Land Use Policy 57 (2016) 800812

Contents lists available at ScienceDirect

Land Use Policy

journal homepage: www.elsevier.com/locate/landusepol

Patterns of agri-environmental scheme participation in Europe:

Indicative trends from selected case studies
Evangelos S. Pavlis a, , Theano S. Terkenli a , Sren B.P. Kristensen b , Anne G. Busck b ,
Georgia L. Cosor c
a
Department of Geography, University of Aegean, University Hill, Mytilene, 81100 Lesvos, Greece
b
Department of Geosciences and Natural Resource Management, University of Copenhagen, sterVoldgade 10, DK-1350 Copenhagen K, Denmark
c
University of Bucharest, Department of Systems Ecology and Sustainability/Research Center in Systems Ecology, Eco-diversity and Sustainability,
SplaiulIndependentei 91-95, Bucharest 050095, Romania

a r t i c l e i n f o a b s t r a c t

Article history: This paper investigates the personal and property characteristics of landowners who use EU Rural Devel-
Received 27 January 2014 opment agri-environmental schemes (AES), as well as their motives for participation or non-participation
Received in revised form in such schemes. The study is based on a questionnaire survey with landowners, in selected study areas in
16 September 2015
the Netherlands, Denmark, Austria, Italy and Greece. Our principal ndings show that AES tend to attract
Accepted 25 September 2015
more the owners of larger farms, who are frequently full-time, younger, post-primary school educated
and agriculturally-trained farmers. The latter ndings are contingent on local geographical particulari-
Keywords:
ties and on subjective factors, farmers individualities, different rural cultures, landscape types, EU and
Agri-environmental schemes (AES)
Subsidy use
national policies and special needs of the study areasall areas where agricultural production is increas-
AES participation ingly marginalized, for different reasons. Subsidy scheme participation motives did not seem to be strictly
Motives economic; they also regarded personal satisfaction. They are all together generally appeared to be place
Farm size specic, since the respondents from peri-urban Northern European areas were more motivated to par-
Farming engagement ticipate in AES than respondents from Central and Southern European areas with marginal potential
for agriculture. Motives for non-participation were also found to be dependent on the level of farming
engagement and on case-area landscape types.
2015 Elsevier Ltd. All rights reserved.

1. Introduction and context
For more than two decades, agri-environmental schemes (AES)
have been among the most important EU policy instruments,
in motivating farmers to improve environmental conditions, in
rural areas (Buller et al., 2000; Latacz-Lohmann and Hodge, 2003;
Vesterager et al., 2016). They are part of the Rural Development
Program (RDP), which supports farmers income and farm viability
and continuity and regulates the impact of agricultural production
in the European environment (European Commission, 2015).
The RDP including the AES represent a challenging policy eld.
On the one hand, they are designed as a uniform and transpar-
ent European policy framework, which should be applicable to all
member states. On the other hand, rural landscapes and farming
Corresponding author.
E-mail addresses: epavlis@geo.aegean.gr (E.S. Pavlis),
terkenli@aegean.gr (T.S. Terkenli), sk@ign.ku.dk (S.B.P. Kristensen), agb@ign.ku.dk
(A.G. Busck), georgialavinia.cosor@g.unibuc.ro (G.L. Cosor).
communities are very diverse across Europe; local conditions for
policy implementation, therefore, vary widely. Hence, there is a
need for more in-depth knowledge about the implementation pro-
cess and uptake of AES under the Rural Development Program,
in different landscapes, under different conditions for agricultural
production, across Europe.
This paper investigates European landowners motives for par-
ticipation in AES, in ve study areas, namely in the Netherlands,
Denmark, Austria, Italy and Greece. The focus of the present study
is on marginal agricultural areas and areas in transition which
are of interest since there is a clear need for targeted interven-
tions in marginal and remote rural areas, which involves either
reversing abandonment or managing a transition to a new land-
scape structure (Pedroli et al., 2015). For the purposes of our study,
we categorized land owners into four types of farmers: full-time,
part-time, hobby farmers and non-farmers. Specically, the survey
attempts to answer the following three questions:
What are the personal and property characteristics of those
landowners who participated in AES?

http://dx.doi.org/10.1016/j.landusepol.2015.09.024
0264-8377/ 2015 Elsevier Ltd. All rights reserved.
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812
Table 1
Key gure for AES implementation in the countries with case studies.
Country % of UAA under AESa Annual EU expenditure per ha supported in
200709 (euros/ha)
Denmark 11 80
Netherlands 4 130
Austria 70 190
Greece 10 340
Italy 19 85
EU-27 22 84
a
The agricultural area under AES represents the physical surface covered by AES, without double counting of areas enrolled in several measures. Source: European Union
(2011).
What are the main motives for (non-) participation in AES?
What are the environmental impacts of AES participation in terms
of farm-level land use changes?
In accordance with previous studies, we would expect smaller
farms to be less likely to participate in AES, and such participation
to be mostly inuenced by family-centred motivations, while there
would be different perceptions of farming and tradition. Further-
more, we would expect a considerable amount of non-prot-driven
motives in AES participation. In addition, we would expect to nd
non-production-oriented motives in land use change, whereas an
increasing number of farmers to be more and more concerned with
environmental issues (e.g. improvement of habitats).
2. Agri-environmental scheme use and farm
developmentdynamics and motives
2.1. The diversity of AES in the European Union
The basic aim of the AES is to encourage the protection and
management of the farmland environment by European farm-
ers, through economic subsidies (Council Regulation 1698/2005).
AES are a part of the Rural Development Programme (RDP)
and are, therefore, co-nanced by the Member States (European
Commission, 2015). They were rst introduced in the EU, in
the 1980s, through the voluntary regulation 797/85, in order to
improve natural and environmental conditions, in agricultural
areas. Until 1992, their implementation remained optional.
The accompanying measures (2078/92/EC), introduced as part
of the MacSharry CAP reform in 1992, were the rst set of manda-
tory AE measures, requiring all member states to implement AES.
Following the subsidarity principle, member states were free to
design measures, which were either targeted at particularly sensi-
tive areas (deep measures), sub-groups of farmers (e.g. livestock
owners), or landscape types (broad measures). Hence, a variety
of measures developed in the EU member states, during the 1990s,
including the protection of grasslands, reduction of fertilizer use
and/or wetland restoration (Buller et al., 2000; Latacz-Lohmann
and Hodge, 2003). In several cases, the measures were a contin-
uation or adaptation of existing national AE programs (Buller et al.,
2000). With the Agenda 2000 reform, in 1999, and the subsequent
Council Regulation (EC) No 1698/2005 on support for rural devel-
opment, AES became a key element of the RDP, under the second
Axis of the CAP. The measures included under the 20072013 EU
RDP were grouped into 12 categories (e.g. organic farming, man-
agement of landscape, pastures and High Nature Value farmlands,
integrated production and other extensication of farming sys-
tems). The EU budget allocated to AES has increased rapidly since
1993 and reached 3026 million Euros in 2010, which is still only
5% of the total CAP budget (close to 56,000 million Euros, in 2013).
In addition, national co-nancing increased the total amount spent
on AES, to around 5.035 million Euros, in 2013 (European Union,
2011).
801
Total expenditure per ha supported (EU and
national co-nancing) in 200709, (euros/ha)
120
450
340
400
190
163
The present research refers only to the recent 20072013 RDP.
The total budget for the 20072013 RDP was 96.3 billion Euros (20%
of the total CAP budget, for the same period). AES are part of Axis
2 Improving the Environment and the Countryside through Land
Management. The budget allocated to AES measures constituted
22% of the total expenditure for rural development, in the 200713
RDP (European Union, 2011).
According to the statistical and economic information report on
rural development, the level of AES implementation varies consid-
erably between member states. On average, 22% of the agricultural
area (UAA) for EU-27 was under AES. However, this varied con-
siderably, from 92% of the UAA in Luxembourg, to less than 5% in
Bulgaria. The most important types of agri-environmental com-
mitments, in terms of area enrolled, were those aimed at the
management of landscape, pastures and High Nature Value farm-
land (39% of the total area committed in EU-27). 14% of the total
agri-environmental area (almost 5 million ha) is classied under
the category other extensication of farming systems, which
includes measures aimed at the reduction or better management
of fertilizers, at plant products protection and at livestock extensi-
cation. Around 8% of the total area committed in the EU-27 was
devoted to organic farming and a similar share was directed to soil
conservation actions (European Union, 2011).
Table 1 is based on the above report and shows the proportion
of agricultural area covered by AES contracts, in 2009, in the coun-
tries where the case studies in the present paper are located. It
reects the diversity of implementation strategies developed, by
country: In Austria, the agri-environmental PUL scheme adresses
many areas and landowners, as illustrated by the very high pro-
portion of agricultural land covered (70%). Italy is close to the
average for EU-27, both in terms of area covered (19%) and annual
EU expenditure (85 euros/ha). The remaining countries (Denmark,
The Netherlands, and Greece) have less area under AES contracts
(between 4% and 11%), well below the EU-27 average. Interest-
ingly, two of these countries (The Netherlands and Greece) have
the highest total expenditure per ha (EU and national co-nancing
combined), amounting to 450 and 400 euros/ha respectively (more
than twice the EU-27 average), which indicates a deep approach
to AES implementation: funds are targeted to specic areas, which
receive substantial subsidies. In the case of The Netherlands, this
is primarily achieved with national top-up funds (72% of the total
expenditure per ha supported), while, in Greece, national top-up
only constitutes only 25% of the total expenditure per ha (European
Union, 2011).
2.2. Factors inuencing participation in AES
Member states have designed and allocated funds to AES, based
on a variety of principles and strategies, as discussed in the previous
section. Whatever the strategy applied, voluntary scheme partici-
pation is common to all countries and land owners and therefore
their decision whether or not to participate is based on a number
of parameters. These are typically classied as either farm charac-
802 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812

Table 2
Land use types in the VOLANTE study areas (Van der Sluis et al., 2013:14).

Lesvos, GR Portono, IT Reichra-ming, AT Roskilde, DK Heerde, NL

Environmental Zone Mediterr. South Mediterr. North Alpine South Atlantic North Atlantic Central
Size area (km2) 108 18 102 212 80
Urban sealed x X x x
Woody crops x X
Waterways X x x
Wetlands x x
Coastal x X
Geomorphological features x X
Native grasslands x X X
Heathland x X x
Scrub x X X x
Woody landscape elements x X x x
Stone walls & terraces x X

Table 3
Farm and land use characteristics of the study areas.

Study area Roskilde (DK) Heerde (NL) Reichraming (AT) Lesvos (GR) Portono (I)

Median farm size (ha) 15 7 37 4 1

Farm size: 0-1/1-10/10-50/>50 ha (% of farms) 7/38/33/22 0/53/36/11 3/8/47/36 4/68/24/4 48/52/0/0
Organic farms (%) 3 0 34 13 0
Properties with grazing animals (%) 38 87 96 30 8
Properties with dairy cows (%) 0 43 70 3 0
Crops (%) 82 21 8 1 14
Permanent grassland (not in rotation) (%) 6 77 32 14 16
Other (forest, perennial crops, etc.) 12 2 60 82 70
Number of respondents (no.) 93 47 73 90 25

Source: survey results.

teristics (production type, size, location), or farmer characteristics
(age, education, level of engagement in agriculture), and have been
studied using sociological/behavioral study approaches or more
quantitative/structural approaches (Wilson, 1997; Burton, 2006;
Mills et al., 2013). The farm and farmer parameters should not be
seen as isolated factors, but rather as interlinked parameters, in a
decision making process. In addition to these local (endogenous)
parameters, exogenous parameters originating at higher spatial
levels (general socio-economic conditions at local and regional
levels, local and regional physical plans, development programs,
infrastructure, etc) constitute important framework conditions,
which dene the room for manoeuvre available to each farmer
(Wilson, 1997).
Monitoring the levels of scheme participation is important for
decision makers, in order to be able to design relevant and attractive
schemes. Therefore, a lot of research has, over the past decades,
investigated the nature of these parameters and their role in the
decision-making process. This section reviews the importance of
different farm and farmer parameters, as investigated in previous
studies, on farmer participation in AES.
2.2.1. Farm parameters
In terms of property characteristics, past studies point to the fact
that smaller farms were less likely to participate in these schemes,
due to a lack of eligible habitats, simply because of the small farm
size (Skerratt, 1994; OCarroll, 1994). In some cases, though, small
farms contain relatively more relevant habitats per ha, as the farm-
ers main motives may not necessarily be prot-driven and they
may also actively pursue the improvement of habitat conditions. In
their meta-study of literature regarding participation in AES, Xavier
et al. (2015) concluded that it is easier for larger farms to participate
in AES. Ruto and Garrod (2009) and Wilson and Hart (2000) found
that large farms were more able to participate in AES. They also
found that grassland farms are more likely to participate in AES
than arable farms. Capitanio et al. (2011) reached similar results,
suggesting that the more a farm specializes in crop production the
less the probability to engage in AES, while the opposite applies for
farms specializing in livestock production. Wilson (1997), likewise,
found that farm size and the amount of remnant semi-natural
wildlife habitats were the most important parameters inuencing
participation in the Cambrian Mountains Environmentally Sensi-
tive Areas (ESA) scheme.1 In a meta-study of AES literature, based
on the assessment of 160 publications and research reports of six EU
member states, Siebert et al. (2006), also detected a general trend
for larger properties to tend to participate more in AES.
2.2.2. Farmer parameters
Several studies have investigated links between farmer type and
engagement in landscape changes, as well as AES participation.
Some of them showed that hobby farmers were responsible for
more landscape changes, often driven by non-production motives
(Primdahl, 1999; Primdahl et al., 2004; Gasson, 1986; Kristensen,
2003). However, Kristensen et al. (2016) show that full-time
landowners of the ve areas studied in this paper were respon-
sible for more landscape changes, than any other group. There is,
thus, not one answer, as to which farmer group is most actively
involved in landscape change, as this depends on contextual factors
(time, place, circumstances and motives) or a mix of several param-
eters. Primdahl et al. (2003) discuss improvement effects, which are
management changes reected in land use and protection efforts
(mainly concerning the use of fertilizer and pesticides). They found
that management and land use changes generally increase through
AES participation. There are some more indications that these
changes improve biodiversity (Kleijn and Sutherland, 2003; Kleijn
et al., 2006; Viaggi et al., 2015). However, there appears to be a
general lack of research on AES effectiveness and it was also dif-
cult to assess such policy outcomes, since there are many other
1
The ESA scheme rewards farmers who maintain traditional agricultural sys-
tems, against rural depopulation, agricultural intensication, etc, through specic
management practices.
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812 803

inuencing factors (Kleijn and Sutherland, 2003; Primdahl et al., more and more farmers are concerned with environmental issues
2003). and such farmers are appreciated by the wider public.
As far as personal characteristics are concerned, past research
shows that farmers with post-primary educational backgrounds
2.2.3. Motives for scheme participation
are more willing to participate in AES, most likely due to the fact
The nature of farmers motives for AES participation is com-
that they can deal more successfully with bureaucracy and are more
plex and dynamic and ranges from socio-economic to cultural and
interested in receiving training and relevant information (Peerlings
personal motives. Motiveseither single or in combinationsare
and Polman, 2009; Siebert et al., 2006; Wilson and Hart, 2000;
expressed as opportunities or constraints (economic reasons,
Wilson, 1997). Peerlings and Polman (2009) showed that AES par-
personal interests, family considerations, social issues, external fac-
ticipation and landscape management are associated with medium
tors, etc), followed by strategic decisions, in specic periods within
or high level educational background. Wilson and Hart (2000) nd
the family life-cycle (Potter and Lobley, 1996; Burton, 2006; Ingram
that those who completed schooling (full-time education) were
et al., 2013). Participation has been shown to be inuenced by
more likely to participate in AES. Wilson (1997) suggests that edu-
farmers identities (Ingram et al., 2013) and, generally speaking,
cation inuences the adoption of more environmentally friendly
by family-centred motivations. Family farm continuity, in particu-
practices or more land acquisition positively, while Siebert et al.
lar, has been observed to be a central and quite signicant motive
(2006) found that specialized knowledge may not always pre-
for AES participation (Ingram et al., 2013). Burton (2004) referred
vent environmentally harmful farming practices. Damianos and
to the symbolic social value of crops and livestock, as well as farm
Giannakopoulos (2002) and Peerlings and Polman (2009) reached
family structure and identity. The role of family in farm decision-
similar results. These studies suggest that, while education alone
making, including motivation to participate in AES, is crucial and
may not strongly determine AES participation, it certainly seems
seems to be signicant in farm business management and plan-
to inuence it. Findings by Skerratt (1994), Moss (1994), Polman
ning (Greiner and Gregg, 2011; Ingram et al., 2013). Moreover, for
and Slangen (2008) and Peerlings and Polman (2009), indicate that
many small farms, economic viability seems to depend on non-
age also inuences participation in AES, since older farmers tend
agricultural sources of income (Ingram et al., 2013; Peerlings and
to be more conservative and less exible, thus, it seems that it
Polman, 2009), while motives for land use change, as a conse-
is less attractive for them to participate in AES. Damianos and
quence of AES participation, could also be non-production-oriented
Giannakopoulos (2002) and Mathijs (2003) also concluded that
(Primdahl et al., 2004).
young and well-educated farmers were more willing to partici-
Mills et al. (2013) reviews the complex set of factors inu-
pate in AES, than farmers without these characteristics. However,
encing farmers activities and attitudes towards the environment.
Defrancesco et al. (2008) found that non-adopters of AES tended to
These factors include nancial motivation, but also personal inter-
be younger. Wilson (1997), also, found contrasting results. In this
ests, psychological (e.g. perceived corruption) and social issues (e.g.
study, he showed that younger farmers were more willing to par-
much greater sense of responsibility towards the environment than
ticipate in AES for conservation reasons, while older farmers were
that of previous generations), farming self-image, time, effort and
also willing to participate in the same schemes, but for mainly eco-
amount of risk, etc (Defrancesco et al., 2008; Berger et al., 2006;
nomic reasons. Thus, the age factor does not seem to be reliable in
Kleijn and Sutherland, 2003; Sattler and Nagel, 2010; Micha et al.,
terms of explaining AES participation.
2015). Consequently, a wide range of complex, and sometimes con-
Ingram et al. (2013) go beyond the discussion of individual char-
icting sociological and psychological factors impact on farmers
acteristics, to describe different types of farmers identities (i.e.
decision-making.
traditional vs. commercial farmers) and farm development path-
We have been inspired by the knowledge generated by the
ways, forming a spectrum from traditional pathways (characterized
research presented in this section, for the design of the concep-
by diversication, low-intensity, family-centrism, environmental
tual framework and research questions of this paper. In essence,
conservation, low inputs, debt and off-farm income) to commercial
we investigate the relationships between a selected set of farm
pathways (characterized by full-time production, food production,
and farmer parameters, on the one hand, and AES participation,
prot-centrism, and abstinence from joining AES). In their study,
on the other, in ve case studies. The farm parameters investigated
traditional farmers have been identied as stewards of the envi-
are: farm size, livestock, type of landscape/land use change under-
ronment and commercial farmers as food producers. Both of these
taken as a response to subsidy participation and type of subsidy
types of farmers seemed to be interested in farm continuity and in
used, while the farmer parameters are: age, marital status, educa-
securing family livelihoods; however, they seem to carry different
tion level, knowledge and use of subsidies. In addition, we study
perceptions of farming and tradition. Nevertheless, according to
the landowners motives for scheme (non-) participation. Motives
Ingram et al. (2013), pathways are constantly subject to change,
for AE subsidy scheme participation have been conceptualized as
making it difcult to predict farmers responses to AES, based on
follows: farm economic considerations, improved property quali-
their choice of farm development pathway. Pathways may change
ties, personal satisfaction, inuence of family/lifestyle reasons and
over time, as a response to life cycle changes; they often result
social network/local community reasons, while motives for non-
in growth and expansion for the farmers in their early life-cycle
participation have been: lack of knowledge, lack of information
stages or those preparing succession, and to marginalisation for
sources, lack of eligible areas, bureaucracy, risks involved, unwill-
retired farmers or those where succession is uncertain (Ingram
ingness to spare land for such schemes, too much ofcial control
et al., 2013). Espinosa-Goded et al. (2010) concluded that farm-
and too small subsidy amounts.
ers may accept lower payments, as long as they could maintain
(some) of their farming activities/management. Micha et al. (2015)
point out the signicance of non-nancial aspects and that different 3. Methodology
types of farmers prefer different policies. Wilson and Hart (2000)
found that Mediterranean farmers may still be caught in a produc- 3.1. Selection of case studies
tivist ethos of farming and feel that they have to catch up with their
northern counterparts (longer experienced in AES) with regard to The research survey was conducted in ve different European
farming efciency and commodity output (Wilson and Hart, 2000). countries, representing different types and conditions of rural
However, the attitude towards AES did not seem to be in conict landscapes; Denmark (Roskilde, DK), Netherlands (Heerde, NL),
with the concern for the environment, but rather the opposite, since Austria (Reichraming, AT), Italy (Portono, IT), and Greece (South-
804 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812

Table 4
Diversity in AES participation in the ve study areas.

IT NL AT GR DK Total AES participation No and (%)

Extensication 3 9 8 21 41 (32)
(subsidies for organic farming, permanent
grassland, farmyard planting)
Livestock 10 2 12 (9)
(subsidies for stock-breeding and mountain
farmers)
Nature/environment 13 43 1 57 (45)
(subsidies for meadow bird protection,
environmental management of cropland and
pasture, ecologically valuable areas and the
Austrian Agricultural Environmental Program)
Landscape 1 5 3 8 17 (13)
(subsidies for less favoured areas other than
mountain areas, landscape elements, set-aside
land)
Total: no. and (% of respondents in each study 1 (4) 21 (45) 65 (89) 10 (11) 30 (32) 126 (36)
area)a
a
Respondents may participate in more than one AES. The 126 scheme activities were distributed across 95 landowners (29% of all landowners in the survey).

Table 5
Implications of AES participation in farm management and land use change, in relations with case study location and agricultural education backgrounds.

Variables Asymptotic signicance (p value)a Result

- Case study location
- Land use change
- Agricultural education
- Land use change
- Case study location
- Specic changes in farm
management and land use
- Age
- Specic changes in farm
management and land use
0 NL: a considerable amount of respondents reported that AES participation had
led to a more diversied landscape, while nearly half of them reported no
change
AT: a considerable amount of respondents converted land to grassland, while
nearly half of them reported no change
DK: converted land from cropland to grassland, while a minority (compared
with other study areas) reported no change
GR, IT: the great majority reported no change
0.004 It was mostly respondents with an agricultural education background who
answered that they effectuated changes in farm management and land use,
while the majority of those lacking such education reported no change
0 AT: CAP reform and its implications on farm management and land use appear
to have led to an increase in the areas eligible for support and to some
extensication of management.
NL, DK: the majority reported no change
0 Young landowners appeared to be more involved in land use changes, a
propensity reducing with age

- Reason for property 0 Changes to farm management and land use were made by respondents who
acquisition bought or acquired their property, mainly because it was a good place to farm
- Specic changes in farm
management and land use

- Size of property 0.028 Most of these changes regarded farms larger than10 ha
- Specic changes in farm
management and land use

a
A signicance level of 95% was used. The chi-square test fullled the criteria of statistical signicance with respect to the minimum number of observations. We regarded
a 0.05 p-value (calculated as the asymptotic signicance of SPSS chi-square test), as a statistically signicant difference and basis for conrmation of a correlation between
variables.

east Lesvos Island, GR). The case studies were selected as examples
of various landscape types in Europe, where farming seems to be
in a transitional or marginal condition, vis--vis predominant land
uses and alternative economic activities.
Roskilde is a municipality in Denmark that can be described as a
peri-urban Northern European landscape, with pronounced urban
growth and high demand (especially over the past 5 decades) for
converting agricultural land to residential, recreational and indus-
trial land. Historically, agriculture has been the main land use, in
this area, but the number of farms has been decreasing. The major-
ity of landowners are part-time or hobby farmers and non-farmers,
while land has been attractive for purchase for recreational and res-
idential purposes. During the last two decades, many farms have
been transformed from full-time to hobby farms and production
and land use have become more and more diversied. There is
an overall decline in the number of full-time farmers and agricul-
tural production is concentrated on fewer and larger holdings. The
majority of properties are used for residential purposes by hobby
farmers, who rent out their land to full time farmers.
Heerde is a middle-sized municipality in the Netherlands,
highly representative of a Northern European agricultural land-
scape under urban pressure, slowly opening up to agro-tourism
and domestic tourism. The farming population here is ageing; many
of the farmers are retired and the number of their successors lim-
ited. Part-time and hobby farmers constitute the dominant farming
category, while, during the last 10 years, their share of Utilised Agri-
cultural Area (UAA) has been increasing. Many of these farmers
depend on off-farm income and tend to diversify their farm produc-
tion, especially after the outbreak of the foot-and-mouth epidemic.
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812 805

Table 6
Statistical relationships between selected parameters and AES participation.

Variables Asymptotic signicance (p value) Interpretation

- Reason for property 0 Respondents who used subsidies tended to have acquired the property, mainly
acquisition because it was a good place to farm, rather than to live on; this nding has a
geographical focus in Austrian and Greek case studies
- Size of property 0 Use of subsidies increases with the size of property

- Age 0.001 Respondents younger than 30 years tended to use AE subsidies more than other age
groups
- Level of engagement in 0 Nearly half of those respondents who used subsidies were full-time farmers, while the
farminga participation of part-time farmers was less; the participation of hobby farmers was
lower, while the participation of non-farmers was almost non-existent. The
participation in AES grows with the level of engagement in farming
- Educational background 0 Respondents of especially vocational educational background seems to have
participated more in a subsidy scheme, while respondents of primary education have
participated less
- Agricultural education 0 Respondents with agricultural education participated more in subsidy schemes, than
those lacking such education
a
100%: Full-time farmer; 50-99%: Part-time farmer; 1-49%: Hobby farmer; 0%: Not a farmer.

Table 7
Main motives for joining subsidy scheme, in each case study location.

Motive Disagree Not important Agree Asympt. signicance (p value) Result

- Farm/economic 7 4 87 0 Mostly applied for AT, GR and NL.

considerations (e.g. increase Applied less in the DK case
of income, reduction of
economic risks, new
investments)

- Personal satisfaction (e.g. 58 8 33 0.027 Strong disagreement in NL, DK, but small
learning new skills, sample
self-challenging,
professional pride,
moral/ethical motives)

They also seem to be choosing to extensify their agricultural activ-
ities, especially during this past decade.
Reichraming is a mountainous municipality, located in the
province of Upper Austria, and has the characteristics of an Alpine
rural landscape. Economically, the area depends mainly on tourism,
agriculture and forestry. Farming is family-based, with a strong
sense of family tradition (family farming). Tourism is developing,
while agriculture is declining, and farming incomes remain low.
In the study area, agriculture mostly involves raising cattle and
producing milk. During the last decade, small farms have tended
to disappear or be absorbed by larger ones. Most of the farmers
are full- and part-time farmers. However, farmers tend to choose
to intensify their agricultural activities, but, at the same time, to
become less labour intensive, allowing for more part-time farm-
ing. Nonetheless, few farms diversify and subsidies are considered
as an important source of income, consequently inuencing land
use and management.
Portono is a small peninsula between Genoa and Sestri Lev-
ante (Liguria) in Italy, a well-known resort and a traditional village,
highly representative of a Mediterranean landscape. The study area
is Portono Regional Park. Nature conservation became important
here, over the past decade, while marketing for tourism purposes
has given rise to second home construction. At the same time, agri-
culture has been marginalised and tends to disappear. The farming
population is rapidly ageing, leaving no successors. They are mostly
part-time and hobby traditional small-scale farmers, gaining most
of their income from other-than-farming activities. There is not
enough motivation to continue farming and young people tend to
migrate to nearby towns, or to turn mainly to the tourism sector
for employment.
The Municipal County of Mytilenelocated in the southeast part
of Lesvos, a middle-sized northern Aegean island in Greecemay
be characterized as a traditional Mediterranean coastal-rural
landscape, slowly opening up to small-scale tourism. Farming is
family-based and of low intensity. The majority of the farmers are
hobby and part-time farmers, mostly living in semi-urban regions,
in the south-eastern part of the island. Farming activity mainly
involves olive cultivation and spreads over the eastern part of the
island; it is of lesser importance on the desertied lands of the west-
ern part of the island. Most of the households in the study area
depend on off-farm income and the largest proportion of the pop-
ulation deals in wholesale and retail trade, construction and public
administration.
Following the typology adopted by Kristensen et al. (2016), the
study areas represent two important landscape types:
i) Peri-urban landscapes (Roskilde and Heerde), dynamic land-
scapes, with conicting land uses of recreation and production.
In both areas, the agricultural potential is quite good, but the
farming sector is nevertheless challenged by erce compe-
tition for land, as farm land is increasingly used for urban
development, or to cater to non-agricultural interests (industry,
manufacture, residential use). Due to its peri-urban location, the
agricultural sector is frequently not as competitive as other eco-
nomic sectors, or able to attract investment, due to alternative
economic opportunities and strict planning regulations.
ii) Landscapes with marginal potential for agriculture (Reichram-
ing, Lesvos and Portono), facing challenges of accessibility,
land abandonment, outmigration of young people, and limited
protability of on-farm activities and decrease of traditional pro-
duction.
Each of the study areas contains urban mixed-use areas, areas
of urban recreation, as well as crop areas, grasslands and forests
806 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812

Table 8
Cluster groups of landowners level of agreement with motives for participation in subsidy schemes.

Cluster 1 34.2 % (25 respondents). For these respondents, the social network/local community motives are not important (76 %). Half of them e
participated in AES for personal satisfaction, to improve qualities on property, and for farm economic considerations
Cluster 2 50.7 % (37 respondents). The principal motivation of the majority of respondents was the economic one, as 97.3 % of them totally agreed
with the farm economic considerations/gains motive, the other motives having been rated as not important
Cluster 3 15.1 % (11 respondents). The majority of the respondents completely disagreed with all personal (100 %), family (81.8 %) or social motives
(90.9 %), while one third (36.4 %) of them somewhat agreed with the economic motive for AES participation

(Table 2). Generally speaking, Mediterranean land use types (IT,
GR) seem to vary the most, since they include a higher diversity
of land use types (e.g. woody crops, coastal areas, geomorpholog-
ical features, stone walls and terraces) than the other study areas,
although they tend to lack waterways.
3.2. Survey methods and techniques
Data were collected with the aid of a structured questionnaire
survey, with a sample consisting of 328 landowners; the survey was
conducted during the period FebruaryMay 2012. An effort was
made to include 100 respondents from each study area, in order to
fulll requirements for standard quantitative analyses, e.g. factor
analysis (MacCallum et al., 1999) and logistic regression (Peduzzi
et al., 1996). However, specic area particularities had to be taken
into consideration and the survey methodology applied in each
study area was adapted to local conditions and data availability,
which varied considerably from case to case (i.e. survey samples
varied from 25 in Portono to 99 in Roskilde), reecting differences
in the size of the study areas and in local land ownership patterns.
In Portono (IT) and Reichraming (AT), where the population was
much smaller than 100, we achieved almost 100% participation.
A few people refused to cooperate, due to mistrust towards the
researchers. In general, an attempt was made to select respondents
as representative as possible of the areas under study, in terms
of local and regional production systems and other geographical
particularities. For the selection of our samples, we relied on local
agricultural register access, using simple random and systematic
sampling techniques. Landowner personal and property character-
istics, which may have inuenced decision-making irrespective
of the biophysical characteristics of the area were also taken in
consideration (Kristensen et al., 2016).
The great majority of the questionnaires were completed in
structured face-to-face 30-45 min interviews, except in Reichram-
ing (AT), where the questionnaires were completed independently
by the respondents. Potential respondents were rst contacted by
telephone, in order to set up an appointment for a face-to-face
interview; some interviews, however, were conducted fully as tele-
phone interviews.
As outlined in Section 2, land owners decision concerning par-
ticipation in AE schemes is frequently understood as the result of
a process involving both farm and farmer characteristics. Further-
more, land owners decisions are guided by motives which can be
classied into a number of categories. The purpose of the ques-
tionnaire survey was to collect information on these parameters,
as described in Section 2.
3.3. Key characteristics of farms and respondents
The study areas selected were quite diverse in terms of farming
systems and land use (Table 3). The median farm size (as opposed to
average farm size) was smallest for the Mediterranean study areas
(1 ha and 4 ha for Portono and Lesvos, respectively) and largest for
Reichraming (37 ha). The majority of farms in Reichraming were
larger than 10 ha, which was the opposite to all other study areas.
Furthermore, in Reichraming, the most common land use was for-
est (60%), which differs from all other areas where forest cover was
much smaller. Perennial crops (olives) were the most common land
use in the Mediterranean study areas (70% and 82% for Portono
and Lesvos, respectively), whereas annual crops (mainly cereals)
were the dominant land use in Roskilde (82%). A large proportion
of properties in Heerde and Reichraming had grazing animals (87%
and 96%, respectively). Both of these areas also featured large areas
of permanent grassland, typically either High Nature Value farm-
land, or high biodiversity interest land. Reichraming differs from
other study areas, in its high proportion of organic farms (34%), fol-
lowed by Lesvos (13%), and at much smaller proportions from the
rest of the case study areas.
In addition, the questionnaire survey revealed that the majority
of respondents:
Grew up on the farm/property,
Had vocational educational backgrounds,
Were between 40 and 69 years old,
Had owned the farm for two or three generations, and
Had acquired property because it was good for farming.
3.4. Data analysis
The conceptual framework which underlies the present study
relies on both the theoretical placement of this research and on
our qualitative and quantitative information, with the main objec-
tive to identify the relationship between, on the one hand, selected
farmer and farm parameters and, on the other hand, land owners
level of AES participation. We, therefore, used the chi-square test
to determine the level of signicance of the correlations between
selected variables and the use of subsidies, an approach which has
widely been used in other studies of landowners participation in
AES (Wilson, 1997; Mills et al., 2013). The data were subsequently
analysed with the SPSS; the statistically signicant results (only)
are presented in Section 4.
Since we assumed that several parameters may jointly inuence
land owner decisions concerning AES participation, we used a mul-
tivariate approach to identify signicant key parameters. Therefore,
a logistic regression analysis was conducted to test how farmer and
farm parameters impact on subsidy use or land use change. This
multivariate analysis tested the relationship between use of AES
and selected farm and farmer parameters, based on ndings from
previous research, as discussed in Section 2.2. Parameters such as
geographical location of the study area, farm size, farm share of
income, educational background, landowner type, age, and place of
farmers upbringing and property purchase, thus allowing for the
identication of statistically signicant relationships (Kristensen
et al., 2016).
Finally, a two-step cluster analysis test was conducted, in order
to group the 95 respondents who had participated in AES, according
to categorical data on motives for subsidy use, and to identify simi-
larities in respondents preferences. The cluster analysis was based
on information provided by those respondents, who had rated
motives for scheme participation on a ve-point Likert Scale (Com-
pletely Disagree, Somewhat Disagree, Not Important, Somewhat
Agree and Totally Agree).
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812 807

Table 9
Main motives for not participating in AES, with respect to the case study location.

Motive Disagree(%) Not important (%) Agree (%) Asymptotic signicance (p value)

- Availability of eligible land 31 25 45 0

- Small area 41 31 28 0.002
- Amount of knowledge 19 35 46 0
- Degree of bureaucracy 15 44 41 0
- Availability of information sources 17 47 36 0
- Degree of control (by ofcials) 24 56 20 0
- Amount of risk 28 46 26 0
- Subsidy amount 25 48 27 0

4. Results characterization of the relationship between the selected variables

4.1. Patterns of AES participation
29% of our respondents had received AE-related subsidies; most
of them were from the Austrian case study. The remainder, who had
not received any AE-related subsidies, were mostly from the Italian,
Greek and Danish case studies.
Different types of subsidies were used in different study areas
(p = 0), as shown in Table 4, along with the distribution of sub-
sidy use per case study, by country. Our research shows that the
most popular subsidies seemed to be used for nature/environment
management (especially for the Austrian Agri-Environmental Pro-
gramme PUL), while the least popular ones were to maintain
grazing of sensitive grassland areas (livestock raising). Both of them
were almost exclusively used by respondents in the Reichram-
ing (AT) study area. Subsidies to extensify agricultural production
mostly involved land owners in Roskilde (DK). The largest pro-
portion of AES participants were found in Reichraming (89% of all
respondents in that study area) followed by The Netherlands (45%).
The participation rates were much smaller in the other study areas.
In Reichraming (AT) and Roskilde (DK), there has been more variety
of subsidies, while in Lesvos (GR) there has been almost only one
type.
A logistic regression analysis of the relationship between, on the
one hand, selected farm and farmer parameters and, on the other
hand, AES participation showed that the percentage of accurately
predicted number of respondents improved from 71%, calculated as
a simple allocation of all cases to the most dominant result (non-
use of subsidies), to 88%, through the use of multiple variables. This
suggests that the model constituted a signicant contribution to the
and AES participation. The variables that were most signicantly
correlated to the use of AES were farm size and presence of grazing
animals on the farm.
4.2. Land use change following AES participation
Many AES have a clear spatial scope and aim to maintain or
improve habitats or threatened types of land use (e.g. alpine mead-
ows, grasslands, wetlands), while others are of a more non-spatial
nature (e.g. reduction of nitrogen use, organic agriculture), which
have mostly farm management implications and only indirect land
use impacts. We investigated the patterns of land use changes,
as a consequence of AES participation and the links with farm
and farmer parameters. The results indicate that there are differ-
ent types of land use change, depending on case study location
(Table 5). All in all, various changes in land use and farm manage-
ment have taken place, as a result of AES participation. Such changes
seemed to involve younger farmers with an agricultural educa-
tion background, who were familiar with participation in subsidy
schemes, and usually referred to a farm size of over 10 ha, as Table 5
indicates.
Table 6 presents ndings about AES participation and their
correlation to the farm and farmer parameters of statistical sig-
nicance.
4.3. Motives for AES participation
The great majority of the respondents (89%) referred to farm
economic considerations and gains (e.g. increase of income, reduc-
tion of economic risks, new investments) (Table 7), as the main

Table 10
Main motives for not participating AES, with respect to the level of engagement in farming.

Motive Disagree Not important (%) Agree (%) Asympt.signif.(p value)

(%)

- Degree of bureaucracy Full-time: 8 29 62481727 0.046

Part-time: 15 37
Hobby: 25 58
Non: 9 64
- Availability of information sources Full-time: 22 39 39344821 0.016
Part-time: 19 46
Hobby:18 33
Non: 5 76
- Amount of risk Full-time: 21 32 47 33 0 15 0.015
Part-time: 27 40
Hobby: 52 48
Non: 10 75
- Amount of land used for AES Full-time: 67 17 17 6 38 75 0.02
Part-time: 44 50
Hobby: 35 27
Non: 25 0
- Amount of subsidy Full-time: 11 42 47331310 0.045
Part-time: 29 38
Hobby: 43 43
Non: 10 80
808 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812
motives for AES participation. However, in Roskilde (DK), as
opposed to the other case studies, economic considerations were
not deemed to be such a strong motive to participate in AES.
Personal satisfaction (e.g. learning new skills, self-challenge, pro-
fessional pride, moral/ethical reasons) appeared to be another
motive for participating in AES, but this result did not seem to be
the case in Roskilde (DK), where half of the participants disagreed,
as well as in Heerde (NL), where almost half of the participants
declared it was not important. Nonetheless, no safe conclusions
could be reached, in terms of geographical dispersion, since the
sample of the participants of the above case studies was small (14
participants for Roskilde and only 7 participants for Heerde).
In addition, a two-step cluster analysis was undertaken with the
aid of the SPSS, in order to categorize motives for AES participation.
The cluster analysis resulted in the three clusters, shown in Table 8.
The likelihood ratio chi-square test conducted on the rela-
tionship between these three landowner clusters and the ve
motives for AES participation (farm economic considerations,
improved property qualities, personal satisfaction, inuence of
family/lifestyle reasons and social network/local community rea-
sons) showed a statistically signicant relationship for all of these
variables. Cluster 1, consisting 34.2 % of respondents, indicated that
community motives were not considered as important, and half
of them declared an agreement with personal satisfaction motiva-
tional motives, as well as with the motive of improving quality and
maximizing the revenue of the property. Cluster 2, consisting of
50.7 % of respondents, revealed a strong agreement with economic
motivation. Finally, cluster 3, consisting of 15.1 % of respondents,
revealed a strong disagreement with personal, social and familial
satisfaction motivational motives.
4.4. Motives for AES non-participation
Turning to the analysis of motives for not participating in the
AES, lack of knowledge of scheme opportunities was a much-
quoted reason for non-participation (46%). Other motives for
non-participation were lack of areas eligible for AES participation
(45%), bureaucracy (41%), lack of information sources (36%), risks
involved (26%), unwillingness to spare land for participating in such
schemes (28%), the small subsidy amount (27%) and, nally, too
many ofcial controls (20%). All of the above motives, were sta-
tistically signicant for non-participation in AES, and correlated,
on the one hand, with different categories of landowner character-
istics (landowner type), and, on the other hand, with geographical
location, at a medium degree of statistical signicance (Tables 911
).
More specically, limited availability of eligible land and small
farm area were mostly observed, in the Greek study area. Most of
the Italian respondents strongly agreed that lack of knowledge was
a reason for non-participation in AES, along with the lack of infor-
mation sources. For Austrian respondents, bureaucracy, the degree
of control (by ofcials), and the amount of risk involved seemed to
be serious constraints for not participating in AES, as was the lack
of information sources. No clear and signicant motives for non-
participation of the majority of the Dutch and Danish respondents
have been detected, since their answers varied. The great major-
ity of the Italian, Dutch and Danish respondents disagreed with the
availability of land and small area ownership, as reasons for not par-
ticipating in AES (or declared it was not important). In addition, half
of the Danish respondents agreed that the amount of knowledge
was not important for non-participation in AES, a viewpoint shared
by Dutch respondents. Bureaucracy did not seem to be an AES par-
ticipation issue, for Danish and Italian respondents. The majority of
the Italian respondents rather disagreed that a high degree of con-
trol (by ofcials) and the amount of risk involved discouraged them
from AES participation, a viewpoint also expressed by the majority
of the Dutch respondents. Finally, a small subsidy amount appeared
not to be an important factor for non-participation, especially for
the Danish study area (Table 9).
The level of agreement with bureaucracy as a reason for not par-
ticipating in AES seemed to grow with the level of engagement in
farming. Lack of information sources seemed to be a strong reason
for hobby farmers non-participation in AES, while it seemed not to
be an important reason for non-farmers especially, but also part-
time farmers, to a lesser extent. The amount of risk involved seemed
to grow with the level of engagement in farming. A considerable
percentage of hobby farmers agreed that the size of their farm area
was small and could not have spared land for these schemes, while
part-time farmers stated that unwillingness to spare land for par-
ticipating in such schemes was not an important reason for not
having entered the respective schemes. The number of full-time
respondents and non-farmers who answered this question was too
small to draw rm conclusions. The importance of the small sub-
sidy amounts for not participating in AES seemed to grow slightly,
with the level of engagement in farming (Table 10).
Lack of knowledge was a reason mentioned as an important rea-
son for not participating in AES, as a possible requirement to enter
such schemes by those lacking agricultural education. Respondents
with an agricultural educational background agreed more than
those without with the bureaucracy problems, as a factor for non-
participation in AES. Respondents with an agricultural educational
background agreed more than those without with the problem of
too much ofcial controls, as a factor for non-participation in AES.
Respondents with an agricultural educational background agreed
more than those without that the amount of risk involved was a
factor for non-participation in AES (Table 11).
5. Discussion
AES participation is one of the main EU policy instruments
for the improvement of natural and environmental conditions, in
rural areas of Europe. In this paper, we investigated the pattern
of AES participation in ve study areas, representative of different
landscape types. We analysed the relationship between the char-
acteristics of the respondents, the use of subsidies and motives for
AES (non) participation. We also investigated the impact of AES
participation on farm management and land use change.
5.1. Study areas and types of AES: different schemes for different
areas
Differences in the types of AES available and in the types of
AES used around Europe abound and also inform our ndings. The
respondents in Reichraming (AT) reported that they used AES, to a
great extent, as did respondents in Heerde (NL). In contrast, espe-
cially respondents in Portono (IT), but also those in Lesvos (GR),
used them the least. Specically, Greek AES participation concerned
extensication; Austrian AES participation concerned nature, the
environment and livestock; Dutch AES participation concerned
nature, the environment and the landscape, whereas Danish AES
use concerned extensication and landscape measures. Gener-
ally speaking, subsidy types seemed to reect the different forms
of agriculture and the environmental conditions and challenges
as characteristic of the landscape types of the case studies. The
Reichraming (AT) case study illustrates this very well. The widely
used PUL scheme can partly be attributed to a successful continu-
ation of a national AES, pre-dating EU membership. This is reected
in the national gures for AES enrollment, whereby 70% of the AAU
of Austria is covered, under the 200713 RDP (European Union,
2011). In addition, the difcult agricultural conditions and the need
for protection of sensitive habitats (e.g. mountain meadows) in
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812
Table 11
Main motives for not participating in AES, with respect to the level of agricultural education.
Motive Disagree
(%)
- Amount of knowledge No: 12
Yes: 32
- Degree of bureaucracy No: 20
Yes: 8
- Too much control No: 24
Yes: 25
- Too risky N 58newline 30
o: 29
Yes: 26
Reichraming (AT)typical for the Alpine regions in Austriahas gen-
erated a large interest in scheme participation. Moreover, many of
the landowners had converted to organic farming, and thus ben-
etted from an AES, which mostly concerned farm management
changes. The Roskilde (DK) case study exemplies a different pat-
tern of AE use, which, in turn, reects a very different agricultural
situation. Roskilde (DK) is home to some of the most fertile soils
in Denmark; the marginal position of agriculture is due to its peri-
urban location, rather than difcult physical conditions. Half of the
AES used here by land owners concerned grassland management
(various types of establishements, extensive management and pro-
tection of grassland). Therefore, scheme participation may, in some
cases, lead to a land use change, while, in other cases, it maintains an
existing land use. Since 2007, AES has only been available to farmers
with land in NATURA 2000 areas. Hence, eligibility is rst and fore-
most linked to NATURA 2000 designations, rather than particular
farm and farmer characteristics, although these may strongly inu-
ence scheme participation, in a subsequent phase. The Lesvos (GR)
case study offers a different perspective. This area is considered as a
marginal rural landscape, from an EU perspective, slowly opening
to tourism. Despite the fact that the area is still highly depend-
ing on olive cultivation as its main land use, AES participation is
not widely spread, mostly attracts organic farmers, and does not
seem to impart land use change. Lesvos farmers do not consider
themselves stewards of the environment, but rather (small-scale)
food producers, and show no special interest in environmen-
tal/landscape conservation. Generally speaking, farmers tend to
oppose ecologists and other groups promoting nature protection
and better management of the farmland environment.
5.2. Farm characteristics
5.2.1. Farm size
Our ndings indicate that AES participation grows with the size
of property. These results suggest that AES were more attractive
to owners of larger properties. Such a trend has also been indi-
cated by other studies (Xavier et al., 2015; GRAIN Report Hungry for
Land, 2014; Mills et al., 2013; Labarthe and Laurent, 2013; Capitanio
et al., 2011; Ruto and Garrod, 2009; Siebert et al., 2006; Wilson and
Hart, 2000; Wilson, 1997). This might be due to the frequent lack of
habitat eligibility, due to to the small property size (Skerratt, 1994;
OCarroll, 1994).
5.2.2. Land use change
Half of our respondents indicated that their land use of their
property has been changed, as a result of AES participation. In
Roskilde (DK), most of the respondents declared that part of their
land has been changed (76%), contrary to Mediterranean study
areas (GR, IT), where almost no land use change was observed. In
the latter study areas, AES concerned landscape protection, con-
servation and maintenance, rather than land use change per se.
809
Not important (%) Agree (%) Asympt. signif. (p value)
32 56 0.013
40 27
55 25 0.03
33 60
65 11 0.05
43 32
13newline 43 0.005
In Roskilde (DK) and Reichraming (AT), most of the area was con-
verted to grassland, whereas the other frequent land use change
concerned a change to a more diversied landscape in Heerde
(NL). Furthermore, in terms of farm management change, only in
Reichraming (AT), an increase in the areas eligible for support and
some extensication of management has been statistically evident.
There is a statistically signicant correlation between age and
land use change, since we found that most land use changes
involved younger farmers, whereas such changes were less com-
mon for older age groups. We also found that land use changes
were signicantly more common among those respondents with
an agricultural education background.
In the Austrian, Dutch and Danish cases, where land use
change was observed, land transitions concerned more produc-
tive approaches to land use change, leading to more diversied or
to different land-cover landscapes (e.g. conversion of cropland to
grassland).
5.3. Farmer characteristics
5.3.1. Land owner type
Our ndings show that subsidy use grows with the level of
engagement in farming (determined by the time spent in farming
activities). These results suggest that, in our study areas of more
or less marginal agriculture, subsidies tended to be more attrac-
tive to full-time farmers (who usually owned the larger farms).
Such a trend has also been partly indicated by previous studies, in
the past, namely that full-time farmers are more likely to join the
schemes, than part-timer farmers (Wilson, 1997). Certain research
results, however, seem to indicate that the contrary may also occur
(Wossink and van Wenum, 2003; Defrancesco et al., 2008; Ruto and
Garrod, 2009; Peerlings and Polman, 2009).
5.3.2. Land owner age and educational background
According to the respondents, the age factor seemed to affect
scheme participation, since there was a tendency for farmers over
70 years old not to join subsidy schemes, as opposed to younger age
groups (below 30 years). Younger landowners of our sample stated
that they were willing to participate in subsidy schemes, more often
than older ones. The latter trend is possibly due to their higher
exibility and capacity for new knowledge acquisition, as well as
adaptation to different and novel farming or market demands and
challenges.
We detected a statistically signicant correlation between
AES participation and the post-primary education level, prob-
ably because scheme participation may demand basic kind of
knowledgebesides management expertise and other agricultural
skillssuch as reading, writing and communication skills. In addi-
tion, we found that the agriculturally educated/trained respondents
tended to participate more in AES, than those lacking such educa-
tion/training, probably because they were more trained in dealing
810 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812

with CAP related issues. These ndings are in line with previ-
ous research ndings, whereby young farmers with post-primary
educational levels seem to have been more likely to enter AES
(Peerlings and Polman 2009; Siebert et al., 2006; Damianos and
Giannakopoulos, 2002; Wilson, 1997; Skerratt, 1994; Moss, 1994).
We also detected the same correlation between subsidy use and the
vocational educational level, as well as with agricultural education.
However, this parameter was geographically linked to Reichraming
(AT), since almost half of the respondents with a vocational edu-
cation came from this study area; the results are, therefore, biased
towards this area.
5.4. Motives for AES participation
5.4.1. Economic and non-economic motives
The main motives quoted for participating in AES appeared to
be economic, and thereby concur with other studies (Xavier et al.,
2015; Ingram et al., 2013; Mills et al., 2013; Brouwer and Lowe,
1998; Whitby, 1996). Indeed, according to Wilson (1997, p. 77),
payments are an important characteristic of any AES, and could
strongly inuence farmer participation in the ESA scheme and
according to Wilson and Hart (2000, p. 2169) nancial reasons are
the main deterrents for participation. In Roskilde (DK), farm eco-
nomic considerations and gains seemed to affect participation in
AES the least, in contrast with the general trend. When discussing
issues of economy, land owners in Roskilde (DK) generally consid-
ered that the compensation was too small to be of signicance for
their overall respective decision (Busck and Kristensen, 2014).
Motives also differed among case studies with marginal land-
scape types. Our ndings suggested difcult conditions and
circumstances (e.g. bureaucracy, risk taken, ofcial control) for
farm enterprises in the Austrian case study, lack of eligible land
for AES participation in the Greek case study, and need for relevant
education and information on AES in the Italian case study. Respon-
dents from peri-urban landscapes (Dutch and Danish) seemed to be
more motivated, than respondents in the case of landscapes with
marginal potential for agriculture.
Personal satisfaction motives associated with self-esteem also
tended to inuence land owner participation in AES. The same
applies to motives of property quality improvement, albeit to
a much lesser extent. Learning new skills, engaging in self-
challenges, professional pride and other moral/ethical motives
were other important motives for AES participation. In contrast,
motives concerning social networking, local community pursuits,
as well as family inuence and lifestyle, as well as the inuence of
property-quality improvement did not seem to be very important.
These ndings partly support Wilsons point that these farmers
have acted not only for pragmatic motives, but also for conser-
vation or nostalgia [. . .] even without great nancial rewards
(Wilson, 1997; Espinosa-Goded et al., 2010), in order to preserve
traditional farm values and cultural capital (Ingram et al., 2013;
Burton et al., 2008; Burton, 2004; Wilson and Hart, 2000), or to
consider environmental issues (Wilson and Hart, 2000; Dupraz
et al., 2003; Berger et al., 2006; Defrancesco et al., 2008; Ruto and
Garrod, 2009; Barreiro-Hurle et al., 2010; Xavier et al., 2015) and
with Primdahls et al. (2004) ndings on the signicance of non-
production-oriented motives for undertaking landscape changes.
6. Conclusions
The AES have been among the most important EU policy instru-
ments, in motivating farmers to improve environmental conditions,
in Europes rural areas. However, their design and implementation
represent a challenging policy eld, due to the diversity of land-
scapes and farming communities, across Europe. This study has
investigated the characteristics of AES participation in ve different
countries, in diverse and more (IT, GR, AT) or less (DK, NL) agricul-
turally marginal locations in Europe, two extensive geographical
areas (Central-Mediterranean and Northwestern Europe), and two
main landscape types (peri-urban and marginal agricultural land-
scapes). Its primary goal was to examine and critically assess
farmers motives for (non-) participation in AES, in correlation with
farmers and farms characteristics.
As far as farm characteristics are concerned, AES seemed to be
most attractive to larger properties or farms. In terms of farm-
ers characteristics, AES participation seemed to favour farming,
as a rather professional engagement, benetting full-time farm-
ers, since the use of subsidies tended to grow proportionally with
the time spent in farming. Agriculturally educated younger farm-
ers appeared to be more prone to participating in AES than older
ones, while the opposite applies to those with primary education
backgrounds, pointing to the fact that education and, in specic
agricultural education, also seems to impact on subsidy use.
According to our research ndings, subsidies led to more land
use changes in our in Northern and Central European (NL, AT) case
study areas, whereas land use changes were less common, as a
result of AES participation, in our Mediterranean study areas (GR,
IT). In general, reported land use changes, as a result of AES par-
ticipation, were more common on the larger farms of the study
areas, in Northern Europe, they involved younger farmers and they
were more common for respondents with an agricultural education
background.
Regarding farmers motives for (non-) participation in AES, farm
economic considerations and gains appeared to be at the top, but
personal satisfaction motives were also important, indicating that
farmers also perceived AES, as a way to accomplish self-esteem,
secure more resources for landscape activities, and fulll them-
selves through farming. Motives for participation differed among
different study areas and landscape types. Specically, respon-
dents of peri-urban Northern European landscapesfeaturing a
variety of economic sectors and opportunities and more devel-
oped diversication of agricultural productionseemed to be more
motivated in AES participation, On the other hand, less motiva-
tion was reported by respondents living in Central and Southern
European landscapes, with marginal potential for agricultureno
longer the dominant economic sectorin less competition with
other economic sectors and less diversied farm production. Hobby
farmers were the group that participated in AES the least, and
appeared to be inuenced by different types of motives, as opposed
to those were more engaged in farming. Their planning or implica-
tion of joining AES in modern agriculture does not seem to comply
with the nature of their farming engagement and, therefore, to
apply for this group. The same applies to non-farmers, as would
be expected.
To sum up, farmers participation in AES seemed to be corre-
lated with characteristics of landscape type, farm location, farm
size, level of farming engagement, and farmer age, education and
type of motivation. Specically, scheme participation seems to
be attractive primarily to larger farm owners, full-time farmers
and more productive-oriented, younger and agriculturally-trained
farmers with post-primary education. Land use change mostly
seemed to involve younger farmers with an agricultural education
background. Farmers were motivated by farm economic consider-
ations and gains, but also by non-economic personal satisfaction
motives. These motives for participationas well as for AES non-
participationseemed to differ from study area to study area, from
Northern to Central and Southern Europe, reecting the different
cultures, policies, economies and geographies. Despite these vari-
ations, some general lessons for future AES design can be derived.
Firstly, a continued effort to promote and spread knowledge about
AES seems fundamental to the increase of scheme participation.
 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812
Different types of farmers use different sources of information dis-
semination about AES opportunities, which, therefore, should be
adapted to the characteristics of these different landowner groups.
Hobby farmers, in particular, may not use the same channels of
communication and information as full-time farmers; it, therefore,
takes different dissemination strategies to inuence their level of
AES participation. Secondly, the amount of bureaucracy involved
in AES participation was found to be prohibitive for some cate-
gories of farmers. An effort to ease the administrative burden seems
important, in order to make future scheme participation attractive
to more European farmers.
Acknowledgments
The research has received funding from the European Union
Seventh Framework Programme (FP7/2007-2013) for the VOLANTE
project under grant agreement no. 460 265104. The authors would
like to thank all the farmers who took part in our survey and the
anonymous reviewers for their helpful comments.
References
Barreiro-Hurle, J., Espinosa-Goded, M., Dupraz, P., 2010. Does intensity of change
matter? Factors affecting adoption of agri-environmental schemes in Spain. J.
Environ. Plann. Manage. 53 (7), 891905.
Berger, G., Kaechele, H., Pfeffer, H., 2006. The greening of the European common
agricultural policy by linking the European-wide obligation of set-aside with
voluntary agri-environmental measures on a regional scale. Environ. Sci. Policy
9, 509524.
Brouwer, F., Lowe, P., 1998. CAP and the Rural Environment in Transition: A
Panorama of National Perspectives. Wageningen Pers, Wageningen.
Buller, H., Wilson, G.A., Hll, A., 2000. Introduction: The Emergence of Regulation
2078. In Agri-Environmental Policy in the European Union. In: Buller Henry,
Wilson Geoff, A., Hll Andreas (Eds.). Ashgate Publishing, Aldershot, England,
pp. 18.
Burton, R.J.F., Kuczera, C., Schwarz, G., 2008. Exploring farmers cultural resistance
to voluntary agri-environmental schemes. Eur. Soc. Rural Sociol. 48 (1), 1637.
Burton, R., 2004. Seeing through the good farmers eyes: towards developing an
understanding of the social symbolic value of productivist behaviour. Sociol.
Rural. 44 (2), 195215.
Burton, J.F., 2006. An alternative to farmer age as an indicator of life-cycle state:
the case for a farm family age index. J. Rural Stud. 22, 485492.
Busck, A.G., Kristensen, S.B.P., 2014. From agriculture to naturea study of drivers
of land use change in a peri-urban landscape. Geogr. Tidsskr. Dan. J. Geogr. 114
(1), 4158.
Capitanio, F., Adinol, F., Malorgio, G., 2011. What explains farmers participation
in rural development policy in Italian southern region? An empirical analysis.
New Medit 10 (4), 1924.
Damianos, D., Giannakopoulos, N., 2002. Farmers participation in
agri-environmental schemes in Greece. Br. Food J. 104 (3/4/5), 261273.
Defrancesco, E., Gatto, P., Runge, F., Trestini, S., 2008. Factors affecting farmers
participation in agri-environmental measures: a northern Italian perspective. J.
Agric. Econ. 59 (1), 114131.
Dupraz, P., Vermersch, D., Henry De Frahan, B., Delvaux, L., 2003. The
environmental supply of farm households. Environ. Resour. Econ. 25, 171189.
Espinosa-Goded, M., Barreiro-Hurle, J., Ruto, E., 2010. What do farmers want from
agri-environmental scheme design? A choice experiment approach. J. Agric.
Econ. 61 (2), 259273.
European Commission, 2015. Rural development 2014-2020. Agriculture and Rural
development.
European Union, 2011. Rural development in the European Union. Statistical and
economic information Report 2011. Directorate-General for Agriculture and
Rural Development.
GRAIN, Report Hungry for Land, May 2014,.
Gasson, R., 1986. Part-time farming: its place in the structure of agriculture. In:
Cox, G., Lowe, P. (Eds.), Agriculture: People and Policies. Allen & Unwin,
London, pp. 7792.
Greiner, R., Gregg, D., 2011. Farmers intrinsic motivations, barriers to the adoption
of conservation practices and effectiveness of policy instruments: empirical
evidence from northern Australia. Land Use Policy 2 (1), 257265.
Ingram, J., Gaskell, P., Mills, J., Short, C., 2013. Incorporating agri-environment
schemes into farm development pathways: a temporal analysis of farmer
motivations. Land Use Policy 31, 267279.
Kleijn, D., Sutherland, W.J., 2003. How effective are European agri-environment
schemes in conserving and promoting biodiversity? J. Appl. Ecol. 40, 947969.
Kleijn, D., Baquero, R.A., Clough, Y., Diaz, M., De Esteban, J., Fernandez, F., Gabriel,
D., Herzog, F., Holzschuh, A., Johl, R., Knop, E., Kruess, A., Marshall, E.J.P., Steffan
Dewenter, I., Tscharntke, T., Verhulst, J., West, T.M., Yela, J.L., 2006. Mixed
811
biodiversity benets of agri-environment schemes in ve European countries.
Ecol. Lett. 9, 243254.
Kristensen, S.B.P., Busck, A.G., van der Sluis, T., Gaube, V., 2016. Patterns and
drivers of land use change in selected European rural landscapes. Land Use
Policy 57, 786799.
Kristensen, S.P., 2003. Multivariate analysis of landscape changes and farm
characteristics in a study area in central Jutland, Denmark. Ecol. Model. 168,
303318.
Labarthe, P., Laurent, C., 2013. Privatization of agricultural extension services in
the EU: towards a lack of adequate knowledge for small-scale farms? Food
Policy 38, 240252.
Latacz-Lohmann, U., Hodge, I., 2003. European agri-environmental policy for the
21st century. Aust. J. Agric. Resour. Econ. 47 (1), 123139.
MacCallum, R.C., Widaman, K.F., Zhang, S., Hong, S., 1999. Sample size in factor
analysis. Psychol. Methods 4, 8499.
Mathijs, E., 2003. Social capital and farmers willingness to adopt countryside
stewardship schemes. Outlook Agric. 32 (1), 1316.
Micha, E., Areal, F.J., Tranter, R.B., Baileya, A.P., 2015. Uptake of agri-environmental
schemes in the less-favoured areas of Greece: the role of corruption and
farmers responses to the nancial crisis. Land Use Policy 48, 144157.
Mills, J., Gaskell, P., Reed, M., Short, C., Ingram, J., Boatman, N., Conyers, S., Carey, P.,
Winter, M., Lobley, M., 2013. Farmer attitudes and evaluation of outcomes to
on-farm environmental management. In: Report to Department for
Environment, Food and Rural Affairs (Defra). CCRI, Gloucester.
Moss, J., 1994. A baseline assessment for a new ESAthe case of the Mourne
Mountains and Slieve Croob. In: Whitby, M. (Ed.), Incentives for Countryside
Management: The Case of Environmentally Sensitive Areas. CAB International,
Wallingford, pp. 153178.
OCarroll, L., 1994. Competition with other environmental designations on a
lowland heaththe case of Breckland. In: Whitby, M. (Ed.), Incentives for
Countryside Management: The Case of Environmentally Sensitive Areas. CAB,
International, Wallingford, pp. 6180.
Pedroli, B., Rounsevell, M., Metzger, M., Paterson, J., and the VOLANTE consortium,
2015. The Volante Roadmap Towards Sustainable Land Resource Management
in Europe.
Peerlings, J., Polman, N., 2009. Farm choice between agri-environmental contracts
in the European Union. J. Environ. Plann. Manage. 52 (5), 593612.
Peduzzi, P., Concato, J., Kemper, E., Holford, T.R., Feinstein, A.R., 1996. A simulation
study of the number of events per variable in logistic regression analysis. J.
Clin. Epidemiol. 49, 13721379.
Polman, N.B.P., Slangen, L.H.G., 2008. Institutional design of agri-environmental
contracts in the European Union: the role of trust and social capital. NJAS
Wagening. J. Life Sci. 55 (4), 413430.
Potter, C., Lobley, M., 1996. The farm family life cycle, succession paths and
environmental change in Britains countryside. J. Agric. Econ. 47, 172190.
Primdahl, J., Peco, B., Schramek, J., Andersen, E., Onate, J.J., 2003. Environmental
effects of agri-environmental schemes in Western Europe. J. Environ. Manage.
67, 129138.
Primdahl, J., Busck, A.G., Kristensen, L.S., 2004. Landscape management decisions
and public-policy interventions. The new dimensions of the European
landscape. In: Jongman, R., Bogers, R. (Eds.), The New Dimensions of European
Landscape. Springer, Wageningen, pp. 103122.
Primdahl, J., 1999. Agricultural landscapes as places of production and for living in
owners versus producers decision making and the implications for planning.
Landsc. Urban Plann. 46 (1/3), 143150.
Ruto, E., Garrod, G., 2009. Investigating farmers preferences for the design of
agri-environmental schemes: a choice experiment approach. J. Environ. Plann.
Manage. 52 (5), 631647.
Sattler, C., Nagel, U.J., 2010. Factors affecting farmers acceptance of conservation
measures a case study from north-eastern Germany. Land Use Policy 27 (1),
7077.
Siebert, R., Toogood, M., Knierim, A., 2006. Factors affecting European farmers
participation in biodiversity policies. Sociol. Rural. 46 (4), 318340.
Skerratt, S.J., 1994. Itemized payment systems within a schemethe case of
Breadalbane. In: Whitby, M. (Ed.), Incentives for Countryside Management:
The Case of Environmentally Sensitive Areas. CAB International, Wallingford,
pp. 105134.
Van der Sluis, T., Kristensen, S.B.P., Frederiksen, P., Cosor, G., Vadineanu, A., Pavlis,
E., Terkenli, T.S., Gaube, V., Vesterager, J.P., 2013. Landscape Change Processes
in Case Study Areas (WP2). ALTERRA, Wageningen.
Vesterager, J.P., Frederiksen, P., Busck, A.G., Terkenli, T., Pavlis, V., Van der Sluis, T.,
Geamana, N., Bucur, M., Vadineanu, A., Kristensen, S.B.P., Gaube, V., 2016.
Dynamics in national Agri-environmental policy under changing EU policy
priorities, past experiences with national Agri-environmental schemes and
strategies for future EU Agri-environmental policy adaptation. Land use policy
57, 764776.
Viaggi, D., Signorotti, C., Marconi, V., Raggi, M., 2015. Do agri-environmental
schemes contribute to high nature valuefarmland? A case study in
Emilia-Romagna (Italy). Ecol. Indic., http://dx.doi.org/10.1016/j.ecolind.2015.
01.017, Available online 4 March 2015.
Whitby, M., 1996. The European Environment and CAP Reform: Policies and
Prospects for Conservation. CAB International, Wallingford, Oxon.
Wilson, G.A., Hart, K., 2000. Financial imperative or conservation concern? EU
farmers motivations for participation in voluntary agri-environmental
Schemes. Environ. Plann. A 32, 21612185.
812 E.S. Pavlis et al. / Land Use Policy 57 (2016) 800812

Wilson, G.A., 1997. Factors inuencing farmer participation in the environmentally
sensitive areas scheme. J. Environ. Manage. 50, 6793.
Wossink, G.A.A., van Wenum, J.H., 2003. Biodiversity conservation by farmers:
analysis of actual and contingent participation. Eur. Rev. Agric. Econ. 30,
461485.
Xavier, B.L.-B., Hubbard, C., Garrod, G., Tolo n-Becerra, A., 2015. What drives
farmers participation in EU agri-environmental schemes?: Results from a
qualitative meta-analysis. Environ. Sci. Policy 54, 19.