The role of contractors in the uptake of precision farming—A spatial economic analysis

Abstract

Contractors will play a vital role in providing farms with access to new precision farming technologies, especially in small-scale farming systems. We investigate the impact of spatial competition among contractors on the uptake of precision farming and the effectiveness of policy interventions, considering alternative spatial price schedules. Conceptual analyses show that a lack of spatial competition among contractors hinders uptake of precision farming technology. The effectiveness of policy interventions to support precision farming also depends on the market structure and contractors’ price schedules. In addition, we illustrate the results in a Swiss case study based on a specific contractors’ service market of plant protection technologies.

1. Introduction

Precision farming is one step toward more sustainable agriculture (Walter et al. 2017). Inputs can be used more efficiently, reducing both farmers’ variable production costs and environmental footprints of farming, for example by reducing losses of nitrogen and pesticide use (Balafoutis et al. 2017; Weersink et al. 2018; Finger et al. 2019). Thus, the adoption and diffusion of precision farming techniques is of paramount political interest. However, the current adoption of precision farming practices differs widely across technologies and countries (e.g. Barnes et al. 2019; Finger et al. 2019). In particular, variable rate application technologies and similar capital-intensive techniques are rarely used on small farms and/or in the small-scale agricultural systems prevalent in Europe as the economic return is considered insufficient. Nevertheless, these technologies have the greatest potential to further sustainable intensification (Garnett et al. 2013). In European countries where small farm systems are common (Barnes et al. 2019; Finger et al. 2019), policy instruments have been implemented to support diffusion in order to counteract the imbalance between the limited adoption among small farmers who perceive inadequate private benefit and the potential public benefit of precision farming. In these systems, contractors who offer the service of bringing machinery onto farms will play a vital role in providing access to new technologies. The combined effects of contractors’ services and policy interventions could encourage widespread adoption, thus aligning the public and private benefits associated with precision farming (e.g. Busse et al. 2014).

In this paper, we investigate the influence of spatial competition between contractors providing precision farming services on (1) the uptake of precision farming and (2) the effects of policies, such as subsidies for precision farming practices.

The spatial competition framework is highly relevant in the context of contractors’ service markets. The high costs of transporting machinery to farms means that the market areas of many precision farming services are localized (Erickson et al. 2017).¹ Furthermore, contractors have to deal with peak workloads during certain periods in the crop cycle (e.g. the vegetation or harvesting periods). This often means that their capacities are stretched to the full and they can only offer their services close to their home base. Therefore, spatial proximity to service providers favors farmers’ access to, and uptake of, precision farming technologies (Khanna et al. 1999; Khanna 2001; Nguyen et al. 2020). At the same time, transport costs associated with contractors’ services may result in spatial market power (Hotelling 1929), which influences the contractors’ pricing strategies and therefore farmers’ uptake decisions. These features of general contractors’ service markets also apply to precision farming contractors’ services, which often involve special machinery and demand expert know-how and skills from the service provider.

Previous literature has documented that contractors may not only facilitate precision farming uptake but indeed be the ones to introduce it in the first place, especially in smaller scale agricultural systems prevalent in many European countries. For example, Reichardt and Jürgens (2009), Nguyen et al. (2020), and Bucci et al. (2019) show the willingness of German, French, and Italian farmers to adopt precision farming via contractors. However, there is still no formal economic framework that considers the spatial nature of these services when analyzing the role of contractors in precision farming uptake and the interactions with policy interventions. In addition, knowledge about the role of contractors in precision farming could be an important factor when assessing the scale and viability of small farm participation in sustainable agricultural intensification and could also help determine how policies could effectively support such participation.

We aim to help bridge this gap by conducting a conceptual investigation into the role of contractors’ services in the uptake of precision farming technologies in a spatial economic framework and the resulting interaction with policy interventions targeted at encouraging the uptake of these practices. Established frameworks for spatial competition and spatial pricing are applied to the context of contractors’ services in the field of precision farming technologies. We investigate two potential price schedules adopted by contractors: spatially nondiscriminatory pricing and discriminatory (uniform) pricing. We analyze the uptake of a given bundle of precision farming technologies within each price schedule and the effect of a subsidy under monopolistic and competitive markets. The results from the conceptual analyses are placed within a case study to obtain a numerical illustration of the relative economic significance of a subsidy in increasing uptake under different scenarios of spatial competition. The focus on technologies to lower pesticide use in our case study contributes to ongoing policy debates in Switzerland and Europe on ways of reducing the impact of plant protection on the environment and human health (e.g. Möhring et al. 2020).

Our analyses contribute to the existing literature by exploring the role of policy interventions in increasing the uptake of precision farming by means of purchasing contractors’ services in cases where the spatial competition between technology providers is relevant. Our findings underline that the effectiveness of policy interventions depends on the market structure and the contractors’ price schedules. This is an issue that warrants more investigation in agricultural markets (Russo et al. 2011; Graubner 2018).

We show theoretical evidence that higher spatial competition among contractors (as opposed to spatial monopolistic power and weak duopoly competition) reduces prices of precision farming services and thus facilitates overall technology uptake. Higher spatial competition also increases the extent to which subsidies for precision farming practices may enhance their uptake. In contrast, spatial monopolistic power can significantly diminish potential public benefits, such as reduced environmental damage, and render public policy intervention ineffective and inefficient. This is an important factor, since transport costs are expected to lead to spatial monopolistic power. In addition, given the same importance of space in the market, a spatially discriminatory pricing scheme (uniform pricing) is associated with higher uptake of precision farming technologies. If policy interventions aim primarily at increasing the uptake of precision farming, it follows that a uniform price schedule is more likely to achieve this goal. Our case study also sheds light on the possible relative economic significance of a policy intervention and a change in the intensity of spatial competition in the context of a specific contractor’s service market.

The remainder of this paper is organized as follows. Section 2 provides background information on precision farming, the market investigated in our case study, contractors’ service in agriculture, and spatial competition in agricultural markets. Section 3 presents conceptual analyses of the uptake of precision farming technologies and the effect of subsidies under spatially nondiscriminatory and discriminatory price schedules. Section 4 contains a numerical illustration of our findings using a case study of precision plant protection contractors’ service markets in Switzerland, and we submit our conclusions in Section 5.

2. Background and related literature

2.1 Background on precision farming

Precision farming involves a range of technologies that are applied throughout the crop cycle and allow for management of high inter- and intrafield spatial variability. This is steered mainly by the global navigation satellite system, which collects spatially explicit field information and enables targeted use of inputs. The European Parliament provides a list describing precision farming technologies and their respective objectives (European Parliament 2014). These technologies can be divided into several categories depending on the roles they play in forming management decisions and their level of complexity, including positioning (e.g. geo-referencing of field information), diagnostic and data management (e.g. soil sampling and yield monitoring technologies), and application (e.g. variable rate technologies) (Khanna et al. 1999).²

Evidence from farm-level surveys has indicated that, among other factors, economic incentives are an important driver of precision farming uptake (e.g. Daberkow and McBride 2003; Kutter et al. 2011; Paustian and Theuvsen 2017). Farmers tend to adopt technologies when they are affordable and cost-effective (Pathak et al. 2019). Therefore, precision farming uptake depends strongly on the characteristics of the technology and the farm (e.g. Khanna et al. 1999; Erickson et al. 2017; Paustian and Theuvsen 2017). In particular, since the initial investments associated with precision farming technologies are very capital-intensive, uptake via investment is limited mainly to large farms, while the expected per-hectare benefits are not high enough to be a worthwhile investment for small farms (Paustian and Theuvsen 2017; Wolfert et al. 2017; Weersink et al. 2018). Compatibility between precision farming equipment and conventional machinery and between different components of precision farming technologies is yet another major barrier for adoption (Kutter et al. 2011; Barnes et al. 2019; Groher et al. 2020). In addition, the lack of know-how and skills required to operate some of the complex precision farming technology and analyze the data collected also discourages adoption and would oblige farmers to make large investments (Khanna et al. 1999; Kutter et al. 2011; Barnes et al. 2019; Blasch et al. 2020). On a personal level, a farmer's decision to adopt precision farming technology also depends on his or her values and motivation. These are influenced not merely by subjective factors such as risk aversion, but also by other aspects such as farm size and production conditions (Pathak et al. 2019).

Despite the barriers to access precision farming technologies, which are particularly notable to small farms and small-scale agricultural systems, sustainable intensification of small farms is an important component of more sustainable agriculture (Garnett et al. 2013; Walter et al. 2017). Even though precision farming is only one out of many steps needed to make agriculture more sustainable, uptake by small farms could contribute to creating large public benefits due to reduced environmental impacts. In addition to providing public benefits, precision farming technologies are also useful to small farmers as they can help reduce input costs (e.g. Kutter et al. 2011; Busse et al. 2014) and facilitate compliance with regulations concerning the environmental impact of farming (European Parliament 2014).

Although various types of precision farming technologies are often inter-related and may be available to farmers simultaneously, previous research has shown that the different levels of technical complexity and measurability of value-adding effects lead to a sequential rather than a simultaneous uptake of bundles of these technologies (Khanna et al. 1999; Khanna 2001; Griffin et al. 2017). In particular, farmers tend to adopt more simple guidance and diagnostic tools before moving on to more advanced applicative tools (McCallum and Sargent 2008; Erickson et al. 2017; Weersink et al. 2018; Finger et al. 2019; Groher et al. 2020). For example, auto-guidance has been the most widely adopted type of precision farming technology, thanks to its directly measurable impact on farms (Balafoutis et al. 2017). In a study of farmers’ sequential choices of different bundle combinations of information-intensive precision farming technologies in Kansas, USA, Griffin et al. (2017) find that as of 2016 yield monitoring tools were the most widely adopted precision farming technology bundle, followed by yield monitoring combined with soil sampling (both are diagnostic tools). On the other hand, variable rate technologies (applicative tools) have the lowest adoption rates and are usually adopted in conjunction with other (diagnostic) tools.

Public interventions can play a role in lowering access barriers to precision farming technologies. This can comprise, for example, (1) provision of facilitating infrastructure and legal frameworks, (2) taxation on environmentally critical inputs, and (3) subsidies (Finger et al. 2019). According to a survey involving farmers and farm managers across five European countries, respondents rate a direct subsidy for the uptake of precision farming technologies and financial support in the form of tax relief to be among the most effective incentives for precision farming technology adoption (Barnes et al. 2019). For example, farmers participating in the PFLOPF project in Switzerland (discussed subsequently) receive subsidies for adopting precision farming technologies.

2.2 Background of case study: precision plant protection in Switzerland

The PFLOPF project in Switzerland (in German ‘Pflanzenschutzoptimierung mit Precision Farming’) is a typical example of the interactions between contractors’ spatial competition, public interventions, and the uptake of precision agriculture in the context of small family farms. This pilot project provides a specific case study to illustrate our economic analysis.

The PFLOPF project is an initiative funded by the Swiss Federal Office for Agriculture which aims to optimize and reduce the use of pesticides by applying precision farming technologies on a limited number of farms. The initiative reflects current efforts being made by policy makers, farmers, and industry in Switzerland to reduce the environmental and health risks caused by pesticide use (e.g. Böcker et al. 2019; Huber and Finger 2019; Möhring et al. 2020). The pilot project runs between 2019 and 2026 in the cantons of Aargau, Thurgau, and Zurich and comprises ca. 60 farms that are provided with different incentive schemes. Participants are subsidized for the use of precision plant protection technologies that meet the project requirements (a per-hectare payment is used to compensate the implementation of specific measures). The measures considered in the project comprise (1) automatic section or single nozzle control on pesticide sprayers, (2) camera hoeing services, both combined with a GPS steering system, (3) site-specific application of crop protection agents based on drone images, (4) use of warning and prediction systems in making pesticide use decisions (see e.g. Möhring et al. 2020), and (4) robot-based weed control and drone-based pesticide application in orchards and vineyards. Farmers can fulfill the requirements by installing their own devices or acquiring services from contractors. Since the use of these measures is associated with high investment costs for the individual (small) farmers, contractors’ services play an important role in the uptake of precision agriculture in Switzerland. Thus, this case study is well suited to illustrate aspects of public interventions when there is spatial competition in precision plant protection technology uptake via contractors’ services. Fig. 1 provides an overview of the study area and the contractors under the PFLOPF project.

As we shall discuss in the conceptual analyses, the spatial market structure determines contractors’ service prices and service areas, which in turn affect farmers’ uptake of precision farming technologies. For instance, additional contractors in the market would expand the total service area, and potentially lower the service prices in the case that they increase the level of spatial competition in the market. This would lead to an increase in farmers’ uptake. Furthermore, in the PFLOPF project, the use of contractors’ services is accepted under different policy incentive schemes that compensate participants for adopting precision farming technologies, for example a direct subsidy to farmers. In the case of an increase in the direct subsidy, farmers would be able to afford relatively higher prices for contractors’ services, which would also lead to higher uptake. We analyze these cases in the conceptual analyses (Section 3) and provide illustrations of the implications for uptake of the technology in the case study (Section 4).

2.3 Contractor markets in agriculture

Contractors’ services have been playing an important role in agriculture, especially over the last few decades (Igata et al. 2008; Defra 2014; Nguyen et al. 2020; Nye 2020). By paying contractors for their services, farmers can outsource certain tasks that may involve capital, labor, or a combination of both. In the case of small farms, outsourcing of specialized technologies, particularly those that involve machinery with high capital cost, may help lower the farm's overall costs (Picazo-Tadeo and Reig-Martínez 2006; Nye 2020).

Contractors not only help to reduce costs, but also facilitate the uptake of modern technologies such as precision farming and thereby contribute to reducing the environmental impact of agriculture by using more efficient machines (Defra 2014; Bucci et al. 2019). In particular, contractors’ services open a way to integrate small farms in agricultural intensification. By bringing the machinery required by precision farming technologies to the farm, contractors’ services allow small farms to access these technologies without having to invest in capital-intensive machinery and equipment. Contractors’ services, which include the implementation of the technologies on the farm and assistance with analyzing data for decision-making, can also contribute to knowledge gain, which helps to lower the technical barriers to precision farming uptake (Busse et al. 2014; Erickson et al. 2017; Groher et al. 2020). Thus, contractors’ services have the potential to mitigate the barriers for small farms to access precision farming technologies and therefore align the public and private benefits of such technologies.

Studies based on the US and European markets (Reichardt and Jürgens 2009; Erickson et al. 2017; Bucci et al. 2019; Nguyen et al. 2020) have documented the viability of accessing precision farming technology through contractors’ services. The relevance of contractors is more pronounced in European agricultural systems, which are usually characterized by small farms and small farm structures (e.g. Busse et al. 2014; Paustian and Theuvsen 2017). Recent studies show an increasing trend toward using contractors’ services for precision farming in some European countries (Bucci et al. 2019; Nguyen et al. 2020). In this context, technology sharing is also extremely important to promote the beneficial use of technology and encourage its diffusion in developing countries (Kirui and von Braun 2018; Finger et al. 2019; von Braun 2019).

There is very little information on the empirical characteristics of agricutural contractors’ service markets and in particular their pricing strategies. However, previous literature has shed some light on how these markets are organized and this underlines the relevance of the contractors’ spatial market structure. Spatial promixity is an important factor when a farmer chooses a contractor to acquire precision farming services (Khanna et al. 1999; Khanna 2001; Nguyen et al. 2020). A farmer can potentially achieve lower costs and receive higher quality of service if he or she works with a local contractor who is familiar with local conditions (Nguyen et al. 2020). This means that the availability of contractors’ services locally can have a strong influence on farmers’ decisions.

The contractors’ spatial competition has a marked influence on the range of working relationships that exist between contractors and farmers. This relationship can range from informal, based on social ties and trust, to fully formal, based on economic factors and convenience (Belton et al. 2021; Nye 2020). Especially for formal relationships, fear of dependency on a contractor for certain agricultural practices or technologies, for instance if the contractor is not available to provide the service, can hinder farmers’ adoption of these practices or technologies (Schneider et al. 2010).³ In cases like this, a competitive contractor market represents a safety net by offering continued service from alternative providers and it can encourage farmers’ adoption (Nye 2020). In addition, increased competition can also motivate contractors to cultivate their informal relationships with their clients and possibly reduce their prices to attract more clients (Nye 2020). Consequently, if the level of spatial competition among contractors is high enough, it could further incentivize farmers to hire contractors’ services.

Fig. 2 conceptually illustrates the relationship between the spatial distribution of contractors and farm access to precision farming technologies. If contractors are few and far between, farms can either be too far from a contractor to be served or face monopolistic prices (panel (A)). High spatial presence of contractors not only ensures that services are available, but can also lower the service prices as the market becomes competitive (panel (B)).⁴

2.4 Spatial competition and spatial price schedules in agricultural markets

The spatial dimension of a market is relevant whenever imperfect competition arises due to spatial interdependence in supply and/or demand (Greenhut, Norman, and Hung 1987). This applies to many agricultural markets where market power exists in the procurement market for agricultural products (e.g. Graubner et al. 2011a,b; Sesmero et al. 2015), or agricultural inputs such as land (e.g. Graubner 2018). In the context of agricultural technology services, contractors, as sellers of these services, possess market power due to their spatially distributed locations and the relevance of distance in farmers’ considerations when seeking services. In any given agricultural market, spatial competition and pricing strategies also interact with market regulations and policy interventions, such as subsidies (Russo et al. 2011; Sesmero 2016; Graubner 2018). Agricultural production generates significant environmental externalities and relies on natural resources. Therefore, the spatial market structure of agricultural markets also has an indirect influence on the environment and natural resource management and depends on farmers’ choices in production, such as land use, and intensity of resource use (Sesmero et al. 2015; Wang et al. 2020).

When modeling spatial competition and spatial pricing, the most relevant strand of literature for our work stems mainly from the model developed in Hotelling (1929) and further generalized in Smithies (1941). In the duopoly model, the market for a commodity is represented by a line segment, buyers are uniformly distributed along the line, and sellers are located at each end of the line segment. Buyers purchase the commodity from one of two sellers with perfectly inelastic demand and are responsible for transportation costs. The transportation cost introduces differentiation of products from the two sellers, who in turn possess market power and can set the price above the marginal cost as opposed to the (Bertrand) price game under perfect competition. The model in Hotelling (1929) is also extended to cover optimal firm location given that the competitor's location is fixed. However, in our study we regard contractors’ locations to be exogenous and only examine the price competition between them. In the Hotelling (1929) setting, price competition takes place in the form of nondiscriminatory spatial pricing,⁵ which means that sellers choose the optimal commodity price at their location to maximize their profit, while taking into account that buyers’ decisions are based on the total price composed of both the commodity price and transportation cost. Under spatially nondiscriminatory pricing, buyers bear the full transportation costs and so the local price faced by a buyer equals the sum of the price charged at the seller's location plus the transportation costs. This is the basic setting of the spatially nondiscriminatory price schedule under noncooperative competition in our conceptual analysis.

Alternatively, in a spatial competition setting, the price game may incorporate spatial price discrimination (Hoover 1937; Greenhut and Greenhut 1975). Instead of setting prices at their firm's location, sellers may choose ‘delivered prices’. A delivered price is the total price faced by a buyer and includes partial or full transportation costs. Under spatial price discrimination, differences in delivered prices do not fully reflect transportation costs as in the free-on-board case, though consumers at equal distance from a firm face the same delivered price (Phlips 1983). One spatially discriminatory price schedule typically considered in the literature on spatial competition is uniform pricing (also referred to as ‘uniform delivered pricing’). Under uniform pricing, transportation costs are fully embedded in the delivered price and averaged over buyers (e.g. Gronberg and Meyer 1981; Zhang and Sexton 2001). Therefore, with uniform pricing, the local price faced by buyers is equal to the delivered price charged by the seller and this price is uniform across the market served by the same seller.

Both nondiscriminatory and uniform pricing strategies have been widely documented in the literature on agricultural markets, where market power exists for both buyers (e.g. processing industries) and sellers (e.g. food manufacturing and retailing industries). Examples include raw milk markets (Alvarez et al. 2000; Graubner et al. 2011b), corn stover markets for biofuel processing (Sesmero et al. 2015), fresh produce markets (Sexton et al. 1991; Durham et al. 1996), and land rental markets (Graubner 2018). The choice of spatial pricing strategies depends on a range of factors, which includes industry characteristics and market conditions, such as the intensity of competition and the magnitude of transportation costs in relation to the value of the product (e.g. Holahan 1975; Capozza and Van Order 1977; Zhang and Sexton 2001; Fousekis 2011). Therefore, the equilibrium outcomes derived under spatial pricing models depend on the assumptions of the model chosen.

In terms of the pricing strategies of a precision farming contractor, previous literature suggests that there is unlikely to be a universal price schedule, but rather, as in other industries, it depends on the market conditions. Specifically, contractors’ spatial competition strategies may depend on the type of technology (e.g. the level of machinery intensity, and thus the magnitude of transportation costs in relation to the value of the contractors’ service) and the form of contracting (e.g. machinery and operation vs. only machinery rental). For example, Zhang and Sexton (2001) summarize that for agricultural markets, nondiscriminatory spatial pricing strategies are more likely to arise when transportation costs are low in relation to the product, whereas uniform pricing is more likely to apply when relative transportation costs are high. Moreover, there may be dynamic changes in the pricing strategies chosen by contractors in countries where the markets for precision farming technologies services are experiencing ongoing expansion. For instance, to fully utilize the capacity of their machineries and expand the service areas, contractors may travel long distances, or even outsource the transport and delivery of machinery to a third party; yet, as the intensity of competition changes over time, their service areas may change (Zhang et al. 2017; Belton et al. 2021). Under such circumstances, the pricing strategies would also vary.

Market conditions and, it follows, contractors’ price schedules are variable and therefore we expect farmers’ uptake decisions, and also the effects of policy interventions, to vary as well across different market scenarios. Consequently, we follow Graubner (2018) and adopt a comparative approach. In the conceptual analyses below, we consider the spatial price schedules that are most commonly applied in agricultural markets, namely nondiscriminatory and uniform pricing.⁶

3. Conceptual analyses of spatial market for precision farming contractors

In the following, we discuss spatial competition models that are relevant in the context of precision farming. Furthermore, we analyze farmers’ uptake under alternative scenarios for price schedules and market structures and the relative rise in uptake resulting from an increase in the subsidy.

We define the precision farming technologies to which our model is applicable. As discussed in the previous section, farmers’ uptake of bundles of precision farming technologies follows a sequential pattern. Based on this adoption behavior, we assume that the precision farming technologies within a bundle are relatively homogeneous and differ distinctly between bundles. Therefore, our model addresses adoption of standalone bundles of precision farming technologies. This means that individual technologies within a bundle are closely related, but as a whole are distinctive from, and not interchangeable with, other groups of technologies. Examples of bundles of technologies include machine guidance (e.g. in our case study, section or single nozzle control combined with GPS steering systems), soil sampling and testing, and variable rate technologies.

As discussed in the previous section, in the context of our study, a farmer's decision to adopt precision farming technology is whether to purchase rental services from contractors rather than to commit to a capital investment.⁷ The purchase decision can apply to a given unit of land; that is, farmers can opt to use precision farming technology on the whole farm or just a part of it. Therefore, we model farmers’ adoption as a binary choice problem, i.e. whether to purchase services from a contractor who offers a particular bundle of precision farming technologies for use on a given unit of land.

We analyze the extent to which different market structures and price schedules affect farmers’ uptake and subsidy pass-through under a given spatial distribution of contractors. While contractors may choose their location and service capacity strategically based on, for example, availability of farming infrastructure, they are not our focus for modeling contractors’ entry and expansion decisions. Rather, we take location and capacity as exogenous. Since we take service capacity to be exogenous and disregard expansion decisions, we can assume constant marginal costs for the contractor, which in this case is equal to average variable costs and includes personnel and machinery maintenance costs associated with each additional unit of service (see Section 3.4 for additional discussion on the assumption of constant marginal costs). Furthermore, in our main analyses we consider noncooperative competition; that is, we assume that contractors located close to each other do not cooperate with regard to price or service area, which would require commitment mechanisms that would fit a dynamic setting.⁸

Since precision farming technologies are expected to lower input use and/or increase yield, with other factors held constant, adoption occurs when the profit changes due to precision farming technology exceed the price charged by a contractor for the service. The change in profit is farm-, field-, and crop-specific, and profit change |$[ {p{\rm{\Delta y}} - {\boldsymbol{w}}{{\bf \Delta }}{\boldsymbol{x}}} ]$| can be estimated via an online savings calculator of a given technology, or based on returns achieved by other farms with similar characteristics. The profit change due to precision farming technology determines a reservation price for the contractors’ service, v. The profit change generated by adopting precision farming can be influenced by technological progress, which may improve the quality of service and therefore increase v.

In our model, a farmer's demand for contractors’ services on a given unit of land is price-inelastic: the farmer purchases the service when the sum of the reservation price and compensation from the subsidy is lower than the service price, that is when |$q < v + b$|⁠.

We illustrate the spatial market for precision farming contractors’ services using the spatial framework presented in Hotelling (1929), and a spatial duopoly to demonstrate spatial competition among contractors, whose locations are taken as given. The market is represented by a line segment with length l and two contractors, A and B, located at each end. Farmers, who are potential buyers of the service, are uniformly distributed along the market. Let d denote the location of a farmer relative to the contractors in the market; we set |${d_{\rm{A}}} = \ 0$| and |${d_{\rm{B}}} = \ l$|⁠.

This setup is apparently a simplification of real-world situations, in particular the empirical setting in our case study. For instance, Fig. 2 shows that contractors do not simply compete in one direction over a line-shaped market. Instead, the service area is two-dimensional, and under competition, it is usually the case that a contractor's service area is contested in some directions, and uncontested in others. Nonetheless, the setup of the conceptual analyses allows us to capture the key features of spatial pricing under different market structures (i.e. spatial monopoly vs. spatial competition), while keeping the models tractable. In terms of the difference in the dimensionality of the market in the conceptual model and the empirical setting, one can regard that in the empirical setting, the contractor acts as a duopolist in the contested directions and sets a price based on spatial competition, and acts as a monopolist in the uncontested directions, applying the same price. Therefore, it also suffices to consider that the contractors are located at the end of the line market in the conceptual model.

3.1 Nondiscriminatory spatial pricing

3.1.1 Formal model

In this subsection, our conceptual analyses are based on the spatial competition models of Hotelling–Smithies competition in the literature, in particular on the results from Mérel and Sexton (2010). First, we consider a spatially nondiscriminatory strategy whereby a contractor charges a price for a service and the farmer bears the transportation costs of bringing that service to the farm. The total cost to a farmer will depend on the distance between the farm and the contractor's location: for a farm at distance d from contractor A, the prices for services paid to contractors A and B would be |${r_{\rm{A}}} + dt$| and |${r_{\rm{B}}} + ( {l - d} )t$|⁠, respectively, with |${r_{\rm{A}}}$| and |${r_{\rm{B}}}$| being the service price charged at each contractor's location, and t being the unit transportation cost. The transportation costs are the sole determinant behind the farmers’ preference for services from contractor A or B.

For contractor|$\ i$|⁠, the profit function is |${\pi _i} = ( {{r_i} - c} )\ {D_i} - {{{OC}}_i}$|⁠, where |${D_i}$| is the farmers’ total demand for the service and c is the marginal cost of the service. |${D_i}$| depends on the price of the contractors’ service in relation to the farmers’ reservation price plus any subsidy payment. The contractor sets the service price that maximizes profit.

We define |$\sigma \ = \ tl$| as the maximum transportation cost across the market, which represents the absolute importance of space in the market (Zhang and Sexton 2001). |$\frac{\sigma }{{v + b - c}}$| represents the importance of space in relation to the maximum net revenue a contractor expects to receive. Both representations of the importance of space (absolute and relative) provide a measure of the intensity of spatial competition in the market: when the importance of space is high, markets are more likely to be separated and contractors have monopolistic power, whereas markets are more likely to be contested when the importance of space is low (e.g. Zhang and Sexton 2001; Graubner et al. 2011a; Graubner 2018 ). Consequently, the importance of space also defines the thresholds between different market structures, in our case from spatial monopoly to spatial duopoly.

3.1.1.1 Spatial monopoly

A spatial monopoly occurs when transportation costs are too high for the entire market to be covered, so markets are separated, and each farm can only be served by one contractor at most. The boundary of contractor i’s service area is marked by the location of the farm that faces a total price that is equal to the reservation price plus any subsidy: |${r_i} + td\ = \ v + b$| (Fig. 3A).

3.1.1.2 Spatial competition

In case of spatial competition, the service areas of the two contractors become contested. There exists a farm that faces the same local price from the contractors, and therefore the farmer is indifferent in purchasing service from either contractor. Given homogeneous services, the contractors compete in a symmetric duopoly, and therefore they set the same price to maximize profit and cover equally sized shares of the market, i.e. l/2 (Fig. 3B,C).

For the sake of completeness, we also discuss a case of nondiscriminatory pricing where both contractors set their price so that the farm that marks the boundary between two service areas faces a total price that is equal to |$v + b$| (Fig. 3B). This special case corresponds to a kink point in the demand curve and, in accordance with Mérel and Sexton (2010), is termed a ‘weak duopoly’.

The equilibrium prices allow us to analyze comparative statics under alternative policy scenarios, particularly the pass-through of a direct subsidy (or an increase in an existing subsidy) to the farmer, |$\frac{{\partial {r_i}}}{{\partial b}}$|⁠. The extent to which a subsidy is passed to the service price influences the market structure and may affect farmers’ uptake decisions. Previous literature has indicated that the effect of competition intensity on the pass-through of policies depends on market conditions, for instance the relative elasticity of supply and demand and the degree of price discrimination (Delipalla and Donnell 2001; Reny et al. 2012; Weyl and Fabinger 2013). In our setting, with inelastic demand and spatially nondiscriminatory pricing, under spatial monopoly, |$\frac{{\partial {r_i}}}{{\partial b}} = \frac{1}{2}\ ;$| that is, half of the subsidy would be passed to the contractor. Under strict duopoly, |$\frac{{\partial {r_i}}}{{\partial b}} = \ 0$|⁠, the farmer retains the whole subsidy. However, weak duopoly, |$\frac{{\partial {r_i}}}{{\partial b}} = \ 1$|⁠, means that the whole subsidy would be passed to the contractor.

Fig. 4 illustrates symmetric equilibrium prices under different market structures and with and without a subsidy payment, against importance of space. In the baseline scenario (i.e. without a subsidy), under strict duopoly (⁠|$\sigma \le \frac{2}{3}( {v + b - c} )$|⁠), equilibrium price decreases as the importance of space decreases. There is an inverse relationship between equilibrium prices and the importance of space over the range of weak duopoly, |$\frac{2}{3}( {v + b - c} ) < \sigma \le v + b - c$|⁠, which is due to the fact that competition is more fierce at remote locations than at those close to the contractor's base (Thisse and Vives 1988). The segment of the importance of space over which competition exists is extended when a direct subsidy b is introduced. In other words, a subsidy would increase the intensity of spatial competition. The prices in the dashed line in Fig. 3 illustrate the pass-through of the subsidy to price. In the strict duopoly segment, the farmer would retain the entire amount of subsidy; in weak duopoly, the entire subsidy would be passed to service price; and in monopoly, half of the subsidy is passed to service price.

3.1.2 Implications for uptake and subsidy pass-through

We summarize the results relevant to contractors’ services for precision farming below. Detailed analytical results are presented in Section A1 in the Appendix. As mentioned at the beginning of this section, our conceptual analyses represent a reasonable simplification of an empirical setting. While conceptual models offer a degree of analytical convenience, there is a subtle difference when it comes to implications for uptake in the empirical setting. Specifically, as shown above, service prices can be lower if levels of spatial competition are sufficiently high. In an empirical setting, this would imply that contractors have larger service areas resulting in higher uptake. In the conceptual model, however, lower service prices do not lead to higher uptake in the market between the two contractors, since the model implies full uptake whenever competition exists. Since our interest focuses on the role of spatial competition in the uptake of precision farming in an empirical setting, the results we present below reflect the relevant implications in the empirical setting (see e.g. Fig. 2), based on the equilibrium prices we derive from the conceptual model.

Result 1 (uptake under nondiscriminatory spatial pricing): As long as competition is sufficiently intense, uptake of precision farming under nondiscriminatory pricing would be higher under spatial competition than under spatial monopolies. However, if spatial competition is only moderate, uptake of precision farming could be lower under spatial competition than under spatial monopolies.

Result 2 (subsidy pass-through under nondiscriminatory spatial pricing): A subsidy would increase the intensity of spatial competition and thus enhance uptake. A subsidy would be most effective with sufficiently high spatial competition (zero pass-through to service price), moderately effective with spatial monopoly (50 per cent pass-through to service price), and least effective with moderate spatial competition (full pass-through to service price).

3.2 Spatial price discrimination (uniform pricing)

3.2.1 Formal model

In this subsection, our analyses draw on results from spatial competition models with spatial price discrimination presented in the literature (e.g. section 3.2.2 in Biscaia and Mota 2013 and section 3 in Zhang and Sexton 2001). Given uniform pricing, all farms face the same price |${w_i}$| from contractor i regardless of the distance to the contractor, and the contractor average transportation cost is embedded in the total price (i.e. farms close to the contractor are discriminated against). From a farmer at location d from contractor i, the price net of transportation cost received by the contractor is |${w_i} - td$|⁠.

3.2.1.1 Spatial monopoly

Monopolistic power means that the most remote location the contractor would serve is determined by the point where the net revenue just covers the marginal cost: |${w_i} - td\ = \ c$|⁠. That is, the maximum demand that the contractor can accommodate is |${D_i} = \frac{{{w_i} - c}}{t}\ $|⁠.

Section A2.1 of the Appendix shows that it is optimal for the monopolist to set the monopoly price as high as possible, i.e. at farmers’ reservation price plus any subsidy, |$v + b$|⁠. Local monopoly power implies that the markets are separated: |${D_i} \le \frac{l}{2}$|⁠, i.e. |$\sigma \ge 2( {v + b - c} ).$|

3.2.1.2 Spatial competition

When the service areas of two contractors are contested, each contractor's profit depends on his or her own price and that of the competitor. The contractor with a lower price |$w_i^ - $| will capture all of the demand to the extent that the contractor still receives a positive net price. If transportation costs prevent the contractor from serving the entire market, i.e. when|${\rm{\ }}w - \sigma \le c$|⁠, there is a residual demand |$l - D_i^ - $| that can be captured by the competing contractor, who acts as a local monopolist. As discussed in the literature on noncooperative games under uniform pricing, when two contractors set the same price, one has a motive to undercut the competitor, and thus matching the competitor's price is not an optimal strategy (e.g. Beckmann 1973). Since the expected profit functions for |${w_i} > {w_{ - i}}$| and |${w_i} < {w_{ - i}}$| are different (see Section A2.1 of the Appendix for the profit functions), the discontinuity of the profit in price implies that there is not an equilibrium solution with pure strategies (Dasgupta and Maskin 1986). Different strategies to address the nonexistence of a pure strategy equilibrium solution are possible. For example, an additional dimension of horizontal product differentiation can be incorporated (e.g. Anderson, Palma, and Thisse, 1989), or a mixed strategy game can be envisaged. The former strategy is not applicable in our context as contractors’ services are homogeneous. Shilony (1981) investigated a model closely related to our framework, though the detailed solutions to the symmetric mixed strategy (expressed in terms of a cumulative distribution function) are not informative in our specific context. However, according to Graubner (2018), for the purposes of our study, it suffices to examine at the upper and lower price limit.

When contractor i charges a higher price than the competitor and captures the residual demand, it is optimal to set the monopoly price; that is, |$w_i^1 = {\rm{\ }}v + b$|⁠. We present detailed derivation of the price limits in Section A2.1 of the Appendix.

When contractor i plays a lower price strategy |${w^ - }$| and sets the price at |${w^0}$|⁠, the share of total demand he or she would capture is given by |$D_i^0 = \frac{{\sigma - ( {v\ +\ b \ -\ c} ) + \sqrt {{{( {v \ +\ b \ -\ c} )}^2} \ +\ 2\sigma ( {v \ +\ b \ -\ c} )\ -\ {\sigma ^2}} }}{{2\sigma }}\ l$|⁠. It can be shown that |$D_i^0 > \frac{l}{2}$|⁠. In Section A2.1 of the Appendix, we show that |$\frac{{\partial D_i^0}}{{\partial ( {\frac{\sigma }{{v \ +\ b \ -\ c}}} )\ }} < 0$|⁠. That is, when one contractor sets prices at the lower price limit |${w^0}$|⁠, the share of the market facing this lower price increases as transportation costs decrease.

As in the spatially nondiscriminatory case in Section 3.1, we analyze comparative statics under alternative policy scenarios, particularly the pass-through of a direct subsidy at the price limits, |$\frac{{\partial {w_i}}}{{\partial b}}$|⁠. At the monopoly price or the upper price limit under spatial competition, |${w_i} = \ v + b$|⁠, and |$\frac{{\partial {w_i}}}{{\partial b}} = \ 1$|⁠. That is, the full subsidy would be passed to price. At the lower price limit under spatial competition (Equation 4), Section A2.2 of the Appendix shows that the share of a subsidy passed to price is positive and is an increasing convex function of the importance of space, |$\sigma $|⁠. As |$\sigma $| increases from 0 to |$2( {v + b - c} )$|⁠, |$\frac{{\partial {w_i}}}{{\partial b}}$| increases from 0 to 1. Fig. 5 illustrates the upper and lower price limits (horizontal and curved lines, respectively) under different policy scenarios. Since subsidy pass-through at the upper price limit depends on the relationship between |$v + b$| and c, the figure presents selected values of the relationship, namely |$v + b\ = \ 1.5c$|⁠, |$v + b\ = \ 2c$|⁠, and |$v + b\ = \ 3c$|⁠.

3.2.2. Implications of discriminatory spatial pricing for uptake and subsidy pass-through

As in the case of spatially nondiscriminatory pricing, the results presented below reflect implications in the empirical setting based on analytical results from the conceptual analyses. Detailed analytical results are presented in Section A2 in the Appendix.

Result 3 (uptake under uniform pricing): Uptake of precision farming under uniform pricing would be higher under spatial competition than under spatial monopolies. Uptake, via the contractor who offers a lower price, rises as the level of spatial competition increases in a mixed strategy game.

Result 4 (subsidy pass-through under uniform pricing): Subsidy pass-through rises as the level of spatial competition increases under uniform pricing. The effectiveness of a subsidy for farmers facing the lower competitive price limit ranges from very high (zero pass-through to service price) to very low (full pass-through to service price) as the level of spatial competition decreases. The subsidy is least effective in encouraging uptake in monopolistic markets (full pass-through to service price).

3.3 Comparison of uptake and subsidy pass-through under alternative price schedules

Fig. 6 shows the share of the market covered under both price schedules in the conceptual models, with and without subsidy increase. The entire market is covered by the competitive segment of the market under both price schedules (⁠|$\sigma \le v + b - c$|⁠). When |$v + b - c < \sigma \le 2( {v + b - c} )$|⁠, the market is separated between two monopolists and is not fully covered under spatially nondiscriminatory pricing, but continues to be competitive and fully covered under uniform pricing. This segment allows a comparison of the interaction between price schedule and market structure, that is monopoly under nondiscriminatory pricing and competition under uniform pricing, with the latter generating a higher uptake. When |$\sigma > 2( {v + b - c} )$|⁠, the market is separated between monopolists under both price schedules and the market area covered by uniform pricing is larger than that under nondiscriminatory pricing. Holding the importance of space constant, an increase in subsidy allows a greater share of the market to be covered when markets are served separately by monopolists.

Fig. 7 shows the share of a subsidy passed to price under both price schedules. Overall, spatial competition reduces the share of subsidy passed to price: under both price schedules the presence of competition means that a smaller share of a subsidy is passed to the price (excluding the special case of the weak duopoly) than in a monopoly market. Over the segment |$v + b - c < \sigma \le 2( {v + b - c} )$| with an interaction of price schedule and market structure (i.e. monopoly under nondiscriminatory pricing and competition under uniform pricing), a subsidy is more effective under uniform pricing (lower price limit) when transportation cost is moderate, and is less effective under uniform pricing when transportation cost is high. Fig. 7 also reveals that in case of uniform pricing with competition, the importance of space is relevant for subsidy pass-through to the lower price limit. By contrast, under nondiscriminatory pricing, subsidy pass-through is constant over each type of market structure. Intuitively, this relates to how the competitive prices are set under the two price schedules. In the uniform pricing setting, the market area covered by the contractor with the lower price depends on this price, and the profit from each location (i.e. price minus transportation costs) depends on farmers’ reservation price, subsidy and the importance of space. Therefore, the contractor's total profit (obtained by integrating profit over the market area) and the price that supports this profit depend on the interaction between the sum of the reservation price plus subsidy and the importance of space. This is reflected in the interaction terms of |$v + b$| and |$\sigma $| in Equation (4). Under nondiscriminatory pricing, each contractor covers a fixed share of the market and the profit from each location is constant as transportation costs are covered by farmers. Therefore, the contractor's total profit, and the price that supports this profit, is determined by the separate effects of the reservation price plus subsidy and the importance of space.

The comparison of the two price schedules is consistent with theoretical findings in the literature on spatial competition: when spatial competition is insignificant, spatially discriminatory pricing is associated with larger output and market area, but could lead to welfare loss for the buyers (Norman 1981). We further summarize the discussions in the results below.

Result 5 (uptake under alternative price schedules): Given an equal level of spatial competition, uniform pricing supports higher uptake than nondiscriminatory spacing pricing with a larger total service area and a larger share of contested markets.

Result 6 (subsidy pass-through under alternative price schedules): Subsidies are most effective under nondiscriminatory pricing when the level of spatial competition is very high, and least effective under nondiscriminatory pricing with a moderate level of spatial competition. If the level of spatial competition is very high or very low, a subsidy is less effective under uniform pricing compared with nondiscriminatory pricing (due to higher pass-through to service price).

3.4 Discussion of conceptual analyses

The results from our conceptual analysis shed light on the role of spatial competition between contractors in the uptake of precision farming by small farms, as well as the effectiveness of policies that promote sustainable agricultural intensification via precision farming operations.⁹ Due to the transportation cost of delivering precision farming technologies, the markets of contractors’ services are localized. Monopolistic power not only inhibits the demand for precision farming technologies to be fulfilled due to higher prices than under perfect competition, but also hampers the pass-through of subsidies aimed at increasing uptake to farmers. Increased availability of contractors’ services promotes spatial competition and thus facilitates the uptake of precision farming technologies and the provision of associated environmental benefits. It also allows farmers to reap greater benefits from subsidies.

The price schedule offered by contractors also matters to the uptake of precision farming technologies and the efficiency of policy interventions. Fig. 6 shows that uptake is high under both price schedules if spatial competition is strong, but farmers can only reap the full benefits from a subsidy under the nondiscriminatory price schedule. However, low levels of spatial competition lead to trade-offs between uptake and the extent to which farmers benefit from a subsidy—uptake is higher under uniform pricing, although a smaller share of the subsidy is transferred to farmers. With a moderate level of competition, uniform pricing is associated with both higher uptake and higher subsidy pass-through to farmers. Therefore, as uniform pricing supports higher uptake of precision farming technologies, it could be more effective in promoting the diffusion of these technologies and sustainable agricultural intensification. Moreover, the extent to which farmers benefit from a subsidy increases with the level of competition under uniform pricing.

Our conceptual analysis is conducted with contractor location and capacity exogenously determined. In the long run, dynamic interactions may occur between market conditions and contractors’ services. For instance, if contractors’ services are not available locally, larger farms may be encouraged to invest in machinery and enter the contractors’ service market. Lowered machinery investment and operational cost due to technological advancement may also encourage the entry of new contractors. Entries of new contractors would increase local spatial competition, and facilitate uptake of precision farming. Improved market conditions may also lead to expansions of existing contractors. Furthermore, technological advancement can lower the cost of machinery and equipment and/or increase the efficiency of operations, which is particularly important during periods of peak workload. Rising degrees of automation can reduce human resource costs associated with the precision farming services business. All these developments can result in changes in the relation between the marginal value product and the marginal cost of precision farming technology and increase contractors’ service capacity. If contractors achieve economies of scale via expansion, marginal cost would decrease instead of being constant. For the sake of model tractability, previous literature on spatial pricing (e.g. Greenhut and Norman 1986) has most commonly assumed constant marginal cost. However, exceptions to constant marginal cost under monopolistic markets have been discussed in Graubner (2020). In particular, with variable marginal cost, monopolistic prices are no longer independent of transportation costs (see the equilibrium monopolistic prices in Sections 3.1 and 3.2). In the case of economies of scale, the optimal monopolistic prices would increase with transportation costs, which implies lower uptake in the context of our study.

4. Case study: An illustration with precision plant protection service in Switzerland

In this section, we further use our case study to illustrate the results of our conceptual analyses in an empirical setting. The results from the conceptual analyses serve as our basis for estimating the uptake of a bundle of precision plant protection technologies via contractors’ services under different policy and spatial competition scenarios. The comparison across different empirical scenarios allows us to assess the relative economic significance of policy interventions under different contractor spatial market structures. As our case study estimates potential uptake of the technology via these contractors over the entire study area, it also provides an extrapolation of the coverage of the project, which is currently in a pilot phase. The illustration underscores the importance of considering spatial competition of contractors’ services in how policies could effectively support the uptake of precision farming technologies.

4.1 Information of contractors under the PFLOPF project

Our case study focuses on the potential uptake of a section or single nozzle control on pesticide sprayers combined with a GPS steering system in the scope of the PFLOPF project: this measure is widely used and is offered by contractors. We do not consider uptake by farmers purchasing their own machinery and equipment, as this is uncommon in Switzerland due to the small acreage for each farm (Groher et al. 2020). As of January 2020, there were ten contractors enrolled in the PFLOPF project¹⁰ who offered plant protection agent spraying services with automatic section or single nozzle control combined with a GPS steering system.

We surveyed the contractors’ pricing and service capacity information (in terms of normal and maximum travel distance) to improve our understanding of their spatial pricing strategies. Table 1 provides summary statistics and Table A1 in the Appendix provides detailed information on contractors’ travel distances. Normal travel distance indicates how far contractors travel to serve most of their clients, and thus gives an indication of their service radius. Normally, each contractor's service radius would depend on the service price. However, since travel distance information is not available from all contractors, we take12.5 km as the average normal travel distance to approximate the total service area. Fig. 8 plots the market boundaries with a radius of 12.5 km and each contractor's prices. This radius shows that about half of the three cantons’ area is covered by the contractors’ service areas, whereby farms outside of the service areas have no, or very limited, access to contractors’ services.

4.2 Scenarios and parameter calibration

We start by using contractors’ normal travel distance to estimate the potential uptake that the current contractors’ service market could support within the study area. Based on information provided by contractors, the majority charge all clients within the service area a flat rate, whereby one stated that extra charges can arise if they have to travel beyond their usual service area. As this suggests that the price schedule within a certain distance is consistent with uniform pricing, we focus on the uniform price schedule in the case study.

Based on the observed market conditions, we calibrate parameters which are then used to estimate uptake in the alternative scenarios. We consider our estimations as back-of-the-envelope calculations because first, the calibration involves applying information from the conceptual analysis, which is based on a one-dimensional market, to a two-dimensional market; second, some markets in the case study are oligopolies rather than duopolies; and third, there is no empirical information on certain parameters such as cost of service, which obliges us to make further assumptions on the relationship between farmers’ reservation price plus subsidy and the cost of service. In spite of this, we use information from the conceptual analysis, insofar as it is reasonable to apply it in an empirical setting, to illustrate the relative economic significance of policy interventions conditional on the market structure.

Since it is unlikely for a contractor to serve farms right in the vicinity of a competitor, even if the farms are located within his or her normal travel distance, we consider the normal travel distance as the boundary of a monopoly market area. That is, a contractor would only travel up to the normal travel distance in directions where there would be no other contractors operating in the area.

In the ‘Baseline’ scenario, we can calculate the potential uptake supported by the contractors, measured by the share of the study area that is covered by their service areas. Specifically, this is calculated by dividing the total area of arable land within the circles by the total area of arable land in the three cantons shown in Fig. 7. We also calculate the share of the aggregate service area that is contested. This is obtained by dividing the area of arable land within the overlapping parts of the circles by the total area of arable land within the circles in Fig. 7.

To estimate uptake under alternative scenarios, we first calibrate parameters in the ‘Baseline’ scenario and then apply the results of comparative statics from the conceptual analysis to infer uptake when a certain parameter value changes. We use the market of two contractors in Thurgau, who together form a duopoly, to calibrate parameters in the ‘Baseline’ scenario. In particular, we create a parameter of ‘empirical importance of space’ based on observed market conditions. This measures the extent to which the market is separated between two contractors. We use this parameter to induce transportation cost, which serves to quantify the comparative statics in alternative scenarios. In Table 2, we summarize the parameters calibrated based on this market. Detailed calculations are presented in Section A4 in the Appendix.

These parameter values serve as the basis for our study of three alternative scenarios reflecting alternative policy and spatial competition conditions. First, in the ‘Policy’ scenario, we consider a rise in the direct subsidy aimed at increasing the highest price farmers are willing to pay for the service. For the purpose of illustrating relative economic significance, we simply assume that the subsidy increase amounts to 10 per cent of farmers’ reservation price plus current subsidy, that is 10 CHF, and calculate the new market area served by contractors under the subsidy. Since |$R\ = \frac{{v \ +\ b \ -\ c}}{t} $|⁠, |$\frac{{\partial R}}{{\partial b}} = \frac{1}{t} $|⁠, and |${\rm{\Delta }}R\ = \frac{{10}}{4}\ = \ 2.5\ $|km. Therefore, with the subsidy, the contractors’ travel distance increases from 12.5 to 15 km. Expansion of the service area supports higher uptake in areas that were previously not served. This intensifies the overall competition throughout the entire market by increasing the share of the market areas being contested and in turn also implies lower service prices.

In the ‘Capacity’ scenario, we estimate how uptake would develop if contractors had higher service capacities. Technological progress can help increase service capacities. For instance, more efficient sprayers would mean that contractors need less time to finish a given task and could serve more customers. This is of great importance during peak seasons when pesticides must be applied within a certain time window. We represent the higher service capacity by extending the service radius from 12.5 to 16 km.¹¹ A higher level of presubsidy spatial competition implies lower importance of space: |${\gamma ^e} = \frac{{18}}{{16}}\ = \ 1.125$|⁠, and |$t\ = \ 3.125$| CHF.

In a final scenario, ‘Capacity & Policy’, we assess the joint effect of a new policy intervention combined with increased capacities. This involves the estimation of potential uptake when a subsidy is applied to markets where the intensity of spatial competition was higher than in the baseline scenario before introduction of the subsidy, that is with a service radius of 16 km. Therefore, the effect of a subsidy also increases: |${\rm{\Delta }}R\ = \frac{{10}}{{3.125}}\ = \ 3.2$| km. Compared with scenario 2, ‘Policy’, an identical subsidy leads to a larger expansion in the service radius, which implies a greater increase in uptake and lower service prices due to a build-up of spatial competition.

4.3 Discussion of case study results

As in the Baseline scenario, we calculate the potential uptake and the share of the aggregate service area that is contested based on the respective service radius. Table 3 summarizes the results based on the respective service radius calculated above and Fig. 9 illustrates the potential uptake according to market structure.

Both policy interventions and capacity expansion due to technological change can increase the level of spatial competition in contractors’ service markets by enlarging their market areas and facilitating higher uptake of precision farming technology. Given the current market conditions (‘Baseline’ scenario), contractors serve approximately half of the total area of the cantons in the PFLOPF project. Yet, the majority of the market is served by only one contractor. The conceptual discussion in Section 2.2 shows that if alternative options are not available, farmers may become worried about being dependent on contractors for the use of precision farming technology and this hinders uptake. A subsidy of 10 CHF per hectare could increase potential uptake by almost ten percentage points, with an increase in the fraction of the market that is served by at least two contractors by more than ten percentage points. If the baseline level of spatial competition (‘Capacity’ scenario) is higher, a subsidy has a stronger effect in increasing the potential uptake, namely by over ten percentage points. Furthermore, the joint effect of subsidy and capacity expansion allows over three-quarters of the total area to be covered by contractors’ services, with the majority facing competitive markets.

Our case study provides a numerical illustration of the extent to which spatial competition among contractors may influence precision farming technology uptake, as well as the effectiveness of policies that support uptake. In addition to the scenarios considered in the simulations, we discuss several aspects not directly addressed in the analysis, but which could help place our results in a broader context. First, farmers may also adopt precision farming technologies (and receive subsidies) by investing in machinery and equipment of their own, whereby this is not accounted for in the case study. An estimate of precision farming technology uptake attributable to contractors’ services can be refined based on the share of farmers who adopt comparable precision farming technologies through individual or joint investment. Thus, the uptake rates should be interpreted as lower bounds of the potential adoption. In this context, increased adoption rates using contractors’ services might also have a self-reinforcing effect by increasing peer learning among farmers. This can lower the knowledge barrier to the adoption of precision farming technologies, especially when this involves an investment. Second, for larger-scale farms in particular, a shortage of contractors’ services may prompt farmers to invest in machinery of their own and become providers of contractors’ service. In other words, in time, dynamic interactions may arise between the level of spatial competition among contractors and uptake and new contractors may emerge, further altering the spatial market structure. A future extension of our work would be to observe these new entrants in the contractors’ service market over time and investigate their motivation and their potential influence on uptake. For our current purpose of evaluating the overall effect of spatial competition intensity change, however, it is sufficient to create alternative scenarios of market structure by varying the service radius. In the long run, the expansion of service capacity could lead to economies of scale. Therefore, a further extension of our work would be to allow for economies of scale by relaxing the constant marginal cost assumption. As discussed in the Introduction, we expect the monopoly price to increase with transportation costs in this case. Finally, contractors’ services may also play a viable role in making use of new technologies for conservation purposes (e.g. as precision conservation). Thus, our analysis could be expanded beyond the more efficient use of inputs.

5. Conclusions

Contractors’ services can give farmers access to precision farming technologies without capital investment. Therefore, they have the potential to facilitate small farms’ participation in sustainable agricultural intensification and to help eliminate the imbalance between the private and public benefits of precision farming that often exist in small-scale agriculture. Due to costs associated with delivering these services, contractors’ local market power can dictate the effectiveness of their services in promoting precision farming uptake. In addition, the local market structure for contractors’ services may interact with policy interventions aimed at promoting precision uptake. This paper provides a conceptual investigation of the role of contractors in the uptake and diffusion of precision farming technologies, as well as the pass-through of a subsidy from a spatial competition perspective. Based on the results from the conceptual analysis, we use a case study of precision plant protection technology contractors’ services in Switzerland to illustrate the extent to which changes in spatial market structure and policy interventions can influence precision farming uptake. Overall, our analyses provide evidence that high local market power may hinder the diffusion of precision farming technologies, as well as the pass-through of a subsidy to farmers. Conversely, spatial competition between contractors is associated with lower prices and higher subsidy pass-through. Our case study also illustrates the possible relative economic significance of policy interventions under different levels of intensity of spatial competition. We conclude that increased intensity of spatial competition in the contractors’ service market can enhance both farmers’ uptake of precision farming technologies and the effectiveness of a subsidy. Against the background of alternative price schedules offered by contractors, which differ according to the presence of spatial price discrimination, our analyses reveal that overall uptake is higher when supported by the uniform price schedule under imperfect competition.

From a policy point of view, our results show that the lack of spatial competition among contractors can have a negative impact on potential public benefits, such as reduced environmental damage thanks to the use of precision farming technologies, e.g. lower losses of nitrogen and pesticides to the environment. Moreover, the lack of spatial competition among contractors can limit the effectiveness of policy interventions, such as subsidies designed to encourage the uptake of precision farming technologies and thus to contribute to sustainable agricultural intensification. Consequently, the optimal choices of policy instruments also need to account for the availability and structure of contractors’ services. We conclude that for the primary goal of increasing uptake, a uniform pricing strategy by the contractors is relatively more advantageous.

Our analyses open an avenue for further research on the interaction between contractors, policies, and precision farming adoption of small farms. This is highly relevant as the vast majority of farms globally are too small to directly invest in new technologies but rely on purchasing services or joint investments. The impact of spatial competition and dynamic effects must be tested empirically and a broad spectrum of policies (subsidies, taxes on inputs, interventions on contractor markets) must likewise be tested.

Acknowledgments

This research is supported by the Swiss National Science Foundation (SNSF), within the framework of the National Research Programme “Sustainable Economy: resource-friendly, future-oriented, innovative” (NRP 73), in the InnoFarm Project, Grant no. 407340_172433. Additionally, we would like to thank the anonymous reviewers for their helpful comments on a previous version of the manuscript.

Data availability

The data underlying this article are available in its online supplementary material.

Footnotes

References