Environmental-economic policy assessment: A farm economic approach

Agricultural Systems 39 (1992) 421-438

h; i Environmental-Economic Policy Assessment: A Farm Economic Approach G. A. A. Wossink, T. J. de Koeijer & J. A. R e n k e m a Wageningen Agricultural University, Department of Farm Management, Hollandseweg 1, 6706 KN Wageningen,The Netherlands (Received 21 August 1991; accepted 30 January 1992)

ABSTRACT Increasing environmental problems in agriculture urge policy-makers to develop instruments to reduce and control the pollution caused by current intensive farming practices. These measures should be both effective from the ecological point of view, which is a public goal, as well as acceptable at the farm level with regard to private goals such as income and continuity of the farm. Therefore, information is required concerning the complex interaction of production intensity, environmental aspects and farm income. This paper concerns extension of the linear programming optimization models employed in farm economics with an environmental component to analyze and evaluate the effects of alternative environmental policy instruments for agriculture. The application presented concerns the potential role of technical innovations and of input levies to reduce biocide use in crop production.

1 INTRODUCTION Usually the primary objectives of the agricultural entrepreneur are continuity of the farm and sufficient income for the farm family. To realize these objectives the farm organization has to be continuously adapted to changing external conditions. In strategic decision-making at farm level, and in modeling this process, environmental pollution is normally not considered, because relevant market incentives are lacking and environmental regulations for agriculture have not been common until recently. Now, it has become evident that agricultural production in its present form 421

422

G. A. A. Wossink, 7". J. de Koe(jer, J. A. Renkema

may incure considerable environmental harm and that governmental intervention is required to counter further ecological damage. The question, however, is which instruments should be chosen for an appropriate environmental policy for agriculture; appropriate in the sense of reducing ecological stress within a reasonable period and without unacceptable income losses for the sector. Assessment of the effects of alternative environmental policy instruments requires insight in the interactions of production practices and intensity, environmental aspects and income. To acquire this insight the economic models used in research are to be extended with parameters for the environmental effects of the production activities considered. The aim of this article is: (1) to indicate the structure and requirements of combined environmental-economic model systems and (2) to present an application at farm level for a levy on chemicals, in order to show the ability of the model to evaluate environmental policy regulations.

2 ENVIRONMENTAL-ECONOMIC MODELING

2.1 Aggregation level and method The first decision in environmental-economic modeling concerns the aggregation level for which the model will be developed. In the present study the individual farm is chosen as the starting point. On this level the actual decisions are made concerning cropping pattern, production intensity, etc. Environmental effects of production, such as nitrogen leaching, can only be assessed in the context of rotation scheme and fertilization practice and the environmental effects are related to specific natural conditions, such as type of soil and groundwater level. For simulating the economic decision-making process at farm level, linear programming and its extensions are frequently employed. These methods present the collection of relevant technical opportunities offered to the farm by separate activities in a process matrix. Programming methods are wellsuited for environmental-economic research because: (a) many activities and restrictions can be considered at the same time, (b) an explicit and efficient optimum seeking-procedure is provided, (c) once formulated, results from changing variables can be calculated easily, and (d) new production techniques can be incorporated easily by means of additional activities in the model. An environmental-economic farm model based on this technique covers--besides the regular items of production such as cropping pattern,

Ent:ironmental-economic policy assessment

423

cultivation operations, labor supply and requirements and investments--an additional component, which incorporates the environmental parameters selected for the cropping activities.

2.2 Environmental criteria Environmental pollution in arable farming is caused by the use of minerals, in particular nitrogen, and chemicals. Depending on the method of application and characteristics of the specific inputs they may have negative environmental aspects, namely: emission into groundwater and surface water, emission into the air and accumulation in the soil. For chemicals, negative aspects for nature, due to their effects on organisms useful for the ecosystem, can be added. Measuring the ecological damage as such is very difficult. Instead, usually, criteria are chosen as indicators of the factual or expected damage. With regard to nitrogen the concentration of nitrate in the groundwater is used as the standard measure. For chemicals the concentration in the groundwater, the emission to the surface water and the toxicity for aquatic organisms are well-known criteria. Governmental intervention by means of environmental policy instruments aims at reducing the level of input used or the level of emission to the environment. Hence, for efficient policy instruments detailed knowledge of input/negative output relationships is needed, both of the current production practices and with regard to new techniques.

2.3 Environmental goals and policy instruments The alternative policy instruments to influence pollutors' behavior can be classified into three categories (Baumol & Oates, 1979): (a) moral persuasion by publicity or social pressure, (b) direct controls, and (c) methods that rely on market processes. At the m o m e n t direct controls are the most widespread form of policy instruments for environmental protection. Direct controls are popular among legislators and regulators, and offer the possibility to realize a predetermined reduction in environmental pollution. Economists, on the other hand, generally are proponents of market instruments, which can be classified into three categories: (a) taxation of environmental damage by means of input levies, (b) subsidies per unit of emission reduction or for emission-control investments, and (c) issue of a limited quantity of transferable pollution licenses. At the m o m e n t direct controlling is the main instrument used in Dutch environmental policy. Examples are the so-called Manure Law and the Admittance Law for biocides. The Manure Law indicates maximum input

G. A. A. 14~ssink, 7-. J. de Koe(jer, J. A. Renkema

424

levels for P2Os specified for types of soil and the crops grown. For N the same approach is in preparation. With regard to the use of chemicals in agriculture it has become clear that the admittance policy does not prevent unacceptable effects. The recent Long Range Plan on Biocide Use (Min LNV, 1990) presents quantitative objectives. The major strategic headline is to reduce the input of agrichemicals (in terms of the weight of active components compared to the average use over 1984-88) by 35% in 1995 and 50% in 2000. Each sector of agricultural production has been given its detailed goals. For arable farming the percentages in Table 1 are the guideline. Policy instruments to achieve the goals of Table 1 are not yet defined, though a levy on chemicals (imposed by weight of active component) was suggested at the political presentation of the Long Range Plan in June 1991. TABLE I Reduction Goals lbr Biocide Use in Arable Farming in the Netherlands

CategopT

Percentage reduction total kg actiL~e cot~lponents a BI' 1995

By 2000

Nematicides Herbicides Insecticides Fungicides Others

46 30 15 15 42

70 45 25 25 68

Total

39

60

"Compared lo the average use over 1984-88. Source: Min LNV (1990).

Further future goals for nitrogen and chemicals relate to the maximum concentration of chemicals and nitrate in groundwater. These restrictions are: maximal 0.5 ~g/liter groundwater for chemical discharge per crop and maximal 50mg NO~/liter groundwater for nitrogen (Min VROM, 1989). Both market instruments and direct control regulations mean a change in the optimal farm organization. Constructing policy regulations to achieve a certain reduction in input use requires detailed information on current production techniques and their alternatives. In particular for price incentives, such as a levy, this is important, as the price elasticity determines the level of the levy to impose.

Environmental-economic polk3" assessment

425

3 MODEL SPECIFICATION AND DATA REQUIREMENTS 3.1 Model structure

The general form of the model is shown in Table 2 and may be interpreted as a standard linear programming problem: Maximize subject to and

{Z = e'x} Ax _< b x> 0

where x = vector of activities; c = vector of gross margins or costs per unit of activity; A = m a t r i x of i n p u t - o u t p u t coefficients (technology-set); and b = vector of constraints. The activities x, out of which the optimal combination is to be chosen by the solution procedure, are shown across the top in Table 2 under five headings: production activities representing different crops and cropping variants per crop, variable operations (choices among using own mechanization or contract work and among methods of control), seasonal labor, 0/1 activities representing new machinery for chemical and mechanical crop care, and activities for the use of the whole range of chemicals. The rows of the matrix indicate the type and form of the constraints included: total land, rotation restrictions, supply of fixed and of seasonal labor, several coupling restrictions and the discharges of chemicals and nitrogen to groundwater. Thus, each unit (hectare) of a production activity x requires inputs represented by its specific vector in the matrix A. Among the inputs the type and quantity of chemicals (in kilograms of active components) related to a production activity is specified. Hence, the gross margin figures c of the production activities do not include the costs of these inputs. These are given by separate activities for every chemical in guilders per kilogram of active component and are linked to the production activities by coupling constraints. In this way the linear programming procedure takes account of listing the total use of chemicals and a levy on these inputs is easily incorporated by an increase in their prices. The model also registers the total input of nitrogen. Here, the costs have not been separated from the gross margins of the production activities, as a levy on N-fertilizer is not considered in the calculations. The leaching figures for chemicals and nitrate into the groundwater are added to every cropping activity as quasi-external data. The assessment of the emission figures is described in Section 3.2. The calculations presented in Section 4 focus on the effects of an input levy on chemical use for arable farming. So, the effects of this instrument on groundwater concentrations of

Objective function

a i.j

+1

+1

Gross margins excl. costs of chemicals

- -

+a~.j

Coupling production activities and chemicals

Discharges of chemicals and nitrogen to groundwater

+1

Coupling production activities and new mach.

d

+1

+ai

+1

+1

Costs per hour

+1

-1

labor l ..... m

activities

] ..... n

Seasonal

Production

Coupling production activities and vat. operations

Seasonal labor in periods of 14 days

Fixed labor in periods of 14 days

Rotation restrictions

Constraints Max. hectares

Activities

Costs per hectare

-- ai. j

-~- a i ,j

-1

-t~ a i, J

1..... p

own mechanization or c o n t r a c t w o r k

m e t h o d s o[' control and

Var. operations:

a

999

b

Annual costs b

-999

a

c

- 999

J o r c r o p c a r e O/ l

N e w machiner.v

TABLE 2 Structure of the Linear Programming Model

Costs per kg a.c.

-1

I ..... q

Chemicals