 108 (16.19%) village (community) river chiefs for Class VI water quality, 89 (13.34%) village (community) river chiefs for Class V water quality, 120 (17.99%) village (community) river chiefs for Class IV water quality, 17 (2.55%) village (community) river chiefs for Class III water quality, 0 (0%) village (community) river chiefs for Class II water quality, and 0 (0%) village (community) river chiefs for Class I water quality.

Table 2

Statistical results of sample characteristics for all variables.

VariablesSample sizeMaximumMinimumMeanStandard deviation
Q 667 2.19 1.49 
T 667 0.50 0.50 
P 667 0.50 0.50 
X1 667 30 3.79 4.32 
X2 667 3.64 1.11 
X3 667 0.77 0.42 
X4 667 3.94 0.90 
VariablesSample sizeMaximumMinimumMeanStandard deviation
Q 667 2.19 1.49 
T 667 0.50 0.50 
P 667 0.50 0.50 
X1 667 30 3.79 4.32 
X2 667 3.64 1.11 
X3 667 0.77 0.42 
X4 667 3.94 0.90 

In terms of control variables, the average number of times in a year that village (community) river chiefs patrol the river reaches in their charge is very small, the frequency of communication between surveyed village (community) river chiefs and the public on river pollution control is high, many village (community) river chiefs have participated in the relevant training of river pollution control, and the extent to which the public support the work of village (community) river chiefs in controlling river pollution is large.

Regression results of DID Model

We use the Ordinary Least Squares Model and the Generalized Ordered Logit Model, respectively, to estimate the parameters of Equation (9). Table 3 lists the regression results of robust standard error. Columns (1) and (3) are the regression results without control variables, and Columns (2) and (4) are the regression results with control variables. Both the time dummy variable and whether to implement ‘One River, One Policy’ have significant positive effects on water quality. The cross-multiplication term between the time dummy variable and whether to implement ‘One River, One Policy’ has a significant positive effect on water quality. Among the control variables, annual river patrol times and public support both have significant positive effects on water quality, but the positive effect of annual river patrol times is not strong. Other factors had no significant effect on water quality.

Table 3

Regression results of the DID Model.

ItemsOrdinary Least Squares Model
Generalized Ordered Logit Model
(1)(2)(3)(4)
T 1.3821*** (0.0748) 1.3591*** (0.0750) 4.3882*** (0.4675) 4.3762*** (0.4683) 
P 0.6284*** (0.1075) 0.6161*** (0.1075) 1.4398*** (0.2732) 1.4356*** (0.2791) 
T × P 0.8954*** (0.1272) 0.8962*** (0.1265) 17.6917*** (0.3639) 17.8217*** (0.3589) 
X1 – 0.0131* (0.0075) – 0.0397 (0.0247) 
X2 – −0.0003 (0.0302) – 0.0008 (00821) 
X3 – −0.0753 (0.0817) – −0.1854 (0.2216) 
X4 – 0.0781** (0.0392) – 0.1979* (0.1069) 
County fixed effect Control Control Control Control 
Constant 0.7638*** (0.0779) 0.4842*** (0.1618) – – 
Sample size 667 667 667 667 
F-statistic 550.92*** 280.98*** – – 
R2 0.7720 0.7758 – – 
Wald χ2 – – 7,628.90*** – 
Pseudo R2 – – 0.4199 0.4243 
ItemsOrdinary Least Squares Model
Generalized Ordered Logit Model
(1)(2)(3)(4)
T 1.3821*** (0.0748) 1.3591*** (0.0750) 4.3882*** (0.4675) 4.3762*** (0.4683) 
P 0.6284*** (0.1075) 0.6161*** (0.1075) 1.4398*** (0.2732) 1.4356*** (0.2791) 
T × P 0.8954*** (0.1272) 0.8962*** (0.1265) 17.6917*** (0.3639) 17.8217*** (0.3589) 
X1 – 0.0131* (0.0075) – 0.0397 (0.0247) 
X2 – −0.0003 (0.0302) – 0.0008 (00821) 
X3 – −0.0753 (0.0817) – −0.1854 (0.2216) 
X4 – 0.0781** (0.0392) – 0.1979* (0.1069) 
County fixed effect Control Control Control Control 
Constant 0.7638*** (0.0779) 0.4842*** (0.1618) – – 
Sample size 667 667 667 667 
F-statistic 550.92*** 280.98*** – – 
R2 0.7720 0.7758 – – 
Wald χ2 – – 7,628.90*** – 
Pseudo R2 – – 0.4199 0.4243 

Note: *, **, and *** indicate significance at 10, 5, and 1% levels, respectively; standard errors are in parentheses; ‘–’ is the default item.

Influence of ‘One River, One Policy’ on river pollution control

Village (community) river chiefs implement ‘One River, One Policy’, which can improve the water quality of the river reaches in their charge. In the RCS, the scheme of ‘One River, One Policy’ is an important basis for river chiefs at all levels to perform their duties, an important starting point for river chiefs at all levels to carry out daily work and river and lake management, and a technical guide for implementing the tasks of the RCS. In all provinces of China, the scheme of ‘One River, One Policy’ should be formulated every 3 or 5 years for all rivers and river reaches in charge of river chiefs at or above the county (district) level. The river chiefs below the county (district) level, namely those at the town (township) level and village (community) level, are also basically required to draw up the scheme of ‘One River, One Policy’ for the rivers and river reaches in their charge. One plan is formed for one river or one river reach. The targets and tasks determined by the whole river are divided into each river reach by levels and sections. In a plan of the whole river, five lists of each river reach are required to be detailed, namely, list of problems, list of goals, list of tasks, list of measures, and list of responsibilities. Tributary schemes are required to be connected to the main stream scheme and reach schemes to be connected to the whole river scheme. To formulate and implement the scheme of ‘One River, One Policy’, the content of the schemes at the municipal and county (district) levels should be further refined on the basis of the provincial scheme, so as to ensure the operability, feasibility, pertinence, and practicability of the schemes at all levels.

In the distribution of responsibilities of river chiefs at the provincial, municipal, county (district), town (township), and village (community) levels, the performance of provincial river chiefs focuses on the examination and approval of system, the performance of municipal river chiefs focuses on coordination and supervision, the performance of county (district) river chiefs focuses on organizational guarantee, the performance of town (township) river chiefs focuses on control and maintenance, and the performance of village (community) river chiefs focuses on inspection and cleaning. The higher the rank of river chiefs, the more abstract their duties are. The lower the rank of river chiefs, the more specific their duties are. The bureaucratic system has basically formed the responsibility system of river pollution control, and its hierarchical operation law has also packaged the tasks of river pollution control. In fact, almost all the specific matters in the scheme of ‘One River, One Policy’ need to be completed by village (community) river chiefs, and the goal of river pollution control depends on the performance of village (community) river chiefs. Therefore, under ‘Responsibility Contracting’, village (community) river chiefs' performance of their duties can greatly improve the water quality of the river reaches in their charge. However, village (community) river chiefs, as the main persons in charge of the Party and local governments at the village (community) level, are under the same ‘Responsibility Contracting’ for the administrative affairs other than the responsibilities of village (community) river chiefs. They not only work extremely hard but are often held to account by higher authorities. Since the implementation of the RCS, most village (community) river chiefs, under the dual situation of high pressure assessment and low-efficiency incentive, have fostered the mentality of ‘fear of responsibility and dare not do’, so they are keen to achieve the goal and task of simple operation, controllable risk, and small accountability, which makes the ‘last kilometer’ of the government order of water control not smooth.

Role of other factors

First, annual river patrol times have a positive impact on water quality. River patrol means that river chiefs at all levels find problems in a timely manner and solve them or submit them to relevant departments or report them to superior river chiefs and request coordination to solve problems. The contents of river chiefs' inspection are as follows: (1) whether the riverbank cleaning is in place, whether there is floating garbage, waste, sick, and dead animals on the river surface, and whether there is obvious sludge, garbage deposition, or obstacles at the bottom of the river; (2) whether the water body of the river is smelly or not, and whether the color is abnormal (such as black, yellow, and white); (3) whether there are missing or damaged marks of the sewage outlet into the river, whether the color and smell of the wastewater discharged from the sewage outlet into the river are abnormal, whether there are obvious abnormal discharge of livestock and poultry farms and sewage treatment facilities into the estuary, and whether there are new sewage outlets into the river; (4) whether there is illegal construction encroaching on the water area, and whether there is dumping of waste soil, slag, and hazardous waste into the river; (5) whether there are illegal cage aquaculture, illegal electric fish, toxic fish, fried fish, and other damage to the water ecological environment. In many provinces in China, it is stipulated that river chiefs should strengthen their patrolling of the river reaches in their charge. Provincial river chiefs should take the lead and patrol river at least once a year, municipal river chiefs not less than once every 6 months, county (district) river chiefs not less than once per quarter, town (township) river chiefs not less than once a month, and village (community) river chiefs not less than once a week. The frequency of river patrol should be increased for the rivers with substandard water quality and more problems. Since village (community) river chiefs are required to patrol river at least 54 times a year, such frequent patrols can undoubtedly improve the water quality of the river reaches in their charge.

Second, public support has a positive impact on water quality. The public support village (community) river chiefs in their work controls river pollution, which is embodied in participating in daily river patrol, participating in river water quality testing, reporting river pollution problems, and so on. This obviously helps to improve the water quality of the river reaches in charge of village (community) river chiefs.

Third, communicating with the public and training for work both have no impact on water quality. For village (community) river chiefs, communicating with the public on river pollution control has a direct impact on mastering the control situation, and having participated in the relevant training of river pollution control has a direct impact on improving the control ability, but neither of them has a direct impact on the water quality of the river reaches in their charge.

Based on whether the ‘One River, One Policy’ is implemented or not, we used the questionnaire data of village (community) river chiefs in nine districts and counties under the jurisdiction of Jinhua City, Zhejiang Province and used the DID Model to empirically analyze the effect of the RCS on river pollution control under ‘Responsibility Contracting’. Our findings are as follows: (1) village (community) river chiefs implement ‘One River, One Policy’, which has a positive effect on the water quality of the river reaches in their charge; (2) annual river patrol times and public support both have positive effects on the water quality of the river reaches in charge of village (community) river chiefs; (3) communicating with the public and training for work both have no effect on the water quality of the river reaches in charge of village (community) river chiefs.

Although the RCS can promote the treatment of river pollution under ‘Responsibility Contracting’, for primary river chiefs, there is still a suspicion of whitewash pollution control behavior, so whether to establish a long-term mechanism of the RCS is still uncertain. To perfect the existing RCS, China should not only construct an effective responsibility mechanism, coordination mechanism, and accountability mechanism but also dissolve the characteristics of the power system in an orderly way in an orderly way, appropriately pursue the construction of the legal system, and gradually strengthen the construction of the moral system. In particular, the emphasis should be placed on increasing the guidance and transmission of benefits, so as to mobilize the enthusiasm of primary river chiefs to perform their duties and stimulate the endogenous driving force of river pollution control, for example, to link the work performance of river chiefs with salary and promotion, to select ‘excellent river chief’ according to the assessment results, to set up special funds to reward, and to improve the financial system to reimburse or subsidize the expenses related to the work of river chiefs.

The paper was funded by The Soft Science Research Base on River and Lake Chief System of the Zhejiang University of Water Resources and Electric Power (grant no. xrj2022008).

I confirm that all the research meets ethical guidelines and adheres to the legal requirements of the study country.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.

Jiajun
G.
(
2019
).
The sustainable development path of river chief system — an analysis based on Smith's Policy-Implementation-Process model
.
Humanities & Social Sciences Journal of Hainan University
37
(
3
),
39
48
.
Jianguo
Z.
&
Ye
X.
(
2017
).
‘The River Chief System’: how is continuous innovation possible? A two-dimension analysis on the basis of both policy text and reform practice
.
Jiangsu Social Sciences
293
(
4
),
38
47
.
Kunrong
S.
&
Gang
J.
(
2018
).
The policy effects of local governments' environmental governance in China — a study based on the evolution of the ‘River-Director’ system
.
Social Sciences in China
269
(
5
),
92
115, 206
.
Lin
Y.
,
Hong
Y.
&
Xiao
Z.
(
2022
).
Effect evaluation and system optimization of River Chief water pollution control: evidence from the Yellow River basin
.
Research of Institutional Economics
20
(
1
),
97
119
.
Manhong
S.
(
2018
).
Analysis on the River Chief System from the view of institutional economics
.
China Population, Resources and Environment
28
(
1
),
134
139
.
Min
R.
(
2015
).
‘River Chief System’: a sample study of inter-departmental collaboration in watershed governance in China
.
Journal of Beijing Administration Institute
3
,
25
31
.
Shuming
W.
&
Mengmeng
C.
(
2011
).
Critique of the system of river-leader based on the perspective of new institutional economics
.
China Population, Resources and Environment
21
(
9
),
8
13
.
Wanhua
H.
,
Tingting
W.
&
Honggui
G.
(
2022
).
Transforming of River Chief System from district unit to basin unit: a quasi-natural experiment based on the policy of River Chief System in the Yellow River basin
.
Journal of Dongbei University of Finance and Economics
6
,
74
84
.
Xinfu
C.
&
Jianguo
Z.
(
2022
).
A study on manifestations, causes and correction paths of the deviation in implementing the policy of River Chief System in Chinese local governments
.
Contemporary Economic Management
44
(
3
),
33
40
.
Ye
X.
(
2017
).
On cross-domain environmental governance — take ‘River-Chief System’ as a sample
.
Social Sciences of Beijing
5
,
108
116
.
Yiwen
L.
,
Jing
W.
&
Hongyu
C.
(
2022
).
Research on the ecological environment governance effect of the ‘River Chief System’ in the Xiangjiang River
.
Soft Science
36
(
3
),
75
82
.
Yunyan
Z.
(
2019
).
River Chief System: advantage and disadvantage, argument and improvement
.
Chinese Journal of Environmental Management
11
(
4
),
93
98
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).