H. Steinfeld and S. Mack
H. Steinfeld is Senior Officer (Programme, Policy and Planning) and S. Mack is Animal Production Officer (Rural Development) in the Animal Production and Health Division, FAO, Rome, Italy
A planning perspective
An approach to livestock-sector planning
Livestock make a major, although largely underestimated, contribution to rural development in developing countries. They produce food, enhance crop production and provide additional economic goods and services as well as cash income. The inclusion of livestock diversifies and increases total farm production and income, provides year-round employment and disperses risk. Sales of livestock products provide funds for purchasing crop inputs and for financing farm investments. Livestock often form the major capital reserve of farming households and, in general, enhance the economic viability and sustainability of a farming system.
Despite these positive contributions to agriculture and economic development, many formal livestock projects have failed to meet their objectives, with the result that donors are becoming more and more reluctant to support such developments. Furthermore, animal production is increasingly being viewed far more critically:
· intensive production systems, particularly in industrialized countries, are seen as a major source of pollution;
· increasing ruminant numbers in developing countries are being associated with the degradation of the rangelands and soil erosion;
· livestock development is said to favour the richer segments of society - both producers and consumers - rather than the most vulnerable;
· livestock are thought to compete directly with humans for cereal grains.
Such controversies and the inherent complexity of livestock production impose constraints that have to be addressed and which pose particular challenges not normally faced by the agricultural planner. Yet it is the very complexity of animal production systems that also offers some of the greatest opportunities for development. Livestock, because of their linkages with the overall farming system, make valuable entry points for wider agricultural development programmes. To exploit these opportunities, an integrated approach that combines both technical and institutional interventions is required.
In many countries, the difficulties associated with increasing sustainable animal production are exacerbated by limited public-sector investment and weak, ineffective support services. Programmes and projects are often poorly designed and inadequately targeted, leading to the inefficient and fragmented allocation of scarce development resources. Policies related to the livestock sector are often incoherent with ill-defined goals and with little or no assessment of their likely impact. The lack of consistent, integrated strategies that focus limited resources on identified and attainable goals remains a major constraint to livestock development. The situation is further complicated in that livestock, especially cattle, represent wealth and status and, as a consequence, are disproportionately owned by policy-makers who have a clear vested interest - an interest that is not necessarily beneficial to livestock development in general.
There is obviously a need for an effective policy and planning framework that will optimize development resources and provide the necessary support and economic environment to allow a country's livestock resources to express its potential. This is reflected in the number of countries that have embarked on preparing strategic plans to develop their livestock subsectors, either with their own resources or, increasingly, with the assistance of agencies such as FAO.
For the agricultural planner, difficulties are related not only to the complexity of livestock production systems but also to an inability to understand how these systems function - this is primarily a problem of quantification and comprehension. One consequence is that development opportunities are often ignored, particularly the potential for using livestock as a catalyst to drive agricultural development. An understanding of the production factors and processes that affect animal production is a prerequisite for livestock development.
Livestock. Animals themselves are the major resource, but their mobility makes them a difficult resource to quantify, especially under the extensive management systems that predominate in much of the developing world. While this may create problems for the government statistician and the tax collector, it is unique in that it allows for the exploitation of feed resources for which there are no alternative productive uses, such as extensive arid, semi-arid and cold savannahs, crop residues and agricultural fallow.
Capital. Livestock ownership is more skewed than ownership of or access to land, and, as a consequence, livestock development, especially if it concerns larger and more expensive species such as cattle, tends to produce benefits of low equity. Many non-livestock owners are often precluded from livestock development because of lack of capital or credit. Yet, in many farming systems, livestock constitute the main, if not the only, capital reserve of farming households and, importantly, one that can be readily realized. As such, livestock serve as a strategic reserve that adds stability to the overall fanning system. In this respect, a mix of species increases stability; for example, cattle represent a long-term investment whereas sheep, goats and poultry are primarily shorter-term investments and sources of petty cash.
Feed. Whereas animals can substitute for capital, purchased feeds can substitute for land, creating "landless" production systems, where land is no longer a production factor. These are not primary production systems in the strictest sense, but more industrial and almost exclusively demand-driven. Conversely, extensive pastoral systems depend exclusively on forage, the availability of which is both seasonal and highly dependent on natural factors, most notably rainfall. Such systems are essentially resource-driven and less responsive to price changes.
Many animal feeds also have alternative uses, either for human consumption or for industrial use. These competing demands are determined by price and availability of the commodity, its use and the value of the end-product. In this respect, the more intensive, demand-driven production systems are an important alternative use of these commodities, and, as such, feed-conversion efficiency becomes an important productivity indicator and management objective.
Land. Escalating ruminant (large and small) populations, agricultural encroachment and decline in traditional authority have put an increasing strain on "open access" feed resources - particularly extensive grazing areas leading, in extreme cases, to irreversible degradation. The conflict between communal ownership of land and private ownership of livestock - the classic "tragedy of the commons" - has resulted in a disequilibrium that continues to threaten the ecological stability of many of these fragile environments. However, an increasing body of evidence suggests that these rangeland ecosystems have adapted and are more resistant to heavy stocking than previously thought, and that opportunistic range management often is an efficient and ecologically sound resource use (Behnke, Scoones and Kerven, 1993). For the livestock planner, these are particularly complex and sensitive issues that must be addressed. Technical options are limited, and certainly solutions cannot be found without cognizance of the wider institutional and social contexts.
Access to water, especially in the extensive rangeland systems, is a fundamental prerequisite for livestock production. Historically, water has governed both access to and Use of these resources and has provided a brake on exploitation. Major environmental issues have recently arisen over the provision of new water sources, however, particularly perennial boreholes.
Labour. Labour is the other important resource for which livestock production has a specific requirement. Typically, livestock production is more labour-intensive and less seasonal than crop production, and can utilize family labour that usually has a low opportunity cost. The distribution of labour, responsibilities and benefits tends to be favourable for women, especially with the smaller animal species, which they themselves may own. Since women are largely responsible for the day-to-day management of the household, any development programme must take their time availability into account.
The production process
Livestock production tends to be more complex than crop production. Production cycles, although seasonally influenced, are less pronounced with livestock. Some species have very short reproductive cycles (rabbits, poultry), while those of others (large ruminants) are far longer. In more intensive production systems, seasonal influences can be offset by modifying the environment through improved nutrition and artificially controlled light and temperature.
Unlike intensive production systems, which produce a single product, many of the production systems common to the developing world produce a range of commodities. These may include a mixture of consumables and services that provide cash, subsistence and inputs (draught power and manure) into the farm enterprise. Since livestock so often have a pivotal role in the overall farm system, any constraint imposed on livestock may also restrict the system as a whole.
Animals also have an important asset function. Many smallholders, particularly in mixed-farming systems, prefer the flow products (milk, draught, manure) rather than the end-products (meat, hides and skins) since selling their animals for slaughter entails the permanent loss of flow products. Only in larger herds or flocks can meat offtake be regarded as a flow product. Flow products generate a regular cash income, unlike end-products or crop revenues, and the importance of regular cash flow, however small, is often underestimated in many development efforts, especially within the smallholder sector.
Given the perishable nature of animal products, development beyond household consumption requires marketing and processing facilities that may not be readily available. Milk, for example, not only needs a continuous outlet, but also transport, storage and processing facilities. This interdependency between animal production and the links in the marketing chain increases as systems intensify.
The close relationship between domesticated animals and humans evident throughout history continues, and many societies have strong socio-cultural values attached to their animals. In most cases, these values reflect specific economic attributes and have important implications for livestock development.
Risk is another important factor in livestock production. Extensive livestock production systems are exposed to particularly high risks as a result of natural factors such as drought and disease, whereas in mixed-farming systems livestock help to reduce the overall risks to the enterprise. As production systems intensify, the production risks from natural causes decrease and are replaced by economic risks, such as price fluctuations and taxes. Producers can mitigate risk through diversification (mixed farming and mixed species), flexibility (choice of stocking rate) and increased productivity. The design of livestock development strategies must take these factors into account through risk analysis (Savvides, 1994).
On the consumption side, livestock products have characteristics that require specie-1 attention in the planning process. Meat, milk and eggs are all perishable products that need refrigeration and careful handling if they are to remain fit for human consumption. Such processes impose certain infrastructural requirements that may be beyond the resources of a given country or even a region. This is a major contributing factor to the small amount of milk products marketed in sub-Saharan Africa, for example.
Typically, livestock products also have a high elasticity of demand but a low elasticity of supply, particularly in land-based smallholder production. In many countries the demand for animal products increases rapidly with expanding urbanization (and changes in consumption patterns), growth in per caput income and a growing population. Because of this demand pattern it has been argued that livestock development tends to favour the higher-income sectors of society - an isolated view, yet one that has deterred potential donors - but does not adequately take account of benefits on the supply side.
All edible livestock products have high nutritive value. They are especially rich in protein, with a favourable composition of amino acids. Milk availability is one of the major contributors to the alleviation of malnutrition among the more vulnerable sections of society, most notably children. The excessive consumption of animal products, which is now recognized as a serious health hazard, is primarily a concern in affluent societies, while in developing countries consumption levels remain low. Indeed, animal fat provides a valuable contribution to the daily energy requirements in these countries.
The FAO Animal Production and Health Division (AGA) is developing a methodology to provide a rational basis for livestock-sector planning that will assist governments in determining their own policies and priorities. The principles described apply equally to a particular segment, species or geographical region of the subsector-milk production, animal health or the subhumid zone, for example - and to the subsector as a whole. The chosen approach can be described as interdisciplinary, as it combines the biological, social and economic sciences, dynamic, in that it uses scenarios to anticipate future development, and systems-oriented.
In general, most livestock development strategies have similar aims, namely, to:
· conserve the natural resource base;
· raise productivity through better utilization of available resources: capital (animals), land and labour;
· expand production where there is a sufficient demand and resources can be utilized at reasonable cost to the environment;
· optimize the allocation of development resources through rational administration and management.
A strategy for livestock development has a number of technical and socio-economic dimensions that need to be addressed if it is to have any relevance or acceptance. To make the planning process simple and transparent, the approach adopted by AGA follows the logical sequence of:
· Diagnosis, An analysis of the current performance of the livestock subsector, its potentials and constraints.
· Prescription. A creative phase in which programmes and policy options are formulated and designed to address identified constraints and potentials.
· Prognosis. A comparative (with and without) assessment of the likely impact and implications of the prescribed programmes and policy options.
The analytical process usually proceeds from the bottom up, that is, from the lowest production unit to its effects on the national economy, and contains an ex ante evaluation of ongoing as well as previous programmes and policy interventions. The process is interactive in that the results of any comparative assessment may require modifications to either the design or further analysis. The final result is the presentation of a series of development options along with a corresponding set of recommended actions and policies required if the options are to be implemented. Implicit in the success of any such exercise is a high degree of national "ownership" resulting from the maximum involvement of nationals in the complete sequence through such elements as rapid rural appraisals (RRAs), national consultancies and workshops and expert working groups.
With regard to livestock policy analysis and formulation, the concept of induced innovation (Hayami and Ruthan, 1985) is applied. According to this concept, the choice of technology is induced by a relative abundance of production factors, such as those introduced above, for example, in the case of livestock production. For a number of the effects that livestock production can have, such as those on the environment and on human health, no markets exist. These effects are referred to as market failures, or externalities, which can be both positive and negative. Where these externalities exist, such as in the case of waste disposal from intensive pork and poultry production, incentives or regulations are part of policies that aim to reflect more accurately the implicit value of a non-marketed commodity, such as human health or the environment. Incentives and regulations affect factor use as they infringe on factor prices and, thus, on technological choice.
The diagnosis phase
Livestock need to be placed within the overall context of the agricultural sector and the national economy as a whole. The first step is usually an inventory of the natural and economic resources available for livestock production: land availability and use (by agro-ecological zone); feed (crop residues, agro-industrial by-products, cereals, forages, dodders); animals (breeds, etc.); water; labour; capital; public infrastructure; and support services (marketing, input supply). Existing production methods and productivity, processing, storage and marketing all need to be assessed in relation to the available resources.
In many developing countries, traditional and commercial production systems exist alongside one another with little or no linkage between them. Industrial systems of poultry, pig and, to a lesser extent, dairy production are demand-driven and responsive to changing market conditions, due, in part, to their short productive cycles and a wide and varying range (in price and availability) of potential feeds. Conversely, traditional ruminant production systems are inelastic in supply and sometimes even respond inversely to price increases, as was evident in the negative beef supply response observed in Latin America (Jarvis, 1986) and in southern Africa (Doran, Low and Kemp, 1979).
A number of macroeconomic policies also have direct and indirect impacts on livestock production. Pricing policies - often through the manipulation of exchange rates - wages and interest rates determine the costs and returns of livestock production. Fiscal policies govern government expenditures on, for example, rural infrastructures and producer or consumer subsidies. Monetary policy determines the availability and cost of agricultural credit, while trade policies (tariffs, quotas and export subsidies) regulate the import (produce and raw material) and export of livestock products and animal feeds. Institutional structures have a bearing on a wide range of support services that affect livestock production, such as marketing (marketing boards), quality and hygiene standards, extension and research, and agricultural education. The balance between private and public ownership of livestock-related services is becoming a critical issue, especially as extended public services prove to be increasingly inefficient and costly to run. These macroeconomic and sectoral policies need to be analysed as to their relevance to livestock production; incentives and/or disincentives to livestock production as they affect relative abundancy of production factors; and adequacy in view of likely future developments.
In addition to these broad economic policies, countries usually pursue more specific sectoral policies that may or may not be interrelated with the overall policy framework. These policies usually refer to issues such as: land tenure; input supplies and pricing (feed industry, veterinary supplies, genetic resources); labour supply and wages; specific livestock credits; marketing of livestock products; product processing facilities; and capacity and performance of extension, training and research institutions.
On a microlevel, policies may be analysed using econometric methods for farm household data, integrated farm household models, mathematical programming or a policy analysis matrix (FAO, 1994).
Livestock and the environment: finding a balance
In cooperation with eight different donors, FAO is leading a multidonor study entitled "Interactions between livestock production systems and the environment global perspectives and prospects". It addresses, as a follow-up to the United Nations Conference on Environment and Development (UNCED), the issues of livestock-environment interaction. The objective of the study is to assess the major positive and negative interactions between livestock production systems and natural resources. The full report will be ready by the end of 1995.
The study's perspective is global, covering a range of representative livestock production systems in both developing and developed countries. Eleven production systems have been identified using a classification of agro-ecological zones and land use, which were then grouped into mixed farming systems, pastoral or extensive grazing systems and systems of landless intensive livestock production
For each of these system groups, livestock-environment interactions are rated for a number of interaction domains, including range utilization, wildlife diversity, crop-livestock interactions, waste disposal from - production (manure) and from processing (slaughterhouses, tanneries, dairies), methane emissions and domestic animal diversity. In addition, more indirect interactions, such as those related to concentrate feed demand and livestock-induced deforestation, are also appraised.
A number of the case-studies presented were developed to illustrate the livestock-environment interactions and impacts that have been observed. These case-studies will also explore approaches that can be implemented either to reduce environmental degradation or to enhance the process of building more stable production systems.
It can be concluded that product) on systems vary considerably in their impact on the environment. Most beneficial effects are found in mixed farming systems, whereas landless systems are facing a range of environmental problems, mostly related to waste and resource use. Pastoral systems are mostly environmentally balanced, but this equilibrium is endangered as demographic pressure persists. Within production systems, production intensity is a major factor determining the nature of livestock-environment interactions.
Livestock production, systems. Within identified farming systems, farm households can be defined according to their available (or lack of) resources, production patterns, productivity and income. Such analysis is usually derived from diagnostic surveys such as RRAs and farm management budgeting (FMB). The breakdown of a livestock subsector into various production systems is justified given the differences that exist in available resources and applied technologies, as well as in primary products and their uses (subsistence or market-oriented), making it easier to deal with typologies. It is at this level where most development efforts are concentrated, and in the majority of countries this usually entails a breakdown of the livestock subsector into production systems broadly based on agro-ecological zones and socio-economic patterns. Livestock production systems are basically a subset of farming systems (Ruthenberg, 1980), although their analysis must comprise the whole farm household.
Available feedstuffs may be classified by their physical properties (dry matter, texture, fibre content), chemical properties (protein, energy) or origin (on-farm, purchased, communal). Inadequate feed intake and/or unbalanced diets are usually the primary production constraint. In addition, feed is also a major cost component, especially in the more intensive production systems. Available feed resources need to be carefully assessed, therefore, with estimated feed requirements in mind.
Livestock resources consist of various species; their breeds and genetic potential, herd composition and ownership are the predominant determinants of a production system. These factors change as production systems become more an integral part of the economy. With increasing commercialization, productivity and demand, there is a move from predominantly resource-driven systems to those primarily demand-driven.
Production systems may also be classified according to their main product(s) - meat, milk, eggs, manure and wool/fibre - as well as by services such as draught, transport and breeding. Systems may be further differentiated based on a scale ranging from "extensive" to "intensive", determined largely by their input-output relationships.
Livestock and their products fulfill different and often complementary functions. Apart from the marketed outputs, these include inputs to crop production (draught power and manure) and home consumption (subsistence), as well as their asset function and socio-cultural importance. At the farm level, therefore, livestock must be viewed in the context of the overall farm-household system, and the complex interactions involved must be understood technically as well as economically and socially. Where livestock are an integral part of the farm household economy, flow products (draught, manure, milk), risk alleviation, capital accumulation and the utilization of family labour with low opportunity costs are frequently the dominant production objectives. This is often in direct contrast to state objectives, which favour end-products (meat, hides, skins), and has been one reason why so many government livestock development initiatives have failed.
Strategy and programme formulation
A development strategy needs to anticipate and identify those forces that drive a particular livestock production system and the subsector as a whole. These can include population growth, market development (urbanization and income growth), technological change and a changing resource base. An understanding of the direction and speed with which these factors may change, both at the sectoral and farm levels, is necessary so that future programmes can respond accordingly. As has been stressed, technological choice depends on the relative abundance of production factors, such as labour, land and capital.
Implicit in this process is the development of "what if" scenarios to anticipate the likely implications of any change. For demand-driven production systems, the variables usually include national and per caput incomes, population and population growth, product prices and price and income elasticities. For resource-driven production systems, trend projections that take account of anticipated changes in factor proportions, advances in technology and likely increases (decreases) in productivity and resource availability are required. Any gap between supply and demand will have to be met either by expanding commercial production or by imports. The construction of scenarios is very often assisted by the use of computerized models (Hallam, 1983); FAO has recently produced an updated and expanded version of this sectoral model.
The first step is to develop the "without" intervention scenario by projecting future production and consumption trends. This will indicate gaps in supply, overuse of natural resources and import requirements, among other things. The next step is the formulation of development pathways (a combination of technical and institutional interventions) that address identified constraints, opportunities and any comparative advantages. Interventions then need to be evaluated under "with" and "without" scenarios, which involves sensitivity analysis of the critical variables as well as an assessment of social and cultural acceptability. Historically this has primarily been a top-down approach that has failed to take account of the aspirations and production objectives of producers, resulting in the poor performance of so many livestock development programmes. It is important, therefore, to ensure the active participation of all interested parties (usually easier said than done!) in the formulation process.
For animal production, possible programmes and policy interventions include the following broad classifications:
· On-farm interventions, adapted to specific agro-ecological conditions and production systems. These interventions commonly aim at increasing the availability or utilization of local feeds, control of economically important diseases (internal and external parasites) and/or improved housing and management.
· Institutional changes, including the structure and function of support services covering input supply, research, extension and training, processing and marketing and credit. Institutional programmes often complement technical interventions and aim at providing a support framework for livestock production that should be both cost-effective and congruent with overall government policies. The concept of "private" and "public" good will increasingly determine who will pay for such services.
· Genetic improvement programmes aimed at improving the livestock resource base. Options include within-breed selection of adapted indigenous breeds, substitution with exotic breeds or cross-breeding. The choice largely depends on the production system, its objectives and the resources at its disposal. Experience has shown serious misjudgement with policies aimed at importing exotic breeds with a corresponding neglect of indigenous breeds in many developing countries. Whatever breeding programme is adopted, equal attention needs to be given to the dissemination of improved genetic material. Dissemination concerns the institutional aspects as well as the choice of biotechnology, such as artificial insemination and embryo transfer.
· Animal health programmes aimed at limiting the impact of disease on animal production. Policy issues concern who will provide and pay for such services. Foremost is the need to control and protect (quarantine) the national livestock resource from major epizootic diseases, such as rinderpest, which is clearly a "public" good. Disease monitoring, veterinary investigation and legislation (public health and meat inspection) also fall within the public domain. On the other hand, control of diseases that cause production losses, for example, helminthiasis, is primarily a "private" good and determined on grounds of cost effectiveness. In this case, institutional policies that encourage the provision of private clinical services are required.
· Processing and marketing policies related to investment in the necessary infrastructure that enables livestock products to safely supply existing demands as well as those of the future. Such specific issues as the design of structures, equipment, training and quality, along with the question of the degree of state intervention (marketing boards, etc.) in the market, must also be addressed.
The predictive phase of the strategy design and planning process examines the various policy options and their implications in light of the government's broader development objectives. Strategy impact assessments usually address issues such as:
· economic efficiency;
· distribution and equity considerations (of both costs and benefits);
· stability (food supply, income, export earnings, etc.) and risk considerations;
· sustainability (environmental, financial and institutional);
· conformity with government objectives.
This is primarily a "bottom-up" procedure looking at the performance of the individual animal through to the production systems and finally the subsector as a whole. At the production system level, interventions can be assessed in terms of their economic, social and environmental impacts on households. It is at this level where interventions that are usually aimed at increasing productivity (mortality, fertility, carcass weights, growth rates, milk yield, etc.) have to be shown to be both technically and financially viable. The results of these production system assessments may well necessitate changes in the off-farm elements of the proposed programme, such as marketing, processing, finance, extension and research.
Programme and policy implementation will also have implications at the sectoral and country levels. And some of these implications will have an impact on programme implementation as well, where they relate to the capacity of the ancillary services and industries to support and service the proposed development. These can be manipulated through price, monetary and trade policies, institutional reform and land regulations, which are mostly under direct government control. Those under private control can only be encouraged through supportive action. The impact of a successful outcome of the proposed programmes also needs to be assessed. Increased production will have an impact on income, consumption and labour (availability and productivity), as well as consequences for the rest of the agricultural system. In its involvement in this assessment process, FAO can draw on its international experience to help avoid mistakes and to benefit from successes elsewhere.
As a result of the impact assessment, a set of development options can be presented to the government for consideration. Each option should be accompanied by a summary description, along with its likely impact and implications. It is the government that will make the final choices as to whether or not on certain strategies should be implemented and whether or not external assistance is necessary; ideally, the major donors would have participated in the review process.
Livestock production has grown faster than agricultural production in most developing countries, and this trend is likely to continue with growth rates over the next 20 years estimated at 4.5 percent per annum. Historically, growth has come primarily from the expansion of livestock numbers rather than an increase in productivity. If this trend continues, it will put tremendous pressure on the available feed resources - even assuming substantial progress in feed conversion efficiency - and this probably will be the major challenge facing livestock planners.
The development of livestock in many developing countries is constrained by minimal public-sector investment and inefficient and poorly coordinated support services, however. This situation can, in part, be attributed to a lack of any consistent strategy for livestock development, which is exacerbated by inadequate analytical tools and a lack of information on which to base decision-making.
Clearly, increased livestock production will depend ultimately on the adoption of appropriate technology, improved support services, market access and infrastructural development to stimulate increased productivity. However, there must be a framework of coherent policies and development strategies that facilitate such development and ensure that the full potential of livestock in developing countries is exploited. FAO, with its unique collection of expertise and development experience, is ideally placed to assist member countries in reviewing and exploring the options for developing their livestock resources.
Behnke, R.H., Scoones, I. & Kerven, C., eds. 1993. Range ecology at disequilibrium, new models of natural variability and pastoral adaptation in African savannas. London, UK, Overseas Development Institute.
Doran, M.W., Low, A.R.C. & Kemp, R.L. 1979. Cattle as a store of wealth in Swaziland: implications for livestock development in eastern and southern Africa. Am. J. Agric. Econ., 61: 41-47.
FAO. 1994. Methods of microlevel analysis for agricultural programmes and policies: a guideline for policy analysts. Rome, FAO.
Hallam, D. 1983. Livestock development planning: a quantitative framework. CAS Paper 12, May 1983. Reading, UK, Centre for Agricultural Strategy.
Hayami, Y. & Ruthan, V.W. 1985. Agricultural development: an international perspective. Revised ed. Baltimore, MD, USA, Johns Hopkins University Press.
Jarvis, L.S. 1986. Livestock development in Latin America. Washington, DC, USA, World Bank.
Ruthenberg, H. 1980. Farming systems in the tropics. 3rd ed. Oxford, UK, Clarendon Press. 424 pp.
Savvides, S. 1994. Risk analysis in investment appraisal: project appraisal, 9(1): 3-18. Surrey, UK, Beech Tree Publishing.