How do markets manage water resources? An experiment

We test how a monopoly, a duopoly and a public monopoly manage and allocate water resources. Stock depletion for the public monopoly is fastest. However, it reaches the optimal stock level towards the end of the experimental sessions. The private monopoly and duopoly maintain inefficiently high levels of stock throughout the sessions. The average quality to price ratio offered by the public monopoly is substantially higher than that offered by the private monopoly or duopoly. A clear result from the experiments is that a public monopoly offers the highest (average) quality to price ratio and has the fastest rate of stock depletion compared to a private monopoly or duopoly. 1 Corresponding author: Aurora García Gallego, Economics Department, Universitat Jaume I, Avda. Sos Baynat s/n, Campus Riu Sec, 12006-Castellón (Spain), Tf. +34964387161, Fax. +34964728591, e-mail: mgarcia@eco.uji.es. Financial support by the Fundación BBVA is gratefully acknowledged. All errors are the authors’ responsibility and opinions expressed hereby are not necessarily shared by the financing institution.


Introduction
The importance of efficient water allocation is being increasingly realised all over the world. Increasing demand on water resources has been one of the main reasons why water shortage is becoming an important issue. For example, recently, The Economist drew attention towards the impending water crisis in China 2 . It is being increasingly argued that one of the main reasons behind the mismanagement of water resources is the non-existence of economic criteria in water allocation. Till now, water has been allocated by government agencies motivated mostly through non-economic criteria.
Agencies such as the World Bank increasingly argue that efficient allocation of water is only possible through market based mechanisms and state examples of functioning water markets in several developing and developed nations.
The problem is difficult given that historically water has been treated as a "social" rather than an economic good. It is well accepted by policy makers that allocating water within the existing systems is inefficient. A need for a market based mechanism to allocate water efficiently is widely gaining acceptance. In its 1993 policy paper (Water Resources Management) the World Bank states that the deterioration and scarcity of fresh water in recent times is due to the "failure to properly consider the economic value of water. Given that water is given little or no economic value it is misallocated and misused." The argument behind this is simple, i.e. understanding the economic value of water leads to its efficient use.
Water is a growing industry worldwide. Recent estimates put the world water market at $300 billion, with the United States accounting for more than half that amount. Two of the fastest growing markets are in water rights and municipal water supply systems. One should note that the development of water markets leads to a better definition of property rights. This by itself eliminates many of the problems related with inefficient allocation of water. The enforcement of property rights becomes difficult if incentives are not properly defined.
Proper definition of property rights and the introduction of markets can alone not solve the problem. Another important issue is the overexploitation of groundwater. For example, it has been shown that competitive water withdrawal can lead to overexploitation (Moench, 1992). Over exploitation is a serious problem in the case of allowing open access to water. Penalties for over exploitation may be one answer.
Another answer may be varying the price of water as aquifer levels reach critical levels.
The price would then reflect the "real" value of water based on future scarcity 3 . Such real time pricing schemes are difficult to implement due to political reasons. Gordon (1954) showed that complete rent dissipation may occur from the exploitation of an open access resource. A single owner, internalizes exploitation externalities, and would be more efficient. The general result obtained in the literature is that there is an inverse relationship between the number of resource extractors and rent accrual. Mason and Philips (1997), for example, provide experimental evidence on the relationship between group size and the standing stock of a common resource.
Experiments have been used by Walker et al. (1990), Walker and Gardner (1992), Gardner et al. (1997) to study common pool resource problems. Walker et al. (1990) and Walker and Gardner (1992) both use experiments to study non-cooperative game common pool resources. Walker et al. (1990) conclude that a high degree of rent dissipation occurred with access limited to eight users. In Walker and Gardner (1992) they show that the common pool resource is destroyed in all cases where no institutions exist to foster cooperative behaviour. Gardner et al. (1997) study strategic behaviour in groundwater depletion within the setting of state governance of groundwater resources in West U.S. They experimentally study the performance of various groundwater property rights and the applicability of game theory in such systems. Limiting entry en such environments improves efficiency.
In a recent study, Murphy et al. (2000) use laboratory experiments to design "smart" computer assisted markets for water. The 'smart' computer assisted market institution was developed by McCabe et al. (1989McCabe et al. ( , 1991. A 'smart' market allows decentralized agents to submit messages to a computer dispatch centre. The centre then computes prices and allocations by applying an optimization algorithm that maximizes the possible gains from exchange. Using California as a case study, Murphy et. al. (2000) test alternative institutional arrangements for a computer assisted spot market. In a thin market characterized by a limited set of trading opportunities, they show that the 3 Such a policy would imply that the price of water goes up in times of water scarcity. `smart' uniform price double auction yield highly efficient outcomes. Co-tenancy seemed to improve efficiency over a monopoly in transportation.
The studies mentioned above clearly show that experiments can be a useful test bed to study alternative property right mechanisms (Walker et al., 1990;Walker and Gardner, 1992), or undertake a direct test of an existing market mechanism (Murphy et al., 2000). Experiments can replicate important characteristics of existing markets and test them in a laboratory setting at a minimal cost to the regulator.
There is another aspect to water markets that has been given little attention. This has to do with the quality of water that each consumer, farm and household, receives.
The quality of water for household use is regulated by quality standards. Quality standards for farm use are much weaker. Any firm supplying water to two different users that differ in their minimum acceptable qualities has an additional dimension to deal with. That is, what is the optimal mix of qualities that the firm should supply to its users?.
In this paper we define a water market that includes standard features such as common pool resource management, depuration costs and water quality. In our experimental market the firm has to also decide the quality that it should supply to households and to farmers. The quality of household water cannot be below a certain minimum standard while the quality of farm water has no such restriction. Our results indicate that a social monopolist offers the highest quality to price ratio and exploits the resource at a faster rate than a private monopoly or duopoly. The most stable stocks correspond to a private monopoly and duopoly. Though, their stock levels are inefficiently high. The average quality to price ratio offered by the public monopolist is the highest.
The paper is structured as follows. In Section 2 we discuss the model. In Section 3 we discuss our experimental design. Section 4 discusses the experimental results.

The model
There are two renewable stocks S H (high quality) and S L (low quality) from which water may be extracted. For the sake of simplicity, we assume that the recharge to the respective basin is deterministic and constant. The inflow to the respective basins is assumed to cease when the storage capacity of the aquifer is reached. That is, once the maximum storable stock is reached, extra water inflow is lost. The return flow of consumed water is assumed to be negligible. Thus, changes in the stocks are exclusively due to extraction and recharge. Extraction costs are supposed to be twice differentiable functions of quantity and stock size. First derivatives are assumed to be, respectively, positive and negative, whereas second derivatives are positive.
We allow for the possibility that the water resources differ in qualities. Quality of water in an aquifer may be lower due to marine intrusion, or due to infiltration of fertilizer from agriculture. Let the qualities be denoted respectively by Q H and Q L , where Q H > Q L >0. The qualities are assumed to be constant over time. However, any intermediate quality may be supplied to the consumers as a result of mixing water from the two sources. Mixing quantities K H and K L of the two qualities results in water whose quality is given by the weighted average: Quality of potable water should weakly exceed the constant minimum quality standard Q min , where Q H >Q min >Q L . Mixed water of quality Q M may, or may not, satisfy the minimum quality standard. This depends on the quantities and the qualities which are mixed. Quality may be improved at a cost. This cost is an increasing function of the difference between the quality before and after purification. Moreover, a given improvement ΔQ of a lower quality is less costly than the same improvement performed on a higher quality. Let the initial quality subject to purification be Q 0 . The purification cost, denoted by C ΔQ (K, ΔQ, Q 0 ), for a certain water quality Q 0 and quantity K=K H +K L requiring a quality improvement ΔQ, is assumed to satisfy the following conditions: Resource flow between the sources and the consumers is coordinated by a centralized knot, which centralizes the decisions about quantity and quality supplied to the consumers. Figure 1 shows the distribution scheme described above.

Inflow
High quality source Low quality source Inflow

Mixing Node
Households Farmers preferences regarding the quality of water. Both types prefer a higher quality of the water to a lower one. Farmers prefer more quantity of each product to less. Households consume water whose quality weakly exceeds a minimum standard. If mixed quality does not satisfy this condition, it will be subject to purification. The purification procedure is assumed to be costly enough such that it is not profitable to improve quality above the minimum standard. Hence, the quality consumed by households is the maximum between the minimum, and the mixed, quality. Thus, Q 0 = Q M and Let the households take the purification cost into account in their utility function.
Further, assume utility functions for the respective consumer-types, U h =U h (K h ,Q Mh ) and U F =U F (K F ,Q MF ) (where K h =K Hh +K Lh , and K F =K HF +K LF ), to be twice differentiable with respect to the quantity and the mixed quality. A farmers' utility is increasing in both arguments. While depending on the purification cost function, the utility function of households might be increasing in the quantity of low quality only up to a certain limit 4 . From twice differentiability of the utility functions it follows that the sum of the functions is twice differentiable, too. The indirect social welfare function V(K H ,K L ), which maximizes consumer surplus for a given quantity of water, can be obtained as a solution to the following problem: As a benchmark for our experimental results, we are interested in the socially optimal solution of water supply. Given the assumptions above, we formulate the program that maximizes social welfare. 5 Without loss of generality, suppose that initially the resource stocks are in the natural hydrological equilibrium, i.e. at the upper bound of the storage capacity. Let ( ) denote recharges of the two water qualities of water, and t 0 the starting time of extraction. Assume that the social rate of discount is δ=1. Thus, the intertemporal objective function is formulated as follows: By means of the resulting current value Hamiltonian and Pontryagin's maximum principle (assuming an interior solution) the two following conditions have to be satisfied in the hydro-economic equilibrium: The conditions in (5) simultaneously determine the steady-state standing-stocks of S H and S L . They basically state that, in the long-run, the marginal social utility, which embod

Experimental design
Our experimental design focuses on studying how different market structures affect water resource management. Upstream firms do not extract from a common resource 8 .
We instead focus on situations where property rights to a given groundwater source are exclusively granted to a single decision making unit. el. Our model also highlights the vertical nature of these markets. In these markets coordination is required between water d supply 9 . urchase up to a certain amount of ies the respective resource price in the economy, should equal the social costs of extraction represented on the right hand side. 6 7 Two water sources supply water of two different qualities. The demand side is represented by two different types of consumers: households and farmers. Water supplied to them may be the result of purification, since households will only consume water whose quality exceeds a minimum lev extraction, purification an Our assumptions concerning consumer utility are qualitatively similar to those in Williams et al. (1986) on multiple commodities which are interdependent in consumption. Two features, which are rather specific to the dynamics of water, are added to the structure: first, buyers are restricted to p 6 In each condition, the first two terms (both positive) represent the marginal cost which results from imental design is based on a previous paper by Georgantzís et al. (2004).
ical integration. extracting a quantity K H (K L ) from the water stock S H (S L ). The third term reflects the shadow price of the resource. 7 Our exper 8 Competitive extraction is not studied in this paper. 9 We do not specifically address issues related to vert each ty eriods. A seller can sell water of high and low quality se the model described above with the following values for the parameters: The specific utility and cost functions used are provided in appendix C.
Applying the above equations, in the steady state of the social optimum a stock size of (S H , S L )=(4.84, 5.01) is obtained associated with the prices (p H , p L )=(102, 86). However, in order to simplify subjects' perceived feedback conditional on their strategies, our design allows only for discrete quantities and prices. Thus, if subjects ad of this steady-state equilibrium stock policy, they must aim at stabilizing stocks S L )= 5). Obviously, this is ach ualit equals the quantity sold in each period. Assuming izing total surplus the quantities, (K Hh , K L matically assigned by the server, respectively, for household (h) and agr 10 Given that their purchases in each period are used to serve their current needs. the ses tudy resource management and wa sion, a deterministic end game horizon was used (a total of 50 periods), of which subjects were informed at the beginning of the experiment. 11 Subjects knew the type of water they were managing. That is, they were conscious about a generic preference by consumers for one good (high quality) over the other. Moreover, they knew that their products were demand substitutes (though not perfectly) and that their extraction cost structures were identical. Subjects received a table with their unit costs depending on the stock size (see the instructions in the appendix). A simulator (made available to them) informs them on the hypothetical costs and gains they would make if they sold all the units of each product for which they are currently submitting post bids. They know that the actual number of units they sell will be known only after they have posted their period bids and that the (automated) demand's reaction to these bids is returned to them on the feedback screen.
Three different market structures are compared. We s ter allocation under a private monopoly, a competing duopoly and the social planner. Three simplest markets are chosen given that we do not have a benchmark to start from. As we mention, the existing experimental literature has looked at problems motivated by specific problems. 12 We instead choose an experimental design that studies the effect of market structure upon resource management and water allocation.
We do not focus on competitive resource extraction at this stage as several water allocation agencies are the sole suppliers of water to their respective areas. We study the following treatments: T1 -Private Monopoly: The monopolist has joint ownership of both sources.
Consumer behaviour is simulated. They reveal perfectly and accept trades at zero surplus. The monopolist posts price-bids for both water qualities. Given these offers, the maximal consumer rent is determined in the simulated centralized downstream market: where w denotes the vector of sealed offers and k denotes the vector of quantities. Thus, the bundle of high quality and low quality water which produces the highest consumer rent is allocated in the economy.
T2 -Private Duopoly: The market is supplied by a non-cooperative duopoly.
Each firm offers one type of water and independently decides on price-bids. There is 11 Given the complex nature of the experiments a sufficiently long time horizon was chosen. T3 -Public Monopoly: The public monopoly decides on both water qualities.
Note, however, that the public monopoly only decides on the price and quality offered, and not the quantity. It is required to meet demand at all stages. Like in all treatments reported here, there is optimal simulated coordination in the downstream part of the market. Subjects act as public monopolists, maximizing total social welfare rather than private th , the maximum quantity each one of them could trade). In the case of atically calculating the optimal consumption of each water quality and the profits.
A history window displays all past outcomes regarding own decisions, i.e.
quantities, payoffs and market prices. In duopoly markets, each subject also receives the clearing price at which the "other" water quality was sold. In each period, subjects are asked to submit their respective reservation prices (offer bids) for each unit of product (from the 1 st unit to the 5 the duopoly, rivals had to post, simultaneously, five sealed offers of each quality water which should equal the minimum price at which they were willing to sell the respective unit. 14 Subjects were told that offer bids had to exceed weakly the cost of the corresponding unit, and offers of subsequent units would have to be non-decreasing.
Once offer bids are submitted, behaviour on the demand side is simulated by a programme autom ir distribution by consumer type for which total consumer surplus is maximized.
After bids were announced, all units of the same product in a period were sold at the same market price (see instructions). This price was the maximum offer bid lying below the lowest willingness to pay on a demand function-like ranking from high to low. All subjects were informed about how the market price is determined. 13 Two possible extensions can be run in this experimental design, the "strangers" protocol and a "coordinated" duopoly. "Strangers" protocol, forming a different subject-pair (duopoly) in each period, would be an interesting extension, the resulting noisy feedback would require allowing for longer experimental sessions. Pilot sessions not reported here indicate the sessions lasting longer than 70 periods are required. In the case of a coordinated duopoly each resource is managed by a different subject, but the interface used is that of the private monopoly with two subjects sitting in front of each one of the PC's. Communication and agreement on the timing of decision submission and the possibility of iterated inspection of the "competitor's" strategy before jointly pressing the "OK" button render this setup highly collusive. However, individual incentives remained uncoordinated and no side payments were feasible. 14 Producers in treatments 1 and 3 had to post five sealed offers for each one of the two water qualities. appendix B) in order to avoid significant differences in individ g the feeling of "unfair" variations of per capita earnings across treatments (apart from the horizontal externality due to competition, each duopolist manages only a part of the market).

Results
The steady state equilibrium stock level for our discrete strategy space version of the model implemented here is 5 for both water qualities. Once this level is reached, the optimal extraction rate is dictated by the rate of the inflow. That is, in each period a firm should aim at selling 3 units of each water type (equal to the amount entering into the tank due to the natural rate of inflow). An efficient management is one in which these predictions are fulfilled over the maximum number of periods possible. Given the specificity of our design and focus, we report data on price levels for each water quality and price-weighted average quality sold to the consumers, which cannot be contrasted to any previous T water resou is that the average stock of water under a private monopoly (T1) and duopoly (T2) is higher than under a social planner (T3). The average stock of the social planner is around 25% lower than under any other market structure. This is true for both high, and low, quality water. A welfare maximizing social planner, not operating under market n be followed by a brief discussion on the co-ordinated duopoly experiments.

Sto anage
m e e o management is one of the problems faced by many countries across the world. Given , we compare how a public monopoly manages its water resources relative to a i e) y n a uo ly F re anagement across the three market structures for the low and high quality stock. All treatme opoly treatment, which is the only one presenting a sustained decreasing tendency. Note, however, that the steady state reached by the public monopolist at the end of the experiment.

High Quality Price
Average Quality/Price

Low Quality Quantity
High Quality Quantity incentives, tends to over extract the (non-competing) resource. Interestingly the average quality to price ratio is the same for a monopoly and duopoly.  T1 T2 T3 T1 T2 T3 T1  T2 T3 T1  T2 T3 T1 T2 T3 T1 T2 T3 T1 T2  Interestingly, in period 50 both the (private) monopoly and duopoly are substantially above this (around 10) steady state optimum. Figure 2 indicates that the duopoly maintains higher stock levels than the private monopoly for the low quality stock. The contrary is true for the high quality stock. A Wilcoxon test on the data indicates that the stocks maintained by the three market structures are statistically different (see Table 2).  in case the quality falls below the "potable" threshold, negatively cess of arises due to the additional objective of maintaining quality above a certain level (in order not to trigger the costly (inefficient) depuration procedure).  Otherwise, all outputs across all treatments are statistically different.

<Figure 6 here>
One sees that the private monopoly and duopoly maintain inefficiently higher water stocks (low and high quality) than the public monopoly. The stocks maintained by the public monopoly are at the optimal level at the end of the sessions. Given that stocks are declining, it is not clear if the public monopoly would have been able to maintain stocks at the optimal level for longer experimental sessions. The average quality to price ratio is highest for the public monopoly and the difference across all treatments is statistically significant. Further experiments need to be run to see how increasing the number of firms, and/or increasing the number of periods, affects stock depletion.
One also needs to understand some fundamental differences under which a public monopoly operates in our experiments. In our setup the social planner is not allowed to set quantities. Instead, as occurs in real world situati e consumption by posting the right price at which an each extra unit should be sold. Incentives for the public monopoly are such that subjects post their bids aiming at simultaneously satisfying the condition for the hydrological equilibrium of the system ("stock recharge equals consumption") and, at the same time, keeping stocks at the desired "not too low" levels. This is done without letting average quality fall below the "potable" standard in order not to trigger depuration. Whether this induces a persistent learning shortcoming or not is not answered by our results so far. Therefore, an interesting extension of the experiments presented here would be to allow for l s, or calling experienced subjects back to take part in another experiment. Williams, A., Smith, V.

Treatment 1
The aim of this experiment is to study how people make their decisions in certain contexts.
Your decisions in the scenario explained below in detail, will be directly related to a monetary reward you will receive in cash at the end of the experiment. Any doubt you may have will be clarified personally to you by one of the organizers after you raise your hand. Beyond these questions, any other communication is strictly forbidden and is subject to immediate exclusion from the experiment.
You participate in a market which is characterized by the following features: • You are the only producer of two commodities: product H and product L. Specifically, product H is water of High quality, while product L is water of Low quality. Products H and L are substitutes, namely, consumers may, to a certain extent, substitute one type of water with the other.
• There are two types of consumers: households and farmers. Although they have different preferences with respect to the two types of water, they all prefer water of high quality (product H) to water of low quality (product L). That is, they are willing to pay more for H than for L. • The market lasts for 50 rounds.

Decision Making
Your only decision as a producer is announcing the minimum price at which you are willing to sell each one of the first five units of each product. Such announcements of minimum prices are called price bids. In order to make your biding decisions, you have to take into account that: 1. The extraction cost per additional unit extracted and by product is included in the "table of costs" bellow. These costs are the same for the two products, and they are expressed in ExCUs, a fictitious Experimental Currency Unit.
2. Taking into account the costs of the table, you have to announce five minimum prices at which you are willing to sell each unit of the first five units of each product. Therefore, your decision making consists of fixing 5 price bids for each product.
3. You should have in mind that, in order not to make any losses, price bids cannot be lower than the corresponding unit costs included in the table of costs.
4. Price bids cannot be decreasing. That is, your bid for the 1 st unit cannot be higher than your bid for the 2 nd unit; the bid for the 2 nd cannot be higher than the bid for the 3 rd unit, and so on and so forth.  • the cost of the 1 • the cost of the 2nd unit: 4 ExCUs • the cost of the 3rd unit: 7 ExCUs • the cost of the 4th unit: 11 ExCUs • the cost of the 5th unit: 18 ExCUs In order not to make any losses, each one of corresponding unit cost. Therefore, in this example, your bid for the 1 st unit should not be lower 2 ExCUs (cost of the 1 st unit); your bid for the 2 nd unit should not be lower than neither 4 Us (cost of this unit) nor your bid for the 1 st unit; your bid for the 3 lower than neither 7 ExCUs nor your bid for the 2 nd unit, and so on for the rest of the units.
In case you sell 5 units of this product, the stock size at the beg be 10 units (7 you kept plus 3 you get in the new round). If, given your bids for the five units, sales are zero, your stock would be 15 units (12 you already had plus 3 you get at the nning of the roun • oxes that appear at your computer The bids you submit have to be integer numb • you may pro fiv ffer pric s, n f the same product will be consumers at a gle price. This price will b our bid fo he st" each product. The num er of its so each which optimal behaviour E the example above, assume that your bids for one of the products are: 10 (for the 1 st unit), 12 nd ), 14 (for the 3 rd ), 16 (for the 4 th ) and 20 (for the 5 th ). Given your bids, the program of a product will be the difference between the market ts of that specific product (your unit income) and the otal profits will be the sum of the unit profits for all Taking again the previous exam t

Decisions
You take decisions on the minimum price at which you are willing to sell each unit of each one of the the two products. You will fill in all the b screen with your price bids (5 bids for product H and 5 for product L). In each box, you will also get information related to the corresponding unit cost.
ers between zero and 2000.
Although pose e di ent e bid all u its o sold to sin e y r t "la unit sold of b un ld period is calculated by a program simulates the of consumers.

xample
In (for the 2 which simulates the optimal behaviour of consumers determines that 3 units of this product will be sold. The price at which you will sell the three units will be your bid for the 3rd unit, that is, 14 ExCUs.

The profits
• Your net profit of selling each unit price at which you sold all uni corresponding unit extraction cost. T

periods.
Example ple, if, at the beginning of a round, your stock size is 12 units, your total profits in that round will be 29 ExCUs, which are decomposed as follows: i) 12 ExCUs for the 1st unit sold (14 ExCUs you receive for that unit minus 2 ExCUs it costs you extracting it).
ii) 10 ExCUs for the 2 nd unit sold (14 ExCUs you receive for that unit minus 4 ExCUs i costs you extracting it).
iii) 7 ExCUs for the 3rd unit sold (14 ExCUs you receive for that unit minus 7 ExCUs it costs you extracting it).
ward at the end of the session will be the sum of your profits The case you sell 5 units.
At the end of each round, the computer screen will show you the total profits obtained in und, including information about unit cost, market price and number of units sold of order to make sure you understood correctly the market described, we will proceed next to ssion of 5 rounds. Please, feel free to make any questions you may have during this Tha • Your monetary re accumulated in 15 rounds (randomly selected by the computer) of the total of 50 rounds, at an equivalence rate of 800 ExCUs=1 Euro. You will be paid in cash at the end of the session.
information • During decision making, the computer will provide you with a table (for each product) on which, conditional to your bid and cost for the corresponding unit, the revenue as the net profit are calculated in each of the five possible scenarios: a) In case you only sell the 1 st unit; b) If you just sell the first two units; …e) In • that ro each product.
• During the experiment, you will have at your computer screen a history of past rounds (market price for each product, number of units sold of each product, total revenue and total profits per product).
In run a pilot se pilot session. The aim is that you should take control of your own decision making. nk you very much for your collaboration. Good luck! The You rewa ll receive in cash at the end of the experiment. Any doubt you may have will be larified personally to you by one of the organizers after you raise your hand. Beyond these r communication is strictly forbidden and is subject to immediate exclusion • H and product L).
ith the other.
in this room, randomly selected by the computer when the session starts.
more for H than for L.

ecision Making
our only decision as a producer is announcing the minimum price at which you are willing to sell each one of the first five units of your product. Such announcements of minimum prices are called price bids. In order to make your biding decisions, you have to take into account that: 1. The extraction cost per additional unit extracted and by product is included in the "table of costs" bellow. These costs are the same for the two products (therefore, costs conditions for you and your competitor are identical), and they are expressed in ExCUs, a fictitious Experimental Currency Unit.
2. Taking into account the costs of the table, you have to announce five minimum prices at which you are willing to sell each unit of the first five units of your product. Therefore, your decision making consists of fixing 5 price bids.

Treatment 2
aim of this experiment is to study how people make their decisions in certain contexts.
r decisions in the scenario explained below in detail, will be directly related to a monetary rd you wi c questions, any othe from the experiment.
You are part of a market which is ruled out by the following characteristics: There are two producers (1 and 2) and two commodities (product Specifically, product H is water of High quality, while product L is water of Low quality. Products H and L are substitutes, namely, consumers may, to a certain extent, substitute one type of water w • You are one of the two producers in this market. At the beginning of the session, the computer will indicate if you are producer 1 or 2. Your competitor will be one (always the same) of the subjects • There are two types of consumers: households and farmers. Although they have different preferences with respect to the two types of water, they all prefer water of high quality (product H) to water of low quality (product L). That is, they are willing to pay • The market will last for 50 rounds.

Example
In the example above, assume that your bids for one of the products are: 10 (for the 1 unit), 12 (f 14 (for the 3 rd ) 6 (fo e 4 th d 2 th ) ve u , m hich simulates the optimal behaviour of consumers determines that 3 units of this product will e price at which you will sell the three units will be your bid for the 3rd unit, that is, h nit wi rence between the market price at your unit income) and the corresponding e sum of the unit profits for all periods.

Example
Taking again the previous exam at the beginning of a round, your stock size is 12 units, otal profits for that round will be 29 ExCUs, which are decomposed as follows: u receive for that unit m In case you sell 5 units, the stock size at the beginning of next round would be 10 units (7 kept plus 3 you get in the new round). If, given your bids for t zero, your stock would be 15 units (12 you already had plus 3 you get at the beginning of the d).

Decisions
You take decisions about price bids at which you are willing to sell your product. You will fill in the boxes that appear at your computer screen wit will also get information related to the corresponding unit cost.  At the end of each round, the computer screen each p • During the experiment, you will have at your computer screen a history of past rounds (own and rival market price, number of units sold, revenue and profits per product).
In order to make sure you understood correctly the market described, we will proceed next to run a pilot se p Th ou very much for your collaboration. Good luck!

Trea
The aim study how people make their decisions in certain contexts.
Your decisions in the scenario explained below in detail, will be directly related to a monetary clari ques from e in a market which is characterized by the following features: t L). That is, they are willing to pay more for H • m prices are alled price bids. In order to make your biding decisions, you have to take into account that: y product is included in the "table of costs" bellow. These costs are the same for the two products, and they are expressed in ExCUs, a fictitious Experimental Currency Unit.
. Taking into account the costs of the table, you have to announce five minimum prices at which you are willing to sell each unit of the first five units of each product. Therefore, your decision making consists of fixing 5 price bids for each type of water.
3. You should have in mind that, in order not to make any losses, price bids cannot be lower than the corresponding unit costs included in the table of costs.
4. Price bids cannot be decreasing. That is, your bid for the 1 st unit cannot be higher than your bid for the 2 nd unit; the bid for the 2 nd cannot be higher than the bid for the 3 rd unit, and so on and so forth.

tment 3
of this experiment is to reward you will receive in cash at the end of the experiment. Any doubt you may have will be fied personally to you by one of the organizers after you raise your hand. Beyond these tions, any other communication is strictly forbidden and is subject to immediate exclusion the experiment.
You participat • You represent a social planner who produces two commodities: product H and product L.
Specifically, product H is water of High quality, while product L is water of Low quality.
Products H and L are substitutes, namely, consumers may, to a certain extent, substitute one type of water with the other.
• There are two types of consumers: households and farmers. Although they have different preferences with respect to the two types of water, they all prefer water of high quality (product H) to water of low quality (produc than for L.
The market lasts for 50 rounds.

Decision Making
Your only decision as a producer is announcing the minimum price at which you are willing to sell each one of the first five units of each product. Such announcements of minimu c 1. The extraction cost per additional unit extracted and b 2 the table that the unit costs decrease with the stock size. At the beginning of the • the cost of the 3rd unit: 7 ExCUs • the cost of the 4th unit: 11 ExCUs • the cost of the 5th unit: 18 ExCUs In order not to make any losses, each one of corresponding unit cost. Therefore, in this example, your bid for the 1 st unit should not be lower s (cost of the 1 st unit); your bid for the 2 nd unit should not be lower than neither 4 of this unit), nor than the bid you fixed for the 1 st unit; your bid should not be lower than neither 7 ExCUs, nor than your bid for the 2 nd unit, and so on for the In case you sell 5 units of this product, the stock size at the beginning of next round would us 3 you get in the new round). If, given your bids for the five units, begi • illing to sell each unit of each one of the two products. You will fill in the boxes that appear at your computer screen with your p r product L). In each box, you get information related to the corresponding unit cost. T id u it have numbers between zero and 2000. • Although you may propose five different price the same product will be a . This price will be your bid for t product. The number of units sold each period is calculated by a program which simulates mal behaviour of consu s. st price at which you will sell the three units will be your bid for the third unit, that is, • to maximize the social benefit per unit sold in this market, which is defined as the difference between the utility level generated by each type of water and all possible combinations of your sales are zero, your stock would be 15 units (12 you already had plus 3 you get at the nning of the round).

Decisions
You take decisions about the minimum price at which you are w rice bids (5 bids for product H and 5 fo will also he b s yo subm to be integer bids, all units of sold to consumers at single price he last unit sold of each the opti mer

Example
In the example above, assume that your bids for one of the products are: 10 (for the 1 unit), 12 (for the 2 nd ), 14 (for the 3 rd ), 16 (for the 4 th ) and 20 (for the 5 th ). Given your bids, the program which simulates the optimal behaviour of consumers determines that 3 units of this product will be sold. The 14 ExCUs.

The aim
As a social planner, your aim in each round is unit consumed and the corresponding unit extraction cost.

The information
• At the beginning of each round, the computer will provide you with a table containing, conditional to the stock size for each consumption of the two products, the corresponding social benefits (measured as the difference between the utility level and corresponding extraction costs) of that round. rmation will include the unit cost, market price and number of units sold of each product. the experiment, you will have on your computer screen a history of past rounds Mon • ly selected by the computer) of the total of In or rectly the market described, we will proceed next to a pilot session of 5 rounds. Please, feel free to make any questions you may have during this n. The aim is that you should take control of your own decision making.
• At the end of each round, the computer screen will show you the social benefits obtained in that round. This info • During (market price for each product, number of units sold of each product an social benefit).

etary reward
Your monetary reward for participating in this experiment will be the sum of the social benefits accumulated along 15 rounds (random 50 rounds, at an equivalence rate of 800 ExCUs=1 Euro. You will be paid in cash at the end of the session. The cost function of producer i (i = H, L) is given by . Given the