Elsevier

Applied Geography

Volume 31, Issue 2, April 2011, Pages 668-676
Applied Geography

Adoption of payments for ecosystem services: An application of the Hägerstrand model

https://doi.org/10.1016/j.apgeog.2010.12.001Get rights and content

Abstract

Many governments are now offering incentive payments to private land owners to adapt their management of the land in such a way as to safeguard or enhance ecosystem service provision. These payments are offered to individual land owners, whose decisions may be influenced by economic rationality, but also other factors. Understanding factors such as social capital and neighbourhood networks is particularly relevant as these can help to create local patterns of high and coordinated uptake. This is important because the delivery of many ecosystem services depends on spatial patterns of interventions at the landscape scale, i.e. at spatial scales of multiple farms. To date little empirical work has been carried out to estimate the extent and relative importance of local land owner networks on entry into ecosystem services payment schemes. This study demonstrates a method to detect possible relationships between farm locations and the time of adoption. Based on Thorsten Hagerstrand’s model of innovation diffusion as a spatio-temporal process, a simulation approach is used to detect spatio-temporal clustering in the uptake of a case study ESP scheme, the Environmentally Sensitive Area (ESA) scheme in Scotland. The analysis reveals clear spatio-temporal uptake patterns at different spatial scales and in different types of rural spaces. It is argued that these findings have relevance for local adaptation of policies, both to liaise more effectively with sections of the farming community, and to achieve better uptake patterns at the landscape scale.

Research highlights

► This paper presents a novel use of the classical Hägerstrand model in contemporary natural resource management. ► It is a suitable tool to examine the spatio-temporal pattern of uptake of voluntary land management schemes/payments to land managers for delivering ecosystem services. ► These patterns are important because many ecosystem services require landscape scale policy interventions, yet policies traditionally target only individual farmers (thus delivering field scale or farm scale interventions). ► These patterns are also important because they provide a proxy indicator for (the strength of) existing neighbourhood networks, through which policies can be communicated more effectively and farmers can be enticed to join more quickly (thus making the policy more efficient). ► Case study application identifies spatio-temporal patterns of uptake in different parts of Scotland. Much stronger patterns are found in mountainous areas and on small islands, which is consistent with the expectation of stronger communities of place and neighbourhood networks in more remote places.

Introduction

In recent years there has been much work to develop models that can inform policy makers on where and how to intervene in the landscape in order to conserve biodiversity and sustain the provision of ecosystem services (e.g. Andersson et al., 2007, Norderhaug et al., 2000, Van der Horst and Gimona, 2005). These models are now extended to take (in addition to the benefits) also the costs of conservation into account (e.g. Messer, 2006, Naidoo et al., 2006, Van der Horst, 2007, Wunscher et al., 2008) so that conservation policies at the landscape scale can be more efficient.

Such methods may be directly suitable to design command and control measures such as the delineation of new conservation areas, or the compulsory purchase of private land. However on many occasions the policy makers seek to achieve conservation goals through collaboration with willing land owners. This more collaborative approach is informed amongst others by the costs of top-down measures such as buy-out and the allure of the concept of multi-functional land use for policy makers who must balance multiple interests (including the vested interests of the farming sector) in a pluralist democratic society. Financial incentives are currently the most widely used policy instrument for conserving biodiversity in the agricultural landscapes of Europe. The design of such incentive policies, traditionally called Agri-Environmental Schemes (AES) in the EU, but now increasingly designed to provide Ecosystem Service Payments (ESP), requires (a) knowledge of potential biodiversity conservation gains from different land management decisions and (b) understanding of what it takes to motivate farmers (and other land managers) to participate in these policies. In recent years there has been a growing body of literature about the conservation gains from agri-environmental schemes, including both empirical studies to assess the impacts on the ground of schemes that have been in place for a number of years (e.g. Feehan et al., 2005, Kleijn and Sutherland, 2003) and simulation or modelling studies to assess the potential impacts of proposed or ongoing schemes (e.g. Haines-Young et al., 2006, Van der Horst, 2007). Also on farmer behaviour there has been a growing body of literature (see further down), focussing strongly on the factors influencing the choices made by individual farmers to join or opt out of AES. These studies employ either survey methods to assess behavioural patterns (of representative samples) of the farmer population, or qualitative approaches to develop a more in-depth understanding of values and motivations underpinning behavioural choices.

Uptake of amongst individual farmers is not by itself sufficient to guarantee the delivery of enhanced habitat structure and connectivity at the landscape level. A well designed AES, combined with a very high uptake by individual farmers, may guarantee the delivery of conservation benefits. However when uptake is not very high or when uptake in some areas is much more beneficial for conservation than in others (i.e. when conservation benefits are spatially heterogeneous), then it becomes important to ensure that uptake takes place in a coordinated way across multiple neighbouring farms that constitute an area of high conservation value. Whilst the successes and failures of collaborative or collective action have been studied for common property regimes such as common grazing land, wildlife and fishery management (e.g. Brainard et al., 1999, Ostrom, 1990), its application is still a rarity within a European agricultural setting that is dominated by privately owned farms (see also Franks and McGloin, 2007, Hodge and McNally, 2000, Hodge, 2001). This is not to say that in existing AES, decision making by farmers is completely individualistic, mutually independent and uncoordinated, but rather that the social settings of farmer decision making have been underplayed both in scheme design and in the evaluation of scheme uptake. Whilst research into farmer motivation has revealed the importance of informal and often local networks (see further down), the aggregate effects of such factors on collective uptake within a locality have been little studied.

In short, there are currently shortcomings in policy design and in empirical research on the spatial pattern of uptake of AES schemes. With regards to current policy design, it tends to focus on farm-level interventions, whilst the provision of many ecosystem services depends on processes that take place at the landscape scale. With regards to empirical research, there have been very few studies of the landscape scale patterns of uptake of AES schemes. The reasons for this lack of attention to spatial heterogeneity and spatial scale in policy appraisal and evaluation, lies perhaps in the traditional dominance of conventional economic thinking in the shaping of policy interventions. The legacy of quantitative geography is of course prominent in regional science, but this heterodox approach has long been ignored by ‘mainstream’ economics.There have only been a few occasions when leading ‘mainstream’ economists have borrowed from the classical tradition of quantitative geography, with Nobel laureates Tinbergen (on international trade) and Krugman (on new economic geography) as perhaps the most notable examples.1 The rapid development of IT in general and GIS in particular, together with a growing attention to the challenge of sustainable resource use, has created a more dynamic research environment for interdisciplinary research since the 1990s, yielding hybrid sub-disciplines such as ecological economics and spatial econometrics. The legacy of quantitative geography is now visible in hybrid concepts such as spatial discounting (Belperio et al., 2002, Hannon, 1994, Kozak et al., in press) and the use of gravity models in travel cost estimates of the non-market value of recreational sites (Brainard, Lovett, & Bateman, 1999). More generally, GIS is now widely used to map values of nature (e.g. Bateman et al., 2002, Bateman et al., 2003, Eade and Moran, 1996, Sherrouse et al., in press, Van der Horst, 2006, Van der Horst, 2007). This paper draws on the work of the Swedish geographer Thorsten Hägerstrand, another ‘classical’ quantitative geographer, to explore its contemporary value in sustainable land use policy. Hägerstrand (1967) used a mathematical model to explore the extent to which innovation diffusion takes place as a spatial process. Observing a clear occurrence of a neighbourhood or proximity effect in the spatio-temporal pattern of adoption, he was able to conclude that for the adoption of subsidised grazing-improvement and other cases of innovation in agricultural management by Swedish farmers in the 1930s, communication within the local farming community was a more powerful agent of diffusion than public announcements which disperse information in a more even pattern across rural space.

Although Hägerstrand’s work was focused on the diffusion of ‘productive’ farming innovations, there is no reason why it cannot be used to examine the diffusion of sustainable farming innovations to produce non-market goods and ecosystem services. In view of the current policy agenda for sustainable and multi-functional rural land use, and the provision of financial incentives to farmers to produce other ecosystem services in addition to growing food, it is relevant to ask how, where and when the innovation diffusion of AES takes place. The aim of this paper is to demonstrate how his model, now largely forgotten as a research tool,2 can be relevant today in studies on the uptake of ecosystem service payments.

In order to illustrate what Hägerstrand’s model could bring to appraisal or evaluation of policies for ecosystem service payments, this paper will utilise a particular AES case study, namely the Environmentally Sensitive Area (ESA) scheme in Scotland. The ESA Scheme is an agri-environmental scheme which aims to protect flora and fauna, geological and physio-graphical features, buildings and other objects of archaeological, architectural or historic interest in an area or protect and enhance the natural beauty of that area through payments made to farmers to maintain or alter their current practices. The ESA Scheme applies only to certain designated areas. These are often areas characterised by traditional agricultural systems and especially areas where this traditional agriculture is threatened by rural depopulation or agricultural intensification (for more details, see Wilson, 1997a, Wilson, 1997b). The ESA Scheme is voluntary, offering incentive payments to entice farmers to join. The ESA Scheme is a European scheme, first implemented in the UK in 1986. In total, 43 ESAs have been designated in the UK and 10 of these are located in Scotland (Fig. 1), covering 19% of Scotland’s land area.

The remainder of this paper is structured as follows. First an overview is provided of the literature on farm adoption of agri-environmental schemes, showing that spatio-temporal patterns of uptake have sometimes been observed but that these have received very limited attention as a key focus of analysis. Subsequently, the use of Hägerstrand’s model to explore these patterns is described in the methodology section. The results of the analysis are then presented, describing observed spatio-temporal patterns and exploring possible factors of influence. Subsequently, the limitations of the method are discussed, along with options for further development and application. The paper concludes with a reflection on the importance of analysing spatio-temporal patterns of uptake for understanding farmer behaviour, and for the effective delivery ecosystem services through voluntary payment schemes.

Section snippets

The ESA scheme; factors known to influence the uptake

An extensive literature exists on farmers’ motivation and factors influencing farmers’ decision to join an AES. Entry decisions have been found to be highly influenced by the consequences for farm income and farmers with fewer financial constraints are much more likely to be influenced by potential conservation considerations (Morris, Mills, & Crawford, 2000). The degree to which the scheme prescriptions fit the existing farm system, determines its ‘popularity’ with farmers, i.e. farmers are

Methodology

The methodology consists of two stages. Firstly, and very similar to Hägerstrand’s original approach, quantitative indicators of spatio-temporal clustering are applied to each ESA. Secondly, the ESAs for which these indicators suggest a possible occurrence of neighbour networks, are subjected to a closer, visual study. A Geographical Information System (ArcView 3.1) is used to display the farm locations on a map (a farmers’s postal address is translated into x,y coordinates using Ordnance

Comparisons between ESAs

Table 1 lists the results of the statistical analysis for each ESA. The coefficient of variation records the degree of temporal clustering about the average entry time. The higher the values for the coefficient of variation, the stronger the temporal clustering is. Mantel p measures the level of spatio-temporal clustering in comparison to a random distribution, with p > 0.5 indicating stronger clustering and p < 0.5 indicating weaker clustering than what could be expected by chance. Table 1

Discussion

By demonstrating the utilisation of Hägerstrand’s spatial model of innovation diffusion in a case study application, the ESA scheme in Scotland, this paper has yielded some strong and diverse evidence of spatio-temporal clustering of farmer decisions to join a voluntary agri-environmental scheme. This discussion section is structured as follows. First the tentative interpretations of the clustering are explored, then opportunities for extending and improving the analysis are identified, and

Conclusions

This paper has sought to illustrate that the model developed by Hägerstrand for Swedish farming in the 1930s, can have relevance for a very contemporary land management challenge; the design of incentive policies to enhance the delivery of ecosystem services in agricultural landscapes. In addition to arguing for the revival of a ‘classical’ but now largely dormant quantitative geography model, this paper is also novel in that it utilises a statistical method (the Mantel test) that is widely

Acknowledgements

I would like to thank Gerard Wynn and Alessandro Gimona for comments on an earlier draft and the ESRC for generous grant support (Res-152–27-0004).

References (50)

  • I. Hodge

    Beyond agri-environmental policy: towards an alternative model of rural environmental governance

    Land Use Policy

    (2001)
  • I. Hodge et al.

    Wetland restoration, collective action and the role of water management institutions

    Ecological Economics

    (2000)
  • K.D. Messer

    The conservation benefits of cost-effective land acquisition: a case study in Maryland

    Journal of Environmental Management

    (2006)
  • C. Morris et al.

    Recruiting the New Conservationists: farmers’ adoption of agri-environmental schemes in the UK

    Journal of Rural Studies

    (1995)
  • R. Naidoo et al.

    Integrating economic costs into conservation planning

    Trends in Ecology & Evolution

    (2006)
  • C. Potter et al.

    The conservation status and potential of elderly farmers: results from a survey in England and Wales

    Journal of Rural Studies

    (1992)
  • D. Van der Horst

    Spatial cost-benefit thinking in multi-functional forestry; towards a framework for spatial targeting of policy interventions

    Ecological Economics

    (2006)
  • D. Van der Horst

    Assessing the efficiency gains of improved spatial targeting of policy interventions; the example of an agri-environmental scheme

    Journal of Environmental Management

    (2007)
  • G.A. Wilson

    Factors influencing farmer participation in the environmentally sensitive area scheme

    Journal of Environmental Management

    (1997)
  • T. Wunscher et al.

    Spatial targeting of payments for environmental services; a tool for boosting conservation benefits

    Ecological Economics

    (2008)
  • E. Andersson et al.

    Measuring social-ecological dynamics behind the generation of ecosystem services

    Ecological Applications

    (2007)
  • I.J. Bateman et al.

    Applying geographical information systems (GIS) to environmental and resource economics

    Environmental and Resource Economics

    (2002)
  • I.J. Bateman et al.

    Applied Environmental Economics.A GIS approach to cost-benefit analysis

    (2003)
  • A.P. Belperio et al.

    Spatial and temporal variability in the Holocene sea-level record of the South Australian coastline

    Sedimentary Geology

    (2002)
  • Blair J.P. and Carroll M.C. Towards a theory of targeting. Applied Geography. (in...
  • Cited by (29)

    • Bridging the divide between intuitive social-ecological value and sustainability in the Manica Highlands of southern Africa (Zimbabwe-Mozambique)

      2019, Ecosystem Services
      Citation Excerpt :

      Valuation of ES as a concept was first highlighted by Westman (1977) as a means to explicitly value, in monetary terms, items in nature that were formerly regarded as difficult to put conventional financial value on. Since then the research field of environmental economics has been well established (see Brauman et al., 2007; Balmford et al., 2008; Yoe et al., 2009; Bullock et al., 2011; van der Horst, 2011; Blignaut et al., 2012; Bennett et al., 2015 – and with dedicated journals, such as the Journal of Environmental Economics and Management), where ES valuation has been developed as a vehicle to integrate ecological understanding and economic considerations to redress the traditional neglect of the value of ES in policy decisions (de Groot et al., 2002; Chee, 2004). The trend towards valuation persists, often as a means to justify the allocation of funds for ecosystem maintenance and restoration in a highly competitive funding world (Van Wilgen et al., 1996).

    • An analysis of conservation practice adoption studies in agricultural human-natural systems

      2019, Journal of Environmental Management
      Citation Excerpt :

      At the same time, researchers appear to be expanding the set of adoption metrics to document or quantify a wider range of aspects shaping adoption and non-adoption. Twenty-two studies gauged some degree of adoption intensity, such as ordinal measures to represent adoption amounts (Vignola et al., 2010) or acreage proportions (Groth and Curtis, 2017), while 14 studies incorporated spatial analysis within a region or farm, such as spatial clustering of farms participating in agri-environment programs (Van der Horst, 2011). The remaining 29 studies took more tangential approach, using a proxy measure to indicate whether a farmer would be a likely adopter, such as perceptions of self-efficacy (McNairn and Mitchell, 1992).

    View all citing articles on Scopus
    View full text