2.1 Impact of Degradation on the Poor
2.2 Poverty Impact on Resource Management
2.3 The Links Between Poverty and Land Degradation - Mixed Empirical Evidence
2.4 Household Level Effects of Degradation
2.5 Conceptualizing the Linkages Between Poverty and Land Degradation
Lipton (1997a) states forcefully that it is irrational to expect people to knowingly behave in ways that destroy resources necessary for their survival or that of their future generations49 unless very strong pressures to do so are present.50 He lists four such pressures generally discussed in the literature.51 These include (1) increases in population as mortality falls but fertility declines lag and (2) declines in common property resources (CPRs). In addition there are international pressures; including (3) interest rate changes and (4) technology transfers [Lipton (1997a)].
49 Often the problems of poverty, population and the environment are intertwined: earlier patterns of development and the pressure of rapidly expanding population mean that many of the poor live in areas of acute environmental degradation [World Bank (1990)].
50 The World Bank maintains a similar position. "The poor do not willfully degrade the environment but poor families often lack the resources to avoid degrading their environment. The very poor, struggling at the edge of subsistence, are preoccupied with day to day survival. It is not that the poor have inherently short horizons; poor communities often have a strong ethic of stewardship in managing their traditional lands. But their fragile and limited resources, their often poorly defined property rights, and their limited access to credit and insurance markets prevent them from investing as much as they should in environmental protection. When they do make investments they need quick results [World Bank (1992 p. 30)].
51 According to the World Bank the main source of pressures generating problems of degradation is thought to lie in rapid population growth. Other pressures come from the widespread use of natural resource intensive technologies; ineffective regulation of common property resources; land tenure systems that do not secure long terms rights to land use; and policies that distort the prices of non-renewable resources [World Bank 1991, in Biot et al. (1995)].
Poverty generates significant incentives to have large families. Traditionally the impact of population growth on natural resources was discussed in terms of "carrying capacity.52" Conceptually, if nothing else changes, then it is assumed that the increasing population will put demands on the resources that can no longer be met without damaging the ability of these resources to support human life. Social and economic factors such as trade, technology, consumption preferences and levels of inequality can alter the carrying capacity. Poor people will often use migration as a coping strategy. However, migration may not always benefit rural environments since the absolute numbers of rural people may continue to increase.
52 Attempts to compare current and projected populations to potential population supporting capacities (PSCs) at certain levels of technology have found that with low input technologies typical of current production practices 1975 populations had already exceeded carrying capacities in several West African countries. The study by Higgins, Kassam and Miken (1983) predicted that 7 of the 8 Sahelian countries will exceed population-supporting capacities by the year 2000. Regional imbalances and environmental damage were greatest in the Sahelo Sudanian zone despite low population densities.
Lipton (1997b) notes that technology generation in agriculture remains exogenous to most of the developing countries and is not driven significantly by their resource saving or other requirements. This is the classic choice of techniques problem highlighted in the literature on industrial development during the 1970s that first made popular terms such as ''technological determinism." This argument holds that the technically efficient techniques are generally developed in the capital-abundant labor-scarce developed countries and generally reflect the factor endowments of these countries.
53 Much of the discussion in this and the following subsection draws heavily from Mink (1993)
The poor generally have access only to areas that have higher risk for health and income generation.54 And they generally lack the resources to reduce the exposure to the risk or to invest in alleviating the causes of such risk. Environmental degradation therefore can affect the health and nutrition status of the poor and lower their productivity. This can happen both directly through, for example, lower yields per unit of labor or land because of reduced soil quality, and indirectly through the reduced physical capacity of labor to produce because of malnutrition and poor health. Even in cases where the poor are healthy labor productivity can be low due to increased time being allocated to less-productive activities such as fuel wood collection and away from agriculture and other income generating activities [Kumar and Hotchkiss, 1988]. In terms of the productivity of the resources that the poor manage, the decline is intricately related to the poverty-population-environment interaction [Mink (1993)]. Where the poor depend on biomass fuel and confront increasing fuel wood scarcity they often shift to using animal dung, fodder and crop residues for fuel. The quantities of these materials that are returned to the soil are thus reduced and its fertility declines.55 Non-replenishment of soil nutrients leads to soil exhaustion as fuel wood supplies diminish and animal manure is increasingly used as a fuel substitute. Poverty forces a trade-off between the immediate demands for fuel for cooking and heating and manure for the land. The time-preference argument suggests that the immediate and urgent needs be satisfied. Mortimore (1989) shows how soil exhaustion occurs when certain nutrients are taken from the soil but are not replenished naturally or artificially with fertilizers. A homogenous crop, usually a cash crop, grown repeatedly on the same piece of land can lead to soil exhaustion.56 Increasing population pressures on land can also lead to shortened fallow periods and this coupled with the farmer's inability to apply variable inputs more intensively because of poverty, can lead to decreased soil productivity. Productivity, especially, in open-access natural resources or of resources under deteriorating common property management may often decline due to over-use.
54 The most debilitating risk is that of drought in semi arid tropical areas. The combination of poverty and drought can have serious environmental consequences that threaten future agricultural productivity and the conservation of natural resources. Poor people are induced to scavenge more intensively during droughts, seeking out wood and other organic fuels, wild life and edible plants, both to eat and to sell. This scavenging aggravates deforestation and damage to watersheds and soil already under stress from the drought. The problem is aggravated in common property pastoral farming where farmers carrying extra cattle as insurance against drought may exploit and over burden the carrying capacity of the land increasing the likelihood of permanent damage. Small ruminants can be exceptionally damaging to resources. Poorer households are generally responsible for raising small ruminants, which are allowed to graze low quality resources especially on open access and common property land [IFAD (1992)].
55 The loss in grain production as a result of diverting dung from fertilizer to fuel use in Africa, the near east and Asia has been estimated at up to 20 million tons per year [Redclift and David (1990)].
56 Given the declining yields on the land and the inability to find the institutional support in terms of fertilizer and access to credit and technology, poor farmers are forced to sell their land and become land less peasants or to encroach on new forest lands [de Graaff (1993)].
Poverty is generally assumed to impose short time horizons.57 Theoretically this results from the poor having high rates of pure time preference which lowers the ability to forego consumption today. This leads to using up savings previously set aside for later consumption and to borrowing if access to credit is available. The implications of a high subjective discount rate are rapid resource extraction to meet present income or consumption needs and low investment in natural resources to improve future returns. Overgrazing of pastures and shortening of fallow periods can result from the high subjective discount rates. Similarly, farmers are less likely to make natural resource investments where returns are expected after a number of years. These factors combine to lead to a wide divergence between private and social discount rates.58 The empirical evidence on whether the poor really do have high rates of time preference is limited and sketchy.59
57 This is not to say that short time horizons are exclusive to the poor.
58 Veloz et al. (1985) in their analysis of a soil conservation project in the Dominican Republic show that soil conservation is profitable on only 20 percent of the land area using private analysis. Alternatively social analysis based on discount rates that reflect the society's inter temporal preferences, indicate that soil conservation is viable in nearly 70 percent of the land area.
59 The ICRISAT study by Pender and Walker (1990) which estimated high rates of time preference through experimental games for a small sample of poor farmers in India is generally cited as an example.
Risk aversion can lead to a short time horizon. To the extent that outcomes in the future become less certain than outcomes closer to the present, people will prefer to trade the more uncertain outcomes for the more certain ones. Risk aversion amongst farmer is widely documented [e.g., Binswanger (1980), Walker (1981), Grisley (1980) and Sillers (1980)]. The results of these studies generally indicate that attitudes of the poor to risk are not distinguished from those of the non-poor by innate or acquired characteristics but by the higher levels of risk faced by the poor and by the greater constraints to coping with these risks. Deteriorating land quality brings not only poorer yields but also greater yield fluctuations and hence higher risk.60 To the extent that access to common property resources serves as insurance for the poor in times of setbacks to the primary sources of income, the decrease in access can increase the risk. Migration can benefit the environment through mitigating risk.61 Individual migration is increasingly seen as an outcome of family decision making, particularly in response to uninsured risks [Stark (1991)].
60 Reardon and Vosti (1997) note that generalized poverty erodes traditional community risk sharing or insurance institutions by over taxing them; forcing the poor to fend for themselves often turning to resource mining and commons dependent strategies.
61 Remittances are an important coping strategy for rural poor [Alderman and Paxson (1992)].
The poor face greater constraints to managing their risks. Their assets and stored production are generally minimal. Their access to credit and insurance is generally limited and or non-existent. Rural credit and insurance markets in developing countries are notoriously fragmented. In most cases there is also a gender bias so that poor women have far less access to mechanisms for managing risk than their male counterparts. If risk is allowed for, the interest rate incentive to deplete is probably sharpened. "Higher interest rates reduce the present value burden of long term future risks relative to that of near term risks (and costs). The land use patterns are therefore shifted towards activities with long-term risks such as possible long-term resource degradation. There is thus a powerful resource depleting incentive created by rising interest rates. Costly credit undoubtedly shifts the composition - of inputs, outputs, techniques, investment, consumption and savings - sharply in a resource depleting direction" [Lipton (1997a)].
The study by Grepperud (1997) concludes that in the relationship between poverty, land degradation and climatic uncertainty it is unclear whether poverty in general induces farmers to manage their resources poorly in the long run. The study by Scherr, Jackson and Templeton (1995) also found no consistent relationship between population density or the frequency with which land is used for productive purposes and degradation of the land. Population growth and poverty, they noted, create both incentives and disincentives for land degradation. There is an extreme dearth of studies that seek to rigorously test these relationships. The lack of appropriate data underlies this paucity. To do this effectively information is required not only on the physical aspects of the land but also of poverty and a host of other factors that need to be controlled for. Such data are not available at the present time. Reliance therefore has to be placed on studies from which the relationships can be inferred.
Most of the available studies look at the problem in terms of the behavior of small-scale farmers and land degradation. Southgate (1988) maintains that small-scale farmers have been the main agents responsible for land degradation activities. He states that market and institutional failure were the primary causes for farmers adopting non-sustainable practices. Pagiola (1995) shows how government price controls on agricultural goods in Kenya have not provided incentives for the small-scale and poor farmers to conserve their land. In some cases this has led to the mining of resources for maximum output. Mortimore (1989), on the other hand, finds evidence of small-scale farmers' willingness to forgo short-term income gains even under price and famine pressure to pursue long term sustainable management strategies. The existence or non-existence of secure land tenure systems might explain the contradiction of results about small farmer behavior. Several studies cite the lack of secure land tenure as the primary reason for poor farmers cultivating their land excessively to exhaustion for the simple reason that they have no vested interest in conserving an asset that they do not own [see, for example, Southgate (1988), Mink (1993), Repetto et al. (1989)].
Change in agricultural practices can have primary and secondary effects on the environment. Von Braun (1997) describes the relationship between agricultural change and the eventual effects at the household level through these environmental effects. Such change has come about in the large part of the world through the adoption of the green revolution type technologies. Agricultural change can also occur where green revolution technologies have not been (as yet) adopted. In the case of the latter, the primary effects on the environment are generally stated to be in the form of desertification, deforestation, watershed degradation, soil erosion and soil fertility decline. The secondary effects can be droughts and floods. These environmental effects can translate into specific effects at the household level. These effects can take the form of impoverishment/productivity decline, migration-related health stress, vector borne disease (if the migration occurs into disease prone areas), communicable disease (when sanitation breaks down), chronic food insecurity, seasonal malnutrition and famines. In the case of the green revolution technology, potential environmental degradation can result from each element in the technology package. It can result from the direct use of each of the technology elements and through indirect effects as well. For example, irrigation can lead to reduced water quantity or quality, salinization, increase in mosquitoes, aquatic snails and blackflies. Inappropriate pesticide use can have harmful household effects. Fertilizer use can result in nitrates leaching into drinking water. At the household level these aspects of potential environmental degradation can translate into diseases such as diarrhea, cholera, typhoid, malaria, schistosomiasis, onchocerciasis, poisoning and diseases of the circulatory system in infants. The secondary effects of the use of such technology can be crowding, sanitation deficiency, diet change and vector control (through inappropriate pesticide use). These can lead to communicable diseases, nutritional diseases and poisoning etc. These household effects imply a reduction in welfare, which under the conventional consumption based methods of measuring poverty, might not show up as such. That is why it is important to include the non-income measures of poverty such as anthropometric measurements in assessments of the poverty status.
Vosti and Reardon (1997) present an interesting conceptual model of the linkages between poverty and the environment that helps to highlight the complexity of the relationships. Poverty is seen to be the product of "asset" components comprising natural resources (private and commonly held), human resources, on-farm resources, off-farm resources, community-owned resources and social and political capital. These links are shown in Figure 2. These determine household and village behavior in terms of income generation, consumption, investment in assets, migration and human fertility, which in turn has implications for use and management of the natural resource component that, - determines the asset components of poverty. How natural resources are used and managed feeds back as a determinant of the asset components of poverty. A set of conditioning factors governs the relationship between the asset components of poverty and household and village behavior and between the household and -village behavior and the natural resource components. These conditioning factors are markets (prices), village and regional infrastructure, technologies (production and conservation), village level asset poverty and population pressures.
This conceptualization leads to innovative policy implications. In comparing traditional productivity investments such as irrigation, fertilizer and modern seeds with conservation investments (such as bunds, terraces, windbreaks and practices such as organic matter application) the study finds that the latter have different requirements and characteristics. Conservation investments need innovative policies beyond just "getting prices right". The three non-price policies suggested by the study are: complimentary public infrastructure investments (such as culverts to divert water flow from farm bunds) that make household investments more profitable to institutional innovations; that improve security and transferability of resource tenure; and, that modify community level arrangements to improve the management of the commons or watershed [Vosti and Reardon (1997)]. In the same book, von Braun (1997) also points out that poor communities lack resources for community level investments such as physical infrastructure, health and education. Policies that strengthen traditional institutions and make them more flexible (particularly in the face of increasing population pressure) can reduce poverty and the dependence of rural poor on resource miming especially in response to draughts and floods.
Defining poverty in the Vosti and Reardon (1997) manner sets a much higher cutoff than the conventional definition. Implicit in this conceptualization is the assumption that sizeable resources62 over and above meeting bare subsistence consumption and production are required by the poor to address issues of resource degradation. While this model provides an interesting tool for conceptualizing some of the complexities involved; it also highlights the trade-off between the depth and detail of understanding and concomitant data requirements; and, the inadequacy of available methodology and resources for measurement.
62 Estimates of the capital costs of prevention vary with the farming system, the methods used and topography. Expenditures of $50-$150 per hectare (sometimes less) for such measures as farm forestry and contouring with vetiver grass or other vegetative barriers are typical; $200-$500 may be required per hectare for structural measures (terracing, land leveling, earth banks and the like) on undegraded lands. Rehabilitation, in contrast, may cost from $500 to several thousand dollars per hectare, depending on the severity of the problem [FAO (1992)].
Figure 2. Poverty and environment links
Source: Vosti and Reardon, 1997.
Duraiappah (1996) also presents an interesting conceptual model for analyzing the many complex inter-relationships between poverty63 and environmental degradation. For simplicity he postulates four possible, though not mutually exclusive relationships64. These are
R1: Poverty leads to Environmental Degradation
R2: Power Wealth and Greed leads to environmental Degradation
R3A: Institutional Failure leads to environmental degradation
R3B: Market Failure leads to Environmental Degradation
R4: Environmental Degradation leads to Poverty63 He defines indigenous poverty as poverty caused by environmental degradation and exogenous poverty as that caused by factors other than environmental degradation.
64 He postulates three crucial initial conditions: 1) no environmental degradation, 2) no indigenous poverty and 3) the possibility of the existence of exogenous poverty.
If only R1 is observed then the poverty induced environmental degradation argument can be accepted. However based on the initial conditions only exogenous poverty can cause this environmental degradation. On the other hand if only R2 is observed then policies adopted under R1 assumptions can be misleading and may in fact exacerbate the degradation process, as demonstrated by Binswanger (1989). In case of either R3A or R3B being responsible for environmental degradation, the solution is theoretically relatively simple - remove or correct the market or institutional failure. If R4 is present two interesting observations arise. First R4 can only be present if it is caused by R1, R2, R3A, or R3B or various combinations of all four. Second, the presence of R4 can set into motion an R1 type of link but in this case it is indigenous poverty, which causes the environmental degradation. This is the R1 feedback or R1FB link.
In the R1, R4 link two outcomes are possible. The first scenario would be that R1 causes R4 and the causality link ends. On the other hand we can get a situation whereby the indigenous poverty caused by R4 sets into motion more environmental degradation by a R1FB relationship. The downward spiral of poverty leading to degradation leading to more poverty [Durning (1989)] is an R1FB type of relationship. The various permutations and combinations of these four relationships highlight the complexity of the relationships.
The model has four contributing forces namely: the power greed and wealth factor; exogenous poverty; institutional failure; and, market failure. It addresses two externalities namely environment degradation and indigenous poverty. The fear of losing land by the poor is a direct function of R2. R3A is also a primary contributor to land degradation in this manner. R1FB can be a contributory factor for soil exhaustion because of two reasons: first from within the sector due to decreases in agricultural productivity, and second from the fuelwood-manure relationship. In the first case, there is evidence of declining agricultural productivity in degraded lands causing indigenous poverty, which in turn forces many of the people to continue to degrade their land further to extract subsistence outputs. The R2 link in the forest sector can cause an R1FB affect in the land degradation category. R2, R3A, R3B and R1FB linkages can cause salinization. In the case of desertification, the primary links highlighted by Durriapah are R2, R3A and R3B.
Durriapah concludes that most environmental protection programs fail because they address only the symptoms while they ignore the causes, i.e. they address only indigenous poverty and ignore its causes.