The Problems of Analysing Markets for Irrigation Water

Water

byBrian Davidson

School of Resource Management,University of Melbourne, Parkville.

A paper presented to the 46th Annual Conference of the Australian Agricultural and

Resource Economics Society, Melbourne, February 2004.

- 2 -

The Problems of Analysing Markets for Irrigation Water

Abstract

The view that the price of water is too low and that water does notflow to its highest value end use, have led many analysts to concludethat a market for water, in which price can adjust to accommodatechanges in the supply and demand for the product, is essential.However, this solution assumes that the market could approachsomething close to perfect competition, purely by making minorchanges within the existing regime of property rights. In this paper itis argued that problems evident in this market stem from a multitude ofmarket failures and characteristics that are particular to it. Tounderstand this case more fully it is necessary to come to terms withthe theoretical formulations of the market for water and the practicaldifficulties associated with it. It is concluded that it may not bepossible to obtain an optimal solution in the market for water, as itcurrently exists. Thus, the solutions to the problems evident in thewater market may need to rely on controlling quantity of water thatflows through it, even though price is an ideal economiser ofinformation.1.

Introduction

It would appear from reading the Wentworth Group of Concerned Scientists (2002)submissions that the traditional methods of allocating water1 in Australia does notachieve all societies’ objectives. The current method of allocating water can be describedas fixed entitlement system that allows for limited trading. Most entitlements to water are Water in this paper is defined to include surface water that is regulated and used for irrigation only. Itdoes not include subsurface supplies or unregulated water captured for private use or water used solely fornon-agricultural uses. However, the issues discussed in this paper may well have an implication on the wayanalysts think about these other forms of water regulation.

1

- 3 -nothing more than a right to access (pump) a specified quantity of water from a river orcanal that has its flow regulated. These entitlements were generally allocated on the basisof geographic location, during or soon after watercourses were regulated. Farmers, themajor holders of these entitlements, paid little or nothing for them originally and it isgenerally accepted that they paid an inadequate price to maintain them. In the beginning,water entitlements were attenuated to the land. After the COAG agreement, the waterand land titles were separated in order to allow trading amongst users. Such a move isbelieved to be a step in the right direction, as it permited the flow of water to its highestvalued end-use.

It could be argued that the current markets for water are ideal, as they serve the interestsof most participants, albeit in a limited way. Farmers can purchase or sell water asrequired, and the government making purchases on its behalf can satisfy the interests ofthe environmental lobby. Yet, to call the current method of trading water in Australia afree and fully flexible market system, as economists believe them to be, would involve ahuge stretch of the imagination. The market for trading water is extremely thin, mainlyinvolving temporary flows. Most water still ends up in the hands of those who have atraditional entitlement to it. Thus, it could be concluded that if water is now allocated toits highest end use, then the fixed entitlement system was achieving most of this anyway.There is no freedom to enter or exit the market, as in many cases any trader must have anexisting entitlement with a water authority. In many cases farmers still pay only afraction of the true price of delivering water. Furthermore, Catchment ManagementAuthorities (the true suppliers of water) cannot readily alter prices, raising them to updateor improve infrastructure or reducing them to sell more water. When these criticisms arecombined with claims that not enough (or no) water is allocated to the environment, thenit is easy to conclude that perhaps water is undervalued. Not allowing the market forwater to operate in the way economists believe they should, has meant that not only is theallocation of water skewed, but that any beliefs regarding what the price of water is orshould be that are determined from that market, will not be correct.

- 4 -If water were traded more freely, without the strictures of the fixed entitlementsunderpinning it, or the constraints placed on the Catchment Management Authorities,would water be allocated so that it maximised social welfare? Would water end up inuses where it is valued most highly? Would the price for it truly reflect its value? Suchquestions are important as they have an implication for the way the water marketcurrently operates, what it is believed is delivered from that market and any changes(improvements) that may be made to it.

In this paper it is argued that a number of theoretical and practical considerations need tobe evaluated before any firm conclusions are reached on the system of allocating water.If these considerations are valid, then it could be the case that the desired outcomes thatare believed to come from the market may not be achieved. The aim in this paper is tospecify these theoretical and practical conditions, to make a judgement call on wetherthey hold and in light of that, suggest an alternative way some of the benefits that arederived from a free market may be delivered to all water users.2.

The Original and Current Systems of Allocating Water in Australia

It is possible to think of different resource allocation systems as a continuum frommarkets that are free of any government intervention through to command economies.Most resource allocation systems are a combination of a market modified by governmentintervention in the form of taxes, subsidies and/or regulations. The purpose in thissection is to specify the system that has been used to allocate water in Australia, prior tothe establishment of a trading system. Without wanting to generalise what is a highlycomplex system, almost all were highly regulated and can be described as a fixedentitlement systems. These systems were modified to include trading during the 1990s.At the heart of an entitlement (or resource allocation) system is the specification of theproperty rights within it. Property rights define the legitimate owner of a commodity orgood or service and who is entitled to enjoy the benefits of ownership. Good propertyrights have well defined rules regarding ownership, use, transferability and enforcement.

- 5 -It as been recognised by commentators (notably Young and McColl 2003 amongst manyothers) that the property rights underlying water allocation in Australia, both now and inthe past has not been ideal.

The history of how water was allocated in Australia is presented in Davidson (1969).Water rights were attached to the land and were, as Challen (2001) has described, aresource quota. In other words, a direct limit was placed on the amount of water a farmercould access. Depending on the system in question, in many years owners may receivemore than their entitlement (if the system is run conservatively as in Victoria) or less (ifthe system is run with less security as in NSW). As farmers contributed only a fraction ofthe costs of supplying water and governments subsidised the costs of constructing andoperating the system, prices played little if any role in the market. To describe thismarket in a simple analytical way, it could be said that the supply and demand schedulesfor water were perfectly inelastic, with aggregate demand placed closer to the origin thanaggregate supply (see Figure 1). It should be noted that this diagram does not includesales water.

FIGURE 1 HERE

Economists were quite rightly critical of the fixed entitlement system. Besides the pointthat in the absence of prices, information within the market cannot flow and the inputcould not be allocated to its most productive and highest value use. In addition,Catchment Management Authorities could not raise prices to update, improve or invest inthe infrastructure that delivers the product to consumers. When these criticisms arecombined with claims that not enough water was allocated to the environment, it is easyto conclude that water was undervalued. However, the fixed entitlement system ofallocating water served its purpose, especially when the water economy was beingdeveloped, as it provides a degree of certainty.

By the early 1980s it was generally recognised that Australia had a mature watereconomy, one in which all readily available supplies of water had been exploited andwater users competed for what was considered to be a resource that was becoming scarce.To overcome the shortcomings of the existing system of allocation, state governments

- 6 -allowed water users to trade water. To accomplish this, it was necessary to separate thetitle for water from the land title. These arrangements were formalised somewhat underthe COAG arrangements in 1994-95.

Godden (1997) and Freebairn (2003) depict how these current arrangements work. Themain change is a recognition that the aggregate demand curve for water is downwardsloping (see Figure 2). The price for water is determined where supply, which isconsidered to be inelastic and fixed at the quantity of water available in the system,equates to aggregate demand. To account for trading it is recognised that users havedifferent demands (see Figure 3). By fixing the amount available along the horizontalquantity axis (thus accounting for an inelastic supply schedule), two competing usersdemand schedules can be imposed on the diagram. If 0Q2 is allocated to user 1 andQwQ2 is allocated to user 2, the market is not in equilibrium. User 2 values the watermore highly than user 1, Q2Q1 of water is exchanged and the equilibrium is reached atQ1 deriving price Pw1.

FIGURES 2 AND 3 HERE

3.

Limitations of the Current System of Allocating Water

The point to recognise is that the existing arrangements are not a freely functioningmarket system. All it really achieves above the old existing system is the ability for usersto trade in a limited fashion and the (as yet unrealised) potential for governments to buyenvironmental flows from existing users. Within the existing water markets the amounttraded is relatively thin. Temporary flows account for most of this trade and permanenttrades are rare. Those who have always had access to it still use most of the water.While activity and prices in the market do rise when water is scarce, trading wouldappear to have little impact on the overall spatial and temporal flows through theirrigation network.

3.1

Is the view of the demand for water the correct one?

This approach to allocating water focussed analysts attention on to the specification ofdemand. The aggregate demand for water is usually assumed to be a linear and

- 7 -continuous function. Yet, as Godden (1997) and Freebairn (2003) acknowledge in theirmodels of the water market, individuals have very different demand schedules, causingthe aggregate demand schedules to lose their smooth continuous nature.To get a smooth specification of aggregate demand requires specification of theSonnenshein-Mantel-Debreu conditions. These conditions assume that all consumershave the same preferences and proportional incomes (Keen 2001). Clearly, as in thewater market, in a situation where the direction and benefits of trade are dependent ondifferences in demand, imposing these conditions is not an option. These limitations canpartially be overcome by estimating demand for water for similar groups of consumers,as Zaman, Davidson and Malano (2004) attempt. Yet within each group of water users,the Sonnenshein-Mantel-Debreu conditions are still implicitly applied.

Furthermore, while it is generally acknowledged that the demand for water is derivedfrom the demand for the products produced from it, little mention is made of the logicalproblems associated with this view. In the demand for a finished good, it is generallyacknowledged that consumers are motivated by a desire to maximise utility. The samething cannot be claimed with the demand for an input, where the buyers are producersmotivated by a desire to maximise profits. Also any of the normal conditions associatedwith estimating consumer demand, such as the Engel, adding up, Cournot, symmetry,negativity and homogeneity restrictions, may not hold for an input. This leads to themore insidious problem of whether the independence of supply and demand exists in thismarket (something that is discussed in more detail below).

3.2 Is the supply schedule perfectly inelastic?

Most economic analysts view the supply of water as being perfectly inelastic and setequal to the quantity of water available for sale2. The advantage of this approach is thatthe identification problem (i.e. trying to determine whether changes in prices are due tochanges in demand or supply or both) is nullified as all price changes are assumed to be

- 8 -the result of changing demand. This does not seem to be an unreal assumption to make,as a certain quantity of water is usually available to distribute. In most free markets thesupply schedule is upward sloping. However, both these situations may not exist in themarket for water. The assumption of an inelastic supply is not only unrealistic, but alsohas important ramifications in the analysis of the market.

Irrigation systems are not dissimilar to other projects that have a large investment ininfrastructure. These are generally classified as decreasing cost industries. Given theovercapacity that is built into the system, these types of industries display decreasingaverage costs. As is well known, if average costs are declining, so are marginal costs andmarginal costs will be lower than average costs. Thus, the supply curve may well bedownward sloping.

This argument could be dismissed by recourse to an assumption that most of the costs ofan irrigation scheme are those incurred in construction and are therefore sunk. It isfurther complicated by the fact that the supply of water in the existing system that allowstrading is derived from existing entitlement holders, not from a Catchment ManagementAuthority. Importantly, the assumption removes analyst’s attention from the supply ofwater, which is surprising given that the product is assumed to be scarce. How it cameabout that the true suppliers of water would be excluded from an analysis can only beguessed at.

Another important dimension to treating water supply as perfectly inelastic is that themany uncertainties that are inherent to the water market are glossed over. For instance, itis difficult to assess the temporal aspects of the market, both within and between seasons,if supply is assumed to be given. Furthermore, the role of uncertainty is difficult tounderstand with a fixed supply schedule. This is especially the case, as the CatchmentManagement authorities tend to eliminate downside risk in the manner in which theyannounce allocations.

While it cannot be proved, it is interesting to note that in most economic assessments of water marketssupply is assumed to be inelastic, while in hydrological studies demand is assumed to be perfectly inelastic

2

- 9 -

Godden (1997), Lim and Dwyer (1999) and Read, Sturgess and Associates and Watson(2001) all concede that the possibility of a natural monopoly exists in the water market.Further evidence can also be found in the early history of running irrigation schemes inAustralia. In the late 1800s and early 1900s many user dominated irrigation trusts weredeclared bankrupt, despite the fact that the government paid the debts they incurred inconstructing them (Davidson 1969).

If the supply curve for water is downward sloping, then three important implicationsarise. First, it might explain why water was always sold at a loss. In such markets thesocially optimal price (i.e. where supply and demand are equal) will be below the averagecost of supplying water. Second, if water supply was privatised, then a monopolist couldenter the market and, by equating marginal costs with marginal revenues, restrict suppliesand force up prices. Third, decreasing cost industries are recognised as being a form ofmarket failure and one reason that may justify government intervention.

3.3 Is supply and demand independent in the water market?

It must be asked if buyers and sellers are truly independent in the fixed entitlement withlimited trading water market. A necessary assumption of a market is that the actions ofproducers determine the supply schedule, while those of consumers determine thedemand schedule. In the existing water market, these are one and the same people.Wijedasa and. Malano. (2002) found that the demand for water differs through out aseason. This means that water users may react differently at different times during theyear. An existing entitlement holder can make decisions to buy water in one period andsell it in another period.

How is it possible to analyse any market using conventional techniques, when supply anddemand are interdependent? Theoretically and practically it cannot be done. Thisconcern could be nullified somewhat if true suppliers of water (the Catchment

Management Authorities) were included in the economic formulations in the market and and equal to the evaprotranspiration rate.

- 10 -demand was modelled from a users perspective, while the water traded was reduced to aresidual balancing activity.

- 11 -

3.4 Does it matter if the market for water is thinly traded?

It should be noted that the existing trade for water only accounts for a small proportion ofall the water allocated in the system. Zaman et al. (2004) report that approximately 20per cent of all the water allocated in 2003 (a relatively dry year) was traded, most of thisbeing temporary trades. Crase, O’Reilly and Dollery (2000) and Freebairn (2003) arguethat the lack of trade reflects uncertainty about future claims for water, the high costs oftransfer and inadequate specification of the property right.

The problem with thinly traded markets is that one gets a false sense of the underlyingforces driving the market. Perhaps the factors that drive the decisions to access waterfrom their entitlements are different to those that drive the full market for water. Further,the possibility of an individual exerting undue influence on the traded water marketexists. In other words, it could be asked: Is the price derived in the market trulyrepresentative of the whole market? Furthermore: can accurate policy decisions for thewhole market be based on the price derived from the traded water market? Maybe not!

3.5.

Is equilibrium even possible in this market?

Given the problems raised above, it must be asked: Is equilibrium even possible in thetraded market for water? It would seem that such a situation might only occur rarely intimes when water is scarce. Mostly, there is excess capacity in the system, especially ifthe supply schedule is downward sloping and where the systems are run conservatively(as in Victoria). It is entirely possible to have the supply schedule far to the right of thedemand schedule (see Figure 4, where the normal conditions regarding the slopes of thesupply and demand are assumed to hold). In such a case, supply and demand cross belowthe quantity axis, resulting in negative prices. An alternative situation can be envisagedwhen water is very scarce and the supply schedule is totally above the demand curve,indicating that along its whole range, consumers do not have the capacity to pay for anyof it. In these cases, supply and demand do not cross at all.

FIGURE 4 HERE

- 12 -It would seem that it is possible to think of situations where equilibrium does not exist.To establish the fact that it does, really requires more research. It should however benoted that this situation only really arises from the fact that the market is thinly tradedand is run to complement an existing fixed entitlement allocation system.

3.6 Is water a stock or a flow?

Intriguingly, it is interesting to note that the horizontal (quantity) axis of most, if not all,descriptions of the market are specified in physical units (i.e. megalitres of water). Yet,the entitlements are all based on flow (i.e. megalitres per annum). Even proponents ofenvironment flows argue in terms of a fixed physical unit of allocation. Watson (2003)addresses this issue by making the point that if the target variables were turned intoinstrumental variables of the flow of water, environmental arguments would make moresense.

If the argument applies to environmental flows, then surely it is applicable to the wholesystem, the farm sector included. Hydrologists control the system by measuring flows.The capacities to move and pump are all specified in terms of a flow. Farmers measurethe amount of water they need as an evaprotranspiration rate, which is a flow. So if flowsare important to those who use and control the systems, why is it that economists andpolicy makers persist in using the stock of available water?

By representing the quantity as a flow may allow policy makers to concentrate the issueof the rights current holders have to water, how they may be altered and whatimplications this has. Presently, with physical units of a stock being analysed, the truenature of service being provided, i.e. the regulation of water, is hidden.

3.7 Is water a right or a good or a service?

With the present structure, existing water holders control the system, as they have a‘right’ to water. However, one of the real problems regarding irrigation in Australia liesin the belief that the provision of water is a ‘right’. In other words, it is believed that

- 13 -people have a right to certain quantity of good quality water for drinking and agriculturalpursuits.

It should be noted that the concept of ‘rights’ is a particularly North American thingembodied in their culture, political mindset and system of distributing natural resources.They are not a particularly Australian ideal, or for that matter, one that comes from aBritish heritage. Rather than enshrining the ideals a society desires in a set of prescribed,inalienable and overriding rights, as they do in the United States, in Australia a muchmore flexible approach is taken. Be that as it may, thinking of water as a right, in theUnited States definition of the term, leads to the conclusion that the costs of providingthat ‘right’ must be ignored. This concept of water rights is questioned, but notchallenged in Australia, and is used by both sides of the current debate on environmentalflows. Farmers and irrigators believe they have a ‘right’ to water, while the

conservationists believe they have a ‘right’ to a clean environment. However, if the costsof providing these rights are considered, then water is not a right, but is a good, whichcan be traded like any other commodity. The belief now embedded in the current systemof allocation is that water availability and its supply is not a ‘right’ that people can insistupon. It is a ‘good’ that comes at a cost and which people can buy and trade it as they seefit. What is tragic is that the owners of the ‘right’ to pump water are the first arbiters ofwho can use the water and yet they pay little for those monopoly ‘rights’.

However, is the provision of water a ‘good’? While it is accepted that it is not a right,perhaps it is not a good either. It could be a ‘service’. According to Petit (1987, p.314)the “…neoclassical approach, … attaches little theoretical importance to the distinctionbetween goods and services.” Yet there is a subtle difference between the two. Goodsdeal with tangible products, while services account for intangible products. In anotherway of looking at this, Noble (1995) argues that services cannot be stored or transferred.So sectors such as finance, transport and (according to Noble) water are all services. Thereal argument is, what is being dispensed by the Catchment Management Authorities andwhat are users buying and trading. If it is water per se, then it is a good. Or is it theability to regulate water, which would make it a service. To treat the water sector as a

- 14 -service industry changes the whole way the market is dealt with. Thinking of water as aservice allows buyers and sellers to think about periods pf adequate service andinadequate service, not whether they get their full entitlement or not. It allowsparticipants in the market to think about temporal questions and channel sizes.

3.8 Does the allocation system achieve its objectives?

The present system should ultimately be judged by its abilities to meet its objectives.That would mean that it delivers water where and when it is needed. Consequently, itmust be asked if it is achieving the savings in water use that are needed to provide morewater to the environment? As it turns out, it is too early to resolve this question and moreresearch needs to be undertaken on it. If water is to be saved so that it can be allocated tothe environment, it could be asked should the savings in water flow be achieved from theusers or from the system? At a more fundamental level, the need for system of allocatingwater should be investigated, as the whole issues is dependent on assessing how scarce iswater?

3.8.1

Has allowing trade led to greater water availability?

It would appear that this question is yet to be addressed by any analysts in any detail, asthe allocation system has not as yet actually had a chance to deliver. However, a numberof issues can be raised which may shed light on this question. It is well known thatopening a market up to trade leads to an increase in overall efficiency. Inefficiencies aretraded away as production occurs in its most optimal place. So if there is greaterefficiency, shouldn’t there be more water available? Well not necessarily.Allowing trade within a fixed entitlement allocation system possibly causes moreproblems than it solves. The rules of the trading scheme mean that only existing accountholders can trade in the market. New entrants face considerable costs in accessing themarket. So the possibility of achieving efficiency gains through transferring the productwould appear limited, especially seeing that not much is traded. Few of the inequalitiesevident in the original distribution of water licences and the inefficiencies derived from

- 15 -them are not addressed by these reforms, except if in the rare case of when a permanenttrade in water entitlements occurs.

Most importantly the over allocation of water in select areas may well have beenenhanced by these reforms. As Young and McColl (2003) point out the new

arrangements provided those who held water licences but had never developed irrigationinfrastructure on their land (i.e. dozers or sleepers) with a valuable resource that theycould trade. Prior to trading, water not used stayed in the reservoir. Some governmentshave attempted to institute rules that suggest that water entitlements should besurrendered if are not used. Yet what is considered use, if it is on-sold, is it used? Itwould appear that allowing trade may have increased the pressure on what wasconsidered to be an over stretched system.

3.8.2 Should the savings in water flow be achieved from the users of the system?It would appear that the bulk of the savings of water are to be made from the users, notfrom efficiency gains from running the systems more efficiently. This belief is based onthe fact that the major change in the fixed entitlement system was made to allow trade.Sure, Catchment Management Authorities had to charge prices that cover their costs, yetthe only way they earn income is from the sale of water. Sell more water, at a fixed priceand their revenues rise. If they sell less, then their prices need to rise to cover the losses3.It could be asked if the users have the capacity to save the required water?

One way of saving water from users is to restrict the allocations, forcing them to buywater on the market, thus increasing the prices for water. This would mean that farmerswould use water more prudently and invest in technology that uses less water. Such anapproach implies that the costs of altering the on-farm water distribution technology arelow. Yet this is not the case. Any move from flood to drip irrigation technology dependsof the marginal increases in water costs and by how much extra value is obtained formthe output. Given that water prices are currently so low, it could be argued that water Such a response is dependent on the true slope of their average cost functions. If they are upward slopingthe response is unambiguous. If the average cost function is downward sloping, then it could rise or fall.

3

- 16 -prices may need to increase by a lot to achieve this, especially if the output is sold on theexport market. Another way of getting the price to rise significantly, is for the

government to step in and force the price up by buying any water that becomes availableand use it for environmental flows. Such an approach is not ideal as what it would do isreward the dozers and sleepers and encourage a degree of speculation into the market thatmay be destructive.

Given these problems, it may be more prudent to think of water savings being derivedfrom the distribution system. Yet, there is little incentive to do this. Hence, it could besurmised that the current system for allocating water may not achieve this objective.3.8.3 Is water really a scarce resource in Australia?

If water is not a scarce resource in this country, then all the efforts made to resolveproblems in this field are misguided. However, some doubt exists about just how scarcewater is in Australia.

It is usual to think of Australia as a dry country, one of the driest continents in the world.It is this sort of thinking that resulted in the development of irrigation schemes in the firstplace. It is the sort of thinking that persists today in comments that underlie beliefsregarding the move to a mature water economy and the need to find water forenvironmental flows. The National Land and Water Resources Audit (2000) wouldappear to confirm this view. Yet, it must be asked if water was once scarce without theprovision of irrigation, why is it still scarce once the service was provided and why do theprices for water remain low?

To understand these questions it is necessary to first understand how water is measured.In terms of the distribution and consistency of rainfall, there is little doubt that Australiais a dry place. Alternatively, in terms of available and useful/useable supplies of waterper head of population, Davidson (1969) suggested that Australia was one of the wettestcountries in the world. In Australia, because of the infrastructure invested in irrigationschemes, there is more water available for use than in any other country in the world,

- 17 -with the exception of Iceland (Lomborg 2001). So if regulated supplies of water arescarce in Australia, one might like to ask how much more is required to regulate therivers and expand the environmental problems that are believed to result from it.The real issue is that the abundance of water results in it being a low cost input, which asa consequence, is not always used as wisely or as efficiently as it should be. Where wateris used to produce high value exotic crops (such as cotton and rice) rather than low valuepastures, water becomes a scarcer resource because exotic crops are more dependent onit. To get the highest returns the least expensive techniques of flood irrigation areemployed. These techniques can be employed because water is readily available at sucha low cost. If water was not so available, its price would be higher and farmers mightconserve it by employing more efficient practices. While these issues are at the heart ofwhy irrigation does not result in a drought-proofed state, they also reveal that water usersdo not consider water to be a scarce resource.

To put it another way, the only way that water could become scarcer is if the demand hadexpanded more than its supply. Given the evidence from Lomborg (2001) and if there isexcess capacity in the system at select times, then it would be impossible to argued thatwater is scarce. Rather, the issue of scarcity arises from the way the system is run on atemporal basis. However, it does seem feasible that demand may have risen, as sleepersare now active in the market, the environmental lobby is requesting more water and inplaces were water was over allocated farmers are requesting more reliable supplies.Supply may not have expanded. Why the price did not rise to equate supply and demandwas because it was regulated at a low level.

This leads to the conclusion that the problem is possibly not one of scarcity, but one ofhow water is allocated. A scarcity problem could be solved by providing more product,say by running a more efficient system. An allocation problem can be tackled by eitherfreeing up prices and/or changing the allocations in the market. This is a politicalproblem, which has economic consequences. Scarcity is an economic problem that haspolitical consequences.

- 18 -

4. A Free Market Approach to Allocating Water

Given the problems with the current method of allocating water, it must be asked if itwould not be simpler to resume all the entitlements to water and let users buy it from theCatchment Management Authorities in an open market? By freeing the market up in thismanner, it could be argued that societies benefits would be maximised. However, suchan approach has a number of flaws that make it unlikely to be ever considered.In this view of the water market supply is not fixed and perfectly inelastic. Godden(1997) suggests that as the marginal costs of water are low and the supply curve isupward sloping. The benefits that accrue from free and open trade in water aremaximised and price becomes the economiser of information, regulating suppliers anddemanders in a transaction free world, without any government intervention4.It could be suggested that moves that allow more trade in water are always beneficial.The belief is that the more open and free that a market is the greater the benefits tosociety are and the lower are the costs of controlling the market. The ideal of a perfectlycompetitive will however, only occur if a strict set of assumptions holds. It could also bethe case that extraneous events in the water market are such that these benefits cannot berealised. These include those already mentioned surrounding the specification of thedemand and supply schedules (see Sections 3.1 and 3.2).

Musgrave (1981) argues that the assumptions of a perfectly competitive market can beclassified into negligibility, domain or heuristic categories. Negligibility assumptions arethose that exclude irrelevant issues. Domain assumptions are those that are used tospecify the conditions under which the theory applies. Heuristic assumptions are thosethat are known to be false, but which are made in order to gain a more general

understanding. While it is generally thought that the assumptions that underlie perfectly

- 19 -competitive markets (i.e. that there are numerous buyers and sellers, the product ishomogeneous, freedom of entry and exit and information is perfect and costless) are inthe negligibility category, this is not the case, particularly in the market for water.Furthermore, as they are rarely investigated or altered they are not of the heuristic typeeither. These are domain type assumptions. Hence, one could question whether theperceived benefits from a perfectly competitive market would ever be realised, either inthe market for water or in any market for that matter. This finding should not be used toquestion whether a freer approach to the market would be beneficial. Rather, all that isbeing said is that a perfectly competitive market structure is unlikely to achieve thedesired benefits, as the domain assumptions are too restrictive.

The existence of a decreasing cost industry (discussed earlier in Section 3.2) is not theonly form of market failure evident in the market for water. Many analysts (see Beare andHeany 2001 amongst many others) have recognised that the provision of water may causeexternalities. Most of the arguments surrounding the issue of greater environmentalflows are based on mitigating the effects of an externality. While market solutions existto correct for market failures, it is generally accepted that the transaction costs associatedwith them in the market may well require a more formal response from the government.Freebairn (2003) and Edwards (2003) both recognise that because of reuse issues (i.e. thatsome water may be returned to the river) and the inability to exclude certain users, watermay display the characteristics of a public good. Finally, it would be difficult to arguethat perfect information exists in this market, as the aim of running the schemes is toovercome uncertainty in the supply of water.

Rather curiously, it should be noted that a free market for water would appear to sufferfrom all the characteristics that have been identified with market failure. The problemthen is that if the Theory of Second Best holds, rectifying one market failure may notresult in an improvement in social welfare. By recognising that significant marketfailures exist, it can be concluded that a freely functioning market in which users buy

To obtain this type of market structure would require users to relinquish their existing entitlements and totrade more openly in a market competing for the available supplies of water, something that would be

4

- 20 -water directly from a Catchment Management Authority will not deliver the benefitseconomists expect from it. Hence, aiming for the idealised free market approach may notbe feasible and some form of government-modified market is required.5.

An Alternative Entitlement System

Given the limitations evident in the fixed entitlement systems, both with limited tradingand in the absence of trading, and with a free market in water, is it possible to specify asystem in which many of the current problems evident in allocating water are overcome?Such a system should endeavour to attain all the flexibility of an open market system inwhich water is used in its highest value use, yet overcomes the failures evident in thattype of market. The system presented below is only a suggestion, offered more thananything to prove the point that many alternatives exist to the seemingly vexed questionof allocating water. It is presented more to provoke thought and to encourage somelateral thinking on the issue of allocating water. This alternative system relies oncontrolling quantities, rather than on attempting to obtain a competitive price to allocatewater.

It should be remembered that at the heart of any allocation system is the regime ofproperty rights that underpin it. Given that it would seem that users fixed entitlements towater prevents greater flexibility in the market, perhaps the government should resume allexisting water licences. This can be done, admittedly not easily, by either reducing theallocation to a minimal amount and putting the rest of it on the market, or by buying themout. Reality would seem to dictate that the supply of water is controlled by CatchmentManagement Authorities. If, as suspected, they are natural monopolies, then it will benecessary to curb their activities with government regulations of some sort or another.This may even imply that governments should take total control of them. So far this islittle different to the free market approach specified above. What is different in thisalternative system is what buyers and sellers would trade in this market and who controlsit.

politically difficult to achieve.

- 21 -

If analysts viewed the water market as a service (see Section 3.7 for more details), thenCatchment Management Authorities would sell the benefits of regulated flows of waterand buyers would purchase them. For instance, a farmer could go to the CatchmentManagement Authority with the desire to plant crop that requires water to survive. TheCatchment Management Authority can accept the contract to supply an amount of waterto keep soil moisture at select levels (or within a set of specified parameters) on that crop.Farmers pump water on to the crop when the Catchment Management Authorities advisesthem to do so. It should be noted that the environment (through an agent like theGovernment) could conduct business in the same manner. Thus, water in this system issold as a service, not as a good. The Catchment Management Authorities are in thebusiness of selling soil moisture, not bulk water. Users are buying moisture. Betweenthe two could be a set of agents who monitor and facilitate the trade between buyers andsellers.

A clear implication of this approach is that the importance of determining the price ofwater, a holy grail if you like of the current debate on water allocation, is no longerimportant. If anything, the price of providing a service is what becomes important underthis system. This opens up the question of whether water should be controlled by a pricemechanism, or by a quantity mechanism. Weitzman (1974) puts forward a compellingcase for when a market, which is dominated by uncertainty and externality concerns,should be controlled by price or quantity.

The advantages of selling the service attributes of water are numerous. Initially, themarket for water is far more flexible and the benefits approximate those of a free market(which is basically what it is). Most importantly, the incentive to sell more water toexisting customers is not as strong. Rather, Catchment Management Authorities haveevery incentive to spread the flow of water as lightly as possible, in order to satisfy asmany customers as possible. Further, the current practice of users’ requesting water andthen cancelling it could be minimised. In addition, the problems of enriching dozers andsleepers are eliminated. Trading water between catchments or even interstate (an issue

- 22 -not addressed above, yet very real in the minds of policy makers) is more readilyavailable, as the real suppliers (the Catchment Management Authorities) are dealt with.Finally, as the Catchment Management Authorities own the water from the dam, throughthe channels and farms and back to the river, the vexing issues of externalities and publicgoods are resolved more easily. Thus, the Catchment Management Authorities areresponsible for any externalities that occur and could attempt to solve it using a Coasianapproach. Any water reuse problems are the preserve of those who own them.

Several problems still exist, especially if the natural monopoly instincts of the CatchmentManagement Authorities are not reined in. Understanding and estimating the demand forwater are still not resolved. Monitoring soil moisture on a catchment basin is not easy,but is also not beyond being technically solved. In addition, the question of valuing theenvironmental flows is still a problem.6.

Summary

It can be concluded that analysts, in assessing a move from a fixed entitlement system toone that allows trading, still need to come to terms with a host of problems that mayaffect the expected outcomes from that process. These include a better understanding ofthe demand conditions in the market and the need to come to terms with the fact that thetrue suppliers in the market are excluded. The trade in water is undertaken by existingwater right holders, thus making it difficult to determine the outcomes from the system.A number of unknowns were identified which need further research. Initially researchneeds to be conducted to illicit the true nature of supply and demand in the market. Alsoneeded is research to determine the true impact of market failures on the market. Thisshould lead to an analysis of the different ways water could be allocated. Finally,research needs to be conducted on whether water should be regulated using quantities,rather than the preferred mechanism of price.

- 23 -Not withstanding these researchable issues, it can be concluded that people do not appearto value water, as its price is low, and that not enough is allocated to environmentalflows. Viewing water as a commodity in a fixed entitlement system that allows trade,does not seem to address these issues adequately. All the fixed allocation system withlimited trading does is enrich existing holders of a water entitlement and force CatchmentManagement Authorities to sell more water to raise revenues. There are alternatives tothis approach. One, which may be used, is to allow Catchment Management Authoritiesto sell water as a service.

References

Crase, L., O’Reilly, L. and Dollery, B. (2000), ‘ Water markets as a vehicle for water

reform: The case of NSW’, Australian Journal of Agricultural and ResourceEconomics, 44(3), 299-321.

Challen, R. (2001), ‘Economic analysis of alternative institutional structures’, in Brennan,

D. (ed.) Water Policy Reform: Lessons from Asia and Australia: Proceedings of anInternational Workshop. ACIAR Proceedings 106, Canberra.

Davidson, B. (1969), Australia Wet or Dry?, Melbourne University Press, Melbourne.Edwards, G. (2003), ‘Water policy: Setting the scene’, Australian Economic Review,

36(2), 193-202.

Freebairn, J. (2003), ‘Principles forr the allocation of scarce water’, Australian Economic

Review, 36(2), 203-12.

Godden, D. (1997), Agricultural and Resource Policy: Principles and practice, Oxford

University Press, Sydney.

Heaney, A. and Beare, S. (2001) ‘Property rights and externalities in water trade’, in

Brennan, D. (ed.) Water Policy Reform: Lessons from Asia and Australia:Proceedings of an International Workshop. ACIAR Proceedings 106, Canberra.Keen, S. (2001), Debunking Economics: The naked emperor of the social sciences, Pluto

Press, Annandale.

Lomborg, B. (2001), The Sceptical Environmentalist: Measuring the real state of the

world, Cambridge University Press, New York.

- 24 -Lim, R.K.H. and Dwyer, T.M. (1999), Infrastructure Pricing, Provision and Process:

Implications for Rural Australia. RIRDC Publication No. 99/162, Rural IndustriesResearch and Development Corporation, Canberra

Musgrave, A. (1981), ‘Unreal assumptions in economic theory: The F twist untwisted’

Kyklos, 34, 361-79.

National Land and Water resources Audit (2000), Australian Water Resources

Assessment, Canberra.

Read Sturgess and Associates and Watson (2001) Water Policy Issues – Stage 2, a report

to the Dairy Research and Development Corporation, Melbourne.Noble, C (1995), Australian Economic Terms, 5th ed., Longman, Melbourne.Petit, P. (1987), ‘Services’, in Eatwell, J., Milgate, M. and Newman, P., The New

Palgrave: A dictionary of Economics, Vol 4., MacMillan, London.

Wentworth Group of Concerned Scientists (2002), Blueprint for a living Continent,

World Wide Fund for Nature.

Wietzman, M. (1974) ‘Prices vs quantities’ Review of Economic Studies, 41, 477-91.Wijedasa, H., H. M. Malano, et al. (2002). Water Trading in the Goulburn-Murray

Irrigation Scheme. Melbourne, CRC-Catchment Hydrology.

Young, M. and McColl, J. (2003), ‘Robust reform: The case for a new water entitlement

system for Australia’ Australian Economic Review, 36(2), 225-34.

Zaman, A., Davidson, B, and Malano, H. (2004) What is driving temporary water trading

in the Goulburn-Broken Catchment? A paper presented to the 46th AnnualConference of the Australian Agricultural and Resource Economics Society,Melbourne, February 2004.

- 25 -Figure 1: A Fixed Allocation System.

Price

Qd QsQuantity waterFigure 2: A Fixed Allocation System that Permits Trading: Aggregate Demand.Price

Qs Pw

Qd Qw

Quantity water

- 26 -Figure 3: A Fixed Allocation System that Permits Trading: Multiple Use.Price

PwQd2 Q1 Q2 Qw

Source: Freebairn (2003)

Figure 4: Disequilibrium in the Market

Price

Qd1Quantity water

Qd QsQuantity water