- Prose Before Hos

Evaluating The Poverty – Environment – Population Nexus

Posted on October 22, 2008 in Images

1. Introduction

The feedback loop formed by the population, environment, and poverty nexus (PEP nexus) creates an analytical framework in which to explain the causes of persistent poverty in the developing world. By including environmental and demographic processes in the development paradigm, theorists have succeeded in bringing attention to elements of underdevelopment previously ignored or disregarded. However, the PEP nexus fails to consider other important elements such as migration, public policy, political economy considerations, and realistic developmental interpretations of property rights and labor markets. Empirical evidence from Nepal and Ethiopia will illustrate that while population and environment do act as significant, but partial, determinants, poverty is complex and influenced by a plethora of factors. Though certainly demographic and environmental pressures contribute to destitution and developmental failures in the developing world, it is an oversimplification to reduce poverty to a cyclical nexus.

2. Examining the PEP Nexus

The PEP nexus attempts to bridge processes relating to environment and demographics that are regarded as fundamental to the persistence of the poverty trap in the developing world. Though earlier authors tried to create a stricter line of causation, the correlation between the three has been best examined by recent authors as a feedback loop. Increased relative poverty, fueled by population increases, both on the household and macro level, hypothetically heighten pressure to convert forests and other land into crop production because of low agricultural productivity and input use (Barbier 2000, 2). Poverty is then recreated, as in developing countries, agricultural land expansion is associated with lower levels of per capita GDP (Barbier 2006, 164).

Aggregate empirical evidence at times supports the interrelated perpetuation of the PEP nexus. World Bank data from 2001 shows that between 5 and 12 million hectares of land are lost annually to severe degradation (Dasgupta 2003, 10). The regions with the highest population growth rates – Africa and Asia – correspondingly experience the highest rates of soil degradation respectively at 65 percent and 40 percent (Dasgupta 2003, 10). From 1980 to 1990, 15 million hectares of tropical forest were cleared annually at a rate of .8 percent a year, mostly for direct conversion to large and medium scale agriculture (Barbier 2000, 2). Thus, most agricultural expansion was for crop production, implying that countries pursued deforestation to answer demand for food arising from increased population pressures (Barbier 2000, 2). Correspondingly, cultivated land is expected to increase 47 percent by 2050 with two-thirds of new land originating from deforestation and land conversion (Barbier 2000, 3). Analysis of this encroaching phenomena is not only timely but necessary for long-term developmental growth strategies.

2.1 Macro and Micro Models of the PEP Nexus

The Barbier and Burger model utilizes a cross-country macro analysis of agricultural land expansion to elucidate the factors influencing land conversion, as conversion is identified as a ‘major cause of forest loss in developing countries’ (2000, 3). In the study, four distinct analytical frameworks are employed, including the environmental Kuznets curve, competing land use models, forestland conversion models, and institutional models. From these frameworks, economic factors including per capita income, agricultural yield, and key institutional factors are determined to be possible factors in agricultural land expansion, creating the following ‘synthesis’ model.

In the equation, the left hand side is equivalent to expansion of agricultural land area, while Yit is per capita income, sit represents country-by-country differences in agricultural sectors and land use, zit is other explanatory variables such as rural population growth, and qi takes into account institutional variables (Barbier 2000, 4). Using data from the Latin America, Asia, and Africa from 1961 to 1994, the model produced several results. First, neither Environmental Kuznets Curve hypothesis nor the GDP per capita are significant explanatory variables ‘in any versions of the model’ and failed joint significance test (8). Secondly, structural variables are ‘strongly robust’ as predictors, as land use patterns, agricultural export share, increasing rural population, and cropland share of land area were all significant. Thirdly, of the institutional variables, only control of corruption displays direct influence on land expansion (2000, 8). Barbier believes there is a fairly intuitive explanation, as government officials may be bribed by landowners to circumvent or ignore land control restrictions (2000, 8). Fourthly, the element that displays the most relevance is terms of trade, where increases in terms of trade appear to directly spur agricultural land expansion (2000, 8). The model shows that structural variables have the greatest explanatory value, concluding that ‘if a developing economy has a sizeable reserve of potential cropland available, increased conversion of this frontier land will occur as agricultural development proceeds in the economy’ and greater dependency on agricultural and raw material exports is fueling land conversion in developing nations (2000, 11).

On the micro level, the Angelsen model tries to capture determinants in deforestation caused by household decisions where Angelsen believes there is significant confusion in the PEP debate (186). In this model, decision making is simplified to variables of household preferences, labor market integration, property regime, and costs of labor – determinants that Angelsen believe show the greatest variation throughout developing countries (187). Costs are revealed in primarily the location cost, measured by the distance between villages and fields. Below respectively are the equations that represent optimal crop, land area under cultivation, and labor input for a household:

This is the basic formulation of household crop optimization, where X is the total production of a house, x is the crop yield per hectare of land, and H is the total land area (188).

Individual households cultivation, represented by H, will use the total agricultural land that forms a circle around the village. From this, b is the distance from the village to the field, bmax is the maximum distance where cultivation takes place, N is the number of households, and m is land use intensity (188).

This equation reveals labor inputs for household — L is the total labor for a household, q is the labor time spent on transport to fields, and b remains distance to the field from the village (189). Though Angelsen presents four models, the ‘full belly’ or subsistence household model will be focused on as it may present the most insight for those likely to be affected by fluctuations in environment and population in developing economies. The households objective is to minimize labor while attaining a subsistence target in a private or communal property rights regime with no access to alternative labor markets (190). Thusly, the production for a household (xH) must equal the subsistence target (cmin) that can only be achieved through increasing the land of cultivation:

This creates expansion where the ‘agricultural frontier is determined by productivity, population, and subsistence requirement’ (Angelsen 193). From this model it can be inferred that increases in the value of production will reduce deforestation if subsistence objectives can be realized from smaller agricultural areas. If large proportions of households in developing economies are using the subsistence strategy, total area of cultivation will be directly proportional to population. Similarly, if requirements for subsistence increase, the agricultural frontier will necessarily be expanded. Most importantly, the previous two inferences mean that population growth or increases in subsistence preferences will directly lead to deforestation, but can be offset if labor or technological inputs can be made proportionality more efficient. Though rigid in its regards to access to labor and trade markets by removing them completely, the subsistence model more accurately depicts rural labor markets where mobility may be sideways at best and distorted trade markets are the rule rather than the exception. Despite the fact that Angelsen is somewhat dismissive of the full belly model, it provides a realistic tool for policy makers where subsistence agriculture often compromises the majority pursuit in developing economies.

2.2 Contribution of the PEP Nexus and Policy Considerations

“Even when studying the semi-arid regions of sub-Saharan Africa and the Indian subcontinent, economists have usually not regarded population, growth, poverty, and the state of the local environmental resource-base as interconnected. Inquiry into each factor has in large measure gone along its own narrow route, with discussions of their interactions dominated by writings which are in the main descriptive, not analytical.” (Dasgupta 1998, 1)

The PEP nexus fills an important gap in developmental literature where economists have failed to recognize the significance of environment and population in assessing causes of the poverty trap. Even highly-regarded and dynamic theorists such as Sen, Easterly, and Collier display a ‘lack of interest in what makes for sustainable well-being or in taking the role of nature seriously’ by discluding causes of environmental dilapidation and population growth in poorer regions (Dasgupta 2003, 5). Economists often ignored demographic considerations in development strategies, especially in the 1980’s during the pursuit of structural adjustment. In a 1986 study of economic consequences of population growth in poor countries, investigators concluded that while development would be ‘faster with slower rates of population growth, there is no cause for alarm over high population rates’ (Dasgupta 1998, 2). It has only been in the past 15 years with the advancement of the PEP nexus that donor agencies have recognized the significance of interrelated environmental and demographic processes on long term poverty reduction and growth strategies.

Policy implications can be expanded upon from the analytical models presented by Angelsen and Barbier. Population pressures can create short-term responses that will have debilitating long-term effects on the environment, especially when there is low mobility of the labor force and local economies are agriculturally dependent (Angelsen 204). The subsistence model presented by Angelsen is helpful in its simplicity and development model application, as it focuses on high population growth, low agricultural productivity, and household struggle to reach subsistence (Angelsen 207). Barbier’s model also presents important implications, especially in engaging in policy that seeks to reduce land conversion through lowering the terms of trade that may invariably be offset by increased institutional corruption (2000, 11). Indeed, in the words of Angelsen, “PEP approaches will make policy makers face some potential and unpleasant conflicts between poverty reduction and limiting deforestation” (211). In a world where agricultural land is the most important source of natural wealth in developing nations, the pressures and processes that continue to fuel poverty and destabilization must be examined (Barbier 2000, 1). The PEP nexus provides a sufficient framework for this by including environmental and demographic issues that have typically eluded previous developmental economists.

3. Empirical Examples in the Developing World

3.1 Nepal

Nepal, a country heavily dependent on forest ecosystems for rural livelihoods, provides excellent insight into the processes that can negatively affect environmental stability and agricultural poverty. Forestry represents the dominant land use system with 29 percent of land covered by forest, and an additional 10 percent covered by shrub (Acharya 150). Moreover, forests are an integral part of the farming system and are heavily utilized for multiple functionalities, such as animal fodder, firewood, agricultural implements, and timber for building (Acharya 2). More than 80 percent of the population depends on subsistence farming, meaning forests are important from a ‘socio-cultural’ and economic standpoint (Acharya 149). Since attempts to nationalize forest areas began in 1957, Nepal has shifted from nationalization to semi-privatization to local management, only to fully return to the community based management system in the past decade after severe environmental depletion.
Shift in policy began in 1957 when the Nepalese government nationalized forests in an attempt to maximize resource utilization to widen the tax base and increase food production (Gautam 136). Subsequently, over the period of nationalization enactment, massive deforestation and environmental ruin occurred (Gautam 136). From 1979 to 1994, Nepal lost approximately 14 percent of forest and shrub cover (Gautam 142). In the past decade, Nepal has moved to a community forestry program that empowers user groups to regulate local forestry, reminiscent of the indigenous systems in place before 1957 (Gautam 146). This has met with several successes. The new system has witnessed the reversal of the deforestation process, institution building and enhanced economic benefits for local people. Gautam describes the approach as “institutions built upon established systems of authority” that allowed for successful monitoring and enforcement mechanisms (146). Indeed, for the first time in five decades, the annual rate of change in forest has increased (Gautam 142).

Though Bluffstone identifies the building of the rural labor market as the main catalyst of the turn around of Nepal’s environment, it becomes a matter of judgment (and thus a considerable issue with PEP nexus) how to evaluate significance where disaggregating results is nearly impossible. Further, over this time period population growth rates remained fairly stagnate at approximately 2.6 percent with total fertility rate being measured at 5.6 percent in 1997 by the UN, suggesting that variations in population stresses were not the cause for environmental instability. As well, no noticeable trend seems to exist between population trends and growth and per capita GDP. Indeed, public policy, not pervasive population or poverty pressures, may be responsible for the environmental fluctuations experienced in Nepal.

3.2 Ethiopia

In the late 1990’s, Ethiopia sustained consecutive years of drought that culminated in 2000 with severe increases in poverty and deprivation. Beginning in 1998, a poor rain season resulted in that years harvest realizing only 60 percent of normal yields (Carter 2007, 846). Continued droughts in 1999 and 2000 led to 90 percent loss of crops and a 75 percent decline from normal yields (Carter 2007, 847). Drawn out and with uneven consequences from a gradual onset, Carter termed the Ethiopian disaster a ‘prolonged event’ of environmental shock (2007, 846). Environmental shocks like these destroy family assets and send households into coping choices that may push them below the minimum asset threshold and into a poverty trap (Carter 2005, 1).

In an economically perfect world, households could rely on insurance and credit to replace lost assets or engage in off-farm labor markets to substitute for lost income. However, these options are rarely available for rural households in developing countries like Ethiopia. Ethiopia, as Carter says is ‘characterized by weak labor markets and nearly absent credit markets’ (2007, 847). Further, hegemonic agrarian structures in Ethiopia may have limited market-based recovery strategies had they existed (Carter 2007, 836). When market and informal assistance mechanisms are absent, coping becomes problematic as households try to maintain level of consumptions by selling valuable resources, making poverty traps more probable.

Interestingly, Carter identifies household participation in social networks as a main determinant of likelihood of escaping poverty traps after environmental shocks. In Ethiopia, 65 percent of households participated in funeral clubs, 67 percent in farm work groups, and 18 percent in religious organizations (Carter 2007, 847). Local social mechanisms assume more importance during recovery periods, when isolated households were more likely to fall under the poverty threshold (Carter 2007, 849). Measuring assets by livestock, Carter identifies social capital as increasing the rate of growth and limiting the rate of loss during and after environmental shocks (2007, 850). Conversely, though Carter acknowledges community membership primarily increases the assets of higher wealth groups, other factors such as foreign aid, labor market access, and availability of food aid does not ‘appear to protect households assets’ and in the case of food aid, ‘in fact has the opposite effect’ (Carter 2005, 3). Building productive social safety nets may effectively keep households above critical poverty thresholds during times of environmental crisis, thus development policy needs to be cognizant how social networks operate (Carter 2003, 853). In addition, it seems negligible if the environmental shock was dependent upon demographics and poverty in the PEP nexus, as witnessed in data on population growth and per capita GDP growth in the 1990’s . Though levels of poverty factored into the ability to respond to environmental shock, causation and strategies engaged in by rural Ethiopian households seem distinctly outside the poverty trap illustrated in the PEP nexus.

4. An Incomprehensive Perspective: Weaknesses in the PEP Nexus

“Even though there probably are only a few pathways to economic prosperity, the number of routes societies can take to experience stagnation – even decay – are many.” (Dasgupta 2003, 1).

Though population and environmental pressures contribute to continued destitution, they do not capture the entirety of an experience that often involves other economic, social, and political concerns. The direct causal relationship stressed by some authors is extremely suspect as well. As Barbier states, frontier land expansion, resource exploitation, and environmental degradation may be associated with poor economic performance and poverty, but is not the cause of it (2006, 165). The PEP nexus would be improved by the inclusion of other demographic and economics elements in its framework, particularly property stability, national policy, migration, and political economy impacts. Confusion over how to evaluate demographic and environmental systems, contradictory conclusions, impractical policy recommendations, and an incomplete understanding of property rights are evident in several PEP nexus expositions.

Local systems in developing countries often rely on community-based management programs that conserve environmental resources and produce economic stability via pooled efforts to lower individual risk. However, the process of economic development can ‘erode traditional methods of control by way of increased urbanization and mobility’ (Dasgupta 1998, 22). For Dasgupta, ‘even a marginal decline in compliance can trigger a process of cumulative causation’, leading to over-exploitation of the commons and thus degrading local environmental resources (1998, 22). Further, when privatization of property rights serves as growth strategy, it is at the expense of communal mechanisms that may successfully manage environmental resources and provide economic safety nets for the rural poor. As identified by Angelsen, privatization can unleash ‘deforestation as a title establishment strategy’ (186).

Social and political considerations could also add to the debate on examining persistent poverty. Dasgupta notes “societal features could be important, but have not yet found expressions in quantitative form on a national basis” and thus are overlooked by economists (2003, 5). Inequalities in wealth among rural households may affect land degradation and deforestation processes, where “rural elites in developing countries are able to steer policies and programs meant to increase rural productivity into capital-intensive investment programs for large farms” (Barbier 2006, 61). In his work, Angelsen openly discounts the roles of policy and government intervention, stating the “lack of empirical evidence on the links between deforestation and government policies magnifies the importance of using an explicit analytical framework when drawing conclusions about this important policy issue” (186). However, this topic has seen recently renewed application as the World Bank and other donor agencies have sought to explain economic underdevelopment within an institutional framework. The case study in Nepal illustrates that empirical case studies do exist and that more may emerge in the near future as developing nations reexamine the consequences of privatization and the structural adjustment programs of the recent past.

Though seeking to explain the same progressions, PEP authors have failed to create a cohesive view on how to appraise environmental degradation cobbled with unrealistic policy proposals. Angelsen believes peasant economies are characterized by ‘partial integration into imperfect markets’, which is critical in a deforestation process highly sensitive to fluctuations in off-farm wage rates (190, 208). However, he later concludes that lower access costs, primarily roads, fuel deforestation, although roads and other infrastructure improvements develop local labor markets that facilitate the non-agricultural employment that hypothetically decreases deforestation in his models (211). There is also much disparity in the type of environmental impact examined in the creation of poverty. For Dasgupta and Angelsen, environmental breakdown is best measured by long term, user created impacts, typically measured in deforestation. For other theorists, external environmental shocks and disasters present the best starting point for investigating environment’s relation with poverty. Still, Angelsen thinks the most relevant issue for the PEP nexus is resources already under user control, which remain relatively unexamined and which would ‘substantially clarify the debate on environmental degradation’ (207). That is, how are population pressures sparking agricultural intensification that lead to soil mismanagement and erosion. Local resources also provide a common good in the form of protective functions and provision of non-timber forest products that is not identified in PEP models, though it may decrease social welfare (Angelsen 187).

5. Conclusion

The PEP nexus, the first concentrated attempt to analyze environmental and population influences on poverty, does an admirable service of introducing serious demographic and environmental concerns into the development paradigm. Aggregate empirical evidence along with advanced macro and micro analytical models provide the relevant backing for inclusion into policy making in reducing poverty in the developing world. However, a more critical perspective on the PEP nexus is necessary, and future refinement of underlying assumptions should shift towards a more sensible view of developmental realities. Indeed, contemporary debate on the persistence of poverty must include examinations of environment and demographics, but would also greatly benefit by considering public policy, property stability, and political economy concerns. In examples presented in Nepal and Ethiopia, the whole picture contains external and unaccounted factors predominating within the PEP nexus. While the PEP nexus has highlighted and heightened the timely debate on environment and population, future models should seek to incorporate other factors to create a broader and more effective model for developmental policy.

Appendix A: Data on Nepal

From Gautam, 2004:

From Acharya, 2002:

Demographic Data Obtained from the UN:

Demographic Data Obtained from the FAO:

Note: Data for total fertility rates and population growth was not available on a logarithmic scale, but my calculations show that population growth was between 2 to 3 percent a year.

Data on Per Capita GDP Growth from ESDS (Actual data from World Bank):

Appendix B: Data on Ethiopia

Demographic Data Obtained from the FAO:

Data on Per Capita GDP Growth from ESDS (Actual data originates from the World Bank):

Sources

Acharya, K.P. (2002). Private, Collective, and Centralized System of Institutional Arrangements in Community Forestry in Nepal. Presented at “The Commons in an Age of Globalisation.”

Acharya, K.P. (2002). Twenty-four Years of Community Forestry in Nepal. International Forestry Review. Volume 4 (2).

Angelsen, A. (1999). Agricultural expansion and deforestation: modeling the impacts of population market forces and property rights. Journal of Development Economics Vol 58, pp 185-218.

Baland, J.M., and Platteau (1996). Halting Degradation of Natural Resources. FAO and Oxford University Press. http://www.fao.org/docrep/x5316e/x5316e00.htm

Barbier (2000). Economic linkages between rural poverty and land degradation. Ecosystems and Environment, Volume 82 (1-3).

Barbier, E (2006). Natural Resources and Economic Development. Cambridge University Press.

Bluffstone, R.A. (1995). The Effect of Labour Market Performance on Deforestation in Developing Countries under Open Access: An Example from Rural Nepal. Journal of Environmental Economics and Management, Vol. 29, 42-63.

Carter, M., Little, D., Mogues, T. and W. Negatu (2007). Poverty Traps and Natural Disasters in Ethiopia and Honduras. World Development, Volume 35 (5).

Carter, M., Little, D., Mogues, T. and W. Negatu (2005). The Long-Term Impacts of Short-Term Shocks: Poverty Traps and Environmental Disasters in Ethiopia and Honduras. BASIS Brief 28, May 2005, Madison: University of Wisconsin.

Dasgupta P (1993). An Inquiry into Well-Being and Destitution. Oxford Clarendon Press.

Dasgupta P (2000). Population Resources and Welfare: An exploration into Reproductive and Environmental Externalities. http://www.econ.cam.ac.uk/faculty/dasgupta/hbkpop.pdf

Dasgupta P (1998). The Population Problem: Theory and Evidence. Journal of Economic Literature.

Dasgupta P (2003). World Poverty: Causes and Pathways.

Dasgupta S, Deichman U, Meisner C and Wheeler D (2005). Where is the Poverty-Environment Nexus? Evidence from Cambodia, Laos PDR and Vietnam. World Development, Volume 33 (4).

De Janvry, A., and Sadoulet (2005). Progress in the Modeling of Rural Households behaviour under Market Failures. in Janvry:. Poverty, Inequality and Development. Kluwer publishing, Chapter 8. http://are.berkeley.edu/~sadoulet/papers/HH-Thorbecke.pdf

Gautam, AP., Shivakoti, and Webb. (2004). A Review of Forest Policies, Institutions, and Changes in Resource Condition in Nepal. International Forestry Review. Volume 6 (2).

Newcombe, K. (1987). An Economic Justification for Rural Afforestation: The Case of Ethiopia. The Annals of Regional Science, Volume 21 (3).

[tags]poverty environment population nexus, pep nexus, pep triangle, causation, environmental degradation, population growth, demographic transition, models, economic models, Dasgupta, barbier, demography, causation of poverty, creation of poverty, relation of poverty to environment and over-population, nepal, ethiopia, agricultural output, feed back loop, Population-Poverty-Environment Linkages, third world, developing economics, policy implication[/tags]