Thoughts on economics and liberty

COASE AND WATER – Nicola Tynan

6  COASE AND WATER – Nicola Tynan

Water is a scarce resource. While this may sound obvious today, a century ago it was not. A recurring criticism of London’s private water companies during the nineteenth century was that they failed to provide a sufficiently large quantity of water for flushing and street-cleaning at a time when water was considered unlimited in supply, if not from the Thames then from Wales. Globally, the majority of institutions for water resource allocation were developed on similar assumptions, though the institutional details differ between countries and even regions within countries (Glennon 2009: 122). For water in rivers and lakes, legal institutions implicitly assumed that surface water could be allocated to consumption uses without a negative impact on the quantity or assimilative capacity of the instream water. Similarly, groundwater rights were often tied to land ownership on the assumption that water withdrawals would remain below recharge rates so use by one landowner would not negatively impact a neighbour.

Though not universally true even in the nineteenth century, for many places the assumption of unlimited water resources was reasonable. More recently, the tide has turned. Driven largely by population growth, water use has increased and water stress – defined as withdrawal in excess of available renewable supply – has increased globally, not only in arid and semi-arid regions.1 Groundwater aquifers are being depleted in many locations – from Mexico, where groundwater pumping has resulted in a clearly observable sinking of Mexico City, to India, the world’s largest groundwater user.2 Shortages of surface water are causing more frequent conflict in the western and southern US and more frequent water use bans in the UK. We now have to face the reality that one person’s use of water often has a negative impact on others, either today or in the future, and, increasingly, the impact is being felt today. Flush toilets capture this change in our understanding of water. We have moved from the competition between toilet brands on the basis of how much they could flush – with names such as ‘Niagara Falls’, ‘The Deluge’ or ‘The Dreadnought’ – to today’s low-flush toilets competing on how efficiently they can flush using the least amount water. Flushing accounts for nearly one third of domestic water use. Households with water meters who pay a volumetric fee have an incentive to reduce water use; low-flush toilets can deliver long-term savings at relatively low cost. Water meters help us move closer to full-cost pricing for domestic water.

  • For a map of global water stress, see http://w w w.w aqueduct/aqueduct-atlas (accessed 3 July 2015).
  • World Bank (2010). For example, according to the Water Governance Facility (2013: 5) ‘governing the groundwater has become a growing challenge in large parts of the country where the water table is steadily sinking’.

It takes time to change indoor plumbing. It takes even longer to change long-established institutions. The problem presented in Ronald Coase’s ‘The problem of social cost’ (1960) is one where the actions of one user have harmful effects on others but where the relevant costs to be considered are the joint costs of preventing the harmful effects.3 This is exactly the situation we face with water resource management today. In improving water resource management policies, we need to determine whether ‘the gain from preventing the harm is greater than the loss which would be suffered elsewhere as a result of stopping the action which produces the harm’ (ibid.: 27). Full-cost pricing and clearly defined rights for all water resources can help make this determination.

Clearly defined property rights

One challenge facing the water sector globally is the weak or now inappropriate definition of property rights. Coase explained that, as long as property rights are clearly defined and transactions costs are low, market transactions will result in the most efficient outcome. He also argued that clearly defining property rights and reducing uncertainty will itself reduce transactions costs.

For surface water, property rights tend to be use rather than ownership rights, often connected to land-ownership (riparian), first use (prior appropriation) or state licence.

  • For further discussion of this tenet of Coase’s argument, see Veljanovski’s introduction to this book.

Even in the western United States, where property rights are seemingly well-defined under a prior appropriation system, there is significant uncertainty because rights were over-allocated in ways that make it uncertain who has the right to use water from a particular source, and there was a failure historically to recognize the value of instream flows. The problem is exacerbated in locations where surface water rights are not tradable, which prevents them being transferred to the highest value user.

Groundwater is more frequently connected to land ownership, in some places as use rights, in others as full ownership rights to the water. Because the difference did not matter too much when withdrawals were below aquifer recharge rates, there is often uncertainty regarding water rights. Robert Glennon highlights this uncertainty within the western US (Glennon 2009: 128):

Property-rights advocates often argue that property owners have an inherent right to drill wells on their property. Restrictions on this right, it is claimed, would violate the takings clause of the U.S. Constitution and require government compensation. But groundwater is not a private resource owned by the overlying landowner. It’s a public resource owned by the state. Citizens can use it, but use rights differ profoundly from ownership rights.

Even use rights can call for compensation if restricted in ways not allowed for in advance. A bigger barrier to the creation of water markets and compensation through the purchase of water rights results from the requirement, in many locations, that landowners use their water themselves.

Water is essential for life. This fact underlies the resistance to water pricing and water trading that has resulted in numerous books, documentaries and public protests attacking ‘water commodification’ in recent years. Making sure that everyone has access to sufficient clean water for survival and general well-being is a crucial policy goal. It is also a goal that some countries have failed to achieve under any institutional structure for water provision. Private participation in domestic water provision has been introduced to improve quality, extend access and improve the efficiency of failing utilities. While poorly implemented policies have made access to water more difficult for some people and communities, private participation elsewhere has improved access for many. In all cases the problem, as explained by Coase, ‘is to devise practical arrangements which will correct defects in one part of the system without causing more serious harm in other parts’ (Coase 1960: 34). To do this we need to ‘compare the total social product yielded by these different arrangements’ rather than focus on a less-than-ideal outcome in one part.

Some instances of conflict surrounding the introduction of private participation in water treatment and distribution have resulted from either ill-defined rights to water or water rights defined in such a way that prior users of water are excluded without compensation. This was a major issue in the notorious case of water privatisation in Cochabamba, Bolivia. Textbook explanations of the Coase Theorem often focus on Coase’s examples of low transactions costs where the initial allocation of property rights does not influence the efficiency of the outcome. For water resources, however, transactions costs can often be high. Where transactions costs are high, Coase argues that ‘the initial delimitation of legal rights does have an effect on the efficiency with which the economic system operates’ (ibid.:16). A human right to a limited quantity of water for essential domestic uses is not incompatible with pricing water in the majority of uses. Indeed, pricing water to prevent it being wasted in low-value uses today may be essential to ensure its availability for higher-value uses in the future.

one consequence of the perceived abundance of water is that water is often treated as a free good, with charges being made only for the infrastructure, energy and other operational costs of treating and transporting it, often with energy costs also subsidised. Clearly defining property rights will raise the price of water in ways that reflect its value as a scarce resource. Pricing water serves to generate information on the value of water in alternative uses, providing information on the cost of replacing one use of water with another. It also gives current holders of rights to use water an incentive to conserve and transfer their rights, increasing transparency while potentially reducing resistance and conflict. Recognizing the role of transactions costs means that it is not only important to determine property rights but also to think about how these rights are assigned. This makes the problem ‘one of choosing the appropriate social arrangements for dealing with the harmful effects’ that will likely differ across countries or watersheds (ibid.: 18). The appropriate social arrangement should be the one that operates at lowest cost when all costs are taken into account. What works as an appropriate social arrangement at one period of time with a given population and technology may not be appropriate at a later date with a larger population, living more densely in urban areas, higher standards of living and new technologies.

Integrated water resources management

Water resource institutions are facing a time of change. Internationally, there has been a move towards integrated water resources management (IWRM) as recommended by the Dublin Statement of the 1992 International Conference on Water and the Environment. IWRM focuses on managing water resources in ways that are economically, socially and environmentally sustainable. Importantly for water markets, the Dublin Statement explicitly recognised water as an economic good in all its uses. This approach fits with Coase’s emphasis on total social benefits: where overlapping legal jurisdictions draw water from the same basin, IWRM focuses on water basin benefits rather than individual user, community or even country benefits (Sadoff et al.: 26–27).

The Dublin Statement recognising water as an economic good noted that ‘access to clean water is a basic right of all human beings’, highlighting the positive connection between treating water as an economic good and improving access to clean water for the poor. Whatever their income level, people are willing to pay relatively high prices for the first litres of water they consume. In most places today, marginal user values for water are much higher for municipal and industrial uses than for agriculture. While many water trades take place between those with similar uses, for instance between two farmers, the fact that water is currently used for low-value agriculture while high-value domestic uses are not satisfied means that more extensive water markets are likely to see water move away from agriculture to domestic, industrial and instream uses.

A number of countries have already adopted IWRM, including developing markets for water trading to various degrees. A recent study (Grafton et al. 2011) compares the performance of water markets in five countries: Australia, Chile, South Africa, the US and China. These are all places sharing the following characteristics to varying degrees:

(1) they are semi-arid regions either experiencing or facing the threat of water shortages; (2) water has different values across uses; and (3) there is sufficiently strong institutional governance and legal capacity allowing for broadly accepted reform. As the authors argue, none of these countries score equally well on measures of efficiency, equity and environmental sustainability and all have room for improvement. However, they each do some things well and can provide guides, if not models, for policy makers elsewhere.


Australia provides a model of a country that has embraced full-cost water pricing comprehensively, while recognising that institutional details will need to adjust over time. In June 2004, the Council of Australian Governments (CoAG) signed the Intergovernmental Agreement on a National Water Initiative (NWI) and established a National Water Commission. With a goal to ‘improve the management of the nation’s water resources and provide greater certainty for future investment’, the NWI built on the prior experience of water rights trading within the Murray-Darling Basin and explicitly embraced water rights, water trading and improved water pricing. Under the NWI, each state or territory is required to clarify and improve the certainty of water rights and to maintain a registry of water titles recording access entitlements, ownership and transfers. on the 10th anniversary of the NWI, Australia’s National Water Commission stated that ‘although the full extent of the National Water Initiative’s aspirations is yet to be realised, we have a framework that 10 years on, is proven and robust.’4 This assessment is reflected in the relatively high scores that Australia receives in Grafton et al.’s integrated assessment (Grafton et al. 2011: 222, 229, 232).

The need for institutional reform in the management of water resources was first acknowledged within the Murray-Darling Basin in the 1980s through an embargo on new licences and projects to replace open channels with pipelines for the delivery of irrigation water. As a result of this early start, water markets are well-established within the Murray-Darling Basin. Entitlements to water from the

http://w w i/nw i-10 -year-anniversar y (accessed 3 July 2015).

basin may be either high reliability, where rights holders can expect to receive their full allocation each year, or low reliability with the possibility of no allocation in dry years. Trades may take one of two forms: permanent transfer of the water right or transfer of a single year’s water allocation. The Murray-Darling Basin has experienced substantial trading, with about 20% of water rights traded at a value of $1.8 billion in 2009 (ibid.: 229–30). A number of brokers operate in the market to reduce transactions costs.5 New trading rules introduced on 1 July 2014 aimed to reduce uncertainty by requiring the reporting of all trade prices and limiting restrictions on trade to four clearly stated circumstances, including impacts on third parties.6

Australia’s NWI has struck an appropriate balance between security of water rights and adaptability to changing circumstances. As statutory rights, water rights can be modified by state governments without compensation. In practice, governments have purchased water rights to increase environmental flows. Such commitments to compensate rights holders are clearly stated in the recent Intergovernmental Agreement and National Partnership Agreement for the Murray-Darling Basin agreed between Commonwealth and New South Wales in February 2014. These agreements state the need for an additional 2,750

gigalitres of water to remain in the Basin annually for ecosystem protection; they agree that the additional instream flow will be achieved through a combination of infrastructure and environmental works aimed at water recovery plus the purchase of water rights up to a maximum of 1,500 gigalitres.7 The state’s strategy has been to purchase permanent water rights to protect instream flows when necessary but to sell temporary use rights when water is surplus to environmental needs.

United States

California’s three-year drought, ongoing in 2015, renewed criticism of the system of rights based on seniority rather than highest value. While failure of the existing system of water rights is recognised and water trading is well established in some places (resulting in market transactions with a value of over $3 billion between 1987 and 2007), in other parts of the state there is strong resistance to moving to full-cost water pricing and transferable water rights (Anderson et al. 2012: 24). Rather than seeing this as an opportunity to transfer water to its highest-valued uses, those who currently hold senior water rights fear that institutional change will result in a loss of rights to water without compensation. Clarifying water rights requires information on how much water is actually used by rights holders. In California’s Central Valley, where over half the

irrigation water comes from wells, some farmers explicitly resist water metering from a fear that this will allow the state to restrict the amount of water they pump, again without compensation.

New housing estates increase water demand. Recognising that Utah’s water was over-appropriated, possibly by as much as 45%, Jerry olds, the state’s engineer from 2002 to 2008, stopped issuing new permits for some basins in the state. He also defined property rights more precisely, to allow transfer and sale of water rights, and to tie development approval to water rights (Glennon 2009: 234). New users, particularly developers, are now required to obtain water rights from those with existing claims. These constraints have not caused Utah to stop development but have required developers to ‘purchase and retire some other water user’s right’ showing that the development is a higher-valued use (ibid.: 237). Although rights transfers can involve significant transaction costs when individual developers are required to seek bilateral deals, because of search costs and uncertainties in the approval process, brokers and an exchange have arisen to lower these costs. This would not have surprised Coase, who encouraged economists to ‘study the work of the broker in bringing the parties together’ (Coase 1960: 18).

The New Mexico city of Santa Fe followed a similar policy requiring developers to acquire water rights from a willing seller before requesting a building permit. In response to developers’ concerns that they would pay for water rights but then might not receive the permit to build for other reasons, the city of Santa Fe established a water bank that allows developers to deposit water rights for future projects (Glennon 2009: 240).

While the western US is making some progress towards integrated water markets the assessment by Grafton and colleagues shows that much remains to be done. In the fast-growing southeast, the need for institutional change has, for the most part, been ignored. Residents of coastal Georgia and South Carolina draw groundwater from the Upper Floridan Aquifer. Heavy pumping in Savannah, Georgia, has reversed the flow of groundwater resulting in salt water contamination of domestic water supply. The state’s Environmental Protection Agency responded with regulations in 2006, 2008 and 2013 to reduce withdrawals by existing permit holders and place a moratorium on additional permits, but did not move towards tradable water rights.

Further south, Florida, Alabama and Georgia have been fighting over water from Lake Lanier since 1990 in what has become known as the tristate water war. Property rights to water from Lake Lanier are ill-defined, giving the city of Atlanta no economic incentive to limit extractions. Even though the water is crucial for ‘sustaining Florida’s $134 million commercial oyster industry’, Florida fishermen have no way to compensate Atlanta for allowing water to continue into the Cattahoochee River from Lake Lanier (ibid.: 29).

Critics of water trading argue that the environment will be the loser as the rich will pay to take water for wasteful purposes. Terry Anderson of Montana’s Political Economy Research Center (PERC) shows that this is not the case even when instream flows are not explicitly protected as they are for the Murray-Darling Basin. Water rights trading allows environmental groups to purchase water rights to protect or enhance instream flows.8 For example, in 2006 the oregon Water Trust kept water in the John Day River to protect Chinook and steelhead salmon by purchasing water rights from a local ranching family (Anderson et al. 2012: 11). By contrast, the regulatory approach creates uncertainty for both rights holders and those wishing to protect fish habitats. In 2014 California’s Water Resources Control Board implemented regulations limiting water use during the summer months to ensure sufficient instream flows for fish in the Sacramento River, curtailing farmers’ rights to water and generating threats of lawsuits against the Board.

PERC’s research highlights the role of ‘enviropreneurs’ – Coase’s brokers – in identifying environmentally beneficial gains from trade and bringing together buyers and sellers. As the value of maintaining instream flows increases, farmers who hold transferable water rights will be encouraged to conserve water to sell some of their allocation and, in some cases, may no longer farm their land (Coase 1960: 4). In Arizona, the Yuma Desalting Plant was completed in 1992 to treat agricultural return flows and reduce the salinity of water in the Colorado River flowing into Mexico. Rather than operate the plant, however, it was cheaper to divert the saline water and obtain flows

-ronald-coase (accessed 3 July 2015).

for the Colorado by paying farmers to fallow unproductive fields (Glennon 2009: 149). Water banking offers another way to realise these gains from trade: senior rights holders can ‘bank’ water that would have been applied to low-value uses, allowing it to be purchased by environmental organizations for higher-valued instream use (Anderson et al. 2012: 8).

South Africa

Before 1998, water rights in South Africa were not clearly defined but were generally connected to land as riparian rights or rights to drill wells. The National Water Act 1998 (NWA) introduced a system of public trusteeship combined with private use rights allocated through licences. South Africa’s water policy is best known for its formal recognition of a right to sufficient water for domestic purposes, a right included in sections 2 and 4 of the NWA.9 While this right to clean water for basic needs has not been achieved for all citizens, there has been significant improvement following the reform of water institutions. According to the World Bank’s World Development Indicators, the percentage of the rural population with access to improved water sources increased from 65% in 1995 to 88% in 2012.

In introducing a radical redefinition of water rights, South Africa recognised that the transition to new institutions can impose losses on some individuals despite an

  • Water supply sufficient for domestic purposes is defined as ‘25 litres per person per day accessible within 200 metres’ (Pienaar and van der Schyff 2007: 185).

overall social gain. With the state now formally custodian of the country’s water resources it has the power to award water use rights, but built into the NWA is the requirement that the state cannot take away water rights without due process and cause. To ease the institutional transition, the NWA included a right to compensation for prior owners of water rights who were negatively affected by the change (Pienaar and van der Schyff 2017: 187).

Based on an integrated water resources management approach, South Africa’s NWA requires that water leases take into account environmental protection. While the NWA was understood to allow for the trading of water leases there is significant uncertainty about the legality of individuals selling or otherwise transferring rights to water. For example, a North Gauteng high court ruling in August 2011 approved the transfer of water rights between farmers but the transfer had been denied by the Minister of Water and Environmental Affairs with subsequent appeal to the Water Tribunal delayed due to the tribunal’s suspension. Such uncertainty over the legality of transferring water rights means that trades have been few, particularly those transferring water between uses (Grafton et al. 2011: 229).

Chile, China and India

Institutional reform of water markets is being undertaken to various degrees elsewhere. Chile has the longest experience of water rights and is often used as a model for water market reform. Strong private property rights in water were established with Chile’s 1981 Water Code and updated with the 2005 Water Code Reforms. Chile’s reform was undertaken before and outside of an IWRM approach. This has resulted in property rights to water that have fewer restrictions on use and transfer than in other countries, generating a fairly substantial market for water rights but less consideration of third-party effects (ibid.: 229, 232). Water rights in Chile have allowed water to move to some higher-valued uses, particularly mining, but concerns remain that water is not going to its most valued uses now that Chile’s five-year drought has reduced overall water availability. Recent calls for reform within the country remind us that in making changes ‘the total effect of these arrangements in all spheres of life should be taken into account’ (Coase 1960: 43).

China has made moves towards allowing trade in water rights in its Water Law of 2002, focusing on trades between municipalities.10 Trades have taken place at directly negotiated prices rather than prices set by market transactions. The Yellow River Conservancy Commission has had the most success in implementing reforms to limit water withdrawals, largely due to an improved monitoring system. Even here, municipalities often violate their limits or withdraw water from tributaries before it reaches the Yellow River rather than purchase rights from other municipalities.

India’s increasingly severe water shortages, considered a crisis by the national government, is driving a review of water institutions (Water Governance Facility 2013: 11). The Supreme Court’s recent interpretation of the public trust doctrine identifies the state as responsible for water as a natural resource, despite the common law tradition that landowners have unlimited rights to extract groundwater from beneath their property. In India, water regulation is the responsibility of states, so the central government issues Model Bills as guidelines. The latest 2011 Model Bill includes a right to water of acceptable quality, specifying 70 litres per capita per day as a minimum, and recommends a separation of land and groundwater rights. The federal government’s 2012 National Water Policy and 2013 draft Framework Law on Water provided further nudges to reform. Despite central government encouragement, however, few states have taken steps towards water markets and IWRM. The state of Karnataka introduced a Ground Water Act in 2011 requiring the registration of existing wells and prior permission for all new wells but a perceived lack of legitimacy has resulted in low levels of compliance. The act was modelled on prior Model Bills so did not include aspects of IWRM included in the 2011 Model Bill. This may partly be due to uncertainty created by the Model Bill itself: the bill has existing water rights expiring after one year but without compensation for lost rights, creating uncertainty that will generate resistance within states considering adopting such regulation. It also fails to make clear whether the trading of water rights is allowable.


The move to IWRM and an acceptance of more clearly defined property rights, water pricing and water markets is happening slowly but the idea has gained a foothold. The countries discussed above are not the only examples; within Europe, Spain’s 1999 Water Law Reform opened the door to water rights trading. The institutional details necessarily differ across countries. This is beneficial because countries differ in terms of water resources, existing institutions and in many other ways that will require different social arrangements to achieve the largest social product. It is also beneficial because competition between, or at least a comparison of, different institutional details provide the information that makes innovation and learning possible.

In the introduction to this book, Veljanovski notes that it took 67 years from Coase’s work for the United States FCC to adopt a spectrum market. Spectrum markets now have broad acceptance although, as expected, the institutional details differ across countries. Applying Coase’s insights and using markets for water resource management faces even stiffer political challenges, but it has the potential to deliver crucial social and environmental benefits.



Sanjeev Sabhlok

View more posts from this author