Report from the Sub-committee on Production, Economics and Employment.

3. Fundamental considerations concerning regulation of pesticide consumption in the agricultural sector

3.1  Introduction
3.2  Prioritisation and goal
3.2.1  The political objective
3.2.2  Valuation of pesticide externalities
3.2.3  The precautionary principle
3.3  Regulation of pesticide consumption
3.3.1  Regulatory basis
3.3.2  Regulatory instruments
3.3.3  Assessment

3.1 Introduction

Uncertainty concerning the effect on health and the environment

The use of pesticides has created a basis for increased productivity and rising yields in the agricultural sector. Farmers thus have a clear economic interest in exploiting the possibilities offered by pesticides. However, the use of pesticides is giving rise to increasing concern about their effect on the environment. There is concern both about possible harmful effects on health as a consequence of pollution of groundwater and products and about the effect of pesticides on fauna and flora. Advantages and disadvantages of the agricultural sector’s use of pesticides must therefore be assessed in a general societal framework, in which the risk of environmental and health effects is weighed against the economic benefits.

Unilateral Danish regulation of access to import conventional products

In an assessment of these issues, it must be decided whether they are transboundary or purely national problems. That applies both in economic analyses, since Denmark, as a member of the EU, is subject to the Common Agricultural Policy, and to the choice of environmental instruments. In the case of cross-border pollution, the problems can usually only be solved effectively if the affected countries coordinate their action. It must also be borne in mind that the possibilities of environmental regulation are subject to international agreements (WTO and EU) that do not permit discrimination against imported products, just as a country may only make requirements concerning production standards (the way products are produced) in respect of products produced in that country. The following analyses are based on unilateral Danish regulation of pesticide consumption and an assumption that Danish consumers and producers have access to purchase conventional foreign products and means of production.

In the following, the question of prioritising between economic benefits and environmental considerations is discussed in relation to the setting of political goals for pesticide consumption. Also discussed, in the context of weighing between economic and environmental considerations, are the problems of fixing relevant political goals and valuing environmental goods, together with the questions of risk and uncertainty, the precautionary principle and irreversibility, as elements of policy planning. The chapter ends with a discussion of regulation of pesticide usage and the choice of regulatory instruments.

3.2  Prioritisation and goal

Overall assessment of economic and environmental impacts
Externalities for the farmer…

As mentioned, regulation of the use of pesticides in agriculture should be based on an overall assessment of the economic and environmental impacts from the use of pesticides. The need for regulation arises from the fact that the user cannot be directly expected to consider, in his production planning, the impact of his production on the surrounding environment because the impact is not reflected – or is only reflected to a limited extent – in the user’s production costs. In other words, environmental damage is an externality for the producer. There is thus no financial incentive for him to reduce the harmful effect on the environment even though, like other population groups, he may naturally have a personal interest in preserving a clean environment.

…Need to be regulated through political restrictions

A possible solution to the problem is to have the producer share the cost imposed by his production on other population groups in the form of damage to the environment so that he has a financial incentive to plan his production with a view to reducing its environmental impact. Such a solution would require government intervention in the form of restrictions on pollution from this source. The classic example is the use of environmental taxes (or quotas) that ensure that pollution does not exceed a societally acceptable level.

Extreme prioritsations is seldom optimal

The basis for such regulation is that the welfare gain from an improved environment varies with the degree of pollution: the greater the pollution, the higher the value attached to an improved environment. Conversely, it is usually relatively cheap to reduce pollution when this is at a high level, whereas it is disproportionately costly to remove the last unit of pollution. It is seldom optimal to choose an extreme prioritisation, such as prohibiting the pollution altogether or one-sidedly maximising the microeconomic return without considering environmental and health impacts. The societally optimal³ level of pollution is defined as the level at which the welfare loss for the producer from further reducing the pollution is exactly balanced by the welfare gain for other population groups. This also means that it will normally be reasonable from a societal point of view to permit some pollution.

Consideration of future generations’ preferences

The procedure is complicated by the fact that the environmental impacts from pesticides can be irreversible, making it necessary also to consider future generations’ preferences when fixing the optimal burden. A lack of exact knowledge concerning the long-term environmental impacts of pesticides (on ecosystems, DNA, allergies, etc.) makes the ideal, environmentally based approach difficult.

This raises a number of questions concerning the formulation of the goal for reduction of the pollution:

The current political goal is to reduce pesticide consumption, but it is in reality the risk of harmful environmental and health impacts that the government wants to reduce. Is it possible to define a measure for the environmental burden that reflects the expected environmental and health impacts of pesticides, and that at the same time satisfies the requirements concerning effective regulation of the agricultural sector’s use of pesticides?

The environmental impact from the use of pesticides in agriculture varies with the natural circumstances. In environmentally sensitive areas, even a small loss of pesticides to the environment can have serious consequences, while other areas are less sensitive. The same applies to a great extent to the impact on flora and fauna. The extent of the damage also varies with the nature of the agents and the objects of the treatment. How are these factors dealt with in policy planning?

In order to establish a societally optimal environmental strategy, it must be possible to compare economic and non-economic quantities. The welfare loss from reducing production can be approximately expressed as the economic loss to society, while the environmental gain will be expressed as better public health, a better environment or a more attractive countryside - all of which are difficult to value in money terms. Is valuation of environmental goods a viable way of deciding environmental strategies, and how is this handled in practice?

In the following, these questions are discussed in relation to the formulation of a societally optimal policy in this area.

3.2.1 The political objective

The political objective includes health and environment

As mentioned, the use of pesticides can have a number of undesirable consequences for human health and the environment. Pesticides can accumulate in soil and water and can harm flora and fauna. The use of pesticides also involves a risk to human health, either in the form of acute poisoning, where a person is, for example, exposed to large doses (while filling a sprayer tank) or in the form of more long-term effects, where persons are exposed to small doses over a long period of time (pesticide residues in food products or drinking water).

Lack of scintific evidence

According to the Danish Working Environment Authority (1986), only a few cases of acute pesticide poisoning have been reported, and Christensen & Schou (1998), in a review of the literature, have not found scientific evidence of a relationship between the general state of health and the occurrence of pesticides in Danish food products or drinking water. There are, on the other hand, examples of an increased risk of cancer in workers in the chemical manufacturing industry, and damage to flora and fauna from pesticides has been observed (Christensen & Schou, op.cit. p. 12). Readers are also referred to the report from the Sub-committee on Environment and Health.

… but concern about harmful effects

Even so, there is growing concern that the harmful effects of use of pesticides in practice will increase. In 1987, this led to the Pesticide Action Plan (Danish Environmental Protection Agency, 1997), the aims of which included reduction of pesticide consumption with a view:

	to protecting people against health risks and harmful effects from the use of pesticides. This applies both to the users of the products and to the general public, who must be protected against ingestion via food products and drinking water,
	to protecting the environment – i.e. both harmless and useful organisms among flora and fauna on and in the soil and in aquatic environments.

It was determined in the action plan that total pesticide consumption was to be halved and that consumption was to be steered towards less harmful agents (box 1).

The reduction was to be achieved partly through advisory activities and partly through intensified research on ways of reducing pesticide consumption. It was also stated that the consumption of agents with particularly undesirable health and environmental properties was to be recorded separately as an element of a system for controlling consumption. In addition, a 3 per cent tax on pesticide consumption was introduced to finance the schemes initiated.

In that very same year, 1987, the rules were changed, bringing an end to approval of products deemed to be particularly dangerous to health and introducing the requirement of a review of agents previously registered and classified by the Toxic Substances Board.

The "Action Plan for Sustainable Farming" from 1991 was followed by a decision to require spraying logs and random inspection of sprayers and to make compulsory the part of the training of farmers etc. that had previously been voluntary. In addition, rules were introduced on the use of pesticides in environmentally sensitive areas.

Box 1
Main content of the Pesticide Action Plan from 1986

Pesticide action plan from 1986

Goals: Measured in relation to 1981-85, the total consumption of pesticides must be reduced by at least 25 per cent before 1 January 1990 and by a further 25 per cent before 1 January 1997 Consumption must be changed towards less dangerous agents. Instruments: Increased advice and research on the use of pesticides and assessment of the effects of that use Tightening of the approval scheme for pesticides by requiring assessment of their human toxicity and ecotoxicological effects. The tightening of the scheme was made retroactive (requiring reassessment of already approved pesticides) A requirement that all farms of 10 ha and over keep a spraying log A requirement that all persons working professionally with pesticides acquire a spraying certificate Introduction of technical standards for spraying equipment A ban on the use of pesticides in environmentally sensitive areas (e.g. wetlands, hedgerows and dykes) and closer than 2 metres of lakes and watercourses A 3 per cent tax on the wholesale value.

Tax increased in 1996

In 1996, the tax on pesticides was increased⁴ and differentiated between types of pesticide (herbicides, fungicides and insecticides). There was no differentiation between the groups of pesticides with respect to their toxicity. The differentiation is thereby presumed to have resulted primarily in the use of cheaper agents rather than in a reduction in the consumption of dangerous agents.

A proxy is a compromise between the ideal and what can be administered

For practical reasons, a decision has been made to link the goal to pesticide consumption in the agricultural sector and to the treatment frequency index, which can be measured with reasonable certainty in practice. However, focusing on the consumption of pesticides instead of the environmental and health impacts means that the goal is distanced from the primary impacts, making it difficult to control the environmental burden. The ideal would be to be able to set the goal in relation to the damage to health and the environment, but owing to the above-mentioned constraints, the choice of proxy will often be a comprise between the ideal and what can be administered in practice.

The problem of measuring the damage sets narrow bounds on the choice of proxy for environment and health. Whereas the impact of pesticides on the natural environment can, for example, be expressed by the frequency of selected flora or fauna in areas adjoining cultivated areas, establishing a measure for the health effect of pesticide usage is a different matter altogether. As mentioned, little is known about the relationship between pesticide usage and human health, and it will often be difficult to quantify the harmful effects of pesticides.

Proxy for harmful effect

Therefore, in practice, the only option is to use approximated measures (proxies) for the harmful effect of pesticides, such as the degree of pollution of the groundwater, the toxicity of the agents, the risk of leaching or – as in the case of the Pesticide Action Plan – the amount of active ingredient. It is obvious that the further the application of the measure is from the real harmful effect, the more difficult it becomes to control that effect. However, as described later, it is also important for the control of pesticide usage to be linked so closely to the production that the farmer/producer can relate to it.

OECD (1997) suggested classifying pesticides environmentally in risk groups based on their mobility in the soil, their biodegradability and their toxicity. When this is combined with information on standard doses per ha, it becomes possible to get an expression of the agents’ potentially harmful effect. However, that does not solve the problem of fixing an acceptable level of use of the agents, since to do that, one must know the relationship between the agents’ toxicity and the damage to health caused by that.

Threshold value for pollution of the groundwater

The above-mentioned problem led to the establishment of threshold values for permitted concentrations of pesticides in the groundwater, which in practice meant that it was not permissible for measurable quantities to be found in drinking water, using the measuring methods available at that time. Such limits are an example of a prioritisation that does not take account of the toxicological risk of harm and that therefore precludes any economic weighing against other solutions, such as treating the drinking water. This problem is taken up below in connection with the discussion of the precautionary principle.

In the discussion so far, it has been implicitly assumed that there is one goal to be met. In reality, there are a number of partial goals that it is considered desirable to fulfil at the same time: the use of pesticides involves a risk to both the environment and health, and there will often be a complicated interaction between different impacts that must be combined into a common goal. For example, the use of insecticides affects the number of prey in the environment and thus the basis of life of higher animal species. At the same time, pollution of the groundwater with pesticide residues can result in a risk of poisoning of humans and animals. What weight should be assigned to the different harmful impacts? Should damage to human health be given a higher weighting than damage to animals and the environment? These questions are taken up in the following section in connection with valuation of environmental goods

3.2.2 Valuation of pesticide externalities⁵

Societal analysis requires valuation of health and environmental goods

As described above, regulation of the use of pesticides in the agricultural sector should, in principle, be based on a societal analysis of the economic and environmental effects of the use of pesticides, the aim being to optimise society’s welfare. Such an analysis is difficult because economic and environmental considerations are not directly comparable: the consumption of marketed goods can be expressed in kroner and øre, while the environmental impact of pesticides, for example, can be expressed by the number of wild plants in a given area, the non-appearance of a given species of bird or the concentration of pesticide residues in the groundwater. We are thus faced with the problem of weighing different environmental goals against each other and evaluating their importance with respect to welfare in relation to the loss of welfare from limiting the consumption of produced goods. The basis for such weighting will often be valuation of the environmental goods.

Economic valuation must reflect the population’s preferences

Economic valuation is based on the fact that policy planners seek to ensure that society’s resources are used in accordance with the population’s preferences. Valuation of environmental goods is an attempt to profit the political decision-makers with information on the strength of the population’s preferences for different environmental goods – expressed by the hypothetical willingness to pay for these goods. In other words, one attempts to measure the market consumption that the population will be willing to forego in return for a specific improvement in the quality of the environment. When the value of an environmental good is thus calculated in monetary units, it becomes possible to compare the gain from producing or preserving the good with the costs involved in that.

The consumer must choose

It is a fundamental economic axiom that scarce goods that have alternative uses cannot be free in the societal sense of the word. That applies both to produced goods and environmental goods and it is based on the fact that, when not all needs or wishes can be satisfied, the consumer has to choose. If more resources (in the form of manpower, capital, etc.) are used for environmental purposes, there will be a societal cost in the form of fewer resources for production of goods and services for consumption. Conversely, it must be reckoned that, all else being equal, increased production of tangible goods will reduce environmental quality.

Cost-benefit analysis

The idea of valuing environmental goods is that, provided prices can be put on such goods, it will be possible to carry out a cost-benefit analysis, in which changes in the value of produced goods are compared with changes in the value of environmental goods and, by that means, assess the welfare consequences of regulating agricultural use of pesticides⁶. The difficulties of such an analysis are compounded by the fact that it should, in principle, incorporate future generation’s preferences and take account of cross-border pollution.

There are, in principle, two possible approaches to such valuation: the preference-based method, in which one seeks to measure people’s willingness to pay for environmental goods, and the non-preference-based method, in which one typically looks at the socioeconomic costs of repairing environmental damage.

Preference-based method

The aim of the preference-based method is to identify people’s willingness to pay for non-market goods. The problem is that most environmental goods are in the nature of public goods, for which a market does not exist⁷. In other words, there is no instrument for pricing the good. This situation is often described as a market failure. When making a valuation on the basis of the preference-based method, one seeks to solve this problem either by interviewing consumers to identify their willingness to pay for the goods in question (direct method) or by observing consumer demand for market goods associated with the consumption of non-market goods (indirect method).

As an example of direct valuation, Dubgaard & Østergaard (1999) mention a Dutch survey carried out to determine consumers’ willingness to pay for biodiversity through scheduling of environmentally sensitive land (Brouwer & Slangen, 1998). Here, a conditional valuation was carried out to establish a monetary estimate of the gains achieved by the rest of the population from cultivation agreements. The willingness to pay covered both the utility value and the non-utility value of plants and birds in the areas in question. Examples given include Oskam & Slangen (1997), who focus on the gains from implementation of a protection programme for landscapes in which farming is the primary activity and a study of preservation of pasture-land (hagmark) in Sweden at the beginning of the 1990s (Drake 1992).

As an example of indirect valuation, Dubgaard et al. (1998a) mention that the value of a recreative area could be fixed on the basis of consumers’ statements about what they are willing to pay to visit the area (the travel-cost method). Measuring recreative values on the basis of differences in house prices in different recreative environments is another example of indirect valuation.

Non-preference-based method

The aim of the non-preference-based method is to estimate the societal value of an environmental good on the basis of the cost of repairing environmental damage. The analysis is based on two assumptions: firstly, that specific political limits have been set for the pollution (e.g. the content of chemicals in drinking water) and, secondly, that the socioeconomic cost of achieving the political goals reflects the value that the consumers assign to the environmental goods in question.

An example of such a valuation is a Danish analysis of the technological and economic feasibility of treating drinking water (Chrintz 1997). The result arrived at is that the cost of treating drinking water by means of carbon-filter analysis is DKK 3/m³. For pesticides that can only be separated by osmosis, the cost is DKK 6-7/m³. It is stated in the analysis that some types of pesticides and their breakdown products cannot yet be removed from drinking water.

Dubgaard et al. (1999) also report a Danish analysis now in progress, the aim of which is to calculate future savings within the drinking water supply system if pesticides are no longer used in Denmark. In this connection, importance is attached to the fact that future losses as a consequence of earlier times’ use of pesticides cannot be prevented by stopping using pesticides in future, since, seen economically, this is a "sunk cost". The aim is thus to devise an alternative cost analysis that covers only the societal savings that can be expected from non-use of pesticides in the future. Such an analysis requires estimation of the extent to which the pesticides that farmers are permitted to use today will cause pollution of the groundwater, and the time horizons involved

Summary

The requirement concerning valuation depends to a great extent on whether economic benefits are to be weighed against environmental values or whether the environment policy goal has been fixed in advance. If the starting point is that pesticide pollution must not exceed certain threshold values (e.g. 0.1 µ/litre groundwater), it is, in principle, of no interest to try to value the environmental gain. In this case, the task is to achieve the agreed goal in the most efficient way, i.e. to find the method that will ensure achievement of the goal at the lowest possible cost. An example of such an analysis is treatment of drinking water versus a ban on the use of pesticides. The method precludes any weighing between economic and environmental considerations.

If, on the other hand, the aim is to weigh economic benefits against environmental values, the environmental gains have to be measured. In this case, the criterion for whether a solution is societally acceptable is that the environmental gains are greater than the economic losses from ensuring them. The above-mentioned analysis of future savings within the drinking water supply system if pesticides are no longer used is an example of a valuation in which the savings are taken as a measure of the value assigned to pesticide-free drinking water by the consumers. By comparing the savings with the costs of stopping using pesticides, an expression is obtained of the cost-benefit ratio, which must be smaller than one in order to be societally acceptable. Such an analysis ensures that the benefits are greater than the costs, but does not guarantee societal optimality.

As mentioned earlier, for societal optimality, the marginal environmental benefits must equal the marginal costs of improving the environment. It is thus necessary to know the value of the environmental gains from changed pollution in order to indicate the optimal solution. This makes special requirements concerning the valuation of environmental goods, and the valuation is made even more difficult by the fact that the analysis must, in principle, cover the impacts on both health and the surrounding environment in order to get a complete picture of the societal benefits from reducing the use of pesticides. It is thus very difficult to indicate the societally optimal solution to the use of pesticides in agriculture.

Lastly, it must be added that the value of environmental goods must be expected to vary with the economic prosperity of the population. High-income population groups generally attach more importance to environmental goods than low-income groups, which underlines the problems of transferring results of valuation studies from one country to another.

3.2.3 The precautionary principle

The precautionary principle is linked to uncertainty concerning - optimal pesticide usage

Measurement of the impact of pesticides on the environment and health is encumbered with considerable uncertainty.

Firstly, it is difficult to fix the optimal use of pesticides in agriculture. There are several reasons for this, including the fact that the timing of the treatment is vital. In periods with severe pest attacks, a delay in treating the crop in question can result in a need for extra heavy dosage to achieve the desired effect, while in periods with few pest attacks it can be difficult to dose sufficiently finely (only part of the crop is attacked, but for practical reasons the entire acreage is sprayed). Furthermore, the farmer’s financial result is uncertain owing to fluctuations in the harvest yield. Production planning and treatment monitoring are thus also encumbered with uncertainty, which will also be reflected in the harmful impact.

- extent of pollution

Secondly, there is uncertainty about the extent to which pesticides used in agriculture end up in the groundwater and the food chain and about when this can be expected to occur. The uncertainty concerning the time frame means that the needs of future generations must be taken into account when assessing the extent of the damage.

- impact on health and environment

Thirdly, there is uncertainty concerning the harmful effect of pesticides on human health and the environment because, in many cases, too little is known about this.

In the following, the precautionary principle is discussed in relation to these uncertainties.

Definition of the precautionary principle

The precautionary principle
… gives priority to the environment

The potential harmful impact from the use of pesticides is causing widespread concern, both in the general public and in scientific and medical circles. With a view to preventing harm to the environment and health, Denmark and the EU have chosen to set restrictive threshold values for the content of pesticides in drinking water and food products⁸. Owing to the uncertainty concerning the effect of harmful substances in the aquatic environment, agreements that consideration for the environment shall have priority in cases in which there is uncertainty about the outcome – the so-called precautionary principle – have been incorporated in a number of international treaties and declarations.

The precautionary principle was originally developed in German environment law and, partly via the North Sea Convention, found its way from there to the Union Treaty. In German, the word used is "vorsorge", which actually means prevention, and the principle means orienting the planning of society’s activities (production, energy supply, etc.) towards preventing environmental problems – especially irreversible problems – before they arise. For a detailed description, readers are referred to Boehmer-Christiansen (1994) and Zimmermann (1990).

The precautionary principle is embodied in international agreements ...

The principle is mentioned in the conclusions from the first North Sea Conference in Bremen in 1984 and is embodied in the ministerial declarations at the subsequent North Sea Conferences (see the box). It has since been embodied in a number of other international agreements (Danish Environmental Protection Agency).⁹

… and in EU policy …

A formulation of the precautionary principle is also embodied in the EU Treaty from 1992 (The Maastricht Treaty), which states: "Community policy on the environment shall aim at a high level of protection taking into account the diversity of situations in the various regions of the Community. It shall be based on the precautionary principle and on the principles that preventive action should be taken, that environmental damage should as a priority be rectified at sources, and that the polluter should pay." It is also stated in article 130 R(3) of the Treaty that in the formulation of Community policy, attention shall be paid to (Danish Environmental Protection Agency, 1998b, p. 6):

	existing scientific and technical data
	the environmental situation in the various regions of the Community
	the advantages and disadvantages of implementing a measure and of not implementing it
	economic and social development in the Community as a whole and balanced development in its regions.

It is thus not stated in the treaties that intervention against damage to the environment must await scientific evidence. However, the precautionary principle has a less far-reaching formulation in the Maastricht Treaty than in the above-mentioned international agreements.

… but only indirectly in Danish legislation

The precautionary principle is not mentioned directly in Danish legislation, but the approach is, for example, embodied in the introductory provisions of Environmental Protection Act: "The purpose of this Act is to contribute to safeguarding nature and environment, thus enabling a sustainable social development in respect of human conditions of life and the conservation of flora and fauna."(Danish Environmental Protection Agency, 1998b, p. 2). The aim of pesticide-free drinking water is another expression of the fact that the precautionary principle has found its way into Danish legislation.

Well-founded uncertainty concerning harmful effect

The crux of the precautionary principle is that well-founded uncertainty concerning the harmful effect of environmentally foreign substances can in itself occasion intervention, even in the absence of certain scientific evidence. The reason for this is the relatively strict requirements made concerning scientific evidence, compared with the complexity of the interaction between environmentally foreign substances’ interaction with nature and health, where it is often difficult to document the possible multivariate relationships. The time factor also plays a role here, in that the relationships can only be documented ex-post, thereby reducing the possibility of avoiding the environmental impacts.

However, the precautionary principle has a broader meaning that also implies a duty to coordinate research and development of new and less environmentally harmful technologies.

Risk and uncertainty in economic analysis

In economic theory one differentiates between risk and uncertainty, risk being used about outcomes that can be described by a statistical probability distribution, while uncertainty is used about outcomes where there is no empirical data on which to build or where the range of outcomes cannot be delimited. Dubgaard et al. (1998b), in describing the problem, used the example that animal tests can be used to derive statistical probabilities for a relationship between the use of pesticides and the occurrence of cancer in animals, but that transferring such results to risk analysis for humans rests on assumptions that cannot be thoroughly tested because human experiments are precluded. The effect of similar treatment on humans is therefore uncertain.

Uncertainty

Optimal environment policy under risk

In the basic environment-based economic model, the environment is treated as a consumption good in line with produced goods and services, and the consumption must be combined in a way that ensures maximum societal welfare. Other assumptions for the model are that the marginal utility falls with rising consumption and that there is a negative relationship between the production of produced goods and environmental goods (damage to the environment increases with rising production). The optimal societal solution is therefore ensured when the marginal damage to the environment is equal to the marginal cost of reducing the pollution. It is assumed in the analysis that the environmental damage can be valued so that damage and cost can be compared.

In practice, there will be uncertainty about the environmental impact. If the probability of a given outcome is known, it is possible to calculate the marginal damage under various assumptions concerning the consumers’ risk assessment. This is illustrated in Figure 3.1, which shows the optimal environment policy under risk with a known probability distribution. The heavily drawn curves show the marginal damage (MD) and the marginal cost (MC) with increased pollution. Here, the MD curve has a probability distribution attached to it, showing the statistical distribution of a given outcome, and the curve E(MD) is the unbiased estimate of the marginal damage.

Figure 3.1
Socioeconomically optimal environment policy under risk
(Figure text:
Forurening = Pollution)

However, the decision-maker may be averse to risk and therefore assign greater importance to deviations above the E(MD) curve than to deviations below it. He will therefore not rely on the simple mean value curve, but on the F(MD) curve. In such case, the optimal solution will be not F, but F*. Measured in relation to the risk-neutral consumer, there will therefore be an income loss corresponding to the marked triangle that the risk-averse consumer is willing to pay for greater certainty. This is a risk premium he pays for maintaining the higher level of certainty corresponding to F(MD). There is no question of real uncertainty, but simply of risk aversion.

Optimal environment policy under uncertainty

In many cases, one is faced with making a decision on environment policy without having any real statistics on which to assess the extent of environmental damage. In economic terminology this is called decision-making under uncertainty, i.e. there is no statistical probability distribution for the extent of the damage. This is illustrated in Figure 3.2, where the societally marginal damage function is only known in the interval 0 – F. In the case of greater damage, the marginal damage function can be vertical (A) or horizontal (B) or somewhere in between (the probability distribution is not known). One is therefore in a situation in which there is no scientific basis for saying anything about the probability of the extent of the damage (except that it is expected to lie between the vertical and the horizontal curve (A and B)). In such a situation, the risk-averse consumer will most probably choose pollution level F, which is the certain solution, but at high marginal cost (Nc), while the risk-taking consumer will choose solution F'. The triangle C in the figure illustrates the extra cost of choosing the pollution level F instead of F', which must be compared with the environmental benefits gained.

Choice of criteria for decisions

DØR (1998) states, with reference to Arrow & Hurwicz (1972), that, under uncertainty, it is rational to base the decision on extreme outcomes that give the minimum or maximum outcome. As an example of decision criteria, DØR mentions "maximin" and "maximax", which express the highest

Figure 3.2
Environmental policy under risk and uncertainty’
(Figure text:
Forurening = Pollution)

achievable welfare with major and minor environmental damage, respectively¹⁰, taking into account the cost of alternative intervention against environmental pollution. The method is an alternative way of looking at the precautionary principle, where decision-makers that are averse to uncertainty prefer the maximin criterion (pessimistic assessment), while risk-taking decision-makers prefer the maximax criterion (optimistic assessment).

DØR (1998) also points out that the maximin and maximax criteria are suitable instruments for identifying research areas that should be given priority in order to gain new knowledge and thus reduce the uncertainty. The crux of the problem is to narrow down the range of outcomes for events under uncertainty. Therefore, according to the report, it is better to concentrate on improving knowledge about the extreme outcomes than on analysing intermediate scenarios.

Consumer sovereignty

Dubgaard et al. (1998b, p. 3f) have analysed the problems concerning decision-making under uncertainty on the basis of the existing literature. The key words are consumer sovereignty, i.e. the right of consumers to remain sceptical about the use of certain substances in food production, even when there is no scientific evidence that the substances in question are dangerous. Buschena & Zilberman (1994) state: "Research in risk perception and risk behaviour shows that it can be difficult to explain people’s assessment of different risk factors on the basis of the assumptions concerning rationality and consistency used in economic theory." Pearce (1994) reported a number of experiments that show that people attribute greater probability to events they can envisage than to events they have difficulty in envisaging. Furthermore, there is often a distorted perception of small probabilities, which are either overestimated or underestimated.

Rational updating of subjective assessment

However, Dubgaard et al. (1998b, p. 4) also refer to research aimed at improving decision-making under uncertainty. The conclusion is that individuals carry out rational updating of their subjective assessment on the basis of information concerning the objective probabilities as further information gradually appears concerning the potentially harmful effects of chemical substances and other potential causes of damage to the environment and health

Irreversibility

Irreversibility comprises

The above-mentioned discussion focused mainly on decision-making under risk and uncertainty. However, choice of environment policy is also affected by irreversibility in the basis for decisions. This can relate to economic factors, in that decisions to initiate major investments in rectifying environmental damage will be binding on future decision-makers. It can also relate to biological factors since failure to intervene can result in accumulated damage that cannot be rectified at a later date. Lastly, it can relate to the political decision-making process, in that, for reasons of credibility, politicians will have difficulty in changing a policy once this has been decided upon. One can thus speak of economic irreversibility, biological irreversibility and lack of decision-making flexibility, all of which can be translated into economic terms.

Economic irreversibility concerns the situation in which combating pollution implies investment in technical plant that will subsequently be in the nature of a ‘sunk cost’, i.e. the investment sum cannot be recovered by selling the investment good. An example is investment in water-treatment plant to remove pesticide residue from drinking water. Once the investment has been made, it will – until the plant is technologically obsolete - reduce the possibility of introducing other forms of protection of the groundwater since the costs of the investment must in all events be paid.

- biological irreversibility

Biological irreversibility relates to the situation in which the damage depends on the accumulated pollution, and in which pollution continuously accumulates in nature because nature’s capacity for regeneration is exceeded. Complete irreversibility is a situation in which the pollution cannot be removed by nature’s own biological processes (e.g. heavy metals), or in which the pollution results in the eradication of species of fauna or flora. In this case, the damage from the accumulated pollution is in the nature of a ‘sunk cost’ since the damage is not automatically rectified when the pollution ends. Intermediate irreversibility is the term used of a situation in which the pollution exceeds nature’s own capacity for regeneration. Pollution of the environment with nitrogen, phosphorus and pesticides is an example of partially reversible pollution in that, here, nature will to a large extent be able to recreate a good environment if the pollution ends.

- lack of decision-making flexibility

Lack of decision-making flexibility relates to the possibility of changing a policy once it has been agreed upon politically. The concept expresses the degree of flexibility in the decision-making process, taking into account both the time factor and the frequency of changes to the policy. One thus speaks of timing inflexibility if a decision must be made now or never be made, while the possibility of postponing a decision gives timing flexibility. As mentioned, inflexibility can also be due to politicians being bound by earlier decisions (irreversibility), while in other cases it will be possible to adjust a policy along the way (reversibility).

In some situations, the possibility of waiting with a decision will have an independent value (option value), that must form part of the decision-making basis. The option concept has been borrowed from the financial sector, where one can take out an option to buy a product at a pre-agreed time and a pre-agreed price. In other words, one ensures that the product can be acquired at the agreed price, but whether the option is actually used will depend on whether the agreed price is lower than the market price. If this is not the case, the purchaser will choose to buy at the market price. For this security, the purchaser pays a premium (the option value), which is the price for having flexibility in the decision-making process.

Option on future choices

Application of the option concept to the environment problem is based on the fact that irreversibility exists, in the form of either biological or economic inflexibility. The option value relates to the possibility of choosing different policies in the future, i.e. there is flexibility in the decision-making process. In the case of biological irreversibility (the environmental damage cannot be rectified), the option value will relate to the fact that limiting environmental pollution today will give future options. The criterion for choosing the option is that the achieved benefit of having flexibility with respect to solutions must be greater than the associated costs.

Decision now or "wait and see

Christensen (1999) has analysed the relationships between uncertainty, irreversibility, decision-making flexibility and option value. The analysis is reproduced schematically in Figure 3.3, where uncertainty, irreversibility and flexibility are ordered hierarchically. If there is certainty concerning the outcome¹¹, the decision-making basis will in all cases be the conventional economic measure NPV(Net Present Value), which expresses the discounted value of future benefits and losses from the policy. If, on the other hand, there is uncertainty about the future outcome, the result will depend on whether there is economic or biological irreversibility and on whether there is flexibility in the decision-making process. The last-mentioned is expressed in the figure by the fact that the decision can be postponed (wait and see) or must be made here and now (now). If there is uncertainty about the outcome and there is economic and/or biological irreversibility, and there is also a possibility of waiting with the decision, there will be an option value that must be included in the decision-making basis (expressed by NPV+option). In the other cases, the decisions will depend solely on NPV.

Figure 3.3
Decision-making basis under uncertainty, reversibility and flexibility
Source: Developed on the basis of Christensen (1999 p. 91f)

(Figure text:
Usikkerhed = Uncertainty
Irreversibel = Irreversible
Reversibel = Reversible
Vent og se = Wait and see
Nu = Now)

As an example of a situation in which an option value affects the decision-making basis, the Economic Council mentions reduction of CO₂ emission in order to prevent future greenhouse effects (DØR, 1998, p. 241). The problem arises from the fact that the damage from CO₂ is assumed to be irreversible. By reducing the CO₂ emission today, one gains an extra option of using new knowledge concerning the damage caused by greenhouse gases at a later date. If new knowledge shows that the damage is great, one has already introduced new limitations. If, conversely, new knowledge shows that the damage is limited, one will have the possibility of easing the policy. The option value thus relates to the possibility of avoiding damage as a consequence the greenhouse effect that would have been lost if one had not intervened against CO₂ emissions earlier on.

Pollution with pesticides is an irreversible process

Transferred to the pesticide problem, the option concept can be illustrated by pollution of the groundwater with pesticide residue. Such pollution is - any rate to some extent - irreversible since it will take many years to re-establish clean groundwater once the damage has been done. However, since there is at present considerable uncertainty concerning the damage caused by such pollution, it may be relevant to maintain flexibility with respect to the choice of policy. The option thus consists in choosing between reducing the pollution today (e.g. by restricting farmers’ use of pesticides) and postponing intervention, with the consequent risk of dangerous substances accumulating in the groundwater. If one chooses the latter solution, one loses the possibility of an environmental benefit because the pollution is extremely irreversible.

The above-mentioned problem is parallel to the precautionary principle, where one chooses to take action against the pollution in order to be "on the safe side". In this connection, it should be noted that option value can exist whether the decision maker is risk-neutral or risk-averse, but that risk aversion will affect the magnitude of the option value. High risk aversion will thus speed up action to limit the pollution if this is irreversible.

Strong, moderate and weak caution

Pearce (1994) and O’Riordan & Cameron (1994) have attempted to concretise the precautionary principle by breaking it down into weak, moderate and strong caution. With the strong precautionary principle, economic considerations have higher priority than environmental considerations. The view is taken that risky/uncertain technologies should normally be prohibited irrespective of cost. With the moderate precautionary principle, importance is attached to eliminating risks in the choice of technology, but use is made of the proportionality principle that the cost must not be disproportionately high, while the weak precautionary principle approaches traditional cost-benefit thinking that risks must be weighed against costs in connection with risk-reducing environmental initiatives.

Dubgaard (1999) has tried to concretise the three concepts still further:

1. The strong precautionary principle covers both risk and uncertainty. The concept prescribes that society shall refrain from activities that involve potential environmental risks – irrespective of the cost, i.e. economic considerations are entirely subordinated to environmental considerations. A total ban on pesticides can be regarded as an expression of a strong precautionary principle.
2. The moderate precautionary principle also covers risk and uncertainty. Intervention is initiated on (confirmed) suspicion of harmfulness, but consideration is paid to the cost. Economic considerations enter the equation through use of a proportionality principle according to which the costs must not be "disproportionately" high. The 0+scenario discussed later can be regarded as an application of the moderate precautionary principle.
3. The weak precautionary principle concerns a situation in which risks must be documented and quantified before a decision is taken on a ban or regulation.

In the American economic literature there is a counterpart to such a precautionary principle in the form of Safety First Rules (Buschena & Zilberman, 1994), which can be interpreted as operational criteria for decision-making under risk. Unlike the expected-utility theory, these rules are in the nature of rules of thumb for cautious decision-making behaviour that assumes the existence of quantifiable risks. The rules therefore give no indication of how one should act in the case of decisions associated with uncertainty in the form of non-quantifiable probabilities. One uses economic optimisation principles with the restriction that risks must be kept within prescribed threshold values.

Veterinary standards, sanitary and phytosanitary rules

Lastly, it should be noted that the precautionary principle is used when setting veterinary standards and sanitary and phytosanitary rules. The aim is to protect human, animal and plant life from diseases. Such regulations are monitored internationally to ensure that they are implemented in a way that minimises the negative impacts on international trade. The regulations must also be based on scientific principles, analyses and risk assessments. Labelling with a view to promoting the sale of "particularly environment-friendly products" (eco-labelling) is also an expression of a precautionary principle – in this case with the aim of influencing consumer behaviour to the benefit of environment-friendly products. Such labelling is permitted provided it does not involve discrimination with respect to the origin of the products.

Summary

harmful impacts of pesticides is that there is often a lack of scientific data on the magnitude of the damage and the probability of the damage occurring. In such a situation, the precautionary principle means that consideration for the environment should be given priority over economic and other societal considerations, i.e. one chooses the safe solution in order to be on the safe side. However, this choice also has a price, which society has to pay.

It has been shown in the foregoing that identifying a societally balanced environment policy depends on the existence of scientifically based knowledge concerning the magnitude of the damage and the probability of it occurring. The choice of policy will, however, depend on the decision-maker’s willingness to take a risk. A risk-averse decision-maker will normally choose a more restrictive environment policy than a less risk-averse decision-maker, even if this implies an economic loss. However, it is important to realise that, in the eye of the decision-makers, the choice will be a balanced one in both situations – environmental considerations are weighed against economic considerations. The difference lies in the fact that the risk-averse decision-maker is more willing to accept an economic loss in order to gain greater certainty concerning protection of the environment.

In many cases, however, well-founded data for pollution and the magnitude of the damage is lacking, and the decision-maker is therefore in an uncertain situation, being without a real basis for a decision. In this situation, the risk-adverse decision-maker will often choose the extreme solution of banning the pollution in order to be "on the safe side", while the more risk-taking decision-maker will perhaps go to the other extreme and ignore the possibility of damage. As stated above, in this situation, the most rational solution would be to base the decision on extreme outcomes that produce minimal or maximum outcomes. However, it is pointed out that research is needed to create more knowledge about the extreme outcomes in order to reduce the uncertainty concerning the outcomes.

Even so, the choice of policy is also influenced by the reversibility in the system. This is because once an investment has been made in an installation to protect the environment or the producers have adjusted to environmental restrictions, it costs to change the policy (economic irreversibility) or because the environmental damage accumulates as a consequence of the pollution exceeding nature’s capacity for regeneration (biological irreversibility). In such situations, the time factor gains particular importance because the possibility of postponing the decision can have an independent value that must be included in the basis for decisions. In this case, too, the decision-maker’s risk aversion will influence the result because high risk aversion will generally make him tend to postpone relaxing restrictions on the pollution.

The precautionary principle must thus be seen in a broader perspective, where both uncertainty and reversibility are included in the basis for decisions and economic and environmental considerations are weighed against each other.

3.3 Regulation of pesticide consumption

Aim: for the farmer to include consideration for the environment in his planning

As mentioned, in the planning of environment policy measures, societal gains and losses from the use of pesticides must be weighed up. In this connection, it is important that the decision concerning the use of pesticides lies with the farmer, whose aim is to optimise the profitability of his farm. The task is thus to formulate a political framework for the farmer’s activities that will encourage him to include consideration for the environment in his planning and thus ensure a societally acceptable solution.

A brief outline is given below of the principles underpinning environment policy measures, including establishment of the regulatory framework and choice of proxies for use in evaluating the effect of the policy. A picture is also given of different regulatory instruments and their suitability for regulating pesticide consumption.

3.3.1 Regulatory basis

The regulation must be targeted on the environmental problem

Regulation must generally be based on the environmental problem that is to be solved. If the problem concerns harmful effects on the population’s health, the regulation must in principle focus on reducing those effects. If, on the other hand, the problem concerns damage to nature, the relevant proxies will be the wild flora and fauna. Since such damage can occur far from the source of the pollution (farmers’ use of pesticides) and the relationships between pesticide usage and harmful impact are often diffuse, one is in practice forced to use proxies for the harmful effect of the pesticides.

Difficult to formulate proxies

Here, one faces the problem that uncertainty concerning the harmful effect of pesticides makes it difficult to define a relevant proxy. If one chooses a proxy close to the ‘injured party’, it will perhaps express the magnitude of the damage reasonably precisely, but the relationship between the source of the damage and the proxy will be diffuse. If, on the other hand, one chooses a proxy close to the source of the damage, the relationships between the proxy and the magnitude of the damage will be uncertain. Since the user must be able to relate to the chosen proxy in the planning of his production, one often, in practice, chooses a proxy close to the source of the pollution, knowing full well that there will be uncertainty concerning the relationships between the proxy and the damage. The goal of a 50 per cent reduction in pesticide consumption in the agricultural sector is an example of such a compromise. The goal is directly related to the production and can be monitored. The problem is that it is difficult to prove the relationship between consumption and damage scientifically, so the goal tends to be perceived as a political choice.

However, there are a number of ways of improving the regulatory basis. In an analysis of pesticide regulation in the agricultural sector, the Danish Institute of Agricultural and Fisheries Economics (SJFI) (1998b) points out that regulation can be made more targeted by including in the regulatory basis the properties of the pesticides (toxicity and risk of leaching) and the geographical distribution of pesticide usage. The reason for the latter suggestion is that the risk of pollution of the groundwater and damage to nature varies greatly with the natural conditions.

3.3.2 Regulatory instruments

Quantitative or tax-based regulation

When choosing regulatory instruments, one differentiates between rules and taxes. The aim of rules is to limit what the user may do, whereas taxes act economically by favouring (or de-favouring) a given behaviour. Another form of regulation is quotas, which are quantitative restrictions on the user’s pollution or production activity. If a fixed quota is imposed on a single user, the regulation is quantitative, but if the quota is made transferable, it can be likened to a tax.

In the choice, account must be taken of both environmental and economic factors (efficiency)

The criterion for the choice of instruments is that the regulation must be efficient. In other words, the environmental goal must be achieved at lowest possible cost. When making the choice it is thus necessary to take account of both the impact on the environment and the economic consequences of the regulation.

When taxes are chosen, the problem is to fix the right level of taxation – the level that will ensure achievement of the environmental goal. Since the user must be expected to optimise microeconomically, taking account of the tax, a tax is an economically efficient regulatory instrument that will ensure socioeconomically optimal use of the resources.¹²

If, instead, one imposes quantitative restrictions on the environmental impact, one will achieve the environmental goal, but the economic result will be uncertain. To be economically efficient, the quantitative restrictions must be adjusted to the individual farm’s or production unit’s economic return.¹³

In the case of environmental regulation, quotas correspond to quantitative control. If the quotas are made transferable, trade in them between efficient and less efficient producers will ensure an economically optimal distribution of the quota restrictions. Quotas are difficult to use when the pressure of damage is differentiated, and to be effective, they must be continuously adjusted to the development of productivity in the sector.

However, in the choice of instrument, attention must also be paid to the administrative costs connected with use of the instruments. A general tax on pesticides is normally easy to administer, it can be imposed at the sales level, and it does not require monitoring of the producers’ behaviour – unlike a quota, for instance, which requires monitoring of the producers’ consumption. The same applies to restrictions on the treated acreage, observance of buffer zones and other quantitative restrictions on the use of pesticides. Experience from the EU’s milk quota also shows that a substantial administrative apparatus is required for a market for transferable quotas to work in practice.

Tax-based regulation is thus administratively less demanding than quantitative regulation and implies revenue for the state. The problem is illustrated in Figure 3.4, in which the MPV curve indicates the producer’s demand for pesticides at different price levels. The optimal consumption before tax (Q) is determined by the price P. If a tax is imposed, the price increases to T and consumption falls to Q’. The tax revenue thus corresponds to the diagonally hatched area, while the producer’s loss corresponds to the entire hatched area. The marked triangle expresses the producer’s loss with full reimbursement of the revenue.

The polluter pays principle

The use of taxes raises the question of reimbursement of the tax revenue to the industry. From an economic point of view, the revenue should be used in a way that achieves the greatest possible welfare for society. There is no direct argument for reimbursement - in other words, the regulation is in accordance with the polluter pays principle. If it is decided that the revenue shall be reimbursed to the producers, it is important for this to take place independently of the production in order not to reduce the efficiency of the regulation. However, it will obviously not be possible to give full compensation through such reimbursement for the simple reason that the total loss for the producers will be greater than the revenue (illustrated by the marked triangle in Figure 3.4.

(Figure text:
Pris = Price
Forbrug = Consumption)

Figure 3.4 can also be used to illustrate the effect of a quota. If the quota is put equal to Q', T will give the shadow price of the product, i.e. the diagonally hatched area indicates the amount the producer is willing to pay for the quota Q’ when the price is P. If the buying rights are initially allocated through a sale of quotas (transferable quotas), the state will receive the proceeds of the sale. If, on the other hand, the quotas are distributed free (for example, on the basis of previous consumption), the revenue will be reimbursed to the producer in relation to the size of the individual quota. In this case, the value of the quota remains in the industry. The full-drawn triangle in the figure expresses the amount the producer is willing to pay to increase his quota from Q’ to Q.

Tax-based regulation

One of the problems of using taxes is that they will normally be linked to the price of the products and that there is no direct relationship between the price of a product and its environmental impact. All else being equal, old agents whose development costs etc. have been fully depreciated can be marketed more cheaply than later agents. It should also be noted that the use of taxes and quotas may be difficult to administer within an open market like the EU if the restrictions on consumption are significant.

Choice of basis for tax

It has been discussed whether it would be better to base the tax on something other than the price of the agents. The following alternatives have been discussed:

1) Burden index. The ideal basis would be to calculate the environmental burden of the individual agents with average use in Denmark. However, use of a burden index leads to the question: Burden on what? In the formulation of a total burden index, one has to prioritise between different environmental impacts – for example, health, groundwater and biodiversity. These impacts have not been fully clarified, and how they should be weighted in relation to each other would to some extent be a political question. If one gives very high priority to one problem, e.g. groundwater, it may push the user towards agents with impacts in other areas.

2) Amount of active ingredient. A tax per kg active ingredient would favour mini-agents over older agents, but here, one has the same problem as in the case of value taxes that there is no direct relationship between weight and toxicity.

3) Standard dosage (treatment frequency index). The standard dosage is fixed by the producer of the agents but is controlled by the Danish Institute of Agricultural Sciences in Flakkebjerg. In some cases, the standard dosage differs between crops. With good plant protection practice, it is possible to use less than the standard dose. As pointed by Rude (1992), a tax on the standard dose might help reduce the intensity of treatment.

Of the above-mentioned bases for taxation, the standard dose seems, on the face of it, to provide a suitable basis for regulation. The system could be handled at the sales level, possibly combined with differentiation of the tax according to the potential damage. A weighted toxicity index could be designed for aggregation of different factors, including the amount of active ingredient. However, if the standard dosage were used as the basis for regulation, the legislative basis would have to be clarified because it is stated in the legislation that an objective and unambiguous tax basis must be established.

The Parliamentary Fiscal Affairs Committee (1998) has pointed out that standard dosage is not a clearly defined concept since it would, for example, vary with the crop. Therefore, for each of the more than 800 approved pesticides, a decision would have to be made concerning the tax basis. In addition, the tax rates have to be approved by law, which would mean considerable administrative work in connection with the appearance of new agents. The Parliamentary Committee states: "It thus seems doubtful whether the treatment frequency index (standard dosage) is preferable to the present value tax as the basis for taxation." However, it should be noted that Norway, for example, bases regulation on standard dosage.

Differentiated tax basis

From the point of view of administration, taxes are easy to handle because they can be imposed at the sales level. Present tax policy has already gone some of the way by differentiating between fungicides, herbicides and pesticides. If pesticides were divided into hazard classes, further differentiation of the tax basis would hardly be likely to cause any great problems. However, tax-based regulation would be less effective with geographically or regionally differentiated regulation of consumption because the possibility of buying pesticides at lower taxes in other parts of the country would make administration and control impossible. Where there is a need, for example, to protect particularly environmentally sensitive areas, differentiated regulation would be better ensured through quantitative regulation, with restrictions on the consumption of pesticides in the threatened areas.

It has been mentioned that instead of regulating pesticide consumption, one could impose a tax on conventional production. Such a measure would require certification of pesticide-free products along the lines of the eco-labelling scheme. However, for international acceptance, the tax would have to be non-discriminatory and taxes on uncertified products would have to be based on objective environmental and health standards (product standards). It is not certain that restrictions based on the way the products are produced (production standards) would be compatible with EU and WTO rules.

Quantitative regulation

Quantitative regulation advantageous with geographically differentiated regulation

The advantage of quantitative regulation is that one can regulate pesticide consumption directly. Quantitative regulation is therefore well suited to situations in which there is a need to differentiate the requirements in accordance with the natural conditions. The threat to the groundwater depends, for example, on the geological conditions. With uniform soil conditions, it can be the total burden in a given area that determines the pollution, but with differing soil conditions, there may be a greater risk of local pollution of groundwater deposits. A ban on the use of pesticides in environmentally sensitive areas is the solution here, but the question is, whether an either/or solution is optimal, seen in relation to the varying conditions that prevail in practice. An alternative solution might be to prohibit the production of crops that require treatment in the most environmentally sensitive locations.

The problem with quantitative regulation is that it is difficult to ensure an economically optimal solution. For quantitative restrictions to be economically effective, they must, as mentioned, be adjusted to the individual’s farm’s economic return, which is extremely difficult, if not impossible, in practice. Furthermore, it requires a substantial administrative apparatus to monitor the pesticide consumption at the individual farms. Quantitative regulation is therefore in general mainly suitable as a supplement to economic regulatory instruments in cases in which differentiated intervention is needed against the use of pesticides in agriculture.

Quota-based regulation

Transferable quotas are an efficient regulatory instrument…

Regulation on the basis of quotas is a form or quantitative regulation that will also ensure an economic distribution of pesticide consumption provided the quotas are made transferable. A national quota can be based on limiting the amount of active ingredient, the number of standard doses, the number of ecotoxicological doses, etc. Another possibility is a quota scheme in which one allots spraying quotas corresponding to a specific acreage fixed at national level. A system of acreage-spraying quotas would make it possible to keep pesticides away from areas designated by county and local authorities as environmentally sensitive areas and drinking-water areas of particular interest. The spraying quotas on the rest of the agricultural acreage could be made transferable in the same way as the set-aside obligations under the EU’s common agricultural policy.

SJFI (1998b) mentions the possibility of issuing spraying permits on the basis of the crop composition (corresponding to the codes of practice used to regulate use of nitrogen) or introducing a prescription system for pesticides under which farmers are only allowed to purchase pesticides for problem crops or in the case of a documented risk of fungal or insect attack. The advantage of this type of regulation is that it can be based to some extent on the existing early-warning system (PC-Plant protection). However, a prescription system makes great demands with respect to administration if minimisation of pesticide consumption based on timely intervention is to be achieved. Another fundamental problem with a prescription system may be achievement of the desired environmental effects unless it is actually designed as a quota for a number of standard doses or similar. SJFI (1998b) mentions, as a possible supplementary measure, prohibiting the use of pesticides in the autumn, when the risk of leaching is greater than in the growing period.

… but difficult to adapt to differentiated regulation

Just as with other forms of quantitative regulation, the use of quota schemes for differentiated regulation of pesticide consumption requires an administrative system that can identify the use of pesticides at a detailed level

Subsidy-based regulation

Subsidy-based regulation
Fixing compensation

Instead of regulating the environmental impact through tax-based or quantitative regulation, it would be possible to influence producer behaviour by subsidy-based regulation – for example, one could compensate farmers financially for not growing crops or for reducing their use of pollutants in environmentally sensitive areas. In order to fix the compensation, it would be necessary to know the loss to the farmer from reducing his production. However, the loss would naturally vary from farm to farm, so it might be difficult to ensure an economically optimal solution in which the cost to the public purse is minimised. Another possibility would be to invite tenders for quotas for production or use of pesticides in environmentally sensitive areas since this would ensure that it was the producers with the smallest losses that would participate in the scheme. The problem with such a tender system is that it might be difficult to fix exactly the level of compensation that would produce the desired reduction in the level of pollution (see the above discussion concerning quotas).

Subsidy-based regulation is not a relevant alternative as an instrument for general regulation of pesticide consumption. First, regulating pesticide consumption through subsidies¹⁴would be a costly solution for society and, secondly, subsidy-based regulation is fundamentally at variance with the polluter pays principle, since the cost of the scheme would fall on the consumer and not the polluter.

Liability

Legislation that opens the way for farmers, importers and/or producers to be made financially liable for any damage caused by pesticides can be regarded as a regulatory instrument, in line with quotas, taxes, etc. The purpose of liability rules is to ensure that the polluters pay the costs of any damage. However, the fact that the polluters have to pay compensation would also (as in the case of tax-based regulation) provide them with an incentive to reduce their use of dangerous pesticides.

The advantage of liability rules (compared with other forms of regulation, including taxes) is that they also provide producers and users with an incentive to find out about the environmental and health impacts of the pesticides used. Since it is the polluters that will pay the cost if they misjudge the damage that can be caused by a pesticide, they will have a clear interest in finding out more about that before they use the pesticide. With traditional regulation, the polluters simply react to adopted rules and taxes. If these prove to be based on too low an estimate of the dangerousness of a pesticide, it is society – and not the polluter – that pays the bill.

However, liability-based regulation has a number of drawbacks compared with traditional regulation. Firstly, it has to be possible to prove a clear correlation between use of the agent and the damage. This can imply substantial administrative costs in connection with bringing a case against the farmer and, at the same time, the strength of the incentive is reduced if there is a considerable probability of the polluter not being found guilty. Furthermore, the fact that it normally takes a long time to get from the act of pollution to the imposition of a compensation order in itself weakens the behaviour incentives.

… a possible supplement to other regulation

This could speak in favour of regarding rules concerning liability for damage caused by pesticides as a possible supplement to traditional regulation and not as an alternative. The instrument is particularly suitable for situations in which pesticides have been used unlawfully or not as prescribed.

3.3.3 Assessment

Assessment
Differentiated or general regulation

Regulation of pesticide consumption in the agricultural sector must as far as possible target the problems that are in focus. If the environmental problems concern the location of the production activities, this should be reflected in the choice of regulatory basis. For example, a general reduction of the use of pesticides on an unchanged treated acreage would presumably have a much less positive effect on flora and fauna than if the same reduction in consumption occurred through the establishment of buffer zones and a ban on spraying in environmentally sensitive areas. Similarly, the regulatory basis must take account of the fact that the risk of leaching to the groundwater differs from one agent to another and from one place to another.

The regulatory basis must reflect the variation in the environmental impact of the pesticide consumption

The choice of regulation (and the aim) must thus be sufficiently detailed to reflect the variation in the environmental impact of the pesticide consumption, both geographically and with respect to the individual agent’s impact. An opposing consideration here is the cost of administering the policy. The more detailed the formulation of a policy, the more it will normally cost to control and administer the policy. The choice of regulatory instrument must thus be based on an assessment of the efficiency of the schemes, taking into account the administrative costs.

Taxes (and transferable quotas)
are efficient instruments for reducing pesticide consumption

Taxes are normally easy to administer, and if the aim is a general reduction of pesticide consumption, a tax is also an efficient instrument, since it ensures that the reduction takes place in the economically most rational way. The same applies to the use of quotas provided these are made transferable. In this connection, the possibility has been pointed out of differentiating the tax according to the harmfulness of the agents and the risk of leaching – and of possibly graduating the tax in relation to the recommended treatment frequency. Another alternative mentioned is the use of liability as a possible supplement to traditional regulation.

Quantitative regulation preferable for differentiated regulation

A general tax on pesticide consumption would not fulfil the requirement concerning differentiated action against pollution in geographically delimited areas. For this it would be necessary to use quantitative regulation, with, for example, a ban on the use of pesticides in environmentally sensitive areas or restrictions on the cultivation of particularly burdensome crops. The problem with such a policy is that it would be difficult to ensure an economically optimal solution and that quantitative regulation is generally very demanding with respect to administration. The choice of political regulatory instruments therefore needs closer consideration, including consideration of the possibility of combining quantitative regulation with economic instruments.