Department of the Environment

About us | Contact us | Publications

About us header images - leftAbout us header images - centreAbout us header images - right

Publications archive

Disclaimer

Key departmental publications, e.g. annual reports, budget papers and program guidelines are available in our online archive.

Much of the material listed on these archived web pages has been superseded, or served a particular purpose at a particular time. It may contain references to activities or policies that have no current application. Many archived documents may link to web pages that have moved or no longer exist, or may refer to other documents that are no longer available.

Second Environmental Economics Round Table Proceedings

Convened by Senator Robert Hill, Minister for the Environment and Heritage, Canberra, 5 July 2000
Environmental Economics Research Paper No. 7
Commonwealth of Australia, 2000
ISBN 0 642 19485 8

Policy mechanism selection for environmental management

Gary Stoneham a
Department of Natural Resources and Environment, Victoria

 

Summary

Bad environmental policies can impose severe costs on the economy, reducing our living standards and often failing to achieve any real improvement in the stock or quality of the environment. This paper highlights several observations that arise form attempts to match the form of intervention with the objectives of environmental policy, the economic characteristics of the environment and the characteristics of the agents involved in environmental management.

Environmental management is a complex resource allocation problem. It is plagued by incomplete information, complex interactions between cause and effect and is burdened by responsibility for the well being of future as well as present generations. There are also many agents, with different characteristics, whose business and personal decisions change the quality or stock of environmental assets. The primary responsibility of policy makers is to meet society’s environmental goals but to do so at least cost and disruption to the wealth objectives of current and future generations. There are four ways to improve our performance in this respect.

1. Introduction

It is widely recognised that free markets fail to encourage adequate investment in environmental management and conservation. The main reasons for this include inadequate property right specification, information and attitude problems and the existence of non-priced goods and services. While environmental management involves complex resource allocation processes, useful insights can be made into the appropriate form of intervention by systematically examining the objectives of environmental management, the participants in the environment, the economic characteristics of the environment and the policy mechanisms available. This paper provides a brief discussion of these factors and their influence on the selection of policy mechanisms for environmental management.

2. The objective of environmental policies

It is not often acknowledged that bad environmental policies can impose severe costs on the economy, reducing living standards and often failing to achieve any real improvement in the stock or quality of the environment. The over-riding objective of most public policy programs is to improve the well-being of society. For ordinary goods and services this can be a complex task but environmental problems introduce many new dimensions such as sustainable development 26 One of the most important characteristics of environmental problems is that there is in-complete information with which to make choices. This occurs because of the complex spatial and temporal interactions between cause (e.g. land clearing) and effects (e.g. salinity) particularly where there are irreversible outcomes (e.g. extinction of species). Where these conditions arise, a Safe Minimum Standard (SMS) or cap approach is sensible (see Bishop 1978 and Holling et. al. 1995). For example, river water quality can be defined in terms of a cap on the salt or nutrient load that would be required to ensure the persistence of a viable aquatic habitat.

Where a SMS approach is appropriate, the criteria that can be used to assist in the choice of environmental policy instruments can be simplified to more tractable and familiar measures including:

3. Participants in the environment

The participants or agents involved in environmental management can be classified in three categories:

Figure 1 (Ingram 1998) illustrates that in the Upper Goulburn River, point-source emitters contribute 22% of the total phosphorus load (8% from fish farms and 14% from sewage plants) and non-point source emitters contribute 74% of the nutrient load (30% from dryland farms and 44% from irrigated farms). In Port Phillip Bay, on the other hand, around 45% of the nitrate load is contributed by point-source emitters (treated sewage), 42% by non-point source and 14% arises from atmospheric sources–mostly mobile emitters (motor cars).

The cost of achieving improvements in the environment varies considerably from firm to firm. In the Upper Goulburn River, for example, there are many dryland farms (contributing 30% of the Phosphorus load) that would have very low abatement costs. Many of these farms could make important contributions to reducing the nutrient load at very low cost. Policy mechanisms that allow these agents to participate in an 'environment economy' will significantly reduce the economic costs of achieving environmental goals particularly when there are high marginal abatement costs associated with the alternatives e.g. fish farms or sewage treatment plants. Similarly, in Port Phillip Bay, it would be important to develop policy mechanisms that allow point source, non-point source and mobile emitters to participate in the nutrient economy.

figure 1 phosphorous emission-Upper Goulburn figure 2-Nitrate emission Port Philip Bay
4. Economic characteristics of the environment

There are four specific characteristics of the environment that influence the selection of policy mechanisms.

4.1 Incomplete and asymmetric information

It is now recognised that many of the important issues facing modern economies can be presented as problems of sharing information in economic systems (Myerson 1999). The environment is clearly one of these problems. To make good environmental choices, a range of information is needed. Two general information problems can be recognised. The first has already been discussed in section 2 where in-complete information and irreversible outcomes suggest a Safe Minimum Standards approach. The second information problem is the asymmetric nature of information needed to make decisions about environmental management. Asymmetric information refers to those situations where one party does not hold all of the information needed to make an informed choice. This can be illustrated by thinking about the problem of dryland salinity where land-use change may be needed. It is clear that landholders, not government, know most about the opportunity cost of changing land-use from activities that cause environmental damage to those that don’t. On the other hand, government or other environmental stewards–not landholders, know most about the intrinsic value of environmental assets such as remnant vegetation and river water quality requirements to sustain native aquatic species etc. Mechanisms that do not reveal this information will not deliver cost-effective outcomes (see Latacz-Lohmann 1997).

4.2 Non-Standard environmental values

For many environmental problems, each unit of conservation effort yields benefits that do not have a standard value. For example, changing land use in different parts of Australia will generate very different environmental benefits. In some areas, there will be large benefits (to other farms) from recharge control, while others have only a localised impact on watertables (see Pannell 2000). Similarly, because of the diversity in habitat, habitat quality and species composition, there are non-standard benefits from activities that conserve remnant vegetation on private land. Recognition of the non-standard benefits nature of environmental management suggests that the location and type of intervention will be very important in determining the effectiveness and costs of environmental management actions.

4.3 Multiple Benefits

Many environmental benefits can arise from one change in natural resource management. Changing land-use on one area of land could yield benefits from weed controls, water table recharge, nutrient control, native vegetation conservation, carbon sequestration etc. These represent joint products and it is not sensible to separately fund the procurement of these benefits from the same area of land. In some, but relatively few, cases there are competing environmental outputs. Failure to anticipate and take account of multiple benefits will increase the overall cost of achieving environmental goals.

4.4 Non-market values

Many of the benefits from conservation of the environment, e.g. biodiversity, are not assigned money values in regular markets even though they may be highly valued by society. Markets typically fail to generate sufficient supply of these goods and services simply because they do not participate in market-based resource allocation systems.

5. Policy mechanisms available for environmental management

A number of policy mechanisms can be employed to achieve environmental improvements. These are briefly reviewed in the following sections

5.1 Command and Control

Command and Control (CAC), refers to the use of regulations to enforce specific actions or planning approaches to achieve environmental goals. For example, governments could legislate to force factories to achieve some benchmark greenhouse emission standard. These regulations impose the same standard on all individuals irrespective of the ability to meet these standards or the cost of doing so. For CAC measures to be cost-effective, governments would need to hold a complete set of information about the individual costs of production and preferences of each firm and individual in society. Governments would have to establish non-uniform (source-specific) standards to force firms to reduce emissions to the point where all firms incur the same marginal cost of control. 27 Clearly this approach will either have very high information and administration costs or will be ineffective because environmental decisions will be based on too little information. CAC can also promote perverse behaviour. For example, legislation to mandate the sale of less polluting motor cars can result in lower running costs encouraging individuals to increase car use. They can also provide incentives to misrepresent marginal abatement costs by producers in order to avoid being identified as low cost and therefore facing strict standards. Further, CAC mechanisms provide no incentive for individuals to continue to reduce emissions, or to invent new approaches once the standard has been attained.

5.2 Market based approaches

The key feature of market based approaches is that individuals who hold information on the marginal cost of environment improving actions make the choice about whether it is financially sensible to implement these actions. This contrasts with the command and control approaches where governments are confronted with the problem of discovering which activities and then which firms, could achieve the environmental goal at least cost. Markets essentially act as a means of exchanging information where individuals respond to the marginal costs and marginal benefits of a wide range of resource use decisions. Market-based approaches essentially rely on changing the incentives faced by individuals in the market (either marginal costs or marginal benefits) so that decisions are made to limit environmental damage.

5.2.1 Taxes and Subsidies

Taxes and Subsidies are per unit payments levied on/paid to, an undesirable/desirable activity respectively. Taxes aim to discourage unwanted behaviour (e.g. harmful emissions) by introducing the marginal social cost into the decision-making arena. For example, the introduction of a tax on emissions equal to the marginal damage cost of these emissions should limit emissions to the social optimum. However, government does not hold information on costs or preferences of producers (asymmetrical information, section 4.1). Therefore, the magnitude of the abatement response to a given per unit tax by different producers is unknown. The aggregate quantity of abatement is uncertain. Because of the non-standard environmental benefit problem (see 4.2) an efficient tax would need to set at non-standard rates to reflect the specific environmental damage being caused. Finally a tax does not place a limit on the total amount of environmental damage. As more firms enter a polluting industry tax revenue will rise but so will pollution. This is particularly important when the input causing environmental problems is not responsive to price changes (e.g. fuel). Subsidies have similar problems.

5.2.2 Tradable Emissions Permits

Tradable Emissions Permits are proving to be a very effective way of reducing the costs of achieving a specified emission target particularly for point source emitters. An increasing number of programs in the USA have used market based systems such as the 'cap and trade', Individual Transferable Quota (ITQ) and marketable development permits. Tradable pollution permits allocate the pollution control burden amongst firms or individuals by employing the market as the information exchange mechanism. In this sense, individual firms, not government, make the decision to reduce environmental damage based on the marginal cost of abatement. Some firms will make no change while others will be able to reduce environmental damage in very cost-effective ways and gain by selling pollution credits. These different responses by different firms simply reflect the fact that there is considerable variation in the cost of abatement between firms. The tradable emission market is constructed to allow the economy to discover these differences in abatement costs and take advantage of abatement actions that are low cost.

A tradable permit approach to environmental control is based on four key elements. It is necessary to firstly identify an acceptable level of environmental damage or exploitation of a natural resource. This is often called a Safe Minimum Standard (SMS) or a cap. The cap represents a translation of the values of society (e.g. a need to reduce acid rain or greenhouse gas) into a limit on environmental damage, emissions or natural resource use. The second element is to specify a legal partial property right that can be held as private property. This represents a share of the cap and is a tradable right to pollute or exploit some resource. Once the cap is identified it is necessary to assign property rights to individuals. Property rights can be distributed in several ways including grandfathering or auctions. Finally, a mechanism to allow trade in property rights between individuals is required. Trade simply allows the market to discover those firms that can make environmental improvements at the lowest cost. There are many important observations to be made about successful tradable permit schemes. These include: a requirement for low transaction costs; tradable emission permits are only suited to point source emitters; a preference for initial emission rights to be auctioned; the importance of speculation; a requirement for minimal government interference; and a need for clear specification of partial property rights over resources 28.

A tradable permit approach offers significant efficiency and cost advantages. The cap will prevent the aggregate load (concentration) of pollutants rising above the acceptable level of exploitation as determined by society. As noted in section 2, this is a useful approach when there is in-complete information and where there is the risk of irreversible outcomes. The definition of legally backed private property rights over the right to pollute creates a situation where producers view this right as an essential factor of production which must be taken into account when making profitable production decisions. Allowing trade in a well defined and easily accessible market reduces the asymmetrical information problem by legally requiring emitters to hold a permit in order to pollute, and thereby encourage firms and individuals to reveal their marginal cost of abatement by bidding for the right (the permit) in a competitive market. The market will increase the probability of achieving a given environmental improvement at least cost by revealing the large differentials in abatement costs across polluters. Environment improving effort will occur where the per unit cost of environmental improvement is least (ie: the return on control effort is highest).

Perhaps the most successful tradable permit scheme is the Sulphur Dioxide scheme operated in the USA. This program operates under the 1990 Clean Air Act in the U.S and is administered by the EPA. It has been estimated that the tradable permits approach to Sulphur Dioxide control has reduced the costs of achieving specified air quality targets set by the EPA by $225-$375million (Schmalensee et. al. 1998). A number of tradable permit programs have now been established around the world, particularly in the United States. These include: Boulder Creek Colorado (ammonia), Chatfield Reservior (phosphorous), New Jersey Chemical Industry Pilot Project (arsenic, cadmium, copper, lead, mercury, nickel, zinc), Tar-Pamlico River Basin Nutrient Reduction Trading Program (Nitrogen), Cherry Creek Basin (phosphorous), Lake Dillon (phosphorous), the EPA’s Emission Trading Program for regulation of air quality and the Wisconsin system of tradable discharge permits (water quality), Collins and Smith (1998).

There are two aspects of tradable permit schemes that are of concern. The first stems from the problem that unrestricted trading of pollution permits can cause hot-spots–places where pollution is higher than the required standard even though overall the level of pollution is acceptable. This occurs because participants trade pollution permits on the basis of marginal costs of abatement rather than on marginal damage costs. Emissions in regions with high marginal abatement costs will tend to increase while emissions in regions with low marginal abatement costs will decline (see Tietenberg 1995). The Sulphur Dioxide trading program in the USA limits this by requiring that trades not lead to exceedance of national ambient air quality standards. Considerable effort has been directed at refining the design of market-based approaches for environmental management to minimise the probability of hotspots occurring. The second concerns the implications of uncertainty in marginal benefits particularly whether the covariance between marginal abatement cost and marginal damage is positive or negative (see Stavins 1996).

5.2.3 Auctions as an environmental policy mechanism

While auctions have been used as far back as 500 BC their application to natural resource allocation has been a relatively recent innovation. The highly successful auctions of mobile phone spectrums in the US and the UK are examples of the application of contemporary economic theory to auction design (Myerson 1999). Auctions are an important economic instrument because they facilitate information revelation from bidders in the auction. In the case of the mobile phone spectrum auctions, multiple-round, simultaneous bidding processes were used to reveal bidders’ valuations of the spectrum. Auctions also have important applications to natural resource problems. There are close similarities between the allocation of access to mobile phone spectrums and the allocation of access to stocks of fish, native timber resources and mineral deposits. Besides allocating access on a competitive basis, a well designed auction will in theory impose a 100% tax on monopoly rents, thereby returning resource rents to government.

Latacz-Lohmann and Van der Hamsvoort (1997) have shown that auctions are also an efficient mechanism for achieving land-use change for environmental goals. The Conservation Reserve Program (CRP), operated by the USDA, uses an auction mechanism to award land management contracts through a competitive bidding process. Unlike the mobile phone and resource stock auctions, land-use change auctions have a price minimisation objective. In Australia, this mechanism has many potential applications where land-use change is required to achieve environmental improvement. Salinity control, nutrient control and conservation of remnant vegetation will all require some form of land-use change. These problems involve asymmetric information problems. That is, landholders know most about opportunity costs if land were to be changed from agricultural production to activities that improve the environment. They understand the financial implications including implicit values for risk, lifestyle factors etc. that would need to be taken into account if land-use were changed. Landholders may also hold implicit values for conservation such that they would be prepared to share costs of making environmental improvements. Government, on the other hand, holds information about the relative value of various elements of the natural estate. Government departments of conservation, for example, hold inventories of flora and fauna location and quality. Good management decisions about land-use change for environmental or biodiversity conservation reasons, will depend on the revelation of information from both private landholders and government departments. Auctions are a policy mechanism that can be designed to specifically reveal this information from individuals and government needed to make good resource use decisions. Because auctions engage individuals in a competitive bidding process, environmental goals can be achieved at least cost–the economic efficiency objective.

There are essentially two aspects of an auction-based approach to land-use change: Individual Management Agreements and a competitive bidding process. Landholders bid in an auction for compensation to make some land-use or management change that has a beneficial impact on the environment (e.g. active management of remnant vegetation). As noted above, the CRP uses a relative preference measure called an Environmental Benefits Index to denote the non-market value of the specific changes included in the Individual Management Agreement. Landholders develop bids that indicate the need to be compensated to make the land-use change. These bids are made in competition with other landholders so that any rents are bid away. That is, successful bidders would receive just enough, but not excessive compensation for making land-use change and would not make excessive profits from the compensation. In fact, an auction would seek to take advantage of land-holders that place bids at less than their own commercial opportunity cost because they wish to make land-use change for other reasons (e.g. private conservation values). Auctions reward cost sharing behaviour.

Of all economic instruments, auctions readily accommodate the non-standard benefits problem. That is landholders bid on the basis of an individual land-use change contract (Individual Management Agreement) and if successful would contract to change land-use in a specific way for a specified period of time. The contract could involve actions such as fencing an area of remnant vegetation to keep stock out, or could involve revegetation of an area identified as high water table recharge. The auctioning organisation would then rank the bidders from the best environmental value for money to the least value for money and sign contracts with the successful parties. In effect, this creates a supply curve of environmental goods and services.

There are many potential applications of auction based approaches to environmental problems in Australia. Specifically, auctions are techniques that are well suited to non-point source environmental problems particularly those associated with inappropriate land use. Considerable effort has already been devoted to the design and development of an auction-based approach to conservation of remnant vegetation in Victoria and the MDBC is now funding a study on auction design for multiple benefits including dryland salinity29. There are compelling reasons to consider the use of auctions for a broader range of environmental problems that involve land-use change.

5.3 Information, education and persuasion

The final policy mechanisms considered in this section are those of information, education and persuasion. Raising the level of human capital is one way of improving environmental outcomes. Inventing more efficient, less polluting motor cars could be one approach to problems associated with air pollution. Promotion, awareness and attitude change are also necessary and effective elements of the policy portfolio. These mechanisms raise human capital generally in the economy with a view to improving the resource use decisions.

6. Which policy mechanisms?

The environment has many similarities with other economic management problems. We want to achieve the environmental outcomes required with as little disruption to current and future standard of living as possible. In another sense however, the environment is a much more complex and challenging public policy problem. Two factors seem to be important in determining which policy mechanisms suit different environmental problems. The characteristics of participants involved in environmental management–whether they are point source, 'non-point source or mobile emitters and the economic characteristics of the environment particularly with respect to incomplete and asymmetric information problems, non-standard benefits, multiple benefits and non-market values. The objectives of environmental policy noted above can only be meet if policy mechanisms are capable of engaging the relevant agents and if they are capable of dealing with the specific characteristics of the environment. Another important consideration in selecting policy mechanisms is the interaction with technical change. The theoretical literature argues that the policy instrument chosen should make a difference in the rate of technical change. The cost saving expected from instruments that encourage technical change are large (see Dowlatabadi 1998 and Tietenberg 1999).

It is clear that no one policy mechanism will be capable of dealing with most environmental problems. Different policy mechanisms will be suitable for certain aspects of environmental management and will be capable of engaging some of the agents that have an impact on environmental outcomes. It would be appropriate, for example to utilise command and control approaches where the optimal level of emission is zero–e.g. where highly toxic substances are produced (e.g. nuclear radiation) and no emissions could be tolerated, see Portney (1978). Command and control approaches will not, however, achieve cost-effective environmental management where there is incomplete and asymmetric information, where there are non-standard and multiple benefit problems or where there are non-point source and mobile emitters that are difficult to monitor. Tradable permit schemes have had undoubted success in reducing the costs of emissions from point source emitters. However, where environmental problems arise from both mobile and non-point emitters, tradable emission permits will not be suitable. It is generally concluded that the transaction costs (identification of the emitter, administration costs, policing and monitoring, for example) associated with including all emitters (point source, mobile and non-point) in a tradable emission scheme are likely to be prohibitive. Auctions appear to be a cost-effective and practical approach where non-point emitters exist. In these circumstances individual management agreements can be developed to identify the specific changes needed from each firm or individual involved (non-standard benefits). Auctions reveal information from all parties needed to make good environmental decisions (asymmetric information) and can be designed to incorporate multiple benefits and non-marker values. Other approaches will be needed for mobile emitters and information education and persuasion will always be an important component of a balanced environmental management portfolio.

In Australia, it appears that we have relied too heavily on too few environmental policy instruments. Improvements can be made in three directions. Firstly we can improve the range of policy instruments available. In this paper, it has been shown that tradable emission permits (TEP) and auctions are new and important additions to the policy instrument tool kit. The second improvement that can be made is in the selection of instruments to ensure that they reflect the characteristics of the environmental problem, the characteristics of the agents involved and harnesses rather than excludes the role that technology can play in addressing environmental problems. TEPs and auctions are useful additions for certain environmental problems and for certain categories of agents. Finally, we can make improvements in the way that different policy tools fit together. This problem can be viewed broadly, for example to ensure that legislation on native vegetation clearing interacts with market based mechanisms as well as within the broad families of policies.

There is also scope to improve integration between the market-based instruments. Stoneham and Chaudhri (2000) for example, have proposed linking of tradable emission market with auctions of land-use change to bring both point and non-point emitters into one ‘environment economy’. Figure 3, illustrates how such a linkage could be made. In this diagram, markets for point source and mobile emitters are created (e.g. carbon, nutrients etc.) through tradable permits. Non-point source emitters are included in the system through auctions of land-use change. For environmental goods and services, where markets do not exist, (e.g. biodiversity from native vegetation on private land) governments signal the relative importance of the various environmental assets using an environmental benefits index. Landholders would construct their 'need to be compensated' bids in light of the offsetting returns they would expect to receive from the generation of emission credits (carbon, salt, nutrients), the return they would expect from the generation of ordinary goods and services (e.g. timber, firewood etc.) and the non-market benefits signaled by an Environmental Benefits Index. Where these benefits arise from one land-use change (multiple benefits), landholders could place very competitive bids because some component of land-use change is funded through the sale of emission credits and the sale of ordinary goods and services. By linking auctions with tradable emission permits, the funding burden for government is reduced. Private firms wishing to expand production and therefore emissions, would either purchase emission permits from the emission market, or will fund the creation of emission credits through land-use change auctions if this represents a less costly strategy. In effect the traded price of emission permits sets a reserve price in the land-use change auction. Cost-effective environmental policy mechanisms could also be attractive for NGOs and private businesses that wish to achieve environmental gains. An efficient set of policy tools, particularly tradeable emission permits and land-use auctions would offer a cost-effective, highly accountable and transparent approach.

 

figure 3 Coordinating policy mechanisms

References

Bishop, J. (1978), ‘Endangered Species and Uncertainty: The Economics of a Safe Minimum Standard’, American Journal of Agricultural Economics, vol. 57, no. 1, pp. 10-18.

Bishop, R.C. (1992), ‘Option Value: An Exposition and Extension’, Land Economics, 58, pp1-15 Collins, D. J. and Smith, S.A.Y. (1998), ‘Economics Instruments in Environmental Regulation–experience by NSW EPA’.

Dowlatabadi, H. (1998), ‘Sensitivity of climate change mitigation estimates to assumptions about technical change’, Energy Economics 20, pp. 473-493

Holling, C.S., Schindler, D.H., Walker, B.W. and Roughgarden, J. (1995), ‘Biodiversity in the functioning of ecosystems: an ecological synthesis’ Chapter 2 in Biodiversity Loss: Economic and Ecological Issues, Perrings, C., Maler, K., Folke, C., Holling, C.S. and Jansson, B. (Eds), Cambridge University Press.

Ingram, B. (1998), ‘Audit of Phosphorus effluent from trout farms in the Goulburn-Broken catchment’, A report to Fisheries Victoira.

Latacz-Lohmann, U. and Van der Hamsvoort, C. (1997), ‘Auctioning conservation contracts: A theoretical analysis and application’, American Journal of Agricultural Economics, vol. 79, no. 2, pp. 407-18.

Myerson, R.B. (1999), Nash equilibrium and the history of economic theory, Journal of Economics Literature Volume 37, pp. 1067-1082.

Pannell, D. (2000), ‘Dryland salinity in Western Australia: are externalities the main cause? Paper presented to the annual conference of the Australian Agriculture and Resource Economics Conference, Sydney.

Portney, P.R. (1978), 'Toxic substance policy and the protection of human health' In Current Issues in U.S. Environmental Policy, edited by P.R. Portney, Baltimore: Johns Hopkins Press, 1978.

Schmalensee, R., Joskow, P., Ellerman, D., Montero, J and Bailey, E. (1998), 'An interim evaluation of Sulphur Dioxide Emissions Trading', Journal of Economic Perspectives 12 (3) pp. 53-68.

Stavins, Robert,N. (1996), ‘Correlated uncertainty and policy instrument choice’, Journal of Environmental Economics and Management, 30 pp. 218-232.

Stoneham, G. and Chaudhri, V. (2000), ‘The role for auctions in tradable emission markets’ Paper presented to the annual conference of the Australian Agriculture and Resource Economics Conference, Sydney.

Tietenberg, T. (1995), ‘Tradable permits for pollution control when emission location matters: What we have learned?’ Environmental and Resource Economics, 30, pp. 95-113.

Tietenberg, T. (1999), ‘Design issues seeking research answers’, Paper presented to Conference on research frontiers in GHG emissions trading, Resources For the Future , Washington DC.

Footnotes

a The author acknowledges the input of Charlotte Duke (NRE, Economics Branch) and Dr Vivek Chaudhri (University of MElbourne, Business School).

26 The Brundtland Report identifies two aspects of sustainable development: concern for the future generation's ability to consume and concern for the stock of biodiversity.

27 Efficiency maximisation and cost minimisation requires equalisation of the per unit cost of controlling pollution between firms, not equalisation of quantity of pollution control between firms.

28 See papers from the Workshop on market-based instruments for environmental protection (1990), The Association of Environmental and Resource Economists, John F. Kennedy School of Government, Harvard University.

29 A joint project by NRE (Stoneham) Melbourne University (Ass. Prof. Chaudhri) and the USDA.