EFFECTS OF GOVERNMENT PROGRAMMES ON SUSTAINABLE AGRICULTURE INTHE PEACE RIVER REGION OF BRITISH COLUMBIA: A LINEARPROGRAMMING ANALYSISbyMAJID KWABENA ADDOB.A., Dalhousie University, 1985M.D.E., Dalhousie University, 1988A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDepartment of Agricultural EconomicsWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAFebruary 1992© Majid Kwabena AddoIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of Agricultural EconomicsThe University of British ColumbiaVancouver, CanadaDate February 1992DE-6 (2/88)iiABSTRACTLand is an essential resource for most types of agriculturalproduction. Its continued productivity forms a significant part ofthe deliberations about sustainable agriculture. While discussingsustainable agriculture, this thesis focuses on governmentagricultural programmes that have influenced agricultural land usein the Peace River region of British Columbia. The general aim isto point out the relevant programmes that impede sustainability ofagriculture.We assume that farmers are continually making decisions aboutthe optimal allocation of land so as to maximise present value ofnet farm incomes. A linear programme (LP) is one of the techniquesof mathematical programming that can be used to maximise farmincomes. It is this technique that we employed to analyze effectsof government agricultural programmes on land use as it pertains tocrop and forage production in the Peace River region, where foragesare assumed to be a derived demand for livestock production.Cultivation practices of summerfallow and continuous cropping areexamined.Parametric linear programming (PLP) is subsequently used toanalyze other optimal land use scenarios by varying the LP'sobjective function coefficients. As well, other cases includingthe elimination or halving of government subsidies are alsosimulated and discussed. Furthermore, an attempt was made toiisimulate two other scenarios that deal with the removal of grainsand summerfallow from lower quality land. The region's soilerosion problem was also simulated.It was found, among other things, that four governmentagricultural programmes--Western Grain Stabilization Act, SpecialCanadian Grain Programme, Crop Insurance and Chemical Rebates--encouraged cultivation of marginal lands, which are moresusceptible to erosion. For example, some 26% (comprising wheatand summerfallow) of the total farm acreage occurred on classes 4and 5 land. Consequently, these programmes, as presentlyconstituted, adversely impact on Peace River region's sustainableagriculture.Cultivation of grains on only good quality lands resulted ina significant reduction of summerfallow and more intensivecropping, which will lead to less soil degradation in the region.This alternative programme also was observed to increase a farmer'sincome by about 4%, which can add a total of over $259,000 to theregion's economy. As well, it can increase pasture to feed morethan 15,600 beef cows, which will be a boon to the livestockindustry in the region.ivTABLE OF CONTENTS PageABSTRACT ^LIST OF TABLES ^ viLIST OF FIGURES viiiLIST OF ACRONYMS, ABBREVIATIONS AND EQUIVALENTS ^ ixACKNOWLEDGEMENTS ^ xiDEDICATION ^xiiCHAPTER1 INTRODUCTION ^1.1^Background 1.2^Problem Statement ^1.3^Outline of Thesis 11452 LITERATURE REVIEW 62.1 Background to Sustainable Agriculture ^ 62.2 Concepts and Definitions ^ 92.3 Neoclassical Economics and the Environment^.^.^. 142.4 Towards Sustainable Agricultural Development^. ^ 173 DESCRIPTION OF STUDY AREA AND PERTINENTGOVERNMENT PROGRAMMES ^ 233.1 Geography and Climate 233.2 Soil and Vegetation 253.3 Agriculture and Land Use ^ 263.4 Government Programmes Affecting PeaceRiver Agriculture ^ 353.5 Review of Pertinent Government Programmes ^ 374 MODEL OF INSTITUTIONS AND LAND USE ^ 434.1 Introduction to Linear Programming ^ 434.2 Applications of Linear Programming to PeaceRiver Agriculture ^ 454.3 Structure of Empirical Model ^ 514.4 The Basic Empirical Model 535 EMPIRICAL RESULTS ^ 655.1 Discussion of Base Case Results ^ 655.2 Results of Simulations Involving Regulations 695.3 Results of Simulations Involving ContinuousCropping ^ 755.4 Duality and Sensitivity Analyses ^ 796^SUMMARY AND CONCLUSIONS ^6.1^Methodology ^6.2^Summary of Results 6.3^Conclusions 82838486BIBLIOGRAPHY ^ 88APPENDICES ^ 94APPENDIX A ACTS ADMINISTERED BY THE FEDERALMINISTER OF AGRICULTURE (BY PROGRAMME) . ^ 94APPENDIX B ACTS ADMINISTERED BY THE BRITISH COLUMBIAMINISTRY OF AGRICULTURE, FISHERIES,AND FOOD (BY PROGRAMME) ^ 96APPENDIX C B.C. LAND INVENTORY SOIL CAPABILITYCLASS DESCRIPTION FOR AGRICULTURE ^ 97APPENDIX D RESULTS OF DUALITY AND SENSITIVITYANALYSES, BY MODEL ^ 98viLIST OF TABLESPage3.1 Average Frost-Free Period in Selected Towns ofthe Peace River Area^ 243.2^Total Precipitation 1951-80 in SelectedTowns of the Peace River Area (mm)^ 253.3^Peace River Land Use: Census Years (Acres)^283.4^Peace River Principal Crop Acreage: 1979-90 (1000)^283.5 Production of Principal Crops in the Peace RiverRegion: 1979 - 90 ('000 Tons)^ 313.6 Peace River Livestock Production: Census Years^313.7 Area in CLI Agricultural LandClass and ALR, 1988 ('000 Acres)^ 333.8^Peace River Land Use by Land Class ('000 Acres)^343.9 Net Direct Payments to B.C. ProducersBy Programme: 1971-89 ($ '000)^ 404.1 Summary Results of the Graham and Anderson 1982Study^ 504.2 Calibrated Results of Land Class Gross Marginof the Model's Instruments ($)^ 605.1 Effects of Government Subsidies on AgriculturalLand Use in the Peace River Region 675.2 Effects of Government Subsidies on Land Class Usein the Peace River Region (As % of TFA)^ 685.3 Effects of Regulations on Land Use in thePeace River Region^ 705.4 Effects of Regulations on Land Class Usein the Peace River Region^ 725.5 Effects of Government Subsidies on Peace RiverAgricultural Land Use Under Continuous Cropping^75vii5.6 Effects of Government Subsidies on Land ClassUse in the Peace River Region Under ContinuousCropping (As % of TFA)^ 765.7 Net Income of Various Simulations^ 796.1 Summary of Alternative Programmes' Impacton Base Case^ 86LIST OF FIGURESviiiPage3.1 Peace River Principal Crop Acreage,1979 - 90^('000) 305.1 Base Case Land Use 665.2 Effects of Government Subsidies onPeace River Land Use 745.3 Effects of Government Subsidies on ContinuousCropping Land Use 78LIST OF ACRONYMS, ABBREVIATIONS AND EQUIVALENTS ALR^Agricultural Land ReserveBCMAFF^B.C. Ministry of Agriculture, Fisheries and FoodCC^Continuous CroppingCCNG^Continuous Cropping with No Government SubsidyCCSG^Continuous Cropping with Some Government SubsidyCLI^Canada Land InventoryCWB^Canadian Wheat BoardFAO^United Nations Food and Agricultural OrganizationFGMDP^Food Grain Market Development ProgrammeGNP^Gross National ProductGRIP^Gross Revenue Insurance ProgramGSP^Gross Sustainable ProductivityIFOAM^International Federation of Organic AgricultureMovementISEW^Index of Sustainable Economic WelfareLP^Linear ProgrammingNG^No Government SubsidyNOG^No Grains on Land Classes 4 - 5NOSF^No Summerfallow on Land Classes 4 - 5NNW^Net National WelfarePLP^Parametric Linear ProgrammingPRSCA^Peace River Soil Conservation AssociationSCGP^Special Canadian Grains ProgrammeSG^Some Government SubsidySOE^State of the EnvironmentixUN^United NationsUNEP^United Nations Environment ProgrammeWCED^World Commission on Environment and DevelopmentWGSA^Western Grain Stabilization Act1 ACRE = 0.405 HECTARE1 IMPERIAL TON = 0.907 METRIC TONNES1 TON/ACRE = 2.242 TONNES/HECTARE1 POUND = 0.454 KILOGRAM1 FOOT = 0.305 METRESxiACKNOWLEDGEMENTS In the name of the Almighty whose continued Grace and Mercyhave guided and protected me from ailments as I pursue myinsatiable quest for knowledge.My sincere thanks to Dr. G.C. van Kooten, who not onlyrendered invaluable supervision of this thesis but also furnishedme with research assistance that profoundly contributed to myunderstanding of sustainable agriculture.I also am indebted to Dr. George Kennedy for his insightfulcomments and suggestions, and for providing me with an enjoyableexperience as his teaching assistant.The third member of my thesis committee, Dr. Art Bomke,meticulously supported this research with intimation and guidancefor which I owe a debt of gratitude.Special thanks to my sweetheart, Princess Ann, for her warmth,affection and support. Indeed, I am forever indebted to my wholefamily and friends, spanning three continents, but especiallyJanet, Zai, Haj, N'ash, Abi, Panin and Kakra and my beloved Auntiefor their prayers, love and financial support that have made mytravels and schooling possible.I am also grateful to the Department for offering me ateaching assistantship, without which I may not have been able tostudy at this renowned university; to Gwynne Sykes for herassistance at all times including finding some pertinent materialsfor me.None of these people share any responsibility for whateverimperfections or peccadilloes that might still bedevil this thesis.DEDICATIONxiiTo my beloved maCHAPTER 1INTRODUCTION1.1 Background Agriculture plays a major role in British Columbia's economy;in 1988, for example, total farm cash receipts exceeded one billiondollars. A notable feature of B.C.Is agriculture is its diversity--over 80 different commodities are produced, including grains in thePeace River Country (Quick Facts About British Columbia 1989,p.20). About half of all agricultural products consumed by BritishColumbians is locally produced.The Peace River region's agriculture plays a significant rolein B.C.'s economy, a fact that influenced our decision to selectthe Peace River as the area for this study. Grains and oilseeds,mostly produced in the Peace River area, contributed $25.2 millionand $24.9 million to B.C's farm cash receipts of $1,063.3 millionand $1,178.8 million in 1987 and 1988, respectively (B.C. Ministryof Finance and Corporate Relations 1989, p.72).In British Columbia, and thus in the Peace River region, farmsare generally classified as either part-time or commercial, withthe benchmark separating them set at gross sales of $25,000(Agriculture Canada 1988, p.8). In 1981, part-time farmscontributed only one percent to the average $30,000 family income(ibid.). Commercial farms (27 percent of B.C.'s farms) contributesome 74 percent to the province's farm cash receipts. Overall, thenumber of farms in the province is declining as monoculture and12mechanization engender relatively larger farms, some of which grossover $250,000 annually (ibid., p.9).In this century, agriculture has generated surpluses thatfacilitated the growth and expansion of the economies of moredeveloped countries (MDCs) such as Canada. In recent decades,however, agriculture has become largely characterized bymonoculture (i.e., specialization) and intensive energy use.Moreover, as Berry (1978) succinctly points out, "modern"agriculture has narrowly linked farm productivity to profits.Invariably, this has induced the view that the farm is like afactory, such that the land is "mined" instead of treated as anatural habitat. Consequently, traditional notions about farminghave all but disappeared, leaving us with increasing landdegradation instead of land nurturing.This new phase of agriculture is "maintained by a highlycomplex system of farm implement and agrochemical industries, ahighly developed marketing system, and government institutions"(Norgaard 1984, p.162). Examples of such institutions in Canada,as in many other more developed countries, are crop insurance,transportation subsidies, and other income support schemes.Needless to say, "modern" agriculture also has been spurred byunprecedented technological innovations, some of which engenderedmore dependence on agrochemicals. For example, Brown and Young(1990) point out that between 1950 and 1989, world fertilizer use"increased from a meagre 14 million tons to an estimated 143million tons" (p.67). Simultaneously, the public has increasingly3been exposed to many chemicals, the side effects of which arehardly known by the scientific community. Some of these chemicalsare toxic substances causing genetic damage (mutagens), birthdefects (teratogens) and cancer (carcinogens).Problems such as soil erosion and degradation, contaminationof water tables, and hazards posed to consumers, farmers and farm-workers by farm chemicals have, in the wake of public concern forthe environment, stimulated the concept of sustainable development.In part, this has been a reflection of humanity's failure toappropriately relate to the environment (van Kooten 1987). BritishColumbia is not immune to these problems. Chlorinated organicsalong with dioxins and furans from bleaching pulp mills, forexample, have contributed to the deterioration of B.C. fishhabitats. In fact, as a province deriving its livelihood from itsnatural resources, sustainable development is imperative.Agriculturally, sustainable development often refers to low-input or eco-agriculture (Batie 1989). Underlying this approach isthe conviction that humankind should change its relationship withthe environment from that of "conquest and exploitation to that ofco-operation and coexistence" (ibid., p.1088). Many incidents andfactors underscore the need for sustainable development. Forexample, agriculturally it has been documented that, due to lack ofcare or inappropriate treatment of land, 14.8 million acres world-wide are lost beyond reclamation, and a further 49.4 million acresannually become so impoverished that they are unprofitable to farmor graze (Postel 1988).41.2 Problem Statement This thesis focuses on government agricultural programmes thathave encouraged and shaped the agricultural land use practices inthe Peace River region. The general aim is to point out relevantprogrammes that encourage land degradation and adversely affectagriculture's sustainability.We assume that farmers are continually making decisions aboutthe optimal allocation of land so as to maximise the present valueof net farm incomes subject to personal (e.g., survival, risk) andinstitutional (e.g., Canadian Wheat Board System) constraints. Alinear programme (LP) is one of the techniques of mathematicalprogramming that can be used to maximise farm incomes. Theoptimization tool of a linear programme (LP) model is, thus,formulated to examine effects of government agricultural programmeson land use as it pertains to crop and forages production in thePeace River region, where forages are assumed to be a deriveddemand for livestock production. Cultivation practices ofsummerfallow in crop rotations versus continuous cropping areexamined.Parametric linear programming (PLP) is subsequently used toanalyze other optimal land use scenarios by varying the LP'sobjective function coefficients. This will shed light on issuessuch as what happens to optimal land use as government subsidiesare cut or halved. Furthermore, an attempt is made to simulate twoother scenarios: the removal of grains and summerfallow from lowerquality land classes.51.3 Outline of Thesis The second chapter is devoted to a review of the literature,and is followed by a description of the study area: the Peace Riverregion (Chapter 3). Chapter 3 focuses, among other things, onagriculture and land use in the region. Moreover, it examines thegovernment programmes that have impacted on B.C.'s agriculture,with particular reference to the Peace River area. Chapter 4begins with a brief description of linear programming (LP),followed by a discussion of previous LP applications to the PeaceRiver region. Subsequently, an empirical LP model is formulatedfor the Peace River area. In Chapter 5, we present and examine theresults of this LP model, along with the PLP simulated scenarios.Finally, Chapter 6 offers a summary and some concluding remarks.CHAPTER 2LITERATURE REVIEW2.1 Background to Sustainable Agriculture Agricultural research has come under increasing pressure torespond to the environmental concerns of the public, as well asfarmers eager to reduce farm operating expenses by resorting tolow-input farming methods. Sustainable agriculture has, thus,become an emerging area of intense research. Althoughenvironmental protection is not a new phenomenon, the 1980s didparticularly focus attention on agriculture's sustainability.Indeed, this concern is no longer subsidiary to short-term economiccalculations.Precipitating this interest in sustainability issues is therealization that agriculture's resource base--the physicalenvironment of land, water, and air--is in peril and, in spite ofmodern agriculture's "successes", it has over-simplified thecomplex natural ecosystem that it evolved from, which has causedmany environmental problems. For instance, it has increased theincidence of recognized pests and diseases (Hodges and Scofield1983).Furthermore, agricologenic (farmer-induced) diseases alongwith soil erosion and salinization have intensified. Reasons givenfor these problems include introduction of foreign factors into the67ecosystem through, for example, the application of syntheticpesticides, and the tendency of farms to "bypass many of theprocesses that are normally associated with soil fertility" (Hodgesand Scofield 1983, p.20).Concern for ecological balance, if any, becomes a secondaryissue in calculations of farm productivity or profitability. Forexample, van Kooten, Weisensel and Chinthammint (1990) concludedfrom their study of soil conservation in Saskatchewan that, thoughfarmers profess interest in soil stewardship, changes in agronomicpractices to achieve it were not observable yet.1 This is in spiteof their finding that stewardship requires only a minimal (lessthan 5 percent) decrease in net returns (ibid., p.112).Environmental quality and conservation have been the topics ofnumerous international conferences. Four such conferences wereorganized under the auspices of the International Federation ofOrganic Agriculture Movement (IFOAM), which exists primarily topromote the goal of eliminating inorganic fertilizers and syntheticpesticides in favour of dependence on organic materials and naturalpest controls.2The IFOAM conferences considered various ways to makeagriculture sustainable through efficient resource use, lowerproduction costs and enhanced environmental protection.1 Stewardship connotes using soil without decreasing its long-term productivity.2The first conference was held at Sissach (Switzerland) in1977, with subsequent ones held at Montreal in 1978, Brussels in1980 and Massachusetts in 1982.8Nevertheless, no one single practice was recommended as a panaceabut suggestions made included integrated pest management as analternative to widespread use of insecticides. Moreover,polycultures, involving crop rotations and intercropping, wererecommended to improve current practices. Andow (1983) indicatedthat crop diversification is "promising as polycultures often yieldbetter than monocultures per unit of land area" (p.94), for reasonsthat include reduced incidence of diseases and better usage of thenatural soil variability.At a 1978 conference in Prince Edward Island (PEI), Hill(1978) rightly pointed out the need to understand that "any speciesincluding our own survives on the basis of nature. It is ourbiology and our interaction with the supporting environment thatdetermines our survival" (p.175). Unfortunately, farm productivityhardly distinguishes farm practices that degrade from those thatsustain the ecology. As a result, modern agricultural indicatorsshow growth without any acknowledgement of its environmental costssuch as soil erosion.In the U.S., where there is not as much dearth of soil erosionresearch, 2.2 billion tons of topsoil was estimated in 1983 to havebeen lost from croplands (Soule, Carre and Jackson 1990). It isfurther noted that "the average official tolerable losses of 5tons/acre for deep soils and of 1 ton/acre for shallow soils wereviolated," (ibid., p.167). 3 Overall, the U.S. is said to have lost3See van Kooten and Furtan (1987) for an overview of thepertinent U.S. studies, along with Canadian ones.9at least a third of its cropland top soil, while the world islosing 14.1 billion tons per annum (Tivy 1990, p.244).In Canada, concern for diminishing arable land and soildegradation were the bases of a study commissioned by the CanadianEnvironmental Advisory Council in 1983. The resulting report byBentley and Leskiw (1985) pointed out, among other things, thatsoil erosion with its associated sediment pollution in water andair, salinization, and soil compaction are causing the destructionof the country's most productive farmlands. In the Prairies, forexample, millions of acres of once-cultivated land have beenabandoned as farmlands (ibid., p.1). Agriculture's sustainabilityis thus a major concern.It must be borne in mind, however, that there is no consensusamong researchers on the severity of soil degradation (van Kootenand Furtan 1987).2.2 Concepts and Definitions World concern for the environment was apparent in the 1970s,as the 1972 U.N. Conference on the Environment and the subsequentgenesis of the U.N. Environment Programme (UNEP) attest. A worldconference was also convened at Nairobi in 1977 to discussdegradation of agricultural lands. The Food and AgricultureOrganization (FAO) at its 21" Session in 1982 followed up byadopting a World Soil Charter, with provision to implement a WorldSoils Policy. This involves, for example, the adoption of goodland husbandry encouraged by a sound institutional framework.10It was not until 1983, however, that the worsening situationprompted the United Nations General Assembly to ask for a WorldCommission to propose long-term strategies to achieve sustainabledevelopment by the year 2000 and beyond. With the release in April1987 of the report of the World Commission, chaired by Gro HarlemBrundtland, Prime Minister of Norway, sustainable developmentbecame a popular concept.The World Commission on Environment and Development (WCED)averred that "the environment does not exist as a sphere separatefrom human actions, ambitions, and needs..." (p.xi). In otherwords, environment and development are inseparable. Pointing outthat many of the development paths of now industrialized countriesare unsustainable, the Commission proposed sustainable developmentas an alternative. This concept was defined as "development thatmeets the needs of the present generation without compromising theability of future generations to meet their own needs" (p.8). Suchdevelopment requires that countries pursue policies that will, forinstance, develop agriculture along ecological principles.Knowing that this might well become political rhetoric, theWCED called for action by proposing many legal principles, amongstwhich was the Principle for Conservation and Sustainable Use:"States shall maintain ecosystems and ecological processesessential for the functioning of the biosphere, shallpreserve biological diversity, and shall observe theprinciple of optimum sustainable yield in the use ofliving natural resources and ecosystems (p.348).In November 1987, the General Assembly of the UN adopted theEnvironmental Perspective to the Year 2000 and Beyond, which was11prepared by the Intercessional Intergovernmental PreparatoryCommittee of the Governing Council of UNEP along with WCED. Thisexercise signalled a political acknowledgement of theinseparability of economic development from the environment. Itmeant, as the World Conservation Union (1989) rightly noted, thatthe "discussion is no longer about the importance of environmental[sic], but rather how development can be implemented within thelimits of sustainable use of resources" (p.14).Sustainable development is a descriptive term that capturesthe growing conviction that development must not be at the expenseof the environment. It is a conviction that not only manypolitical leaders of different stripes are articulating, but manyprogrammes the world over (including those of internationalorganizations, such as the World Bank) are incorporating.The federal government of Canada released a Green Plan in 1990that acknowledges sustainable development as a goal that Canadiansshould collectively strive to achieve. Described by the thenFederal Minister of the Environment, Robert de Cotret, as "the mostimportant environmental action plan ever produced in Canada" (p.3),the Green Plan commits the federal government to an environmentalexpenditure of $3 billion over a period of five years, in additionto the annual budget of $1.3 billion.This comprehensive plan, supported by over 40 federaldepartments and agencies, recognizes the need for a concertedeffort to address environmental problems. As part of the GreenPlan's implementation, the federal government has introduced the12Canadian Environmental Assessment Act that obliges the governmentof Canada to "integrate environmental considerations into itsproject planning and implementation processes" (p.19). Currently,there are over 50 statutes with environmental implications inaddition to legislations, with attendant regulations, by theprovinces and Territories. With over 100 initiatives, the GreenPlan--embodying economic incentives and stronger enforcementmechanisms--can make a significant contribution to sustainabledevelopment in Canada and, possibly, in the rest of the world.In 1989, the U.S., Congress legislated an annual calculationof gross sustainable productivity (GSP), along with conventionaleconomic indicators such as gross national product (GNP), beginningin 1990 (Brown 1990, p.9). Moreover, Careless (1990) notes that a1988 joint study by the U.N. Environmental Programme and the WorldBank proposed the notion of sustainable income, which expands theconventional economic definition of capital to include stocks ofnatural endowments. Other countries--including Indonesia--andmajor organizations such as the United Nations are currentlyrevising their indices as well, while Japan and Norway initiatedtheir revisions in the early 1970s (ibid.).Generally in the revised national accounts, or what may becalled environmental accounting, deductions are made fordepreciation of environmental assets. Sustainability can (in thiscontext) be narrowly defined as the balance between environmentaldepletion and investment in the environment.13Some countries, such as Canada, have resorted to periodicalreports under the theme of State of the Environment (SOE). Ofcourse, this underscores the realization that incorporatingenvironmental indices into national accounts, albeit with greatdifficulties, has become paramount to the survival of humankind.British Columbians have increasingly become aware thatincreases in the province's standard of living are threatened byenvironmental degradation. Consequently, the British ColumbiaRound Table on the Environment and the Economy chaired by ChuckConnaghan has been initiated to establish a framework in whichenvironmental accounting can be commenced and sustainabledevelopment spurred.Agriculturally, sustainable development presents challengesand opportunities for farmers, consumers and governments, not tomention academics. It compels us to re-appraise our institutionsto ensure that those that hinder eco-agriculture are rectified.This re-orientation has sparked new research generally known in theagricultural economics literature as sustainable agriculture,defined by Hileman (1990) as:a systems approach to farming that seeks to develop amultiyear practice that takes advantage of whatever isproduced or can be produced on the farm, includingnaturally occurring beneficial biological interactions toensure soil fertility and to keep losses from pests,weeds, and animal diseases within acceptable levels. Theaims are adequate productivity and profitability,conservation of resources, protection of the environment,and assured food safety (p.27).It may have been noticed that different terms were used todescribe sustainable agriculture. Indeed, there are many such14terms that are used to denote the same idea. For example,"alternative agriculture", "low-input agriculture", "organicagriculture" and "eco agriculture" are sometimes usedinterchangeably with sustainable agriculture even though each mayhave some specific production requirements not shared with theothers. None of these terms should be misconstrued as aresurrection of past or incompatible agricultural practices;rather, they signify attempts to synthesise contemporary andancient methods to ensure that agriculture takes cognizance of theenvironment.Sustainable development draws its intellectual wisdom frommany disciplines including ethics and, as such, is amenable todiverse interpretations, with their concomitant disparateramifications. As Batie (1989) rightly observes, some of itsadvocates even question the desirability of economic growth.2.3 Neoclassical Economics and the Environment Economists have attempted to incorporate the environment intothe framework of neoclassical economics. For example, Turner(1988) suggests a working definition of sustainable development as"maximising the net benefits of economic development subject tomaintaining the services and quality of natural resources overtime" (p.352). Nonetheless, as he points out, there are twovariants of economic modelling of environmental resources, "onemore revisionist than the other in terms of the modificationsrequired in the neoclassical blueprint" (p.354). The revisionists15generally incorporate entropy limits into their economic analyses.It is worth noting that neoclassical economists commonlydichotomize the focus on environment into nature (for amenity usessuch as Parks), and what is imprecisely referred to as pollution(Fisher and Peterson 1976). It is the latter aspect of theenvironment that is the source of public exasperation, and thenucleus of this discussion.Early economic thinkers about the environment, such asMarshall, spoke about the environment as external to production. 4Subsequently, the notions of private cost and social cost wereenunciated which, for policy purposes, implied that imposing taxes(often called Pigouvian) to equalize marginal social cost andmarginal private cost will resolve the pollution problem. Hence,dealing with the "externality problem" became an extension of pricetheory in economics.Given the collapse of major command economies in 1989 and thedemise of the Soviet Union, many people believe that the efficiencyof markets in allocating resources has been vindicated. Privateenterprise though is found wanting in so far as resolving theincreasing environmental calamity is concerned. In fact, Krutillahas indicated that "private market allocations are likely topreserve less than the socially optimal amount of the naturalenvironment" (Fisher, Krutilla and Cicchetti 1972, pp.605-606).4See Fisher and Peterson (1976) for a survey of environment ineconomic analysis.16This is exacerbated by the fact that some environmentaldegradation is very difficult, if not impossible, to reverse.Others, including Nobel laureate Arrow (1985), have alluded to thelimitations of the price system in dealing with environmentalproblems, while even the notion of externality has been critiqued(e.g., Scitovsky 1954 and Bator 1958).Another major argument advanced against total reliance onprivate markets to solve environmental problems is the notion ofoption demand, defined by Krutilla (1967) as "a willingness to payfor retaining an option to use an area or facility that will bedifficult or impossible to replace and for which no closesubstitute is available," (p.780). As well, there is the conceptof existence demand, sometimes referred to as intrinsic demand,described by Johansson (1990) as the derivation of satisfactionfrom the pure fact that an asset is available for other peopleliving now or unborn. The individual in this case may or may notutilize the asset or its services. Markets for these demands arenot impeccably operational as they are bedeviled by difficultiesassociated with public goods such as the free rider problem.An equally important factor, perhaps, is the fact thatneoclassical economists mostly operate in the positive economicsarena, with little or no reference to normative economics.Environmental economics, with its roots in neoclassicaleconomics, continues to grow with, as Winpenny (1991) notes, theunderlying rationale of market failure due to factors such aspublic goods. As well environmental economists tend to accept that17"the workings of markets do not always achieve the most efficientallocation of resources due, for instance, to the existence ofmonopolies and imperfect information among consumers" (ibid., p.2).Techniques of economic valuation such as cost-benefit analysisand cost-effectiveness analysis are grappling with environmentalaccounting and its complexities. These include ignorance about theeffects of programmes or projects, and the task of valuation andquantification of costs and benefits even if known. The increasinguse of Environmental Impact Assessment by countries, privatecompanies and international organizations are attempts to ensurethat development and the environment are not separated.Sustainable development, albeit with its presentamorphousness, means to address the need for humanity to have aharmonious relationship with the environment. It is a new thinkingthat requires us to eliminate the chasm between nature andhumankind. This new thinking also understands, among other things,that "our economic cosmos is not one of uniform circular motion ofcommodities among men but one of elliptical orbits throughinterdependent ecological sectors" (Daly 1968, p.400). Its a newthinking that draws insights from many disciplines and accepts thepremise that unfettered private enterprise does not adequatelyprotect our environment.2.4 Towards Sustainable Agricultural DevelopmentThe intractability of environmental problems has not preventedresearchers such as Smith (1968) and Burt (1981) from grappling18with optimality of intertemporal natural resource allocation, whilesome, including Fisher, Krutilla and Chicchetti (1972), havegrappled with both theoretical and empirical questions associatedwith environmental preservation. Others, such as Potter andChristy (1962), and Ruttan (1971), have put their confidence intechnology to redress environmental predicaments. Ruttan (1971),for instance, argues that:The advance of science and technology has enabled modernsociety to achieve a more productive and better balancedrelationship to the natural world than in the ancientcivilizations or in the earlier stages of Westerncivilization (p.708).Acknowledging, however, that environmental deterioration hadreached an alarming proportion even two decades ago, Ruttan urgedthat institutional and technical change should be redirected toenhance the performance of the ecosystem. This is especiallypertinent in view of how our institutional web of laws and policieshave undervalued the environment.Ruttan's remedial measures include concentration oninstitutional changes to achieve decentralized decision-making suchthat property rights will be established within what he calledenvironmental subsystems, wherein market forces will directproduction and services. He also views a re-organization of thesocio-political environment as imperative, since the private sectorhas no incentive to invest in the research needed to spur thechanges required and the public sector traditionally lacks supportfor such research.The need to integrate economics into an ecological paradigm is19reiterated by Daly (1968), who avers that "the entire physicalenvironment is capital on which the hierarchy of life depends"(p.397). Imploring that elements of this environment (air, waterand soil) should be cared for in the same way that we care forother machines, he outlined a model he calls the "total economy",which is an extension of Leontief's input-out model thatincorporates the human economy into the economy of nature. In itssimplest version, his model has four quadrants representing humanand non-human sectors. While Daly's model has practical problemsassociated with it (e.g., measuring the non-human sector), it isyet another attempt to grapple with the notion of sustainabledevelopment.Calculation of GSP and other environmentally-oriented indicesare attempts to address this problem by countries and/ororganizations. For example, in Japan a Net National Welfare (NNW)index is calculated annually to adjust the country's GNP. NNWprimarily corrects for environmental problems, such as watercontamination, by making deductions for pollution and losses due tourbanization (Careless 1990, p.9). Norway, noted as the premierinitiator of environmental accounting, regularly reevaluates bothits renewable and non-renewable resources to take cognizance ofenvironmental degradation and resource enhancement in the nationalaccounts.Synthesizing Daly's 1968 ideas, Daly and Cobb (1989) developedyet another model that defines a welfare index based on social,economic and environmental considerations, which they called the20Index of Sustainable Economic Welfare (ISEW). This index appraisesmany factors including automobile use and loss of valued habitats(e.g., wetlands). Though their index is a remarkable one, itscomplexity and scope makes it difficult, if not impossible, topractically compute; hence, it may never be conjectured inenvironmental accounting.With respect to agriculture, Hileman (1990) notes that thereare a variety of methods that contribute to sustainability ofagriculture, including crop rotations, biological pest controls,and an amalgam of livestock and crops. The appeal of croprotations is that they help to control pests and diseases. In theState of Washington, where some farmers actively practicealternative agriculture, Hileman notes that wheat root pathogens,for example, have been eliminated with three-year rotationsinvolving barley and peas or lentils.As well, she observes that "grain yields following a legumeare usually greater than yields of grain planted as a continuousmonoculture, no matter how much fertilizer is applied" (Hileman1990, p.33). Manure and forage legumes also reduce water and winderosion, while non-native predator pests and/or sterilized insectpests are part of integrated pest management programmes.In recognition of the viability and the need for sustainableagriculture, the 1990 U.S. Farm Bill includes proposals, dubbed"positive incentives," to eliminate restrictions on polycultures,and give farmers the flexibility needed to grow different cropswithout loss of government benefits (Hileman 1990).21Contending that mechanical philosophy leaves the impressionthat technology, whether existing or not, will redeem us, vanKooten (1987) argues for a public philosophy for agriculture, whichhe integrates into a notion of stewardship, akin to coevolutionarydevelopment as enunciated by Norgaard (1984). This, he says, isthe only relevant solution to our environmental degradation. Hebuilds this notion into a socio-economic model for prairieagriculture, that encompasses religious communalism and requireswhat he calls "Agricultural Practices Councils" to be establishedin agricultural regions. These would have authority to enforcesustainable agriculture guidelines. His suggestions, radical asthey may be, attempt to operationalize the notion of eco-agriculture by encouraging decision making at the local level asopposed to provincial or national level.Upon reviewing prairie agriculture, van Kooten and Kennedy(1990) noted that soil degradation involving soil erosion andsalinity is the main problem. They suggested practices such ascrop diversification, greater reliance on management (flexcropping)and other applicable methods of low-input agriculture to rectifythe problem.The Peace River region manifests prairie agriculture and itsattendant erosion problems. Indeed, the B.C. Ministry ofAgriculture, Fisheries and Food (BCMAFF) noted in its Annual Reportof 1984 that the Peace River soils are among the most erodible inNorth America (p.19). Hence, sustainable agriculture in the PeaceRiver area, and for that matter in the rest of B.C., is essential22if agriculture is to continuously contribute to the growth of theeconomy and increase standard of living in British Columbia.CHAPTER 3DESCRIPTION OF STUDY AREA AND PERTINENTGOVERNMENT PROGRAMMESAs much as government programmes have encouraged what is grownin the Peace River region, the influence of the natural habitat(including climate, soils and vegetation) certainly can not beignored. As a result, this section presents a general overview ofthe study area. It is also intended to familiarize readers withthe area, and to present the framework for the formulation of alinear Programme model.3.1 Geography and Climate The Peace River region is located in the north-eastern part ofthe province of British Columbia. It is bounded on the north andsouth by latitudes 58 °N and 55oiN, respectively, while the Albertaborder is the eastern boundary and the Rocky Mountains border it onthe west. This tends to isolate it from the rest of BritishColumbia. The total land area of the region, as reported by theStatistics Canada 1986 Agriculture Census, is some 48.2 millionacres. This region is interspersed with timber, rocks, lakes andrivers.The Peace River demarcates the area into the North, with FortSt. John as the main urban centre, and the South, with Dawson Creekas the paramount urban centre. The northern plateau has an2324altitude not exceeding 1,000 feet above sea level in some parts,while the south is generally some 2,500 feet above sea level(Graham and Lopez 1976, p.9).Climatically, the area resembles much of the Prairies, but ithas generally milder summers and relatively colder winters. Duringthe winter warm Chinook winds that blow can lead to snowmelt and,consequently, soil erosion problems; cold arctic air may alsoinduce frosts in the summer. The frost-free period variessignificantly amongst areas; for example, Dawson Creek and FortSt.John have 78 days and 115 days, respectively, as Table 3.1shows.Table 3.1 Average Frost-Free Period in Selected Towns ofthe Peace River AreaStation Height(feet above M.S.L.)Frost-Free Period(Days)Dawson Creek 2,148 78Fort Nelson 1,253 106Fort St. John 2,280 115Source: Environment Canada (1982). Canadian Climate Normal,Volume 6, Frost: 1951 - 1980.Total precipitation in the region shows that summer is thewettest, with July having an average precipitation of over 70 mm inDawson Creek, Fort Nelson and Fort St. John (see Table 3.2). Theregion is classified as having moderate precipitation. Theprincipal streams and rivers are inaccessible due to deep valleysand so obtaining water for even domestic use could be a problem.25The farming community therefore tends to depend on dams or"dugouts".Considering the Peace River region's climate, it has beenobserved that the three main climatic factors--rainfall,temperature and wind--vary significantly within and between years.Hence, agricultural production is more uncertain relative to mostother regions in British Columbia (Graham and Lopez 1976, p.12).Table 3.2 Average TotalSelected Towns ofPrecipitation 1951-80 inthe Peace River Area (mm)Dawson Creek Fort Nelson Fort St. JohnJanuary 36.1 24.9 35.6February 28.9 19.5 27.3March 30.6 24.4 29.7April 19.0 16.7 21.5May 35.2 41.7 38.9June 72.8 69.1 68.0July 70.8 84.3 77.1August 68.9 61.2 60.3September 41.5 41.6 39.2October 30.9 24.3 27.7November 29.6 22.7 31.2December 39.4 21.4 36.1Year 503.7 451.8 492.6Source: Environment Canada (1982). Canadian ClimateNormals, Volume 3, Precipitation: 1951-19803.2 Soil and VegetationMany soils are found in the region. Classification of these26soils at the Great Group Level include Dark Gray Solod and OrthicGray Luvisol (see Agriculture Canada 1986). The soils generallyhave moderate drainage capabilities but rapidly drained ones suchas Neumann soil are also present in the region. Soil formation inthe area is low due partly to the region's cold climate and thefact that the soils generally are shallow.Indigenously, the Peace River's vegetation consists of Park orWoodland. The wooded areas have an undergrowth that is excellentfor grazing livestock in the summer months. There is a denseforest containing merchantable white spruce and lodgepole pine inthe South. Through burning, vegetation in a significant part ofthe open areas has become typical prairie.Soil erosion, especially that by water and wind, bedevil thearea. This is due in part to the impervious subsoil. Sheet andgully erosion are present even on the slightest slopes of thesummerfallowed lands. In fact, the area has been identified ashaving "the highest erosion risk in all reporting regions inBritish Columbia" (Van Vliet and Hall 1991, p.2). Consequently,the area's T-value (tolerable soil loss) is low; it is about 2tons/acre. Largely as a result of this problem and other non-landsustaining practices, the Peace River Soil Conservation Association(PRSCA) was established in 1986 by some grain farmers in the area.3.3 Agriculture and Land Use Settlement in the Peace River region was undertaken by furtraders in the 1790s. The subsequent Klondike gold rush, coupled27with the introduction of the railway and the construction of theAlaska Highway in 1942, stimulated further settlement.Agriculture, the backbone of the region's economy, was largelysubsistence prior to the first World War. Production occurredaround trading posts such as Dawson Creek and consisted of fruits,vegetables, a bit of grain and livestock (B.C. Department of Landsand Forests 1952, p.15).Livestock farming initially constituted the major land use inthe area but, by 1939, grain farming (dominated by wheat until the1970s) had become the primary activity. Since grain farming waspractised by people mostly from the southern Prairies, theagronomic methods used in the Peace River region were akin to theprevailing southern prairie practices. In view of varying landfertility and climatical inconsistencies, grain quality and yieldvary enormously. Wheat yields initially averaged about 30 bu/acreon the black soils of the Peace River area, and about 50 percentless in the gray-wooded (gray luvisolic) areas. In 1990, wheatyielded 40 bu/acre (Statistics Canada, Grain Trade in Canada).Land use in the Peace River area shows that, of the region's48.2 million acres, area in farms constituted 1.7 million acres in1971 and 2.2 million acres in 1986 (Table 3.3). It is also shownby Table 3.3 that improved farmland increased in a decade (1971-1981) by 171,048 acres, or 22.5%, while, in the same period,unimproved land declined by 58,421 acres, or some 6.4%.5 However,5Agriculture Canada classifies unimproved land as nativepasture or hay land that had not been cultivated, grazing and wasteland, rocky land, marsh, etc.Table 3.4^Peace River Principal Crop Acreage: 1979-90 ('000)Year Wheat Barley Oats Canola Total As % of Total AcreageA1979 71.3 111.8 38.9 267.4 489.4 14.6 22.8 8.0 54.61980 142.5 171.9 43.8 138.7 496.9 28.7 34.6 8.8 27.91981 94.4 197.8 53.0 62.4 407.6 23.2 48.5 13.0 15.31982 134.2 193.4 54.5 139.4 521.5 25.7 37.1 10.5 26.71983 142.9 155.1 54.5 198.1 550.6 26.0 28.2 9.9 36.01984 142.9 167.9 46.8 207.9 565.5 25.3 29.7 8.7 36.81985 151.7 172.1 46.8 173.7 544.3 27.9 31.6 8.6 31.91986 88.0 138.1 46.8 110.1 383.0 23.0 36.1 12.2 28.71987 97.9 121.1 54.5 108.9 382.4 25.6 31.7 14.3 28.51988 93.5 94.6 54.5 108.9 351.4 26.6 26.9 15.3 31.01989 115.7 116.0 62.5 76.8 371.0 31.2 31.3 16.8 20.71990 111.3 116.0 78.1 99.1 404.5 27.5 28.7 19.3 24.5Ave. 115.5 146.3 52.9 140.9 455.7 25.4 32.3 12.1 30.228Table 3.3 Peace River Land Use: Census Years (Acres)1971 1976 1981 1986Total LandArea 48,199,680 48,199,680 48,199,680 48,199,680ImprovedFarmland 761,107 1,853,256 932,155 1,008,644UnimprovedFarmland 907,966 793,130 849,545 1,168,067Area inFarms 1,669,680 1,060,126 1,781,704 2,176,711Sources: BCMAFF (1982), pp. 16-17; BCMAFF (1980), pp. 15-16;Statistics Canada, Agriculture Census. B.C. 1986,Catalogue # 96-112.W = wheat; B = barley; A = oats; C = canolaSource: Author's calculations from B.C. figures obtained fromStatistics Canada, 1986 Agriculture Census, Catalogue 0 96-112;Catalogue # 22-002; Catalogue # 22-201. Based on Statistics Canada'sindication that, in 1986, acreages for wheat, barley, oats and canolawere 89%, 86%, 78% and 99%, respectively, of the B.C. total.improved farmland decreased by some 49.7% from 1976 to 1981, whileunimproved farmland increased by 7.1%. Between 1971 and 1986,improved and unimproved farmland increased by 32.5% and 28.7%,29respectively. In this period (1971-1986) area in farms increasedby 30.4%, as agriculture continues to expand into more remote areaswith the support of various government programmes.Acreage of the principal crops (wheat, oats, canola andbarley) in the Peace River region is presented in Table 3.4. Itshows that, between 1979 and 1990, total acreage averaged 455,700acres. Barley had the largest average share (32.3%), of the totalacreage, followed by canola (30.2%), wheat (25.4%), and oats(12.1%). Acreage for all four crops vacillated during this period(see Figure 3.1). For example, barley ranged from a high of197,800 acres in 1981 to a low of 94,600 acres in 1988. Similarly,canola varied from its high of 267,400 acres in 1979 to a low of62,400 acres in 1981; wheat's acreage spanned from a high 151,700acres in 1985 to only 71,300 acres in 1979. The acreage of wheatshows the largest oscillation followed by canola, while oatsexhibits hardly any swings (see Figure 3.1).These swings may be attributable to grain prices and CWB quotaincentives that encourage expansion of grain acreage (see section3.4). Between 1979 and 1990, the largest total acreage for thesefour primary crops in the Peace River occurred in 1984 with thelatter part of the 1980s accounting for the lowest acreages (Table3.4). The on-going grain "subsidy war" between the U.S. and theEuropean Community, which commenced in 1985 and has resulted inrecord low world grain prices certainly was instrumental indepressing grain acreages in the Peace River region.Fig. 3.1: Peace River Principal CropAcreage: 1979-90 ('000)-200-150-100: ■11.1.-uuro, 200-150-< 100--e- Wheat —I— Barley^Oats —+— Canola30300^300250- -25050- -50Or^197901980^1982^1985Years 1987^1990In terms of production, barley generally ranked as the premiercrop between 1979 and 1990 followed by wheat. They averaged about148,800 tons (about 41%) and 107,290 tons (about 29%), respectively(Table 3.5). The average total production of these four crops was368,400 tons, with the largest of 488,900 tons in 1980 and thesmallest of 319,900 tons in 1981. Canola's average output was thethird largest (about 16%), though it generally had the secondlargest average acreage (about 30%) in the region.Year W B A C As % Total ProductionB^.^A1979 80.1 119.7 43.8 133.7 377.3 21.2 31.7 11.6 35.41980 166.9 208.5 51.7 61.9 489.0 34.1 42.6 10.6 12.71981 85.4 167.2 42.5 24.8 319.9 26.7 52.3 13.3 7.71982 90.7 179.5 55.7 54.5 380.4 23.8 47.2 14.6 14.31983 125.6 175.4 57.6 62.2 420.8 29.8 41.7 13.7 14.81984 101.0 134.6 42.1 62.2 339.9 29.7 39.6 12.4 18.31985 80.4 109.0 31.8 31.6 252.8 31.8 43.1 12.6 12.51986 82.4 164.9 49.0 54.6 350.9 23.5 47.0 14.0 15.51987 104.0 129.9 58.5 62.2 354.6 29.3 36.6 16.5 17.51988 114.8 121.3 67.9 64.4 368.4 31.2 32.9 18.4 17.51989 122.6 138.4 70.2 37.1 368.3 33.3 37.6 19.1 10.11990 133.4 136.5 86.2 42.1 398.2 33.5 34.3 21.6 10.6AVE. 107.3 148.8 54.7 57.6 368.4 29.0 40.6 14.9 15.6Total31Table 3.5 Production of Principal Crops in the Peace RiverRegion: 1979 - 90 ('000 Tons)W = wheat; B = barley; A = oats; C = canolaSource: Author's calculations from B.C. figures obtained fromStatistics Canada, 1986 Agriculture Census, Catalogue 96-112;Catalogue 0 22-002; Catalogue 0 22-201. We assumed that, ofB.C.'s total production, Peace River accounted for: wheat 89%,barley 86%, oats 78% and canola 99%; these were the respectiveshares in acreage in 1986.Table 3.6 Peace River Livestock Production: Census Years1971 1976 1981 1986Dairy cows 1,692 1,999 2,699 1,669Beef cows 11,037 25,764 31,825 31,767Calves 9,744 21,820 28,559 29,001Horses 4,090 4,833 7,723 8,017Pigs 12,759 5,123 15,125 12,192Sheep 5,672 9,807 8,256 7,028Sources: BCMAFF (1982), pp. 36-41; BCMAFF (1980), pp. 25-28;Statistics Canada, Agriculture Census, B.C. 1986,Catalogue # 96-112.Livestock accounts for not an insignificant amount of land usein the Peace River region. It is especially prevalent in the areascharacterized by native pastures and low quality grain. Hogsinitially were the major livestock raised, but beef cow productionLivestock32is now the main activity of the non-grain farmer. For example, in1986, 31,767 beef cows and 12,192 pigs were raised (Table 3.6).In view of intense pressure on farmlands, by developments inurban areas for example, "legislation was put in place in 1974(under a New Democratic Party government) which placed the majorityof the higher quality land in the province in an Agriculture LandReserve (ALR)" (Agriculture Canada 1983, p.21). However, provisionwas made to allow some of the reserve land to be exempted fromagriculture use upon request to the Land Commission.The area of land in each Canada Land Inventory (CLI) class andthe proportion in ALR in the Peace River region is indicated byTable 3.7. The quality of land (i.e., capability to support fieldcrops) is inversely related to land class (see Appendix C for adescription of land classes). It must be pointed out that the CLIclassifications have some anomalies. For example, "the improvedrating does not indicate that it is either technically oreconomically possible or desirable to drain or irrigate the land(Agriculture Canada 1983, p.22). The ALR generally excludes landin CLI classes 5 - 7; this is indicated by the relatively minimalproportion of these land classes in the ALR (Table 3.7).Based on the 1986 area in farms and the proportion of improvedCLI in the ALR, the author estimated that over 500,000 acres ineach of land classes 1 - 4 were under cultivation in the PeaceRiver region in 1988 (Table 3.7). Class 5 land cultivated was94,200 acres, while less than 15,000 acres of classes 6 and 7 landwas under cultivation (Table 3.7). The "small" area of classes 5 -337 land under cultivation stems from the "biases" in thedesignation of ALR noted above. More "poorer" land classes arecultivated than our calculations suggest.Table 3.7^Area in CLI Agricultural LandClass and ALR, 1988 (1000 Acres)Class CLI ALR CultivationUnimpr Impr1 9.4 9.5 8.8 515.52 299.2 299.0 279.9 520.63 902.7 925.6 832.5 500.24 1,239.0 1,323.3 1,250.2 525.45 4,162.6 4,934.4 835.8 94.26 1,465.1 1,412.8 38.0 14.97 3,705.8 2,870.3 29.7 5.9Total 11,783.8 11,774.9 3,274.9 2,176.7Notes: CLI = Canada Land InventoryALR = Agriculture Land ReserveImpr = ImprovedUnimpr = UnimprovedImproved and unimproved land each has an additional40 million acres of water and other unclassifiedareas such as urban and unmapped zones, giving each atotal of some 52 million acres.* Author's 1986 Estimate. These were obtained by taking CLIimproved farmland proportion in ALR, multiplied by theregion's total area of farms (2.2 million acres) in 1986.A description of the land classes ispresented in Appendix C.Sources: Agriculture Canada 1983, p.32Agriculture Canada 1988, pp.24 - 25Table 3.8 Peace River Land Use by Land Class (,000 Acres)Land Class1 - 3 4 - 5 6452.8 412.2 3.250.8 83.3 2.3147.1 410.0 6.8460.8 1,163.0 26.01,111.5 2,068.5 38.2ActivityGrain and ForageGrassland and PastureForested rangeForested landsTotal34In 1978, acreage by land class in the Peace River was outlinedby the B.C. Select Standing Committee on Agriculture as shown inTable 3.8. It shows that grains are cultivated on classes 1 - 6land, which suggests that some grains in the Peace River regionare to be found on marginal land.Source: Graham and Anderson (1982), p.49Grain production has not ceased to be the mainstay of theeconomy of the region. In fact, most business activities in urbancentres such as Dawson Creek and Fort St. John depend upon thisprimary activity. Nevertheless, other economic activities rangingfrom forestry to mining provide avenues for off-farm employment.Summerfallow accounts for a large acreage in the Peace Riverregion; it averaged over 200,000 acres (about 21%) between 1974 and1982. Newly broken land averaged about 21,000 acres during thesame period (Agriculture Canada 1983B). Large family farmsdominate grain production in the region. Of the 700 grain farms,35 percent were over 960 acres and they account for 75 percent oftotal output (Mcguire 1985, p.4).3 53.4 Government Programmes Affecting Peace River AgricultureUnder Canada's constitutional provisions, both the federal andprovincial governments have jurisdiction over agriculture. Thegovernment of British Columbia, through its Ministry ofAgriculture, Fisheries and Food, administers 37 Acts (see AppendixB), while Agriculture Canada oversees 47 Acts (see Appendix A) onbehalf of the federal government (Agriculture Canada 1988, p.48).Other provincial and federal ministries or departments, such as theB.C. Ministry of Forestry and Lands and Environment Canada, alsooffer programmes that impact on agriculture.Federal programmes tend to focus on support services foragriculture and production efficiency, in addition to sustainingfarms across the country. However, the federal government offersfinancial programmes, such as crop insurance, to protect farmersagainst losses due to factors including natural hazards and marketconditions. Mortgage credit is even extended to farmers throughthe Farm Credit Corporation.The federal government has sometimes also given out money inan ad hoc fashion. A case in point is an announcement in November1991 of $800 million--farmers want at least double that amount--tohelp farmers cope with an on-going "subsidy war" started in 1985between the US and the European Community (EC), which has resultedin record low grain prices. Since 1985 the federal government hascontributed $12.6 billion to support farmers, while since 1986 theEC and the US government have paid $182 billion and $285 billion totheir respective farmers (The Vancouver Sun, 1 November 1991, A15).36Provincial programmes are similarly oriented, and range fromoverseeing marketing boards and agencies, regulations to maintainhigh quality, extension services and financial assistance tofacilitate farm investment and development.Furthermore, joint federal and provincial institutions areoccasionally established. A major recent example is the five-year(1985 - 1990) $40 million Agri-Food Regional Development SubsidiaryAgreement (ARDSA), consisting of programmes for productivityenhancement, resource development and commodity development(Agriculture Canada 1988, p.47).In the next section, we examine the major programmes of bothlevels of government that have shaped or developed the Peace Riverregion's agriculture. As noted earlier, the Peace River area isthe primary domain for the cultivation of cereals and oilseeds inB.C. As one of B.C.'s ten major commodities, grains accounted forsome four percent of the province's total farm cash receipts in1986. The total production of grains in the Peace Riverconstitutes only two percent of western Canada's output of the fourmajor grain crops--wheat, oats, barley and canola (AgricultureCanada 1988, p.74).In the Peace River region, there was an annual acreageincrease of some 21,000 acres that apparently came from marginal orunimproved land (see section 3.3). Government programmes have beenlargely responsible for the conversion of land into the productionof grains, and the continued financial viability of grain farms inthe Peace River region. Wheat grown in the Peace River region37tends to be one grade lower in terms of quality relative to therest of the prairies, but average yields are comparable. About 80percent of the output in the region is sold through the CanadianWheat Board (CWB). Since the primary mode of transport for thegrains is rail, changes to the Crow Rate transportation scheme willimpact on Peace River's grain production and expansion.3.5 Review of Pertinent Government Programmes Since its inception in 1935 and its designation in 1943 as thesole purchaser of prairie grains, the CWB has operated with a quotascheme that encourages conversion of unimproved land into grainproduction. As van Kooten and Kennedy (1990) point out, "CWB quotaallocations are tied to the amount of improved land, therebyencouraging farmers to improve marginal land and increase theamount of grain they are permitted to deliver to the marketingsystem" (p.748).The CWB system discourages grain producers from sellingdirectly to the domestic feed market. In fact, livestock farmershad to pay premiums as high as $40/ton, until the 1974 new FeedGrains Policy was enacted to make feed grain competitive with cornin Montreal (Agriculture Canada 1983B). This policy favouredEastern Canada over Western Canada, but it did not last long, beingreplaced by other forms of feed assistance to eastern livestockproducers (e.g., Feed Freight assistance). Coupled with the factthat grain fed to livestock is excluded from calculations forgovernment benefits under the 1949 Agricultural Stabilization Act,38diversification of farms to include livestock for example andcontribute to practices of sustainable agriculture has beenhindered, if not discouraged.Besides delivery of grains to the CWB, growers can selldirectly to feed users, with assistance from the Livestock FeedBoard of Canada's administered Feed Freight Assistance Programme(FFAP). With the exception of the Peace region itself, all areasof B.C.'s regional feed grain market qualify for assistance, andthis assistance was extensively used in the early 1980s. Of thetotal expenditures of the feed freight subsidy programme, B.C.'sshare increased from 18% in 1976/77 to 39% in 1981/82 (ibid., p.5).Livestock production in the Peace River region was hampered by theFeed Freight Assistance Programme.These increased expenditures for the provincial governmentwere due largely to the relaxation of inter-provincial sales offeed grains. Factors such as the superior quality of prairie feedgrains have adversely affected sales of the Peace River region'sgrains, the feed freight subsidy not withstanding. For example,more than 220,460 tons moved to southern B.C. from the Peace regionprior to 1972, but, since 1974/75, the average has only been about33,069 tons (ibid., p.6). In response to the declining use ofprovincial feed grains, the B.C. government introduced the FeedGrain Market Development Programme (FGMDP), which provides an extrapayment per ton for grain sold locally as feed. In 1989, about$1.8 million was paid under this scheme, an increase of some$20,000 over 1988 payments. While FGMDP may have boosted more39consumption of local grains, it may also have supported grainproduction on marginal lands.The provincial government also subsidizes B.C. Rail movementof grains from the Peace River region, comparable to the Crow Rate.A direct result is high farm gate prices for grain producers and aninherent inducement to cultivate unimproved land, which potentiallysubjects more areas to land degradation.Another government programme that has encouraged grainproduction and cultivation of marginal land is the Western GrainStabilization Act (WGSA) of 1976, which paid a net high of $5.2million in 1986 to grain producers in B.C. (see Table 3.9). ThisAct, intended to stabilize net income, was amended in 1988 to coveran additional nine crops, including sunflower, but not leguminouscrops that can also enhance soil fertility. It has now beenreplaced by the Gross Revenue Insurance Programme (GRIP).The WGSA's implications for sustainable agriculturaldevelopment were grim, since it not only fostered cultivation ofunimproved land (van Kooten 1990), but no sustainable productionpractices of good husbandry such as grains cum legume rotation wererecognized. Crops covered under this scheme induced the farmer tocultivate monoculturally. Moreover, the WGSA hampereddiversification of farms. In the Prairies, for example, paymentsof WGSA alone skew production costs against diversifying wheatfarms even though such an effort can be profitable.In addition, the 1986 introduction of the Special CanadianGrains Programme (ScGP) to offset low international grain pricesTable 3.9 Net Direct Payments to B.C. ProducersBy Prociramme: 1971-89 ($ 1000) Year WGSA CI CR FUEL SCGP1971 0 787 82 0 01972 0 669 153 0 01973 0 75 73 0 01974 0 218 4 0 01975 0 -242 4 0 01976 -150 -604 6 84 01977 -168 235 3 895 01978 522 -265 6 986 01979 1,410 1,779 7 824 01980 -488 464 0 642 01981 -570 3,843 0 513 01982 -386 3,773 0 521 01983 -438 5,690 0 426 01984 1,188 7,878 0 443 01985 3,116 10,132 0 748 01986 5,217 10,871 0 1,287 01987 7,702 4,895 0 856 5,5001988 3,222 3,004 0 1,010 6,6121989 130 7,638 0 724 240paid over $966 million country-wide in 1987, of which $5.5 millionwent to B.C., mostly to the Peace River area; the correspondingpayments for 1988 were $1.1 billion and $6.6 million, respectively(see Table 3.9). This programme also stimulates extensive andintensive cultivation irrespective of any underlying calamitousenvironmental consequences.Notes: WGSA = Western Grain Stabilization Act;CI = Crop Insurance; CR = Chemical Rebates;SCGP = Special Canadian Grain ProgrammeSource: Agriculture Canada (1990). Farm Income and Financial Conditions and Government Expenditures.The Crop Insurance Act was enacted in 1959 to provide all-riskcoverage to farmers throughout the country. It undoubtedlyprecipitates reliance on monoculture, as it reduces risks41associated with single-crop farming. Indeed, monoculture croppingpractices have become more widespread in the last 10 to 15 years inthe Peace River region (Van Vliet and Hall 1991). By providingcoverage for monoculture involving summerfallow practice, itindirectly hastens land deterioration, since fallow speeds up soilerosion by wind, snowmelt and rain, and could contribute to soilsalinity.Moreover, crop insurance stimulates farmers to cultivateeligible crops on unimproved lands. Coverage was at 70 percent andis now 80 percent of average yields. The limited number of cropsthat are insurable do not include crops that can assist in soilnurturing, and so such crops have high risks associated with theircultivation. Hence, only little amounts of soil-enriching cropsare cultivated in the Peace River region. Consequently, cropinsurance (as presently constituted) has had profound adverseimplications for eco-agriculture in British Columbia's grain belt.In 1989, the federal government established a Federal-Provincial Agricultural Committee to propose terms of reference foragricultural policy in Canada. Following this committee's report,which proposed four Policy Principles including increasedenvironmental sustainability, another committee has suggested twonew programmes to stabilise income for grain farmers: the GrossRevenue Insurance Program (GRIP) and the Net Income StabilizationAccount (NISA). However, as Gray et al (1990) point out, GRIP hasthe potential of becoming the most important AgriculturalLegislation, yet it does not cover forage and pasture land. As42constituted, GRIP will not foster sustainable agriculture; in fact,GRIP will inhibit it not withstanding the committee's proposal tolimit "total seeded acreage for each producer at 110 percent overthe previous three year average" (ibid., p.35).The government programmes examined above are only some of themultifarious institutions that have shaped agriculture in the PeaceRiver area. Whether ad hoc or not, these programmes have tended tothwart the practices of alternative agriculture, a problem thatsome programmes are occasionally instituted to resolve. A case inpoint is the Prairie Pothole Project intended to fosterconservation of the natural habitat for waterfowl. As well,Agriculture Canada recently (1989) introduced the Permanent CoverProgramme (PCP) with a budget for the first phase of $19.5 million.This has helped more than 4,000 farmers in the prairies to convertsome 300,000 acres of "marginal" lands into permanent cover such asgrass, according to Agriculture Canada's Prairie FarmRehabilitation Administration.Instead of such ad hoc attempts to spur sustainableagriculture, what is needed is a re-appraisal of existinggovernment programmes or institutions to make them friendly to theenvironment. This is imperative because current land degradingagricultural practices in the Peace River region manifested by anescalation of monoculture and an increasing usage of unimprovedmarginal lands, are largely the result of government programmesthat, among other things, inadvertently engender landdeterioration.CHAPTER 4MODEL OF INSTITUTIONS AND LAND USE4.1 Introduction to Linear ProgrammingOptimization techniques, in both their static and dynamicaspects, have been widely applied to problems in many disciplinesincluding agriculture. These problems are divided into those withconstraints (equality and inequality) or without constraints.Solving unconstrained optimization problems was assisted bycalculus expounded by Newton and Leibnitz in the seventeenthcentury (Hazell and Norton 1986). Lagrange found solutions toequality-constrained problems in the eighteenth century, while itwas not until the 1940s that Neumann and Dantzig outlinedprocedures to solve inequality-constrained optimization problems(ibid.).Equality constrained optimization problems are often solved byconverting them to unconstrained ones using the Lagrangean method.The associated Lagrange multipliers--referred to ag shadow pricesby economists--indicate by how much the objective function willincrease or decrease with a unit relaxation of the constraint.Thus, an unbinding constraint has a zero shadow price.Optimal solutions must satisfy the necessary condition (i.e.,the first derivative vanishes), and the sufficient condition (i.e.,the Hessian matrix is negative (positive) semi-definite for a4344maximum (minimum)). Satisfaction of the former but not the latterleads to a saddle point (i.e., the solution is neither a maximumnor a minimum). Furthermore, the optimal solution must satisfy thenon-negativity constraint.Resource allocation problems are usually analyzed byeconomists using optimization methods. In terms of linearprogramming, both the objective function and constraints must belinear. Mathematically, this is often expressed as:(4.1) Max Z En C X-1S. t.(4 . 2)(4.3)^Xwhere: i = 1,....,Mj = 1,....,NIn such a formulation, Z is the objective function andC1,....,Cnare the objective constants, while X1,....,Xnare referredto as the instruments; Aii are the technical constant coefficients;B1,....,B, are the available levels of the resource. Underlyingthis expression are three quantitative aspects: an objectivefunction that can either be maximized or minimized, different waysto achieve the desired objective, and limited availability ofresources (Heady and Chandler 1958, p.2). The opportunity set--theset of instruments satisfying the MN non-negative constraints (4.2and 4.3)--is defined by the following closed convex set(Intriligator 1971, p.74):45(4.4)^X={ X e EnIAXsB, )(0}By the Weierstrass theorem, there exists a boundary solution ifthis opportunity set is compact.6 However, this does not mean thatthere is always a unique solution; there could be many solutions.Some of the assumptions underlying such a model are:1) the problem is deterministic (i.e., Cj, Au, and B.coefficients are known constants);2) resources employed are homogeneous;3) activities are additive; and4) inputs have perfectly elastic supply.Since its inception in 1947 by Dantzig as a method to plan theactivities of the U.S. Air Force (Dorfman, Samuelson and Solow1958), LP has become a major research tool in agriculturaleconomics. It enables researchers to empirically discuss problemsrelated to farm management and spatial location.4.2 Applications of Linear Programming to Peace River AgricultureLinear programming has been used to analyze Peace Riveragriculture in various ways. Nisbet (1962) applied LP to proberesource use in the area and Craddock (1970) used LP to examine theeconomic efficiency of grain production in Canada, whichnecessarily included the Peace River region. Using 1966 as the6 This theorem says that a continuous function defined over acompact set attains a maximum over the set.46base year, he selected wheat, oats, barley, rye, mixed grain andcorn (where applicable) as the crops in his model.There were eight versions of Craddock's model; models 1 to 3presumed an annual Canadian wheat export demand of 420 millionbushels, 350 million bushels, and 300 million bushels,respectively; model 4 was similar to model 2 except corn importswere not allowed; models 5 and 6 were the same as models 1 and 2,respectively, but for the assumption that there was no federal feedfreight subsidy; models 7 and 8 were characterized by theassumption that acreage adjustments, because of low demand forwheat, are made only on the Prairies.Unfortunately, Craddock (1970) generally, lumped the entirePeace River region with Alberta for the simple reason that farmingin the region bears greater resemblance to the Prairies thanBritish Columbia. Nevertheless his models 1 and 4 suggested that56 percent of Peace River region's land was inefficient inproducing cereals; models 5 and 6 showed that land in BritishColumbia and Eastern Canada were competitive in the production ofgrains vis-a-vis the Southern Prairies (ibid., p.22).Holtby (1972) discussed the most profitable use ofagricultural resources for the median farmer. His analysisconsidered fifteen different activities: four crop rotations; fourfeedlot activities; conventional rearing of lambs; confinementrearing of lambs; cow-calf; cow yearling; pasture finishing ofbeef; farrow to finish swine; and finishing swine. He imposed thefollowing restrictions: 480 acres of cultivated land, financial47capital of $70,000 (minus cost of land), and 780 hours of labourdivided into quarterly periods. In spite of data limitations, hismodel suggested pasture finishing of beef, rather than cropproduction, as the most profitable activity.Livestock was found to be a viable activity that farmers couldincorporate into their enterprises in a farm planning study byAwmack (1974). Graham and Lopez (1976) embodied uncertainty intheir analysis to determine an optimum income plan for the PeaceRiver farmer. They found generally that "a combination of crop andlivestock enterprises are (sic) more stable than an optimum planwhich is diversified in crop enterprises only" (p.96).Graham and Anderson (1982) used LP to formulate a regionaldevelopment model of the Peace River region that focused largely ongrains, forage and livestock. Their study was done in the contextof "developing proposals for a long-term agricultural developmentplan for B.C. agriculture" (ibid., p.1), a project initiated by theB.C. Ministry of Agriculture in 1976. However, their objective wasnot to evaluate sustainable development in the region.The criterion they used to evaluate the alternative strategiesconsidered in the study was net farm income; their objective was tomaximize net farm income. The alternatives they considered werethe following:1) double stockers in the region (from 8,000 to 16,000);2) backgrounding with lower feeder and calf prices;3) double beef cow herd size;4) expand cow herd and cultivated acreage;485) increase land under cultivation by an extra 60,000acres, while other activities are held constant (i.e.,an examination of acreage increases only);6) increase cultivated acreage by an extra 100,000 acres(i.e., examine potential impacts of cultivating moremarginal lands);7) increase the number of yearlings sold off pasture in thefall;8) expand feedlots; and9) increase the dairy herd size.In their analysis, alternative 1 increased net income perrancher by $1,500.00, with a net gain to the region as a whole of$135,000.00 (see Table 4.1). For alternative 2, high prices oflivestock were crucial for profitability; when lower prices wereassumed (relative to 1978 prices), a net loss of $3.2 million wascalculated for the region, as against a profit of $3.5 million forthe base case.The beef sector is a major contributor to the Peace Rivereconomy. Consequently, when beef prices were depressed, the region"suffered and showed a negative aggregate net return for the totalagricultural economy" (Graham and Anderson 1982, p.121). Anincrease in net return from $3.5 million to $4.9 million wasassociated with a doubling of beef herd size. However, thisrequired an additional 48,775 acres of classes 4 and 5 land. (SeeAppendix C for land class description). Furthermore, increased useof 4,147 acres of class 6 land was needed, in addition to moreusage of crown range. Profitability of the livestock sector in theregion means grain farmers may convert their marginal land to49forage and diversification of farms to include livestock may becomean attractive venture; either practice augurs well for sustainableagricultural development in the region.When they simultaneously doubled beef herd and increasedcultivated acreage (alternative 4), the researchers calculated anincrease in net return of over $1.63 million. By assuming anaverage farm size of 1,000 acres per new farmer, 60 new farmerswere capable of achieving a net return of $155,000.^Thisalternative also required bringing classes 4 and 5, and even class6, land into production (see Table 4.1). Nevertheless, Graham andAnderson found this alternative to yield the highest return. Ofcourse, its ecological implications were not taken into account.Alternative 5 resulted in a volume increase of all grains(except oats) and a significant decrease in alfalfa sales, areduction they linked to the imposition of a stagnated beef herdsize. They also found cultivating more marginal land (alternative6) lucrative since the required additional capital investment of $5million had a return of 13 percent. Though not stated, thisalternative's profitability may be partly attributable togovernment programmes (see section 3.4) that support grainproduction on marginal land.Expansion of feedlots (alternative 8) to increase net farmincome in the Peace River region was not profitable according tothe Graham and Anderson study. Alternative 9, on the other hand,had the effect of increasing the net benefit to the region by$102,000 (Table 4.1). However, this led to the reduction in the50Table 4.1 Summary Results of the Graham and Anderson 1982 StudAlternative Increase in Peace River'sPlan Net Income ($'000)^Prereauisite 1^ 135^additional 4,000 acres ofhay land.2^-3,200^low prices of livestockcause loss.3^1,400^increase class 4 and 5land by 48,775 acres;increase class 6 land by4,147 acres.4^1,600^additional 60,000 acres;cultivate land class 6.5 178^increase class 4 and 5land by 25,685 acres;reduction of improved hayland by 34,486 acres.6^259^39,000 acres of class 6land.7 -98^high price structureneeded to expand yearlingproduction.8^-2,260^decline in sales of oats,barley, and alfalfa.9^ 102^double dairy herd size to750 cows.sale of grains (oats and barley) and alfalfa, due to the extra feedrequirements.Though these other studies indicate that sustainable practicesare viable in the Peace River region, none looked at theimplications of farming methods in the area--shaped by government51programmes--for sustainable agricultural development. This is whatthe current study attempts to do.4.3 Structure of Empirical Model The LP model formulated for this study is designed toprimarily examine the effects of government agricultural programmeson land use in the Peace River region which, in turn, will shedsome light on sustainable agriculture in the area.Recall that it was argued in section 1.2 that land use and itsattendant issues, such as land degradation, are essential factorsin a critical analysis of agriculture's sustainability. Indeed, anotable factor in the world's growing sustainability concerns island degradation due, not insignificantly, to "poor" agriculturalland use. Hence, sustainable agriculture necessarily includesalleviating, if not eliminating, land use practices that lead toland degradation including soil erosion rates that are above aregion's T-values (tolerable soil loss). In the Peace River area,for example, the T-value is 2 tons/acre (see section 3.2).The model is a partial equilibrium one, with the coefficientsof the objective function determined exogenously. The commoditiesendogenous in the model are the main crops (wheat, oats, barley,and canola) and forage (fescue) cultivated in the region.Envisaged by the model is a crop farmer who may not produce anylivestock directly. Nevertheless, forage production is undertakenby the farmer as a derived demand by the livestock sector of thePeace River economy, and to contribute to sustainability by52decreasing soil erosion. The other side of the farming coin in thePeace River area are farmers who engage in livestock production,with little or no crop production. This dichotomous farmingpractice accounts for at least 90 percent of the farming activityin the region (Graham and Lopez 1976, p.40).Crop production is undertaken on classes 1 - 5 land. Theland classes were assumed to decline in their capabilities tosupport field crops as the class number increases, akin to the CLIclasses described in Appendix C. Production should take place onlyon classes 1 - 3 land, but government programmes have made landclasses 4 and 5 marginally cultivable. A land class constraint isimposed to ensure that land use does not exceed the availableacreage.Some of the activities in the model were constrained using theguidelines enunciated in Guide to Farm Practice in Saskatchewan(1987). For example, it suggested a lower bound for grains at 60percent of the cultivated acreage, and an upper bound of 30 percentfor summerfallow. Fixed proportions for other activities were alsostipulated. Thus, the model specified at least a joint productactivity, if not a multicrop activity.Crop rotation is a sustainable farm practice that farmersoften follow to control weeds and soil erosion. Since grains havesimilar soil requirements, forages and legumes are usuallycultivated by farmers to enhance soil fertility and/or increaseincome. To reflect this approach to farming, we built proportionalrotations into the model. However, no particular rotation was53built into the model a priori. Any rotation options would bediscerned from the model's results. Underlying this representationis the assumption that acreage available to the rotational crops isnot fixed; for example, unbroken land can be broken to increaseacreage for any or all crops by the farmer.The CWB delivery quota system was reviewed in 1970 and 1978 togive farmers flexibility over land use for their crops and anequitable share of the delivery opportunities. Concomitantly, thiswas to enable the CWB to efficiently bring into country elevatorsat the right time the quantity and quality of all the kinds ofgrain required to compete effectively in the market (Wilson 1979,p.238). Delivery quotas, expressed by CWB as bushels per acre, areincluded in the model using a calculated acreage of totalassignable quota comprised of total acreage of grains (wheat, oats,barley and canola), summerfallow and forage.7 Given the non-emptiness of the constraints, there exists an optimal solution thatis also a global one (Meister, Chen and Heady 1978).4.4 The Basic Empirical Model The model posits the maximization of returns to wheat, oats,barley, canola, forage, summerfallow, and unbroken land for a7 The Canada Grains Council (Statistical Handbook 1985, p.184)indicates that assignable quotas are determined using a 4-partformula as follows: (1) land seeded to wheat, oats, barley, rye,flaxseed and canola; (2) summerfallow; (3) miscellaneous crops; (4)perennial forage (up to 1/3 of the area in the other threeclassifications). Though using this formula, the quota in themodel is defined without (3) as well as rye and flaxseed.54farmer with 1,400 acres comprised of land classes 1 to 5. Forageis assumed as a derived demand by the livestock sector, whileunbroken land allows the phenomenon of new breaking and itsattendant cultivation of marginal land to occur. Production ofgrains on such marginal land facilitates unsustainability ofagriculture in the region. As well, expansion of grains at theexpense of forage (e.g., fescue) in crop rotations enhances soilerosion that adversely impacts on sustainable agriculture. This isa practice that is increasingly becoming a feature of the region(Van Vliet and Hall 1991).Mathematically, the model's objective function can beexpressed as equation (4.5).(4 . 5 )^Max Eiwhere:is the gross margin of ith instrument on jth land class; andthe instruments are wheat, oats, barley, canola, forage,summerfallow and unbroken land.^In other words, the aboveobjective function was specified as follows (equation 4.6):(4 . 6 ) Max Rw1W1+ Rw2W2 + Rw-W-.5 Rw4W4 R w5W5 Ra1A1▪ R a2A2 Ra3A3 Ra4A4 Ra5A5 Rb1B1+ Rb2B2+ Rb3B3 + Rb4B4 + Rb5B5 R c1C1+ R c2C2 R c3C3▪ R04C4 + R c5C5 + R^+ R f2F2 + R f3F3 + R f4F4+ R f5F5+ RsfiSF1 R s f2SF2 R s f3SF3 R5f4SF4+ Rsf5SF5 4- Ru/U1 +Ru2U2 +Ru3U3 +Ru4U4 +Ru5U555where:W = wheat; A = oats; B = barley; C = canola; F = forage;SF = summerfallow; U = unbroken land;R1 .....,R5 = gross margin of wheat on land classes 1 - 5;R0,....,R6 = gross margin of oats on land classes 1 - 5;Rb"... .,/%5 = gross margin of barley on land classes 1 - 5;Rd,....,R6 = gross margin of canola on land classes 1 - 5;Rfl,....,Rf5 = gross margin of forage on land classes 1 - 5;R, .....,R. = gross margin of unbroken land on land classes 1-5;Rsfl I • ' • , Rsf5 = gross margin of summerf allow on land classes 1-5;Wi,....,W5 = wheat on land classes 1 - 5;A1 .....,A5 = oats on land classes 1 - 5;B1,. ..^barley on land classes 1 - 5;.,C5 = canola on land classes 1 - 5;F1 .....,F5 = forage on land classes 1 - 5;..,SF5 = summerfallow on land classes 1 - 5; andU1,...^unbroken land on land classes 1 - 5.This acreage is typical for a representative farmer in thePeace River area, with net income in the $50,000 - $99,999category. There were some 106 such farmers in 1981, while some 75farmers in the net income category of $100,000 - $249,999 had anaverage farm size of over 2,300 acres; another 24 farmers, makingover $250,000 cultivated an average farm size of over 5,000 acres(BCMAFF 1984, p.97). The average farm size in 1981 was about56836.62 acres. In 1986, the corresponding numbers (of wheat farmersalone) were 91, 98, and 44, respectively (Agriculture Canada 1988,p.18).In order to take account of land quality differences in thePeace River area, the 1,400 acres was disaggregated--usingpercentages calculated from Table 3.7--into the following:(a) land class(b) land class(c) land class(d) land class(e) land class(f) land class(g) land classThe acreage of land classes 1 and 4 were entered as equations(4.7) to (4.10) in the model. The remaining land (equation (4.11))was assumed to be class 5.(4.7)^Wl +Al +Bi+Cl+Fl+SFi+U1s 331. 52(4.8)^W2-4- A2 -1- B24- C2 -1- F2+8F2 -1- U2s334 • 88(4.9)^W3A- A3+B3+ C34- F3+SF3 4- U3s 321.721 = 331.52 acres;2 = 334.88 acres;3 = 321.72 acres;4 = 337.96 acres;5 = 60.62 acres;6 = 9.66 acres; and7 = 3.78 acres.57^(4.10)^W4 + A4 + B4 + C4 + F4 + SF4 + U4 s337.96^(4.11)^W5+A5+B5+C5+F5+SF5+U5s74.06The following constraints also were added based on discerniblefarm practices in the area or, in some cases, on the agriculture ofthe Southern Prairies (when Peace River information was notavailable). The Peace River region has similar farm practices asthe prairies (see Chapter 3).Summerfallow is at least 20 percent of the grains cultivated(equation 4.12):(4.12) v-■52.d.i.i SF.-0 . 2 (Wi +Ai+ Bi+ Ci) 0Grains constitute at least 60 percent of total farm acreage(TFA) (equation 4.13).(4.13)^E5 0.4 ( Wi + A i + B i + C2.) - 0 . 6 ( SFi + Fi+Ui) 0Grains produced by Peace River farmers are deliverable to theCWB on quota basis. This works out to be an assignable quota (seesection 4.3) of at least 530 acres in our model (equation 4.14).585 5i Wi +A i +Bi +C2.+F2.-FSF.2 30(4.14)Wheat is at least 25 percent of acreage planted to grain asthe historical acreage shows in Table 3.4 (equation 4.15).(4.15) E5i10.75W1-0.25(A1+B1+C1) 0=Oats acreage is not less than 11 percent of total grainsacreage based on historical data from Table 3.4 (equation 4.16).(4.16)^E5i=i 0.89AI-11 (Wi+Bi+Ci) 0Barley is at least 29 percent of grains acreage (see Table3.4) (equation 4.17).(4.17) E5 0.71Bi-0.29 (Wi+Ai+Ci) 1;)i=1Canola's lowest bound is not less than 24 percent of grainsacreage as shown in Table 3.4 (equation 4.18).(4.18) E5 0.76C1-0.24 (Wi+Ai+Bi) I'J1.1.59Forage is not less than nine percent of grains and forageacreage (equation 4.19).(4.19) E5 0.91FI-0.09 (Wi+ Ai+ Bi+In accordance with farm practices, grains were also specifiedto be at least equal to summerfallow on each land class to ensurethat the latter's acreage does not exceed that of the former.Crop and labour relationships in the Peace River regionsuggest that a quarterly breakdown of demand for labour isreasonable. However, gross margins specified in the objectivefunction make deductions for labour costs; hence, no labourrestraints are imposed on the model.Gross margins (GM), as indicated by BCMAFF, allocates moneyfor interest payments, overhead and other indirect costs as well assome return for the farmer. The GMs per acre in 1989 for wheat,barley, oats, and canola--all on summerfallow--were given bySaskatchewan Agriculture and Food (1989) as $53.64, $84.76, $66.15and $105.18, respectively, while the cost of maintaining an acre ofsummerfallow was stated as $25.82. The GMs for wheat, barley,canola, and oats--on stubble--were $40.52, $53.45, $60.10 and$48.68, respectively. For forage, the BCMAFF's GM for fescue overa 6-year period was converted to a net present value, with adiscount rate of 15%; this gave us a GM of $52.50 per acre.Unbroken land was valued at one-half of the GM for forage on landTable 4.2^Calibrated Results of Land Class Gross Marginof the Model's Instruments ($)Land ClassesInstrument 1 2 3 4 5Wheat on: fallow 53.64 52.84 33.69 21.48 13.70stubble 40.52 39.92 25.45 16.23 10.35Oats on: fallow 66.15 65.17 41.55 26.49 16.89stubble 48.68 47.96 30.58 19.50 12.43Barley on: fallow 84.76 83.50 53.24 33.95 21.65stubble 53.45 52.65 33.57 21.41 13.65Canola on: fallow 105.18 103.61 66.07 42.13 26.86stubble 60.10 59.21 37.75 24.07 15.34Forage 52.50 51.72 32.98 21.03 13.41Unimproved land 26.25 26.25 26.25 26.25 26.25Summerfallow -25.82 -25.82 -25.82 -25.82 -25.8260class one (i.e., $26.25 per acre) (see Weisensel, Rosaasen andSchoney 1991).A function was posited to determine a declining gross margin(as a function of land class) of the instruments in the model,except unimproved and summerfallowed land which were assumed not todecline with land class.8 This function assumed that the above GMswere for land class one. Estimated GMs for other classes of landare presented in Table 4.2.Linear programming's non-negative constraint is representedas equations (4.20) to (4.26).8The following was run using GAUSS to determine gross marginas a function of land class: X=seqa(0.1, 3.0, 10); GMi = Yokexp(-0.15*X); where i = instruments in the model; Y = gross margin ofthe instruments as obtained from Saskatchewan Agriculture and Food,or from BCMAFF; GM = gross margin.61(4.20)^W1 , W2 W3 W4 W5 °( 4 .21)^A1 , A2, A3 , A4 , A5 , O(4.22)^B1, B2, B3, B4, B51(4.23)^C1/ C21 C3/ C41 C51 k0(4.24)^I'1l2/ 31 141 F 5 k 0(4.25)^ST1, ST2, SF3 , SF4 , S F 5 0(4.26)^, U2, U3, U4 , U5 k 0The objective function (4.6) and the constraints (4.7 to 4.26)form the study's Base Case. This is the scenario that depicts thePeace River's current land use practices. It is compared with thefollowing eight simulated scenarios:621) no government subsidy (NG);2) some government subsidy (SG);3) continuous cropping (CC);4) continuous cropping with no government subsidy (CCNG);5) continuous cropping with some government subsidy (CCSG);6) regulation of no summerfallow acreage on land classes 4and 5 (NOSF);7) regulation of no grains acreage on land classes 4 and 5(NOG); and8) imposition of an erosion constraint on the Base Case (ER).The NG case involves the elimination of the average governmentsubsidy per acre paid to Peace River farmers through WGSA, SCGP,Crop Insurance and Chemical Rebates (see Table 3.9), which wascalculated at $20.35 per acre.9 This subsidy rate was subtractedfrom the GMs of the Base Case, except forage from which only athird of this amount was deducted (in accordance with CWB paymentscalculations). Similarly, for SG one-half of the subsidy/acre wasdeducted from all the gross margins, except one-sixth of thesubsidy was deducted from the gross margins for forage.Continuous Cropping (CC) was also simulated. This followssuggestions by soil scientists that CC increases output over time(e.g., Rennie 1986). The objective function coefficients of the CCcase were varied parametrically to obtain two other cases:9It was obtained by summing the average payments of thesegovernment programmes between 1979 and 1989 per grains and forageacreage in the Peace River region in 1986.63continuous cropping with no government subsidy (CCNG) andcontinuous cropping with some government subsidy (CCSG); both wereobtained in the same way as NG and SG, as explained above.Simulations of soil-saving practices were made with two otherscenarios. The first exercise was an imposition of a regulation onthe Base Case disallowing summerfallow on land classes 4 and 5(results under NOSF). The second involved another regulation thatprohibited cultivation of grains on land classes 4 and 5 (resultsunder NOG). These scenarios may spur conversion of marginal lands,that are more vulnerable to land degradation and have been broughtunder grains cultivation, into either forage or unimproved land byperhaps seeding it with permanent cover such as grass. It is worthnoting, as section 3.4 indicated, that Agriculture Canada hasintroduced the Permanent Cover Programme to help farmers convertsome of their land (marginal land) under crop production topermanent cover. This is an effort to enhance sustainableagriculture. However, in this study, financial incentives to seedland to permanent cover are not modelled.In the final simulation, a soil erosion constraint was addedto the Base Case model in order to address the problem of overallsoil erosion in the region. This simulates regulation of soil lossto ensure that farming in the region does not unduly cause soilerosion. We assumed that wheat, oats and canola generate 3tons/acre soil loss, while barley, summerfallow and forage (fescue)cause 3.4 tons/acre, 5 tons/acre and 2.2 tons/acre soil loss,respectively. It was further assumed that the entire farm could64not lose more than 4,200 tons of soil, which supposes that eachacre of the farm generated soil loss of 3 tons/acre. This value isapproximately what is commonly accepted by the American Society ofAgronomy as tolerable soil loss (van Vliet and Hall 1991), which ishigher than the 2 tons/acre envisaged as the Peace River region'stolerable soil loss (see Section 3.2).It must be borne in mind that estimation of individual soilloss is a complex exercise because of factors such as thedifficulty of discerning the effects on current crops from previousagronomic practices. Not surprisingly, there is a dearth ofresearch about soil loss due to particular crops. Thus, the soillosses specified above may or may not be adequate. Be that as itmay, we used the above values to specify a soil erosion constraintfor the entire farm acreage; it is given by equation 4.27 below.(4.27) 3 (W.a / 1+A•+C.) +3 4B +2 2F+5SF1s4,200CHAPTER 5EMPIRICAL RESULTS5.1 Discussion of Base Case ResultsThe simulations were run using an interactive software packagecalled LINDO. The results are presented as Tables 5.1 through5.7 and Figures 5.1 - 5.3. They show acreage cultivated in thePeace River area for: wheat, oats, barley, canola, forage,summerfallow and unbroken land for land classes 1 to 5. The landclasses were assumed to decline in their capabilities to supportfield crops as the class number increases, akin to the CLI classesdescribed in Appendix C. The tables and figures also depict thesimulations discussed in section 4.4.Representation of current summerfallow production in the PeaceRiver area is shown by the Base Case results. Figure 5.1 showsthat canola has the largest acreage (26%), followed by barley(21.6%), wheat (18.6%) and oats (8.2%). This hierarchical order ofacreage is in accordance with years such as 1983 - 85 and 1988 (seeTable 3.4). Acreage for forage constituted some 7%, whilesummerfallow's share was about 15%. Unimproved land had the leastacreage (3.4%).As shown in Table 5.1, the Base Case also indicates thatcanola had the largest proportion (35%) of the total grainsLINDO is an acronym for Linear Interactive DiscreteOptimizer.65^rFig.5.1: Base Case Land UseSummer-fallowUnimprovedLand(3.4%)(14.9%) (18.6°1°) wheat(7.4%)Forage 1•111111•11•1111=11•••••••••••111•••••••111•1■•••••••■(8.2%)OatsCanola(26.0%) (21.6%)Barleyacreage, followed by barley (29%), wheat (25%) and oats (11%). Thetotal acreage of grains accounted for some 74% of the total farmacreage (TFA).Table 5.2 shows that some 13% of Base Case grains arecultivated on land classes 4 and 5. Consequently, an equal amount(13%) of summerfallow takes place on classes 4 and 5 land.Naturally, all unimproved land occurs on land class 5; this is6667Table 5.1^Effects of Government Subsidies on AgriculturalLand Use in the Peace River Re ionTOTAL AS % OF GRAINSBASE^NG^SGTOTAL AS %BASE^NGOF TFASGWheat 25.0^25.0 20.0 18.6 12.0 14.6Oats 11.0^10.0 10.0 8.2 6.0 7.3Barley 29.0^20.0 20.0 21.6 12.0 14.6Canola 35.0^50.0 50.0 26.0 30.0 36.5SF 20.0^20.0 20.0 14.9 12.0 14.6Forage 7.4 5.9 7.2U 3.4 22.1 5.3TOTAL ACREAGEBASE NG SGWheat 260 168 204Oats 115 84 102Barley 302 168 204Canola 364 420 511sub-total 1041 840 1021Forage 103 83 101SF 208 168 204U 48 309 74TFA = Total Farm Acreage; SF = Summerfallow;NG = No Government Subsidy; SG = Some Government Subsidy;U = Unimproved land.about 64% of land class 5's total acreage. The net return for theBase Case was $74,155 (see Table 5.7). This is well within theincome bracket for farmers cultivating such a total farm acreage(TFA) in the region (see section 4.4).Compared with the Base Case, the results for NG (Table 5.1)indicate that, as a percent of TFA, the acreage for wheat, oats,barley, summerfallow and forage all declined: by 35.5%, 26.8%,44.4%, 19.5% and 20.3%, respectively. The acreage for canolaincreased by 15.5% and that of unimproved land increased by morethan five-fold. Similarly, compared with the Base Case, the SGresults (Table 5.1) show analogous directional changes for the68Table 5.2 Effects of Government Subsidies on Land ClassUse in the Peace River Re ion As % of TFA1LAND CLASSES4 52 3WheatBase 0.0 0.0 5.6 12.1 1.0NG 0.0 0.0 8.3 3.7 0.0SG 0.0 0.0 2.5 12.1 0.0OatsBase 0.0 0.0 8.2 0.0 0.0NG 0.0 5.6 0.4 0.0 0.0SG 0.0 0.0 7.3 0.0 0.0BarleyBase 0.0 21.6 0.0 0.0 0.0NG 0.0 12.0 0.0 0.0 0.0SG 0.0 11.1 3.5 0.0 0.0CanolaBase 23.7 2.4 0.0 0.0 0.0NG 23.7 6.3 0.0 0.0 0.0SG 23.7 12.8 0.0 0.0 0.0ForageBase 0.0 0.0 7.4 0.0 0.0NG 0.0 0.0 5.9 0.0 0.0SG 0.0 0.0 7.2 0.0 0.0SummerfallowBase 0.0 0.0 1.9 12.1 1.0NG 0.0 0.0 8.3 3.7 0.0SG 0.0 0.0 2.5 12.1 0.0Unimproved landBase 0.0 0.0 0.0 0.0 3.4NG 0.0 0.0 0.0 16.8 5.3SG 0.0 0.0 0.0 0.0 5.3TFA = Total Farm Acreage; NG = No Government Subsidy;SG = Some Government Subsidy.instruments in the model, albeit not in the same proportion. Forexample, the share for forage, as a percent of TFA, declined byonly 2% while canola's acreage increased by 40%. The sub-total forgrains fell by about 19% under NG and by 2% with SG as Table 5.1indicates.No grains were cultivated on class 5 land under both NG and SGscenarios, as shown by Table 5.2. Under the Base Case, acreage forwheat generally occupied land classes 3 and 5 as did summerfallow,69perhaps suggesting wheat and summerfallow in the same rotation. Inall cases (Base, NG and SG), canola's cultivation is on classes 1and 2 land, while farming of oats and barley occurred on classes 2and 3 land; all of forage's acreage was on class 3 land. Acreagefor unimproved land occupied all class 5 land plus some 70% ofclass 4 land under NG.5.2 Results of Simulations Involving Regulations Regulations that prevent farmers from summerfallowing andcultivating grains on land classes 4 and 5 led to more than five-fold increase in the acreage of unimproved land; acreage for allthe other instruments declined (Table 5.3). Total grain acreagedecreased by about 19% with the former regulation (NOSF) and by 14%with the latter regulation (NOG). Prohibiting grains (NOG) on landclasses 4 and 5 leads to the highest conversion of marginal landsto unimproved land, or permanent cover, and the least summerfallow:6.4% of TFA (see Table 5.3). The decline in summerfallow and theincrease in acreage for unimproved land, as a result of theseregulations, are particularly welcome in the Peace River region dueto the fact that it has been identified as the most erodible regionin B.C. (see section 3.2).70Table 5.3 Effects of Regulations on Land Use in thePeace River RegionTOTAL AS % OF GRAINS^TOTAL ASBASE^NOSF^NOG RASE^NOSF% OF TFANOG ERWheat 25.0^25.0 25.0^18.6^15.0 16.1 17.3Oats 11.0^11.0 11.0 8.2 6.6 7.1 7.6Barley 29.0^29.0 29.0^21.6^17.4 18.6 20.1Canola 35.0^35.0 35.0 26.0 21.0 22.4 24.3SF 20.0^20.0 10.0^14.9^12.0 6.4 13.9Forage 7.4 5.9 6.4 6.9U 3.4^22.1 23.1 10.1TOTAL ACREAGEBASE^NOSF^NOG ERWheat 260^210^225 242Oats 115 92 99 106Barley 302^244^260 281Canola 364 294 314 340sub-total 1041^840^898 969Forage 103 83 89 96SF 208^168^90 194U 48 309 323 141TFA = Total Farm Acreage; SF = Summerfallow; U = Unimproved land;NOSF = No Summerfallow on Land Classes 4 - 5;NOG = No Grains on Land Classes 4 - 5; ER = Erosion Regulation.Agriculture Canada's Prairie Farm RehabilitationAdministration (PFRA) suggests that 160 acres converted frommarginal lands to permanent cover provides pasture that can feed100 beef cows. Since this pasture will last only three to fivemonths, there will be a need to conserve feed which may have to beproduced from land classes 5 and 6.This means that NOG's conversion of 275 acres to unimprovedland (see Table 5.3) can help feed about 172 cows, with possiblefurther grazing on class 6 rangelands. For the entire region,25,025 acres can be converted to unimproved land and help feed15,640 beef cows. This will be a boost to the region's livestockindustry--which Graham and Anderson found to be a major contributorto the Peace River economy (see section 4.2)-- and, perhaps, farm71diversification. This is not done voluntarily, perhaps, becausegovernment payments skew production costs against diversifyinggrain farms to include livestock; farmers may also not likeinconvenience of cattle.As well, the NOG scenario indicates that there could be a 3.8%increase in farmers' income (Table 5.7). Under the NOSF scenario,however, income falls by about 12% (see Table 5.7), but the long-term benefits from reduced erosion and the attendant sustainabilityof farms will more than likely compensate farmers even ifgovernments do not offer financial incentives to spur the necessarychange in farming practices. It is worth noting that on-farm costsof water erosion alone in the region were estimated to be about $9million in 1984 (noted by Van Vliet and Hall 1991, p.2).Another effect of NOG and NOSF is that forages acreage shiftsfrom class 3 land to class 4 land and, with all the unimproved landaccumulated on classes 4 and 5 land (see Table 5.4), sustainableagriculture seems to be better enhanced with these types ofregulations. Summerfallow acreage switches to land classes 1 and3 under the NOG and NOSF scenarios, with the latter appearing toestablish some rotational relationship between barley andsummerfallow on class 2 land. However, NOSF still leaves about2.4% of the total grain acreage on land class 4 (see Table 5.4below).Table 5.4 Effects of Regulations on Land Class Use in thePeace River Re ion As % of TFA1LAND CLASSES4 52 3WheatBase 0.0 0.0 5.6 12.1 1.0NOSF 0.0 0.0 13.6 1.4 0.0NOG 0.0 0.0 16.0 0.0 0.0ER 0.0 0.0 7.6 9.7 0.0OatsBase 0.0 0.0 8.2 0.0 0.0NOSF 0.0 0.0 6.6 0.0 0.0NOG 0.0 6.5 0.5 0.0 0.0ER 0.0 3.3 4.3 0.0 0.0BarleyBase 0.0 21.6 0.0 0.0 0.0NOSF 2.6 12.0 2.8 0.0 0.0NOG 0.0 1.2 17.4 0.0 0.0ER 0.0 20.1 0.0 0.0 0.0CanolaBase 23.7 2.4 0.0 0.0 0.0NOSF 21.0 0.0 0.0 0.0 0.0NOG 22.5 0.0 0.0 0.0 0.0ER 23.7 0.6 0.0 0.0 0.0ForageBase 0.0 0.0 7.4 0.0 0.0NOSF 0.0 0.0 0.0 5.9 0.0NOG 0.0 0.0 0.0 6.4 0.0ER 0.0 0.0 6.9 0.0 0.0Summerf allowBase 0.0 0.0 1.9 12.1 1.0NOSF 0.0 12.0 0.0 0.0 0.0NOG 0.0 0.0 6.4 0.0 0.0ER 0.0 0.0 4.2 9.6 0.0Unimproved landBase 0.0 0.0 0.0 0.0 3.4NOSF 0.0 0.0 0.0 16.8 5.3NOG 0.0 0.0 0.0 17.8 5.3ER 0.0 0.0 0.0 4.8 5.3TFA = Total Farm Acreage;NOSF = No Summerfallow on Land Classes 4 - 5;NOG = No Grains on Land Classes 4 - 5;ER = Erosion Regulation.7273As a result of the erosion regulation (ER) imposed on the BaseCase model, acreage for all the grain crops and summerfallowdeclined by an average of 7% and 6.7%, respectively. Acreage forforages also decreased by 6.8%, but that of unimproved landincreased by about 193.8% (Table 5.3). This means ensuring thatsoil loss in the region was tolerable resulted in perhaps the mosterodible lands under cultivation being converted to permanentcover. Thus, there is more pasture to support the region'slivestock industry, and yet less need for forages.Net returns fell by 1.1% with the erosion regulation (Table5.7). However, less erosion also means a reduction in the costsassociated with erosion; this can compensate farmers in addition tomaking the farms environmentally sustainable. Note that, aspointed out earlier, on-farm costs of water erosion alone in theregion were estimated at $9 million in 1984.Comparison of all six scenarios (Base Case, NG, SG, NOG, NOSFand ER) shows that canola had the highest acreage under the firstthree cases (see Figure 5.2). Unimproved land had the highestacreage under NOG and NOSF, but it accounted for the least acreageunder the Base Case.In summary, the simulations indicate that farmers can reducethe amount of marginal land under cultivation and fostersustainable agriculture. Regulations such as disallowing grains onpoorer lands were found to decrease acreage for summerfallow andgrains, which various government programmes had encouraged onmarginal lands in the first place, in favour of unimproved land./400-0a)oi- 300-