Agricultural Policy Measures Addressing Non-Trade Concerns
Historical archive
Published under: Stoltenberg's 1st Government
Publisher: Ministry of Agriculture
Guidelines/brochures | Date: 16/10/2000
Agricultural Policy Measures Addressing Non-Trade Concerns
by Arild Vatn
| Discussants:
|
1. Introduction
1>>
>Agriculture takes place under very different conditions
throughout the world. Certainly this has to do with variations in
natural conditions - a factor of great importance for an activity
like agriculture. The sector is integrated in different ecological
systems. Its form and position further reflect variations in the
social and political evaluation of its role in society. While
agriculture is a multifunctional undertaking, this reveals that the
very substance of multifunctionality will vary from place to place,
from country to country.
The strong trend of globalization over the last couple of decades, has brought the issue of agriculture's various fnctions into the international debate. To the extent that public goods provided by agriculture are related to the production of food and fiber itself there is a potential antagonism between welfare enhancement and free trade. Further, there may - even in the case of separability in production between public and private goods - exist situations where support aimed at the provision of public goods has spill-over effects on the competitiveness of commodity production. We thus observe an intense debate over what is a reasonable and legitimate policy within the field.
My intentions with this paper are the following. First, I will make a visit to welfare theory, which constitutes an important basis for judging what are reasonable and legitimate policies. Here the concept of 'market failure' will be specifically evaluated. Second, I will try to characterize some of the multifunctional aspects of agriculture and discuss how well fitted our theory is to analyze these issues. Some of the controversies, stems in my mind from different understandings of the problems at hand. Finally, on the basis of these considerations, I will discuss some principles for the development of a reasonable policy within the field of multifunctionality. I use 'reasonable', because the level of complexity is of such a high degree that to talk about optimal policy is to create an illusion.
In a short paper like this, a comprehensive treatment is preposterous. My main aim is therefore to emphasize a set of perspectives and illustrate how the analysis in my mind should be carried out.
2. Welfare theory and the concept of
'market failure'
Welfare theory is an important - but not the only -
theoretical basis for evaluating economic policies. It plays a
major role in the discussions over the gains of freer trade. Under
the assumption that all goods can be traded in markets (i.e.
exclusiveness and rivalry in consumption), that products are
homogeneous, resources are mobile and transactions costs
>
2>> >are zero, it is shown that competitive
markets result in the highest level of aggregate welfare if
measured as the sum of individual welfare. Given the assumptions,
trade barriers or production subsidies are reducing aggregate
welfare. State 'intervention' is, however, permitted if markets
fail to stand up to the defined presumptions (Bator 1958). Public
goods (non-exclusiveness and non-rivalry) is the prime case where
regulation or public supply is accepted (Boadway and Bruce 1984;
Randall 1983).
It is thus a very 'heroic' type of market construct which functions as a bench-mark for evaluating policy within economics. Let us term it the 'market vision'. The vision is not only an ideal model. It also tends to empower our ideas about 'real' markets with the capabilities of the same smooth resource allocation as this theoretical construct. When we talk about markets, we tend to think in terms of the vision. 3>> In my opinion this perspective does not create a good understanding of the role of the state nor the functioning of the public sphere. It does not perceive the role in defining the "rules of the game" properly. Given real world complexities, there is a need for reasonable weighting of different needs and interests that goes far beyond constructing solutions within the image of the still unattainable market vision.
Instead of defining political issues with reference to 'market failure' - giving markets an exalted position - one should accept that markets are a very diverse set of structures, diverging to a varying degree from the 'vision'. This variation is intelligible due to variations in commodity characteristics, transaction costs, but also in culture and defined rights. Further, this varied set of institutional arrangements is just some among many others able to fulfill human needs and to produce the necessary trade-offs between various interests. There are two important issues at stake here.
The first issue is about realism and consistency. If the market vision is far from attainable, it becomes a poor standard for policy evaluation. According to Coase (1960) 4>> and Dahlman (1979) one should instead make comparative analysis on the basis of attainable structures and realistic assumptions about goods, markets, and I would add human capabilities. To put externalities and public goods into a box called market failure, becomes problematic as soon as one realizes that any market fails the 'vision'.
The second issue is about formulating rules over what is a proper or legitimate role for the collective or the state. According to Bromley (1989), the standard procedure is able only to identify three types of policies: increasing efficiency, redistributing income and supporting 'rent seeking'. Of these only the first two are considered legitimate. There is however a fourth possibility - which may even be a dominant state responsibility. That is to create economic opportunity by changing institutions. This implies to change the rules of the game such that new sets of interests or needs get protection by becoming a part of the social welfare function. This is different from standard efficiency enhancement because it defines new standards for what is to become efficient. It is different from mere income distribution because it is about creating new possibilities. It is different from 'rent seeking' because it has become legitimate through a constitutionally defined set of political procedures. Certainly, this kind of institutional processes are necessary in a world differing from the 'market vision'.
Bromley (ibid, p. 128) formulates it the following way: "To take any collective action at all is to presume a particular social welfare function, for collective action is simply a response to one or more groups' perception of what ought to be done. Moreover, collective decisions are undertaken with the explicit notion in mind of altering the position of one group of individuals with respect to another group. This recognition needs not be viewed with suspicion, as it often is by some economists."
When deciding about international trade issues, one faces a problem here since they appear in an institutional setting where there is no legitimate political body to make the necessary balancing of needs and interests - to establish a social welfare function. From the standard perspective, making trade easier would be evaluated as efficiency enhancing while forming rules in a trade restricting manner would be considered supportive of the interests of the few at the cost of the many. It would be mere 'rent seeking'. It could, however, also be a way to legitimately reallocate economic opportunity. In to-days institutional setting the evaluation of this has to be based on agreements between the involved states concerning the rules for defining what interests or needs should get protection.
This goes beyond the concept of market failure. It is about defining which needs or interests markets or other institutional structures should serve. Thus these kinds of policy choices are prior to whatever structure is chosen. The challenge is to find structures that are able to serve the chosen goals as well as possible. Policy has to come before economics since institutions define what will become efficient. As soon as one accepts that there is a continuous need for social choices, one is forced to accept that the social welfare function is an essential part of policy making . > 5>>>
On the basis of this, I will emphasize that the role of economics must be more about explaining mechanisms than calculating the optimal choice. It must be more about understanding what specific problems surfaces from the character of the issues at hand than to make analysis on the basis of the list of 'heroic' assumptions. This is a challenge since it forces us to be critical to the way we formulate the analysis and the utilization of standard technical devices that has become part of our repertoire as economists. 6>> Even changing the assumptions one at a time - as is often the case in the 'second best' literature - may yield insights of only limited value. 7>>>
The points made so far are general. Still, their importance is enhanced when we move from the analysis of standard commodity markets to studying an issue like multifunctional agriculture. In this case all standard assumptions as listed at the beginning of this section seems challenged. Goods cannot be traded in markets, they are not homogeneous, resources are not (fully) mobile and transaction costs are far from zero. Further, many of the goods are jointly supplied or jointly consumed. That may be a part of their inherent characteristics even. Finally, the goods we study here seem often to be relational - i.e. the value of one good is dependent upon the status of another or the value of one farmer's investment in the production of a public good depends on the investments made by his neighbors.
3. The characteristics of multifunctionality in agriculture
Moving into a world as outlined above is
challenging. My aim here is not to cover all important
relationships. Instead I will outline a perspective and make some
illustrations of what kind of conclusions that can be drawn.
3.1 Multifunctional aspects of
agriculture
The idea that agriculture affects the well being of
the society beyond the value of its production of food and fibre is
not a new one. The WTO agreement from 1994 this issue surfaced in a
paragraph on so-called 'non-trade concerns' (Noewegian Ministry of
Foreign Affairs 1994).
8>> Here both food security and environmental
protection are mentioned. The concept of multifunctionality has
also gained position lately, especially within the OECD (OECD 1998a
and 1998b). Language is not unimportant here. While the concept
non-trade concerns may be understood to drive a wedge between
production and a set of 'social issues', the concept of
multifunctionality may be perceived to establish a link between the
production of food and fibre and the production of public
goods.
The debate over non-trade concerns or multifunctionality is basically a debate over the legitimacy of various goals and measures within agricultural policy. The definitions of these concepts and their position in the upcoming WTO agreement will define what in the end become legitimate economic opportunities for the member countries. The perspective underlying the definition of multifunctionality becomes crucial to the result. I will especially emphasize the following:
Will multifunctionality at all become part of a new agreement - i.e. will it be defined as a legitimate economic opportunity?
If so, which elements or goods will be accepted as part of the concept?
How will the relationships between agricultural production and the production of public goods be perceived?
The answer to these issues will have to evolve through reasoned discourse between the states involved. These states have different interests, and the conclusion will have to be reached through negotiations. 9>> In my view, the most difficult issue concerns defining the relationships between commodity production and the production of collective goods (bads). The stronger the connections, the more difficult it will be to keep trade and 'non-trade concerns' apart. Certainly, there is an inherent drive stemming from the aim of WTO to try to keep a distance between the two.
From what is said above, it should not come as a surprise that there exists no generally accepted list outlining what the multifunctional aspects of agriculture are. The following list is developed on the basis of aspects mentioned in the literature. Still, I do not consider it to cover all aspects that in the end ought to be considered. It is only put forward as a basis for the subsequent analysis:
Environmental effects
Landscape
-Biological diversity
-Recreation
-Aesthetics
Cultural heritage
Pollution (changes in matter cycles; genetic pollution etc.)
Food security
Availability
Aspects of food quality/sanitary conditions
Rural concerns
Rural settlement
Local economic activity
Social concerns
The list covers both public goods and bads, negative and positive 'external effects'. It is open for debate whether multifunctinality should only cover so called positive 'external effects' of agriculture - i.e. elements like pollution should be handled as a separate problem.
In my view the distinction between positive and negative 'external effects' is a difficult one. It cannot be defined on pure physical grounds, as seems to be the position in most of the literature (Vatn and Bromley 1997). As is emphasized also by Scheele (1999), the distinction can only get meaning though a defined set of rights. Thus the demarcation line between what is positive and what is negative has to be part of the political agreement itself. Again it is a political decision to define what becomes an economic opportunity and for whom.
The following analysis does not demand a certain definition of the concept - i.e. what is inside or outside. It does however take the position that the various elements of a multifunctional agriculture are interlinked. They are - at least to a very large extent - inherent features of the agronomic system, and cannot be viewed as separate or independent elements which one can freely combine.
3.2 Public and private goods - a model
The linkages between private and collective goods in
agriculture can be illustrated with the help of the following
figure.
The agricultural system produces both private and public goods. The volume and type (incl. qualitative aspects) are determined both by the level of inputs and the way inputs are combined - i.e. the method of production. Further there may also exist linkages between the output of private and public goods. These linkages may work both ways. As an example production of private goods may be part of the realization of food safety or rural viability, while public goods like biodiversity may affect the productivity of food or forest production systems.
Figure 1.Agriculture and the production of private and public goods
All linkages may in principle be both positive and negative. Further, there will normally exist linkages also between different levels and qualities of public goods. Finally, many of the public goods listed above may be produced without a connection to agriculture. Rural settlement is one example, aesthetic landscape effects may be another. Actually there are two important issues at stake here.
First, one must consider whether the productions of each functions is directly linked to agriculture or if the good can be separately produced. In relation to that one must also ask if the quality is the same independent of how the good is accomplished. This issue is easily illustrated by looking at the issue of landscaping. Is the product differently valued if it is part of an agricultural system compared to an open landscape produced without any connection to agriculture at all?
Second, the aspect of costs is important. How can the kind of public goods studied here be cheapest provided? Several of the goods in the above list may be produced outside agriculture. Many of them may further be produced cheaper this way if separately delivered. This does not mean that agriculture in total is the most costly solution though. In general, if goods can be jointly produced, there is a potential for cost savings (Hoel and Moene 1993; Shumway et al. 1984). This means that even in a case where there exists a less expensive way to separately produce each function in the above list outside agriculture, this industry may be able to deliver the cheapest total package if jointness is sufficiently present. This is often overlooked in the literature since focus dominantly has been on one aspect at a time. Given that agriculture already produces some public goods, the extra cost of adding others to the list may be low. The relevance of this conclusion seems to be heavily dependent on the type of production methods used, though.
In relation to the question of what is the best and cheapest solution, the cost of formulating and operating a policy is also important. Is it possible to produce the public goods by redirecting agricultural policy through changes in input/output prices, restrictions on technology etc., or is it better to use instruments directly focussed on the production of the public good. Here the issues of competition, jointness, and complementarity in production and consumption is important. The same goes for the trade-off problem between precision and administration costs.
3.3 Competition vs. jointness in production
The relationships inherent in Figure 1 are complex. That is especially the case if we acknowledge the long list of public goods and the potential interlinkages between them. In developing some basic principles, simplifications are necessary. I would like to start by looking at the potential relationships between agricultural production (y) and the production of a specific public good (z) based on the use of the same production factor, in this case the use of land. Figure 2 illustrates three classes of situations - i.e. three different 'production possibility frontiers' (PPFs).
Figure 2.Competition and jointness in production of private (y) and public goods (z).
The first relation is the standard one where production of y and z compete in their use of the factor land. Increased production of one implies a reduction in the production of the other. One example here would be land used for production of food and land used as space for wildlife, footpaths etc. Here the problem is to decide how much land to use for y and for z.
The second relationship is the extreme opposite. Here, production of the private good in all relevant ranges (OR) leads to a (potential) increase in the production of the public good. We experience joint production, i.e. the two goods are a non-separable outcomes of the same production process. The level of one product (z) does not necessarily need to be proportional to the level of the other (y) as is also indicated in the figure. 10>> One example of joint production would be production of food and the creation of open landscape. Another might be food security.
The use of the PPF concept is a bit difficult in this case. Understanding the curve in panel (2) as a standard production possibility frontier, the whole area boarded by the frontier covers the potential set of solutions - i.e. a standard production possibility set. Our interpretation is instead that of a functional relationship. 11 The form of the relationship has implications for policy. If z is a function of y, the sole problem becomes one of choosing the right level of y. If the relationship is less fixed, separate policies for the production of z may also be warranted.
Panel 3 of figure 2 shows a third - or intermediate - type of relationship. 12>> Up till S there is a positive relation between the production of y and z. To the right of S the two goods compete. Examples of this kind of connection might be biodiversity in the pine regions of the temperate zone. Up till a certain level agricultural production results in increased biodiversity - i.e. the effect of converting forests to open agricultural space. However, the more intense the area is utilized, we will observe less variation in plant cover, removal of wetlands and vegetation between fields etc. Further, forest biotopes might be reduced to a level below what is considered preferable. We move into a region where y and z compete as to the use of land. The questions of what an optimal policy might be, will vary between the regions to the left and right of S. 13>>>
Which kind of frontiers or production relationships that exists, will depend on the type of production methods used (see also Buckwell 1989; Russell 1993; Howitt 1995). This complicates the analysis - not least when searching for an optimal or reasonable policy. Figure 3 illustrates the potential relationships between the same private (y) and public (z) good with two different methods used for producing the private good.
Figure 3. Production possibility frontiers for different production methods
This opens up for the exploitation of new possibilities, as relationships of this type may imply that (part of) the policy ought to be directed towards the methods themselves.
I would also like to draw attention to the fact that what constitutes a positive link at one level of resolution, may be negative at another. Let us use the cultural landscape as an example. Two levels of resolution may be (a) the overall openness of the landscape, and (b) the mosaic of different landscape forms, elements etc. which this landscape consists of. While the relationships between the private and public good concerning (a) may in general be positive, they may be competing if we look at the conservation of the public good elements of level (b).
3.4 Jointness in consumption
Jointness does not only exist in production. In our
case it is important also in consumption. Let us go back to the
example with land used for food production or for footpaths. While
these are competing in resource use (land), they are to a large
degree joint in consumption. The value of the path depends on which
landscape it is a part of. If it is openness, the pattern of
fields, and hedges etc. that together creates the surroundings
necessary to make the path interesting, one cannot distinguish the
value of the path from the character of the landscape. While a lot
of products are complementary in consumption - like 'eggs and
bacon' - it is distinctive for this case that each 'consumer'
cannot chose the combination of elements freely. At best she can
chose among a specific sets of combinations, i.e. various
landscapes. This is what distinguishes joint from complementary as
defined here.
In this case the interconnectedness between private and public goods, and between different types of public goods, stems from the fact that they are relational - i.e. the value of one good depends on the level and form of another. This is typical for goods where quality aspects are of importance.
3.4 Non-homogeneous goods
The final characteristic I would like to draw
attention to in this case, is the concept of
non-homogeneity. While we in most analysis of market
transactions assume goods to be homogeneous, this can hardly be the
case for many of the goods produced by agriculture - actually both
private and public goods. This is the basic feature of food quality
and sanitary conditions. It is however typical also for most
landscape values where variation is an important part of what
creates value.
The existence of non-homogeneity raises two important issues. First, it increases the need for information substantially. Second, and even more important, it influences the working of markets as allocation mechanism more in general since the presumption about markets as competitive tends to break down.
4. Policy measures
4.1Trade and the environment
The increased focus on the multifunctional role of
agriculture has evolved parallel with the development towards a
freer trade in agricultural products. The question is raised: Does
free trade impede the possibilities for supplying/preserving the
various public goods related to agriculture? Is it neutral in that
respect, or does it instead increase the potential for welfare
enhancement even within the realm of the environment? (Esty 1994).
This debate relates partly to the effect of freer trade on the use
of resources. Partly, it opens up for a debate on the possibilities
for different countries to develop national policies for protecting
the environment and increasing or preserving the production of
public goods.
It is interesting to observe that so far the debate about environmental effects of trade has focussed mostly on so-called 'negative externalities' like pollution and not so much on the provision of public goods. In the case of agriculture one line of argument is that liberalized trade in most countries will result in reduced product prices and by implication less environmental stress since lower product prices imply lower production and less intensive use of inputs. Further, environmental gains would follow indirectly from a redistribution of production between countries such that more would be produced in countries with the best natural conditions, higher efficiency in production etc. (for some of these arguments see Esty 1994; Anderson and Strutt 1996; Runge 1998).
The last line of reasoning is partly in resemblance with the so-called Kuznets curve argument. Trade increases growth. Up till a certain level there is a negative correspondence between growth and environmental quality. However, after a certain level of GNP per capita is reached, it is argued that the environmental stress decreases.
This reasoning is heavily debated, both on theoretical and empirical grounds (Arrow et al. 1995; Jackson 1998). Accordingly we observe the argument that growth, and by implication trade, is a general threat to the quality of future environments (see Esty 1994, and Anderson et al. 1995 for an overview of the arguments). Further, it is argued that free trade complicates the possibility for each country to impose environmental measures locally (Haavelmo and Hansen 1991).
While it is broadly accepted that reduced product prices will tend to diminish the use of inputs that may cause environmental damage (land, fertilizers, insecticides etc.), it is also argued that the relationships are neither very strong nor necessarily unidirectional. Further, increased concentrations following tougher international competition may lead to negative ecological effects, especially in agriculture (Runge 1998). Increased transportation is another argument against the existence of a positive link between trade and environmental quality (Esty 1994). Finally, it is claimed that freer trade enhances the speed of technological change, where especially questions related to genetically modified organisms has become a central issue lately.
Evaluating the observed trends, several authors conclude that policies specifically directed towards addressing environmental issues are necessary. Arrow et al. (1995) make the claim in general, while Buckwell (1989), Dragun (1998), and Hodge (1992), are among those positioning themselves similarly concerning potential gains and losses from a 'deregulated' agriculture. I support this conclusion, and would further like to add:
-As long as there exists relationships between free trade and the creation of 'externalities', one cannot isolate the one area of policy from the other. Such a two stage analysis will have to become incoherent and inefficient (Vatn and Bromley 1997).
-While free trade may or may not contribute to an overall reduction in pollution, it cannot produce incentives securing the provision of public goods from agriculture. They must either be directly paid for (the case of competition with other resource use), or indirectly provisioned through the level and form of agricultural produce (the case of joint production). It is this last conclusion that is the most controversial.
In relation to this it is important to observe that the existing greening of the agricultural policy both in the US and the EU seems to evolve more as an response to old problems like 'overproduction' than to genuine care for the environment (Potter 1998). There is a substantial need for redirecting policy. There is a need to formulate a policy much more consistently facing the future problems related to the environment and public goods provision. This may even demand the creation of a 'new farmer' with an identity less related to commodity production and more focussed on public goods.
Following the above argument, I am still not in favor of a strategy that just 'decouples' support from agricultural production. As long as there is a connection between the provision of public goods and the production of food, there is a future role even for differentiated product prices. In countries where agriculture is not able to compete with imports, securing the existence of a certain level and composition of agricultural production may be an important element in securing its multifunctional role.
Production costs vary over the globe both due to variations in natural conditions, labor and capital costs, skills etc. This forms the basic argument for gains in trade. It also implies that free trade may have rather substantial effects on agricultural production especially in more marginal regions. It is no surprise that one finds the largest resistance against liberalization in these types of regions. Let me proceed with a simple formalization of the above arguments.
4.2 Case I: Production of private and public goods are
competing
What could be a reasonable policy for securing the
supply of the kind of public goods considered here in countries
where agriculture faces problems with being internationally
competitive? Let us start with the situation where food production
is competing with the use of resources either in the production of
other commodities or public goods. Let us further assume that the
production of food can be separated from the production of all
other goods. In this case the economic argument is that each
product should receive payment equal to its marginal value. The
ceteris paribus implication of price support on
agricultural products is that there will be too much production of
these goods at the expense of others. Further, if these other goods
are of the public kind, they will be under-supplied if producers do
not receive any payment for them at all. This effect will be
enhanced if food production is supported, and can only be
counteracted with an extra payment for producing the collective
good. The optimal solution is illustrated in Figure 4.
Figure 4. Optimal pricing of a private (y) and public (z) good. Py is the price of the private good, Pz of the public.
Given that the optimal combination of y and z are y* and z*, the relative prices of y and z should equal the negative of the slope of the PPF in that point. While the public good must be nationally supplied, the private good may be imported if national production does not compete at the price of Py. This is the standard case where free trade in commodities and public support of collective goods produces the wanted combination.
4.3 Case II: Jointness in production
If the production of the public good is linked to
the production of the private, the choice of policy measures
becomes different. In a country whose agriculture is
internationally competitive, one may not need any specific policy
concerning the jointly produced public good at all. In a country
where national production of the private good is not able to
compete with imports, the situation is different. Figure 5 can be
used to analyze this situation.
The upper part of the figure (part I) shows total costs and gains at sector level. TIy(wm) denotes total income from producing the private good at world market prices. TCy is the total costs accompanied with that production in our country. Finally, TIz depicts the total gain from the public good z following from the production of the private good. We see that the total value of this good is decreasing beyond point S. This implies that the two goods y and z are competing in that region (confer panel 3 in figure 2). The curvature of TIz will also be influenced by changes in the marginal value of z as z increases (e.g. the position of S will most probably deviate from that of figure 2).
Figure 5.Optimal pricing in the case of joint production of private (y) and public (z) goods.
The lower part of Figure 5 (part II) gives a principal depiction of the marginal values. While marginal income of y - MIy - is constant at world market price P 1, marginal costs - MCy - are increasing. Marginal costs of producing y is below world market prices only up till y', as its agriculture is not very competitive in the production of y. With no additional measures, production of y will be y' while the additional value of producing z is V'z. This is, however, not optimal. There is a region to the right of y' where the marginal income from the production of z - MIz - is greater than the separate loss following the production of y (MCy - MIy). The combined marginal income from producing both y and z - MIy,z - crosses MCy first at y* which is the optimal level of production considering the joint effects of y and z. The accompanying total value of z is V*z. An increase in the price of y till p 2 will produce the optimal solution.
Maybe the most surprising result of this analysis is that the optimal solution is not in the region where the two goods y and z compete over the use of land. This is because the optimal production of y in isolation, given world market prices, lies to the left of S. In this case only a positive contribution from the production of z can motivate higher production than y'. Certainly the form of TCy and TIz is crucial to the conclusion.
4.4Case III: Jointness in consumption
Jointness in consumption also poses some challenges
to analysis. Let us go back to case I (Figure 4). While the goods y
and z were competing in production, they may be joint in
consumption, as previously illustrated with footpaths through an
agricultural landscape.
In this case, the preferred combination of fields and footpaths cannot in general be obtained by market allocation of food production and a specific payment for footpaths. First, for a country whose agriculture is not competitive internationally, there will not exist an agricultural landscape to 'consume' visually. Second, even when agriculture is competitive, the internationally determined prices may not produce the kind of agriculture that gives a landscape that one would like to consume.
In this kind of situation, two types of measures need to be considered: a) payments to secure the existence of agriculture (magnitude), and b) payments or various restrictive measures to influence what form of agriculture will be undertaken.
Point a) relates to the so-called participation constraint. Farming must generate enough revenue to keep enough businesses in operation. 'Decoupled' support (lump sum transfers) may be one way of accomplishing that. It can, however, hardly function as the sole measure since it does not create any incentive to keep the landscape in the wanted shape. That is not the case with (undifferentiated) price support either. If open landscape is important, as in this case, payment per acreage is a relevant supplement. If animals constitute an important part of what makes the landscape valuable, payments per head may be another option. Differentiated product price support may also be a relevant measure in 'constructing' the kind of landscape that is wanted. Again the degree and form of jointness drives the conclusion.
In this example only two elements were considered, the production landscape and the footpath. The next step of analysis would be to introduce other 'non-agricultural' elements of importance like hedges, other kinds of vegetation corridors, stone walls, waterways etc. to increase the visual and biological value of the landscape. Here the question is whether it is rational to pay for such elements one by one or if it is better to combine more coarse forms of payment like the ones mentioned above with restrictions on agronomic practices and landscape changes. Here, we approach the issue of regulation and administrative costs - i.e. transaction costs.
4.5 Transaction costs and precision
Transaction costs are defined as the cost of
information gathering, of making agreements and monitoring and
policing these agreements (Dahlman 1979). Assuming transaction
costs to be zero, makes any policy viable. We can measure
quantities and qualities, tax or pay with no cost. We can thus
reach any goal we want with 100 % precision.
The world of zero transaction costs is certainly a strange one. Still, in most analyses these costs are assumed not to influence the conclusions. This is a rather curious position, though, since with zero transaction costs there will be no difference in efficiency between any structure - be it 'markets' or 'regulatory approaches'. It is only when one accepts the cost both of market transactions and political regulations to be positive that structure may make a difference. Since transaction costs are hard to assess, it still becomes difficult to distinguish between policy options.
What does this imply for our analysis? First, it means that not all policies that seem favorable will be worth pursuing. Second, it implies that we have to make trade-offs between transaction costs and precision (Vatn et al. 1996). This means that the famous Tinbergen (1952) conclusion of one measure for each goal actually does not hold. It may be favorable to reduce precision somewhat to make the administrative and regulatory costs less demanding.
This has important implications for our analysis. Since it will be too costly to pay/tax or regulate each and every element of the public goods aspects related to agriculture, we must search for tenable compromises. This is hard, because we have to leave the 'market vision' as a reference point, making the analysis more realistic, but much more difficult to structure.
5 Sketching a solution
Even though I have tried to simplify the analysis as
much as possible, the picture of the multifunctional agriculture
and the related policy options has become a complex one. A policy
in this field has to take into account that agriculture both have
positive and negative effects on the environment, that public goods
may be a joint product with food production, but that it may also
be competing. Still, if it is competing in production, it may be
joint in consumption. The goods seem to be largely characterized by
non-homogeneity and with a high degree of spatial variation
(Scheele 1997). Finally, we need to consider the effect of
transaction costs.
To get much further we need more thorough assessment of the empirical relationships involved. Still, I find it possible to sketch some principles that I believe will be important for future policy formulation:
Recoupling has to supplement or even substitute decoupling. From my perspective recoupling not only means coupling payments to the production of public goods - which is the normal use of the concept in to-days debate over agricultural policy. I also have in mind a redirection of agricultural practices in a way such that public good provisioning becomes much more of an inherent or integrated part of agriculture itself. This means recoupling private and public goods, and will imply a radical change in the perception and role of agriculture.
In that process I find it necessary to combine three types of measures:
payments for public goods/positive 'externalities' and taxes/regulations to reduce negative 'externalities'
the use of regionally differentiated product prices
the use of management contracts/requirements to further secure the provisioning of public goods
Informative measures and locally based public good provisioning campaigns is also needed to develop the necessary insights and personal commitment among farmers to natural conservation.
While a) focuses on the direction of change as an overall principle, point b) focuses on the instrumental type measures necessary. Point c) emphasizes the need for constructing/recreating a culture in the farming society where nature conservation in a broad sense becomes a significant part. I do not believe that it is possible to direct all choices in farming with the use of economic, judicial or other technical instruments. Complexity and local variation is far to great for that.
The above list is in general based on the assumption that directing all types of choices involved in a complex utilization of natural processes is far to demanding (transaction costs). An alternative path of development is to try to create positive links between farming and public good provisioning. Thus, recoupling becomes the core of the whole restructuring of farming. It is more of a holistic approach than the kind of piecemeal strategy following from traditional policy advice. In doing this, one has to consider the whole list of public goods and bads associated with agriculture and be aware of the linkages. One must also consider what is best left out as a responsibility for agriculture.
Even though recoupling implies constructing a win-win situation, this condition cannot be created without the use of various technical instruments. Concerning b1) the major challenge will be to balance the search for high precision with keeping transaction costs down. Even though it in some cases will be possible to pay directly for the good, it will in many instances be necessary to find 'proxies' for the public good or bad to which payment/taxation can be attached.
In the list under b), regionally differentiated prices is the most controversial. This should not be so. To the extent that the public good is jointly produced together with the private, this is the most rational solution as long as production costs and public good values vary across the globe. The potential for reduced transaction costs is also part of my argument.
A potential neative effect of this proposal could be increased pollution in areas with prices above those of the world market (increased nutrient losses, pesticide use etc.). First of all, there is no reason why varied product prices across the globe, with its associated pattern of production, should yield more pollution than a solution with a uniform price. Concerning pollution, local conditions like recipients, climate and soils are as important as product prices. Secondly, to the extent that the production of private and public goods are joint, increased pollution may be a desired cost to pay for obtaining a higher level of the public good. Remember that as the situation is defined in this case, there are two public effects linked to food production, one positive and one negative. The optimal solution will be found considering the net effect of the two, not just by considering the negative 'externality'. It must be emphasized, though, that all arguments made here are principally formulated. They do not give specific justification neither to the overall level nor the variation of existing price supports across the globe.
The proposed use of management contracts (b3) is justified for two reasons. First, it will in my mind be the main measure obtaining management practices which are well fitted for public goods provisioning. Thus the management contract has to define core features of these practices. Second, desired precision can hardly be reached through the use of measures of type b1) and b2). The complexity is far too great for that. Thus some kind of contract is necessary especially to define the qualitative aspects of goods paid for under b1).
The final element - the informative part - is necessary to create commitments, insight and positive attitude to the whole redirection of agricultural policy. The changes will be substantial, and there will be many possibilities for farmers to to circumvent the rules if they so want. Thus, it is very easy to get into a kind of 'hide and hunting' game between farmers and authorities. The analysis of Lowe and Ward (1997) shows that there exist a great potential in the farming communities themselves for redirecting behavior as long as one is able to establish resemblance between farmers' insights and visions, and the goals of public policy. This is maybe the greatest challenge for the future, also for the farmers' society.
References
Anderson, K. and A. Strutt, 1996. On Measuring the Environmental Impact of Agricultural trade Liberalization. In Bredahl, M.E., N. Ballenger, J.C. Dunmore and T.L. Roe (eds): Agriculture, Trade, and the Environment: Discovering and Measuring the Critical Linkages, pp 151-172. Westview Press, Boulder.Anderson T.C., C. Folke and S. Nystrøm, 1995. Trading with the Environment. Ecology, Economics, Institutions and the Environment. Earthscan.
Arrow, K., B. Bolin, R. Costanza, P. Dasgupta, C. Folke, C.S. Holling, B.O. Jansson, S. Levin, K.G. Mäler, c. Perrings and D. Pimentel, 1995. Economic growth, carrying capacity, and the environment. Science, 268:520-521.
Bator, F.M., 1958. The Anatomy of Market Failure. Quarterly Journal of Economics, 72:351379.
Boadway, R.W. and N. Bruce, 1984. Welfare Economics. Blackwell, Oxford (UK), Cambridge (US).
Bromley, D. W., 1989. Economic Interests and Institutions. The Conceptual Foundations of Public Policy. Oxford: Basil Blackwell.
Buckwell, A., 1989. Economic Signals, Farmers' Responce and Envireonmental Change. Journal of Rural Studies, 5(2):149-160.
Coase, R.H., 1960. The Problem of Social Cost. The Journal of Law and Economics, 3:144.
Dahlman, C.J., 1979. The Problem of Externality. The Journal of Law and Economics, 22:141162.
Dragun, A.K., 1998. The Agricultural Production of Environment. Some Nordic issues. Swedish Journal of agriccultural Resource, 28:49-55.
Eggertsson,T., 1990. Economic behavior and institutions. Cambridge Survey of Economic Literature.
Esty, D.C., 1994. Greening the GATT. Trade, Environment, and the Future. Institute for International Economics, Washington D.C.
Gravelle, H. and R. Rees, 1981. Microeconomics. Longman, London and New York.
Haavelmo, T. and S. Hansen, 1991. On the strategy of trying to reduce economic inequality by expanding the scale of human activity. In Goodland R., H. Daly, S. El Searfy and B. von Droste (eds): Environmental Sustainabile Economic Development: Building on Bruntland, pp 41-50.
Heady, E., 1952. Economics of Agricultural Production and Resource Use. Prentice-Hall, inc., New York.
Hodge, I., 1992. Supply control and the environment: the case for separate policies. Farm management, 8(2):65-72.
Hoel, M. and K.O. Moene, 1993. Produksjonsteori. Universitetsforlaget, Oslo.
Howitt, R.E., 1995. How economic incentives for growers can benefit biological diversity. California Agriculture, 49(6):28-33.
Jackson, T., 1998. The Unfinished Symphony. An evolutionary perspective on the conception of sustainable development. Paper presented at the 2 nd> Conference of the European Society for Ecological Economics. 23 s.
Kuhn, T.S., 1970. The Structure of Scientific Revolutions. Chicago, University of Chicago Press.
Lowe, P. and N. Ward, 1997. The Moral Authority of Regulation: the Case of Agricultural Pollution. In Romstad, R., S. Simonsen and A. Vatn, (eds): Controlling Mineral Emissions in European Agriculture. Economics, Policies and the Environment, pp 59-72. CAB International, Oxon UK.
Lynne, G.D., 1988. Allocatable Fixed Inputs and Jointness in Agricultural Production: Implication for Economic Modeling: Comment. American Journal of Agricultural Economics. 70(4):947-949.
Norwegian Ministry of Foreign Affairs, 1994. St.prp nr 65 (1993-94). On the results of the Uruguay Round (1986-1993) and the agreement of ratification of the creation of World Trade Organization (WTO) etc. Oslo, Norway.
OECD, 1998a. Meeting of the Committee for Agriculture at Ministrial level, 5-6 March 1998. Agriculture in a changing world: Which Policies for Tomorrow. Internett address: http://www.oecd.org//news_and_events_release/nw98-22a.htm.
OECD, 1998b. Multifunctionality: A framework for policy analysis. AGR/CA(98)9, Directorate for Food, Agriculture and Fisheries, Committee for Agriculture, OECD, Paris.
Potter, C., 1998. Against the Grain. Agri-Environmental Reform in the United States and the European Union. CAB International, Oxon UK.
Randall, A., 1983. The Problem of Market Failure. Natural Resources Journal, 23(1):131148.
Runge, C.F., 1998. Emerging issues in agricultural trade and the environment. Working paper W98-3, Center for International Food and Agricultural Policy, University of Minnesota. .
Russell, N.P., 1993. Efficiency of rural conservation and supply control policies. European Review of Agricultural Economics. 20:315-326.
Scheele, M., 1997. The Decomposition Approach: Spatially Differentiated Analysis and Implementation of Environmental Strategies. In Romstad, R., S. Simonsen and A. Vatn, (eds): Controlling Mineral Emissions in European Agriculture. Economics, Policies and the Environment, pp 41-58. CAB International, Oxon UK.
Scheele, M., 1999. Environmental Services Provided by Agriculture. The Setting of Environmental Targets and Reference Levels. Paper presented at the workshop on 'Non-Trade Concerns in a Multifunctional Agriculture'. Sanner, Norway, 9-11 March 1999.
Shumway, R.C., R.D. Pope and E.K. Nash, 1984. Allocatable Fixed Inputs and Jointness in Agricultural Production: Implication for Economic Modeling. American Journal of Agricultural Economics. 6614):72-78.
Tinbergen, J., 1952. On the Theory of Economic Policy. In: Contributions to EconomicAnalysis, no 1. Chapters 1-5. North-Holland, Amsterdam.
Vatn A. and D.W. Bromley. 1997. Externalities - A Market Model Failure. Environmental and Resource Economics9:135-151.
Vatn, A., L.R. Bakken, M. Bleken, P. Botterweg, H. Lundeby, E. Romstad, P.K. Rørstad and A. Vold, 1996. Policies for Reduced Nutrient Losses from Norwegian Agriculture. Integrating Economics and Ecology. Norwegian Journal of Agricultural Sciences, Supplement no 23.
I would like to thank Eirik Romstad, Per Kr. Rørstad and Viil Søyland for comments and support while developing the ideas of this paper. I would also like to thank Leo Jaeok, Michel
Pellaux, Leo Maier and Hans-Christian Beaumond for comments given at the Sanner workshop.
Back to text
2>Transaction costs are defined as the costs of gathering information, making decisions/contracts etc., and controlling/policing. Zero transaction costs thus implies costless information, i.e. full information.
linkurl_blankBack to text
3> Some would argue that this is a far too restrictive evaluation of standard practices where risk and asymmetric information etc. are important issues, all implicitly implying that transaction costs are positive. My point here is that the problems with most of these studies are that as soon as information is costly, it becomes impossible to define an optimum both privately and socially. You will never be able to know what is the right allocation between gathering more information or choosing between the options available given the existing state of knowledge.
linkintHoveddel1t3Back to text
4> Coase is maybe most famous for his proposition that given zero transaction costs, the distribution of property rights does not influence the optimal allocation of resources - the so called 'Coase theorem'. A much more interesting part of his famous paper covers what is a reasonable strategy to follow when transaction costs are positive.
linkintHoveddel1t4Back to text
5 >My analysis has bearings on the dichotomy 'market' vs. 'policy failure'. To go deep into this issue, will take the analysis too far away from the main problem of this paper. I can, however, offer my conclusions. The consequences of my analysis seems to be two fold. First, since policy/collective choice is necessary to establish any institutional structure, 'policy failure' cannot be evaluated at the same level as 'market failure'. They are just not alternatives. Second, in the literature the concept of policy failure seems also to derive its point of reference from the 'market vision', which certainly is an inconsistent way of treatment given the perspective of this author.
linkintHoveddel1t5Back to textKuhn (1970) talks about the 'exemplars' of a science. In economics that may be things like smooth cost and indifference curves, separability, the existence of an equilibrium and a specifiable margin etc.
linkintHoveddel1t6og7
linkintHoveddel1t6og7Back to text7>As the reader may already have understood, my critique of the second best literature is not only based on this observation. It goes deeper, since again the reference point for the analysis is the 'market vision'. Still, there are many important insights in second best theory that are very helpful. One of the basic conclusions (Boadway and Bruce 1984) even supports parts of the argument presented here: If there is a deviation from first best, one cannot be sure that a policy based on first best assumptions will lead towards the attainment of higher welfare.
linkintHoveddel1t6og7Back to text
8> In the 'Agreement on Agriculture', article 20.
linkintHoveddel1t8Back to text
9 >Confer what is said previously on the problem of establishing a social welfare function or making general agreements on which interests or needs to protect at the international level.
linkintHoveddel1t9Back to text
10> I use the definition of jointness in production as offered by Gravelle and Rees (1981). Jointness in production thus means non-separability in inputs. Heady (1952) uses the concept in a similar way as he discusses joint products in agriculture like wool and meat from sheep production. Not all authors limit the concept to this kind of technical interdependency. One example is Shumway et al. (1984) who also define situations where fixed inputs (like land) creates 'interdependencies' between products to be joint. Lynne (1988) argues that this only adds confusion to the use of the concept. I agree with Lynne's argument.
linkintHoveddel1t10Back to textAn alternative concept to use could be that of the trade-off frontier. It would bring the focus to the frontier, while emphasizing that there neccesarily is a trade-off involved would be wrong.
linkintHoveddel1t11Back to textOn may perceive panel 3 to be a combination of panel 1 and 2. This, however, demands both to be looked at as either functional relationships or production possibility frontiers.
linkintHoveddel1t12Back to textSome may comment that only the region to the right of S is of interest. As will be made clear later, this is not necessarily the case.
linkintHoveddel113Back to text