and Business, University of Guelph, Guelph, Ontario. 2Former research associate, Department of Crop Science,. 3Former rese. University of Guelph, Guelph, ...
Table of Contents
A Case Study Approach to Comparing Weed Management Strategies under Alternative Farming Systems in Ontario D. Peter Stonehouse,l
S. F. Weise,2 T. Sheardown, and C. J. Swanton
R. S. Gill3
lAssociate prqfessor, Department ofAgricultural Economics and Business, University of Guelph, Guelph, Ontario. 2Former research associate, Department of Crop Science, University of Guelph, Guelph, Ontario. 3Former rese arch associate, Department ofAgricultural Economics and Business, University of Guelph, Guelph, Ontario. 4Associate p rofe ssor, Department of Crop Science, University of Guelph, Guelph, Ontario. Received 12 January 1994, accepted 14 December
1995
When research was initiated into comparing alternative method‘s of managing weeds in Ontario ‘s major field cash crops, nofield trial data existed. Twenty-five farmers were therefore surveyed for their production data on corn, beans andfall-seeded cereal grains, including weed management practices, input costs and wage rates, yields and product prices. Nine farmers were class$ed “conventional ” because of their heavy dependence on synthetic herbicides, which were routinely broadcast on the three focus crops. Nine farmers were classiJied as “reduced input ” tfthey placed reduce dependence on herbicides for at least one of the focus crops. Seven organic farmers placed zero reliance on herbicides, using instead substitutes such as crop rotations, smother crops, soil tillage and timeliness offield operations. Although organic farmers spent the most time and money on weed control, their overall direct costs of production were lowest for all three focus crops. Crop gross margins were highest on organic.farms, partly because of lower production costs, but also because of higher product prices along with comparable crop yields. Linear programming model results jtir whole-farm analyses revealed highest netfarm incomes on organic farms and lowest on conventional farms, in part due to lower overhead costs on organic farms, and in part due to greater enterprise divers@ation and to greater self-sufficiency in material inputs. These case stud)) results need broader-scale testing to venfi the conclusion that organic or reduced-input methods of weed management offer viable alternatives to conventional approaches. Avant cette recherche, il n ‘existait pas de don&es concretes pour effectuer une comparaison des dif ferentes methodes de controle des mauvaises herbes au sein des recoltes a benejkes majeures de 1 ‘Ontario. 051sondage sur 25 fermiers a done ete eflectue pour determiner leur production de mai’s, de haricots et de grains cerealiers d hutomne ainsi que leurs habitudes de controle des mauvaises herbes, leurs cotits d ‘entree et de gages, leurs rendements et leurs prtk de produits. Neuf cultivateurs ont Pte classes R conventionnels )) puisqu ‘ils dkpendaient fortement sur les herbicides de synthPse r&ulit+ement repartis sur les trois rtkoltes cibles. Neuf cultivateurs ont Ptk class& comme (t entrees-reduites )) car ils dependaient moins sur les herbicides pour au mains une des rtkoltes cibles. Sept agriculteurs de produits organiques ne se jiaient pas aux herbicides et utilisaient a la place, des substituts comme la rotation de la recolte; des recoltes etouflantes; la culture de la ten-e et 1‘opportunite des opkrations du terrain. Bien que les agriculteurs de produits organiques passaient le plus de temps et depensaient le plus dhrgent sur le contrGie des mauvaises herbes, 1‘ensemble de leurs touts directs de production etait le plus bas pour chaque recolte. La marge brute des recoltes etait plus elevee pour les fermes organiques, en partie grace aux bas cotits de production, tnais aussi grcice aux pti de produits plus eleves tout en ayant des rendements comparables. Les resultats du modeVe de programmation lineaire pour la totalite de la ferme ont revele que les fermes organiques avaient g&G-e les revenus les plus Pleves tandis que les fermes conventionnelles avaient g&t&e les revenus les plus bas en partie a cause dune diversification accrue des entreprises et de I’auto sufisance accrue en materiaux d’entrees. Les resultats de cette etude de cas ont besoin de preuves plus dktaillees pour vPrt$er la conclusion que les methodes organiques ou a entrees-reduites de contrble des mauvaises herbes sont une alternative viable voire meilleure que les methodes conventionnelle. Canadian Journal qf’Agricultura1Economics 44 (1996) 81-99 81
82
CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS
THE PROBLEM SETTING AND OBJECTIVES OF STUDY In response to growing public and farming sector concerns about sustainability, resource degradation and environmental damage, many new policies and programs with an environmental orientation have been introduced by all levels of Canadian government, especially since the early 1980s (Stonehouse and Bohl 1990; Stonehouse 1994). These policies and programs rely heavily on education, extension, demonstrations, technology transfer and financial incentives, in keeping with a well-established voluntary compliance approach to solving conservation problems in agriculture (van Vuuren and Stonehouse 1995). One such program launched in 1988 by the Ontario government, “Food Systems 2002 Pest titled Management Research Program (198893),” had the stated aim of reducing the use of chemical pesticides in Ontario agriculture by one-half by the year 2002. Synthetic herbicides were estimated to be contributing over 60% of total pesticides usage in Ontario agriculture in the 1980s (Moxley 1989). Factors that concern farmers who might otherwise be willing to reduce or even eliminate chemical herbicides for control of weeds in field crops include: . the extent to which such a shift might reduce crop yields and/or increase variability in crop yields; and . the effects on farm business revenues, production costs and profits, including variability and risk effects. What seems to be encouraging farmers to reduce or eliminate synthetic pesticides is the perceived potential for positive effects on the environment, in terms of improvements to land, air and water quality, wildlife habitats, biodiversity, food safety and human health. Whether these perceptions of farmers are well-founded or not is not the focus of this study. Suffice it to say here that the scientific literature, while continuing to recognize the merits of synthetic pesticides in economically expanding food output, is increasingly acknowledging some detrimental sideeffects on the environment, ecology and
human health (see, for example, Edwards et al 1990; Fox et al 199 1; Pimentel and Lehman 1993; Paoletti et al 1993). With increasing emphasis being placed on resource management and environmental protection, there is an emerging interest in finding ways to farm profitably without being quite so heavily dependent upon chemical inputs. Given the importance of productivity and economic viability to farmers in Ontario, any search for alternative farming systems that are less heavily dependent on chemicals must address their effects on yield response rates and profitability. Farming systems is defined here as the conceptual and practical approach to crop and animal production in general, and to weed control in major field crops in particular. Many alternative farming systems using different weed management techniques can be envisaged. These techniques may range from heavy dependence on chemical herbicides to reduced reliance on chemical herbicides to zero-herbicide approaches, which rely on substitutes for herbicides. Examples of substitutes for synthetic herbicides include mechanical tillage operations, sequences of crops in rotations, use of “smother” crops such as rye or buckwheat, timeliness of field cropping operations, scouting and handweeding, and cornposting of livestock manures to destroy weed seeds. Each weed management system has different on-farm and off-farm implications and consequences. Conventional systems based on Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA) herbicide recommendations may be extremely effective in controlling weeds and in achieving high crop yields, but may also result in trace residues in foods or in groundwater or downstream watercourses. A number of studies have undertaken comparisons between conventional and reduced-herbicide approaches to weed control. It has been demonstrated that equal or superior results, mostly in terms of profitability, can be achieved with reduced rates of herbicide applications (e.g., Musser et al 198 1; Zavaleta et al 1984; Lybecker et al 1984; Colvin et al 1986; Lybecker et al
COMPARING
WEED
MANAGEMENT
1988; Baldwin et al 1988). These so-called “reduced herbicide input” systems (subsequently referred to as reduced-input systems) may reduce pollution costs and health hazards to farm operators (Madden 1988). Reduced-input systems, employing both herbicides and cultivations to control weeds, are examples of flexible strategies that have generally been found to produce less variation in returns over an assumed range of input and product prices (e.g., Snipes et al 1984; Wilcut et al 1987a; Wilcut et al 1987b; Monks 1989). Flexible strategies in this weed control context are defined as those allowing for different combinations of herbicides and cultivations or other herbicide substitutes to be used according to such variables as weed types and densities, crop growth stage and condition, and soil conditions. Organic systems in general eliminate all synthetic inputs, such as fertilizers, pesticides, feed additives, growth promotants, antibiotics, etc., from use on the farm. In particular, for weed management purposes, organic systems do not use any synthetic herbicides. Organic systems must therefore rely to an even greater extent than reduced-input farmers on crop rotations, smother crops, tillage and timeliness of field operations and other management procedures for weed control. Organic systems eliminate the potential for ground and surface-water pollution and health hazards from synthetic herbicides because such inputs are proscribed. Inconsistent results have been obtained from two-way (organic versus conventional) or three-way (organic versus reduced-input versus conventional) farming systems comparisons. A number of studies have concluded that reduced-input systems based on integrated pest management and organic (or zero-herbicide) approaches to farming and weed management appear to be superior to conventional methods (e.g., Bridges and Walker 1987; Goldstein and Young 1987; King et al 1986). Other studies have reported exactly the opposite (e.g., Berardi 1978). Some studies have reported nonstatistically significant differences among alternative methods (e.g., Sahs et al 1989; Helmers et al
STRATEGIES
83
1986). There have been no studies so far in Ontario providing empirically based comparisons across alternative farming systems, either in general technical and economic terms or for weed control efficacy in particular. The main objective of this study is to provide a case study basis for making comparisons of weed management strategies across alternative farming systems in order to furnish some preliminary results under Ontario conditions for interested farmers and others. Specifically the objectives are to: . compile resource usage, crop yield and financial data for weed management in Ontario’s principal cash field crops (grain corn, beans and fall cereal grains), for three alternative farming systems representing different levels of dependence on herbicides for weed control (i.e., conventional, reduced-input, and organic); and . evaluate these data from the technical and economic viewpoints, so as to provide comparisons of crop yields, resource use rates, productivities and profitability across systems. It is hypothesized, by extrapolation from previous research work elsewhere, first that organic systems would use resource inputs at the lowest rate (with the notable exception of labor) and conventional systems at the highest rate. Second, it is hypothesized that output rates and gross revenues would be comparable for conventional and reducedinput systems, but lower for organic farms. It is hypothesized, third, that profitability levels would be highest for reduced-input systems, based on comparable output rates but lower input rates relative to conventional systems. Fourth, it is hypothesized that profitability levels would be lowest for organic farms, for which lower input costs would not be sufficient to compensate for lower output rates and revenues relative to the other two systems. This study is designed to provide case study documentation of actual performance levels achieved by Ontario farmers using alternative weed management approaches. Anticipated benefits from providing some preliminary comparative results include:
CANADIAN
JOURNAL OF AGRICULTURAL
assisting individual farm decision makers interested in alternatives to conventional approaches to weed management in major cash field crops; expanding the core of extension advisory information for use by OMAFRA Representative Branch staff and others responsible for assisting and guiding farmers on weed management; and assisting those responsible for formulating new policies or modifying existing policies, the better to address the joint concerns of farm business viability, environmental protection and farmer/farm worker safety. RESEARCH
METHODOLOGY
In the absence of any secondary data for farming systems comparisons under Ontario conditions at the time of initiating this study in 1989, reliance is placed on primary data It is fully recognized that an sources. approach using commercial farm data from three alternative farming systems may not provide completely objective results because of differences among systems, and among farms within systems, in: . natural resource endowments (climate, soil, topography, etc.); . combinations of crop and livestock enterprises; . size and scale of operation, degree of and other economic indebtedness, aspects; and human capital endowments, in terms of information available and uses, and management skills and abilities. In particular, it is recognized that all organic farms would have livestock enterprises as an integral part of a mixed farming system designed to minimize dependence on imported synthetically produced inputs (e.g., agrochemical pesticides, fertilizers, animal feeds, feed additives, etc.). In contrast, it is expected that at least some of the conventional and reduced-input farms would have no livestock enterprises, and that most would have fewer enterprises than the organic farms. In addition, it is expected that all conl
ECONOMICS
ventional and reduced-input farms would rely more or less heavily on synthetically produced inputs from off the farm. In order to reduce some of the more important differences, the selection of farms is based the following criteria: at least 35 hectares tillable land (owned or rented, not necessarily all devoted to cash crops); designed to eliminate small, part-time or hobby farms; . at least 2500 corn heat units, 1 but not more than 2900 corn heat units; designed to narrow the range of climatic variability across farms and farming systems; and . some tillable land allocated to at least one of grain corn, beans (e.g., soybeans, white beans, kidney beans), and fall cereal grains (e.g., winter wheat, rye, spelt); designed to ensure that each farm, regardless of systems approach, grew one or more of the focus crops for this study, so that direct weed management and cost comparisons could be made. In addition, the selection of farms is based on their being able to provide at least five years of data related to crop production and weed control procedures, including resource usage rates and costs, and crop yields and prices. Twenty-seven cooperators were selected for the analysis. Nine conventional and nine reduced-input farmers from the centralsouthwestern district of Ontario were selected with the assistance of OMAFRA Representative Branch staff. Reduced-input farmers were distinguished from conventional farmers by herbicide usage rates. Farmers applying herbicides at recommended OMAFRA rates were classified as conventional. Those using herbicides at less than OMAFRA-recommended rates were defined as reduced-input farmers. Nine organic farmers were selected with the assistance of the Organic Crop Improvement Association and the Ecological Farmers Association of Ontario. It should be noted that relatively few farmers used organic methods in Ontario (less than 1% of the 70,000 or so farmers in 1989), that even l
COMPARING WEED MANAGEMENT STRATEGIES
fewer organic farmers met the selection criteria above and that not all of the organic farmers who were approached were willing to participate in the study. This resulted in only a small sample of nine organic farmer participants. For reasons of consistency, only nine conventional and nine reduced-input farmers were selected. There is no knowledge of whether or to what extent each sample is representative of its constituent part of the Ontario farming population. Data were collected through an in-depth questionnaire on an on-farm interview with all nine conventional farmers, and all nine reduced-input farmers, but with only seven organic farmers. The remaining two organic farmers withdrew because of time constraints. In order to concentrate attention on weed management aspects of the farm business, each farm’s analysis was based on detailed specifications for each of the three focus crops for all production processes, and especially the labor, machinery and materials inputs, and the crop yields associated with the requisite weed management approach. An enterprise budget approach was used to determine the gross margins (gross revenues less total direct production costs) per hectare for each of the three focus crops, and averaged for each of the three farming systems. Intersystems comparisons were then performed. More comprehensive comparisons were then made, using whole farm analysis. In this study, linear programming (LP) techniques were used to determine how net farm incomes compared across farming systems. LP models were developed for four farms in each farming systems alternative, the four farms being randomly selected from each parent subgroup.” The whole farm LP model in each of the 12 cases was used to simulate the 1989 actual level of net farm income (net returns to equity capital, risk and management). Crop and livestock enterprise activity levels were constrained to their actual 1989 levels, as were labor requirements and enterprise allocations; capital equipment needs, enterprise allocations and costs; and production operating costs for the farming system used (1989 was the most recent year for
85
which full data were available). LP techniques were selected because of their ability to identify and estimate the opportunity costs associated with constraining crop and livestock production and marketing activities to the 1989 actual levels. The underlying presumption was made that the 1989 simulated outcomes for each farm did not necessarily represent a net farm income maximum outcome (i.e., a comer point solution). In that case, shadow values would be generated by the LP model. Shadow values would reflect (as dual prices) the values of marginal products attached to any constrained resources or (as reduced costs) the marginal net farm income changes to the objective function value attached to constrained activities. In particular, any positive reduced costs would indicate the per-unit activity contributions to be made to net farm income by profitable constrained activities. In contrast, negative reduced costs would indicate detractions from net farm income occasioned by unprofitable activities forced into the LP solution at minimum levels. The objective function for the LP models was set to maximize total net farm income (gross margins from focus crop plus other crop and any livestock enterprises less farm overhead costs). Since overhead expenses included opportunity cost returns to operator and any unpaid family labor, net farm income represented residual returns to owner’s equity capital invested in real estate, management and risk taking. The columns in the model mainly comprise the individual crop production, crop marketing, and livestock production and marketing activities (Figure 1). Each crop and livestock activity reflects usage of land, labor and machinery resources. Crop production activities are linked to crop marketing or livestock production activities through crop transfer equations. Livestock feed requirements are delineated through the technical input-output coefficients along the crop transfer equations. Overhead expenses for the farm are not allocated across enterprises, but are combined into a single “pay overhead costs” activity (Figure 1). Capital investment
Objective Function
Crop 1 Production
Crop 2 Production
Crop 1 Sales
Crop 2 Sales
Livestock Production and Sales
Direct Costs/ha
Direct Costs/ha
Sales Price/tonne
Sales Price/tonne
Gross Margin/head
Righthand Sides
Pay Overhead costs N
Max. net farm income
5
b hectares
Land Constraint
1
1
Labor, l-15 April 16-30 April
a a
a a
a
a
b hours
a
a
b hours
16-31 Ott 1 Nov-31 Mar Machinery,
l-15 Apr
b hours b hours
a a
I 16-31 Ott
a
Crop Transfer, Crop 1 Crop 2
a
a
-a
a a
-a
I
0 tonnes 0 tonnes
Crop Constraint, Crop 1 Crop 2 Livestock Constraint
1
= =
b hectares b hectares
=
b head
Pay Overhead Constraint
b $ costs
Figure 1. Generalized structure of farm LP models for 1989 simulations. The a values represent technical input-output coefficients. sent constrained resource availability or activity level coefficients.
^_._
_^ .l^l.- ” I
._.^l_-..-_l..*-l.-_ - -__I_.-- -
b hours
_“..__ - -..-^- ._“.
__^.-_~s.-___._ _ “~l._-._^.l_.lll-“-_l
~
The b values repre-
._ -1-
-
.---
”
-
COMPARING
WEED
MANAGEMENT
Table 1. Land base and focus crop hectarages of participating Conventional
87
STRATEGIES
farms, average by farming system, 1989 Reduced input
Organic
9
9
7
Total land base (av. ha) Standard deviation Coefficient of variation (%)
260.70 122.87 47.13
214.86 92.03 42.83
151.11 92.3 1 61.09
Tillable land base (av. ha) Standard deviation Coefficient of variation (%)
244.72 132.16 54.00
195.14 92.82 47.57
119.89 78.44 65.43
Focus crop hectarages: Grain corn (av. ha) Standard deviation Coefficient of variation (%)
99.10 29.65 29.92
65.89 40.66 61.71
7.47 10.06 134.67
Beans (av. ha) Standard deviation Coefficient of variation (%)
47.36 31.19 65.86
50.81 36.52 71.88
2.31 3.18 137.66
Fall cereal grains (av. ha) Standard deviation Coefficient of Variation (%)
42.78 30.47 71.22
31.19 24.23 77.69
29.70 30.70 103.36
77.3%
76%
32.9%
Number of farms
Proportion of tillable land under focus crops
costs for machinery and equipment are incorporated in the overhead expenses through the use of annual amortization costs, as proxies for depreciation and interest on capital invested. Additional ownership costs for machinery are for repairs and insurance, and a share of any equipment housing costs. Other overhead expenses include those for interest paid on loans, general insurance, banking, accounting and legal fees, property taxes, rental charges, utilities costs and miscellaneous overheads. Six categories of rows represent land, labor and machinery resource constraints; crop yield transfers from production to livestock feeding or crop marketing activities; crop areas and livestock numbers constraints; and payment of overhead costs. Labor and machinery constraints are specified for each two-week period during the growing season (April to October), and for each operation of the cropping activities, in order to identify any
time periods with labor and/or machinery time shortages and their shadow values per hour. EMPIRICAL RESULTS DISCUSSION
AND
Comparison of General Characteristics Conventional farms have the largest average land base, at 260.70 ha per farm, followed by reduced-input farms at 2 14.86 ha/farm, and organic farms with the least at 15 1.11 ha/farm (Table 1). Standard deviations and coefficients of variation indicate a somewhat higher degree of variability for the organic farm sample. This does not necessarily signify anything important, because of lack of information about the representativeness of the sample farms. Both conventional and reducedinput systems record between 90% and 95% on average of their total land bases as tillable land, considerably greater than the 79% recorded for organic farms
88
CANADIAN
JOURNAL
OF AGRICULTURAL
The focus crops comprise a much higher proportion of tillable land base on both conventional and reduced-input farms than on organic farms. The higher coefficients of variation for organic systems stem directly from a majority of organic farms reporting zero tillable land allocated to grain corn or bean crops, as well as from wide variations in hectarages allocated among farms that did report either of these focus crops. The explanation proffered by organic growers for low or zero hectarages of corn and bean crops was lack of markets for organically produced corn and beans (i.e., sales outlets could not be found). In contrast, organic farms typically carried a much wider range of crops, had greater allocations of land to pasture and hay crops that were used mainly for supplying feed to their livestock, and relied for the majority of their revenues on livestock enterprises rather than on cash cropping. This may have reflected any one or all of an approach to weed management that relied on crop rotations, cover crops, cultivations, timeliness of field operations, etc., as substitutes for synthetic herbicide methods; a response by organic farmers to niche market opportunities in specialty crop areas; the overall philosophy of organic farmers, to rely on mixed livestock and cropping enterprises, and an emphasis on self-sufficiency in seeds, plant nutrients, livestock feeds and replacement livestock; or a response by organic farmers to inherent risks in farming by depending heavily on diversification of resource use and product lines. The diversified approach to land use by organic farmers and the small hectarages of winter wheat on organic farms led to the perceived need to combine all fall-seeded cereal grain crops together on these farms. It is recognized that none of the conventional or reduced-input farmers grew any spelt or rye, so that winter wheat on these farms is being compared with a range of cereal crops on l
l
l
l
ECONOMICS
organic farms. It is furthermore recognized that yields for rye and spelt are generally expected to be lower than those for wheat, and that this could have negatively influenced the outcomes on organic farms. This may have led, in turn, to a bias against organic systems in making comparisons at the focus crop enterprise level. It is, however, felt important to make the comparisons to reveal differences across weed management systems in cultural practices and production costs for a group of crops all seeded at the same time of the year. Focus Crop Enterprise Comparisons Several differences among farming systems for weed control methods should be noted (Table 2). Organic farmers relied more heavily on cultivations and hand weeding than the other two types of farmers. Organic farmers were the only ones to compost manure (with the farmer-stated objective of destroying disease organisms as well as weed seeds3); the other two types of farmers applied untreated manure. Conventional and reduced-input famlers used herbicides, while organic farmers applied none. Organic farmers committed more labor resources to weed control than the other two types of farmers. This is reflected in higher labor costs on organic farms. For all farming systems, unpaid operator and family labor costs are included, using an opportunity cost-based method of valuing unpaid labor. In each farm’s case, the farmer identified the opportunity cost value of unpaid labor. Higher total weed control costs, both absolutely and as a percentage of total direct production costs, reflect the heavier applications of resources to weed control on the organic farms, despite the zero expenditures on herbicides. In contrast, other direct production costs were lower on organic farms, due mainly to lower seed costs, and to zero fertilizer and (nonherbicide) synthetic pesticide materials and applications costs. This renders overall direct production costs, including total costs for all (paid and unpaid) labor, lowest on organic farms for all three focus crops and highest on conventional farms (Tables 3 to 5).
Table 2. Weed control methods, and labor requirement
and cost averages, by farming system and by focus crop, 1989 Corn
Conventional Number of farms Pre-plant cultivationsa ($/ha) Post-plant cultivationsa ($/ha) Manure compostingb ($/ha) HerbicidesC ($/ha) Weed scoutmg ($/ha) Hand weeding ($/ha) Total weed control costsd ($/ha) Total weed control labor (hour/ha) Other direct production costsd ($/ha) Total direct production costse ($/ha) Total weed control costs as %
Reduced input
9 34.06 6.55
9 30.98 14.62
55.04 3.92
51.37 4.35 0.47 101.70 2.07 319.18 420.88 24.16
99.57 2.08 389.60 489. I7 20.35
Fall cereal grains
Beans Organic 4 45.71 53.93 20.00 2.32 1.49 123.45 6.21 180.67 304.12 40.58
Conventional 33.22 5.06 58.96 3.17 100.41 1,75 219.93 320.34 31.34
Reduced input
Organic
8 34.79 21.18
3 48.60 54.72
9 22.41 0.73
8 6.19 9.94 -
1.84 14.85 120.01 6.20 188.16 308.17 38.94
4.52 1.33 0.55 29.54 0.91 308.90 338.44 8.73
9.84 3.00
56.79 3.09 0.47 116.32 1.61 200.7 1 3 17.03 36.69
Conventional
28.97 0.64 239.56 268.53 10.79
a Includes machinery operating costs and labor costs; 100% of cultivation costs attributed to weed control. b 100% of manure cornposting costs attributed to weed control. c Includes herbicide materials costs, machinery and labor application costs. d Includes all materials, labor, and machinery operating costs. e For total direct production costs, see Tables 3 to 5 below.
”
Reduced input
“”
Organic 6 45.92 1.34
2.74 2.80 52.80 2.37 207.83 260.63 20.25
90
CANADIAN JOURNAL OF AGRICULTURAL
ECONOMICS
Table 3. Summary: Grain corn enterprise budget data, averages by farming system Conventional Number of farms
Reduced input
Organic
Organicd
9
9
4
4
Five-year av. yield (t/ha)” Standard deviation (t/ha) Coefficient of variation (%)
6.32 1.32 20.89
7.13 1.36 19.07
6.62 1.30 19.64
6.62 1.30 19.64
Five-year av. price ($/t)a Standard deviation (s/t) Coefficient of variation (%)
119.21 19.59 16.43
121.29 20.62 17.00
146.70 32.75 22.32
120.25 -
Five-year av. gross revenue ($/ha) 1989 av. total direct production costs ($/ha)b 1989 standard deviation ($/ha) Coefficient of variation (%)
753.4 1 489.17
864.80 420.88
974.15 304.07
796.06 304.07
110.44 22.58
89.53 21.27
42.19 13.88
42.19 13.88
Five-year av. gross margin ($/ha)c Standard deviation ($/ha) Coefficient of variation (%)
264.24 219.94 83.23
443.92 239.25 53.90
667.08 325.29 48.78
491.99 -
a Significantly different according to Kruskal-Wallis test, a = 0.05. b Includes all labor costs, with unpaid operator and family labor valued at its opportunity cost. c Five-year av. gross margin = five-year av. yield * five-year av. price less 1989 av. total direct production costs. d Gross revenue and gross margin calculated using mean prices for conventional and reduced-input systems.
Crop yield differences are not great among the three farming systems, although these differences are revealed to be significant for all three focus crops at the 5% level of confidence according to the Kruskal-Wallis test for completely randomized designs where data are not normally distributed (Keller et al 1994). Reduced-input farms achieved the highest average yields for corn (Table 3) and cereal grains (Table 5), while organic farms achieved the highest yields for beans (Table 4). With reasonably comparable yields, but with premium prices for products and lower total direct production costs, organic farms received highest gross margins for all three focus crops. Product prices are statistically significantly different, at the 5% level, for all three focus crops according to the Kruskal-Wallis test. This suggests rejection of the second hypothesis: in particular, organic system revenue levels are not lower than those for other systems.
It needs to be emphasized that the lower production costs are far more critical than the higher product prices for two of the three focus crops. With premiums on product prices eliminated (column 4, Tables 3 to 5), gross margins on organic farms would have averaged $49 1.99/ha for corn, $527.53/ha for beans, and $170.05/ha for cereal grains. Organic farms would therefore have continued to obtain highest gross margins for grain corn and bean crops, but no longer for fall cereal grain crops. Fall cereal grain crops were largely exported by organic farmers to European markets, mainly as spelt and rye at considerable price premiums relative to winter wheat prices. Recall that organically grown spelt and rye are expected to yield considerably less than winter wheat. With production costs for fall cereal grain crops almost the same on organic and reducedinput farms, the premium prices for organic
COMPARING WEED MANAGEMENT STRATEGIES
91
Table 4. Summary: Bean enterprise budget data, averages by farming system Conventional Number of farms
8
Reduced input 8
Organic 3
Organic d 3
Five-year av. yield (t/ha)” Standard deviation (t/ha) Coefficient of variation (%)
2.37 0.42 17.72
2.50 0.53 21.20
3.03 0.69 22.77
3.03 0.69 22.77
Five-year av. price ($/t)a Standard deviation (s/t) Coefficient of variation (%)
272.69 54.30 19.91
278.93 33.56 12.03
289.6 39.89 13.77
275.81 -
Five-year av. gross revenue ($/ha) 1989 av. total direct
646.27 320.34
697.33 3 17.03
877.49 308.17
835.70 308.17
53.54 16.71
74.11 23.38
45.86 14.88
45.86 14.88
325.94 202.14 62.02
380.30 172.92 45.47
569.32 166.82 29.30
527.53 -
production costs ($/ha)b 1989 standard deviation ($/ha) Coefficient of variation (%) Five-year av. gross margin ($ha)c Standard deviation ($/ha) Coefficient of variation (%)
a Significantly different according to Kruskal-Wallis test, a = 0.05. b Includes all labor costs, with unpaid operator and family labor valued at its opportunity cost, c Five-year av. gross margin = five-year av. yield * five-year av. price less 1989 av. total direct pro-
duction costs. d Gross revenue and gross margin calculated using mean prices for conventional and reduced-input systems. products became a more critical factor for this focus crop than the lower costs of production. The above findings showing lower resource use rates and production costs for only some crops on organic farms provide only guarded support for the first hypothesis. In particular, it is not clear whether labor wage rates and costs were higher on organic farms, even though these were revealed to be higher for weed management purposes alone. Farming systems may also have influenced risk loadings, as measured by variability in crop yields and prices, production costs and gross margins. Coefficients of variation for crop yields do not indicate much difference in variation across farming systems (Tables 3 to 5). Those for crop prices indicate greater variability for grain corn and fall cereal grains for organic systems. Given both lower and less variable total direct production costs on organic farms, coefficients of
variation for gross margins are comparable for fall cereal grains but lower than those for conventional and reduced-input systems for grain corn and bean crops, despite the higher crop price variability on organic farms. An alternative way of comparing gross margins across farming systems would be to extrapolate from the per-hectare figures in Tables 3 to 5 to a total focus crop enterprise size base (i.e., total gross margins per focus crop). These hypothetical gross margin-perhectare figures (Table 6) are obtained by multiplying the average number of hectares across farms within each farming system allocated in 1989 to each focus crop (per Table 1) by the requisite gross margin-perhectare figures by focus crop and by system (per Tables 3 to 5). For example, multiplying average grain corn area of 99.10 ha for conventional farms (Table 1) by average gross margin/ha of $264.24/ha (Table 3) gives
92
CANADIAN
JOURNAL OF AGRICULTURAL
ECONOMICS
Table 5. Summary: Fall cereal grains enterprise budget data, averages by farming system Conventional Number of farms
Reduced input
Organic
Organic d
9
7
3.50 1.27 36.75
4.02 1 .04 25.86
2.84 0.90 3 1.69
2.84 0.90 3 1.69
Five-year av. price ($/t)a Standard deviation (s/t) Coefficient of variation (%)
150.93 23.25 15.58
152.37 24.26 15.92
215.77 70.54 32.69
15 1.65 -
Five-year av. gross revenue ($/ha) 1989 av. total direct production costs ($/‘ha)b 1989 standard deviation ($/ha) Coefficient of variation (%)
528.26 338.34
612.53 268.53
612.79 260.63
430.69 260.63
80.43 23.76
81.94 30.5 1
76.33 29.29
76.33 29.29
Five-year av. gross margin ($/ha)c Standard deviation ($/ha) Coefficient of variation (%)
189.82 116.71 61.49
344.00 231.79 67.38
352.16 228.14 64.78
170.05 -
Five-year av. yield (t/ha)a _ Standard deviation (t/ha) Coefficient of variation (%)
6
6
a Significantly different according to Kruskal-Wallis test, a = 0.05. b Includes all labor costs, with unpaid operator and family labor valued at its opportunity cost. c Five-year av. gross margin = five-year av. yield * five-year av. price less 1989 av. total direct production costs. d Gross revenue and gross margin calculated using mean prices for conventional and reduced-input systems.
$26,186.18, the total gross margin for grain corn (Table 6). takes
This alternative comparison method into account differences in farm size,
tillable land base and scale of operation among the three farming systems, but only for the focus crops. Using this method clearly disadvantages the organic farms, with their much smaller degree of dependence on the three focus crops, particularly grain corn and beans. This is consistent with previous mention that organic farmers depend more on livestock than on cash-cropping for their gross revenues, and that few sales outlets were to be found for organically produced corn and beans. Whole Farm Business Analyses The previous analyses can demonstrate only comparative input use rates, costs and contributions by the focus crop enterprises to the
coverage of overhead charges across systems. An alternative and more complete analysis can be afforded by LP model comparisons on a net farm income basis for whole farm businesses. Such an approach reveals the greater enterprise diversification on organic farms, as well as the opportunity costs associated with constraining enterprise production levels to be different from their net farm income maximizing levels. Organic farms also emphasize holistic approaches, self-containment and self-sufficiency in resource use and intermediate products such as animal feeds and crop seeds. For conventional and reduced-input farms, the whole farm business analysis approach again reveals the opportunity costs attached to enterprise activity constraints. This approach also indicates the extent to which enterprise specialization was practised, and the extent to which the focus crops were depended
COMPARING
Table 6. Extrapolated
WEED
MANAGEMENT
93
STRATEGIES
focus crop enterprise gross margins, by farming system, 1989a Conventional
Reduced input
Organic
cv Grain corn Beans Fall cereal grains Total, all focus crops
26,186.18 15436.52 8,120.50 49,7 13.20
29,249.89 19,323.04 10,729.36 59,302.29
4,983.09 I,3 15.13 0,459.15 6,757.37
a Elements in Table 6 are composed of the products of average crop hectarages, by focus crop and by farming system (per Table l), and per-hectare gross margins, by focus crop and by farming system (per Tables 3 to 5). Table 7. LP model results for the four conventional
Total farm gross margin ($) Other gross margina ($) Overhead expenses ($) Net farm income ($) Net farm income ($/tillable ha) Corn (ha) Soybeans (ha) Winter wheat (ha) Barley (ha) Spring canola (ha) Winter canola (ha) Spring wheat (ha) Total tillable land (ha) Beef feeders (head) Shadow values:b Corn ($/ha) Soybeans ($/ha) Winter wheat ($/ha) Barley ($/ha) Spring canola ($/ha) Winter canola ($/ha) Spring wheat ($/ha) Beef feeders ($/head)
farms
Farm Cl
Farm C2
Farm C3
Farm C4
Mean average
90,492 10,000 134,200 -33,708 -76 303.64 60.73 80.97 0 0 0 0 445.34 0
39,139 0 20,785 18,354 227 26.32 26.32 20.23 8.09 0 0 0 80.96 0
63,444 0 73,389 -9,945 -27 141.24 93.08 80.94 0 12.14 36.42 32.37 396.19 0
21,913 4,863 42,489 -15,713 -127 101.21 0 22.27 0 0 0 0 123.48 150
53,747 3,716 67,716 -10,253 -1 74.85 45.03 51.10 2.02 3.04 9.1 I 8.09 261.49 37.5
212.30 140.10 216.39 0 0 0 0 0
635.16 45 1.68 368.07 381.52 0 0 0 0
205.95 121.43 225.79 0 -14.28 199.83 -71.90 0
96.16 0 -2.45 0 0 0 0 171.57
287.39 237.74 201.95 381.52 -14.28 199.83 -7 1.90 171.57
a Custom hire and land rental earnings. b Mean average shadow values calculated with zero weight attached to those farms not engaged in a specific enterprise. for the majority of business revenues. This in turn helps to highlight the contrast in approaches to farming in general across farming systems. Analytical results from the LP models are presented for each of four farms in each upon
of the three farming systems (Tables 7 to 9). In each table, the principal economic and technical findings are listed for the simulation of the actual 1989 farm situation. Economic results include the shadow values for those crop and livestock activities con-
94
CANADIAN
JOURNAL OF AGRICULTURAL
Table 8. LP model results for the four reduced-input
Total farm gross margin (S) Other gross margina ($) Overhead expenses ($)
Net farm income (S) Net farm income ($/tillable ha) Corn (ha) Soybeans (ha) White beans(ha) Kidney beans (ha) Winter wheat (ha) Oats (ha) Red clover (ha) Total tillable land (ha) Shadow vaIues:b Corn ($/ha)
Soybeans ($/ha) White beans ($/ha) Kidney beans ($/ha) Winter wheat ($/ha) Oats ($/ha) Red clover ($/ha)
ECONOMICS
farms
Farm RI1
Farm RI2
Farm RI3
75,316 16,404 80,348 8,372 46 106.03 31.16 25.09 0 17.80 0 0 180.08
84,2 11 0 81,238 2,973 17 37.65 37.25 46.96 0 50.00 0 0 171.86
25,836 2,206 21,171 6,87 1 136 12.14 8.00 12.14 6.07 12.14 0 0 50.49
120,339 0 37,626 82,713 317 82.26 83.40 0 0 54.25 36.42 4.86 261.19
75,675 4,653 55,096 25,232 129 59.52 39.95 21.05 1.52 33.55 9.1 1 1.21 165.91
357.40 276.67 709.66 0 449.12 0 0
544.77 346.99 691.56 0 898.86 0 0
635.16 583.5 1 470.38 955.66 354.0 1 0 0
559.07 413.66 0 0 517.87 283.63 293.52
524.10 405.2 1 623.87 955.66 554.97 283.63 293.52
Farm RI4 Mean average
a Custom hire and land rental earnings. b Mean average shadow values calculated with zero weight attached to those farms not engaged in a specific enterprise.
strained to be at their 1989 simulated levels. The shadow values represent the marginal change to net farm income that would have been obtained had the crop and livestock activities not been restricted to their 1989 actual levels. The four conventional farms modeled averaged a total farm gross margin of $53,747 in 1989 (Table 7). Other gross margin, generated from land rentals and custom hire work for neighboring farmers, averaged $3,7 16 per farm, and total farm overhead costs averaged $67,7 16 per farm. Total net farm income averaged a negative $10,253, leaving a zero contribution available to cover returns to equity, management and risk taking. Total net farm income per tillable ha ranged from a low of -$127 for farm C4 to a high of $227 for farm C2. Average 1989 tillable land base of 261 ha for these
four farms was allocated 28% (75 ha) to grain corn, 17% (45 ha) to soybeans, and 20% (5 1 ha) to winter wheat. Other crops grown included spring cereal grains and canola. Only one of these four conventional farms had any livestock; three of the remaining conventional farms not modeled also had livestock present. The shadow values indicate that, in all cases where corn and soybeans were grown, a positive marginal contribution to net farm income would have resulted from lifting constraints on activity levels, Winter wheat produced positive shadow values in three out of four cases. While barley and winter canola had positive shadow values on farms C2 and C3, respectively, spring canola and spring wheat had negative shadow values on farm C3. The beef feeders on farm C4 generated a positive shadow value of $17 1.57 per head.
COMPARMG
WEED
MANAGEMENT
95
STRATEGIES
Table 9. LP model results for the four organic farms Farm 01 Total farm gross margin (S) Other gross margina ($) Overhead expenses ($) Net farm income (S) Net farm income ($/tillable ha) Corn (ha) Soybeans (ha) Winter wheat (ha) Buckwheat (ha) Flaxseed (ha) Oil radish (ha) Spelt (ha) Rye (ha) Oats (ha) Barley (ha) Mixed grains (ha) Hay (ha) Pasture (ha) Total tillable land (ha) Poultry, hens (head) Dairy cows (head) Beef cows (head) Shadow values:b Poultry, hens ($/lo0 head) Dairy cows ($/head) Beef cows ($/head) Corn ($/ha) Soybeans ($/ha) Winter wheat ($/ha) Buckwheat ($/ha) Flaxseed ($/ha) Oil radish ($/ha) Spelt ($/ha) Rye (S/ha) Oats ($/ha) Barley ($/ha) Mixed grains ($/ha) Hay ($/ha) Pasture ($/ha)
I 14,487 12,397 99,089 27,795 110 29.94 0 30.35 0 0 10.11 30.35 20.23 28.32 54.65 0 54.63 10.12 253.21 4,000 0 10
1,639.40 0 219.20 240.82 0 290.43 0 0 -136.26 743.16 523.5 1 299.48 316.89 0 -465.49 -142.61
Farm 02
Farm 03
Farm 04
128,608 0 62,288 66,320 474 10.93 0 0 0 0 0 16.79 10.43 13.76 0 48.56 29.54 10.12 140.13 0 85 0
83,938 18,380 48,562 53,756 403 0 0 18.21 0 2.02 1.21 12.14 28.33 20.23 0 0 48.56 2.83 133.53 0 0 40
3,882 8,289 30,875 -18,704 -337 3.24 4.04 0 1.62 0 0 0 4.05 8.09 0 0 34.40 0 55.44 0 0 48
0
0 0 550.10 0 0 926.42 0 600.1 1 -142.17 1,009.12 512.01 986.63 275.45 0 -104.06 -159.35
0 0 52.05 -423.90 490.98 0 -36.44 0 0 0 -297.68 10.81 0 0 56.71 0
1,364.28 0 456.22 0 0 0 0 0 905.77 -188.24 61.28 0 62.15 -275.19 -120.98
Mean average 82,729 9,767 60,204 32,292 163 11.03 1.01 12.14 0.41 0.51 2.83 14.82 15.76 17.60 13.66 12.14 41.78 3.24 145.58 1,000 21.25 24.50
1,639.40
1,364.28 273.79 91.05 490.98 608.42 -36.44 600.1 1 -I 39.22 886.0 1 137.40 339.55 296.18 62.15 -197.0 1 -140.99
a Custom hire, land rental, firewood, maple syrup earnings. b Mean averag e shadow values calculated with zero weight attached to those farms not engaged in a specific enterprise.
The four reduced-input farms modeled averaged $75,675 for total farm gross margin,
$4,653 for other gross margin, and $55,096 for overhead expenses, so that net farm
96
CANADIAN
JOURNAL OF AGRICULTURAL
income averaged $25,232 (Table 8). Both gross margins and net farm income averages were higher for reduced-input farms than they were for conventional farms, while the overhead expenses average was lower. All four farms had positive net farm incomes per tillable ha, ranging from $17 for farm RI2 to $3 17 for farm R14. The range of crops grown on reduced-input farms was wider than that on conventional farms, and none of the reduced-input farms modeled had any livestock (three of the nine reduced-input farms in the study had livestock). The shadow values were positive for every crop grown on each of the farms and, in general, higher than those of the conventional farms. As with the conventional farms, there was wide variation in the marginal contributions to net farm income shown across reduced-input farms. The four organic farms modeled averaged higher total farm gross margins at $82,729 and other gross margin (from custom hire, land rentals, firewood and maple syrup sales) at $9,767 (Table 9) than their conventional and reduced-input counterparts. Overhead expenses, at $60,204 average on the organic farms, were higher than those on reduced-input farms but lower than those on conventional farms. Net farm income averaged $32,292 on organic farms, compared with $25,232 on reduced-input farms and -$10,253 on conventional farms. Net farm income per tillable hectare extended from a low of -$337 for farm 04 to a high of $474 for farm 02. By this measure of profitability, organic systems are shown to have the highest variability across farms. These results provide equivocal support for the third hypothesis only: reduced-input system profitability is shown to be higher than that for the conventional system. The results lead to rejection of the fourth hypothesis: organic system profitability is revealed not to be the lowest among the three weed management systems. Several reasons can be suggested for the above findings, especially for the result that organic system profitability is found to be highest. Every one of the organic farms had at least one type of livestock, and the diversity of crops grown was wide, much wider than
ECONOMICS
on the other two systems’ farms. Not all those crops grown on organic farms made a positive marginal contribution to overall net farm income, as shown by the shadow values (Table 9). Hay, pasture, oilseed radish and buckwheat (all grown variously for livestock feed production, soil conservation, plant nutrient retention and/or weed control) generally produced negative shadow values. Although not contributing in a positive measurable sense to net farm income, these crops with negative shadow values were presumably important components of the overall crop rotation system. Through their roles as soil conditioners, plant nutrient retainers or weed controllers, such crops presumably made indirect contributions to net farm income, despite the negative shadow values indicated.4 Diversified enterprises, along with long crop rotations, cultivations, attempts to increase soil organic matter levels and to be as self-sufficient as possible in plant and animal nutrients, were all strategies used by organic farmers. Such strategies served as alternatives to the synthetic pesticides and fertilizers used on reduced-input and conventional farms. Given the lower costs of production for the focus crops found on organic farms, these same strategies may also be indicating the potential for more costeffective methods of producing crops, despite the higher weed management costs. Lowest overhead costs on reduced-input farms would appear to indicate potential for more costeffective methods of farming in general. CONCLUSION Based on a small sample of 25 farms, whose representativeness of the general population is not known, only preliminary conclusions can be drawn from the results obtained. No claim is being made here about the general applicability of the results or conclusions. First, conventional farms in this study tended to be more specialized operations with a smaller range of crop enterprises than either reduced-input or organic farms, but also tended to operate larger hectarages of tillable land and land in total. Second, overall costs of weed control and labor require-
COMPARING WEED MANAGEMENT STRATEGIES
ments for weed management were highest on organic farms in the sample, but have likely been overestimated in this study, by virtue of attributing all cultivation (and, on organic farms, all manure cornposting) costs to weed control. Third, there are equivocal rankings among farming systems for average crop yields, with reduced-input systems in the sample outranking other systems for both grain corn and fall cereal grains, but organic systems in the sample ranking first in bean yields. However, this conclusion in particular should be viewed as tentative because only three organic farms reported producing bean crops. Further testing is required to support or reject this finding. Fourth, despite higher weed control costs on a per-hectare basis, organic farming systems in the sample generated higher gross margins on a per-hectare basis across all crop enterprises evaluated than the other two systems. While some of the superior gross margins outcomes were due to product price premiums earned by organic farmers in the market place, total direct crop production costs were lower on the organic farms, especially in the case of grain corn. Using identical crop prices across systems, organic farmers would still have generated superior gross margins for corn and bean crops on a per-hectare basis. The inferior performance of organic farms in fall cereals grains using identical product prices can be partially attributed to the lower crop yields expected from both spelt and rye, which were emphasized more than winter wheat on the organic farms. Fifth, both conventional and reducedinput systems generated superior gross margins to those on organic farms when measured on a gross margin per focus crop enterprise basis, by virtue of their much larger tillable land areas devoted to the three focus crops. Focus crop enterprise gross margins were higher on reduced-input farms than on conventional farms. Sixth, net returns on a whole farm basis were highest on the four organic farms modelled, being 28% higher than those on four reduced-input farms, and considerably higher than the negative returns on average obtained
97
on four conventional farms, despite the much smaller land base employed on organic farms. Highest net farm incomes on a per-tillable hectare basis were also found on the organic farms, and also the greatest variability in this measure of profitability. Broader-scale testing is required to confirm or reject these findings. Much of the superior net returns performance on organic farms in the sample may have been attributed to the wider range of enterprise lines and, in particular, to the inclusion of livestock enterprises. Additional factors were the lower levels of direct costs of production on organic farms, which may themselves have been a function of wider diversification. Organic farms appeared to gain from cost savings on inputs purchased from off the farm, such as fertilizers, pesticides, and feed additives, and also the premium prices received for organically produced crops. The significance of the preliminary results and conclusions lies in the implication that reduced-input and organic methods appear to be viable alternatives to conventional farming methods. Reduced application rates of inputs seem to be strongly implicated. Other factors, such as greater diversification of enterprise lines on organic farms, also seem to be implicated. The extent of contributions to overall viability from these factors is not clear from the present study results, and would require further research. NOTES ‘Corn heat units (CHU) are defined as a maturity indexing system relating air temperature to corn development (towards maturity). Daily CHU can be calculated according to the formula: 10.0)-0.084(TMAX-10.0)2
1
2 where
T
= the daytime
maximum temperature, in Celsius, in Th41N = the nighttime minimum temperature, degrees Celsius. Seasonal totals are obtained by summing all positive values for daily CHU calculations.
d$%es
98
CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS
2Scarcity of research resources made it impossible to develop LP models for all 25 farms. 30ther objectives of cornposting include stabilizing the plant nutrient content of the manure, especially the nitrogen component, and reducing the bulk and weight of manure for easier handling purposes. 4The true extent of contributions to net farm income from enterprise activities generating negative shadow values could be revealed only with more elaborate LP models in a dynamic setting, showing more direct and complete linkages among crops in the rotational sequence, and between crop and livestock enterprises. ACKNOWLEDGMENT Funding support from the Ontario Ministry of Agriculture, Food and Rural Affairs “Food Systems 2002” program is gratefully acknowledged. The helpful cooperation of 25 participating farmers is sincerely appreciated. The constructive criticisms and helpful suggestions of three anonymous Journal reviewers on earlier drafts of this paper are gratefully acknowledged. REFERENCES Baldwin, L. R., R. Oliver and T. N. Tripp. 1988. Arkansas experience with reduced rate herbicide recommendations. Abstract in Environmental Legislation and its Effect on Weed Science, Proceedings of the 4 1st Annual Meetings of the Southern Weed Science Society, Nashville, Tennessee, January. Berardi, G. M. 1978. Organic and conventional wheat production: Examination of energy and economics. Agro Ecosystems 4: 367-76. Bridges, D. C. and R. H. Walker. 1987. Economics of sicklepod (Cassia obtusifolia) management. Weed Science 35: 594-98. Colvin, D. L. B. J. Brecke and W. L. Currey. 1986. Weed control and economic evaluation of full season-double cropped, conventional and minimum tillage peanuts. In Weed Science and Risk Assessment, Proceedings of the 39th Annual Meetings of the Southern Weed Science Society, Nashville, Tennessee, January. Edwards, C. A., R. Lal, P. Madden, R. H. Miller and G. House, eds. 1990. Sustainable Agricultural Systems. Ankeny, Iowa: Soil and Water Conservation Society. Fox, G. C., A. Weersink, G. Sarwar, S. Duff and B. Deen. 1991. Comparative economics of
alternative agricultural production systems: A review. Northeastern Journal of Agricultural and Resource Economics 20 (4): 124-42. Goldstein, W. A. and Douglas L. Young. 1987. An agronomic and economic comparison of a conventional and low-input cropping system in the palouse. American Journal of Alternative Agriculture 2 (2): 5 l-56. Helmers, G. A., Michael Langemeier and Joseph Atwood. 1986. An economic analysis of alternative cropping systems for east-central Nebraska. American Journal oj- Alternative Agriculture 1 (4): 153-58. Keller, G., B. Warrack and H. Bartel. 1994. Statistics for Management and Economics. 3rd. ed. Belmont, California: Duxbury Press. King, R. P., D. W. Lybecker, E. E. Schweizer and R. L. Zindahl. 1986. Bioeconomic modelling to stimulate weed control strategies in continuous corn (Zea mays). Weed Science 34 (6): 972-79. Lybecker, D. W., R. P. King, Edward E. Schweizer and Robert L. Zimdahl. 1984. Economic analysis of two weed management systems for two cropping rotations. Weed Science 32 (1): 90-94. Lybecker, D. W., Edward E. Schweizer and R. P. King. 1988. Economic analysis of four weed management systems. Weed Science 36 (4): 84649. Madden, P. 1988. Low-input/sustainable agricultural research and education: Challenges to the agricultural economics profession. American Journal of Agricultural Economics 70 (5): 1167-72. Monks, C. D. 1989. Economic comparison of snap bean control system. Abstract in Weed Science and Risk Assessment, Proceedings of the 39th Annual Meetings of the Southern Weed Science Society, Nashville, Tennessee, January. Moxley, Jacqueline. 1989. Survey of pesticide use in Ontario, 1988: Estimates of pesticides used on field crops, fruits and vegetables. Economics Information. 89-08. Toronto: Ontario Ministry of Agriculture and Food, Economics and Policy Coordination Branch. Musser, W. N., B. N. Tew and J. E. Epperson. 1981. An economic examination of an integrated pest management production system with a contrast between e-v and stochastic dominance analysis. Southern Journal of Agricultural Economics 13 (1): 119-25. Paoletti, M. G., T. L. Napier, 0. Ferro, B. R. Stinner and D. Stinner, eds. 1993. Socio-economic and Policy Issues for Sustainable Farming Systems. Padova, Italy: Cooperativa Amicizia S.r.1.
COMPARING
WEED
MANAGEMENT
Pimentel, D. and H. Lehman, eds. 1993. The Pesticide Question: Environment, Economics and Ethics. New York and London: Chapman & Hall. Sahs, W. W., Glenn A. Helmers and Michael R. Langemeier. 1989. An economic comparison of the profitability of organic and conventional production systems for east-central Nebraska. Paper presented at the National Conference on Policies, Organic/Sustainable Agriculture Washington, D.C., 16-l 7 March. Snipes, C. E., R. H. Walker, T. Whitwall, G. A. Buchanan, J. A. McGuire and N. R. Martin. 1984. Efficacy and economics of weed control methods in cotton (Gossypium hirsutum). Weed Science 32 (1): 95100. Stonehouse, D.P. 1994. Canadian experiences with the adoption and use of soil conservation practices. In Adopting Conservation on the Farm: An International Perspective on the Socio-economics of Soil and Water Conservation, edited by T. L. Napier, S. M. Camboni and S. A. El-Swaify, pp. 369-96. Ankeny, Iowa: Soil and Water Conservation Society. Stonehouse, D. P. and M. J. Bohl. 1990. Land degradation issues in Canadian agriculture. Canadian Public Policies 16 (4): 4 18-3 1.
STRATEGIES
99
van Vuuren, W. and D. P. Stonehouse. 1995. Policy instrument effectiveness in water quality control programs aimed at agriculture. In Potential Applications of Economic Instruments to Address Selected Environmental Problems in Canadian Agriculture, edited by A. Weersink and J. Livemois, pp. 11l-38. Ottawa: Agriculture and Agri-Food Canada, Environment Bureau. Wilcut, John W., Glenn R. Wehtge and Michael G. Patterson. 1987a. Economic assessment of weed control systems for peanuts (Arachis hypogaea). Weed Science 35 (3): 433-37. Wilcut, John W., Glenn R. Wehtge and Robert H. Walker. 198713. Economics of weed control in peanuts (Arachis hypogaea) with herbicides and cultivations. Weed Science 35 (5): 71 l-15. Zavaleta, L. R., B. Eleveld, M. Kogan, L. Wax, D. Kulman, and S. M. Lim. 1984. Income and risk associated with various pest management levels, tillage systems, and crop rotations: An analysis of experimental data. Agricultural Economics Research Report. Urbana-Champaign: University of Illinois, College of Agriculture, Department of Agricultural Economics, Agricultural Experiment Station.
