Agricultural policies in industrial countries and their environmental impacts : applicability to and comparisons with developing nations

THE WORLD BANK POLICY PLANNING AND RESEARCH STAFF Environment Department Agricultural Policies in Industrial Countries and Their Environmental Impacts: Applicability to and Comparisons with Developing Nations Ernst Lutz and Michael Young February 1990 Environment Working Paper No. 25 This paper has been prepared for internal use. The views and interpretations herein are those of the autho(s) and should not be attribted to the World Bank, to its affiliated organizations or to any individual acting on their behalf. Ernst Lutz is a Senior Economist in the Eywironment Department of the World Bank. Michael Young is a Natural Resource Economist with the Commonwealth Scientific and Industrial Organization (CSIRO) in Canberra, Australia. The authors are listed in alphabetical order; senior autheTship is not assigned. The authors are indebted to Jock Anderson, John Cleave, Herman Daly, John English, Vijay Jagannathan, Raymond Noronha, and William Magrath for valuable comments and discussions. The views expressed, however, are those of the authors and should neither be attributed to any other individual, nor to the World Bank. Departmental Working Papers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that are circulated to encourage discussion and comment; citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Because of the informality and to present the results of research with the least possible delay. the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. . ii - For developing countries agricultural output growth between 1965 and 1986 has been an impressive 3.3 percent per annum. There is mounting evidence, however, that some of the growth is developing and industrial countries alike has occurred at the expense of long-term resource degradation. The question which this paper addresses concerns the types of environmental and resource-related agricultural policy adjustments that could properly address the costs of resource degradation and increase long- term prospects for sustainable development. The paper reflects experiences in both industrial aid developing nations. Where possible, lessons and insights of relevance to policy-making in developing countries are sought. Conceptually, agricultural production will be most efficient when the costs borne by producers include external costs. Policies should try to reduce the divergence between the private and social costs of resvurce use. In actuality, there are large differences between conceptually pure policies and a?tual ones. A natural resource accounting framework which recognizes the difference between the income generation and the depreciation of resource stocks and which more accurately measures time, sustainable income would enable policy successes and failures to be judged more reliably. One important lesson that has emerged from industrial countries is that e-gricultural and environmental policies should be "integrated". It may be costly to determine specific linkages in precise quantitative terms, but the qualitative impacts can be thought through, at low cost. Applicability of integrated policy-making to developing countries is therefore possible and is to be encouraged. The paper reviews conceptual possibilities for internalizing externalities and where possible provides specific examples or experiences. It discusses the Polluter-Pays-Principle, taxes/subsidies as well as regulations and the associated enforcement issues. In general terms, a contribution of both the incentives' and the regulatory approach are needed. In developing countries one will have to rely more on incentives than in industrial n&tions, since the capacity to enforce regulations there is more limited. Other possible issues of relevance for environmentally sound agricultural development are also discussed, including income/tax reform, research priorities, land tenure security and its possible effects on long- term investment, possible ways of enhancing environmental quality, and the limits of extension work when environmental and financial objectives diverge. The paper concludes that improvements are possible particularly in the following areas: (a) the development of policies which internalize external costs; (b) the preparation of natural resource accounts and other information which enable decision makers to appreciate the environmental costs of the policy options under consideration; (c) identification of the opportunity costs and impacts on economic growth of different resource pricing policies; (4) improving the enforcement of regulations and reducing corruption; (e) phasing out input subsidies and introducing policies and - iii . introducing policies and property right arrangements which reflect the marginal cost of resource use; and (f) making income taxation policies net tral to the environment. The paper also encourages research, particularly on the physical impacts of projects and policies as well as their economic valuation. AGRICULTURAL POLICIES IN INDUSTRIAL COUNTRIES AND THEIR MVIRONKEMA IMPA.CTS: APPLICABILITY TO AND COMPARISONS VITH DEVELOPING NATIONS TABLE OF CONTENTS za INTRODUCTION ......................................1 CONCEPTTALIATION.. ...................................2 POLICY FORMULATION: A CASE FOR "INTEGRATED" POLICY MAKING ............3 INTERNALIZING EXTERALITIES ........................................... 5 The Polluter-Pays Principle .........................................5 Taxes ...............................................................7 Input Subsidies......................................................8 Other Subsidies .....................................................9 Regulations and Enorcement ........................................10 ROLE OF OTHER MEASURES OF RELEVANCE FOR ENVIRONMENTALLY SOUND AGRICULTURAL DEVELOPENT ............T..... 11 Income/Tax Concessions ............................................. 1 Improving Information ..............................................12 Improving Research .................................................12 Land Tenure/Credit Policies and Long-term Investment ...............13 Extension, Self-interest and Conservation Ethic ....................13 Preventing Declines in and/or Enhancing Environmental Quality ......14 CONCLUSIONS...........................................................14 Table 1: PESTICIDE IMPORTS OF DEVELOPING COUNTRIES, 1967-1986......16 Table 2: ESTIMATED AVERAGE RATE AND VALUE OF PESTICIDE SUBSIDIES... 17 Table 3: HYPOTHESIZED POTENTIAL OF OPPORTUNITIES TO PROMOTE SUSTAINABLE DEVELOPMENT AND ACHIEVE OTHER ENVIRONMENT RELATED GOALS...............................18 REFERENCES.........................................................19 -1- AGICTMBAL POICIES IN 111MT . 0M A THEIR =M&IRONWffALIMPACT= APPLCABXLXTY TO AND C011PARISONS =1T DIMELWPXNG NAT=ON Over the last two decades agricultural output in developing countries has expanded at an impressive rate. For developing countries as a whole, the expansion rate between 1965 and 1986 was 3.3 percent per annum. Even during the recessionary period between 1980 and 1986 the average rate of production expansion was 3.6 percent per annum. During 1980-86 growth in agriculture was fastest in manufactured goods exporting countries (5.0 percent) and worst for the sub-Saharan African countries (1.2 percent).1 The relatively strong performance of developing countries overall has been achieved by an expansion of the cultivated area and the adoption of new varieties along with the use of new techniques and high- energy inputs such as fertilizers and pesticides. Some policy improvements may also have played a role. In industrial market economies agricultural production expanded at an annual rate of 0.9 percent between 1965-80 and a strong 2.5 percent from 1980-86.2 There is mounting evidence that some of the growth in the agricultural sector has occurred at the expense of long-term resource degradation. In the United States, for example, it has been estimated that the off-farm cost of soil erosion is between $3 and 13 billion per annum with the best single value estimate being $6.1 billion in 1980 dollars (Clark and others, 1985). In the United Kingdom 200 million pounds in capital costs and an annual expenditure of 10 million pounds will be required over the next 20 years to bring all drinking water within the European Community's standard for maximum permissible concentrations of nitrate (OCD, 1989). Repetto and others (1989) analyzed depletion of timber, petroleum and top soil in Indonesia over the period 1970-84. While GDP grew during that period at an average annual rate of 7.1%, the growth of income corrected for depletion of the three resources was only 4.0%. Moreover, in four of the fourteen years analyzed, the corrected rate of economic growth was negative. For Mali, Bishop (1989) estimated annual income losses due to soil erosion to amount between 0.6% and 2.4% of 1988 agricultural GDP. When the impact of erosion in one year on yields in the following ten years is considered, the losses range between 4% and 16% of 1988 agricultural GDP. Costs of this sort are not included in national accounts and, hence, in economies where resource depletion/degradation and pollution issues are important, decision makers are not being accurately informed about the efficacy of their policies. A natural resource accounting framework, which recognizes the difference between true income generation and the depreciation of resource stocks (Abmad, E1 Serafy, and Lutz, 1989), and which more accurately measures true, sustainable income, would enable policy successes and failures to be judged more reliably. Research efforts are under way to build such accounts through geographic information systems which might enable that policy impacts be identified by area. J Source: World Development Report, 1988. ,/ The reasons for the expansion and its impact on world market prices and developing countries are well known (Zietz and Valdez, 1986). - 2 - The question which this paper addresses is: Given that growth has been achieved at some (as yet perhaps unaccounted) costs, what types of environmental and resource-related agricultural policy adjustments could properly address these costs and increase long-term prospects for sustainable development? The focus of the paper is on long-term economic development since some short-term policies and approaches to growth may not lead to sustained development. The paper reflects experiences in both industrial and developing countries. Where possible, lessons and insights of relevance to policy making in developing countries are sought. CONCEPTUALIZATIQN It is probably accurate to say that in the past the environmental effects of many policies have often been neglected in agricultural policy making, as elsewhere. Moreover, the cumulative and ultimate effects of a policy change on resource use and development are often overlooked. What is needed now is that environmental concerns receive proper, balanced consideration during the development of pricing and other policies which affect resource use (OECD, 1989). As far as resource pricing policies are concerned, a set of principles for the design of an incentive system is needed. Collectively, and providing the emergent policies are consistent with these principles, the outcome should be "right", not in the narrow sense of short-term economic efficiency but in terms of efficiency that accounts for the resultant structural changes and includes external effects. In addition, the transitional arrangements to this preferred policy framework themselves should be in line with the overall objectives. From a theoretical point of view, agricultural production will be most efficient when the costs borne by producers include any negative external costs, and when inputs are acquired at their social opportunity cost. That is, the policy framework associated with production should seek to correct for market failures and remove any divergence between the private and social costs of resource use. Conceptually, and providing that all other related product and input prices are subject to the same discipline, this would result in resource use and production at a level which is optimal from a (broad) efficiency point of view. Essentially, this would ensure an optimal allocation of resources.3 The differences between conceptually pure policies and those that actually exist are large; and when one looks at existing policies, it is at times difficult to clearly see their objectives and rationale. In industrialized countries producer subsidy equivalents for many products and in many countries exceed 30 percent (OECD, 1988). Taking second-round and indirect effects into account the Centre for International Stndies (1988) reports that the full liberalization of agricultural trade coupled with the removal of internal subsidies would increase real income in each of the World Bank's six developing country categories, produce an aggregate gain / A better allocation of resources is very important but it would of course not, by itself, be sufficient to reach a sustainable path of development as Daly (1989) and Pezzey (1989) have pointed out, since, in order to achieve that, one would have to tackle the more difficult problem of scale of production, population and resource use in relation to limited resources. - 3 - of US$28 billion, reduce the need for foreign aid by US$17 billion and reduce external public debt of developing countries by 2.4% per year. In a similar study Tyers and Anderson (1988) observe that the gains of farmers in developing countries from higher food prices would be nearly four times as large as the loss to consumers in those nations. Distortions are also prevalent in many developing countries (USDA, 1988) but with cases of taxation still more prevalent than subsidization (Lutz and Saadat, 1988), motivated generally by the objective to raise revenues. In Brazil, for example the producer subsidy equivalent for beef is -33% while within the European Community it is +44% (USDA, 1988). In agricultural policy analysis, most approaches -- be they partial or of multi-market type -- have not considered environmental impacts that might be associated with a set of policies on the output or input side. In some cases, these impacts may be negligible. In others, such as fertilizer, pesticide, or livestock subsidies, these may be significant. In our view, environmental effects should not, a priori, be assumed to be negligible and excluded from agricultural policy analysis (or policy reform packages). Rather, we feel, that analysts should, on a case- by-case basis, consider whether environmental effects might be significant, and if so, decide whether inclusion in the analysis would be merited (either qualitative/descriptive or quantitative). It seems that this view is fully in line with that supported by the Development Committee (1987, 1989). Existing policies are usually the outcome of a complex interplay of lobbies, government ministries, the prevailing wisdom of what the right policies are, etc. (Rausser and Foster, 1989), and this process is occurring in an environment of limited information and considerable uncertainty. On top of the existing forces and factors, environmental concerns are now emerging and are, at first glance, "complicating" the picture, but this is so only if one looks at things from a short-term perspective. In agricultural policy making as elsewhere, one is faced with the fact that one is operating in a second-best world. This means that it is possible that a certain specific policy reform, however well intended and justified, may make things worse rather than better when there are large distortions elsewhere in the system. Ideally, one would want to use a general equilibrium model to test out a proposed policy change and the nature of the adjustment process. But few such models exist, and they are costly to build. Should one therefore resort to a "do nothing" attitude? No, rather, in spite of possible drawbacks, we believe that one should try to work toward partial changes in the hope of moving from the current state to an improved second-best world. This includes both avoiding changes in the wrong direction and implementing "safe" changes in the right direction. The discussions below should be seen in this light. POLICY FORMULATION: A CASE FOR "INTEGRATED" POLICY MAMING One important lesson that has emerged from industrial countries is that agricultural and environmental policies should be "integrated". Integrated policy development and implementation requires conscious trade- offs to be made amongst competing objectives. Integration requires that full account is given to environmental objectives during the formulation of agricultural policies and, similarly, that environmental policies must reflect a recognition of their potential impact on agricultural output, incomes and prices. It requires recognition and acceptance of the fact that objectives are rarely fully complementary and are often contradictory (OECD, 1989). It may be somewhat idealistic to think that decision-making by policy makers is fully objective and equitable, and that they want to know all the consequences of their decisions. But whether that is or is not the case, economists undertaking policy analysis should supply policy makers with integrated analyses which include consideration of the likely environmental effects and impacts of policy decisions on overall welfare and distribution of income. Integration can be supported by the creation of institutional arrangements (including administrative procedures), and the formulation and implementation of policies which help to better achieve related objectives. Unintegrated policies are often characterized by the belated recognition of their consequences for the objectives of other sectors. Integration suggests the development of policies that are preventative and anticipatory rather than reactive (OECD, 1989). To the extent possible likely structural changes, investment and changes in land-use practices should be anticipated. But also, given the limitations of correctly anticipating changes, the policy set should be capable of adaptation to changing circumstances. Some Lessons/Experiences from Industrial Countries. Two lessons from industrial countries stand out. First, the use of one program or policy for the siaultaneous achievement of several objectives may result in a sub-optimal outcomes with unintended effects. Ideally, each policy goal should be supported by an independent policy instrument. Within the European Community, agricultural price support, achieved via tariff barriers and market intervention, was used to promote self-sufficiency, maintain farm income and preserve traditional landscapes. Production surpluses, the inequitable transfer of income to rich farmers, extensive agricultural pollution and declines in landscape quality all resulted. With great political pain, the Common Agricultural Policy is now being slowly reformed into one where more direct instruments are used to support specific objectives. If this had been done at the start, many of the present problems would not have arisen, and those which did arise could have been solved with much less difficulty. A fear which remains is that the new, more direct policies may simply be super-imposed over the old inappropriate ones. Second, the development of integrated policies can be supported by thinking through the effects of policy on investment and, through this, on structural change. This is particularly important from an environmental point of view since some problems appear as lagged results to slow structural changes. In the Netherlands, for example, a preferential tariff agreement which favors the use of cassava as an animal feed has facilitated a dramatic expansion in intensive animal production. Some 10 to 15 years after the introduction of the agreement the disposal of animal waste, the reduction of nitrate pollution, and the control of ammonia emissions have become major problems. - 5 - Anlicability of Integrated Poliev Making in LDCs. Integrated policy making may, "at full maturity," require a well-developed institutional set-up. However, something can be done by any policy maker or adviser even at relatively low costs, namely to ask whether a proposed agricultural policy change will have an environmental impact, and if so, whether the impact might be cumulative. It may be costly to determine the specific linkages in precise quantitative terms, but the qualitative impact can bt thought through and may well be obvious. Examples of obvious cases are pesticide subsidies, which have added to pollution and health affects (Repetto, 1985), and the tax schemes in Brazil which have promoted rain forest clearing (Binawanger, 1989). Therefore, the integrated policy making approach, in more or less developed form, can and should be recommended for countries at any level of development. INTERNALIZING ETERNALITIES An important recommendation is that a policy concerned with true growth and sustainable development should try to internalize the external costs arising from production activities. While in this paper we only deal with agricultural policies, it is obvious that internalization of external costs is needed in all sectors. Occasionally, the policies which apply to secondary industry will influence agricultural development. Policies which affect the quantity of residual cadmium in phosphate fertilizers provide one example of such an indirect effect. The Polluter-Paa PrinCIde Internalizing external costs essentially means making the polluter pay for them. This principle is often referred to as the "Polluter-Pays Principle."4 The Principle's emphasis is on allocating the cost of pollution control and prevention measures. If applied throughout all countries it would ensure that no country, through the use of differing pollution control and prevention strategies, could obtain an unfair trading advantage. Its rationale is based on the economic advantages of improving efficiency in the allocation and use of resources by internalizing pollution control and prevention costs. g/ While the idea of making the polluters pay has been around for a long time, the OECD formally adopted a limited version of the principle in 1972. The formal agreement states: "The principle to be used for allocating cost of pollution prevention and control measures to encourage rational use of scarce environmental resources and to avoid distortions in international trade and investment is the so-called "Polluter-Pays Principle". The Principle means that the polluter bears the expense for carrying out the above-mentioned measures decided by public authorities to ensure that the environment is in an acceptable state. In other words, the cost of these measures should be reflected in the cost of goods and services which cause pollution in production and/or consumption. Such measures should not be accompanied by subsidies that would create significant distortions in international trade and investment." This Principle, reaffirmed in 1974, goes on to state that: "there may be exceptions or special arrangements, particularly for transitional periods provided that they do not lead to significant distortion in international trade and investment." (OECD, 1986). - 6 - An important consideration is the relationship between property rights and the Polluter-Pays Principle. Theoretically, if one can assume completely specified, exclusive property rights, which are costlessly transferable and costlessly enforced, voluntary exchange would lead to the efficient allocation and use of resources (Coase, 1960; Randall, 1983). That is, the initial distribution of property rights is irrelevant to the efficient allocation and use of resources. But in the world in which we live, most environment related property rights are not specific nor exclusive. Moreover, the private, social and political information and transaction costs associated with reallocating property rights are so high that the initial distribution of property rights virtually determines the nature of resource use. The Polluter-Pays Principle, as defined by the OECD avoids the need for concern about property rights by taking the position that, in practice, they are determined by law and government policy. If a government introduces a pollution prevention and/or control program, then its actions imply that the polluters affected by the program no longer have the right to pollute and should be required to pay the costs of the control program. The case of agricultural price support arrangements presents a barrier to the successful implementation of the Principle. Price support policies reduce the incentive for farmers to grow products only in areas which do not have pollution problems. Thus, for the full benefits of this Principle to flow to the environment, direct market support policies might have to be adjusted. An exception to the rule of trying to make polluters pay could be made during the transition period when a pollution prevention and control program is introduced. Occasionally, short term transitional assistance may be necessary to achieve more speedy adjustment and acceptance of a policy change. The ultimate aim, howver, is to equate the private and social costs of resource use and development. The rationale is based on the economic advantages of achieving efficiency in the allocation and use of resources by internalizing pollution control and prevention costs. This may require policy change. The OECD report on agriculture and environment (1989) notes that "in several cases, compliance with this agreement will require countries to modify current policies." Those countries which are applying it, report gains associated with input levies to finance pollution prevention through targeted extension programs and transitional subsidies to stimulate investment prior to the introduction of strict regulations. In the Netherlands, for example, taxes on manufactured animal feed, and regulations and taxes on animal manure disposal are causing industries to develop new technology and relocate in areas with greater unused environmental capacity (Young, 1989). Anlicability to-UDCs. The question on whether the Polluter-Pays- Principle, an economically sound policy prescription adopted by OECD countries, should also be recommended for developing countries, is a difficult one. Some might argue that trying to internalize external costs - 7 - is inequitable where the large majority of farmers are poor. But probably a large share of inputs which result in externalities are consumed on the larger, commercially-orienced farms. Also, one needs to keep in mind that, where negative effects occur, some of these are borne by the poor themselves. Pesticide poisonings are a case in point. In addition, one could argue that the Polluter-Pays-Principle supports long-term efficiency and sustainability which should not be sacrificed because of (valid) poverty concerns. Instead, poverty should be addressed directly and in parallel with environmentally sound policies. Therefore, while internalizing external costs may be more difficult in developing than in industrial countries, the Principle is still valid and, when pursued, is likely to lead to more efficient land use practices and, in the long-run, greater economic growth. There is, however, another dimension that should be considered in applying the Polluter-Pays-Principle to developing countries. Since a large majority of individual agents as well as the countries themselves are poor, there may be a role for international mechanisms including grant funding to help achieve environmentally sound production in an economically efficient way. Such funding might be particularly appropriate in the case of activities with long pay-off periods and where positive international externalities are associated with them. Taxes Taxes on inputs can be used to encourage farmers to adopt agricultural practices which are environmentally favorable and make them pay for at least part of the cost of pollution prevention and control. This, for example, has been advocated by Weinschenck (1987). After a systematic discussion of various policy alternatives and their implications Weinschenck concluded that a tax on yield-increasing inputs, particularly nitrogen, "would not only reduce surplus production [in the EC] but would also encourage a less intensive and ecologically more desirable use of the land" (p. 49). An interesting point in this regard is that profit maximizing behavior by itself, even in the absence of taxes, would lead to reduced input levels "under most conditions" (Harper and Zilbermann, 1989, p. 692). Some Lessons from Industrial Countries. Input charges are used in Sweden to contribute to the cost of reducing water pollution from fertilizers and damage to the environment from pesticides. In the Netherlands, a levy on manufactured feed inputs is used to finance research and extension costs aseociated with pollution from intensive animal husbandry. These charges are relatively low, and are only used to finance research, extension, development activities and some clean-up costs. A commitment to the involvement of farmers in the allocation of the money collected, and a guarantee that all revenue collected will be spent on activities of ultimate benefit to the polluting industry, has been important in gaining political acceptance of these charges. Mixed transferable quota/tax schemes have also been advocated in a number of industrial countries but, as yet, none have been implemented. -8- Input Subsidies Fertilizer and pesticide subsidies have not been common in industrialized countries. An exception is Australia, where fertilizer was subsidized until 1986 to encourage adoption and usage. Irrigation water, however, is commonly subsidized, and where the soils are saline, this tends to aggravate salinity problems. Marginal cost pricing for all purchased inputs is now seen as a pre-condition to the efficient use of natural resources in most industrial countries. Few, however, are implementing marginal cost pricing as a matter of course. Progess is being made in a number of them by charging property rights for water and other resources so that private markets emerge for the rights to use scarce resources. The most common example of this is the introduction of transferable water rights. The Case of Pesticide Subsidies in Developing Countries, Pesticide imports of developing countries have increased rapidly up to 1980. They increased about five-fold (in volume terms) between 1967 and 1980 (Table 1). Significant reductions occurred in 1981 and 1982, and the level reached in 1980 was subsequently surpassed only in 1984. What the import figures imply for actual consumption is difficult to gauge, since production figures of pesticides in developing countries as a whole are not available. It is however, likely that production capacity for the chemicals in question has increased and that consumption therefore increased faster than what the import figures indicate. Pesticide usage, while it can make an important contribution to short-term productivity gains, does have negative effects on human health, non-targeted species, soil and water. It also causes increased pest resistance. John Nickel (1989), the CIAT Director, said, for example: "The Rice Program has been working with a number of countries to assess pesticide usage and has found that over-application of insecticides is one of the chief causes of high production costs, destruction of natural enemies, and environmental pollution (p.5)". In many developing countries pesticide subsidies were introduced to encourage the adoption of a technology that was initially unfamiliar to farmers. However, even after pesticide use was widespread, many of these subsidies have been maintained at considerable financial costs to governments. Repetto analyzed pesticide policies in a sample of countries (Table 2) and found that the average rate of subsidy in those countries was 50 percent. Both fiscal and economic reasons suggest a gradual reduction of the rate of subsidy on inputs in situations where output prices are relatively undistorted. But where externalities associated with input use exist, one could even go one step further and, depending on the size of those effects, recommend a partial or complete elimination of subsidies also in cases where agricultural output is taxed. Where output prices are relatively undistorted, consideration of the negative side effects of input use could argue for imposing tariffs or taxes such that farmgate prices would represent real costs including external and long-run costs. The removal of a pesticide subsidy and/or the imposition of a pesticide tax would be expected to reduce pesticide consumption (depending on the price . 9 - elasticity and the degree to which substitutes exist). Users would have an increased incentive for judicious usage and for the adoption of alternative methods such as the Integrated Pest Management (IPM) approach. Depending on the farmers' responses and options, agricultural output may or may not be negatively affected by the policy change. A good example of positive reform is the case of Indonesia, where, because of reduced pesticide subsidies, fiscal savings have amounted to US$150 million annually over the last four years (Brady, 1989). Simultaneously with the subsidy reduction more pest resistant rice varieties became available, and the IPM method was encouraged through the extension network. Other Subsidies In several industrial countries, because of government reluctance to enforce regulations, farmers are being offered subsidies as an inducement to encourage them to adopt non-polluting agricultural practices. One example is the United States Conservation Reserve Program which pays farmers not to crop highly erodible soil for a ten-year period. The many U.S. cost-share programs which subsidize the adoption of best management practices is another. This policy approach is contrary to the Polluter-Pay Principle. Also, paying farmers not to pollute may also make them perceive that they have a right to pollute and/or that they should only attempt to adopt non-polluting activities when they are paid to do so. The permanent acquisition of cropping rights, would overcome the above problems. If at all, incentives should only be offered for those practices which are non-polluting and enhance environmental quality when farmers act in a way that improves the landscape and provides other positive benefits, such as enhanced wildlife habitat. Where there is no clear economic mechanism available to pass the costs of doing this on to those members of society who benefit from the improvements, incentives could be offered. Management agreements are often used to provide such incentives.5 5/ In entering into such management agreements it is desirable, first. that the agreement is made in a manner which guarantees that the desired environmental benefits will be realized. This means, for example, that if the aim is to prevent cropping then the cropping rights should be permanently acquired. Second, the payment should not exceed compensation for the reduction in expected net farm income caused by the adoption of the practices required under the agreement. Third, the agreements should not be structured in a way and for a length of of time which enables farmers to hold the desired benefit to ransom as a means of increasing their compensation (OECD, 1989). Most of the management agreements which exist in Europe, however, only run for a specific period of time such as five years and hence put farmers in a very advantageous position. Either a) annual payments should be limited to reimbursement for the cost of necessary practices such as mowing, or, b) there should be economic penalties for failing to renew an agreement. * 10 - Regulations and Enforcement When fully complied with or enforced, regulations, which require farmers to adopt environmentally favorable practices, internalize the costs of pollution prevention and control and give the correct economic signals to farmers. Theoretical economists often object to regulations on the grounds that they seem to ignore the concept of optimality. However, after consideration is given to administrative costs, transaction costs, information needs, and the frequency with which input and product prices change, regulations can be more efficient than taxes and subsidies (Dasgupta, 1982). Regulations can be looked at as a special case of a tax where no tax is applied up to a certain threshold level. When that level is exceeded, the actual tax is equal to the expected penalty multiplied by the probability of getting caught. If the probability of being caught for violating a regulation is close to zero, then compliance is also likely to be low. This points toward the centrality of proper enforcement if the regulatory approach is to succeed. Some Lessons/Experiences from Industrial Countries. Examples where regulations have been used to internalize costs include the banning of insecticide spraying while crops are in flower in Belgium and the prohibition of the aerial spraying of pesticides in Sweden. Several countries regulate agricultural practices such as manure spreading and cultivation of erodible soils. Land use controls are also used to prevent cropping on steep slopes and along the edge of a water course in a number of countries. Industrial country experience with regulations to restrict undesirable investments and prevent easily identified activities has generally been favorable. Controls on the construction of new animal houses in sensitive water catchments and the prohibition of aerial spraying in Sweden have worked. However, Dutch controls over the amount of manure which can be spread per hectare, and requirements that it be plowed in within 24 hours, are proving difficult and, in many cases, impossible to enforce. The problem is that is virtually impossible to estituate how much each farmer has spread on a site. Several industrial countries are beginning to define the types of agricultural practices necessary to protect the environment, and to require farmers to prepare management plans which indicate how these environmentally desirable practices will be adopted. Examples of the trend towards the use of management plans can be found in the Netherlands where animal manure management plans are required; in Denmark where farmers are being required to prepare fertilizer plans; and in the UnitGa States where, under new conservation-compliance provisions, farmers who wish to continue to receive price support and crop highly erodible soils will be required to prepare soil conservation plans before 1990. In the Netherlands, these plans have the additional advantage of making it relatively simple to levy a tax on farmers who produce surplus animal manure. Provided there is a will to enforce regulations, such plans can also simplify enforcement procedures (OECD, 1989). A pre-condition for this approach is a well developed institutional and physical infrastructure; administrative costs of the regulatory approach can be substantial. - 11 - A related issue of growing importance is the standards set for pesticide registration and food residuals. A number of developed countries are moving to make registration criteria more stringent and reduce maximum permitted residuals in food and water destined for human consumption. As this occurs, countries are introducing new non-price tariff barriers. Examples include the European Community ban on the import of US beef produced with growth hormones and the US ban on Australian beef containing organo chlorine residues (Young, 1990). Countries which set less stringent standards are losing markets to those which match them. Developing countries should follow developments in this area closely in order to avoie possible setbacks in terms of market access. Ahplicability in Develoing Countries. The regulatory approach, while it has been successfully applied in a number of specific instances in industrial countries, is less applicable in developing countries because the institutional capabilities are generally weak, enforcement is difficult, and monitoring expensive. Often the literacy skills of farmers and farm workers are also limited. But the regulatory approach is clearly needed in some areas such as those of pesticides, where highly toxic ones should be (and are often are being) banned and where other restrictions are needed along with instructions for proper handling and application. Generally, the standards which apply in developing countries are much lower than those found in OECD countries. Those developing countries which wish to continue to export products to developed countries may soon be forced to set standards which are equivalent to those found in industrial countries. ROLE OF OTHER MEASURES OF RELEVANCE FOR ENVIRONMENTAILY SOUND AGRICULTURAL DEVELOPMENT Aside from resource pricing policies that should take account of external and long-run effects, there is an array of other policies and factors that can make agricultural production more sustainable. Income/Tax Concessions An example of a well-intended policy with some unanticipated effects is the tax concessions for forestry in the United Kingdom, which led to widespread planting of pine trees that resulted in declines in wildlife habitat, landscape amenity values and agricultural land. Another example is tax concessions for land clearing and development in Australia, which were introduced in 1950. When their adverse effects on the rate of land clearing became apparent in the early 1980's, they were removed. Recently, the OECD (1988) has concluded that income and capital taxation policies should be neutral to environmental and agricultural objectives. Binswanger (1989) has pointed out that income tax and capital market policies in Brazil have had important environmental implications: These policies have accelerated the pace of settlement in the Amazon and led to the deforestation of land in premature, uneconomic and/or environmentally unsustainable ways. Other Latin American countries have subsidized livestock production on large estates through tax incentives and other means. As a result, tropical forests have been cleared for grazing purposes, even though carrying capacity is low and pasture deterioration rapid. The incentives make the activity financially profitable even though it is questionable on social, economic, as well as environmental grounds. - 12 - Imrovn Information The nature of the information about product performance which is supplied with inputs, can be very important. In the case of pesticide use, many companies perceive that they should guarantee 100 percent control of plant pests. Moreover, in a number of OECD countries liability laws force companies to recommend application rates which give control in all circumstances and conditions. Often the result is a breakdown in natural control mechanisms and the need for more pesticides. By changing the advice given to farmers the Danish government has identified a variety of ways to halve application rates, significantly improve the environment and raise farm income (Young, 1990). The development of agricultural policies which account for their environmental impacts does require additional information. This is particularly the case when most of the impacts are off-site, lag behind the practices which cause them, and are cumulative. The gathering of additional information can obviously be costly. On a case-by-case basis, those costs need to be compared to the potential benefits that can be reaped from improved policies. Imnroving Research The research priorities, agenda and attitudes of today can contribute significantly to the production practices of the future. Obvious examples are the breeding of pest-resistant varieties and development of biological pest control measures. In this regard John Nickel (1989) speaking from the CIAT perspective, said: "The availability of a large, well characterized germplasm collection makes possible genetic solutions to many problems rather than relying too heavily on agricultural chemicals. This "seed versus bag" philosophy doeanot ean 6 that we believe the world will be able to feed itself without the use of inputs, or that we think small farmers should not use them. Indeed small farmers need to use inputs, especially fertilizers, to get the most from their small parcels and maintain soil fertility. It does mean6 that we attempt to develop technology that is not overly dependent on purchased inputs; that we try, whenever possible, to find genetic rather than chemical solutions to production constraints. This is particularly applicable with regard to resistance to insects and diseases." Given the uncertainty as to whether enough food can be produced for growing populations, it is understandable that production objectives have had priority over others in agronomic research. There is now increasing recognition that environmental/sustainability concerns should be given a higher weight than has been hitherto the case. This requires a re- balancing of research objectives. Some of it is no doubt going on but more can and should be done. As York (1988) put it: "Still more progress must be made, and all agricultural research should embrace sustainability as a fundamental precept" (p. 1). A/ Underlining in original. - 13 - Land Tenure/Credit Policies and LonAg-term Investment Land rights are crucial to how a cultivator uses and treats the land. It has been hypothesized that the greater the security of tenure, the more a cultivator is motivated to make investments and adopt production techniques that are beneficial to the long-term productivity of the land. Whether this is indeed the case is not clear; the available evidence is inconclusive and/or contradictory. All developed countries offer secure tenure to landholders. Historical evidence from Australia indicates that short-term leases and restrictions on property rights were a cause of land degradation (Young, 1983). In developing countries more work needs to be done in this area, and environmentally relevant questions should be given more attention (erosion levels, investments to maintain the long-run productivity of the soil, etc.). Intuitively the relationship between security and degradation is obvious, but in many cases, much more than the creation of a more secure tenure system is needed. Land rights and credit are interlinked: The higher the degree of security of tenure, the higher the creditworthiness of the farmer and therefore the better the access to credit. Feder and Onchan (1987) have found empirical support for this theoretically plausible argument. In the OECD countries mortgages over land titles are common. To the extent that investments in the land are personally profitable to the cultivator (as compared to social profitability), access to credit may be important. Larger farmers tend to have more easy access to formal credit markets than small ones, whose access may be limited to the informal market where implicit interest rates tend to be high, not only by market imperfections but also by high retailing costs and risk assessments. The higher interest rates for small farmers puts pressure on them to invest in activities with short pay-off periods. Extension. Self-Interest and Conservation Ethic The more conventional approach to controlling agricultural pollution is to rely on extension services to persuade farmers to adopt environmentally favorable practices. This approach, however, is only successful when the proposed change does not decrease net farm income. In Sweden, for example, reducing fertilizer application levels to that which could be expected to maximize net farm income has been found to increase profits by 100-150 Skr per hectare (OECD, 1989). But when the desired change is not in the direct financial interests of the farmer, this advisory approach is usually insufficient (and understandably so) to persuade farmers to change their practices, and pollution usually continues. Actual behavior of a farmer depends on a number of factors including the economic incentives he/she is facing, his/her conservation ethic and income level. The higher a farmer's income level, the more likely it is that environmentally benign actions are taken by those with a positive attitude toward conservation. Lynne, Shonkwiler and Rola (1988) have found empirically that: "The [conservation] intentions will generally . 14 - be highly correlated with actual behavior if the farmer has sufficient income actually to implement the desires and has a situation needing conservation attention" (p. 14). Also, they found that: "Economic incentives will increase effort, but responsiveness will differ with the strength of conservation-related attitudes" (p. 18). Lynne, Shonkwiler, and Rola (1988) do not directly deal with the plight of poor farmers in developing countries. However, observed behavior in is consistent with their results, namely, that in general, poor farmers tend to heavily discount benefits in the longer-term future, since pressing current needs have priority. The conclusion is that poverty needs to be addressed simultaneously with environmental concerns. It is also clear that short-term financial productivity gains are needed for the poor to adopt techniques that are environmentally benign (and that these must be low cost). Recent work by Jagannathan (forthcoming) provides an interesting additional perspective on these issues. He found for the case of Java that economic opportunities, in particular, employment opportunities in a situation where daily or seasonal migration is possible, are equally important for tl poor as physical assets. Preventing Declines in and/or Enhancing Environmental Ouality The policies of many industrial countries reflect a growing interest in landscape quality, wildlife preservation and the retention of species diversity. Few mechanisms exist to internalize these benefits and, hence, they tend to be undersupplied by private landholders. Moreover, once lost, they are not easily restored. Changes can be irreversible. To avoid these losses and enhance the social benefits associated with the development of these assets, some industrialized countries are providing grants for landscape improvement, creating new property right mechanisms, and reimbursing farmers for the costs of activities which enhance environmental quality. Easements to protect heritage values and retain future options are also being introduced. In Australia an attempt is being made to force a change in each farmer's sense of responsibility for the environment by establishing land care groups. Examples of these conflicts are now widespread in parts of Africa where private and government parks are increasing in number. In most of these cases property rights are modified to give managers exclusive control of access to and use of the natural resources. Many OECD countries have avoided this trend, preferring to make access to such areas available to all, but in areas where game poaching is a problem and government expenditure on conservation is constrained by the lack of tax revenue there appears to be little alternative. CONCLUSIONS Environmental factors, as elsewhere, have not been given enough weight in agricultural policy making in the past. At the same time, however, one must acknowledge that much has been done and is being done that is in line with sustainable development. Nevertheless, improvements are possible, particularly in the following areas: - 15 - (a) The development of policies which internalize external costs; (b) The preparation of natural resource accounts and other information which enables decision makers to appreciate the environmental costs of the policy options under consideration; (c) Identification of the opportunity costs and impact on economic growth of different resource and product pricing policies; (d) The phased reduction of input subsidies and their replacement with charges and property right arrangements which reflect the full marginal costs of resource use; (e) Improving the enforcement of regulations and reducing corruption; and (f) Making income taxation policies neutral to agriculture and the environment. Progress in several of these areas will be dependent on research that sheds additional light on the physical impacts of projects and policies on the resource base as well as on the economic valuation of these. Table 3 highlights our impressions of the relationships between each of the above opportunities for improvement and several important social goals. - 16 - PESTICIDE IMPORTS OF DEVELOPING COUNTRIES. 1967-1986 Current Pri,ce Constant XI Vsle Index WHl.O (US$ M) (1977-100) (US$ M) 1967 210 64 328 1970 295 62 476 1972 320 66 485 1973 467 67 697 1974 659 76 867 1975 876 102 874 1976 858 111 773 1977 979 100 979 1978 1,178 94 1,253 1979 1,225 96 1,276 1980 1,531 102 1,501 1981 1,540 111 1,387 1982 1,485 119 1,248 1983 1,642 125 1,314 1984 1,930 128 1,508 1985 1,804 128 1,409 1986 1,891 127 1,489 Source: FAO Trade and Production Yearbooks, Various Issues. a/ Since a price index for LDC pesticide imports is not available, and since a U.S. dollar based index is needed, the prices for pesticides paid by U.S. farmers as shown in the FAO Production Yearbook was chosen as the deflator. * 17 - Table 2 ESTIMATED AVERAGE RATE AND VALUE OF PESTICIDE SUBSIDIES Subsid as Percentae PAstIcide Rate_of_Full Value Per Countr Retail_Costa Value Capita (US$ Millions) Senegal 11 4 0.7 Egypt 17 207 4.7 Ghana 33 20 1.7 Honduras 71 12 3.0 Colombia 56 69 2.5 Ecuador 59 14 1.7 Indonesia 18 128 0.8 Pakistan neg1. neg1. China 81 285 0.3 Source: Repetto (1985). .le• 點回瞋」 粤取粅峰邸瞧J鯽嗓鷹矓J肥馴軍馴 卹鱸紅單盛劇匯口曉駟躍L讓L煙翅 ---－一－--－一個劍口圈U蝕他－”一－--------－一• 黝劉蠶騙 細自臼細細個血劉煙江細略麒由由縫 取縫取個叢點胡U細取四細鰓 劉為U搜勵•鵝•h么觔症盔盤勵自瞌釀盒直眸細 口■圈勵口口口口口■口■口口口口口開口口口口闕口口口口口 內際ly，曉加加計h群 h細．磚加＋引陸什中令令 ，劇叩劇開歸叫鬨選痲 頂口細O鰓居唱個滷口個O個鑪．＋◆＋唔唔令 點畸鬨•，糒么。•紹啊悶坤此＋，鸞寺中劊陣引陸 讓開O潤，二二。e 頂壯•t常幼tlo•口D 00夸令 ,h“一。鰓七狗眸比 •鰓h．鏟1..+，實啼哺寺◆令中 其。念劉必闖•勵攤鰓騙1 Co.t,r細么閱潤辭。令寺令寺寺劊時 C言．•t•X用抑化馴匕d開七．•．挪1．引時◆啼今引時 勵db：。e勸中處鷗亂。•• •d勵必闖．么鰓〝中tlo斗引時令令◆唔令令 袖由個鸞離寫發01101.．騙柚開言常誠 to細r亂．纏觔總•‘勵〞亂寫劇開馴閱比＋＋今寺會＋＋令令十 憫口■••■••■■••■口■■口口口口■■•自■■■■口■■■ D且劇l.ot．細押鹹勾妒。。雲h•d鯧黝•切，Uohl。層．．也“•七 縫黠••七．th．細黝•r．•緘加此•緘h磚細習幹亂。•二 才劇隊幼記．。誠總d細必么劊細g••藝面實幼6幼．tr•。•ltl劇蟹織 p．常細d編洶誌比•”1•，••七戶1加細．•寫．磁團嗎州。 ◆寺申，。t綱鴉壇邊遞紛i•r寫•C‘細。 ++‘幼叫賺加開噥辭鑲． +,&，幼細“鑲． 19 - RRFERINMM Ahmad, Y., S. El Serafy, and E. Lutz (editors): "Environmental Accounting for Sustainable Development", World Bank, 1909. Binswanger, H. P., "Brazilian Policies that Encourage Deforestation in the Ama on," Environment Department Working Paper, No. 16, World Bank, April 1989. Bishop, J., and J. Allen, "The On-site Costs of Soil Erosion in Mali: A Natural Resource Accounting Approach," Environment Working Paper, World Bank, forthcoming. Brady, N., Remarks at the Symposium: "Making Tropical Agriculture Sustainable: The CIAT Experience", World Resources Institute, April 12, 1989. Cabs, R., and S. Johnson, ONatural Resource Accounting Systems and Enviromental Policy Modeling". Paper presented to the Annual Meetings of the Western Economic Association, Lake Taboo, Nevada, June 22, 1989. Contra for International Economics (1988). Macro-egg = ic Consequences of Farm &Mort ligies, Contra for International Economics, Canberra. Clark, E.H., Haverkamp, J.A. Chapman Eroding goilse The Off-Farm I=acts. The Conservation Foundation, Washington, DC, 1985. Coase, R. "The Problem of Social Cost". Journal of LU and Economics, Vol. 3(l), 1960, pp.1-44. Daly, H., "Steady-State Versus Growth Economics: Issues for the Next Century," World Bank, processed, 1989. Dasgupta, P.: "Ibe Control of Resources.", Basil Blackwell, Oxford, 1982. Development Committee, "Environment, Growth, and Development", Paper No. 14, The World Bank, April 1987. Development Committee, "World Bank Support for the Environment: A Progress Report", Paper No. 22, The World Bank, September 1989. Environmental Liaison Contra, "Monitoring and Reporting the Implementation of the Internatioual Code of Conduct on the Use and Distribution of Pesticides (the FAO Code)," Final Report, October 1987. FAO Trade and Production Yearbooks, various issues. Fader, G. and T. Onchan, -Land Ownership and Farm Investments in Thailand, American Journal of A"icultural Remmics. May 1987, pp. 311-320. Jagannathan, V., "Java: Poverty and Environment Linkages*, Environment Working Paper, World Bank, forthcoming. - 20 * Harper, C.R., and D. Zilbermann, "Pest Externalities from Agricultural Inputs", American Journal of Agricultural Economics, Vol. 71, No. 3, August 1989. Lutz, E. and Y. Saadat, "Issues Relating to Agricultural Pricing Policies and Their Analysis in Developing Countries," Agricultual Economics, Vol. 2, 1988, pp. 19-37. Lynne, G.D., J.S. Shonkwiler, and L.R. Rola: "Attitudes and Farmer Conservation Behavior", American Journal of Agricultural Economics. Vol. 70, No. 1, February 1988. Nickel, J., "Making Tropical Agriculture Sustainable: The CIAT Experience", World Resources Institute, April 12, 1989. OECD, "Agricultural and Environmental Policies: Opportunities for Integration", OECD, Paris, 1989. OECD, -OECD and the Environment", OECD, Paris, 1986. Pezzey, J., "Economic Analysis of Sustainable Growth and Sustainable Development," Environment Department Working Paper No. 15, World Bank, March 1989. Randall, A. "The Problem of Market Failure", Natural Resources Journal, Vol 23(1), 1983, 131-148. Rausser, G. and W. Foster, "Political Preference Functions and the Market for Policy Reform," Paper presented at World Bank Seminar, Processed, March 1989. Repetto, R., "Paying the Price: Pesticide Subsidies in Developing Countries," World Resources Institute, Washington, D.C., 1985. , W. Magrath, M. Wells, C. Beer and F. Rossini, "Wasting Assets: Natural Resources in the National Accounts," World Resources Institute, 1989. Tyers R., and K. Anderson: "Liberalizing OECD Agricultural Policies in the Uruguay Round: Effects on Trade and Welfare", Journal of Agricultural Economics Vol 39, 1988, pp.197-216. USDA (1988) Estimates of Producer and Consumer Subsidy Equivalents: Government Intervention in Agriculture, 1982-86. ERS Staff Report No. AGES 88 0127. Weinschenk, G., "The Economic or Ecological Way? Basic Alternatives for EC's Agricultural Policy," European Review of Agricultural Economics, Vol. 14, 1987, pp. 49-60. World Bank, World Development Report, 1988. York, E.T., "Improving Sustainability with Agricultural Research," Environment, Vol. 30, No. 9, November 1988. - 21 - Young, M.D. "Relationships between Land Tenure and Arid Zone Land Use". In What Future for Australia's Arid Lands, Australian Conservation Foundation, Melbourne, 1983. Young, M.D., "The Effects of Supply Control Policies,* CAP .Bietin, 1988. Young, M.D. AgXiculture and tha Enyironment: OECD Polie Exnerience and American opportunities, Forthcoming Economic Research Service Staff Report, Washington, DC 1989. Young, M.D. "International Trading Arrangements, the Intensity of Resource Use and Environmental Quality." In Water Ouality and Agriculture, J. Bradon ad., Lynne Rienner, Boulder, 1990 (forthcoming). Zietz, I. and A. Valdez, "Costs of Protectionism to Developing Countries: An Analysis for Selected Agricultural Products," World Bank Staff Working Paper, No. 769, 1986.