SEA Working Paper 00/07

Economic and Sociological Factors Affecting Growers’ Decision Making on Herbicide Resistance

David J. Pannell^A and David Zilberman^B

^AUniversity of Western Australia, Nedlands WA 6907, Australia
^BUniversity of California, Berkeley

1. Introduction

Economic and sociological aspects have received little attention in the published literature on herbicide resistance, with the only published studies to date being those of Schmidt and Pannell (1996a, 1996b), Gorddard et al. (1995, 1996) and Orson (1999). The published literature on economic and social aspects of insecticide and fungicide resistance is a little larger (e.g. Hueth and Regev, 1974; Taylor and Headley, 1975; Moffitt and Farnsworth, 1981; Regev et al., 1983; Knight and Norton, 1989; Clark and Carlson, 1990; Peck and Ellner, 1997; Smale et al., 1998) but contains little that can be readily applied to the case of herbicide resistance apart from general concepts.

The management problem for herbicide resistance has a number of differences to these other types of resistance. One important practical difference is that there exists a vastly greater range of control methods available for weeds than for insects and micro-organisms, so that substitution out of chemical control options is much more likely to be economically viable. Indeed, many farmers already incorporate a variety of control methods, in tandem with selective herbicides. Another difference is the importance of spread. Resistant insects and diseases are commonly much more mobile than resistant weeds, and so have much greater problems of communal resistance management, whereas herbicide resistance is mainly a problem for private, individual farmers.

Our chapter proceeds in three parts. Firstly, we discuss issues relating to the phase when farmers are considering adoption of changed farming practices, either to delay the onset of resistance, or to deal with its arrival. We review the key factors affecting the speed and nature of farmers’ responses to possible changed management practices and discuss the implications of this information for adoption of management practices for herbicide resistance.

Secondly, we employ a detailed bioeconomic model to examine herbicide resistance in the context of a complex mixed farming system. We examine the impact of resistance on the profitability of farming in a case study, and examine the changes in farm management that resistance requires or encourages.

Thirdly, we examine the policy implications of herbicide resistance. Is there a need for government intervention to manage the resistance problem for the common good? In what ways might this intervention occur?

2. Herbicide Resistance and the Adoption of Changed Farming Practices

Before addressing some of the specific issues relating to resistance we will present some of the major concepts related to technological change and agriculture. In the discussion of these concepts, we will illustrate key points with historical examples relating to adoption of mechanical and chemical weed control methods. Following this, we will discuss specific issues relating to the adoption of herbicide resistance management practices.

2.1 Introduction of New Technologies

It is useful to distinguish between two processes: innovation, which is the introduction or development of new technologies, and adoption, which is actual usage of these new technologies. The induced innovation hypothesis, originated by Haymi and Ruttan (1970) and formulated by Binswanger (1974), argues that innovations are strongly influenced by profitability considerations, particularly "demand-driven" innovations. Demand-driven innovations, in the context of farming, would be innovations generated by or at the behest of farmers in direct response to farming problems perceived or recognised by farmers. They may save scarce resources, increase yields or enable production of new products that are more profitable. "Supply-driven" innovations, on the other hand, are generated from scientific research. They may not necessarily be addressing a problem recognised in advance by farmers, but may nevertheless generate benefits to farmers and/or others in the community, for example, by revealing a hitherto unrecognised constraint on yields.

In traditional society, manual weeding was (and often still is) the dominant form of weed control. The development of mechanized weeding through disks and cultivators was driven both by demand factors, namely the increased scarcity of labor, and by supply factors, such as the invention of the tractor.

Mechanical weeding, however, still requires substantial machinery and labor time and entails energy costs. During the 20^th century, costs of inputs such as labor, machinery and energy increased relative to the prices of the major agricultural commodities, so mechanical weeding became more expensive per unit value of production. This created a potential for benefits from alternative weed control technologies, which was met through the development of chemical herbicides. Environmental factors also drove the demand for chemical solutions, as concerns about soil erosion made cultivation less desirable. The availability of chemical herbicides made it feasible to introduce no tillage and minimal tillage systems, allowing earlier seeding of crops. Recently there have also arisen concerns that heavy tillage releases carbon dioxide into the atmosphere (Marland et al., 1999). The attractiveness of low tillage activities may be enhanced by policies that provide economic incentives for soil carbon sequestration, as part of an effort to curtail climate change (Marland et al., 1999).

The introduction of chemical herbicides was also driven by supply factors. First, development in weed science and chemistry made chemical herbicides cheaper and safer. Furthermore, the use of chemicals to control other pests provided opportunities for complementarity. Marginal application costs of herbicides are low if farmers already use spraying machinery to protect against other pests, such as insects.

While the use of chemicals for control of other pests has not grown significantly since 1973, at least in the United States, the use of herbicides has increased drastically (Miranowski and Carlson, 1993).

The creation of an innovation is only part of the process of technological change. In order for change to occur, the product must be adopted by the producer.

2.2 Adoption and Diffusion

We can distinguish between adoption, which is the extent to which an individual producer is using the new technology, and diffusion, which is the percentage of farmers that have adopted a new technology or the percentage of land on which the innovation is used.

Adoption and diffusion have been studied by both sociologists and economists. The empirical evidence shows that the pattern of diffusion usually follows a sigmoidal or S shaped function over time (Rogers, 1995). That is, diffusion may be represented by a function of the form P(t) = K/[1 + exp(-a - bt)], where P(t) is the diffusion level at time t, K is the long-run upper limit on the level of diffusion, the slope coefficient, b, reflects the speed of uptake of the new technology, and the intercept, a, reflects aggregate adoption at the start of the period. K, in most cases, is smaller than one since diffusion of innovations is generally not complete. In a landmark study Griliches (1956) demonstrated that the three parameters are strongly affected by economic profitability. Producers or regions for which adoption of a new technology is relatively profitable are likely to adopt the technology earlier and to a greater extent. For example, in the case of herbicides, regions that have relatively high costs of energy or labour are likely to more rapidly adopt herbicides, which save on these inputs.

Non-financial influences also play important roles in the diffusion process, including factors like the quality of communication channels, social acceptability, social infrastructure and farming subcultures or farming styles (Vanclay, 1997). These are often conceptualised by economists in terms of their influence on incentives for or against adoption, and in this way most sociological factors can readily be considered within the sort of economic framework normally applied to financial influences on adoption (e.g. Lindner, 1987). Sociologists use different language and different conceptual frameworks to discuss these issues (e.g. Vanclay, 1997) but the underlying logic of the two disciplinary approaches is, in most cases, identical.

2.3 Explanations of the Sigmoidal Diffusion Curve

One explanation for the S shaped diffusion curve is the process of imitation. There are a small number of early adopters but then the number of imitators grows exponentially and the diffusion process enters its take off period. Since the population is finite beyond a certain point, the diffusion rate slows until the level stabilizes. Eventually, there may be disadoption of the technology following the introduction of more advanced technology. This pattern of dynamic behavior has occurred with various weed control technologies. There was a transition from manual to mechanical weeding. Then there was diffusion of chemical herbicides, and within this category there has continued to be diffusion of new herbicide products at the expense of earlier herbicides. Following the failure of herbicides due to herbicide resistance, sigmoidal diffusion curves for replacement technologies and practices will again be apparent.

An alternative explanation to the S shaped diffusion curve was presented by David (1969) and Stoneham (1981) who argued that the reason for the initially gradual diffusion is heterogeneity. The causes of heterogeneity are differences in farm size, human skills and knowledge, machinery capacity, crop yields, soil types, disease risks and other variables. Producers are motivated to choose technologies that maximize profits and, because the profitability of any new technology will vary widely from farmer to farmer, this provides different incentive levels for adoption. For example, a technology that requires indivisible investment in equipment (purchase of a tractor or computer) is more likely to be adopted earlier by larger producers because the fixed cost can be spread over a larger volume of operation. If a new technology earns a extra dollars per unit of output and requires a fixed investment of k dollars per period, only producers with a volume greater than k/a will adopt the new technology. Thus there would be a threshold farm size separating adopters and non-adopters. Over time, the threshold level is likely to decline because of experience and learning by both the farmers and, if relevant, the manufacturers of the technology. Learning on the part of farmers results in an increased a , and learning on the part of manufacturers may reduce the cost of producing the technology, and thus decrease k. Either way, the threshold farm size for adoption declines over time, resulting in an increased level of diffusion. Tractors and other farm machinery are likely to be examples for which this threshold model is relevant. Larger farms adopted the technology first, increasing production volumes and allowing tractors to become more affordable so that smaller farms later joined the ranks of adopters.

However, the threshold model does not fully explain the dynamics of new technology diffusion, even for technologies such as tractors that conform to the model (an indivisible, high-cost item). For example, earlier adopters of such a technology may have a significant economic advantage over non-adopters. Some would have sufficient excess productive capacity to enable them to purchase extra land, leading to an increase in average farm size. On the other hand, Olmstead and Rhode (1993) suggests that farmers form partnerships so that they can afford mechanical innovations. This enables several small farmers to adopt expensive machinery and thus accelerates the diffusion process. In some areas with small farms, private contractors established firms that provided mechanical services to farmers who could not afford to purchase the machinery.

No doubt, both of the above explanations apply simultaneously to some extent for virtually all innovations. Overlaid on these explanations (and consistent with them) is the approach that conceptualises adoption as being primarily a process of dynamic learning and refinement of decision making over time (e.g. Lindner, 1987; Abadi and Pannell, 1999). This approach recognises that growers vary in their risk preferences and their perceptions of riskiness of an innovation, and that their perceptions are affected by information from various sources at different times. Sources of information can include experience gained from on-farm trials of the innovation, but also off-farm sources such as extension agents, neighbouring farmers or the media. Heterogeneity of perceptions, perhaps due to different levels of experience or different sources of information, is another factor that would contribute to the sigmoid shape of the diffusion curve.

2.4 Influences on the Speed and Level of Diffusion

The incentives for or against adoption may be influenced by many factors, including those outlined below.

1. Land quality. Caswell and Zilberman (1986) argued that differences in land quality and soil conditions (e.g. slope of land and water holding capacity of soil) are also important in explaining adoption of modern technologies. There is strong evidence to support this empirically (see Green, 1995). Similarly, adoption of low tillage practices occurred at locations with lower soil quality (measured by depth of top soil – Wu and Babcock, 1998). Adoption of herbicides was complementary to the lower tillage practices and thus was particularly favoured in situations where soil erosion was a problem and topsoil was shallow.
2. Prices. Increases in the price of fuel and labor tend to increase the adoption of herbicides (Miranowski and Carlson, 1993; Carlson and Wetzstein, 1993b). Increases in commodity prices or support programs that are based on yield tend to increase the intensification of farming and thus increase adoption and use of herbicides (Helms et al., 1987; Miranowski et al., 1991). Higher sale prices also tend to increase cultivation of land in locations of lower soil quality (de Gorter and Fisher, 1993) and this may further increase use of herbicides.
3. Government policies. Government policies that aim to improve environmental policies also serve as inducements for adoption of alternative practices. Since there has been an increased emphasis on soil erosion prevention, US governments have required certain environmental practices as conditions for participation in commodity programs that raise product prices. This has led to increases in the area of land with no or minimum tillage, and thus increased the reliance on herbicides (Uri, 1998; Wu and Babcock, 1998). Note, however, that these programs also include provisions to encourage reductions in the use of some herbicides, such as atrazine. Thus, their overall effect is to increase the use of herbicides but to modify them towards herbicides that are judged to be more environmentally benign. The global tendency towards freer agricultural markets and reduced government supports for agriculture may actually lead to a reduction in the area of low tillage practices. On the other hand, the introduction of policies that subsidize soil carbon sequestration may increase the use of soil conserving practices and, consequently, the use of herbicides.
4. Location. Initial studies of diffusion linked diffusion rates to locations (Rogers, 1995). For most technologies the evidence points to higher adoption rates by producers that are closer to the nearest urban center than producers that are further away (Rogers, 1995). That may be due to a lower transportation cost and/or closer contact with dealers and distributors of the technology or extension agents. In some cases, however, remote regions are more likely to adopt technologies such as herbicides if they save inputs (labor and energy) that are more costly in these remote locations.
5. Uncertainty and risk. Uncertainty and risk are major obstacles for the adoption of new technologies. Farmers to a large extent are risk averse (see Bar-shira et al., 1997) meaning that they are willing to give up some average profit in order to reduce fluctuations in, or uncertainty about, profit. In many cases adoption of new technologies is gradual so farmers can experiment with them and have more reliable assessments of their impacts. One of the main activities of marketing specialists and extension agents is to reduce farmers’ uncertainties about new technologies. This is achieved, for example, by the establishment of demonstration plots, demonstration activities, money back guarantees or warranties. At the time of the introduction of herbicides, there must have been substantial uncertainty among farmers regarding their performance both in terms of their effectiveness and their possible side effects, and this would have affected the pace of their diffusion. Effective marketing plans that educate farmers about the properties of a new technology and reduce their uncertainty about it will likely result in higher adoption rates. The high initial adoption rates of transgenic glyphosate-resistant crops in the United States suggest that Monsanto was able to reduce much of farmers’ uncertainty about the product by its marketing activities.
6. Observability and trialability. These issues are closely related to the issue of uncertainty. If a technology is highly observable, the diffusion process via the imitation mechanism is likely to proceed more rapidly. One can well imagine that the high observability of tractors enhanced their rate of spread through farming communities (although it clearly does not over-ride other considerations in a farmer’s final decision to adopt or reject an innovation). Perhaps even more important is the potential to trial the innovation on a small scale, so that information can be obtained, uncertainties reduced and skills developed without the risk of large financial costs if the technology turns out to be uneconomic or fails due to inexperience. Innovations such as a new herbicide are highly trialable; its impacts on weeds are readily observable, it can be readily compared to existing herbicides, and it can be used effectively on a very small scale.
7. Adjustment costs. The high rate of adoption of genetically modified crop varieties in North America and, now, South America can likely be explained by the fact that these are technologies that require very little adjustment cost. They do not require modification of the production system or a significant amount of learning by the farmer. The farmer simply replaces one seed with another and continues to use herbicides with which they are familiar. One of the key obstacles to the adoption of new technology is that can they require investment of time and effort from the farmers. The farmer does not consider only the direct financial cost of adoption but also any non-monetary costs. One reason that the adoption of IPM has been relatively slow is the relatively high adjustment cost it entails (Wiebers, 1992). High adjustment costs also explains the slow rate of adoption of modern irrigation technologies (Green et al., 1996). Because of the need to invest in adjustment, older farmers are less likely to invest in learning the adoption of new technologies than are younger farmers (Green, 1995). One rationalisation of the observation that individuals with higher human capital tend to adopt new technologies earlier is that they probably have relatively low learning costs. The adoption of mechanized farming practices and mechanized weeding was slow because of farmers' need to learn to operate new machinery, with the result that complete adoption of tractors took about 40 years in the United States (Cochrane, 1979). On the other hand, adoption of herbicides has been less problematic, especially for farmers who already sprayed against insects and therefore required little up-front investment in machinery. This helps explain the rapid increase in herbicide use in the 1980s. Adoption of complex resistance management strategies may be slowed by the adjustment cost that they entail, including purchase costs of any new machinery that is necessary.
8. Public acceptance. Adoption of new technologies may also be affected by public acceptance considerations. Adoption rate may be lowered if farmers perceive that by adopting a new technology they may endanger their capacity to sell their product at a high price, or that they may encounter legal problems or protests. Some of the controversy regarding genetically modified varieties is likely to reduce the area planted to such varieties. Farmers will engage in extra activities that will enable them to adopt new technologies in a manner that will be more socially acceptable, as long as the overall benefit is greater than the cost.
9. Education. More educated farmers tend to adopt advanced technology earlier. Putler and Zilberman (1988) have shown that education was a key explanatory variable for the adoption of computers by farmers in Tulare County, California. Education has also been shown to be a key factor explaining adoption of various forms of IPM (Wiebers, 1992) and soil testing. (Wu and Babcock, 1988). However, farmers with less education may adopt more sophisticated technologies when they rely on private consultants or cooperative extension. Huffman (1974) have shown that the productivity of extension increases when farmers are less educated. Wolf (1996) documented the evolution and growth of private consulting in pest management.
10. Extension and consulting advice. Zilberman et al. (1994) have argued that, increasingly, decisions regarding pest management are made based on significant input from professionals, particularly in the United States and other industrialised agricultural nations. Smaller farms rely on dealers or independent consultants and some of the bigger farms have in-house experts. Furthermore, government extension is shifting much of its efforts towards education and certification of consultants. In the US, these experts are mostly concerned about insect control but, coversely, in Australia they are mostly focussed on weed control. No doubt, the role of consultants in weed control will increase. This trend suggests that educational activities by companies and universities regarding herbicide resistance and weed control in general should be increasingly directed at consultants. Indeed, in Australia, where herbicide resistance is widespread, conscious decisions have been made by government agencies and universities to target consultants as information providers.
11. Marketing. A key to the diffusion of a new technology is providing information about its existence and properties. A significant amount of the budget of chemical producers is dedicated to marketing their products through activities such as advertisements, demonstrations and trade shows, money-back guarantees and warranties, and product stewardship through salespeople (Kotler, 1997). The marketing literature views adoption as a hierarchical and gradual process with several stages. First is product awareness where farmers may become aware of a new product through advertisements and discussions with other farmers. The second stage is information search where farmers may go to a retail outlet to learn about new products. That is followed by evaluation, purchasing decision, and post-purchase evaluation. Advertising is crucial for product awareness to trigger interest in a new product. Kotler (1997) divided potential adopters into two segments: (1) early adopters who are influenced by advertisements and (2) imitators who are influenced through word of mouth of previous adopters.
    
    Successful marketing of a new herbicide entails identifying the market segments where this pesticide is most appropriate, developing a marketing strategy using the right pitch to gain potential buyers' attention, and training dealers to recognize and deal with buyers' concerns and to help them adjust to the new product. Activities of salespeople have to be supported by effective demonstrations, educational material, and warranty programs (Kotler, 1997).

This by no means exhausts the list of factors that have been found or proposed to influence adoption, but it includes the main types of factors of influence mentioned in the literature (although not necessarily with the same language).

2.5 Resistance Management

While there is a significant economic literature on the adoption of integrated pest management (see survey by Carlson and Wetzstein, 1993a), there are no existing studies focussed on adoption of resistance management practices in agriculture and only few on the economics of such practices. Most of the economic studies are normative and/or conceptual. They derive economically optimal pest control strategies taking into account resistance build up and compare them to alternative strategies that may be more pervasive. One notable example is a study by Hueth and Regev (1974). They argue that the optimal resistance management must be analyzed within a dynamic context. Pesticide resistance build up is a dynamic phenomenon and once resistance reaches a high level, chemical treatment will become ineffective. Therefore, they modeled vulnerability to a pesticide as a non-renewable resource and argued that overuse of a pesticide would lead to depletion of this vulnerability. They argued that concern for build up of resistance should reduce application of chemicals.

Decision makers may apply pesticide sub-optimally, disregarding resistance build up considerations. In most cases that will result in over-application of certain chemicals. There are several possible reasons for high chemical usage. First, decision makers may be unsophisticated, using myopic or short-term decision making rules rather that dynamic, long term ones. A second possible reason is that producers may have high discount rates because of their financial situations and thus tend to give a relatively low weighting to long-term benefits relative to short-term costs of their present activities. This may particularly be the case with poor farmers whose economic survival is potentially under threat. It may also be the situation in an economy where access to credit is very limited and interest rates faced by farmers are very high. A third possibility is that long-term planning may actually result I little difference in management compared to short-term planning. A case study presented later shows that this is a realistic possibility for herbicide resistance in at least one farming system. Fourth, the adoption of strategies to slow resistance will be slow when farmers expect substitute pesticides to become available over time, replacing those lost to resistance.

Extension and other outreach activities may educate farmers to modify their behavior towards more sustainable practices when ignorance is the cause for ignoring resistance considerations. In agriculture of the developed world, the previously noted increase in reliance on consultants for pesticide advice is likely to modify behavior so that resistance will be taken into account. In some farming regions, however, it seems possible that farmers may use consultants and be exposed to sophisticated advice on insect control problems but may be less well informed about weed control and herbicide resistance. On the other hand, the converse is more likely to be true in other environments (e.g. Australia). Therefore, it is important to develop educational opportunities that will enable farmers to consider the resistance problem for all types of pesticides.

Education will not modify pest application practices, however, if farmers' behavior is dictated by high discount rates. In these cases, pest control strategies will best be modified if informed farmers have easier access to credit under better terms so that their discount rates will decline and they will give more weight to future consequences of their activities. Thus, in many cases, improvement of credit channels is essential for more emphasis on long term perspectives in pest management. Neither education nor cheap credit may be sufficient if the prime cause for non-adoption of herbicide resistance prevention strategies is that the long run economically optimal strategy is to exploit the herbicide resource to exhaustion.

It has been noted in the literature on insecticide resistance that a lack of attention to resistance build up may also occur when this build up is affected jointly by the activities of many producers. In cases where there is sufficiently rapid movement of pests between farms, individual farmers are aware that their individual actions will not prevent the build up of resistance (which depend on the activities of the overall behavior of the community) and thus that individual investments to prevent resistance will incur costs without generating benefits that can be captured by the investor. In this case there is likely to be a "tragedy of the commons". Namely, each individual considers only their private, short-term interests and over-applies chemicals such that there is excessive build up of resistance. Possible policy mechanisms for preventing such over application are discussed in the fourth section of this chapter.

It seems that regional dependency is less important for the build up of herbicide resistance than some cases of insecticide resistance, because of the relatively low mobility of most weeds relative to some insects. Furthermore, much of the use of herbicides occurs on large operations where herbicide resistance problems are contained within the boundaries of the operation. In these cases, there is less inter-dependence amongst operators and more prospect of private herbicide decision being consistent with the interests of the community.

Most studies of resistance have considered only one chemical and developed optimal resistance strategies, assuming that this chemical will be the only one available in the long run. However, if producers expect to see new categories of pest control appearing over time, then every such category can be viewed as a backstop technology that provides relief from the problem associated with the build up of resistance to existing chemicals. The potential development of new pest treatment mechanisms in the future leads producers to reduce their emphasis on resistance build up consideration in the present. This is analogous to a decision to ignore the exhaustibility of fossil fuels because of a perception that there is a high likelihood of developing alternative energy sources. A major issue here is the speed of development of alternative treatment types relative to the speed of resistance build up. Furthermore, uncertainty regarding the development of alternative pest control technologies provides another reason for adoption of a resistance management strategy. It may be that at a certain point we run out of new forms of pest management. This risk is enhanced by the occurrence of resistance to multiple herbicides in particular weeds, particularly if the mechanism for multiple resistance is non-target site (e.g. Tardif et al. 1997). In the case of herbicides, it seems that the rate of discovery of chemicals with new modes of herbicidal action has already fallen to a very low level. Obviously, assessment of the rates of resistance build up versus rates of development of new treatments is an empirical issue that must be studied further. A conservative approach may suggest emphasizing resistance prevention strategies.

That suggests that whenever a new form of treatment is being introduced (e.g. new technologies developed using biotechnology) it is important to consider the potential for resistance build up and, if possible, develop management strategies that will economically slow such build up. This may not always be technically possible, as illustrated in the next section. There may also be difficulties obtaining cooperation from chemical producers. For example, delay of resistance build up may require the integration of chemicals (and other approaches) that are promoted by different chemical companies. Sellers of a new chemical, especially in the early stages of production and patent protection, have every incentive to under emphasize the drawbacks of their product and to aggressively capture a significant market share that will trigger the diffusion process. On the other hand the sellers of incumbent products may not wish to be accommodating to the introduction of new chemicals that may displace their own products. This suggests an important role for extension specialists and independent consultants in developing resistance management strategies.

2.6 Adoption of Resistance Prevention Strategies

Considering the nature of herbicide resistance and its management, it is possible to make some observations about adoption of resistance prevention strategies based on lessons from the general adoption literature.

In particular we note that herbicide resistance, as a farm management issue is:

• complex;
• potentially expensive or difficult to prevent;
• a source of considerable uncertainty; and
• difficult to observe until herbicide resistance is advanced.

Compexity increases uncertainty, adjustment costs and, possibly, direct costs. Complexity in herbicide resistance management arises from several sources. Firstly, farmers faced with resistance can benefit greatly from understanding the ecological and biochemical theories that explain the occurrence of resistance, but these theories may be difficult to fully understand for non-scientists. Secondly, the weed treatment options that can be used to substitute for herbicides and delay resistance are themselves complex to use relative to herbicides, and normally must be used in combinations in order to achieve adequate weed control. Thirdly, the number of potential combinations of treatments to be considered is vast, especially when the issue is considered as a dynamic problem spanning a number of years. In this case, the options include not just changes in weed kill methods but also, potentially, changes in land use; for example, pasture may be included in place of a crop to broaden the available range of weed control options.

The cost-effectiveness of herbicides is revealed by their ubiquitous use through cropping systems of agriculture in the developed world. If a farmer is considering changing to a lower intensity of herbicide use in order to delay the development of herbicide resistance, he or she will be very conscious that in most cases such a move will increase short-term costs. The change will therefore be unattractive unless the farmer is convinced that there are sufficient offsetting benefits. There may actually not be sufficient offsetting benefits in some situations.

Even if there are sufficient offsetting benefits to justify early adoption, it may be very difficult for the farmer to determine this with sufficient certainty in time for preventative strategies to be put into place. For some examples of herbicide resistance development, full resistance is apparent within a small number of years (e.g. five years or less in many recorded cases for annual ryegrass, Lolium rigidum, in Australia). Even if the innovative technologies are triallable, few years of trialling are available before the potential for preventative action has past.

This problem is worsened by the difficulty of trialling some types of treatment recommended to delay resistance. The difficulties include the following:

(a) Where treatments are implemented in combinations (as they generally must be) it can be very difficult to determine from field observations alone the individual impacts of each element of the combination in order to assess their individual worthiness for inclusion in an integrated management system.

(b) Some treatments have impacts that are relatively difficult to observe even if implemented in isolation. For example, increasing the crop seeding rate affects the seed production of both crops and weeds (in opposite directions), and the latter is rather difficult to observe quantitatively in the field without tedious collection and counting of weed seeds in both standard and high seeding rate plots.

(c) The effectiveness of some alternative weed treatments is very sensitive to weather conditions or the quality of implementation, and so trials give highly variable results from time to time. Even if a treatment is beneficial in the long run, it may not appear so in a short-term trial, or it may take a long time before its value can be determined with adequate confidence.

For all of these reasons, rapid adoption of integrated weed management systems, involving combinations of unfamiliar, complex, and expensive treatments that are difficult to trial, is unlikely to occur until it is essential.

On the other hand, the nature of herbicide resistance is that, once it has developed, farmers have no choice but to alter their weed management systems. Thus, following the onset of resistance, most of the problems involved in encouraging farmers to change to some alternative system evaporate. The problem for farmers then becomes, which of the many possible alternative systems should best be adopted? The decision support system outlined in the next section is the outcome of one attempt to provide guidance on this question.

3. Herbicide Resistance as a Farming Systems/Resource Management Problem

There has been increasing interest in non-chemical weed control options in response to the threat of herbicide resistance (e.g. Boerboom 1999; Nalewaja 1999) but little work on their optimal economic integration into farming systems. Stewart (1993) and Schmidt and Pannell (1996a) applied early versions of the RIM (Ryegrass Integrated Management) model (Pannell et al., 1999), to explore the economic implications of herbicide resistance in Western Australia. RIM has since been substantially enhanced and expanded, and the 1999 version will be the basis for the analyses presented here. We will first present a very brief description of the model (readers are referred to Pannell et al., 1999, for more details) and then a selection of model results that illustrate key points about the economics of resistance management. Results are all based on a set of parameters specified for the eastern wheatbelt of Western Australia.

3.1 Model Description

RIM is a multi-period simulation model representing the biology, technology and farm-level financial aspects of ryegrass weed management in dryland farming systems of southern Australia. Its general features include:

• a 20 year time frame;
• representation of up to 7 different crop or pasture based enterprises in a user-specified sequence;
• 35 different weed control options that can be selected in any technically feasible combination and in any technically feasible sequence;
• detailed representation of relevant biological relationships;
• representation of relevant financial details, including input costs, output prices, machinery costs, interest rates, yields, and environmental costs for certain treatments;
• potential for the user to adjust all biological and economic parameters;
• ability to specify the herbicide resistance status of the ryegrass weeds with respect to each of seven herbicide groups.

The treatment options are shown in Table 1. There are 12 selective herbicide options, 5 non-selective herbicide options and 18 non-chemical options. They are listed in chronological order, based to the time when they would be implemented by farmers. Some of these treatment descriptions may not be clear. Detailed explanations for each are provided by Pannell et al. (1999). For those chemicals for which trade names are used, chemical names are provided in an appendix.

Table 1. Weed treatment options included in the RIM model.

* N = Non-selective herbicide, S = Selective herbicide, B = "Biological" treatment (non chemical)

The biological processes, variables and relationships represented include:

• the density of living, ungerminated seeds in or on the soil through each year;
• a germination pattern of weed seeds over time within each year;
• mortality or removal of seeds, seedlings, or adult plants or prevention of seed production through various processes, including both natural and human-applied mechanisms;
• inter and intra-species competition impacts on crop yields, and weed seed production;
• damage to crop production from weed control treatments, including phytotoxic damage from herbicides, and mechanical losses due to treatments such as green manuring;

3.2 Results

1. Impact of herbicide resistance on profit and optimal treatment strategies

Consider a comparison between two otherwise similar paddocks of arable land that differ in their herbicide resistance status. One has no resistance and, for illustrative purposes, is assumed to be immune from resistance development in future regardless of the pattern of herbicide use. The other has a severe herbicide resistance problem. Table 2 shows a comparison of economic returns and optimal treatment strategies for the two cases. In deriving these results:

• (a) it has been assumed that the ryegrass weeds in the second case are resistant to all "fop", "dim", triazine and sufonyl-urea herbicides;
• (b) the sequence of crop types has been left the same for both cases, a lupin, wheat, wheat rotation, which is widely practiced in the modelled region;
• (c) the "optimal" treatment strategies have been identified by a process of trial and error. They may not be globally optimal, but are at least close to this.

Table 2. RIM model results with and without herbicide resistance for lupin-wheat-wheat crop rotation

Several aspects of these results stand out.

• The onset of severe resistance results in substantial economic losses.
• After the onset of resistance, it is possible to maintain the continuous cropping rotation with a substantially alterred weed management system.
• The altered system involves a greater diversity of treatment types. Each of these is individually less effective than herbicides, so a greater number of treatments must be employed. The "with resistance" strategy is more strikingly consistent with what would be considered "Integrated Weed Management" (IWM).
• The "optimal" strategy with resistance present involves only a slightly greater average density of weeds. This is consistent with survey results that have found weed densities in farmers’ paddocks with herbicide resistance are, on average, no greater than in non-resistant paddocks. Thus the economic difference between the scenarios is not primarily due to differences in weed density, but to differences in total treatment costs.

2. Benefits and costs of conserving the herbicide resource

Evidence for herbicide resistance in ryegrass in Western Australia is that use of herbicides on ryegrass is similar in nature to extraction of a non-renewable resource. There is a limited stock of "herbicide susceptibility" and each usage of herbicide moves the farmer inexorably towards herbicide resistance, from which there is no return. Because resistant ryegrass plants are apparently no less fit than susceptible ryegrass, there is no reason to expect that cessation of herbicide use will reverse the process of resistance development to any extent.

Given the high cost of resistance development (Table 1) the question arises of whether it is economically preferable to pre-empt the onset of resistance by adopting a greater range of treatments at an early stage, thereby reducing the speed at which the stock of herbicide susceptibility is used up. There are several aspects of this question that need to be considered:

Just because the cost of resistance is high, it does not necessarily follow that early adoption of IWM is economically preferred.

Factors in favour of early adoption of IWM would include (a) preserving herbicides would help to avert a major increase in weed numbers late in the planning period, and (b) if the farmer obtains experience and expertise with IWM prior to the onset of resistance, the risk of losing control of the weed population at the time of resistance onset is reduced.

Factors against early adoption include: (a) if the weed population density is low at the time when resistance occurs, it is easier to maintain the weed population at low levels with the use of non-chemical treatments (e.g. increasing the crop seeding rate is more effective if the weed population is low), and (b) farmers can earn more interest from income that is earned sooner rather than later (i.e. the value of later benefits must be discounted relative to early benefits).

Existing evidence indicates that early adoption of IWM does not increase the total number of herbicide applications that are possible before resistance is fully developed. It just allows the farmer to spread out the fixed number of applications over a longer period.

Considering all these factors together, it appears unlikely that there would be a strong economic incentive either for or against early adoption of IWM. The results in Table 3 confirm that this is the case. Assumptions behind these results include:

• The allowed numbers of applications of herbicides from different groups prior to onset of resistance were: "fops" and "dims", 4; sulfonyl ureas, 2; triazines, 4.
• The lupin-wheat-wheat rotation is maintained in both secnarios.
• In order to avoid a strategy involving neglect of weed management at the end of the planning period, a constraint is imposed that final weed seed density must not exceed the starting weed seed density.

Table 3. RIM results with and without early adoption of Integrated Weed Management.

One of the considerations mentioned earlier was not captured in these model runs: that there may be reductions in risk if the farmer develops experience with IWM prior to onset of resistance onset. Were this factor to be properly valued, it is likely that there would be little difference in the economic values of early and late adoption of IWM.

This result should be considered illustrative only. It applies to a particular example and may not be transferable to others, depending on their biology and economics. Nevertheless, it does illustrate the potential for exploitation, rather than conservation, of the herbicide resource to be the optimal long-term strategy for farmers.

3. Implications of herbicide resistance for optimal land use

Results presented so far have all been for a particular cropping rotation. Severe resistance raises the prospect that changes in land use may be beneficial. In particular, in Western Australia, it has been argued that inclusion of an occasional pasture phase can be of benefit to weed management by allowing a greater range of weed treatments, employing methods that do not rely on selective herbicides, such as grazing by livestock. Of course, the economics of such a strategy depend not just on weed management considerations, but also depend on relative costs and returns of the alternative enterprises. Table 4 shows results for four additional strategies involving inclusion of pasture phases. They are directly comparable to the strategies shown in Table 3 except for the inclusion of the pasture phases. In each case, the pasture phases last for three years, which is considered the time necessary to reduce ryegrass seed densities substantially. The scenarios with two pasture phases include two groups of three years, within the 20-year planning horizon.

Table 4. Profit (A$, equivalent annual value over 20 years) estimated by the RIM model following inclusion of pasture phases

Given the costs and prices included in the model, profit from a strategy involving a single pasture phase is either similar to or slightly less than a strategy involving no pasture. Two pasture phases are less profitable for both scenarios regardless of the timing of IWM adoption.

Of course these results are dependent on product sale prices. The main product of pastures in the farming system modeled is wool, harvested from merino sheep. The 1999 wool price, which was used in the analysis, is at a relatively low level historically. If it were to increase, the economic attractiveness of pasture phases would rise accordingly.

4. Implications for farmer decision making

The following observations highlight the implications of these modeling results for likely farmer decision making.

The management of herbicide resistant weeds is complex, involving a greater range of more costly and unfamiliar weed treatments than farmers are initially used to. One would expect farmers to take some time to establish their preferred weed management strategy after the onset of resistance, following a period of trial and, inevitably, error.

There appears to be no compelling case for reducing the reliance on herbicides in order to delay the time when they will be lost to resistance. Even in cases where early adoption is in fact beneficial, it will be very difficult for farmers to determine that this is true. In the face of major uncertainties about such a change, and no compelling arguments in favour of it, most farmers are likely to maintain a more or less traditional, herbicide-based weed management system until forced to change by the full development of resistance. There is evidence that this is indeed what farmers tend to do (e.g. Peterson 1999).

4. Policy Implications of Herbicide Resistance

The threat of herbicide resistance, given its potentially severe economic consequences, raises the question of what government policies should be implemented in response. Economists recognise two morally respectable categories of possible reasons for governments to intervene in markets: market failure (efficiency related) and unacceptable distributional consequences (fairness related).

Market failure refers to several clearly defined circumstances where the operation of a free market may not lead to the most efficient outcome. Pannell (1994) reviews the full range of market failures and outlines their relevance to weeds in general. Several types of market failure may be relevant to the problem of herbicide resistance.

4.1 Externalities From Spread

Spread of resistance from farm to farm or from farm to the non-farm environment may occur by several mechanisms, causing "externalities" - costs external to the farmer who has the power to take action to reduce the risk of spread. Spread of resistant weed seeds can occur either naturally (e.g. by water, wind or animals) or carried by humans (e.g. in hay, or as a contaminant in purchased seed). It is also possible for resistance to spread via pollen blown by wind (Timmons et al., 1995) or carried by insect pollinators (Scheffler et al., 1993). Potential government policies in response to this include regulation of seed sellers, or of movements of agricultural produce containing resistant seeds. In practice, in Australia, given that virtually all farmers are using herbicides, they are almost all developing resistance problems of their own, regardless of any imports. Thus it has been difficult to detect any cases of farm-to-farm spread, because they are masked (and rendered irrelevant) by on-farm resistance development. Exception to this in agricultural areas where animal grazing is the dominant enterprise with the consequence that herbicides are used relatively infrequently. There are anecdotal stories of farmers using herbicide for the first time in a field and finding resistance, either because they sowed Lolium rigidum seed that was resistant (Lolium is a prized feed to animal producers, e.g. Abadi Ghadim and Pannell, 1991) or imported resistant seed in hay. Similarly, resistant Lolium rigidum seed has been sold in Spain and introduced resistance problems (P. Boutsalis, pers. comm., 1999).

For some cases, it is possible that externality problems from resistance spread may be important. Some of the issues associated with designing a mechanism to control problems with "tragedy of commons" due to spread are treated in the environmental economics literature (see Segerson, 1999). In general, when resistance build up is a shared problem within a community, the solution requires some form of collective action. Some economists have designed policy mechanisms intended to provide incentives for such collective action to occur.

4.2 Externalities From Practices Adopted in Response to Herbicide Resistance

Two of the weed control practices that are brought back into consideration by the onset of resistance are practices from which agriculture has been moving away: mechanical cultivation and burning (Schmidt and Pannell, 1996a). Externalities are part of the reason for the loss of favour of these practices. This is particularly the case for cultivation, as concerns about erosion and soil health have driven a world-wide trend among crop producers in developed countries towards various versions of low- or no-tillage production systems (Cornish and Pratley, 1987). Herbicides have been critical to the success of this trend, replacing previously high intensities of cultivation as the main method of weed control. Resistance clearly poses a threat to this trend. A potential government response is to ensure that farmers who do increase their level of cultivation are made to bear the burden of the external costs that result from greater movement of soils off the farm. Theoretically, an efficient way to do this is to impose a tax on soil loss. In practice, there are considerable difficulties in monitoring and evaluating the extent of such losses with sufficient accuracy to form the basis for imposing a tax. Economists recognise these difficulties as "transaction costs" and they are likely to be very high for most non-point sources of pollution, such as eroded soil particles. It may be necessary, then, to consider second-best approaches which are theoretically less efficient than a tax, but may be more efficient when transaction costs are considered. Possibilities include regulations to prohibit or limit use of certain farming practices, or subsidies to encourage particular practices.

4.3 Uncertainty/Information Failures

Herbicide resistance is technically complex and hard to observe until it has reached a high level, so misperceptions, ignorance and high levels of uncertainty are the norm for farmers facing herbicide resistance for the first time. Even once resistance has occurred, farmers may find it relatively difficult to reach a full understanding of its biological nature and of the effectiveness of the many possible combinations of available treatment options. Rogers (1995) identified triallability as a key characteristic of new technologies affecting their speed and level of adoption. Pannell (1999) argued that the potential to learn from trials is reduced for "sustainability" problems such as herbicide resistance, relative to more traditional agricultural practices with quick response times. This suggests the need for public support for research and extension, which have been subjects of substantial investment in Australia since herbicide resistance became prominent. Government involvement in such research and extension may also be justified by its "public good" nature. Pannell (1994) briefly describes the two types of public goods. Their most important feature in the current context is that part of the benefits resulting from the research and extension is of a nature that it cannot be traded in markets (consider the benefits from reduced soil particle loads in waterways).

4.4 Distributional Consequences

Pannell (1994) noted that

There is nothing in economic theory which helps us to objectively evaluate the relative merits of different distributions of wealth, although in practice redistribution is often one of the goals of government policy and action. In these cases the professional contribution of economists is limited to:

• quantifying the distributional effects of government action,
• quantifying the impacts of government action on efficiency, and
• assessing the performance of the action with regard to several rules of thumb which attempt to capture (or perhaps shape) community attitudes regarding what is fair and equitable.

It is therefore difficult to generalise about the implications of distributional impacts for policy formation, other than to observe that it further complicates the consideration of potential policy options. For example, the potential to apply a tax on soil loss to address external costs caused by herbicide resistance may be considered "fair" by some, since it allocates the cost to the person who is the source of that cost (i.e. it is consistent with the allocative rule of thumb called "polluter pays"). On the other hand, such a tax may be considered unfair, since it imposes an additional financial burden on a group of farmers who have already suffered financial losses from the development of herbicide resistance. The solution to this dilemma is necessarily a subjective judgement, made within a political context.

5. Conclusion

The existing literature on adoption of innovations by farmers provides a number of findings and insights that are highly relevant to farmers’ responses to the herbicide resistance problem. We have noted that herbicide resistance has a number of characteristics that make it rather unlikely for farmers to rapidly adopt strategies that would delay the onset of full resistance. Rather, they seem more likely to wait for resistance to occur, and then adjust their management practices accordingly.

Using a detailed bio-economic model, we have provided insights into the nature of these adjustments for a particular case study, relating to the farming system that provides the most extreme example of herbicide resistance development in the world. The model confirms the expectation that, with resistance present, profits are lower and the optimal weed management system is more diversified. It includes a greater range of non-chemical treatments than is the case for farming systems without herbicide resistance. Details of the exact changes were provided, although in practice these will vary from case to case.

Finally we discussed a range of policy implications of herbicide resistance. It appears that the externality problem from spread is, in many cases, not as severe as for resistance to insecticides or fungicides, but it may be an issue in some cases. The problem of uncertainty and information failures was discussed as an issue to which governments might justifiably respond, and the distributional implications of resistance were also considered as a policy issue.

Clearly herbicide resistance poses very serious challenges not only to farmers and scientists, but also to extension agents, consultants, and policy makers. Sensible responses to many of these challenges will require a good appreciation of the social and economic issues that we have presented here.

Appendix: Chemical Names

Simazine® simazine

Glean® chlorsulfuron

Hoegrass® diclofop methyl

Fusilade® fluazifop-butyl

Select® clethodim

Spray.Seed® paraquat + diquat

Gramoxone® paraquat

References

Abadi Ghadim, A.K. and Pannell, D.J. (1991). The economic trade-off between pasture production and crop weed control, Agricultural Systems 36(1): 1-15.

Abadi Ghadim, A.K. and Pannell, D.J. (1999). A conceptual framework of adoption of an agricultural innovation, Agricultural Economics 21: 145-154.

Bar-Shira, Z. Just, R. and Zilberman, D. (1997). Estimation of farmers' risk attitude: an econometric approach, Agricultural Economics 17: 211-221.

Binswanger, H.P. (1974). A microeconomic approach to induced innovation, Economic Journal 84, 940-958.

Boerboom, C.M. (1999). Nonchemical options for delaying weed resistance to herbicides in Midwest cropping systems, Weed Technology 13: 636-642.

Carlson, G.A., and Wetzstein, M.E. (1993a). Pesticides and pest management, In: G.A. Carlson, D. Zilberman, and J.A. Miranowski (eds.), Agricultural and Environmental Resource Economics, New York: Oxford University Press, pp.268-318.

Carlson, G.A., and Wetzstein, M.E. (1993b). Firm decisions and behavior in pest management on a regional level, In: G. A. Carlson, D. Zilberman, and J. A. Miranowski (eds.), Agricultural and Environmental Resource Economics, Oxford University Press, Oxford.

Caswell, M.F. and Zilberman, D. (1986). The effects of well depth and land quality on the choice of irrigation technology, American Journal of Agricultural Economics 68: 798-811.

Clark, S.J. and Carlson, G.A. (1990). Testing for common versus private property: The case of pesticide resistance, Journal of Environmental Economics and Management 19(1): 45-60.

Cochrane, W.W. (1979). The Development of American Agriculture: A Historical Analysis, University of Minnesota Press, Minneapolis.

Cornish, P.S. and Pratley, J.E. (eds.) (1987). Tillage: New Directions in Australian Agriculture. Inkata Press, Melbourne.

David, P.A. (1969). A contribution to the theory of diffusion, Stanford Center for Research in Economic Growth, Memorandum no. 7.

de Gorter, H. and Fisher, E.O. (1993). The dynamic effects of agricultural subsidies in the United States, Journal of Agricultural and Resource Economics 18: 147-159.

Gorddard, R.J., Pannell, D.J. and Hertzler, G.L. (1995). An optimal control model of integrated weed management under herbicide resistance, Australian Journal of Agricultural Economics 39: 71-87.

Gorddard, R.J., Pannell, D.J. and Hertzler, G.L. (1996). Economic evaluation of strategies for management of herbicide resistance, Agricultural Systems 51: 281-298.

Green, G.P. (1995). Technology Adoption and Water Management in Irrigated Agriculture, Dissertation, Agricultural and Resource Economics, University of California, Berkeley, Ca.

Green, G., Sunding, D., Zilberman, D., Parker, D., Trotter, C., and Collup, S. (1996). How does water price influence irrigation technology adoption? California Agriculture 50: 36-40

Griliches, Z. (1957). Hybrid corn: An exploration in the economics of technological change, Econometrica 25: 501-522.

Hayami, Y. and Ruttan, V.W. (1970). Factor prices and technical change in agricultural development: The United States and Japan, 1880-1960, Journal of Political Economy 78.

Heiman, A., and Muller, E. (1996). Using demonstration in new product diffusion - the theory of demonstration duration, Journal of Marketing Research, 33: 422.430.

Helms, G.L., Bailey, D. and Glover, T.F. (1987). Government programs and adoption of conservation tillage practices on nonirrigated wheat farms, American Journal of Agricultural Economics, 69: 786-795.

Hueth, D. and Regev, U. (1974). Optimal agricultural pest management with increasing pest resistance, American Journal of Agricultural Economics 56(3): 543-552.

Huffman, W.E. (1974). Decision making: The role of education, American Journal of Agricultural Economics 56: 85-97.

Knight, A.L. and Norton, G.W. (1989). Economics of agricultural pesticide resistance in arthropods, Annual Review of Entomology 34: 293-313.

Kotler, P. (1997). Marketing Management : Analysis, Planning, Implementation, and Control, 9th edition. Prentice Hall, Upper Saddle River, New Jersey, 789 pp.

Lindner, R.K. (1987). Adoption and diffusion of technology: an overview, In: Technological Change in Postharvest Handling and Transportation of Grains in the Humid Tropics, B.R. Champ, E. Highley and J.V. Remenyi (eds.), ACIAR Proceedings No. 19, pp. 144-151.

Orson, J. (1999). The cost to the farmer of herbicide resistance, Weed Technology 13: 607-611.

Marland, G. McCarl, B. and Schneider, U. (1999). Soil carbon: Policy and economics, In: N. Rosenberg, R.C. Izaurralde, and E.L. Malone (eds.), Carbon Sequestration in Soils: Science, Monitoring, and Beyond, Proceedings of the St. Michaels Workshop, December 1998, Battelle Press, Columbus and Richland.

Miranowski, J.A., and Carlson, G.A. (1993). Agricultural resource economics: An overview, In: G.A. Carlson, D. Zilberman, and J.A. Miranowski (eds). Agricultural and Environmental Resource Economics, New York: Oxford University Press, pp. 3-27.

Miranoswki, J. A., Hrubovak, J. and Sutton, J. (1991). The effects of commodity programs on resource use, In: N. Bockstael and R. Just (eds.), Commodity and Resource Policies in Agricultural Systems, Springer-Verlag, New York.

Moffitt, L.J. and Farnsworth, R.L. (1981). Bioeconomic analysis of pesticide demand, Agricultural Economics Research 33(4): 12-18.

Nalewaja, J.D. (1999). Cultural practices for weed resistance management, Weed Technology 13: 643-646.

Olmstead, A.L. and Rhode, R. (1993). Induced innovation in American agriculture: A reconsideration, Journal of Political Economy 101: 100-118.

Pannell, D.J. (1994). Economic justifications for government involvement in weed management: a catalogue of market failures. Plant Protection Quarterly 9(4): 131-137.

Pannell, D.J. (1999). Uncertainty and Adoption of Sustainable Farming Systems, Paper presented at the 43rd Annual Conference of the Australian Agricultural and Resource Economics Society, Christchurch, New Zealand, January 20-22 1999. http://www.general.uwa.edu.au/u/dpannell/dpap9901f.htm (50K)

Pannell, D.J., Stewart, V.A.M., Bennett, A.L., Monjardino, M., Schmidt, C.P. and Powles, S.B. (1999). RIM 99 User’s Manual, A Decision Tool for Integrated Management of Herbicide-Resistant Annual Ryegrass, University of Western Australia, Nedlands, Australia.

Peck, S.L. and Ellner, S.P. (1997). The effect of economic thresholds and life-history parameters on the evolution of pesticide resistance in a regional setting. The American Naturalist 149(1): 43-63.

Peterson, D.E. (1999). The impact of herbicide-resistant weeds on Kansas agriculture, Weed Technology 13: 632-635.

Putler, D.S. and Zilberman, D. (1988). Computer use in agriculture: Evidence from Tulare County, California, American Journal of Agricultural Economics 70: 790-802.

Regev, U., Shalit, H. and Gutierrez, A.P. (1983). On the optimal allocation of pesticides with increasing resistance: The case of alfalfa weevil, Journal of Environmental Economics and Management 10: 86-100.

Rogers, E.M. (1995). Diffusion of Innovations, Free Press, New York.

Scheffler J., Parkinson, R., and Dale, P.J. (1993). Frequence and distance of pollen dispersal from transgenic oilseed rape (Brassica napus). Transgenic Research 2: 356-364.

Schmidt, C. and Pannell, D.J. (1996a). Economic issues in management of herbicide resistant weeds. Review of Marketing and Agricultural Economics 64: 301-308.

Schmidt, C. and Pannell, D.J. (1996b). The role and value of herbicide-resistant lupins in Western Australian agriculture. Crop Protection 15: 539-548.

Segerson, K. (1999). Flexible incentives: A unifying framework for policy analysis, In:F. Casey, A. Schmitz, S. Swinton, and D. Zilberman (eds.), Flexible Incentives for the Adoption of Environmental Technologies in Agriculture, Kluwer, pp. 79-95

Smale, M., Singh, R.P. and Dubin, H.J. (1998). Estimating the economic impact of breeding nonspecific resistance to leaf rust in modern bread wheats. Plant Disease 82(9): 1055-1061.

Stewart, V.A.M. (1993). Economics of integrated strategies for managing herbicide resistant ryegrass under continuous cropping, Undergraduate dissertation, Faculty of Agriculture, University of Western Australia.

Stoneman, P. (1981). Intra firm diffusion, bayesian learning and profitability, Economic Journal 91: 375-88.

Tardif, F.J., Preston, C. and Powles, S.B. (1997). Mechanisms of herbicide multiple resistance in Lolium rigidum, In: Weed and Crop Resistance to Herbicides, Kluwer, Dordrecht.

Taylor, C.R. and Headley, J.C. (1975). Insecticide resistance and the evaluation of control strategies for an insect population, The Canadian Entomologist 107(3): 237-242.

Timmons, A.M., O'Brien, E.T., Charters, Y.M., Dubbels, S.J. and Wilkinson, M.J. (1995). Assessing the risks of wind pollination from fields of genetically modified Brassica napus, Euphytica 85: 417-423.

Uri, N.D. (1998). Conservation tillage and the use of energy and other inputs in U. S. agriculture, Energy Economics 20: 389-410

Vanclay, F. (1997). The social basis of environmental management in agriculture: A background for understanding Landcare, In: S. Lockie and F. Vanclay (eds.) Critical Landcare, Key Papers Series 5, Centre for Rural Social Research, Charles Sturt University, Wagga Wagga, pp. 9-27.

Wiebers, U.-C. (1992). Economic and Environmental Effects of Pest Management Information and Pesticides: The Case of Processing Tomatoes in California, PhD Dissertation, Tecnische Universität Berlin, Fachbereich Internationale Agrarentwicklung, Institue für Agrarbetriebs-Und Standortsokonomie.

Wolf, S.A. (1996). Privatization of Crop Production Information Service markets: Spatial Variation and Policy Implications, Dissertation, University of Wisconsin-Madison.

Wu, J. and Babcock, B.A. (1998). The choice of tillage, rotation, and soil testing practices: economic and environmental implications, American Journal of Agricultural Economics 80: 494-511.

Zilberman, D., Sunding, S., Dobler, M., Campbell, M. and Manale, A. (1994). Who makes pesticide use decisions: Implications for policymakers. In: W. Armbruster (ed). Pesticide Use and Product Quality, Glenbrook: Farm Foundation, pp. 23-39.

Citation: Pannell, D.J. and Zilberman, D. (2001). Economic and sociological factors affecting growers’ decision making on herbicide resistance. In: D.L. Shaner and S.B. Powles (eds.) Herbicide Resistance and World Grains, CRC Press, Boca Raton, pp. 251-277. (SEA Working Paper 00/07, Agricultural and Resource Economics, University of Western Australia).

The SEA News home page is now at http://www.crcsalinity.com.au/newsletter/sea/

Copyright © 2000 David Pannell and David Zilberman
Last revised: August 14, 2008.