The chosen water management mechanism will directly or indirectly determine the success or failure of the South-to-North Water Transfer Project. Due to the importance and complexity of the project and based on national conditions and the water situation, it is necessary to adopt a quasi-market mechanism. Using the advanced experience in China and abroad for reference, this study takes the central route of the project as an example and conducts experiments by undergraduates in the laboratory. Under the simplified experimental environment, three basic rules of a quasi-market mechanism are contained and integrated: the combination of governmental macro-control with a market mechanism; the combination of water-supply management with water-demand management; and the combination of routine management with emergency management. By observing the decision-making behaviour of the experimental participants, researchers constantly changed experimental conditions to simulate the real water market realistically; finally, researchers drew conclusions based on the repeated experiments.

Introduction

To solve the problems of an uneven temporal-spatial distribution of water resources and the periodical fluctuation of water supply, more than 20 large inter-basin water transfer projects have been built in China, including the Luan River to Tianjin Water Transfer Project, the Yellow River to Qingdao Water Transfer Project, the East-to-West Water Transfer Project, the Eastern River to Shenzhen Water Transfer Project, etc. Seen from a general survey of these projects, low allocation efficiency of water resources often occurs due to outdated management following a planned economic system. From construction to operation and management, these established water transfer projects are arranged, controlled and managed by central government and local governments. Simply because the governments take on all things, unclear definition of water rights between governments at all levels and local water users, and too low water tariff without indicating the real cost of water transfer, leads to the dilemma that the more water that is transferred, the more water is wasted. All these failed experiences indicate that the existing pattern of operating and managing water transfer projects cannot comply with the new form of market mechanisms. International experience in water resources management indicates that more attention should be paid to changing water-supply management to water-demand management, or combining them both.

The South-to-North Water Transfer Project (SNWTP) is a super large infrastructure project to reallocate water resources in China. Because there is much less rainfall in northern China and rivers there dry up day by day, this project is planned, designed and constructed to divert water from the Yangtze River to the Yellow River, the Huai River and the Hai River, which passes through four river basins from a humid region to a semi-arid zone, and this project includes a western route, a central route and an eastern route. By the year 2050, the total water transferred will be 45 billion cubic meters and it will greatly promote economic development and improve the eco-environment in North and North-West China.

The SNWTP is the largest water transfer project in the world, and its oncoming difficulty of operation and management is unprecedented. According to the timeline of the project, the eastern route of the SNWTP was put into operation by the end of 2013 and the central route of the SNWTP completed after the flood season in 2014. We lack similar managerial experience in operating such a large-scale water project, and we also cannot import foreign experience completely into China. In early 2002, the basic idea of managing the project was determined, including a combination of national macro-control, quasi-market operation and user participation, and the project takes non-profit as its orientation. Corresponding to this idea, a clear definition of water rights was determined between the central government and local governments according to each investment share in the project; the local governments include four users (Beijing City, Tianjin City, Hebei Province and Henan Province). Based on such good market conditions, the largest water market in the world will be established. Today, the final and the most important step of the project is urgently to plan, design and perfect this quasi-market mechanism, but some issues are still unresolved. The main problems for water resources allocation of the SNWTP boil down to two issues.

It is unclear how the government will implement macro-control measures in the process of water resources allocation

Water is both the source of life and the base of the ecosystem. A water transfer project not only distributes water resources but also affects water economy, water politics and water security. ‘The Planning of the South-to-North Water Transfer Project’ (2002) focused on clarifying the relationship between state and local governments and indicated the allocation scheme of water resources, general rules for resolving conflicts, necessary measures for water-quality protection, the policy of water pricing, etc. Twelve years later, the ‘Water Management Regulations for the South-to-North Water Transfer Project’ (2014) shows clearly that a system of annual water transfer will be implemented. However, there are still no operable and detailed rules and regulations on water resources allocation, how to put annual water transfers into practice, how to coordinate the relationship between the waterhead area and the water intake area, how to ensure governmental macro-control measures, or how to coordinate the interests between the central government and local governments, and how to coordinate the interests among four local governments, and so on. Taking the central route of the SNWTP as an example (see Figure 1), it has a total diversion length of 1,267 km and will divert 13–14 billion cubic meters of flow from the Danjiangkou reservoir to Beijing, and this route connects Beijing City, Tianjin City, Hebei Province and Henan Province with more than seventy outlets. Because the central route links the Yangtze River, the Yellow River, the Huai River and the Hai River, climatic factors and anthropogenic factors will make the allocation of water resources more complex. On the occasion that the northern water intake area and the southern waterhead area confront common wet seasons or dry seasons, or water demand in the water intake areas increases while water supply in the waterhead area decreases, it is unclear how to combine the governmental macro-control measures with water market mechanism, especially when an imbalance of the supply-demand relationship or even an emergency event (extreme flood or drought) occurs. This problem will be a challenge and it will draw the attention of the world.

Fig. 1.

The central route of the South-to-North Water Transfer Project.

Fig. 1.

The central route of the South-to-North Water Transfer Project.

It is unclear how to bring the quasi-market mechanism into practice

China is a country with a long history of planned economic systems, and natural resources allocation generally depends on a mandatory plan, which is effective in times of abundant resources. Nowadays, a traditional mandatory plan becomes more and more invalid due to dwindling non-renewable resources. Water resources are no exception; water-supply management for water transfer projects is insufficient and water-demand management should be supplemented. As for the SNWTP, on one hand, it is even more complicated than any other projects at home and abroad. At times of any contradiction between profit and nonprofit during the project operation, the nonprofit orientation of the project prevailed, so we should always adopt both governmental macro-control and market mechanisms in order to put the project into benign operation, and only in this way can the project ensure rational allocation of water resources.

On the other hand, the SNWTP cannot be entirely operated in accordance with market mechanisms and nor can it absolutely pursue benefit maximization, because water resources allocation relates to the national economy, people's livelihood, regional sustainable development and social stability. As a result, a quasi-market mechanism was put forward at the start of the project in 2002. The eastern route of the project entered the testing stage in 2013 and some problems have occurred: because of the high price of the water that has been transferred, the demand for water reduces and there has been a lack of integration of the transferred water with local water sources in the water intake area. As a result of such problems the poor outcomes seen in existing transfer projects occur again.

About the quasi-market mechanism

From the perspective of international experience, developed countries with developed market mechanisms such as the USA continuously explore institutional reform of water management based on sound water rights principles. One successful model is the mechanism of the water bank in the North-to-South Water Transfer Project in California, USA. This innovation overcomes the shortcomings of single water-supply management. When a contradiction between water supply and water demand occurs due to the periodical fluctuation of hydrological and climatic factors, a rigid and fixed mode of water resources allocation can not balance the tension in the water intake area. However, through a management platform based on a water bank and with auxiliary benefit-regulation measures (called water-demand management, and including stimulating the potential supply-demand relationship and facilitating exchanges from areas with a water surplus to water deficient areas by consultation and negotiation), a flexible and rational reallocation mode of water resources was born. Moreover, Vernon Smith (the founder of experimental economics) and his research group tested a smart water market institution using laboratory experiments based on empirical data from a water bank (Murphy et al., 2000). This smart water market institution was computer-assisted designed by Vernon Smith (McCabe et al., 1991) and is close to the quasi-market mechanism proposed in the planning of the South-to-North Water Transfer Project in China (Zhang, 2002). In spite of different national institutions, advanced management experience is worth learning.

Because the institution of a planned economy has lasted for a long time in China, the allocation of water resources has traditionally followed the mandatory-planned management mode. With the contraction of the water supply and with water demand becoming more and more serious, the concept of a quasi-market mechanism was put forward in 2000 (Hu & Wang, 2000), and then the former Minister of Water Resources (Wang, 2000) further enriched it: ‘water is a scarce and valuable resource, so water rights are valuable, transferable and tradeable’. To date, both the government and scholars have recognized the importance of combining governmental macro-control with market mechanisms. Many scholars have put forward constructive ideas on the water market from different perspectives: Luo et al. (2006) established a transaction model of the water market and analyzed the market equilibrium; Li et al. (2006) established a game model on incomplete information of water rights transaction based on the auction mechanism; Zhang (2006) established a multi-objective model of optimal allocation of water resources for the eastern and central routes of the SNWTP; Dong et al. (2007) proposed a quasi-market model of water resources allocation for the central route; Zhang (2002) and Zhang & Lv (2007, 2008) proffered a transaction mode for the SNWTP inspired by the water bank of the North-to-South Water Transfer Project in California, USA; Wang et al. (2008) proposed an option mode for the northern route of the SNWTP. Today, it is not certain which mode of the quasi-market mechanism is more appropriate, or how to integrate these ideas into the quasi-market mechanism and operate it in practice.

Considering the strong uncertainty of water supply for the central route of the SNWTP, once an imbalance of water supply and water demand occurs, reducing the available water supply to all users in the water intake area by the same percentage is not a rational configuration. Therefore, we should improve water-supply management and make more accurate predictions of the water supply, and we also should implement water-demand management to reallocate limited water resources to efficient areas and efficient departments. From the above, three basic rules should be contained and integrated within the quasi-market mechanism of the SNWTP. These are: the combination of governmental macro-control with a market mechanism (the core of the quasi-market mechanism); the combination of water-supply management with water-demand management; and the combination of routine management with emergency management. This cognitive premise establishes the theoretical foundation for the subsequent laboratory experiments of the quasi-market mechanism.

Experimental research

Generally speaking, a momentous institutional reform progresses slowly and generates huge social influence; however, the reform effect can only be evaluated after its implementation and risks within the institutional design cannot be reduced in advance. Similarly, the quasi-market mechanism of the SNWTP is no exception, and we are not sure of the operability of the quasi-market mechanism in practice in advance.

At present, from the world's point of view, experimental economic research has become a necessary link between policy-making and institutional innovation, and research on market mechanism design is especially popular. By simulating market transactions, changing the rules of market exchange and observing decision-makers' behaviour, researchers improve market exchange rules and decrease the risk of a poor institutional design in an economical way (Smith, 1982, 1989, 2002; Brown et al., 2004). Experimental economics research has been widely used in the field of market mechanism design and policy-making including resources (energy) management and finance policy reform (Weiss, 1997; Capra et al., 2002; Cason & Friedman, 2003; Morgan et al., 2006; Ding & Li, 2007; Jiang, 2008; Wu, 2008). Taking the smart experimental research (Murphy et al., 2000) for reference, this paper considers the central route of the SNWTP as an example and examines the feasibility of the quasi-market mechanism so as to mitigate the risk of a poor water market institution design for the SNWTP in advance and in an economical way.

Experimental background

As a long route and inter-basin water transfer project, there are various water-supply and water-demand combinations in the SNWTP resulting from periodical hydrological variation and non-periodic social requirement from both the waterhead area and the water intake area, so water resources allocation of the central route is even more complicated and difficult. No matter whether there is sufficient or insufficient water supply in the waterhead area via the central route of the SNWTP, a local or whole imbalance of the supply-demand relationship in the water intake area may occasionally occur. Through the platform of water reallocation and by way of negotiation with both the water deficiency and water surplus sides, the goal of allocating water resources efficiently and neatly will be achieved.

In this experimental research, the application premise of the quasi-market mechanism is expressed as Figure 2.

Fig. 2.

Application premise of the quasi-market mechanism: a case for the central route of the SNWTP.

Fig. 2.

Application premise of the quasi-market mechanism: a case for the central route of the SNWTP.

Experimental principle

A market is an arena in which commercial dealings are conducted. The basic role of a market mechanism is to stimulate and guide market participants to allocate resources effectively and optimally. Unlike in previous experiments by computer simulation, this research uses ‘real people to conduct experiments’. Participants who succeed in the experiments get a reward, which contributes to simulating real market behaviour as far as possible. Thus, to some extent, the effect of institutional design can be observed in a laboratory before it is put into practice.

As shown in Figure 3, the aim of the experimental study was to find out how the quasi-market mechanism comes true under the simplified economic environment of the central route of the SNWTP. Two kinds of objects were studied: experimental observation objects included participants' decision-making and trade results, observed and analyzed by researchers in each experiment; the experimental control objects included the economic environment, market incentives (reward), distribution laws and market language (nodes of water sellers and water buyers, expected water price from water sellers or water buyers, pre-sale volume of water sellers, pre-order volume of water buyers, assignment of contract, etc.). The experimental control objects could be changed repeatedly, and researchers observed the process of participants' decision-making, evaluating trade results and further analyzing factors influencing participants' behaviour, contributing to improving the design of the quasi-market mechanism in the later experiments.

Fig. 3.

Experimental principle of the quasi-market mechanism of the SNWTP (adapted from Du (2008)).

Fig. 3.

Experimental principle of the quasi-market mechanism of the SNWTP (adapted from Du (2008)).

The goal of the SNWTP is to ensure water resources allocation is efficient and consistent with the governmental intention; accordingly, the goal of the laboratory experiments was to pursue overall maximized or satisfactory benefit of market participants in line with the target of governmental macro-control. The allocation of water resources in the experiments was based on an experimental management centre (which acted as the function of governmental control; all governmental control measures were operated through this platform). Water rights transfer of the SNWTP was simulated through the platform, which was responsible for information collection, processing and issuance, and also worked as the market clearing centre and control centre by reference to the rules of bargain, auction and bidding (Chatterjee & Samuelson, 1983; McAfee & McMillan, 1987; Nicolaisen et al., 2001). We adopted a Bayesian model based on double auction (Chatterjee & Samuelson, 1983; Li et al., 2006) as the distribution laws for the market mechanism in the experimental research. The water market consisted of multiple buyers and multiple sellers, and multiple-objective optimization models were presented as follows: 
formula
1
Subject to 
formula
2
 
formula
3

Equation (1) is the objective function; it represents the maximized (or satisfactory) total benefit of all market participants in line with the target of governmental macro-control. Equations (2) and (3) are individual rationality constraints for a water buyer and water seller, respectively.

In Equations (1)–(3), bj represents the bid price per unit of water rights as a water buyer (j = 1,2,…N) and Vj represents valuation per unit of water rights as a water buyer; here we suppose j and Vj belong to independent and identically distributed variables, both of them submitting to distribution G(·) and density g(·) among [0, δ]. The bidding strategy of water buyer j is expressed as bj(Vj), a water buyer's benefit is expressed as ; similarly, si represents the offer price per unit of water rights as a water seller (i = 1,2,…M), , Ci represents valuation per unit of water rights as a water seller, and we also suppose i and Ci belong to independent and identically distributed variables, both of them submitting to distribution F(·) and density f(·) among [0, ω]. The bidding strategy of water seller i is expressed as si(Ci); a water seller's benefit is expressed as .

As for the transaction rules, we adopted high and low matching principles (Nicolaisen et al., 2001) in the experiments, whilst for the clearing rules (Li et al., 2006) we determined the clearing price P according to Equation (4): 
formula
4

where and represent the lowest bid price of water buyer bj and the highest offer price of water seller si, respectively.

Experimental design and process: a case study

Each laboratory experiment of the quasi-market mechanism of the SNWTP included experimental preparation and experimental process. Preparation for the experimental design consisted of the experimental principle, design for experimental control conditions, technical preparation and participants' preparation. The experimental process was organized and controlled by the experimental manager, who performed the function of managing a transaction platform and facilitating water rights transfer between water supply and water demand, so as to achieve rational allocation of water resources (as in Figure 4).

Fig. 4.

Experimental design and process of the quasi-market of the SNWTP.

Fig. 4.

Experimental design and process of the quasi-market of the SNWTP.

The experimental participants consisted of undergraduates from the School of Geographic Science at Northeast Normal University, who played the roles of water suppliers and water users of the central route of the SNWTP. These roles included water buyers and water sellers, and some participants played the double roles of water users and suppliers (who can transfer surplus water to other water users needing more water, according to different experimental conditions).

The experimental process was operated as follows:

  • First, before an experiment began, all participants read the experimental instructions in which the researchers had made a detailed description of the market mechanism and the rules of market exchange.

  • Then, under the initial condition set up by the researchers, each participant submitted their supply or demand information to the experimental management centre at the specified time intervals, whilst the water-supply information of each water seller included the node which they were in, how much water they wished to sell and at what price they wished to sell it. Similarly, each experimental participant playing the role of a water buyer also submitted their information: the node which they were in, how much water they wished to buy and what price they wished to offer.

  • Next, the experimental management centre was in charge of information reception, optimization management and necessary consultations with water buyers and water sellers to make the experiment go smoothly.

  • Finally, researchers drew conclusions by observing the decision-makers' behaviour and analyzing the experimental results, no matter whether an experiment succeeded or not. Each experimental participant who succeeded in a transaction received a corresponding monetary reward, which contributed to ensuring the authenticity of their transaction behaviour, to some extent.

It should be noted that two assumptions frequently occur for the central route of the SNWTP: reliable water supply decreases due to climatic change and hydrological fluctuation; and water demand changes due to the adjustment of the industrial structure or water saving in the water intake area. Should either of these two conditions happen, it is possible that some water users would not have enough water whilst, meanwhile, some water sellers (or some water users with adequate water) have a surplus. Under these circumstances, without the stimulus of a price leverage and information communication mechanism between the water intake area and the waterhead area, a dynamic and flexible reallocation of water resources cannot be spontaneously achieved, especially for the central route of the SNWTP.

Based on the above understanding, the experimental background assumed that an unbalanced water supply-demand contradiction occurred in the whole water intake area and the previous water-supply management could not handle this situation. In addition, transport links were taken to be zero cost in the whole experimental study because water transferred from the waterhead area could flow by itself along the whole central route of the SNWTP.

Water buyers and water sellers consisted of twelve undergraduates as participants to complete the work, and each participant was assigned to one node (outlet) in the water intake area. According to each initial condition assumed by researchers, every participant established their bidding strategies and then submitted their supply information or demand information to the experimental management centre, which dealt with the information scheduling, collecting, sorting and two-way adjustments to buyers and sellers. The data about water supply-demand submitted to the experimental management centre are shown in Tables 1 and 2. From the perspective of cost and benefit, water buyers wanted to buy water at a low price as far as possible and water sellers wanted to sell water at a high price as far as possible. Laboratory experiments in this study were no exception and, in this case study, asymmetric information, lack of communication and participants' overconfidence led to failed trades. In fact, these kind of trade results occurred often in the experimental research.

Table 1.

Initial buyer bids.

Buyer Node location Bid price (yuan/m3Volume of water demand (m3
B11 Tianjin City 3.9 
B12 Beijing City 3.8 
B13 Xingtai City 3.5 
B14 Shijiazhuang City 3.5 
B15 Handan City 2.9 
B16 Baoding City 2.6 
Buyer Node location Bid price (yuan/m3Volume of water demand (m3
B11 Tianjin City 3.9 
B12 Beijing City 3.8 
B13 Xingtai City 3.5 
B14 Shijiazhuang City 3.5 
B15 Handan City 2.9 
B16 Baoding City 2.6 
Table 2.

Initial seller bids.

Seller Node location Offer price (yuan/m3Volume of water supply (m3
S11 Zhengzhou City 3.8 
S12 Hebi City 3.9 
S13 Xuchang City 4.1 
S14 Xinxiang City 4.4 
S15 Anyang City 4.5 
S16 Nanyang City 4.8 
Seller Node location Offer price (yuan/m3Volume of water supply (m3
S11 Zhengzhou City 3.8 
S12 Hebi City 3.9 
S13 Xuchang City 4.1 
S14 Xinxiang City 4.4 
S15 Anyang City 4.5 
S16 Nanyang City 4.8 

As for this result, the manager in the experimental centre negotiated with all water buyers and water sellers, informed them of risks and of the possibility of transaction failure, and shared supply-demand information with them, and then each participant readjusted their bid strategy respectively after careful thinking, and re-submitted an application to the experimental management centre several times, until a trade result was satisfactory. The data about water supply-demand submitted to the experimental management centre are shown in Tables 3 and 4. Based on the market clearing rule following Equation (4), the market clearing price of water transfer was determined at CNY ¥3.6/m3; the buyers (including B21, B22, B23 and B24) and the sellers (including S21, S22, S23 and S24) made deals.

Table 3.

Final buyer bids.

Buyer Node location Bid price (yuan/m3Volume of water demand (m3
B21 Beijing City 4.2 
B22 Tianjin City 3.9 
B23 Xingtai City 3.7 
B24 Shijiazhuang City 3.6 
B25 Handan City 3.5 
B26 Baoding City 3.3 
Buyer Node location Bid price (yuan/m3Volume of water demand (m3
B21 Beijing City 4.2 
B22 Tianjin City 3.9 
B23 Xingtai City 3.7 
B24 Shijiazhuang City 3.6 
B25 Handan City 3.5 
B26 Baoding City 3.3 
Table 4.

Final seller bids.

Seller Node location Offer price (yuan/m3Volume of water supply (m3
S21 Nanyang City 2.9 
S22 Xinxiang City 3.2 
S23 Xuchang City 3.5 
S24 Hebi City 3.6 
S25 Zhengzhou City 3.8 
S26 Anyang City 3.9 
Seller Node location Offer price (yuan/m3Volume of water supply (m3
S21 Nanyang City 2.9 
S22 Xinxiang City 3.2 
S23 Xuchang City 3.5 
S24 Hebi City 3.6 
S25 Zhengzhou City 3.8 
S26 Anyang City 3.9 

According to our observation and analysis of the participants' decision-making, we found that under the material incentive, the experimental participants tried to make their decisions on the basis of benefit pursuit and rational judgement, on the one hand, although each participant wanted to pursue benefit maximization based on their initial condition so as to get more reward; on the other hand, rational judgement led them to a reasonable bid strategy, because everyone knew that they would get nothing from a failed trade. In this way, the trade results got more and more efficient and rational, which contributed to simulating the water market more realistically, and those who succeeded in water rights transactions received their reward at the end of the experiment.

Discussion and conclusion

As the largest and the most complex water transfer project in the world, the SNWTP, especially the central route, is difficult in operation and management. With regard to this topic, it is clear that a quasi-market mechanism will be put into practice, but it is unclear how it will come into play and how the government will implement macro-control measures for water resources allocation. Thus it is essential to test the quasi-market mechanism in advance.

Above all, in the case of this experiment, we give the quasi-market mechanism three basic rules: a combination of governmental macro-control with market mechanism (the core rule); a combination of water-supply management with water-demand management; and a combination of routine management with emergency management. The goal of the experimental research was to test how these rules are carried out by the participants. In order to get straight to the point, the premise of the experimental research assumed that traditional water-supply management for the SNWTP is not enough to reallocate the limited transferred water, so it is urgent to supplement water-demand management; furthermore, integrating governmental macro-control with the market mechanism is dominant, and the third rule (the combination of routine management with emergency management) is regarded as being contained in the first two rules in the experiments. Based on this, we placed emphasis on observing how the quasi-market mechanism operates and what influences participants to make their decision in laboratory experiments.

In the process of each experiment, under a simplified economic environment and based on the designed initial condition, each participant (acting as a water buyer or a water seller) performed their own role inspired by reward, and then submitted their bid strategy until a trade result came out. All experimental control objects can be changed and repeated. Insights and knowledge gained from both successful and unsuccessful trade results and from observing participants' behaviour in decision-making included the following three points:

  • (1) Water resources allocation (or reallocation) may be achieved based on a quasi-market mechanism by means of facilitating potential water suppliers and water users. From the perspective of cost and benefit, benefit pursuit is the original motivation of each market participant, so a market mechanism is necessary to improve the allocation efficiency of water resources. Especially under the advantageous condition of clear definition of water rights in the SNWTP, all three basic rules about the quasi-market mechanism are made to serve this new condition; however, if maximized benefit or satisfactory benefit does not coincide with the goal of governmental macro-control, benefit pursuit alone is not enough and this trade result cannot be accepted.

    Seen from the simplified economic environment, the three basic rules prove to be appropriate but they need detailing, and we cannot give a clear answer on how to judge exactly the ‘degree of satisfaction’ about the quasi-market mechanism. From the perspective of the theoretical study, the objective function in the experimental research is the total maximized benefit of all market participants in line with the target of governmental macro-control, but from the perspective of the operational level, the experimental management centre has the power to accept a result called the ‘satisfactory total benefit’ of all participants in the experimental research. Two resulting problems emerge: whether we must insist on the maximized total benefit of all market participants; and what the criterion is for judging the satisfactory total benefit of all market participants in practice. Decisions can be made by the experimental manager in laboratory experiments, but what will happen in real practice, where one result could perhaps relate to four users with more than seventy outlets of water intake area? In our opinion, if a trade result conforms to a state of ‘multiple aspects win’ or ‘multi-party benefit’ in line with the target of governmental macro-control, this kind of result can be regarded as an acceptable scope and we need not make excessive demands on whether the benefit is maximized or not. Of course, this decision-making power should be performed by the administrative department of the SNWTP, which plays the role of the governmental macro-control.

  • (2) As seen from some unsuccessful trade results, unsolved problems still existed, the volume of water supply and water demand did not match, or consultation with buyers and sellers failed, and trade results were not satisfactory, etc. All unsuccessful cases show the difficulty and importance of combining governmental control with market mechanism, and the role of both bridge and link played by the experimental management centre is very crucial. Consultation and negotiation are so important that they sometimes even determine the trade result, especially when water buyers or water sellers stick to their own views and the consultation becomes deadlocked. Thus, timely publishing of information, detailed negotiatory rules and adequate negotiatory skills are very important to avoid the dilemma caused by asymmetric information, lack of communication and participants' overconfidence. As we have shown, a quasi-market mechanism that combines governmental macro-control with a market mechanism is not easy to set up and cannot be perfected through only some laboratory and field experiments; water resources allocation for the SNWTP depends more on carrying forward water rights institutions in China.

  • (3) The experimental research on institution design for the SNWTP in advance and in an economical way is significative. Of course, there is a limitation to the authenticity of undergraduates who act as the participants in the quasi-market experiment for the SNWTP. If field experiments on the quasi-market can be carried out, and true water suppliers and water users are chosen as the experimental participants from both the waterhead area and the water intake area in the central route of the SNWTP, then the experimental results will be closer to true quasi-market conditions and better than the laboratory experiments. In fact, the founder of experimental economics, Vernon Smith, mostly chose graduates as participants in his experimental research, and reasonably scientific experimental results were obtained thereby, such as the valuable suggestions made for the water market design of the North-to-South Water Transfer Project in California, USA. Although the undergraduates chosen in this research had studied hydrology and economics, it was still hard to avoid some error in the role-playing as water suppliers and water users, but market incentives (monetary rewards) in the experiments may make up for this, and the authenticity and scientific validity of this research from experimental design to experimental process have been ensured as far as possible, and hence, we trust this research will provide a reference for establishing a quasi-market mechanism for the SNWTP.

Acknowledgements

This study was supported by the National Planning Office of Philosophy and Social Science (10YJA790257) and the National Natural Science Foundation of China (41271555). We thank them and also the undergraduates who were the participants in the laboratory experiments in this research.

References

References
Brown
M.
Falk
A.
Fehr
E.
(
2004
).
Relational contracts and the nature of market interactions
.
Econometrica
72
,
747
780
.
Capra
M.
Gomez
R.
Holt
C.
Goeree
J.
(
2002
).
Learning and noisy equilibrium behavior in an experimental study of imperfect price competition
.
International Economic Review
43
,
613
636
.
Cason
T.
Friedman
D.
(
2003
).
Buyer search and price dispersion: a laboratory study
.
Journal of Economic Theory
112
,
232
260
.
Chatterjee
K.
Samuelson
W.
(
1983
).
Bargain under incomplete information
.
Operation Research
31
,
835
851
.
Ding
W.
Li
M.
(
2007
).
Research on experimental economics and its application in electricity market
.
Modern Electric Power
24
(
1
),
88
92
.
Dong
Y.
Jiang
Y.
Wang
H.
Yin
J.
(
2007
).
Market model of water distribution of the South-to-North Water Transfer Project
.
South-to-North Water Transfers and Water Science & Technology
5
(
1
),
10
12
.
Du
N.
(
2008
).
Experimental Economics
.
Shanghai University of Finance and Economics Press
,
Shanghai
.
Hu
A.
Wang
Y.
(
2000
).
Transformation of the allocation of water of public policy: quasi-market and political democracy consultation
.
China Soft Science
5
,
5
11
.
Jiang
Y.
(
2008
).
Power market trade mechanism based on experimental economics
.
Master Thesis, North China Electric Power University, Beijing
.
Li
C.
Wang
X.
Fan
W.
Zheng
X.
(
2006
).
Double auction Bayesian model and design of mechanism in water rights market
.
Resources and Environment of the Yangtze River Basin
4
,
465
469
.
Luo
H.
Li
L.
Wang
M.
Liu
G.
(
2006
).
Quasi-market design and its market equilibrium of tradable water rights
.
Journal of Hydraulic Engineering
4
,
492
498
.
McAfee
P.
McMillan
J.
(
1987
).
Auction and bidding
.
Journal of Economic Literature
25
(
8
),
699
738
.
McCabe
K. A.
Rassenti
S. J.
Smith
V. L.
(
1991
).
Smart computer-assisted markets
.
Science
254
,
534
538
.
Morgan
J.
Orzen
H.
Sefon
M.
(
2006
).
An experimental study of price dispersion
.
Games and Economic Behavior
54
,
134
158
.
Murphy
J. J.
Dinar
A.
Howitt
R. E.
Rassenti
S. J.
Smith
V. L.
(
2000
).
The design of smart water market institutions using laboratory experiments
.
Environmental and Resource Economics
17
(
4
),
375
394
.
Nicolaisen
J.
Petrov
V.
Tesfatsion
L.
(
2001
).
Market power and efficiency in a computational electricity market with discriminatory double auction pricing
.
IEEE Trans on Evolutionary Computation
5
(
5
),
504
523
.
Smith
V. L.
(
1982
).
Microecnomic system as an experimental science
.
American Economic Review
72
(
5
),
923
955
.
Smith
V. L.
(
1989
).
Theory, experiment and economics
.
Journal of Economic Perspective
3
(
1
),
151
169
.
Smith
V. L.
(
2002
).
Method in experiment: rhetoric and reality
.
Experimental Economics
5
,
91
110
.
Wang
S. C.
(
2000
).
Water rights and water market: economic methods of realizing the optimal allocation of water resources
.
China Water Resources
11
,
458
465
.
Wang
H. M.
Wang
H.
Qiu
L.
Tong
J.
(
2008
).
Study on the option contract in water allocation of the eastern route of South-to-North Water Transfer Project
.
China Population Resources and Environment
18
(
2
),
44
48
.
Weiss
J.
(
1997
).
Market Power Issues in the Restructuring of the Electricity Industry: an Experimental Investigation
.
Harvard Business School, Harvard University
,
Cambridge, MA
.
Wu
M.
(
2008
).
Experimental Economics Studies on the Effectiveness of the Emission Trading Mechanism
.
Shanghai Jiao Tong University
,
Shanghai, MA
.
Zhang
G.
(
2002
).
Some reviews about water transfer scale for the South-to-North Water Transfer Project
.
China Water Resources
2002
(
10
),
120
123
.
Zhang
Y.
(
2002
).
Construction of water market of the South-to-North Water Transfer Project
.
Water Resources Development Research
3
,
4
7
.
Zhang
P.
(
2006
).
Optimization research on water resources of the South-to-North Water Transfer Project
.
PhD Thesis, Hehai University, Nanjing
.
Zhang
Y.
Lv
D.
(
2007
).
Research on combining water-supply management with water-demand management of the South-to-North Water Transfer Project
.
South-to-North Water Transfers and Water Science & Technology
5
(
3
),
3
5
.
Zhang
Y.
Lv
D.
(
2008
).
Design for water contract of water bank in the central route of South-to-North Water Transfer Project
.
South-to-North Water Transfers and Water Science & Technology
6
(
3
),
12
18
.