TY - JOUR
AU - Artell,, Janne
AB - Abstract In order to integrate ecosystem services (ES) in designing agri-environmental policy, we investigated both the demand for, and supply of, ES from agricultural environments in Finland. Using the discrete choice experiment method, we measured citizens’ willingness to pay (WTP) for four different ES and analysed farmers’ compensation request (willingness to accept [WTA]) for producing these services. Biodiversity and water quality gathered the highest WTA of farmers, but also the highest WTP of citizens. Overall, the average WTA exceeded the WTP for almost all attributes and levels, but 20–27 per cent of farmers were willing to produce the ES with the compensation lower than citizens’ WTP. 1. Introduction Agricultural production faces versatile and often conflicting expectations. These include considerations related to the production of various ecosystem services (ES), such as food, pollination, landscape and climate services. Policy-makers should be able to integrate these different expectations into acceptable and applicable agri-environmental policy. This task will become increasingly difficult in the future, because in Finland, as in many other European countries, the public sector suffers from a fiscal sustainability gap. This paper explores and provides tools for integrating citizens’ and farmers’ preferences and values (related to agricultural production) into the design of agri-environmental policies to obtain more environmental benefits with lower costs for taxpayers. One solution to these challenges would be agri-environmental policy based on citizens’ and farmers’ values and preferences (Burton and Schwarz, 2013; Hasund, 2013; Schroeder et al., 2013) In the framework of ES, the primary goal of agriculture is to produce provisioning services, such as food. However, it is commonly recognised that agricultural environments also deliver cultural (such as enjoyment from landscape and recreation), regulating (such as control of climate and diseases) and supporting services (such as nutrient cycles) (de Groot, Wilson, and Boumans, 2002; Gobster et al., 2007; Power, 2010; van Zanten et al., 2014). Some of these services, and possible dis-services, are unintended side effects of the production of provisioning services. The purpose of agri-environmental policy is to develop incentives towards agricultural management that supports a broader range of ES (Prager, Reed, and Scott, 2012; Rey Benayas and Bullock, 2012). To support the efficient design of agri-environmental policies, knowledge of the value of the services provided by agricultural ecosystems is required. The current agri-environmental policy in the European Union (EU) is designed to encourage farmers to participate in voluntary agri-environmental schemes and to compensate them for the additional costs incurred by the implementation of agri-environmental measures as well as the income foregone due to any loss of profit (e.g. reduced production). The current schemes do not demand or ascertain the production of public goods or ES, i.e. farmers are not paid for achieving environmental outcomes but for implementing management practices. For example, farmers are currently compensated for providing water protection zones instead of outcomes of water protection, such as decrease in nutrient run-offs from their farm. Instead of management-oriented policies, new forms of policies in which farmers obtain income from the production of public goods, i.e. environmental outcomes, have been suggested. In some previous studies, these new policy initiatives have relied on environmental and other indicators of scheme success and have been discussed under the term of result-oriented or result-based policy (e.g. Burton and Schwarz, 2013; Herzon et al., 2018). Here, we stress that the design and legitimisation of such policies also require knowledge of how the various ES are valued by the final beneficiaries, i.e. citizens (van Tongeren, 2008; ENRD, 2010). Emphasis on benefits entails analysing how citizens weight the services and how they perceive their trade-offs, for example, how valuable possible improvements are in landscape compared to improvements in water quality. To underline the information on the values of beneficiaries as an important part of policy design, we use the term of benefit-based policy. Benefit-based policy implies that, in policy design, the environmental outcomes are in focus as in results-based policy, but it also emphasises the value of outcome for citizens. Instead, in the traditional cost-based policies, the focus of policy design is in compensations for farmers to cover the cost of environmental management practices. From the farmers’ point of view, the question concerning the feasibility of benefit-based policy is whether the compensation corresponding to the production of benefits is enough to motivate them to supply new types of service demanded by citizens. Previous studies on policies focusing on ES have demonstrated the importance of demand and supply information (Lima Santos et al., 2016). Few studies have empirically contributed to the design of benefit-based policies from both demand and supply perspectives, although such integrated analysis might provide a strong consultation basis in policy-making (e.g. Zasada, 2011; Castro et al., 2014; Nieto-Romero et al., 2014; Huang et al., 2015). Although agricultural ES are often supplied in multiple-service bundles, preferences are usually identified for a single provisioning, regulating or cultural service. Most previous studies have only addressed demand, neglecting the supply side, and have also concentrated on a single or occasionally on a few ES (Chen et al., 2017). Our paper aims to respond to these limitations by exploring both the supply of, and demand for, a bundle of ES from agriculture. The overall preferences of citizens and farmers for agri-environmental policy objectives in the form of ES are derived using the discrete choice experiment (CE) method. The CE method reveals citizens’ willingness to pay (WTP) for agricultural ES. The same method is used to evaluate the willingness of farmers to provide ES. In this case, farmers consider the amount of compensation needed, i.e. their willingness to accept (WTA), to produce the environmental outcomes in terms of ES. This study covers both the demand and supply sides of ES from agricultural land. First, we measure citizens’ WTP for four different ES from agricultural environments. Second, we analyse farmers’ WTA for producing the four services. In both, we apply coordinated CEs in such a way that the results can be used for aggregation to reveal the policy priorities for benefit-based future policies. 2. Previous studies on ecosystem service demand and supply from agri-environments Agro-ecosystems are human-managed ecosystems that play a crucial role as both a provider and consumer of multiple ES (Swinton et al., 2007; Zhang et al., 2007). They are socio-ecological systems that are multifunctional, including functions for food and fibre provision, and greatly interact with, and depend on, surrounding natural ecosystems (Huang et al., 2015). The provision of goods and services is a direct result of ecosystems influenced by farming activities, where the latter externally modify, improve or degrade the ES provision of agro-ecosystems (Dale and Polasky, 2007; Power, 2010; Zhang et al., 2007), but do not directly provide them. Agro-ecosystems provide a range of provisioning, regulating and cultural services to human society (Swinton et al., 2007; Power, 2010; Huang et al., 2015), while, due to their strong dependence on natural, unmanaged ecosystems, these systems require other regulating and supporting services to be productive. Given certain management practices, agro-ecosystems may also generate dis-services, i.e. negative effects from farming activities, such as nitrogen leaching and pesticide drift, the loss of habitat or sedimentation of waterways (Zhang et al., 2007). Integrated approaches suggest that the supply of, and demand for, ES should be analysed together in order to identify supply-demand mismatches that lead to the unsustainable and/or non-efficient management of ecosystems. Several frameworks for the integrated assessment of ES supply and demand are available in the literature (Wei et al., 2017). In all these frameworks, supply is measured in biophysical terms defined as ‘the components of a provided ecosystem based on biophysical properties, ecological functions and social properties in a particular area and over a given period’ (Wei et al., 2017: 16). These frameworks ignore the social and economic part of supply, i.e. how physical supply is affected by the acts and practices of farmers and by policy or market responses. Farmers face the trade-off between production of provisioning services, i.e. food and fibre, and the provision of regulating or cultural services to society (MEA, 2003; Gordon, Finlayson, and Falkenmark, 2010; Rodríguez et al., 2006; Power, 2010). Empirical studies that have quantified trade-offs are limited in number (Baldi et al., 2015). In order to govern the trade-offs and to target sustainable practices, it is imperative to better understand the interdependencies between various ES (Baldi et al., 2015) and to account for the views of farmers on aggregated/bundled ES (de Groot et al., 2002; Raudsepp-Hearne, Peterson and Bennett, 2010). The demand or the benefit side can be addressed by using non-monetary indicators (e.g. people’s perceptions of the importance of ES) and/or by using economic indicators derived from real or hypothetical markets (Turner, Morse-Jones, and Fisher, 2010; Martín-López et al., 2012). Usually, economic valuation of demand aims at revealing the WTP of citizens or beneficiaries in general for certain ES. The supply or cost side is related to farmers’ willingness to adopt management practices and farming procedures (e.g. organic farming or extensive management) that can promote ES, such as amenities, as well as soil and water protection (Zasada, 2011). An extensive list of studies have referred to the farmer uptake of voluntary agri-environmental measures and the factors that determine farmers’ willingness to implement such measures and consequently to supply ES (Grammatikopoulou, 2016; Siebert, Toogood and Knierim, 2006). The willingness to supply ES can also be measured in terms of farmers’ WTA a certain level of payments to adopt specific management practices. The outline of demand and supply will entail the identification (profile, preferences and valuation of ES) of beneficiaries, as well as that of providers, to ensure the socially efficient management of ES, solving the problems of under-provision or mismatching of ES (Pagiola, Arcenas and Platais, 2005). Stated preference studies including contingent valuation, conjoint analysis, CEs and contingent ranking (Huang et al., 2015) have been employed in assessing ES from agro-ecosystems. For agricultural ES, which are often examined in the framework of agri-environmental schemes, CEs can account for the complex characteristics (Bennett and Blamey, 2001; Hanley, Mourato and Wright, 2001) of the service in the sense that multiple options and several attributes are considered. Some studies have aimed at deriving a comprehensive picture of citizens’ preferences for agricultural ES. Novikova, Rocchi and Vitunskienė (2017) applied a CE in Lithuania to explore the preference of residents for the reduction of underground water pollution, preservation of biodiversity and sustenance and improvement of agricultural landscapes at the national scale, which revealed heterogeneity of preferences. A CE and latent class choice modelling were used to examine the demand for a range of agri-environmental services in Thailand from multifunctional agriculture (Sangkapitux et al., 2017). Dupras et al. (2018) applied contingent valuation and CE methods to value the impact of farming practices on landscape aesthetics in Canada. WTP for landscape aesthetics as well as water quality and fish diversity were found to be at a high level. The WTP for enhanced biodiversity of small forest patches in agricultural landscapes was examined in a study by Varela et al. (2018) through a CE. On the supply side, studies have mainly focused on farmers’ perceptions of ES rather than on economic assessment of the compensation required to produce the ES. Bernués et al. (2016), Smith and Sullivan (2014) and Xun et al. (2017) explored farmers’ knowledge of ES, interaction among them, perceptions of value and their relationships with certain practices. An interesting outcome of these studies is that, although farmers place a high value on ES, they perceive them to be only moderately manageable. Several studies have employed CE applications in eliciting farmers’ choices. Aslam, Termansen and Fleskens (2017) and Espinosa-Goded, Barreiro-Hurlé and Ruto (2010) revealed that farmers prefer to remain in a ‘business as usual’ state, showing a strong aversion to drastic changes in current activities. Some CE studies have concluded that the level of compensation is related to and differentiated according to farmers’ current management practices as well as the attributes of the new scheme (e.g. Espinosa-Goded et al., 2010; Vedel, Jacobsen and Thorsen, 2015; Villanueva et al., 2017). Broch and Vedel (2012), Christensen et al. (2011) and Ruto and Garrod (2009) have highlighted the relationship between the required compensation level and the scheme’s flexibility and administrative burdens. Previous literature has also indicated challenges in using WTA measure due to WTP/WTA disparity (Tunçel and Hammitt, 2014). Villanueva et al. (2017) revealed considerable heterogeneity among farmers in their preferences for agri-environmental schemes, which to a large extent could be explained by the specifics of the agricultural system (the type of joint production), but also by farm/farmer characteristics and farmer knowledge and perceptions. Broch et al. (2013) examined the relationship between farmers’ willingness to provide ES and the spatial heterogeneity associated with ES demand. WTA deviates in accordance with the ES in question, as revealed by Broch and Vedel (2012) who found that farmers accept a lower level of compensation when the aim is to protect biodiversity and groundwater relative to recreation. Latacz-Lohmann and Schreiner (2019) used an integrated approach and examined consumers’ WTP and producers’ WTA for higher animal welfare standards by using similar CEs for both respondent groups. However, related to ES, the literature still lacks studies that account for both demand and supply and conclude with holistic suggestions for policy-making. One example comes from Finland where both citizen and landowner preferences for one agricultural ecosystem service (landscape improvements) have been examined using a voluntary scheme (Grammatikopoulou, Pouta and Salmiovirta, 2013). The study concluded with clear suggestions for a locally implemented landscape value trade scheme. Target- or result-based schemes are structured based on the ES framework and on the evidence that ES include values that are measurable and visible in a demand-supply market context. This is one of the rare studies that empirically address both parts, i.e. demand and supply, to assist in the design of benefit-based measures. National-level studies are either ongoing or lacking. 3. Methods and data 3.1. Identifying ES for valuation We began the selection of agricultural ES for valuation by applying the Common International Classification of Ecosystem Services as a basis (CICES, 2016). CICES is a continuously developing European-wide classification system that can also be used for valuing ES. To select relevant ES provided by agricultural environments from the CICES classification, a literature review and the expert judgement of agricultural economists and ecologists were used. The selected services included food, agro-diversity, bioenergy, pollination, habitats for animal nursery and reproduction, pest control, soil productivity, cultural heritage, the existence of species and ecosystems, the recreation environment, landscape, water quality and climate change mitigation. In the valuation of agricultural ES with the CE method, it is not possible to include all of the various services agricultural environments provide. To choose the attributes for the CE from the 13 ES mentioned above, the following steps were performed by the project group: Analysis of the importance of the ES for citizens based on previous survey data (n = 800) (Pouta and Hauru, 2015) Evaluation of the importance of agri-environmental ES by stakeholders from the administration and NGOs (n = 6) Stakeholder (n = 7) discussion of the relevant ES based on step 2 A summary by researchers (n = 9) of steps 1–3 and analysis of market and non-market services, as well as final and intermediate services Evaluation by valuation experts (n = 10) of the questionnaire and the CE Attribute selection for the pilot study Pilot study (n = 202) A decision by researchers on the attributes in the valuation task of the final survey. The ES selected for the CE were landscape, the existence of species and ecosystems, water quality due to agriculture and climate change mitigation. In developing these selected ES into measurable attributes and their levels, the project group of environmental economists, ecologists and agri-environmental policy experts (n = 12) searched for concrete indicators that could be affected by farming practices and consequently targeted with agri-environmental policy. It was important to find reasonable attribute levels for both citizens and farmers separately, while making them as compatible as possible. The selected attributes and their descriptions for both citizens and farmers are presented in Appendix A. The different levels for the attributes are listed in Table 1 where level 0 represents the current state, i.e. the status quo option. Table 1. Attributes of agri-environmental policy programmes and their levels Ecosystem service . . Citizen survey . Farmer survey . Biodiversity Level 0 Present area (TRB), 0 species protected Present area of TRB Level 1 Area is increased by 30%, 100 species protected Area increased by 5 hectare Level 2 Area is increased by 60%, 200 species protected Area increased by 10 hectare Landscape: animals Level 0A Seldom seen Cattle, sheep and horses graze for under 3 months Level 1A Often seen during summer Cattle, sheep and horses graze for over 3 months Level 2A Often seen during summer and the unfrozen season Cattle, sheep and horses graze for over 6 months Landscape: plants Level 0P 3 species 3 species Level 1P 4 species 4 species Level 2P 5 species 5 species, of which one is a scenic plant (sunflower, corn etc.) Climate change mitigation Level 0 0% decrease in current emissions At least 20% of the area under cultivation with perennial plants Level 1 10% decrease in current emissions At least 40% of the area under cultivation with perennial plants Level 2 30% decrease in current emissions At least 60% of the area under cultivation with perennial plants Water quality effects Level 0 60% of surface waters in good or excellent condition The estate’s current nutrient flow Level 1 70% of surface waters in good or excellent condition 70% of the current nutrient flow Level 2 80% of surface waters in good or excellent condition 40% of the current nutrient flow Cost/Agri-environmental payment Levels €5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 130, 160, 190, 300, 500/taxpayer/year, during 2017–2026 €50,100, 200, 350, 550, 800 hectare/year during 2021–2027 Ecosystem service . . Citizen survey . Farmer survey . Biodiversity Level 0 Present area (TRB), 0 species protected Present area of TRB Level 1 Area is increased by 30%, 100 species protected Area increased by 5 hectare Level 2 Area is increased by 60%, 200 species protected Area increased by 10 hectare Landscape: animals Level 0A Seldom seen Cattle, sheep and horses graze for under 3 months Level 1A Often seen during summer Cattle, sheep and horses graze for over 3 months Level 2A Often seen during summer and the unfrozen season Cattle, sheep and horses graze for over 6 months Landscape: plants Level 0P 3 species 3 species Level 1P 4 species 4 species Level 2P 5 species 5 species, of which one is a scenic plant (sunflower, corn etc.) Climate change mitigation Level 0 0% decrease in current emissions At least 20% of the area under cultivation with perennial plants Level 1 10% decrease in current emissions At least 40% of the area under cultivation with perennial plants Level 2 30% decrease in current emissions At least 60% of the area under cultivation with perennial plants Water quality effects Level 0 60% of surface waters in good or excellent condition The estate’s current nutrient flow Level 1 70% of surface waters in good or excellent condition 70% of the current nutrient flow Level 2 80% of surface waters in good or excellent condition 40% of the current nutrient flow Cost/Agri-environmental payment Levels €5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 130, 160, 190, 300, 500/taxpayer/year, during 2017–2026 €50,100, 200, 350, 550, 800 hectare/year during 2021–2027 Open in new tab Table 1. Attributes of agri-environmental policy programmes and their levels Ecosystem service . . Citizen survey . Farmer survey . Biodiversity Level 0 Present area (TRB), 0 species protected Present area of TRB Level 1 Area is increased by 30%, 100 species protected Area increased by 5 hectare Level 2 Area is increased by 60%, 200 species protected Area increased by 10 hectare Landscape: animals Level 0A Seldom seen Cattle, sheep and horses graze for under 3 months Level 1A Often seen during summer Cattle, sheep and horses graze for over 3 months Level 2A Often seen during summer and the unfrozen season Cattle, sheep and horses graze for over 6 months Landscape: plants Level 0P 3 species 3 species Level 1P 4 species 4 species Level 2P 5 species 5 species, of which one is a scenic plant (sunflower, corn etc.) Climate change mitigation Level 0 0% decrease in current emissions At least 20% of the area under cultivation with perennial plants Level 1 10% decrease in current emissions At least 40% of the area under cultivation with perennial plants Level 2 30% decrease in current emissions At least 60% of the area under cultivation with perennial plants Water quality effects Level 0 60% of surface waters in good or excellent condition The estate’s current nutrient flow Level 1 70% of surface waters in good or excellent condition 70% of the current nutrient flow Level 2 80% of surface waters in good or excellent condition 40% of the current nutrient flow Cost/Agri-environmental payment Levels €5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 130, 160, 190, 300, 500/taxpayer/year, during 2017–2026 €50,100, 200, 350, 550, 800 hectare/year during 2021–2027 Ecosystem service . . Citizen survey . Farmer survey . Biodiversity Level 0 Present area (TRB), 0 species protected Present area of TRB Level 1 Area is increased by 30%, 100 species protected Area increased by 5 hectare Level 2 Area is increased by 60%, 200 species protected Area increased by 10 hectare Landscape: animals Level 0A Seldom seen Cattle, sheep and horses graze for under 3 months Level 1A Often seen during summer Cattle, sheep and horses graze for over 3 months Level 2A Often seen during summer and the unfrozen season Cattle, sheep and horses graze for over 6 months Landscape: plants Level 0P 3 species 3 species Level 1P 4 species 4 species Level 2P 5 species 5 species, of which one is a scenic plant (sunflower, corn etc.) Climate change mitigation Level 0 0% decrease in current emissions At least 20% of the area under cultivation with perennial plants Level 1 10% decrease in current emissions At least 40% of the area under cultivation with perennial plants Level 2 30% decrease in current emissions At least 60% of the area under cultivation with perennial plants Water quality effects Level 0 60% of surface waters in good or excellent condition The estate’s current nutrient flow Level 1 70% of surface waters in good or excellent condition 70% of the current nutrient flow Level 2 80% of surface waters in good or excellent condition 40% of the current nutrient flow Cost/Agri-environmental payment Levels €5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 130, 160, 190, 300, 500/taxpayer/year, during 2017–2026 €50,100, 200, 350, 550, 800 hectare/year during 2021–2027 Open in new tab In the farmers’ CE, some status quo (level 0 in Table 1) attribute levels were farm-specific. For example, the status quo level for the area of traditional rural biotopes (TRB) was given as the farmers’ current TRB area, which had been enquired in a preceding part of the survey. In addition, the reduction of nutrient runoff and current agri-environmental payment were case-specific. Crop producers and those in animal husbandry had different landscape attributes in the CE. The crop producers’ landscape attribute was crop diversity and that of the animal husbandry farmers was the length of the grazing season. The questionnaires were targeted at the groups of farmers based on their main production line that was obtained from the national farmer register, along with the contact information. 3.2. Surveys and CEs The citizen survey started with questions about personal relationship with agriculture and then proceeded to questions concerning attitudes towards agri-environmental issues, importance of different agricultural ES and how well Finnish agriculture has succeeded in producing ES. Next, citizen survey introduced a new benefit-based agri-environmental policy to the respondents by informing them that, in the hypothetical new programme, farmers would be paid for producing environmental benefits. The survey explained that the new agri-environmental programme would be financed with income tax. Depending on the extent of the programme, the cost to taxpayers would vary, but all taxpayers would participate in financing the programme. Respondents were informed that the current programme also causes expenses to citizens, amounting to approximately €40 per individual per year. This cost was based on expert judgement. Consequentiality was enforced by stating that the information from the choice tasks would help decision-makers to revise the agri-environmental programme. The farmer survey began with socio-demographic questions and background information on the farm. These were followed by questions about current agri-environmental compensation and attitudinal questions concerning agri-environmental schemes, as well as current and potential production of ES on the farm. The survey then suggested that the current agri-environmental scheme, which compensates additional costs and income foregone resulting from applying agri-environmental measures, would be replaced by a benefit-based agri-environmental scheme. The proposed programme would replace all other environment-related compensation currently paid to the farmers. The current agri-environmental scheme has been in place for 20 years, and 90 per cent of the farmers are included. There has been a strong public discussion of the in-efficiency of current agri-environmental practices in producing environmental outcomes. Consequentiality, important for incentive compatibility (Vossler, Doyon and Rondeau, 2012), was enforced in the farmer survey by informing about the need to renew the agri-environmental scheme for the next Rural Development Programme period starting in the EU in 2021. Both surveys informed the respondents about the suggested new agri-environmental scheme and the attributes as well as their levels (Table 1). Following the introduction of the attributes and the new benefit-based programme, the respondents in both citizen and farmer surveys were presented with six choice tasks. Each choice task had three alternatives: the status quo alternative, described as maintaining the current programme, and two alternatives with improvements in the state of ES. The alternatives were described with four ES attributes. Attributes had three different levels: status quo level as well as lower improvement and higher improvement. Status quo levels of different attributes also appeared in non-status quo alternatives. There was also a monetary attribute (cost for citizens and compensation for farmers) associated with each alternative. The status quo alternative was identical across choice tasks. Examples of choice tasks for citizens and farmers are presented in Appendix B. To allocate the attribute levels to the choice tasks in both citizen and farmer CE, we used efficient experimental designs. Efficient designs are used to generate parameter estimates with standard errors that are as low as possible and thus to obtain the maximum information from each choice situation (see Rose and Bliemer, 2009). The generation of efficient designs requires the specification of priors for the parameter estimates. In the design of the pilot surveys, we employed zero priors. In the final studies, however, we employed a Bayesian D-efficient design using Ngene (v. 1.0.2), taking 500 Halton draws for the prior parameter distributions, and parameter estimates obtained from the pilot study were used as priors. Bayesian efficient designs take into account the uncertainty related to the parameter priors. In the design of the CE for citizens, we used a Bayesian prior only for the number of cultivated plant species in the landscape and fixed priors for the other attributes. In the design of the farmer survey, we used Bayesian priors for all other attributes except for the bid level. In total, 36 choice tasks were generated and blocked in six subsets, which resulted in six choice tasks per respondent. For the citizens’ survey, four versions of the design were created using four different cost scales (€5–300, €5–500, €40–300 and €40–500). The design of the four versions was identical, aside from the varying cost scale. However, in this paper, the effect of differing cost scales is not examined. In the farmer survey, the compensation scale was €50–800. The final designs of the citizen and farmer CEs had D-errors of 0.08829 and 0.057962, respectively. Our survey design aimed at defining meaningful attribute levels and changes that were reasonable for both respondent groups. We also aimed to avoid vague qualitative descriptions of attribute levels. Although correspondence between samples was sought, the selected levels might have led to differences in the amount of change between citizens and farmers. For most of the attributes, we can conclude that the level of research information available is not comprehensive enough to guarantee the information bases to define the measures on farm level that would lead with certainty to particular environmental outcomes. The correspondence of attribute levels was also analysed by ex-post expert judgement. The levels for landscape were found to correspond with each other rather well. However, it was impossible to reliably compare the water quality effects based on existing knowledge, because one cannot directly deduce the ecological condition of waters from a reduction in agricultural nutrient runoff. The ecological status of surface waters is primarily defined based on biological quality factors (phytoplankton, other aquatic plants, fish and benthos). In addition to biological quality factors, nutrients, water quality and hydromorphological factors are also considered. There was also some uncertainty in the climate change attribute concerning the correspondence of citizens’ and farmers’ attribute levels. According to the expert knowledge, we can assume that the lower level of climate attribute in the farmer survey corresponded quite well with the citizen survey, but there is considerable uncertainty in the correspondence of the higher level of climate attribute. The uncertainty relates especially to peatland fields, where the carbon balance is very sensitive to different management practices. Another source of uncertainty is the end use of biomass from perennial plants. In the biodiversity attribute, the levels in the citizen survey could have been obtained if those farmers who are currently managing TRB increased their activity. Farmers producing TRB in areas that have not been managed by traditional methods in the past would not automatically lead to significant increases in biodiversity as the natural conditions may not be suitable for creating these ecosystems. Furthermore, there is also uncertainty in the knowledge on how establishing a traditional biotope in a typical field area would enhance the protection of endangered species. 3.3. Data 3.3.1. Citizen data The survey data of citizens (aged between 18 and 74) were collected using an Internet survey in the spring of 2016. The sample was drawn from the Internet panel of an independent market research company, Taloustutkimus, comprising over 30,000 respondents who have been recruited to the panel using random sampling to represent the population (Taloustutkimus, 2017). A pilot survey (n = 202) was used to test the questionnaire, especially the attributes and levels in the CE. For the final study, a random sample of 8,391 respondents was selected, of whom 2,066 completed the survey, resulting in a response rate of 25 per cent. Comparison of the socio-demographics of the sample with the population indicates that the proportion of females was lower, the respondents were slightly older and more highly educated and the proportion of people with children was a little higher compared with the population based on one-sample z-test (Table 2). However, most of these differences were small. Table 2. Descriptive statistics (n = 2,066) . Sample . Population (age 18–74)* . z-test p-value . Proportion of females, % 44 50 0.000 Mean age, years 53 48 0.000 Proportion of people with a higher educational level, % 37 24 0.000 Proportion of people with children (<18 years) in the family, % 26 24 0.000 Proportion of people living in Southern Finland, % 52 52 0.928 . Sample . Population (age 18–74)* . z-test p-value . Proportion of females, % 44 50 0.000 Mean age, years 53 48 0.000 Proportion of people with a higher educational level, % 37 24 0.000 Proportion of people with children (<18 years) in the family, % 26 24 0.000 Proportion of people living in Southern Finland, % 52 52 0.928 *Statistics Finland (2015). Open in new tab Table 2. Descriptive statistics (n = 2,066) . Sample . Population (age 18–74)* . z-test p-value . Proportion of females, % 44 50 0.000 Mean age, years 53 48 0.000 Proportion of people with a higher educational level, % 37 24 0.000 Proportion of people with children (<18 years) in the family, % 26 24 0.000 Proportion of people living in Southern Finland, % 52 52 0.928 . Sample . Population (age 18–74)* . z-test p-value . Proportion of females, % 44 50 0.000 Mean age, years 53 48 0.000 Proportion of people with a higher educational level, % 37 24 0.000 Proportion of people with children (<18 years) in the family, % 26 24 0.000 Proportion of people living in Southern Finland, % 52 52 0.928 *Statistics Finland (2015). Open in new tab 3.3.2. Farmer data The quantitative farmer data were collected in January 2017 using an Internet survey. The sample was drawn from the farm business register of the Agency for Rural Affairs. An e-mail invitation was sent to 5,000 farmers. The sample consisted of 3,449 farms with crop production as the main production line and 1,551 farms focused on animal husbandry. After two reminders, we received 591 usable responses. The response rates of the crop producers and animal husbandry farmers were 13 and 11 per cent, respectively. The questionnaire was tested before the main study in a pilot survey (n = 98, response rate 10 per cent) and in several expert interviews. Descriptive statistics of the farmer sample are compared with the whole population, i.e. all farmers in Finland, in Table 3. Most of the statistics for the sample are close or equal to the population, and the representativeness of the sample was thus satisfactory. Table 3. Descriptive statistics of the farmer sample (n = 591) and the whole farmer population of Finland (N = 49,982) . Sample . Population* . Mean age, years 52 51 Mean acreage of agricultural land, hectare 31 54 Organic farming 9% 9%** Participating in an agri-environmental scheme 89% 88% Crop production 45% 35% Other plant production 13% 27% Greenhouse production 0% 2% Outdoor production 1% 3% Milk production 12% 15% Beef production 4% 6% Other cattle husbandry 1% 1% Pig production 3% 1% Poultry production 1% 1% Other grazing livestock 5% 5% Mixed production 9% 4% . Sample . Population* . Mean age, years 52 51 Mean acreage of agricultural land, hectare 31 54 Organic farming 9% 9%** Participating in an agri-environmental scheme 89% 88% Crop production 45% 35% Other plant production 13% 27% Greenhouse production 0% 2% Outdoor production 1% 3% Milk production 12% 15% Beef production 4% 6% Other cattle husbandry 1% 1% Pig production 3% 1% Poultry production 1% 1% Other grazing livestock 5% 5% Mixed production 9% 4% *Natural Resources Institute Finland. **Finnish Food Safety Authority, Evira/Finnish Organic Food Association Pro Luomu. Open in new tab Table 3. Descriptive statistics of the farmer sample (n = 591) and the whole farmer population of Finland (N = 49,982) . Sample . Population* . Mean age, years 52 51 Mean acreage of agricultural land, hectare 31 54 Organic farming 9% 9%** Participating in an agri-environmental scheme 89% 88% Crop production 45% 35% Other plant production 13% 27% Greenhouse production 0% 2% Outdoor production 1% 3% Milk production 12% 15% Beef production 4% 6% Other cattle husbandry 1% 1% Pig production 3% 1% Poultry production 1% 1% Other grazing livestock 5% 5% Mixed production 9% 4% . Sample . Population* . Mean age, years 52 51 Mean acreage of agricultural land, hectare 31 54 Organic farming 9% 9%** Participating in an agri-environmental scheme 89% 88% Crop production 45% 35% Other plant production 13% 27% Greenhouse production 0% 2% Outdoor production 1% 3% Milk production 12% 15% Beef production 4% 6% Other cattle husbandry 1% 1% Pig production 3% 1% Poultry production 1% 1% Other grazing livestock 5% 5% Mixed production 9% 4% *Natural Resources Institute Finland. **Finnish Food Safety Authority, Evira/Finnish Organic Food Association Pro Luomu. Open in new tab 3.4. Statistical models A mixed logit model (MXL) takes into account respondent heterogeneity by allowing parameter values to vary across the respondents according to a pre-specified distribution. MXL is a highly flexible model and enables efficient estimation when there are repeated choices by the same respondents (Revelt and Train, 1998). The MXL model also resolves the problem of the independence of irrelevant alternatives (IIA) as it does not require this assumption. In the modelling of both the demand for and supply of ES, monetary variables were treated as continuous variables and the other attributes were coded as dummy variables. We also included an alternative specific constant for the status quo (ASC SQ), having the value 1 when a respondent chose the status quo alternative and 0 otherwise. In the estimation, the distributions must be imposed for each of the random parameters. All programme attributes and the alternative specific constant (ASC) for the status quo were treated as random variables with normal distributions. The cost and compensation parameters were specified as fixed. Specifying cost as a random parameter can cause problems in the estimation of WTP, as WTP is the ratio of the attribute’s coefficient to the price coefficient (Hensher, Rose and Greene, 2015). When both coefficients are allowed to vary, the distribution of WTP is quite complex as it is no longer just the scaled distribution of the attribute’s coefficient (Train, 2003). Selecting the distribution for the price coefficient is not straightforward and can lead to WTP distributions that do not have defined moments or they can be heavily skewed (Hole and Kolstad, 2012), implying extremely high WTP. This is why we used fixed parameters for cost and compensation, even assuming that there is no heterogeneity among the respondents in relation to price is somewhat unrealistic. 4. Results The results of MXL models for both citizens and farmers are presented in Table 4. In the citizen model, most of the ecosystem service parameters were statistically significant and of the expected sign, excluding the number of cultivated plant species in the landscape. There were no clear tendencies in the choice of policy alternatives, as the ASC SQ, i.e. the current programme, was not significant. The cost was significant and negative, meaning that an increase in the cost decreases the utility. Level 2, i.e. greater improvement, was preferred for animals in the landscape, climate regulation and water conditions. However, for biodiversity, level 1 was preferred. Biodiversity (level 1) and water conditions (level 2) had the greatest effects on utility. Table 4. Demand for, and supply of, agricultural ES . . Citizens . Farmers . . . Mean . Standard . Mean . Standard . . . . deviation . . deviation . ASC (SQ) 0.239 0.776** (0.00) (0.380) Cost/Agri-environmental payment −0.009*** 0.005*** (0.147) (0.000) Biodiversity Level 1 0.707*** 1.171*** −1.572*** 1.757*** (0.066) (0.080) (0.208) (0.221) Level 2 0.368*** 2.227*** −2.068*** 1.903*** (0.086) (0.102) (0.243) (0.270) Landscape: animals Level 1 0.411*** 0.977*** 0.025 1.822*** (0.071) 0.096 (0.208) (0.212) Level 2 0.587*** 1.331*** −0.798*** 2.446*** (0.076) 0.083 (0.232) (0.276) Landscape: plants Level 1 0.082 0.928*** — — (0.063) (0.094) Level 2 0.068 1.161*** — — (0.064) (0.084) Climate change mitigation Level 1 0.297*** 1.421*** −0.421** 1.239*** (0.076) (0.081) (0.191) (0.211) Level 2 0.417*** 1.689*** −1.141*** 1.752*** (0.079) (0.094) (0.222) (0.232) Water quality effects Level 1 0.434*** 1.517*** −1.346*** 1.769*** (0.072) (0.081) (0.237) (0.211) Level 2 0.719*** 0.877*** −1.370*** 2.137*** (0.071) (0.103) (0.250) (0.261) N 2066 456 log likelihood −11473.417 −2119.340 LR chi2(10)a(8)b 1723.67 496.25 Prob < chi 0.000 0.000 . . Citizens . Farmers . . . Mean . Standard . Mean . Standard . . . . deviation . . deviation . ASC (SQ) 0.239 0.776** (0.00) (0.380) Cost/Agri-environmental payment −0.009*** 0.005*** (0.147) (0.000) Biodiversity Level 1 0.707*** 1.171*** −1.572*** 1.757*** (0.066) (0.080) (0.208) (0.221) Level 2 0.368*** 2.227*** −2.068*** 1.903*** (0.086) (0.102) (0.243) (0.270) Landscape: animals Level 1 0.411*** 0.977*** 0.025 1.822*** (0.071) 0.096 (0.208) (0.212) Level 2 0.587*** 1.331*** −0.798*** 2.446*** (0.076) 0.083 (0.232) (0.276) Landscape: plants Level 1 0.082 0.928*** — — (0.063) (0.094) Level 2 0.068 1.161*** — — (0.064) (0.084) Climate change mitigation Level 1 0.297*** 1.421*** −0.421** 1.239*** (0.076) (0.081) (0.191) (0.211) Level 2 0.417*** 1.689*** −1.141*** 1.752*** (0.079) (0.094) (0.222) (0.232) Water quality effects Level 1 0.434*** 1.517*** −1.346*** 1.769*** (0.072) (0.081) (0.237) (0.211) Level 2 0.719*** 0.877*** −1.370*** 2.137*** (0.071) (0.103) (0.250) (0.261) N 2066 456 log likelihood −11473.417 −2119.340 LR chi2(10)a(8)b 1723.67 496.25 Prob < chi 0.000 0.000 Note: ** and *** denote variables that are significant at the 5% and 1% levels, respectively.MXLs in the preference space for citizen and farmer data. aDegrees of freedom in the citizen model. bDegrees of freedom in the farmer model. Open in new tab Table 4. Demand for, and supply of, agricultural ES . . Citizens . Farmers . . . Mean . Standard . Mean . Standard . . . . deviation . . deviation . ASC (SQ) 0.239 0.776** (0.00) (0.380) Cost/Agri-environmental payment −0.009*** 0.005*** (0.147) (0.000) Biodiversity Level 1 0.707*** 1.171*** −1.572*** 1.757*** (0.066) (0.080) (0.208) (0.221) Level 2 0.368*** 2.227*** −2.068*** 1.903*** (0.086) (0.102) (0.243) (0.270) Landscape: animals Level 1 0.411*** 0.977*** 0.025 1.822*** (0.071) 0.096 (0.208) (0.212) Level 2 0.587*** 1.331*** −0.798*** 2.446*** (0.076) 0.083 (0.232) (0.276) Landscape: plants Level 1 0.082 0.928*** — — (0.063) (0.094) Level 2 0.068 1.161*** — — (0.064) (0.084) Climate change mitigation Level 1 0.297*** 1.421*** −0.421** 1.239*** (0.076) (0.081) (0.191) (0.211) Level 2 0.417*** 1.689*** −1.141*** 1.752*** (0.079) (0.094) (0.222) (0.232) Water quality effects Level 1 0.434*** 1.517*** −1.346*** 1.769*** (0.072) (0.081) (0.237) (0.211) Level 2 0.719*** 0.877*** −1.370*** 2.137*** (0.071) (0.103) (0.250) (0.261) N 2066 456 log likelihood −11473.417 −2119.340 LR chi2(10)a(8)b 1723.67 496.25 Prob < chi 0.000 0.000 . . Citizens . Farmers . . . Mean . Standard . Mean . Standard . . . . deviation . . deviation . ASC (SQ) 0.239 0.776** (0.00) (0.380) Cost/Agri-environmental payment −0.009*** 0.005*** (0.147) (0.000) Biodiversity Level 1 0.707*** 1.171*** −1.572*** 1.757*** (0.066) (0.080) (0.208) (0.221) Level 2 0.368*** 2.227*** −2.068*** 1.903*** (0.086) (0.102) (0.243) (0.270) Landscape: animals Level 1 0.411*** 0.977*** 0.025 1.822*** (0.071) 0.096 (0.208) (0.212) Level 2 0.587*** 1.331*** −0.798*** 2.446*** (0.076) 0.083 (0.232) (0.276) Landscape: plants Level 1 0.082 0.928*** — — (0.063) (0.094) Level 2 0.068 1.161*** — — (0.064) (0.084) Climate change mitigation Level 1 0.297*** 1.421*** −0.421** 1.239*** (0.076) (0.081) (0.191) (0.211) Level 2 0.417*** 1.689*** −1.141*** 1.752*** (0.079) (0.094) (0.222) (0.232) Water quality effects Level 1 0.434*** 1.517*** −1.346*** 1.769*** (0.072) (0.081) (0.237) (0.211) Level 2 0.719*** 0.877*** −1.370*** 2.137*** (0.071) (0.103) (0.250) (0.261) N 2066 456 log likelihood −11473.417 −2119.340 LR chi2(10)a(8)b 1723.67 496.25 Prob < chi 0.000 0.000 Note: ** and *** denote variables that are significant at the 5% and 1% levels, respectively.MXLs in the preference space for citizen and farmer data. aDegrees of freedom in the citizen model. bDegrees of freedom in the farmer model. Open in new tab In the farmer model (Table 4), the ASC for the status quo was significant and positive, indicating that the respondents preferred the status quo, not the alternatives with increased ecosystem service production together with a certain amount of compensation. The share of serial non-respondents (i.e. respondents always choosing the status quo option) was 22 per cent of the farmer respondents. On the other hand, the share of those respondents who did not choose the status quo option in any choice set was 20 per cent. Level 1 of the landscape attribute was not statistically significant, but level 2 was. In all other attributes, the lower levels were also statistically significant and the signs of the attributes were negative, as expected. In the biodiversity improvement and climate change mitigation attributes, level 1 was preferred to level 2. This result is in accordance with the expectation that larger changes require higher compensation. In the nutrient flow attribute, the coefficients of the levels 1 and 2 were very close to each other, indicating that the respondents did not react to the differences in the required reduction in nutrient run-off. The standard deviations for all random parameters were statistically significant in both the citizen and farmer models. This implies that there is heterogeneity between the respondents over the mean parameter estimates (Hensher et al., 2015). For citizens, level 2 of the biodiversity attribute had the largest standard deviation, indicating that preferences for greater improvement of this attribute varied the most among the respondents. However, the standard deviation was lowest for the greater improvement in water quality, whereas this was the attribute that had the highest level of heterogeneity amongst farmers. Overall, the standard deviations were slightly lower in the citizen model. The CE for citizens included two landscape attributes, grazing animals in the landscape and diversity of cultivated plants, whereas farmers had either of these two attributes based on their main production line. The farmer model presented in Table 4 is a joint model for crop producers and animal husbandry farmers1, and the coefficient of the landscape attribute can thus be interpreted as an average of lengthening the grazing season and increasing the number of plant species in cultivation. Crop producers comprised 69 per cent of the respondents, which corresponds well with their share of all Finnish farmers, and thus, increasing plant diversity dominates the result. The citizens’ WTP and farmers’ WTA for different attributes and their levels were calculated based on the MXL models. The results are presented in Table 5. WTP ranged between 31 and 76 euros per taxpayer per year. WTP was highest for a greater improvement in water quality effects and animals in the landscape, as well as for a lower improvement in biodiversity. Farmers’ WTA figures ranged between 81 and 397 euros/hectare/year. The WTA was lowest for the lower improvement in climate change mitigation, i.e. increasing the area cultivated with perennial plants, and highest for a greater improvement in biodiversity, with a 10-hectare increase in the TRB area. Table 5. Willingness-to-pay (€/year) and willingness-to-accept (€/hectare/year) estimates from MXL models (95 per cent confidence intervals*) Ecosystem service . . Citizens’ WTP . Farmers’ WTA . . . (€/year) . (€/hectare/year) . Biodiversity Level 1 75.71 302.24 (61.07–89.55) (214.96–387.71) Level 2 39.38 396.93 (21.28–55.45) (307.88–510.04) Landscape: animals Level 1 44.04 — (28.90–58.30) Level 2 62.87 153.46 (49.00–78.76) (63.97–249.68) Landscape: plants Level 1 — — Level 2 — — Climate change mitigation Level 1 31.81 80.88 (15.38–47.66) (11.35–153.37) Level 2 44.66 219.43 (28.75–60.47) (130.99–305.24) Water quality effects Level 1 46.43 258.74 (31.60–61.68) (162.99–348.91) Level 2 76.98 263.42 (62.53–91.29) (168.57–367.01) Ecosystem service . . Citizens’ WTP . Farmers’ WTA . . . (€/year) . (€/hectare/year) . Biodiversity Level 1 75.71 302.24 (61.07–89.55) (214.96–387.71) Level 2 39.38 396.93 (21.28–55.45) (307.88–510.04) Landscape: animals Level 1 44.04 — (28.90–58.30) Level 2 62.87 153.46 (49.00–78.76) (63.97–249.68) Landscape: plants Level 1 — — Level 2 — — Climate change mitigation Level 1 31.81 80.88 (15.38–47.66) (11.35–153.37) Level 2 44.66 219.43 (28.75–60.47) (130.99–305.24) Water quality effects Level 1 46.43 258.74 (31.60–61.68) (162.99–348.91) Level 2 76.98 263.42 (62.53–91.29) (168.57–367.01) *Calculated with the Krinsky and Robb method. Open in new tab Table 5. Willingness-to-pay (€/year) and willingness-to-accept (€/hectare/year) estimates from MXL models (95 per cent confidence intervals*) Ecosystem service . . Citizens’ WTP . Farmers’ WTA . . . (€/year) . (€/hectare/year) . Biodiversity Level 1 75.71 302.24 (61.07–89.55) (214.96–387.71) Level 2 39.38 396.93 (21.28–55.45) (307.88–510.04) Landscape: animals Level 1 44.04 — (28.90–58.30) Level 2 62.87 153.46 (49.00–78.76) (63.97–249.68) Landscape: plants Level 1 — — Level 2 — — Climate change mitigation Level 1 31.81 80.88 (15.38–47.66) (11.35–153.37) Level 2 44.66 219.43 (28.75–60.47) (130.99–305.24) Water quality effects Level 1 46.43 258.74 (31.60–61.68) (162.99–348.91) Level 2 76.98 263.42 (62.53–91.29) (168.57–367.01) Ecosystem service . . Citizens’ WTP . Farmers’ WTA . . . (€/year) . (€/hectare/year) . Biodiversity Level 1 75.71 302.24 (61.07–89.55) (214.96–387.71) Level 2 39.38 396.93 (21.28–55.45) (307.88–510.04) Landscape: animals Level 1 44.04 — (28.90–58.30) Level 2 62.87 153.46 (49.00–78.76) (63.97–249.68) Landscape: plants Level 1 — — Level 2 — — Climate change mitigation Level 1 31.81 80.88 (15.38–47.66) (11.35–153.37) Level 2 44.66 219.43 (28.75–60.47) (130.99–305.24) Water quality effects Level 1 46.43 258.74 (31.60–61.68) (162.99–348.91) Level 2 76.98 263.42 (62.53–91.29) (168.57–367.01) *Calculated with the Krinsky and Robb method. Open in new tab As the WTP and WTA estimates were not directly comparable (WTP was for taxpayer per year but WTA was for farmer per hectare), we aggregated the total WTP and WTA estimates (Figure 1). For the aggregation of citizens’ WTP, we used the Finnish population over 18 years (4,431,392 in 2016). The aggregated WTP for different ES ranged from 141 million euros to 341 million euros. For the aggregation of farmers’ WTA, the total area enrolled in the current agri-environmental scheme (2,278,500 hectares) was used. The marginal, per hectare WTA figures were multiplied by this number to produce the aggregated WTA per year for certain attributes. Total area enrolled in the current scheme was used for aggregation, as it was the best and most justified estimate. However, it is possible that by using benefit-based policy, the environmental benefits could be obtained from a smaller area and therefore at lower cost as the measures would be undertaken where they actually are beneficial. Fig. 1. Open in new tabDownload slide Aggregated WTP and WTA and 95 per cent confidence intervals for different attributes. * In the survey of citizens, the landscape was divided into two attributes, whereas in the farmer study, livestock producers had an attribute concerning the grazing period and crop producers concerning the number of different plants cultivated in one season Fig. 1. Open in new tabDownload slide Aggregated WTP and WTA and 95 per cent confidence intervals for different attributes. * In the survey of citizens, the landscape was divided into two attributes, whereas in the farmer study, livestock producers had an attribute concerning the grazing period and crop producers concerning the number of different plants cultivated in one season Figure 1 presents the aggregated WTP and WTA for each attribute and level. The high demand for biodiversity, in particular on the lower level, faces a high demand for compensation from the farmers’ side. Citizens’ WTP for biodiversity does not follow the monotonicity assumption, as the WTP for level 1 is higher than for level 2. This could be due to the fact that some of the respondents may have considered 60 per cent increase in TRB area as too large and been concerned about the area left for food production. However, as the current area of TRB is only 1 per cent of the total area of agricultural lands, the area would remain low even with the higher increase. The compensation request for water quality benefits is about two or three times as much as the citizens’ WTP. WTPs and WTAs for climate regulation on a lower level, as well as landscape benefits from grazing animals, approach each other. In the landscape attribute, the citizens’ aggregated WTP for a lower level even exceeds the farmers’ compensation demand. It should be noted, however, that the farmers’ WTA for the lower level landscape attribute was linearly interpolated from the larger level change as the lower level landscape attribute was not statistically significant in the farmer model. In this case, we perceive that this method produces an acceptable, conservative estimate for WTA for a lower landscape change. Figure 2 presents the aggregated WTP and WTA for three scenarios. The first scenario sets all the attributes on their lower level. It reveals that the aggregated compensation request of farmers is about twice as high as the aggregated WTP of citizens. The WTP of citizens increases only slightly for the scenario that raises all the attributes to their highest level. Instead, farmers perceive the burden of increased service provision, showing an approximately 50 per cent higher demand for compensation than in the lower level scenario. The scenario with the highest net benefits was formulated by selecting those attribute levels in which the difference between the aggregated WTA and WTP was as low as possible. This means the lowest level for other attributes except water conservation. For this scenario, the costs also exceeded the benefits. The comparison of aggregated WTP and WTA suggests that none of the scenarios should be implemented as such. Fig. 2. Open in new tabDownload slide Scenario comparisons based on annual aggregated WTP and WTA estimates Fig. 2. Open in new tabDownload slide Scenario comparisons based on annual aggregated WTP and WTA estimates To analyse whether a share of the farmers would be willing to provide the services with the compensation level corresponding to the citizen-aggregated WTP, Figure 3 presents farmers’ individual-level WTA estimates for the highest net benefit scenario. The estimates were calculated with individual-level coefficients. The figure also presents the aggregated WTP of citizens divided by the number of hectares in the agri-environmental scheme to make them comparable with the WTA per hectare in monetary terms. The aggregated WTP per hectare for the scenario with the highest net benefits was 444.6 euros. About 27 per cent of the farmers were willing to produce ES for this compensation per hectare. In policy planning, it would be important to focus on these farmers who would be willing to produce the requested services with the lowest compensation demand. Fig. 3. Open in new tabDownload slide Distributions of the farmers’ individual-level WTA measures (black curve), citizens’ individual-level WTP measures (dashed curve), citizens’ mean aggregated WTP (grey line) and the percentage of farmers with a lower WTA than this mean WTP (dotted line) Fig. 3. Open in new tabDownload slide Distributions of the farmers’ individual-level WTA measures (black curve), citizens’ individual-level WTP measures (dashed curve), citizens’ mean aggregated WTP (grey line) and the percentage of farmers with a lower WTA than this mean WTP (dotted line) Figure 3 also shows individual-level WTP estimates for citizens by presenting the share of respondents willing to accept a certain payment level. Instead of presenting traditional demand and supply curves and market balance, it explores how well farmers’ WTA and citizens’ WTP for a policy programme match. The WTA and WTP measures are equal (600 €/hectare) for a 37 per cent share of citizens and farmers. However, even though a larger share of farmers would be willing to supply ES for this amount of compensation, it would not be legitimate from the citizens’ point of view, as the scheme would be paid with taxes and less than half of the citizens are willing to pay this much. 5. Discussion and conclusion This study examined the demand for, and supply of, agricultural ES on a national level in Finland. We used MXL models to analyse citizens’ WTP for four agricultural ES and farmers’ requested compensation (WTA) for producing them. The study revealed that there is a clear demand for higher levels of ES produced by agriculture. The demand was highest for better water quality and a more diverse landscape. On the supply side, farmers preferred the status quo, i.e. the current programme. This was reflected in high WTA values, indicating that farmers require greater amounts of compensation in order to improve the production of ES. However, it is promising that the ES with the highest requested compensation were also those with the highest citizens’ WTP. Overall, the comparison between annual aggregated WTP and WTA estimates revealed that the costs of the programme exceeded the benefits in all scenarios. However, a proportion of farmers, i.e. 20–27 per cent, depending on the details of the programme, were willing to produce the ES for the compensation that the citizens were willing to pay. The results presented here are the first national results on the supply of, and demand for, key ES from agri-environment. If we reflect them with a local case study from Finland concerning landscape attributes in agricultural environments (Grammatikopoulou et al., 2013), we observe considerable differences. In Grammatikopoulou et al. (2013), citizens’ WTP for landscape attributes exceeded the actual costs caused by the provision of landscape attributes, but we did not observe the same tendency here when WTP was compared with the compensation demand, i.e. WTA. This implies that there is a need to evaluate the farmers’ WTA in relation to the actual costs of providing the services. However, we obtained similar results in that farmers were most willing to provide other services than those that were most demanded by citizens. There is also a possibility that a farmer takes into account his or her own benefit from the public good while making choices. From our survey attributes, a farmer would be most likely to derive utility from the changes in landscape and TRB, i.e. the attributes that have very local effects. Instead, water quality and climate effects spreading to a wide spatial area are more complex and are also related to choices of the other farmers in the area as well as other agents further away. Our focus group discussions with farmers showed that their own interest was mainly in improving the growth potential of soil. Due to high WTA values obtained, it is rather unlikely that farmers' would have deducted their own utility from the WTA. However, the farmers own utility from different ES on their own lands is an interesting future research topic. High WTA values for farmers are partly driven by the difficulties in providing biodiversity services. In this study, the focus of biodiversity services was on TRB, which are hotspots of biological diversity and threatened species. The high compensation request related to these highlights the importance of finding new and more easily manageable solutions for providing biodiversity on agricultural land. The WTA estimates could be slightly increased by non-participation in the policy due to protest behaviour. The possible protest respondents were not excluded from the analysis because of the difficulty in ensuring the protest status of any respondent group, despite the typical attitude questions included in the survey. The serial participation and non-participation may also imply problematic behaviour from the modelling point of view if these respondents have decided to choose the status quo or one of the proposed policy alternatives despite the actual attribute levels. However, the shares of serial non-participation and serial participation were almost equal and are assumed to cancel out most of each other’s impact on the WTA estimates. Due to the study questions and design, we used both WTP and WTA measures, although an empirical divergence is often observed between WTA and WTP while measuring the same environmental change and the WTA approach is possibly problematic due to incentive compatibility issues (Lloyd-Smith and Adamowicz, 2018). WTA estimates have typically been higher in cases with less familiarity of the environmental good (Tunçel and Hammitt, 2014). In our case, farmers were also unfamiliar with the new type of benefit-based measures. In farmer decision-making, there was considerable uncertainty about the methods that would produce the given attribute levels. This probably caused farmers to support the status quo alternative in many choice sets and consequently led to high WTA estimates. It is also possible that in this application of a payments for ecosystem services (PES) scheme, in which farmers are asked to give up production possibilities, i.e. private good, there exists strategic behaviour and some respondents have overstated their WTA (Lloyd-Smith and Adamowicz, 2018). Earlier CE studies on landowners’ WTA for ES have concluded that strategic behaviour is possible (Vedel et al., 2015). Nevertheless, we consider that the general recommendation to use WTA in cases where it is institutionally feasible (Johnston et al., 2017) is applicable here, although we cannot rule out strategic behaviour and fully guarantee incentive compatibility. We recommend future research looking for solutions for this issue in the case of CE. If agri-environmental policies are moving towards benefit-based direction, there is a need to find policies that balance the demand for, and supply of, different ES. As our results demonstrated that citizens’ WTP does not cover the compensation need of farmers if WTP and WTA are examined on an average level, the results do not encourage the policy towards large-scale provision of ES as such. However, the public support for the supply of ES could be targeted for the quarter of farmers that are willing to supply these services for compensation that is equal to or lower than citizens’ WTP. Targeting the policy to these farmers might decrease the total area under AES scheme, but could still compete with the current policy or even outperform it with regard to environmental outcomes if farmers with good prerequisites for ecosystem service production are found. However, significant uncertainties related to a benefit-based policy and the information requirements of farmers related to choosing methods that produce particular environmental outcomes need to be resolved before changing the policy regime. The results of this study could also be useful in developing the current policy scheme based on compensating the costs, by focusing on the ES having the greatest demand. Compensation based on the additional costs and income losses resulting from agri-environmental measures, however, may not lead to the most efficient outcome in terms of the overall supply of the desired ES. In this sense, payments based on observed and measured environmental benefits are more likely to lead to improved cost-effectiveness and efficiency. However, the implementation of a benefit-based policy scheme would require a fundamental shift in policy structures. Acknowledgements We are grateful to the Ministry of Agriculture and Forestry of Finland and the Academy of Finland for funding this research (grant number 310205). 6. Appendix A. Attributes of agri-environmental policy programmes in CE The table summarises how different attributes were described to the two different respondent groups (citizens and farmers) in the survey. Ecosystem service . Attribute for citizens . Attribute for farmers . Attribute description . Attribute description . Biodiversity TRB and endangered species TRB Mowed or grazed semi-natural grasslands (meadows, leas and pastures) can provide a habitat for several endangered species TRBs are biotopes shaped by traditional land use (e.g. meadows and pastures). Mowed or grazed TRBs can provide a habitat for several endangered species Landscape Typical agricultural landscape Diverse agricultural landscape Grazing animals and crops grown in open fields affect the diversity of the landscape Crop producers: diversity of crops increases the recreational value of the agricultural landscape Animal husbandry farmers: A higher number of grazing animals increase the recreational value of the agricultural landscape Climate change Climate effects Climate change mitigation mitigation A decrease from current emissions Agricultural greenhouse gas emissions contribute to climate change. The greenhouse gas emissions can be reduced by various cultivation practices and capturing greenhouse gases Agricultural greenhouse gas emissions contribute to climate change. Greenhouse gas emissions can be reduced by increasing the acreage of perennial plants Water quality due to Water quality effects Nutrient flow agriculture Proportion of surface waters in a good or excellent state About half of the nutrient runoff to waters comes from fields. This is affected by the amount of fertilisers used, cultivation practices and annual weather conditions The amount of nutrient runoff depends, for instance, on the fertilisers used. Nutrient runoff can be monitored from ditches with an indicator that would be installed without cost for farmers Ecosystem service . Attribute for citizens . Attribute for farmers . Attribute description . Attribute description . Biodiversity TRB and endangered species TRB Mowed or grazed semi-natural grasslands (meadows, leas and pastures) can provide a habitat for several endangered species TRBs are biotopes shaped by traditional land use (e.g. meadows and pastures). Mowed or grazed TRBs can provide a habitat for several endangered species Landscape Typical agricultural landscape Diverse agricultural landscape Grazing animals and crops grown in open fields affect the diversity of the landscape Crop producers: diversity of crops increases the recreational value of the agricultural landscape Animal husbandry farmers: A higher number of grazing animals increase the recreational value of the agricultural landscape Climate change Climate effects Climate change mitigation mitigation A decrease from current emissions Agricultural greenhouse gas emissions contribute to climate change. The greenhouse gas emissions can be reduced by various cultivation practices and capturing greenhouse gases Agricultural greenhouse gas emissions contribute to climate change. Greenhouse gas emissions can be reduced by increasing the acreage of perennial plants Water quality due to Water quality effects Nutrient flow agriculture Proportion of surface waters in a good or excellent state About half of the nutrient runoff to waters comes from fields. This is affected by the amount of fertilisers used, cultivation practices and annual weather conditions The amount of nutrient runoff depends, for instance, on the fertilisers used. Nutrient runoff can be monitored from ditches with an indicator that would be installed without cost for farmers Open in new tab A. Attributes of agri-environmental policy programmes in CE The table summarises how different attributes were described to the two different respondent groups (citizens and farmers) in the survey. Ecosystem service . Attribute for citizens . Attribute for farmers . Attribute description . Attribute description . Biodiversity TRB and endangered species TRB Mowed or grazed semi-natural grasslands (meadows, leas and pastures) can provide a habitat for several endangered species TRBs are biotopes shaped by traditional land use (e.g. meadows and pastures). Mowed or grazed TRBs can provide a habitat for several endangered species Landscape Typical agricultural landscape Diverse agricultural landscape Grazing animals and crops grown in open fields affect the diversity of the landscape Crop producers: diversity of crops increases the recreational value of the agricultural landscape Animal husbandry farmers: A higher number of grazing animals increase the recreational value of the agricultural landscape Climate change Climate effects Climate change mitigation mitigation A decrease from current emissions Agricultural greenhouse gas emissions contribute to climate change. The greenhouse gas emissions can be reduced by various cultivation practices and capturing greenhouse gases Agricultural greenhouse gas emissions contribute to climate change. Greenhouse gas emissions can be reduced by increasing the acreage of perennial plants Water quality due to Water quality effects Nutrient flow agriculture Proportion of surface waters in a good or excellent state About half of the nutrient runoff to waters comes from fields. This is affected by the amount of fertilisers used, cultivation practices and annual weather conditions The amount of nutrient runoff depends, for instance, on the fertilisers used. Nutrient runoff can be monitored from ditches with an indicator that would be installed without cost for farmers Ecosystem service . Attribute for citizens . Attribute for farmers . Attribute description . Attribute description . Biodiversity TRB and endangered species TRB Mowed or grazed semi-natural grasslands (meadows, leas and pastures) can provide a habitat for several endangered species TRBs are biotopes shaped by traditional land use (e.g. meadows and pastures). Mowed or grazed TRBs can provide a habitat for several endangered species Landscape Typical agricultural landscape Diverse agricultural landscape Grazing animals and crops grown in open fields affect the diversity of the landscape Crop producers: diversity of crops increases the recreational value of the agricultural landscape Animal husbandry farmers: A higher number of grazing animals increase the recreational value of the agricultural landscape Climate change Climate effects Climate change mitigation mitigation A decrease from current emissions Agricultural greenhouse gas emissions contribute to climate change. The greenhouse gas emissions can be reduced by various cultivation practices and capturing greenhouse gases Agricultural greenhouse gas emissions contribute to climate change. Greenhouse gas emissions can be reduced by increasing the acreage of perennial plants Water quality due to Water quality effects Nutrient flow agriculture Proportion of surface waters in a good or excellent state About half of the nutrient runoff to waters comes from fields. This is affected by the amount of fertilisers used, cultivation practices and annual weather conditions The amount of nutrient runoff depends, for instance, on the fertilisers used. Nutrient runoff can be monitored from ditches with an indicator that would be installed without cost for farmers Open in new tab B. Example of the choice set for citizens . Current programme . Alternative X . Alternative Y . TRB and endangered species Present area, 0 species protected Area is increased by 60%, 200 species protected Area is increased by 30%, 100 species protected Typical agricultural landscape Grazing animals Seldom seen Often seen during summer Often seen during summer Plants in cultivation 3 species 5 species 4 species Climate effects Decrease from current emissions 0% 0% 30% Water quality effects Proportion of surface waters in good or excellent condition 60% 80% 60% Cost/taxpayer/year, during 2017–2026 €40 €70 €130 My choice ○ ○ ○ . Current programme . Alternative X . Alternative Y . TRB and endangered species Present area, 0 species protected Area is increased by 60%, 200 species protected Area is increased by 30%, 100 species protected Typical agricultural landscape Grazing animals Seldom seen Often seen during summer Often seen during summer Plants in cultivation 3 species 5 species 4 species Climate effects Decrease from current emissions 0% 0% 30% Water quality effects Proportion of surface waters in good or excellent condition 60% 80% 60% Cost/taxpayer/year, during 2017–2026 €40 €70 €130 My choice ○ ○ ○ Open in new tab B. Example of the choice set for citizens . Current programme . Alternative X . Alternative Y . TRB and endangered species Present area, 0 species protected Area is increased by 60%, 200 species protected Area is increased by 30%, 100 species protected Typical agricultural landscape Grazing animals Seldom seen Often seen during summer Often seen during summer Plants in cultivation 3 species 5 species 4 species Climate effects Decrease from current emissions 0% 0% 30% Water quality effects Proportion of surface waters in good or excellent condition 60% 80% 60% Cost/taxpayer/year, during 2017–2026 €40 €70 €130 My choice ○ ○ ○ . Current programme . Alternative X . Alternative Y . TRB and endangered species Present area, 0 species protected Area is increased by 60%, 200 species protected Area is increased by 30%, 100 species protected Typical agricultural landscape Grazing animals Seldom seen Often seen during summer Often seen during summer Plants in cultivation 3 species 5 species 4 species Climate effects Decrease from current emissions 0% 0% 30% Water quality effects Proportion of surface waters in good or excellent condition 60% 80% 60% Cost/taxpayer/year, during 2017–2026 €40 €70 €130 My choice ○ ○ ○ Open in new tab Example of the choice set for animal husbandry farmers. . Current programme . Alternative X . Alternative Y . Grazing season Cattle, sheep and horses graze for under 3 months Cattle, sheep and horses graze for over 3 months Cattle, sheep and horses graze for over 6 months Climate effects At least 20% of the area under cultivation with perennial plants At least 40% of the area under cultivation with perennial plants At least 60% of the area under cultivation with perennial plants Share of the perennial plants from the arable area Water quality effects Your farm’s current nutrient flow Nutrient flow decreased to 70% Your farm’s current nutrient flow Reducing the amount of nutrient runoff with the measures chosen by the farmer. Measurement from the main drain Traditional rural biotopes Current area Area is increased by 10 hectares Area is increased by 5 hectares Agri-environmental payment, €/hectare/year, during 2021–2027 Your current agri-environmental payment per hectare €100 €350 My choice ○ ○ ○ . Current programme . Alternative X . Alternative Y . Grazing season Cattle, sheep and horses graze for under 3 months Cattle, sheep and horses graze for over 3 months Cattle, sheep and horses graze for over 6 months Climate effects At least 20% of the area under cultivation with perennial plants At least 40% of the area under cultivation with perennial plants At least 60% of the area under cultivation with perennial plants Share of the perennial plants from the arable area Water quality effects Your farm’s current nutrient flow Nutrient flow decreased to 70% Your farm’s current nutrient flow Reducing the amount of nutrient runoff with the measures chosen by the farmer. Measurement from the main drain Traditional rural biotopes Current area Area is increased by 10 hectares Area is increased by 5 hectares Agri-environmental payment, €/hectare/year, during 2021–2027 Your current agri-environmental payment per hectare €100 €350 My choice ○ ○ ○ Open in new tab Example of the choice set for animal husbandry farmers. . Current programme . Alternative X . Alternative Y . Grazing season Cattle, sheep and horses graze for under 3 months Cattle, sheep and horses graze for over 3 months Cattle, sheep and horses graze for over 6 months Climate effects At least 20% of the area under cultivation with perennial plants At least 40% of the area under cultivation with perennial plants At least 60% of the area under cultivation with perennial plants Share of the perennial plants from the arable area Water quality effects Your farm’s current nutrient flow Nutrient flow decreased to 70% Your farm’s current nutrient flow Reducing the amount of nutrient runoff with the measures chosen by the farmer. Measurement from the main drain Traditional rural biotopes Current area Area is increased by 10 hectares Area is increased by 5 hectares Agri-environmental payment, €/hectare/year, during 2021–2027 Your current agri-environmental payment per hectare €100 €350 My choice ○ ○ ○ . Current programme . Alternative X . Alternative Y . Grazing season Cattle, sheep and horses graze for under 3 months Cattle, sheep and horses graze for over 3 months Cattle, sheep and horses graze for over 6 months Climate effects At least 20% of the area under cultivation with perennial plants At least 40% of the area under cultivation with perennial plants At least 60% of the area under cultivation with perennial plants Share of the perennial plants from the arable area Water quality effects Your farm’s current nutrient flow Nutrient flow decreased to 70% Your farm’s current nutrient flow Reducing the amount of nutrient runoff with the measures chosen by the farmer. Measurement from the main drain Traditional rural biotopes Current area Area is increased by 10 hectares Area is increased by 5 hectares Agri-environmental payment, €/hectare/year, during 2021–2027 Your current agri-environmental payment per hectare €100 €350 My choice ○ ○ ○ Open in new tab Footnotes 1 Separate models were estimated for crop and animal husbandry farmers to test whether there was a difference in the response to the landscape attribute. The results were very similar: the smaller change (level 1) in attribute level was insignificant, but the larger change (level 2) had a negative, significant coefficient. We will analyse the heterogeneity of farmers’ responses in the CE in detail in the forthcoming paper. Review coordinated by Carl Johan Lagerkvist References Aslam , U. , Termansen , M. and Fleskens , L. ( 2017 ). Investigating farmers’ preferences for alternative PES schemes for carbon sequestration in UK agroecosystems . Ecosystem Services 27 : 103 – 112 . Google Scholar Crossref Search ADS WorldCat Baldi , S. , Prado , A. , Gallejones , P. et al. . ( 2015 ). Modeling trade-offs among ecosystem services in agricultural production systems . Environmental Modelling and Software 72 : 314 – 326 . Google Scholar Crossref Search ADS WorldCat Bennett , J. and Blamey , R. (eds). ( 2001 ). The Choice Modelling Approach to Environmental Valuation . Cheltenham, UK, Northampton, MA, USA : Edward Elgar . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bernués , A. , Tello-García , E. , Rodríguez-Ortega , T. et al. . ( 2016 ). Agricultural practices, ecosystem services and sustainability in high nature value farmland: Unraveling the perceptions of farmers and nonfarmers . Land Use Policy 59 : 130 – 142 . Google Scholar Crossref Search ADS WorldCat Broch , S. W. and Vedel , S. E. ( 2012 ). Using choice experiments to investigate the policy relevance of heterogeneity in farmer agri-environmental contract preferences . Environmental and Resource Economics 51 ( 4 ): 561 – 581 . Google Scholar Crossref Search ADS WorldCat Broch , S. W. , Strange , N. , Jacobsen , J. B. et al. . ( 2013 ). Farmers' willingness to provide ecosystem services and effects of their spatial distribution . Ecological Economics 92 : 78 – 86 . Google Scholar Crossref Search ADS WorldCat Burton , R. J. F. and Schwarz , G. ( 2013 ). Result-oriented agri-environmental schemes in Europe and their potential for promoting behavioural change . Land Use Policy 30 : 628 – 641 . Google Scholar Crossref Search ADS WorldCat Castro , A. J. , Verburg , P. H. , Martín-López , B. et al. . ( 2014 ). Ecosystem service trade-offs from supply to social demand: a landscape-scale spatial analysis . Landscape and Urban Planning 132 : 102 – 110 . Google Scholar Crossref Search ADS WorldCat Chen , Y. , Zhang , Q. , Liu , W. et al. . ( 2017 ). Analyzing farmers’ perceptions of ecosystem services and PES schemes within agricultural landscapes in Mengyin County, China: transforming trade-offs into synergies . Sustainability 9 : 1459 . Google Scholar Crossref Search ADS WorldCat Christensen , T. , Pedersen , A. B. , Nielsen , H. O. et al. . ( 2011 ). Determinants of farmers’ willingness to participate in subsidy schemes for pesticide-free buffer zones – A choice experiment study . Ecological Economics 70 ( 8 ): 1558 – 1564 . Google Scholar Crossref Search ADS WorldCat Common International Classification of Ecosystem Services (CICES) ( 2016 ). CICES-V4.3. https://cices.eu/resources/. Accessed 7 April 2016. Dale , V. H. and Polasky , S. ( 2007 ). Measures of the effects of agricultural practices on ecosystem services . Ecological Economics 64 : 286 – 296 . Google Scholar Crossref Search ADS WorldCat de Groot , R. S. , Wilson , M. A. and Boumans , R. M. J. ( 2002 ). A typology for the classification, description and valuation of ecosystem functions, goods and services . Ecological Economics 41 : 393 – 408 . Google Scholar Crossref Search ADS WorldCat Dupras , J. , Laurent-Lucchetti , J. , Revéret , J.-P. et al. . ( 2018 ). Using contingent valuation and choice experiment to value the impacts of agri-environmental practices on landscapes aesthetics . Landscape Research 43 ( 5 ): 679 – 695 . Google Scholar Crossref Search ADS WorldCat European Network for Rural Development and European Commission (ENRD) ( 2010 ). Report on the Contribution of the European Network for Rural Development to the Public debate on the Common Agricultural Policy after 2013 (13/07/2010). http://www.eufinantare.info/Documente/synthesisENRDcontributionsFINAL-130720101. Accessed 22 January 2018. Espinosa-Goded , M. , Barreiro-Hurlé , J. and Ruto , E. ( 2010 ). What do farmers want from agri-environmental scheme design? A choice experiment approach . Journal of Agricultural Economics 61 ( 2 ): 259 – 273 . Google Scholar Crossref Search ADS WorldCat Gobster , P. H. , Nassauer , J. I. , Daniel , T. C. et al. . ( 2007 ). The shared landscape: what does aesthetics have to do with ecology? Landscape Ecology 22 : 959 – 972 . Google Scholar Crossref Search ADS WorldCat Gordon , L. J. , Finlayson , C. M. and Falkenmark , M. ( 2010 ). Managing water in agriculture for food production and other ecosystem services . Agricultural Water Management 97 : 512 – 519 . Google Scholar Crossref Search ADS WorldCat Grammatikopoulou , I. , Pouta , E. and Salmiovirta , M. ( 2013 ). A locally designed payment scheme for agricultural landscape services . Land Use Policy 32 : 175 – 185 . Google Scholar Crossref Search ADS WorldCat Grammatikopoulou , I. ( 2016 ). Addressing the demand for and supply of ecosystem services in agriculture through market-based and target-based policy measures. Doctoral dissertation. Natural resources and bioeconomy studies 35/2016 . Helsinki, Finland : Natural Resources Institute Finland (Luke) . Hanley , N. , Mourato , S. and Wright , R. E. ( 2001 ). Choice modelling approaches: a superior alternative for environmental valuation? Journal of Economic Surveys 15 ( 3 ): 435 – 462 . Google Scholar Crossref Search ADS WorldCat Hasund , K. P. ( 2013 ). Indicator-based agri-environmental payments: a payment-by-result model for public goods with a Swedish application . Land Use Policy 30 : 223 – 233 . Google Scholar Crossref Search ADS WorldCat Hensher , D. A. , Rose , J. M. and Greene , W. H. ( 2015 ). Applied Choice Analysis , 2nd edn. Cambridge, UK : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Herzon , I. , Birge , T. , Allen , B. et al. . ( 2018 ). Time to look for evidence: Results-based approach to biodiversity conservation on farmland in Europe . Land Use Policy 71 : 347 – 354 . Google Scholar Crossref Search ADS WorldCat Hole , A. and Kolstad , J. ( 2012 ). Mixed logit estimation of willingness to pay distributions: a comparison of models in preference and WTP space using data from a health-related choice experiment . Empirical Economics 42 ( 2 ): 445 – 469 . Google Scholar Crossref Search ADS WorldCat Huang , J. , Tichit , M. , Poulot , M. et al. . ( 2015 ). Comparative review of multifunctionality and ecosystem services in sustainable agriculture . Journal of Environmental Management 149 : 138 – 147 . Google Scholar Crossref Search ADS PubMed WorldCat Johnston , R. J. , Boyle , K. J. , Adamowicz , W. et al. . ( 2017 ). Contemporary guidance for stated preference studies . Journal of the Association of Environmental and Resource Economists 4 ( 2 ): 319 – 405 . Google Scholar Crossref Search ADS WorldCat Latacz-Lohmann , U. and Schreiner , J. A. ( 2019 ). Assessing consumer and producer preferences for animal welfare using a common elicitation format . Journal of Agricultural Economics 70 ( 2 ): 293 – 315 . Google Scholar Crossref Search ADS WorldCat Lima Santos , J. , Madureira , L. , Ferreira , A. C. et al. . ( 2016 ). Building an empirically-based framework to value multiple public goods of agriculture at broad supranational scales . Land Use Policy 53 : 56 – 70 . Google Scholar Crossref Search ADS WorldCat Lloyd-Smith , P. and Adamowicz , W. ( 2018 ). Can stated measures of willingness-to-accept be valid? Evidence from laboratory experiments . Journal of Environmental Economics and Management 91 : 133 – 149 . Google Scholar Crossref Search ADS WorldCat Martín-López , B. , Iniesta-Arandia , I. , García-Llorente , M. et al. . ( 2012 ). Uncovering ecosystem services bundles through social preferences . PLoS ONE 7 ( 6 ): e38970. OpenURL Placeholder Text WorldCat Millennium Ecosystem Assessment (MEA) . ( 2003 ). Ecosystems and Human Well-being: A Framework for Assessment . Washington, DC, USA : Island Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nieto-Romero , M. , Oteros-Rozas , E. , González , J. A. et al. . ( 2014 ). Exploring the knowledge landscape of ecosystem services assessments in Mediterranean agroecosystems: Insights for future research . Environmental Science and Policy 37 : 121 – 133 . Google Scholar Crossref Search ADS WorldCat Novikova , A. , Rocchi , L. and Vitunskienė , V. ( 2017 ). Assessing the benefit of the agroecosystem services: Lithuanian preferences using a latent class approach . Land Use Policy 68 : 277 – 286 . Google Scholar Crossref Search ADS WorldCat Pagiola , S. , Arcenas , A. and Platais , G. ( 2005 ). Can payments for environmental services help reduce poverty? An exploration of the issues and the evidence to date from Latin America . World Development 33 ( 2 ): 237 – 253 . Google Scholar Crossref Search ADS WorldCat Pouta , E. and Hauru , K. ( 2015 ). Citizens’ evaluation of the importance of ecosystem services. In J.-P. Jäppinen and J. Heliölä (eds), Towards A Sustainable and Genuinely Green Economy. The value and social significance of ecosystem services in Finland (TEEB for Finland) – Synthesis and roadmap. The Finnish Environment 1en/2015 . Helsinki, Finland : Ministry of the Environment , 66 – 69 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Power , A. G. ( 2010 ). Ecosystem services and agriculture: Tradeoffs and synergies . Philosophical Transactions of the Royal Society, B: Biological Sciences 365 : 2959 – 2971 . Google Scholar Crossref Search ADS WorldCat Prager , K. , Reed , M. and Scott , A. ( 2012 ). Encouraging collaboration for the provision of ecosystem services at a landscape scale – Rethinking agri-environmental payments . Land Use Policy 29 : 244 – 249 . Google Scholar Crossref Search ADS WorldCat Raudsepp-Hearne , C. , Peterson , G. D. and Bennett , E. M. ( 2010 ). Ecosystem service bundles for analyzing tradeoffs in diverse landscapes . Proceedings of the National Academy of Sciences of the United States of America 107 : 5242 – 5247 . Google Scholar Crossref Search ADS PubMed WorldCat Revelt , D. and Train , K. ( 1998 ). Mixed logit with repeated choices: households' choices of appliance efficiency level . The Review of Economics and Statistics 80 ( 4 ): 647 – 657 . Google Scholar Crossref Search ADS WorldCat Rey Benayas , J. M. and Bullock , J. M. ( 2012 ). Restoration of biodiversity and ecosystem services on agricultural land . Ecosystems 15 : 883 – 899 . Google Scholar Crossref Search ADS WorldCat Rodríguez , J. P. , Beard , T. D. , Bennett , E. M. et al. . ( 2006 ). Trade-offs across space, time, and ecosystem services . Ecology and Society 11 ( 1 ): 28 . Google Scholar Crossref Search ADS WorldCat Rose , J. M. and Bliemer , M. C. J. ( 2009 ). Constructing efficient stated choice experimental designs . Transport Reviews 29 ( 5 ): 587 – 617 . Google Scholar Crossref Search ADS WorldCat Ruto , E. and Garrod , G. ( 2009 ). Investigating farmers' preferences for the design of agri-environmental schemes: a choice experiment approach . Journal of Environmental Planning and Management 52 ( 5 ): 631 – 647 . Google Scholar Crossref Search ADS WorldCat Sangkapitux , C. , Suebpongsang , P. , Punyawadee , V. et al. . ( 2017 ). Eliciting citizen preferences for multifunctional agriculture in the watershed areas of northern Thailand through choice experiment and latent class models . Land Use Policy 67 : 38 – 47 . Google Scholar Crossref Search ADS WorldCat Schroeder , L. , Isselstein , J. , Chaplin , S. et al. . ( 2013 ). Agri-environment schemes: Farmers’ acceptance and perception of potential ‘payment by results’ in grassland – a case study in England . Land Use Policy 32 : 134 – 144 . Google Scholar Crossref Search ADS WorldCat Siebert , R. , Toogood , M. and Knierim , A. ( 2006 ). Factors affecting European farmers’ participation in biodiversity policies . Sociologia Ruralis 46 : 318 – 340 . Google Scholar Crossref Search ADS WorldCat Smith , H. F. and Sullivan , C. A. ( 2014 ). Ecosystem services within agricultural landscapes – farmers' perceptions . Ecological Economics 98 : 72 – 80 . Google Scholar Crossref Search ADS WorldCat Statistics Finland ( 2015 ). www.stat.fi. Accessed 15 October 2016. Swinton , S. M. , Lupi , F. , Robertson , G. P. et al. . ( 2007 ). Ecosystem services and agriculture: Cultivating agricultural ecosystems for diverse benefits . Ecological Economics 64 : 245 – 252 . Google Scholar Crossref Search ADS WorldCat Taloustutkimus ( 2017 ). Internet Panel. http://www.taloustutkimus.fi/in-english/products_services/internet_panel/. Accessed 25 January 2017. Train , K. ( 2003 ). Discrete Choice Methods with Simulation . New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Tunçel , T. and Hammitt , J. K. ( 2014 ). A new meta-analysis on the WTP/WTA disparity . Journal of Environmental Economics and Management 68 ( 1 ): 175 – 187 . Google Scholar Crossref Search ADS WorldCat Turner , R. K. , Morse-Jones , S. and Fisher , B. ( 2010 ). Ecosystem valuation. A sequential decision support system and quality assessment issues . Annals of the New York Academy of Sciences 1185 : 79 – 101 . Google Scholar Crossref Search ADS PubMed WorldCat Van Tongeren , F. ( 2008 ). Agricultural Policy Design and Implementation: A Synthesis. OECD Food, Agriculture and Fisheries Working Papers, No. 7 . OECD Publishing . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC van Zanten , B. T. , Verburg , P. H. , Espinosa , M. et al. . ( 2014 ). European agricultural landscapes, common agricultural policy and ecosystem services: a review . Agronomy for Sustainable Development 34 : 309 – 325 . Google Scholar Crossref Search ADS WorldCat Varela , E. , Verheyen , K. , Valdés , A. et al. . ( 2018 ). Promoting biodiversity values of small forest patches in agricultural landscapes: Ecological drivers and social demand . Science of the Total Environment 619–620 : 1319 – 1329 . Google Scholar Crossref Search ADS PubMed WorldCat Vedel , S. E. , Jacobsen , J. B. and Thorsen , B. J. ( 2015 ). Forest owners' willingness to accept contracts for ecosystem service provision is sensitive to additionality . Ecological Economics 113 : 15 – 24 . Google Scholar Crossref Search ADS WorldCat Villanueva , A. J. , Rodríguez-Entrena , M. , Arriaza , M. et al. . ( 2017 ). Heterogeneity of farmers' preferences towards agri-environmental schemes across different agricultural subsystems . Journal of Environmental Planning and Management 60 ( 4 ): 684 – 707 . Google Scholar Crossref Search ADS WorldCat Vossler , C. A. , Doyon , M. and Rondeau , D. ( 2012 ). Truth in consequentiality: theory and field evidence on discrete choice experiments . American Economic Journal: Microeconomics 4 : 145 – 171 . Google Scholar Crossref Search ADS WorldCat Wei , H. , Fan , W. , Wang , X. et al. . ( 2017 ). Integrating supply and social demand in ecosystem services assessment: a review . Ecosystem Services 25 : 15 – 27 . Google Scholar Crossref Search ADS WorldCat Xun , F. , Hu , Y. , Lv , L. et al. . ( 2017 ). Farmers’ awareness of ecosystem services and the associated policy implications . Sustainability 9 ( 9 ): 1612 . Google Scholar Crossref Search ADS WorldCat Zasada , I. ( 2011 ). Multifunctional peri-urban agriculture – a review of societal demands and the provision of goods and services by farming . Land Use Policy 28 : 639 – 648 . Google Scholar Crossref Search ADS WorldCat Zhang , W. , Ricketts , T. H. , Kremen , C. et al. . ( 2007 ). Ecosystem services and dis-services to agriculture . Ecological Economics 64 : 253 – 260 . Google Scholar Crossref Search ADS WorldCat © Oxford University Press and Foundation for the European Review of Agricultural Economics 2020; all rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Demand and supply of agricultural ES: towards benefit-based policy
JF - European Review of Agricultural Economics
DO - 10.1093/erae/jbz044
DA - 2020-06-15
UR - https://www.deepdyve.com/lp/oxford-university-press/demand-and-supply-of-agricultural-es-towards-benefit-based-policy-bOczNBcuQW
DP - DeepDyve
ER -