The economics of land degradation in the Third World

THE WORLD BANK - 0 POLICY PLANNING AND RESEARCH STAFF Environment Department The Economics of Land Degradation in the Third World Douglas Southgate May 1988 Environment Department Working Paper No. 2 This paper has been prepared for internal use. The views and interpretations herein are those of the authors and should not be attributed to the World Bank, to its affiliated organizations or to any individual acting on their behalf. This paper has been prepared by Douglas Southgate, a consultant to the Environment Department of the World Bank. He is a resource economist and an associate professor in the Department of Agricultural Economics and Rural Sociology at Ohio State University. The author gratefully acknowledges comments on an earlier draft by Jerry Warford, David Pearce, Richard Ackermann, Dennis Anderson, and Jose Olivares. Departmental Working Papers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that is circulated to encourage discusson 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 author and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Dire tors or the countries they represent. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city or area, or of the authorities thereof, or concerning the delimitation of its boundaries or its national affiliation. 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 errog. This paper contains an analysis of the causes of land degradation in the third world as well as a description of some of its consequences. Because developing country fragile lands are populated mainly by small farmers, this paper's analysis of the causes of land degradation focuses on that group's interaction with the natural environment. Certain agricultural development strategies (e.g., excessive substitution of capital for labor) result in the colonization of environmentally fragile hinterlands by displaced peasants. Also, small farmers clear too much land and do too little to arrest erosion from existing farmland because of distorted prico signals and tenurial conditions. Also discussed in this paper are some of the downstream costs of land degradation in the third world. Nutrients attached to eroded soils impair water quality while sedimentation impedes hydroelectricity production, the development of irrigated agriculture, water transportation, and efforts to control flooding. Summarized at the end of the paper are insights into the design of projects and the formulation of policy to arrest soil erosion and related environmental disturbance gained from this paper's analysis. THE ECONOMICS OF LAND DEGRADATION IN THE THIRD WORLD TABLE OF CONTENTS Ra82 1. INTRODUCTION 1 II. SMALL FARMERS AND LAND DEGRADATION 2 III. AGRICULTURAL DEVELOPMENT AND SMALL FARMERS' COIDNIZATION OF FRAGILE ENVIRONMENTS 3 IV. LAND TENURE AND SMALL FARMERS' DECISIONS REGARDING IAND USE AND SOIL MANAGEMENT 5 V. ECONOMIC INCENTIVES AND SMALL FARMERS' DECISIONS REGARDING SOIL CONSERVATION AND LAND CLEARING 8 vI. SOIL EROSION'S DOWNSTREAM COSTS 11 VII. ARRESTING IAND DEGRADATION IN THE THIRD WORLD 13 REFERENCES 15 - 1 - THE ECONOMICS OF LAND DEGRADATION IN THE THIRD WORLD I. INTRODUCTION 1. Considerable alarm has been expressed about the consequences of soil erosion and related environmental disturbance in the third world. Eckholm (1976), USAID (1979), and Ward (1979) view la..d degradation as a threat to agricultural productivity and as a cause of rural poverty. Furthermore, the consequences of soil erosion are not suffered only by those who inhabit fragile lands. Sfeir-Younis (1985) documents that many of the benefits of water resource development in the third world are jeopardized because of poor watershed management. Also, natural resource depletion in the countryside accalerates rural-to-urban migration, which strains the capacity of developing country cities to furnish services and employment. 2. In spite of these concerns, the economics of land degradation in less developed countries (LDCs) have not been closely investigated. In affluent countries, study of the farm-level economics of erosion control often involves the use of sophisticated dynamic optimization models (Burt; McConnell). By contrast, microeconomic analysis of LDC farmers' choices among resource development options is at an incipient stage. Similarly, never in an LDC has anyone attempted the type of national-level assessment of land degradation's impacts completed in the United States (Clark at:al.; Crosson and Stout; USDA/ASCS). 3. In many LDCs, data limitations preclude analyzing farmers' land use and soil management decisions with even the simplest empirical models. For the same reason, conducting a thorough study of soil erosion's downstream costs is difficult in much of the third world. However, since LDC land degradation problems can be oxpected to remain severe for some time to come, the demand for economic analysis of those problems will remain strong. Furthermore, as experiments, field surveys, and othei data- gathering efforts yield fruit, it will be possible to comment with increasing precision on the trade-offs involved in controlling erosion in LDCs. 4. It is important that future studies of the economic3 of land degradation in the third world consist of more than the application of conceptual frameworks developed in affluent countries to new geographical settings. In affluent countries, microeconomic analysis of erosion control concerns itself with choices among soil management strategies made by those who own their farms and lack the option to bring "idle" land into production. By contrast, LDC agriculturalists, particularly those trying to survive where soils are highly erodible, often lack formal title in the fields where they plant crops or graze livestock and frequently can either use or claim unoccupied land. Microeconomic analysis of land degradation in LDCs must take into account these facts. 5. This paper contains a discussion of land degradation problems in LDCs. Because third wodd fragile lands are populated mainly by small - 2 - farmers, emphasis is placed on explaining that group's interaction with the natural environment. Their colonization of fragile hinterlands and their relative disinclination to adopt erosion control measures are shown to be a consequence of agricultural development strategies, the structure of land tenure, and price signals. To clarify differences between social and private interests in soil conservation, the downstream costs of soil erosion are also described. This paper's analysis of small farmers' interaction with the natural environment and its description of the downstream costs of that interaction yields implications for the design of conservation projects and for the formulation of policy, which are summarized at the end of the paper. II. SMALL FARMERS AND LAND DEGRADATION 6. Small farmers, a group defined in this paper to include pastoralists with small herds, are often held responsible for soil erosion problems in LDCs. For two reasons, such a judgment is not universally valid. First, as Hamilton and King (1983) stress, erosion from land not used for crop or livestock production is severe in many watersheds. Second, land degradation also occurs on large farms. 7. However, small farmers make a large direct contribution to land degradation in many watersheds. Furthermore, their indirect contribution is often significant. For example, practicing erosive agriculture in one place accelerates the rates at which soil and gravel are scoured from streambanks and roadsides downhill. Directly and indirectly, then, small farmers are primary agents of land degradation in developing countries. 8. One is reluctant to explain the rural poor's depletive use and management of land by claiming that, because of ignorance or the force of tradition, members of that group fail to perceive or to pursue their own best interests. There is, after all, no basis for arguing that small farmers behave any more or any less like rational maximizers than other groups. Instead, their tendency not to adopt conservation measures, for example, is explained by tenurial conditions and price signals that cause the short term economic sacrifices individual farmers associate with erosion control to exceed the present value of the future benefits to them of soil conservation. 9. Solution of many of the third world's worst land degradation problems necessarily involves action by small farmers. In order that projects, policies, and other conservation initiatives elicit that action, however, the reasons why that group settles on fragile lands and tends not to use conservation measures must be understood. Reflecting the focus of past microeconomic analysis of soil conservation, most of this paper's discussion deals with the tenure regime and economic incentives influencing small farmers' choices regarding land use and erosion control measures. Before turning to that topic, however, the forces inducing the rural poor to colonize humid lowlands, steep hillsides, and other fragile environments in the third world are discussed. Since small farmers' location where soils are erodible is often a more fundamental cause of land degradation - 3 - than their selecting inappropriate conservation measures, the past failure of natural resource economists to analyze colonization more carefully is a major oversight. III. AGRICULTURAL DEVELOPMENT AND SMALL FARKERS' COLONIZATION OF FRAGILE ENVIRONMENTS 10. Observing that populations are growing in most LDCs and that environmental problems are present in the third world, one might infer that the former phenomenon causes the latter. A neo-Malthusian perspective is sometimes justified since some of the world's most eroded landscapes are encountered in LDCs where the natural environment cannot sustain all of the existing population. Explaining all natural resource degradation in the third world as a consequence of population growth, however, is hardly reassuring. Half or more of the typical LDC's population is under twenty years old, which implies that, even if fertility rates decline, the number of births in poor countries will continue to rise. The prospects for future population growth are strongest in Africa because fertility rates have not fallen there, as they have in much of Asia and Latin America. If the neo- Malthusian perspective is correct, then continued environmental degradation is inevitable in the third world, particularly in Africa. 11. Natural resource conservation might ultimately depend on eliminating the positive difference between fertility and mortality rates. However, human pressure on third world fragile environments is much more complex than a simple neo-Malthusian interpretation suggests it to be. Rather than being caused purely and simply by population growth, natural resource degradation often occurs as peasants, displaced by economic change in the countryside that is in turn precipitated by various causes, migrate to environmentally fragile hinterlands. 12. Economic change in the countryside doei not always displace peasants. For example, Hayami (1981) cites studies showing that small and medium-sized Asian farmers have adopted high-yielding varieties of rice more rapidly than have large Asian farmers (IRRI) and that use of improved wheat varieties in the Indian Punjab constitutes a neutral technological change with respect both to scale and to factor use (Sidhu). If anything, Hayami argues, the "green revolution" has increased demand for agricultural labor in Asia, partially counterbalancing other forces that limit economic opportunity for the rural poor. 13. Often in history, though, large numbers of peasants have been displaced as modes of production have changed, as large landowners have responded to new opportunities in agricultural markets, and as pre- capitalist rural orders have broken down (Hayami and Ruttan). For example, so that England's growing textile industry could be supplied with wool, much of the English countryside was converted into enclosed pasture during the seventeenth and eighteenth centuries. In the process, tens of thousands of peasants, whose families had long held customary rights to use manor land, were evicted (Allen). -4- 14. The social costs of agricultural change have often been contained because decreasing agricultural demand for labor was offset by improving employment opportunities elsewhere in the economy. Indeed, Binswanger (1984) contends that the latter trend usually helps cause the former since substitution of capital for labor generally occurs when demand for labor outside the agricultural sector is strong. For instance, the decline in the American workforce employed on farms has coincided this century first with an increase in industrial demand for labor and more recently with expansion of the services sector. Similarly, Commander and Peek (1986) document that, during the 1970s, most of the peasants displaced by the simultaneous dismantling of Ecuador's semi-feudal rural order and the continuing modernization of its agriculture either found non-farm employment in rural areas or found jobs in cities. 15. However, not all displaced peasants find non-agricultural employment. Consequently, economic change in the countryside typically results in some colonization of hinterlands by small farmers. In a sense, European settlement of the Western Hemisphere is an example on a grand scale of this response. Most of Europe's arable land was occupied long before Columbus sailed west from Spain. Consequently, a portion of that continent's displaced peasantry emigrated to agricultural frontiers in the Americas between the 1500s and early 1900s. In terms of the number of people involved, however, current colonization of humid lowlands, steep hillsides, and other environmentally fragile hinterlands in Africa, Asia, and Latin America is a more important migration. 16. The scale of movement to third world agricultural frontiers has been influenced by distorted price signals. In the typical LDC, small farmers bear most of the costs of policy-induced distortions in markets for credit, agricultural inputs, and crops and livestock (see the fifth section of this paper). Consequently, their ability to pay for prime agricultural land is limited. Economic competition among different classes of farmers for scarce natural resources is also influenced by the tenure regime typically encountered along a third world agricultural frontier. In tropical America, for example, agricultural colonists are given property rights in the land they deforest (Goodland; MacDonald; Nations and Komer). Given this option, many small farmers forgo attempting to bid for more productive or better situated land. 17. Blaikie (1985) contends that agricultural colonization of hinterlands is also observed because large farmers influence the market and tenurial environment. He point out that there are signals throughout the agricultural "establishment" in most LDCs (broadly defined to include research centers, donor agencies, and governmental bureaucracies in addition to agri-businesses and large farms) to develop large-scale farming systems that are not labor-intensive (Posner and McPherson). As large farmers adopt such systems, more peasants are displaced. In addition, Blaikie argues that concentrated economic interests promote the rural poor's settlement on small parcels in order to reduce wages paid to unskilled workers in agriculture and other sectors. Inasmuch as the family of a laborer who circulates periodically to large farms and urban areas in - 5 - search of employment can help support itself on a small farm, that laborer will be willing to work at a lower wage. De Janvry (1981) argues that recognition of this consequence of creating a class of small landowners in the countryside helped motivate land reform in the vicinity of Puebla, Mexico. Taussig (1978) speculates that land reform in Colombia's Cauca Valley was similarly motivated. 18. The results of an agricultural development strategy that involves migration by large numbers of peasants to erodible hinterlands are usually inefficient from a social perspective. Even if coercive eviction of peasants does not occur, the outcome of bidding between small and large farmers for scarce natural resources will not be optimal because no one competing for land considers the off-farm consequences of soil erosion. Obviously, large farmers bidding for high quality land never take into account the social and environmental costs of pushing the rural poor up steep slopes or into humid lowlands when calculating their maximum bid. Neither do the prices small farmers offer for the same land reflect, for example, the downstream impacts of soil erosion from marginal hinterlands, where they must move if they are unable to compete for prime agricultural land. 19. To summarize, analysis of the rural poor's colonization of fragile environments leads away from a simple neo-Malthusian position. Small farmers do not settle on steep hillsides and humid lowlands just because their numbers are increasing too quickly. Instead, understanding of the greatest frontier settlement in all of human history requires an evaluation of agricultural development strategies. IV. LAND TENURE AND SMALL FARMERS' DECISIONS REGARDING LAND USE AND SOIL MANAGEMENT 20. Colonization by small farmers of erodible hinterlands will not end soon. Implementing agricultural development strategies that create less social dislocation and hence less pressure on fragile environments would take time, even if the necessary commitment from governments and private parties could be easily arranged. In addition, completely eliminating colonization would not be optimal since, with proper management, erodible lands can be farmed on a sustainable basis. 21. Fragile lands in the third world are often not farmed in a way that conserves soil, however. In part, farmers' reluctance to adopt conservation measures is a consequence of distorted price signals (see next section). In addition, their suboptimal use and management of natural resources is explained largely by the tenure regime they face. They rarely possess "nonattenuated" (or, unlimited) private property rights (Cheung) in the resources they use. Also, they live quite often where traditional systems of common property tenure are breaking down or where open access resources (i.e., resources that anyone can use) are available. Explained in this section are the consequences for land use and soil management of these tenurial conditions. - 6 - 22. Attenuated Private Proerty. Because systems for enforcing nonattenuated private tenure are costly to run, attenuation of private property rights is common in the third world. As a result, efficiencies in the use and management of land are lost. 23. Consider, for example, how a temporal limit on a farmer's tenure affects his decisions regarding erosion control. More often than not, the benefits of soil conservation, which comprise enhanced land productivity in the future, can only be gained through some sacrifice in the near term. That sacrifice can take the form of expenditures on erosion control structures, reduced income because less land is planted to a profitable yet erosive crop (e.g., maize), or both. Clearly, if a farmer does not own a parcel in perpetuity but rather possesses only a leasehold, he will be unwilling to incur these short term costs for the sake of benefits realized after the terminal date of that leasehold. The same holds for farmers whose legal claim to land is precarious. Recognizing the risk of future dispossession, they tend to disregard the benefits of conservation realized only after the passage of several years. 24. Suboptimal management of soil can also be a consequence of non- temporal attenuation of property rights. In many tenure regimes, individuals hold only use rights in land. That is, they are entitled to exclude other potential users from a parcel so long as they employ that parcel in a particular way. For example, agricultural use rights are the norm in Latin American hinterlands. As indicated in the preceding subsection, farmers who colonize Latin American agricultural frontiers are often given property rights in land they deforest (I.e., formal tenure is a reward for converting forests into pasture or cropland). In addition, because application for formal tenure is invariably a time-consuming process for a poor person in Latin America, systems of informal property rights, in which agricultural use rights are a basic feature, tend to predominate along the region's agricultural frontiers (Southgate at Al.). Where this is the case, any colonist will place little weight on the benefits of a conserving land use (e.g., forest management or even delayed deforestation) if adopting it jeopardizes his tenure. 25. Common Property and Its Breakdown. At least since the publication of Gordon's (1954) seminal article explaining why open access resources are chronically over-utilized, some natural resource economists have tended to confuse those resources with common properties. Ciriacy-Wantrup and Bishop (1975) emphasize that, far from being available to anyone, common properties can be used only by members of a well-defined group (e.g., those living in a particular village or those belonging to a particular clan or tribe). Typically, members of the privileged group must observe certain resource use rules. By contrast, open access resources are not properties at all. As the label suggests, they can be used by anyone. 26. Common property has been used by various cultures throughout history, including the present, to manage natural resources on a sustainable basis (Ciriacy-Wantrup and Bishop). However, there is always the risk that a common property will be over-utilized because any single - 7 - user will sense that the costs of resource depletion are shared by all users while the benefits are internalized (Demsetz). For example, one herder will calculate that the costs of declining pasture quality associated with his livestock over-grazing a common field are shared by all herders with access to that field. By contrast, he will capture all the benefits of over-grazing, which comprise weight-gain, milk production, etc. If all herders act on the basis of this kind of calculation, then the common field will be degraded. 27. However, individuals with access to common properties do not consistently opt for activities that deplete natural resources. This is because each of them knows that their actions at one date will influence the future behavior of others who use the common property. To continue with the example of the common pasture, any individual herder knows that his decision to allow his livestock to over-graze at one time will probably enhance the inclination of others to let their animals over-graze subsequently. Axelrod (1981) has studied the behavior of those who face such circumstances, finding that tacit cooperation tends to emerge among participants in a "repeated zero-sum game." The implication of his research for the management of common properties can be summarized as follows: to the extent that his resource use decisions constitute a signal influencing others' resource use decisions in the future, each individual will tolerate some near term sacrifice for the sake of conservation that benefits the group, to which he belongs. 28. While they may serve a stable community as a viable framework for resource management, common property regimes can break down because of a number of intrusions from the outside world. Demsetz (1967) documents how Amerindians in North America divided what had formerly been common hunting grounds into private reserves soon after the arrival of European fur traders. Also, the outside world's refusal to recognize a communal tenure regime usually causes that regime's speedy demise. MacDonald (1981) documents that Amerindians in eastern Ecuador quickly converted communally owned fallow lands into privately owned cattle pasture during the 1970s, when agents of the national government informed them that the former were liable to be occupied and claimed by agricultural colonists. This land use change undoubtedly accelerated soil erosion. However, these examples, rather than demonstrating that common property is a defective arrangement per se, indicate that it cannot withstand certain outside pressures. In the absence of those pressures, common property lands can be managed well. 29. Open Access Resources. MacDonald's study of recent events in eastern Ecuador indicates that the demise of common property does not always result in improved resource management. By contrast, open access is an arrangement that invariably results in resource depletion. As Gordon (1954) first showed, a competitive group will chronically over-utilize an open access resource because any economic returns to resource conservation will be perceived by the group as "excess profits," in the sense defined by Marshall (1961). Consistent with Marshallian analysis of long run competitive equilibrium, that group can always be expected to respond to those returns by exerting more pressure on the resource. Thus, any individual who considers practicing conservation knows that any gains will - 8 - be dissipated by increased depletive pressure exerted by other resource users. Similarly, a government's attempt to improve the resource without altering its open access status will not yield any lasting return. 30. Just as common property and its breakdown are pervasive phenomena in LDCs, open access resources are found frequently in the third world. No serious effort is made in the typical LDC to exclude anyone from mangrove swamps and other coastal habitats, even when they are, nominally at least, state properties. Consequently, when a valuable fish species (e.g., shrimp larvae) is found in them, over-fishing is inevitable. Similarly, much of the most severely eroded land in the third world is, in effect, open access pasture, even though it too, like coastal habitats, is often demarcated as government property. The failure of government to resolve the discrepancy between its nominal status as a property owner (either by making a more serious effort to manage its lands or by ceding open access lands to the private sector) is at the root of many LDC environmental problems, including many of the third world's most serious land degradation problems. 31. A Mix of Private Properties and Open Access Resources. The economic literature on "externality" problems contains an extensive discussion of how the open access status of some resources not only influences how those resources are managed (Gordon) but also the management of other resources (Randall). For example, if agricultural land is privately owned but waterways (or, more specifically, waterways' capacity to absorb nonpoint source pollutants washed from farmland) are not, then soil erosion will be excessive. Also, allocating one's labor to the claiming of "idle" open access land, which is always an option near igriculture's extensive margin, increases the opportunity costs of labor used in the management of privately owned land. As a result, the application of conservation measures to existing farmland is discouraged (Southgate and Pearce). Thus, converting open access resources into properties is important for promoting not only their conservation but also the conservation of resources already possessed by individuals or groups. V. ECONOMIC INCENTIVES AND SMALL FARMERS' DECISIONS REGARDING SOIL CONSERVATION AND LAND CLEARING 32. The use and management of a parcel of land have much to do with its legal status (i.e., with whether or not it is, for example, a private property, owned by a group, or an open access resource). In addition, how small farmers alter land use and soil management in response to a change in economic incentives depends on the tenure regime. 33. In this section, five price signals facing LDC small farmers are evaluated: the opportunity cost of labor, interest rates, commodity prices, exchange rates, and prices for non-labor inputs to agriculture. Differences between economic incentives facing that group and incentives facing other farmers are highlighted. In order to emphasize that the influence of price signals on natural resource development is conditioned by tenurial conditions, the impacts of change in or distortion of different incentives both on the application of conservation measures to privately - 9 - owned farmland and the conversion of "idle" land for crop and livestock production are described. A more detailed treatment of the topics discussed in this section is found in Southgate and Pearce (1987). 34. The Onortunity Cost of Labor. Since labor is the principal factor endowment of LDC small farmers, its opportunity cost is an important determinant of resource development decisions made by that group. Because of widespread underemployment, it is sometimes assumed that unskilled rural labor in the third world carries no opportunity cost. Without denying that this assumption is valid in some contexts, one can nevertheless identify two reasons why labor inputs to erosion control and land clearing can be scarce for a small farmer. First, labor's scarcity value is not constant throughout the year. During planting and harvesting seasons, for example, time and effort carry an appreciable opportunity cost. Second, few small farmers are constrained to choose just between working on their own farms and leisure. That they and their families frequently work off the farm, often living away from the household for weeks or months at a time, influences on-farm labor allocation. 35. A change in market forces or government policies (e.g., an increase in r.on-agricultural demand for labor or enforcement of a minimum wage law) that results in an increase or decrease in labor's opportunity cost affects small farmers' decisions regarding the use and management of land. For example, an improvement in off-farm employment opportunities will, in general, result in less effort being devoted both the to erosion control and land clearing. However, wage increases also stimulate the adoption of conservation measures (e.g., reduced tillage) that reduce the labor-intensity of farming. 36. Interest Rates. Whenever there is interference with a price mechanism, market disequilibrium results. The costs of ensuing shortages or surpluses are not distributed evenly among all consumers and producers. Typically, the poor, especially the rural poor, incur a large share of those costs. Control of interest rates, which is widespread in the third world, is no exception (Shaw; McKinnon; Adams at al.). 37. For two reasons, formal financial intermediaries, facing excess demand for loans at pegged rates, tend to direct credit toward large borrowers. First, the ratio of loan service costs to loan volume is minimized by doing so. Second, those borrowers are generally more creditworthy. By contrast, small farmers are costly to serve and are risky clients since they possess little collateral and since their marginal socioeconomic status renders uncertain the returns they gain from farming. Thus, they suffer disproportionately from credit shortages (Gonzalez-Vega). They may be refused credit outright. Alternatively, they may have to incur high "transactions costs," such as application fees, repeated trips to the lender (in order to usher a loan application past various stages in the review process), and mandatory deposits in accounts paying below-market interest. 38. Without interest rate reform, infusions of external funds do not always greatly alleviate credit shortages for small farmers. Frequently, - 10 - donor agencies require extensive documentation about the purposes to which disbursed credit is being put and about the groups receiving that credit. Cuevas and Graham (1984) show that documentation expenses amount to a significant portion of total costs for banks serving rural Honduras. Alone, this impact reduces the supply of rural credit, especially for small farmers. 39. Being largely excluded from formal credit markets, small farmers are obliged to finance their investments with funds obtained from informal sources. Such funds tend to carry relatively high interest rates (informal credit markets are usually beyond the purview of government regulation) not just because small farmers are less creditworthy borrowers than are those who gain access to formal markets but also because lenders in informal markets are generally too small to capture all economies of scale in, for example, risk pooling. 40. Because they rely more than other groups on informal credit markets in which real interest rates are high, small farmers are, in general, discouraged from making near term sacrifices for the sake of future gains. Among those activities are the application of conservation measures to existing farmland and the clearing of new land for crop or livestock production. 41. Commodity Prices. Just as policy-induced distortions in rural financial markets result in small farmers' paying high real rates of interest, policy-induced distortions in markets for agricultural commodities result in their receiving low prices for their crops and livestock. Influenced by urban clienteles, both affluent and poor, third world governments use price controls and other policy instruments to keep food cheap. Even a stated policy of agricultural support prices rarely benefits small farmers. Most governments lack the financial resources to effect such a policy. When the public sector makes one of its sporadic interventions in commodity markets on behalf of a support price, small farmers, who lack access to timely and reliable market information, are usually the last to know. Consequently, they are unlikely to sell commodities at the support price (Timer et al.). 42. Receiving low prices for crops and livestock, small farmers in the third world are discouraged from investing in natural resource inputs to agriculture. In particular, they are less willing both to apply conservation measures to existing farmland and to clear new land for agricultural production. 43. At least as important as a general increase or decrease in commodity prices, however, is a change in relative prices. Land degradation problems will be exacerbated if a change in policy or market forces reduces the relative profitability of soil-conserving land uses. For example, because fuelwood and charcoal prices are controlled in some African cities, incentives in the surrounding countryside to adopt agroforestry systems that yield fuelwood in addition to other commodities or to establish fuelwood/charcoal plantations are weakened (Anderson). To say the same thing, removal of price controls would promote agroforestry and plantation development. - 11 - 44. Overvalued Domestic Currency. The linkage between exchange rate policy and land resource development is the same as the relationship between pricing policy and the use and management of land. Setting official exchange rates above market exchange rates, which in the past has been standard policy (like interest rate and commodity price controls) in many developing countries, discourages the production of agricultural exports. This, in turn, diminishes derived demand for agricultural land, which discourages individuals both from managing existing farmland well and from pushing out agriculture's extensive margin. 45. In addition, to the extent that soil-conserving perennial crops (e.g., coffee and tree fruits) are tradeable while erosive, subsistence crops (e.g. maize, beans and root crops) are not, overvaluing domestic currency discourages soil conservation. For instance, Velos at al. (1985) found that currency overvaluation in the Dominican Republic discourages tha* country's small farmers from switching to agroforestry since a sizable share of such a system's revenues are obtained from the sale of coffee and citrus fruits, which are exported. An opposite relationship between exchange rates and soil conservation is observed in the savannah areas of West Africa, where a major export crop, groundnuts, is highly erosive. 46. Prices of Non-Labor Innuts to Agriculture. In addition to reducing prices paid for tradeable commodities, currency overvaluation lowers the prices farmers pay for imported agricultural inputs. Non-labor input prices can also be reduced through direct subsidization. 47. A major class of non-labor inputs is agricultural chemicals. Making them cheaper reduces the cost to farmers of soil erosion because they substitute for natural soil fertility lost through erosion. It should be remembered, however, that this kind of substitution is inhibited by non- price allocation that typically accompanies subsidized input prices in LDCs. For example, shortages caused by quotas on cheap fertilizer imports restricts small farmers' ability to apply more fertilizers to eroded lands. 48. Currency overvaluation and other policies also reduce the price of farm machinery. Section III's analysis suggests that, as large farmers respond to low machinery prices by substituting capital for labor, agricultural colonization of environmentally fragile hinterlands by displaced peasants is bound to increase. At the same time, low machinery prices encourage land clearing by large enterprises. VI. SOIL EROSION'S DOWNSTREAM COSTS 49. Up until this point in the paper, the reasons why small farmers in LDCs are more likely than other groups to aggravate land degradation problems have been highlighted. The tenure regime, distorted price signals, and some agricultural development strategies induce them to settle in environmentally fragile hinterlands while discouraging them from using conservation measures. - 12 - 50. In this section, small farmers' colonization of fragile lands as well as their failure to adopt erosion control measures are shown to affect society as a whole. The focus of this discussion is on "physical externalities." It should be remembered, however, that some of the "off- farm" consequences of environmental degradation are more purely "social." Blaikie (1985) affirms, for example, that concern over the consequences of migration from badly eroded rural areas explains the ambitiousness of the Republic of South Africa's soil conservation program. Whether or not this is indeed true, the governments of most developing countries lack the financial and organizational capacity to undertake comprehensive, nationwide efforts simply to eliminate environmental degradation as a cause of rural to urban migration. Instead, they are limited to initiating conservation projects in order to improve downstream water quality, to decrease the severity of floods, and to reduce sedimentation of reservoirs and navigable waterways. 51. Water Qualit Imnacts. Data on soil erosion's water quality impacts are scarce in most LDCs. In light of evidence about water pollution in affluent countries, however, those impacts are undoubtedly serious. In the United States, where emissions from non-agricultural sources greatly affect water quality, diffuse runoff from farms is reckoned to be very high relative to total loadings of several important pollutants. As much as 98 percent of all suspended solids and three quarters of total dissolved solids in the nation's waterways originate on agricultural land. Similarly, agricultural nonpoint source pollution accounts for up to 90 percent of all nitrogen and phosphorus loadings while creating as much as 70 percent of biochemical oxygen demand (Clark etal.). 52. Erosion and Flooding. Erosion affects both the frequency and severity of flooding in two ways. First, because topsoil holds water better than other strata, water yield from non-eroded watersheds tends to be less variable than water yield from eroded watersheds. To say the same thing, peak flow from a non-eroded region shortly after a storm is less than peak flow would be if topsoil layers in the same region were thinner. Second, the frequency with which water breaches streambanks is a positive function of the amount of eroded materials accumulated in stream beds. 53. Various authors caution against over-stating the magnitude of the first impact, which leads to exaggeration of the associated downstream benefit of erosion control (Hamilton and King). In particular, they warn that minor reductions in water yield variability should not be regarded by decision makers as a change in total flows of water. To emphasize that erosion does not influence total water yield, one is not being too simplistic to indicate that water flows downhill in eroded areas as well as in noneroded areas. 54. If retention of topsoil in a watershed does not greatly affect stream flow from a watershed, then erosion control eases flooding problems mainly because it reduces the rate at which sediments accumulate in stream beds. As stream beds build up, water flows onto floodplains, streams change their course, and roads and bridges are washed out more frequently. - 13 - All three events are troublesome for those living near or having to cross rivers and streams. 55. Erosion and Reservoir Sedimentation. A major concern for reservoir managers around the world is that sedimentation causes loss of storage capacity. To the extent that reservoirs are intended to hold a certain volume of water, displacement of storage capacity with eroded materials obliges the relevant authority to design larger dams, thereby extending the boundaries of the reservoir (i.e., permanently flooding more land). Also, loss of active storage capacity (reservoir volume above discharge tunnels to irrigation systems and/or intakes to generating stations) reduces the value of services obtained from a reservoir. Displacing active storage capacity with sediments impinges on one's latitude for storing water during periods of peak stream flow and/or low demand for hydroelectricity, irrigation water, and potable water in order to release water when it is valued more highly. 56. Sediments entering a reservoir are a problem not simply because they displace storage capacity. A portion of eroded materials pass through the reservoir and into generators and irrigation systems, thereby damaging turbine blades and other machinery and clogging irrigation canals. Also, sediments tend not to stay in one part of a reservoir. Particularly in the narrow, steeply-sloped reservoirs found in mountainous parts of the third world, rock, gravel, sand, and soil travel down reservoirs toward dams, where they can clog turbine access pipes as they slough off against dams. Because recuperation of a dam/reservoir complex after major sloughing has occurred (e.g., through dredging) is quite costly, they are frequently abandoned after such an incident. 57. Erosion and the Sedimentation of Navigable Waterways. Looking farther downstream from watersheds, one observes another adverse impact of erosion/ sedimentation. Whether it is in Burma, where deforestation up- country has caused sediment load in the lower Irawaddy River to increase, or in Argentina, where $20 million is spent each year on dredging of the Rio Plata (Sfeir-Younis), the link between erosion upstream and draft for ships in downstream navigable waterways is amply demonstrated. VII. ARRESTING LAND DEGRADATION IN THE THIRD WORLD 58. Although small farmers incur many of the costs of soil erosion and related environmental disturbance in the third world, LDC governments' initiatives to arrest land degradation reflect a preoccupation with soil erosion's "external effects," particularly its impacts on water resource development. Because a large share of the benefits of water resource development are jeopardized by sedimentation, nutrient loading, and other forms of nonpoint source pollution (Sfeir-Younis), erosion control projects are typically undertaken upstream from reservoirs, harbors, and other critical waterways. 59. Small farmers are widely perceived as major agents of natural resource degradation. Consequently, substantial emphasis is placed in the - 14 - typical conservation project on influencing their interaction with the natural environment. This emphasis is sometimes misplaced. Soil loss from land used by small farmers does not amount to a large share of downstream sediment yield in some watersheds (Hamilton and King). Even when the activities of the rural poor result in the displacement of large amounts of soil, it is not always correct to infer that a major change in their decisions regarding land use and soil management would appreciably improve downstream water quality or significantly reduce downstream sediment yields. For example, channel scour and other non-agricultural forms of erosion often increase as overland loads of sediments delivered to streams and rivers decline (Mahmood). Recognizing these possibilities, those responsible for the integrated development of land and water resources in a watershed or river basin are obliged to consider the alternative of redesigning or relocating dams and other water resource development projects. 60. Even when focusing a conservation initiative on small farmers is appropriate, the beneficiaries of water resource development, who are primary supporters of improved watershed management, might be dissapointed with the results. The initiative might fail because the targeted group finds ways to evade punitive fiscal or regulatory instruments intended to discourage erosive soil management and land use. Alternatively, subsidies and technical assistance offered to encourage conservation might not fully counterbalance the price signals and tenurial conditions that discourage erosion control. 61. Alone, conservation projects are unlikely to end land degradation in the third world. Ultimately, population growth must be curtailed so that all human beings can enjoy a satisfactory standard of living given the natural resources at hand. As indicated in this paper, however, land degradation comes about not simply because of an imbalance between the human population and the environment's carrying capacity. Agricultural development strategies pursued in some LDCs, the tenure regime facing the third world's rural poor, and distorted price signals all militate against the conservation of soil and other natural resources. 62. Reform of the decision-making environment responsible for LDC small farmers' suboptimal use and management of land is hardly a straightforward matter. As indicated in the fifth section of this paper, removing a distorted economic incentive (e.g., allowing market forces to determine the prices of agricultural commodities) without altering tenurial conditions can induce both the application of erosion control measures to existing farmland and an expansion in agriculture's extensive margin. Clearly, effective conservation policy in the third world requires an integrated analysis of the factors causing the rural poor to colonize fragile lands and to do too little to arrest erosion. The need for such an analysis will become more clear as the costs associated with land degradation continue to rise. - 15 - REFERENCES Adams, D., D. Graham, and J. von Pischke (eds.). Undermining Rural Developlent with Cheap Credit. Boulder: Westview Press, 1984. Allen, R. "The Efficiency and Distributional Implications of Eighteenth Century Enclosures," Economic Journal 92 (1982) 937-953. Anderson, D. The Economics of Rural Afforestation in Ecologically Threatened Areas. Washington: World Bank, 1986. Binswanger, H. "Agricultural Modernization" (staff working paper number 673), World Bank, Washington, 1984. Blaikie, Piers. The Political Economy of Soil Erosion in Developing Countries. London: Longman Group, 1985. Burt, 0. "Farm Level Economics of Soil Conservation in the Palouse Area of the Northwest," American Journal of Agricultural Economics 63:1 (1981) 83- 92. Cheung, S. "The Structure of a Contract and the Theory of a Non-Exclusive Resource," Journal of Law and Economics 13:1 (1970) 49-70. Ciriacy-Wantrup, S. and R. Bishop. "'Common Property' as a Concept in Natural Resources Policy," Natural Resources Journal 15 (1975) 713-729. Clark, E., J. Haverkamp, and W. Chapman. Eroding Soils: The Off-Farm Impacts. Washington: The Conservation Foundation, 1985. Commander, S. and P. Peek. "Oil Exports, Agrarian Change, and the Rural Labor Process: The Ecuadorian Sierra in the 1970s," World Development 14:1 (1986) 79-96. Crosson, P. with A. Stout. Productivity Effects of Cropland Eros2q :in the United States. Baltimore: Johns Hopkins University Press, 1983. Cuevas, C. and D. Graham. "Agricultural Lending Costs in Honduras," in Adams et al. (eds.), Undermining Rural Development with Cheap Credit. Boulder: Westview Press, 1984. De Janvry, A. The Agrarian Question And Reformism in Latin America. Baltimore: Johns Hopkins University Press, 1981. Demsetz, H. "Toward a Theory of Property Rights," American Economic Review 57:2 (1967) 347-359. Eckholm, E. Losing Ground. New York: W.W. Norton, 1976. Gonzalez-Vega, C. "Credit Rationing Behavior of Agricultural Lenders," in Adams et al. (eds.), Undermining Rural Development with CheA Credit. Boulder: Westview Press, 1984. - 16 - Gordon, H. "The Economic Theory of a Common Property Resource: The Fishery," Journal of Political Economy 62:2 (1954) 124-142. Hamilton, L. and P. King. Tropical Forested Watersheds: Hydrologic and Soils Response to Major Uses or Conversions. Boulder: Westview Press, 1983. Hayami, Y. "Induced Innovation, Green Revolution, and Income Distribution: Comment," Economic Development and Cultural Change 30:1 (1981) 169-176. Hayami, Y. and V. Ruttan. Agricultural Development: An International Peragective Baltimore: Johns Hopkins University Press, 1985. International Rice Research Institute (IRRI). Changes in Rice Farming in Selected Areas of Asia. Los Banos: 1978. MacDonald, T. "Indigenous Responses to an Expanding Frontier: Jungle Quichua Economic Conversion to Cattle Ranching," in Whitten (ed.), Cultural Transformations and Ethnicity in Modern Ecuador. Urbana: University of Illinois Press, 1981. Mahmood, K. Reservoir Sedimentation: Impact. Extent. and Mitigation. Washington: World Bank, 1987. Marshall, A. PrinciRles of Economics(9th ed.). London: MacMillan, 1961. McConnell, K. "An Economic Model of Soil Conservation," American Journal of Agricultural Economics 65:1 (1983) 83-89. McKinnon, R. Money and Capital in Economic Development. Washington: Brookings Institution, 1973. Myer, N. "The Present Status and Future Prospects of Tropical Moist Forests," Environmental Conservation 7:2 (1980) 101-114. Nations, J. and D. Komer. "Indians, Immigrants, and Beef Exports: Deforestation in Central America," Cultural Survival Quarterly 6:2 (1982) 8- 12. Posner, J. and M. McPherson. "Agriculture oa the Steep Slopes of Tropical America," World Development 10:2 (1982) 341-354. Randall, A. Resource Economics. New York: John Wiley, 1987. Sfeir-Younis, A. "Soil Conservation in Developing Countries," Agriculture and Rural Development Department, World Bank, Washington, 1985. Shaw, E. Financial Deepening in Economic Development New York: Oxford University Press, 1973. - 17 - Sidhu, S. "Economics of Technological Change in Wheat Production in the Indian Punjab," American Journal of Agricultural Economics 57:2 (1974) 217- 226. Southgate, D. F. Hitzhusen, and R. Macgregor. "Remedying Third World Soil Erosion Problems," American Journal of Agricultural Economics 66:5 (1984) 879-884. Southgate, D. and D. Pearce. "Natural Resource Degradation under Agricultural Colonization: A Causal Analysis" (discussion paper), Department of Economics, University College London, 1987. Taussig, M. "Peasant Economics and the Development of Peasant Agriculture in the Cauca Valley, Colombia," Latin American Perspectives 18:5 (1978) 62- 98. Timmer, C., W. Falcon, and S. Pearson. Food Policy Analysis. Baltimore: Johns Hopkins University Press, 1983. U.S. Agency for International Development (USAID). Environmental and Natural Resource Management: A Report to Congress. Washington: U.S. Government Printing Office, 1979. U.S. Department of Agriculture, Agricultural Stabilization and Cooperative Service (USDA/ASCS). National Summary Evaluation of the Agricultural Conservation Program, Phase I. Washington: U.S. Department of Agriculture, 1978. Veloz, A., D. Southgate, F. Hitzhusen, and R. Macgregor. "The Economics of Erosion Control in a Subtropical Watershed: A Dominican Case," Land Economics 61:2 (1985) 145-155. Ward, A. Progress for a Small Planet. New York: W.W. Norton, 1979.