Someone has to pay … any volunteers? Voluntary versus regulatory approaches to environmental protection in agricultural landscapes of Australia David J Pannell School of Agricultural and Resource Economics, University of Western Australia, and CRC for Plant-Based Management of Dryland Salinity

Abstract

A range of policy mechanisms for promoting changes in private land management to enhance environmental protection are considered. The mechanisms involve varying emphases on voluntarism and regulation. The paper examines four key questions relevant to the choice of policy approach: What drives change in private land management? Who should pay? What are the target outcomes? How well do particular policy mechanisms perform? Key conclusions include that each of the policy mechanisms can be appropriate for particular circumstances for particular environmental issues. A key challenge is to develop policy frameworks that are sufficiently flexible to accommodate the most appropriate policy approach for given environmental, economic and social circumstances. The most appropriate approach can vary considerably between circumstances, even for the same environmental problem. Policies that are designed without sufficient flexibility, sophistication or information can often be ineffective and/or inefficient.

Environmental objectives have increasingly featured in agricultural policies over the last 30 years across OECD countries (OECD, 2003). This paper is about the relative merits and limits of different policy mechanisms for enhancing or protecting the environment, with a focus on their relevance to rural areas of Australia. The title implies a dichotomy between voluntary and regulatory approaches, although in fact there is a spectrum of approaches. I will be considering mechanisms from across the spectrum.

I do not have any philosophical or ideological preference for or against any of the policy options. My preference is for an approach that is appropriately targeted, well implemented and achieves its objectives at least overall cost.

Table 1 shows six broad categories of policy mechanisms relevant to changing land management for environmental protection in rural areas. Changing private land management is not the only option for environmental policy, and it is not always the best option. Other options are discussed briefly in the discussion of salinity policy later in the paper, but most of the paper focuses on the options in Table 1.

Consider which of these six options we might consider to be voluntary or regulatory policy measures. A narrow interpretation might limit our consideration of voluntary measures to the first one or two categories (extension, broadly defined, and accreditation). However, all of the first four are voluntary in the sense that participation by landholders is optional and at their own discretion. There is, of course, a distinction between those approaches that provide a positive or negative financial incentive to participate and those that do not. Even regulations that set standards for acceptable levels of environmental protection can be meaningfully viewed as voluntary in the sense that compliance is not assured – people may choose to violate the standard and bear the risk of suffering a penalty.

A narrow interpretation of regulation would focus on the setting of standards backed by penalties for non-compliance. However, there are other policy approaches that involve compulsory participation (e.g. pollution taxes, tradable pollution permits), and virtually all approaches may involve some form of regulation in establishing the basis for their operation.

Table 1. Broad categories of policy mechanisms for encouraging change in land management to enhance environmental protection.

One can see that the distinction between voluntary and regulatory approaches is not clear-cut; the policy options have varying degrees of voluntarism and regulation. Rather than attempting to resolve this ambiguity, I suggest that it is better to examine each of the six categories listed in Table 1, without getting bogged down with definitions.

The real question is when to use each type of policy mechanism. One cannot say that one of the approaches is better or worse than the others as a general rule, but one can say so for particular mechanisms in particular cases. The choice of policy option should depend on the answers to a range of questions, including the following.
• What drives change in private land management?
• Who should pay?
• What are the target outcomes?
• How well do particular policy mechanisms perform?

The remainder of this paper addresses these four questions, and discusses their consequences for the choice of policy mechanism.

What drives change in land management?

Research has given us a good understanding of the factors that inhibit voluntary, non-incentive programs such as the National Landcare Program from achieving more decisive and extensive changes in land management (Pannell 1999a, 2001a). In Pannell (1999b) I summarise the main factors into the following five areas.

(a). Lack of profitable options: There are land degradation problems for which the responses required to mitigate them would be unprofitable even in the long run (e.g. dryland salinity in many low-to-medium rainfall areas). If the required changes are not excessively costly, this may not be a crucial problem, but if large-scale adoption of unprofitable changes is required, the change is unlikely.

(b). Long time scales: Some environmental problems are long-term in nature and practices designed to stop land degradation can be slow to take effect. This means that those farmers who are forced by circumstances to give priority to short-term profits are unable to adopt preventative measures even if they would eventually be profitable enough to offset the up-front costs and interest.

(c). Uncertainty: For many land degradation problems, the value of the proposed management changes is highly uncertain (Pannell, 2003). Farmers are uncertain about: (a) the long term profitability of the treatment, (b) its impact on the environment, and (c) whether their investment will be wiped out (e.g. by drought or by movement of salt and water from another property).

(d). Problems with trialling the options: Farmers prefer to trial new practices on a small scale before adoption. For some practices, trials are not effective in protecting the environment unless they are large-scale. Even where trials are undertaken, the results can be very difficult to observe. For example, if an area of trees is planted, its off-site impacts on water tables are difficult to measure.

(e). Off-farm impacts: A farmer who is the source of problems for someone else may not factor in those problems when considering a new practice. Secondly, if a farmer perceives that a problem is caused by another farmer, the incentives for taking action are reduced. For example, the trees you plant may only help someone else, or they may be killed by a rising saline water table caused by other farmers.

Of the five issues listed above, (a) is probably the most important. Many factors influence farmers’ decisions about adoption of new landcare practices, but where major, large-scale changes are required, commercial reality is likely to be the decisive influence. To a greater or lesser extent, almost all farmers are willing to make financial sacrifices for the good of their land or the environment, but they also must give priority to remaining in business and meeting other family and social objectives.

Lindner (1987) in a wide-ranging review of the literature on adoption and diffusion of agricultural innovations concluded that the objectives of individual farmers figure centrally in the adoption and diffusion process. He found that,

“there is compelling empirical support for this emerging consensus that the final decision to adopt or reject is consistent with the producer’s self interest.” (p. 148)

“Self interest” in this context is considerably broader than merely “profit”. It may, for example, include objectives related to risk, leisure and environmental protection. Nevertheless, profit is an important element of “self-interest”. There is evidence that even for innovations oriented towards resource conservation, economic considerations are the most important determinants of actual adoption decisions. For example, Sinden and King (1990) concluded that,

“While the stewardship motivation and personal factors encourage perception and recognition of a problem, economic factors promote actual adoption.” (p. 179)

Similarly, Cary and Wilkinson (1997) found that,

“Generally, the best way to increase the use of conservation practices to overcome land degradation … will be to ensure the practices are economically profitable.” (p. 20)

Apart from exceptional individuals, NRM outcomes from the National Landcare Program have been limited to cases where the net cost of change to landholders is low. NRM problems that require large-scale change at high cost (e.g. dryland salinity) are not likely to be effectively addressed by voluntary, non-incentive policy measures.

The implication is that voluntary, non-incentive policy measures need to be well targeted to issues and regions based on information about the private benefits and costs to farmers of the proposed management changes.

Who should pay?

The policy mechanisms in Table 1 involve different distributions of the responsibilities for paying. Particularly in the light of the previous section, it is important to appreciate that unless the land use changes are beneficial to the farmers (relevant to mechanisms 1 and 2) or can be made so (mechanism 4) then, one way or another, somebody would have to pay for their implementation.

“While society as a whole may benefit from preserving biodiversity, someone must pay to provide those benefits. If habitat is preserved by condemning private land, the private landowner pays. If it is conserved by purchase of private land with public funds, taxpayers pay. If it is preserved via a system of development rights, developers and future property owners pay. Reasonable persons may differ regarding which groups are more morally deserving of bearing or escaping the burden of payment. But someone must pay.” (Boyd and Simpson 1999, p. 92).

Much of the heat in policy debates is over the question of who should pay. Some people assert that ethical considerations imply that farmers should be willing to implement environmental works, because their failure to do so will result in substantial adverse impacts on each other, on the non-agricultural community and on future generations. (I will refer to this as the “narrow” view of environmental ethics.) The ethics are, in reality, more complex than this (Pannell 2000).

An ethical principle we might espouse is avoidance of harm to others. Presumably this lies behind the narrow view on environmental ethics. However, it is a principle that can cut both ways. It might equally well be argued that the community should not impose costs on farmers for the benefit of a much larger group of non-farmers. Such an argument might be complicated by considerations of differences in wealth and income between the two groups, the rights of the two groups, and the relative sizes of on-farm and off-farm benefits from on-farm treatments. Nevertheless it remains a subjective issue about which different individuals may reasonably disagree. I am not arguing that this broader perspective on the ethics of salinity should take precedence over the narrow view. I am merely pointing out that this alternative perspective exists, that it is not objectively “wrong”, and that it requires us to consider the issue carefully.

I also strongly suspect that if proponents of the narrow view were fully informed about the degree of harm to individual farmers that would result from application of their prescription in least some situations, they would reconsider. The degree of welfare loss and social dislocation from forcing farmers to plant sufficient non-commercial perennials to make a substantial difference to salinity would be so large as to be socially and politically unacceptable.

A second, related, criticism of the narrow view is that it seeks to impose (or perhaps it expects) a strong ethical response by others, while freeing the proponent of the arguments (or society as a whole) from the need for any similar response. In my view, a more defensible and desirable ethical position for the community as a whole to take is that the community should develop policies that protect the environment in a balanced way and are as effective as possible per dollar spent. In other words, the ethical responsibility for ensuring that at least the non-agricultural impacts of environmental protection are managed appropriately rests with governments (on behalf of the whole community) rather than with farmers. It is up to governments to design and implement policies that are effective and fair. This is not to say that farmers are immune from responsibility, but that it is up to government to apply effective mechanisms to ensure that farmers meet whatever are considered to be their responsibilities. A policy that relies on farmers to comply voluntarily with ethical principles with which they do not agree would not be effective and therefore, from this perspective, not ethical.

In a similar vein, Richards (2000) concludes that when uncertainty is high, goals are ill-defined and could be changed, and measurement is costly, more onus should be placed on the government to achieve environmental protection rather than with private firms through economic or regulatory policy instruments.

If ethics provides no clear solution to the question of who should pay, perhaps economics might. After all, economics offers principles such as “polluter-pays” and “beneficiary-pays” to guide just such issues. Unfortunately there is nothing in economic theory that helps us to objectively evaluate the relative merits of different decisions about who pays. For such questions, the contribution of economists is limited to:
• quantifying the distributional effects of alternative policies,
• quantifying the impacts of alternative policies on efficiency, and
• assessing the performance of the policies with regard to rules of thumb that attempt to capture (or perhaps shape) community attitudes regarding what is fair (“polluter-pays” and “beneficiary-pays”).

These are important contributions, but they do not avoid the need for subjective decision making about who should pay. Under the “user-pays principle” or “beneficiary-pays principle”, the beneficiary of a good or service should bear the costs of its provision. It is not a principle in the sense of a scientific principle, but rather a suggestion of what is fair. Ironically, it has been associated with the ideology tagged as “economic rationalism”, although there is nothing in economic theory to favour it as an approach for decisions on distributional issues. The “polluter-pays principle” is in direct conflict with the user-pays approach. It is similarly lacking in any basis in economic theory, whatever one may think of it as a guiding ethic.

There are problems in trying to rigorously implement either rule. For many environmental issues, it is difficult to accurately identify and quantify the benefits and costs for either the polluters or the beneficiaries of mitigation. The user-pays approach would dictate that members of the community should pay in proportion to their benefits from the environmental protection. The polluter-pays approach requires costs to be borne in proportion to the damage caused. Meeting the information requirements of either rule is impossible in any practical sense, although approximations may be feasible.

Irrespective of the wishes of government or the community, the market will also have an influence on the distribution of benefits and costs. For example, if farmers' production costs go up due to legal requirements to protect biodiversity, the farmers may or may not be able to pass on the increase to consumers of their products. It depends on how responsive consumers are to price changes. If consumers of their products are too responsive, dramatically cutting consumption as prices rise, farmers lose more than they gain by attempting to pass on the extra costs. In a free market, the distribution of costs between farmers and consumers is completely outside government control as it depends entirely on the responsiveness of product supply and demand to price changes, and these depend on producers’ cost structures and consumers’ preferences, not on government policy.

In summary, neither ethics nor economics solve the problem of who should pay for environmental protection. They offer relevant considerations and helpful information, but the final decision is necessarily made subjectively, in the political arena, based at least in part on views about rights and fairness. These views are somewhat flexible over time, driven by community attitudes and expectations, politics and power. The obvious implication for the choice of policy mechanism are that the choice needs to be made with an awareness of political realities and community attitudes.

What are the target outcomes?

Designing policy in response to science

In my view, there has been insufficient attention paid in Australia to scientific, economic and social evidence about environmental problems and their context before settling on the broad design of environmental policy, or selecting specific policy mechanisms. It seems that the design of Australia’s main policies for the rural environment over the past 15 years (the National Landcare Program, the Natural Heritage Trust, the National Action Plan for Salinity and Water Quality) have tended to reflect currently fashionable ideas or philosophies. The philosophies, which have been of varying merit, include farmer participation in facilitated groups, integrated catchment management, partial subsidies for small-scale on-ground works, regional natural resource management planning, and market-based instruments.

One problem for environmental policy makers is that environmental, economic, and social circumstances vary so widely throughout the country. The chances of being able to successfully base a major environmental policy program around one broad philosophy or approach is very low. As an example, consider dryland salinity.

The biophysical, economic and social realities that ought to determine the design of salinity policy include the following.

• Successful prevention of groundwater rise on a large scale would require vastly greater adoption of perennials than we have seen so far (National Land and Water Resources Audit, 2001; Hatton and Nulsen, 1999; Hatton and Salama, 1999; Dawes et al., 2002)

• Adoption of perennials on that scale would require that they are economically competitive with traditional agriculture.

• Profitable perennials are lacking in most regions (Kingwell et al. 2003), so lack of farmer capacity is usually not the key constraint to managing salinity.

• The on-farm economic benefits from salinity prevention are low relative to the direct costs and benefits of perennials (Bathgate and Pannell, 2002).

• The economic value of off-farm benefits from on-farm salinity management are generally much smaller than previously appreciated. Salinity is often best managed on a local scale, rather than a catchment scale, particularly in WA (Pannell et al. 2001) but also in many areas of other states (Pannell, 2001b; National Land and Water Resources Audit, 2001; National Dryland Salinity Program 2004).

• Over large areas, shallow water tables are not practically preventable (National Land and Water Resources Audit, 2001; National Dryland Salinity Program 2004).

• In some regions, there are increasing numbers of landholders who derive their main income from sources other than farming. They may not be willing or able to devote time and resources to more intensive land management if that is what is required for environmental protection (Barr et al., 2000).

Implications of these findings include the following (Pannell, 2001b; National Dryland Salinity Program 2004).

• Protection of most agricultural land (and some dispersed public assets) from salinisation will primarily depend on the development of profitable perennial land use option. This requires R&D as a key element.

• Engineering has a bigger role in salinity management that we used to think.

• Not all land can or should be protected from salinisation.

• Options for making productive use of salinised resources are very important.

• The role for policy mechanisms to directly influence on-ground salinity management (e.g. market-based instruments, regulation, subsidies) should be limited to high priority locations, not the whole landscape.

• Integrated catchment management is often not the most appropriate response to salinity.

Ridley et al. (2004) attempted to integrate the latest research results from various disciplines to identify appropriate policy responses for different circumstances. The policy options they considered were incentives (my category 3 from Table 1), penalties (5 or 6), plant-based R&D (4), extension (1), salt interception, on-site engineering responses and no action. The last three are for cases where land-use change on private land is not the most appropriate or important response, which is in fact the case for salinity in a number of circumstances.

Recommended policy responses for recharge areas with salinity impacts on waterways

Ridley et al. recommended different policy responses for four sets of salinity impacts: 1) recharge areas with salinity impacts on waterways; 2) recharge areas with salinity impacts on terrestrial assets (infrastructure and high-value biodiversity assets); 3) recharge areas with salinity impacts on agricultural land, flood risk and remnant vegetation on farms (that with lower value than in the second category) and 4) salt-affected agricultural land. To illustrate the approach and the logic, Table 2 shows their recommendations for water resources. The recommendations presented aim to provide logical and transparent arguments as a broad guide and a basis for further debate.

Where salinity has impacts on waterways, the important factors driving the choice of policy approach for protection of water resources are: 1) salt input from groundwaters into the waterway; 2) responsiveness of groundwaters; 3) the volume of surface flows of fresh water entering the waterway and 4) the farm-level economics of perennial plant-based options. Recommended approaches for different combinations of these factors are shown in Table 2.

The most frequently recommended approach within the table for this set of assets is penalties to prevent loss of fresh runoff that would provide both dilution of salinity and a volume of flow. (This corresponds most closely to category 5 in Table 1.) Where the potential input of salt from groundwaters is low and fresh runoff is low, no action is usually the most appropriate response. Where perennial plant-based options are more profitable than traditional agriculture, the relatively cheap approach of extension to promote the existing options is suggested. Salt interception schemes (where economic) are suggested where the salinity threat is high but groundwater responsiveness to revegetation is low. Five categories have R&D for plant-based options as part of the recommended approach (cases 3, 5, 6, 11 and 12) to protect water resources. See Ridley et al. (2004) for further discussion and explanation.

Incentives to grow existing plant-based options are only an appropriate response in cases 2 and 5 (high groundwater response, perennials slightly less profitable than annuals). This response would be appropriate for areas within the Great Dividing Range with local flow systems, in those parts of the landscape where the profit shortfall for farmers comparing perennials with annuals is less than the off-site benefits from perennials. In case 5, case-by-case analysis would be needed to assess whether incentives, development of plant-based options or a mixture provides the greatest net benefit.

Table 2 is included here to make several points:
• The most appropriate policy response to a given environmental problem may be highly sensitive to the circumstances where the problem is occurring.
• Policies that are designed without sufficient sophistication or information can miss the mark. The main mechanisms being applied under the National Action Plan are extension and incentives, but these are only appropriate tools in a minority of circumstances.
• It is important to consider the full range of policy mechanisms. For example, technology development (represented in Table 2 by plant-based R&D) has been badly neglected under past environmental policy programs, yet can be a very important option in some cases.

Table 2. Suggested policy responses for recharge areas with salinity impacts on water resources (i.e. stream salinity), including consideration of potential loss of flows.

A Penalties would be applied to discourage conversion of annual-based agriculture to higher water using systems such as forestry. The penalty would reflect the loss of water values for downstream users (irrigation, domestic use, environment).
B Requires analysis to determine whether the net effect of perennials is positive or negative. If positive, incentives would be paid to encourage establishment and management of existing perennial options.
C Plant-based R&D in areas of poor profitability may make the system economically competitive, or may reduce the on-farm cost of land-use change to the extent that the incentives needed to prompt the change become small enough to be worth paying.
D Low responsiveness to groundwater and high potential for fresh runoff is unlikely to occur at the same location.

How well do particular policy mechanisms perform?

Information-based approaches (categories 1 and 2 in Table 1) have been discussed already. I proposed that they can be most effective where the changes being promoted are in the private best interest of landholders. The evidence of Landcare is that they can also be effective in achieving altruistic change where costs low enough, though usually on a relatively small scale (e.g. Kington and Pannell, 2003).

Beneficiary-pays approaches (category 3) are the basis for the largest current rural environmental policies in Australia, although a degree of cost-sharing is expected, with landholders bearing well over half the direct and indirect costs in most cases. Most environmental programs in the US and Europe have been strongly based on beneficiary pays (e.g. Latacz-Lohmann and Hodge, 2003; OECD, 2003), using subsidies on a scale that are unimaginable in an Australian context. I suspect that much of that spending has been ineffective, inefficient, or both. For example, the large amounts spent to subsidise farmers who implement approved practices have produced very little in the way of measurable water quality improvements (Alan Randall, pers. comm., 2004). In most parts of the United States, nutrient discharges from point sources (e.g. wastewater treatment plants) have been substantially reduced, but farm runoff remains a major cause of nutrient-related problems (King and Kuch, 2003) and water quality issues overall remain serious (United States Environmental Protection Agency, 2000).

A distinction that applies to both beneficiary-pays and polluter-pays policies is whether they target acceptable levels of pollutants (performance-based policies) or the type of practices or equipment that can be used (design-based policies). In principle, performance-based policies are preferred, where practical, as they more directly target the desired outcomes. Design-based policies are likely to be less efficient as the polluter is usually in a better position than the regulator to judge the least-cost method of achieving a particular level of pollution abatement. However, as noted below, poor observability or measurability of outcomes may limit the practical applicability of performance-based policies in agriculture.

There is some evidence that the particular mechanism used to deliver assistance under a beneficiary-pays scheme can make a big difference to the performance of the scheme. A tender-based approach for targeting assistance was investigated in the BushTender trial scheme in Victoria. It was found in the trial that an improved allocation process for funding to protect remnant vegetation, with greater attention to maximising outcomes per dollar spent, can dramatically increase the resulting environmental benefits (Stoneham et al. 2003). Compared with the approach used in typical fixed-price subsidy schemes to encourage fencing of remnant vegetation, and using a particular system for rating the environmental outcomes, the BushTender approach improved the estimated cost-effectiveness of expenditure by approximately seven fold (Stoneham et al. 2003). On the other hand, such schemes are more expensive to implement, so the question of whether they are more cost-effective overall requires further evaluation.

As noted earlier, technology development (category 4) has been largely neglected in an agricultural NRM context, although it has been extremely successful in non-NRM contexts. Logically, it has a much greater role than it has been given in the past, especially for tackling large, expensive problems like dryland salinity. Based on the success of production-oriented R&D for agriculture, I believe there are strong grounds for optimism about its potential contribution to the environment. For example, scientists believe that substantial improvements in the range and scope of profitable perennials are achievable. The current lack of profitable perennials reflects a low investment in their development historically rather than intractability of the task.

A factor that tends to favour category 4 over the other incentive-based approaches (3, 5 and 6) is the difficulty and expense of monitoring and enforcement of some environmental policies in agriculture. The diffuse nature of most agricultural pollution implies high measurement costs for performance-based policies since it is difficult to trace the exact origin of emissions and thus assign liability. In addition, emission levels depend not only on potentially observable inputs, such as whether perennial plants have been established, but also on unobservable ones, such as the timing of application of potentially polluting inputs and the care with which they are applied (Weersink, 2004). This limits the practical extent of monitoring and enforcement that can be practiced (OECD 2001). Category 4 (technology development) sidesteps this problem by ensuring that practices that generate environmental benefits are directly attractive to landholders, so that there is no issue of enforcement.

Among the revenue-neutral or polluter-pays approaches (5 and 6), there is limited experience in agriculture in Australia or elsewhere. There is more experience in pollution problems outside agriculture. Harrington and Morgenstern (2004) compared the performance of these mechanisms for addressing industrial pollution. They contrasted the performance of economic instruments (EI) (meaning pollution taxes and tradable pollution permits) and command and control (CAC) (meaning legally specified standards enforced by penalties for non-compliance) in the US and Europe. There has been a rapid grown in the use of economic policy instruments, particularly in the US.

Harrington and Morgenstern (2004) studied six particular environmental problems, as shown in Table 3. In each case they compared a policy from the US with one implemented in one or more European countries. In summary, the findings were as follows:
• Almost all of the programs contained some blend of economic instruments and command and control, although there was wide variation in the relative emphases.
• Both approaches can “work”, in the sense of reducing pollution, if implemented well.
• Despite the perceived firmness of CAC, abatement was just as high under EI.
• Firms prefer CAC because of its perceived lower cost to them (even though in fact the costs to the firms were no greater). In the US, business preference for CAC may be because firms felt better able to combat the implementation of CAC schemes through legal challenges. “Regulatory complexity combined with litigiousness delayed the implementation of most regulations far beyond the schedules envisaged by congress.” (p. 15).
• Economic instruments were more efficient, in the sense of achieving a given level of environmental protection at lower cost to the community as a whole, partly because they provide greater incentives for innovation over time. For example, the Swedish NOx tax induced experiments in boiler operations that led to substantial reductions in emissions.

Table 3. Six environmental problems studied by Harrington and Morgenstern (2004).

It is notable that most of the schemes above were applied to industries with few, large point sources (e.g. power utilities), rather than many diffuse pollution sources. It would be much more difficult to apply these policies to industries based on many small diffuse sources, such as extensive agriculture. King and Kuch (2003) providing a sobering analysis of the poor performance of nutrient credit trading schemes for waterway protection from farm runoff, concluding that they have not worked and should not be the main component of the US strategy. Apart from issues of technical effectiveness, the non-use of polluter-pays approaches strongly reflects the political power of agriculture in those countries. Some perceive the beneficiary-pays rural environmental programs of the US and Europe as being primarily designed to deliver financial assistance to farmers, now that price support is out of favour.

Another notable observation by Harrington and Morgenstern (2004) is that, in almost every case, the palatability of the economic instruments was improved by compensating the polluters in some way for the imposition of “polluter pays”. “In nearly all cases, governments eliminate the burden of [economic instruments] by returning fees to the firms. For example, in France, revenues collected through NOx discharge fees subsidised the firms’ abatement investments. … In the United States, where the [economic instrument] of choice is a tradable permit, the permits have been given away rather than auctioned off.” (Harrington and Morgenstern, 2004, p. 17).

A risk with pollution tax approaches is that land managers may not be responsive to the tax, if there are not already suitable, sufficiently low-cost, manage changes available to them. In such a case, the main effect of a tax is to collect revenue, rather than protect the environment, and this would likely further degrade its political acceptability.

In Australian agriculture, political resistance to these types of schemes is also likely to be significant. Consider the difficulties that governments have faced in regulating clearing in New South Wales and Queensland. Enforcing bans is also difficult and expensive, but crucially important if they are to be worthwhile. Given the dispersed nature of agriculture, people who feel that their rights have been infringed by new regulations may attempt to assert those rights by violating the regulations or by more subtle means. An example was clearing bans in water resource catchments of Western Australia in the late 1970s. Even though farmers were fully compensated for the bans (some say they were over-compensated), this did not overcome their public outrage at the bans. Their long-term resentment was not adequately considered in the design of the policy. Farmers were not required to exclude stock from the vegetation, and they continued to clear by stealth by grazing new growth. In the 1990s, a survey found that land for which compensation had been paid was often in poorer condition than other comparable remnant vegetation. Much of it was not contributing effectively to protection of water quality.

Expectations are important drivers of political acceptability. In my judgment, polluter pays approaches are most likely to be politically acceptable when they are used to prevent some activity that is not currently common practice in an industry. For cases where a practice is well established as an industry norm, some form of beneficiary pays approach may be needed in order for the policy to clear the political hurdle. This is the basis behind the suggestion in Table 2 that “penalties” (policy category 5) be used to prevent establishment of perennials in locations where their adverse impacts on water yields would outweigh their positive impacts on salinity. This relates to perennials being grown in locations where they are not currently present. Conversely, “incentives” (policy category 3) were suggested for cases where perennials are not currently grown and are not economically competitive at the farm level, but would provide sufficient off-site benefits to outweigh the on-site costs.

Based on the above discussion and other material, Table 4 summarises some of the strengths and weaknesses of each policy mechanism.

Conclusion

A number of conclusions emerge from this discussion. Firstly, it is not possible to conclude that either voluntary or regulatory approaches are, in general, superior. Each can be appropriate for particular circumstances for particular environmental issues, although polluter-pays approaches are likely to be politically and administratively difficult in many agricultural situations.

There is no dichotomy of choices between voluntary and regulatory approaches. Rather there is a spectrum of options that involve varying emphases on voluntarism and regulation.

A challenge for non-incentive voluntary policy approaches is findings that the private economic costs and benefits of environmental practices is a key influence on their adoption by farmers, especially where land use change is desired at large scale.

Neither ethics nor economics provides a clear solution to the problem of who should pay for environmental protection. They offer relevant considerations and helpful information, but the final decision is necessarily made subjectively, in the political arena, based at least in part on views about rights and fairness.

It is important to consider the full range of policy options, as the option that is currently in favour in policy circles may not be the most appropriate for a given problem. As a generalisation, technology development has been neglected as a policy tool for the environment in Australia.

Within any one of the categories given in Table 1, there is sometimes more than one policy mechanism available (e.g. standards versus pollution taxes, or simple subsidies versus tenders). In general, each of the mechanisms can be effective in achieving environmental protection, but they can vary widely in their cost effectiveness or economic efficiency.

A key challenge is to develop policy frameworks that are sufficiently flexible to accommodate the most appropriate policy approach for given environmental, economic and social circumstances. The most appropriate approach can vary considerably between circumstances, even for the same environmental problem. Policies that are designed without sufficient flexibility, sophistication or information can be ineffective.

Table 4. Strengths and weaknesses of environmental policy mechanisms

References

Barr, N., Ridges, S., Anderson, N., Gray, I., Crockett, J., Watson, B., and Hall, N. (2000). Adjusting for Catchment Management: Structural Adjustment and its Implic