The use of socio-economic and environmental indicators in assessing the effectiveness of EU agri-environmental policy


Nov 9, 2013 (3 years and 5 months ago)



The use of socio
economic and environmental indicators in assessing
the effectiveness of EU agri
environmental policy

Geoff A. Wilson (*) and Henry Buller (**)

* Department of Geography, King’s College London, Strand, London, WC2R 2LS; e

** Countryside and Community Research Unit, Cheltenham and Gloucester College of HE, Francis
Close Hall, Cheltenham, GL50 4AZ; e


Since the mid
1980s, ‘indicators’ have been increasingly suggested as a possible
means to evaluate the effectiveness of various policy mechanisms. In this paper, we
shed some critical light on the use of indicators, by investigating in detail

advantages and disadvantages of using the indicator approach in the assessment of
environmental policy effectiveness in the European Union. We review the
theoretical and conceptual bases of the indicator approach, highlighting the
complexity of i
dentifying appropriate indicators and the difficulties in identifying
means to ‘measure’ them. We analyse four levels of indicators (uptake, output, actor
related and procedural indicators), and argue that while indicators continue to have
a central role t
o play in the evaluation of European agri
environmental policy, that
role is essentially limited and the notion of what constitutes an indicator needs to be
thoroughly re

Key words

Policy indicators; agri
environmental indicators; policy succes
s; policy effectiveness;
Regulation 2078


1. Introduction

Since the mid
1980s, agri
environmental policies (AEPs) have been implemented in
all European Union (EU) countries (Whitby, 1996; Potter, 1998; Buller
et al.
, 2000).
Most of these have been put in

place as a response to EU regulations (e.g. 797/85;
2078/92; Agenda 2000), and many share the common aims of extensification of
agriculture, farm income support, and environmental protection. Yet, although
hundreds of agri
environmental schemes are now op
erating in various EU localities,
costing European taxpayers about EURO 4 billion/year (with predicted budgets of
about EURO 8 billion/year by 2006), information on the ‘effectiveness’ of these
schemes has remained relatively sparse. This has been partly a

result of the fact that
most EU policies were put in place without adequate and compulsory monitoring
frameworks attached to their implementation (CEC, 1997; Potter, 1998), but it also
reflects the diversity of rationales underlying AEP as a whole leading

to, in many
cases, a notable lack of incentive to measure scheme success (Brouwer, 1999; Lowe
et al.
, 1999; Buller
et al.
, 2000; Wilson and Wilson, 2001).

During the 1990s, increasing concern was voiced from different quarters about the
lack of evaluatio
n of AEP success, particularly in view of the Agenda 2000
propositions which initially sought to substantially increase the agri
part of the Common Agricultural Policy (CAP), and in view of the fact that
agricultural support policies have mul
tiple and sometimes contradictory
environmental effects (e.g. BirdLife International, 1994; CEC, 1997; Baldock, 1997;
Potter, 1998; Parris, 1999). Brouwer and Crabtree (1999b, 2), for example, argued
that “methods are … needed to indicate the effectiveness

of policy response through
the agri
environment programmes … Indicators therefore are required to judge
whether a reduction in production
linked agricultural support would be beneficial or
harmful to the environment”, while Parris (1999, 27) has reminded
us that “for most
OECD countries the systematic collection of basic agri
environmental data and
measurement of agri
environmental indicators is only beginning”.

Although the European Commission put in place mechanisms such as Regulation
746/96 that laid
down the rules for the monitoring of schemes implemented under
environment Regulation 2078, there has been considerable debate surrounding
the means with which to ‘measure’ AEP success with regard to both socio
and ecological parameters (Brou
wer, 1995). While most monitoring has included
limited ecological assessments on the relative recovery/re
establishment of lost or
degraded wildlife habitats, discussions in the agriculture and environment sector in
the mid 1990s were also strongly influen
ced by general debates on how to measure
the effects of wider environmental policy strategies aiming at ensuring sustainable
development through the use of ‘environmental indicators’ (see in particular OECD,
1993, 1997, 1998; Moldan and Billharz, 1994; Ham
et al.
, 1995; Moxey
et al.
1998; Parris, 1999). Hence, the notion of ‘agri
environmental indicators’ to measure
the effectiveness of AEP and as a tool for regulation increasingly gained ground
among national and transnational policy

The aim
of this paper is to discuss the use of such indicators in assessing the
effectiveness of EU AEP. The paper is divided into two parts. First, we will discuss
more broadly the notion of ‘indicators’ and consider conceptual and theoretical
debates surrounding

the utility, accuracy and general applicability of indicators.


Second, we will discuss empirical evidence to discuss problems and opportunities of
the indicator approach in assessing the effectiveness of AEPs. Following both Moxey
et al.
’s (1998) and Lowe

et al.
’s (1999) calls for improved discussions of indicators
that take human agency and institutional structures into account, we will focus
specifically on indicators aimed at monitoring the response of the agricultural sector
to AEP policy targets, with

particular reference to how farmers have adjusted farming,
farm management practices and environmental thinking on the basis of agri
environmental scheme participation. Based on our own experience with attempting to
use indicators to ‘measure’ policy effe
ctiveness, we will cast a relatively critical eye
on the utility and applicability of the indicator approach, and we will conclude the
paper by arguing for and against the use of indicators as a future tool for AEP

2. Agri
environmental indic
ators: some insights into the conceptual debates

Indicators are generally described as “a parameter [a property that is measured and
observed], or a value derived from parameters, which points to, provides information
about, describes the state of a phen
omenon/environment/area, with significance
extending beyond that directly associated with a parameter value” (OECD, 1998,
107). Environmental indicators are, therefore, a vehicle for summarising, or otherwise
simplifying, and communicating information abou
t something that is of importance to
environmental decision
makers (Moxey
et al.
, 1998). They are a conceptual attempt at
identifying optimal amount of information under conditions of high uncertainty (i.e.
incomplete information settings on the state of t
he environment or human
environmental behaviour, or relating to the policy ‘implementation gap’; Barrett and
Fudge, 1981; Hogwood and Gunn, 1984; Winter, 1990; Adriaanse, 1993; Romstad,
1999). Indicators, thus, reduce the number of measurements and paramet
ers that
normally would be required to give an ‘exact’ presentation of the situation, and they
simplify the communication process through which results of measurements are
provided to users. They should (in theory at least) provide managerially significant

information about patterns or changes in the state of the environment or changes in
human activities that affect the environment (Dabbert
et al.
, 1999), and should enable
better targeting of information and action, particularly with respect to AEP (e.g.
ilson, 1997a; Potter, 1998; Buller
et al.
, 2000). Moxey
et al.

(1998) emphasise that
the increasing need for indicators also reflects broader shifts in society, with
increasing pressures to measure and assess the performance of public authorities by
al references in relation to actual policy outcomes, i.e. improved inter
organisational control within increasingly complex systems of governance.

Before providing specific examples of the utility of the indicator approach in the
assessment of AEP in the

EU in the second half of this paper, this section will discuss
four conceptual issues that have to be considered when using the indicator approach
in environmental policy evaluation. First, we need to acknowledge that indicators are
based on the belief of

quantifiable environmental impacts and human environmental
behaviour. Second, the direct link between indicators and policy design and
implementation needs to be addressed. Third, conceptual debates surrounding the
choice of appropriate indicators need to

be considered, and fourth, the issue whether it
is possible to develop a hierarchy of indicators in the effort to pick the ‘right’
indicators for policy assessment needs to be discussed.


2.1 A quantifiable environment?

Indicators are based on quantita
tive notions that assume that policy success can be
‘measured’ (Rose, 1991). As a result, indicators have been widely used in disciplines
such as economics that rely largely on the quantification of human action (e.g.
environmental accounting; physical nat
ural resource accounts) or in natural science
disciplines such as ecology or climatology that largely assume that environmental
effects can be modelled and quantified (see McKenzie
et al.
, 1992; OECD 1998). As
a result, much of the theoretical work on indi
cators has come from these disciplines,
including contributions by economists (e.g. Norgaard, 1990; Farrow and
Krautkraemer, 1991; Asheim, 1994) focusing on issues of resource scarcity and
assuming rational and measurable human decision
making processes, o
r by natural
scientists preoccupied with developing suitable physical environmental indicators to
measure changes in the state of the environment (e.g. McKenzie
et al.
, 1992; Dabbert
et al.
, 1999).

Conceptual approaches for the development of environmenta
l indicators by the
Organisation for Economic Cooperation and Development
(OECD) highlight this
preoccupation with ‘the measurable’. The OECD (1998) sees indicators as a tool for
making and for assessing countries’ environmental performance to be
for quantitative environmental performance reviews, in the EU often using data
compiled by EUROSTAT questionnaires on the state of the environment and on
pollution abatement and control expenditure provided by different member states. The
OECD adopts
the (arguably simplistic) view that the main environmental concerns
related to agriculture can be reduced to nitrogen and phosphorus run
off from
excessive commercial fertiliser and pesticide use and intensive livestock farming
(OECD, 1993, 1997, 1998; see

also MAFF, 1998). As a result, the OECD places great
emphasis on quantifiable indicators such as changes in the intensity of use of N and

fertilisers, livestock densities, or the intensity of pesticide use (see Technical
Annex ‘Agriculture’ in OECD,
1998; Brouwer, 1999).

There are, however, evident problems related to the notion of quantifiable
environmental change. Indicator approaches focus on
entities that can be
expressed in numbers where data is more readily available. As we will dis
cuss below,
this may lead to the neglect of more complex indicators where immediate quantitative
data may not be available, such as ‘loss of scenic beauty of agricultural landscapes’,
‘changes to nature and extent of wildlife habitats’ or, arguably most im
‘changing environmental management practices and attitudes of farmers’. In this
context, the Council for the Protection of Rural England has recently emphasised that
some environmental characteristics and human environmental responses are
ntly better suited than others to quantification (CPRE, 1995). This leads Lowe
et al.

(1999, 271) to question the OECD approach by arguing that “the OECD
indicator framework perpetuates a policy outlook which, by abstracting farming from
its social an
d environmental context, is part of the fundamental problem” (see also
IEEP, 1998).

Indicators can, therefore, never be a perfect quantification of an agri
problem. Although the belief in quantifiable environmental change may give the


rance of objectivity, it renders invisible the social choices their collection and
collation entail, leading Lowe
et al.

(1999, 264) to further argue that “the focus on
environmental outcomes, which is reinforced by the preoccupation with quantitative
cators, may lead to a disregard of the lessons being learned from regulatory policy
by target groups and the need for wider institutional reforms”. In the agri
environmental context discussed below, this means that measurable indicators (e.g.
changes in fe
rtiliser and pesticide use) and quantifiable farm uptake data have been
overused as possible means to measure the ‘success’ of agri
environmental schemes,
while intangible socio
cultural and qualitative indicators

have been neglected (for the
UK see, for
example, MAFF, 1993, 1994, 1997; National Audit Office, 1997; or
House of Commons Agriculture Committee, 1997).

2.2 Indicators and policy

Indicators have to relate directly to stated environmental policy goals and objectives
(OECD, 1998; Jesinghaus, 1
999). Indicators can, therefore, only be conceptualised in
the context of policy design and implementation. As such, they have become an
integral assessment tool of EU agricultural policy reform (Brouwer and Crabtree,
1999b), and the main objective of indi
cators is to assist policy
makers in their
evaluation of policies and, in particular, to gain an understanding of the
‘implementation gap’ between policy formulation ‘at the top’ and policy acceptance
at the grassroots level that often hinders successful p
olicy implementation (Barrett
and Fudge, 1981; Hogwood and Gunn, 1984; Winter, 1990). This is important insofar
as so
called ‘policy relevance’ has become a key criteria for selecting environmental
indicators by various organisations (e.g. MAFF in the UK;
OECD; United Nations).
However, this implies that policy goals and objectives have to be clear. As we will
discuss in detail below, this is rarely the case with respect to many environmental
policies such as Agri
environment Regulation 2078/92/EEC. With re
gard to research
on this specific policy, indicators have, therefore, been most commonly linked to
explicit quantitative objectives of schemes implemented under the Regulation (e.g.
uptake targets; cost
benefit; environmental thresholds; set
aside targets)
, but have so
far been insufficiently linked to less clearly prescribed qualitative policy objectives
such as the influence of agri
environmental schemes on various actors’ action and
thought (e.g. ‘ways of using agricultural land which are compatible with

and improvement of the environment’ [Article 1b of Agri
environment Regulation
2078] or ‘education and training of farmers in types of farming compatible with the
requirements of environmental protection and upkeep of the countryside’ [Article

If we accept the notion that indicators and policy are a continuously reinforcing
feedback system (OECD, 1997, 1998), three policy
related problems emerge. First,
the indicator approach may
the design and implementation of policies that
re easily quantified and may, therefore, increase the often existing implementation
gap. In other words, more easily quantifiable physical and economic environmental
targets may be over
emphasised to the neglect of less easily prescribable socio
cultural p
olicy goals. Second, any published indicator exerts a pressure on policy


For example, alternative behaviourally
grounded conceptions of scheme success have emphasised
changes in landholders’ environmental actions as equally important indicators to measure policy
ccess’ (e.g. Morris and Potter, 1995; Wilson, 1996, 1997a; Lobley and Potter, 1998).


makers to improve the figure but not necessarily to improve the underlying problem
(Jesinghaus, 1999). Regulation theorists such as Campbell (1987) or Romstad (1999),
therefore, argue

that indicator relevance should go beyond simple policy relevance to
include broader socio
cultural needs and expectations. Third, while indicators might
trends, they do not necessarily
them (Lowe
et al.
, 1999). Measuring
AEP ‘effectiveness
’ may just become a simple description of change without
addressing the issue of causality. If it were possible to estimate the possible policy
situation, then the comparison between observed outcomes and the policy
situation may offer the possibil
ity to quantify policy effects (Lehmann and Schmid,

2.3 Criteria for choosing appropriate indicators: the ‘5 Rs’

That indicators are based on notions of quantifiable environmental change and that
they have to relate directly to overtly stated or
hidden policy goals and objectives has
important implications for the selection of ‘appropriate’ indicators. Indicators are
expected to be consistent, reliable, unambiguous, transparent, and they should have
predictive capacities and analytical soundness (
OECD, 1998; Moxey
et al.
, 1998;
Romstad, 1999; Peco
et al.
, 1999). As the discussion above illustrates, they should
particularly be
policy relevant
. These different demands and
expectations on indicators can be summarised into five categorie
s (the ‘5 Rs’):
indicators should be
epresentative and

As indicators are simplifications of often complex processes, they need to remain

with regard to what they are aiming to ‘measure’ (Hammond

, 1995;
Romstad, 1999). While organisations such as the OECD argue that this reliability
criteria is crucial for indicator measurability (in other words, that indicators should
lend themselves to being linked to economic modelling, forecasting and info
systems), such measurability still varies greatly among indicators and across different
geographical areas (Brouwer and Crabtree, 1999a). In addition, the non
straightforward links between agriculture and the physical environment, the resulting
causality of environmental parameters, and that linkages between agriculture
and environment are far more complex than many other economic activities, suggest
that the reliability criteria for indicators is often the most difficult to achieve.

A commo
n claim is that indicators have to be
, especially in their requirement
for policy performance analysis. Policy
makers’ preference for certain indicators can
be explained as a trade
off between quality and cost. As a result, indicator relevance
ies from one country to another (or even from one region to another), and often
they have to be interpreted in the appropriate context, taking into account different
ecological, geographical, social, economic and institutional features. With regard to
ical environmental indicators, for example, the relevance of an indicator may
depend on the site
specific environmental state, and the optimal indicator set will
most likely differ from one location to another. In particular, indicator relevance may
vary f
rom actor to actor, as different actors in the AEP implementation process seek
different goals and objectives with regard to scheme outcomes (e.g. CPRE, 1995;
Department of the Environment, 1996; OECD, 1997). Hence, while environmental
NGOs, for example, m
ay seek to promote environmental sustainability through AEP
(e.g. FoE, 1992; Environmental Challenge Group, 1995; World Resources Institute,


1995), agricultural economists generally emphasise the importance of financial
benefits of schemes with a resulting

focus on cost
benefit indicators and ‘value for
money’ (e.g. Garrod
et al.
, 1994; Garrod and Willis, 1995).

A further key requirement for indicators is that they are
. An indicator
that can not be used by other researchers is meaningless. Ide
ally, they should be
theoretically well founded in technical and scientific terms, they ought to be based on
international standards and international consensus about their validity, they should be
adequately documented and of known quality (with an emphas
is on the national and
international level rather than the local level; see below), be updated regularly in
accordance with ‘reliable’ procedures, and should be based on consistent (and
reproducible) methodologies. Indicators should also be

in that they
ought to provide the basis for inter
regional and international comparison. They
should, therefore, have a threshold or reference value against which to compare
changes in environmental state or human environmental activity, and such threshol
should ideally be defined in relation to legal, scientific or policy
related norms (Parris,
1999). Finally, indicators should be
. In other words, they ought to be
simple and easy to interpret and able to show trends over time, and they should

readily available at reasonable cost
benefit ratio (Romstad, 1999).

It is obvious that no researcher will ever achieve to fulfil the ideal ‘5 R’ criteria,
prompting Peco
et al.

(1999, 151) to argue that “it is impossible to construct a general
) system of agri
environmental indicators for the purpose of evaluating the
impact of the CAP on the environment”. As findings from our own research will show
(see Tables 1 and 2 below), selecting appropriate indicators, therefore, always has to
be a trade
off between what is feasible, affordable and manageable within a research
environment that is most usually restricted by limited time, money and personnel.

2.4 A hierarchy of indicators?

Selection of appropriate indicators is further hampered by the f
act that the hierarchical
importance of indicators varies depending on the spatial focus of investigation, the
expected use of an indicator (or groups of indicators) in answering specific objectives,
and the policy context. Thus, some indicators are better

able to reveal implementation
deficits than others (cf. Winter, 1990). Typological, spatial and conceptual indicator
hierarchies can be identified, that all need to be taken into account when interpreting
research results.

Typological indicator hierarchi
refer to the pressure
response (PSR) model
developed by the OECD (1993, 1997, 1998) which corresponds well to the profile of
EU member states’ statistical services (see also United Nations, 1996; Eurostat,
Pressure indicators
are related to

human activities, processes and patterns (e.g.
agricultural activities), and their importance is highlighted by current efforts to
establish a harmonised European System of Environmental Pressure Indices (ESEPI)
(Jesinghaus, 1999).
State indicators
, meanw

give information on the state of the
environment and natural resources arising from pressure indicators, while
provide information on policy responses and changes in awareness and
behaviour of communities and individuals. Human ac
tivities exert environmental
pressures that lead to changes in the state of the environment, and resulting response


by society in form of policies and/or changed environmental management practices,
which, in turn, may change human pressures on the environm
ent (Brouwer and van
Berkum, 1996; Lowe
et al.
, 1999). Although the OECD (1998) has acknowledged that
there can be no universal set of AEP indicators, they nonetheless suggest to focus on
13 indicators that are believed to be sufficient core indications on

changes in the
environmental state of agricultural landscapes and changes in human response
mechanisms. Thus, OECD
pressure indicators
include information on nutrients,
pesticides, water use and land use changes,
state indicators
focus on soil quality,
ter quality, greenhouse gases, biodiversity, wildlife habitats, and landscape change,
response indicators
stress farm management practices, farm financial resources
and socio
cultural aspects (including, for example, education, training, or issues of

rural/urban migration).

The PSR typological model has been widely criticised for being too simplistic in its
assumptions of environmental causality, i.e. that the three types of indicators
response) form discreet conceptual units that can

be easily separated
from each other (Moxey
et al.
, 1998). The problem of unequivocal causality relates to
the fact that, often, the multitude of environmental influences are only partly related
to processes measured through indicators (policy
on and polic
off scenarios and
trajectories of farmers’ decision
making processes in AEP being an obvious case
point). The model has been further criticised for its positivist theoretical assumptions
that are based on the belief that both physical indicators and i
ndicators of human
environmental action and thought can be quantified (Lowe
et al.
, 1999; see also
Section 2.1 above). Particular problems are seen with state and response indicators.
For state indicators, no framework has been established yet for how to o
problems of lack of linearity and immediacy (Oñate
et al.
, 2000), i.e. that the problem
may lag behind the source of disturbance (N leaching into soil and groundwater being
an obvious example). Although state indicators may be more closely linked t
environmental policy interests, they are more costly to produce and often suffer from
complex demands on the nature of measurement. Yet, the most severe shortfalls are
seen with regard to response indicators (Brouwer and Crabtree, 1999a). Parris (1999,
0) for example, concedes that “although the importance of socio
cultural issues in
the analysis of agriculture and the environment … is generally accepted, no precise
definition of the policy issues nor the relevant indicators have yet been established”.

patial indicator hierarchies

highlight the problem of the need for multiple spatial
scales of indicator frameworks. In this context, Moxey
et al.

(1998) ask how it is
possible in indicator research to preserve the fine resolution detail required for
ng local level actions, while avoiding information overload at the broader global
level? The OECD suggests to use international indicators in combination with specific
national indicators (i.e. moving away from the ‘5 R’ criteria mentioned above). This is
supported by Moldan and Billharz (1994) who argue that different indicators of policy
performance are required at different hierarchical levels, and that indicators useful for
international comparison will inevitably differ from those for monitoring region
al or
local environmental change. While recent trends suggest that the focus is increasingly
on using national indicators for use in international work (e.g. ACANZ, 1993; IFEN,
1997; OECD, 1998)

a problem that we will also encounter in our analysis below

recent work has also insisted that any list of indicators must be peculiar to each
system, and that the value of indicators must be calculated at the county or farm scale,
rather than at the national or international level (e.g. Peco
et al.
, 1999). The

debate on


internationalisation or localisation of the indicator approach is, therefore, still on

A third approach, and one that we adopted in the research programme from which this
paper is drawn (Schramek
et al.
, 1999), is founded on a


that proposes four levels of increasing complexity (Table 1). Our analysis
of agri
environmental indicators presented in the following section, identifies the
strengths, weaknesses and potentials of these different levels in providing
assessments of the ‘success’ of AEP. However, before doing this, we need to consider
the broader presuppositions that underlie the use of indicators in evaluating,
specifically, AEPs.

Table 1:

Parameters for the evaluation of agri
environmental s


Level 1.

Uptake and outputs


numbers of contracts/farms


areas under contract



Level 2.

Outcomes and Effects


effects upon farm management practices


effects upon agricultural production


effects upon the environment


effects upon farm incomes


effects upon agricultural labour

Level 3



impact upon farmer attitudes and knowledge


impact upon the social processes of



impact upon the activities of associated actors

Level 4



role of management institutions


efficiency of the policy
making process


upon regulatory procedures

(Source: adapted from Buller, 2000)

3. How effective at doing what? The strengths and limitations of agri
environmental policy indicators

A recently completed research project funded by the DGVI of the European
Commission so
ught to explore appropriate methodologies for the implementation and
effectiveness of AEP in Europe (Schramek
et al.
, 1999). In this part of the paper, we
consider the results of this research and, more specifically, what it tells us about the
use and misu
se of agri
environmental indicators in informing and guiding policy
making. As we have stated above, the identification of a policy’s objectives is a
fundamental first step if effective evaluation procedures are to be put into place and
useful results are
to be obtained. This presupposes, first, that the objectives are clearly
stated, second, that they are coherent and, third, that a relatively direct relationship
exists between the recognition of a given problem, the mechanisms evoked to deal
with it, and
its eventual resolution. One of the problems of constructing a suitable


evaluation strategy for AEP is that these presuppositions are not always wholly born

Article 1 of Regulation 2078 lays down the three goals of EU AEP: to accompany the
changes to

be introduced under the market organisation rules; to contribute to the
achievement of the Community’s policy objectives regarding agriculture and the
environment; and to contribute to providing an appropriate income for farmers (CEC,
1992). These reflect

different concerns. As an accompanying measure to CAP reform,
environmental aid schemes comply with the basic aim of that reform, namely to
reduce agricultural over
production within the EU and thereby lessen the overall costs
of agricultural support
. The Agri
environmental Regulation also reflects growing
concern over the environmental consequences of intensification, notably with respect
to water pollution, biodiversity loss and landscape change. A third goal has been to
support and maintain extensi
ve farming practices, not only against intensification but
also, particularly in upland and southern regions, against agricultural decline and the
environmental and social consequences of under
production. Here, then, are three
very different objectives th
at reflect macro
economic, environmental and socio
economic concerns that seek de
intensification in certain cases and maintenance, even
intensification, in others, that ultimately relate to very different rural spaces and
agricultural systems.

Thus, any

evaluation methodology is necessarily going to be very complex because
not only are the implementation procedures very different in the various EU member
states, and the types of farm system targeted and the range of landscape of
environmental problems en
countered highly variable, but also the very objectives of
AEP are, themselves, multiple and by no means restricted to a single environmental
policy agenda. AEP cannot be seen simply as a strategy for achieving environmental
objectives within agriculture,
particularly as such objectives might arguably be more
effectively secured by alternative strategies. Indeed, as a number of commentators
have suggested (see, for example, Buller
et al.
, 2000), EU AEP has been singularly
unsuccessful in addressing the cent
ral integration issue of farm pollution. Its
‘success’, if it exists, lies elsewhere.

Despite the recent demand for effective instruments capable of evaluating AEP, two
operational problems to the development of a valid indicator framework for agri
nmental measures exist. The first of these has been the consistent lack of base
line data on environmental ‘states’ prior to the start of contractual agri
agreements (i.e. referring to the issue of indicator
mentioned in Section
.3). As most EU Member states only began to implement agri
measures in the 1990s, insufficient reflection was given to the eventual evaluation of
the policy some years further on (Buller, 2000). Second, although it is becoming
increasingly co
mmonplace to evaluate the AEP implementation process in terms of
indicators, the vast bulk of these assume that successful implementation will engender
positive changes in environmental conditions. Yet, any analysis of the goals as well as
the implementati
on of EU
led AEP reveals that, for the most part, the resultant
schemes do not, in fact, seek
, be they in agricultural practice or in
environmental quality. Rather, they seek to
existing practices considered
both to be environmentally frie
ndly and necessary for the maintenance of sustainable
rural/agricultural communities by encouraging farmers
to make changes to
existing management techniques and thereby
to increase environmental pressures.


Given that so much of the indicator ratio
nale developed by the OECD (1993; 1997;
1998) and the European Environment Agency (e.g. EEA, 1998) (amongst others) is
guided by the desire to record the effects of agricultural de
intensification upon
environmental improvement, there is a risk here that t
he long
term and, ultimately
more profound, effects of environmental integration are being missed. The evaluation
of AEP thus needs to take into account alternative parameters of scheme success.

3.1 Output and outcome indicators

Referring back to Table
1 (above), we propose a hierarchy of indicator levels that take
us from ‘outputs’ through ‘outcomes’ and effects to a set of less tangible components
of policy ‘success’. The first of these levels, indicators of uptake, also known as
‘output indicators’, a
re commonly derived from the contracted area, numbers of
contractants and proportions of budgets spent and/or allocated (and combinations of
these), because of the relative simplicity and low cost of their construction as the only
really universal evaluati
on criteria across Europe (see also Section 2.3 above). This
has largely been the methodology employed by the Commission in their own initial
evaluation of the application of Regulation 2078/92 (CEC, 1997). As the Commission
itself claims in a later docume
nt (CEC, 1998, 123), “the most relevant indicator of
application is the area to which measures apply”. As a means of evaluating the
policy’s achievements, uptake figures provide a reassuringly high ‘success’ rate with
some 22.6 million hectares of European

farmland being placed within agri
environmental schemes and well over a million European farmers engaged, to greater
and lesser degrees, in the drive towards reconciling the needs of agricultural
production with those of environmental sustainability.

ever, such an approach has a number of limitations in a cross
national context.
Output indicators based upon uptake rarely account for participant
, usually
differentially defined in terms of farm, types of farm practice, specific geographical
ocations or budgetary envelopes (see, for example, Wilson, 1997b). Further, broad
assessments, as presented by the European Commission’s own evaluation study
(CEC, 1997; 1998), do not account for different

or levels of participation
within indi
vidual schemes. Agri
environmental aid schemes currently operating are
often comprised of a number of ‘tiers’, with higher tiers often requiring, in exchange
for higher premiums, genuine changes in farm management practices, while the lower
tiers frequentl
y require only the maintenance of existing practices (see, for example,
MAFF, 1993; 1996; Wilson, 1997a; and Lobley and Potter, 1998, for the UK context).
In many countries, including the UK, it is the lower tiers that attract the greatest
number of partic
ipants. An additional problem concerns the criteria for defining
successful uptake. Are schemes with high levels of uptake more ‘successful’? Clearly,
they attract proportionally more farmers and they succeed in placing significant areas
under contract. As

such, their ‘impact’, in spatial as well as participation terms, is
undoubtedly greater. However, we cannot automatically assume that the generally low
up rates associated with, for example, arable schemes necessarily constitute
scheme failure. In as
sessing uptake, we need to be aware of the differential impact of
participant numbers. While the coherent protection of large areas of upland grasslands
might depend upon a high uptake rate, improving water quality, particularly around
sensitive areas, mig
ht be effectively achieved through the participation of only a small


number of highly targeted eligible farms. In such cases, the use of uptake indicators is
arguably of little value as a means of assessing scheme effectiveness.

A second level of indicato
rs, requiring more precise temporal and spatial reference
points, is concerned with the outcomes and impacts of implementation in terms of
farm management techniques, farm income, farm labour, and, indeed, the wider
environment. Given that agri
tal schemes specifically seek to engage
farmers in the employment of environmentally sound farm management practices,
such indicators are the common fare of agri
environmental scheme assessments
(CEC, 2000). Research carried out for DGVI by the current aut
hors and others
specifically sought to identify and test indicators of this kind (Schramek
et al
., 1999;
Wilson and Hart, 2000). A summary of the most common indicators of this group,
drawn from various studies, is presented in Table 2.

Table 2:

Common a
environmental indicators identified and tested

by recent research into agri
environmental scheme evaluation

Changes to stocking density

Changes to pasture land rotation

Quantity of pesticides used

Quantity of nutrients used

Change in farm usage of in

Proportion of land under grass

Yield changes for specified crops

Changes in use of farm land (arable to grassland….)

Changes in energy balance on farm

Changes in labour

Changes in gross profit per farm

Emissions to the atmosphere per farm

Number of li
vestock removed

Area of target land cover type

Dates and times of harvests and grass cutting

Hedgerow densities and changes to hedgerow densities

Changes to field margins

Levels of pesticide residue in produce and environment

Concentration on N in water (r
elation to soil depth and crop type)

Specific species of wild plant

Changes in vegetation density in grassland

Density of specific invertebrates

Bird nesting sites

Nesting pairs

Other identified animal species

Loss of soil under certain cultivation techniq

(Sources: CEC, 1998; Schramek
et al.
, 1999; Oñate et al., 2000)

The application of outcome and impact indicators to the evaluation of agri
environmental schemes has generally revealed positive environmental benefits arising
from contractualisation (Sc
et al.
, 1999). Oñate
et al.

(2000), whose work
derives from the same research project as our own, distinguish ‘improvement’ effects
and ‘protection’ effects and demonstrate, from a specific comparison of Spanish and
Danish agri
environmental schemes
, that clear improvement effects were found and
that protection effects “may well be an important long
term result” (p. 278). Other
evaluations draw broadly similar conclusions. For example, the exercise undertaken


by the French Ministère de l'Agriculture
(1998) reveals, first, a notable increase in
specific landscape management actions, such as hedgerow planting and water channel
clearing (particularly in those areas already noted for the quality of their agricultural
environments) and, second, some limite
d change in the use of pastures and chemical
inputs, though rarely in the major arable regions of the country.

However, despite their widespread use, such outcome and effect indicators contain a
number of significant limitations, over and above the opera
tional issues associated
with agri
environmental scheme evaluation identified above. First and foremost,
many outcome indicators reveal little more than compliance to contractual
obligations, highlighting problems associated with the policy ‘implementation

mentioned above (see also Peco
et al
., 1999; Buller
et al.
, 2000; Oñate
et al
., 2000).
Indicators that reveal declining stocking densities or the maintenance of grasslands,
where farmers are receiving payments to undertake such actions, are ultimatel
y of
little long
term value in assessing the degree to which environmental considerations
are being integrated into agricultural policy. Second, in many instances, the
environmental benefits of changes to farm management techniques are extrapolated
and ass
umed, rather than specifically revealed and documented. Hence, conversion of
arable land to grassland, like conversion from ‘conventional’ to organic farming, are
universally, and often uncritically, taken as being of positive environmental benefit.

the employment of such indicators provides only a shallow and some would
say ‘optimally inaccurate’ (Moxey
et al.
, 1998) evaluation of scheme effectiveness.
They are temporally and spatially limited and, in their uni
dimensionality, reveal little
of the u
nderlying processes at work. In their assessment of such indicators, Andersen
et al.

(1999) conclude that the farm management effects identified “cannot be linked
directly to specific stated targets or success criteria of individual schemes, mainly

such criteria do not exist” (p.143).

While the two sets of indicators identified above have undoubtedly contributed to the
generation of an increasingly systematic and comparative evaluation methodology of
AEP, and have provided policy makers with an imp
ortant tool in the legitimisation of
their intervention, they remain, within the context of an intended drive towards
achieving more sustainable forms of agriculture, limited in their ability to account
either for variations in implementation or for the ul
timate durability of sustainable
practices. We have argued above (see Section 2.1) that critical for the understanding
of both of these are assessments of farmer participation and the sociological,
economic and institutional contexts and rationales that in
fluence that participation (i.e.
cultural and qualitative indicators). These constitute the third and fourth sets of
evaluation criteria that we wish to explore here.

3.2 Actor and procedural indicators

In looking at farmer participation as a mean
s of assessing and of evaluating AEP
performance and effectiveness, how do we account for farmer participation? At one
level of analysis, we might seek to understand how and why farmers participate in
schemes (and why they do not participate) by considerin
g their characteristics, their
motivations and their attitudes towards individual schemes, or indeed to AEP as a
whole. Such an approach commonly yields typologies of farmers and allows us to
differentiate likely participants from likely non
participants (
see, for example,


Brotherton, 1989; Morris and Potter, 1995; Billaud
et al.
, 1996; Wilson, 1996; 1997a;
CNASEA, 1997; Lemery
et al.,

1997; Schramek
et al.
, 1999; Wilson and Hart, 2000

Nevertheless, considerable care needs to be taken in drawing hard an
d fast
conclusions regarding the ‘success’ of schemes from these apparent participation
factors. While certain farm and farmer characteristics do correlate with participation
in schemes, this information is too often highly scheme specific and does not alw
hold up when used in a cross
scheme or cross
national comparison (see our discussion
of the ‘5 Rs’ in Section 2.3). Further, statistical correlations, however significant, do
not in themselves indicate causality. While large farms might show a statisti
significant correlation with scheme participation in some areas, one cannot conclude
that having a larger farm is, in itself, a reason for, or indeed an influence on,
participation (Wilson and Hart, 2000). Many such variables are beset with
ality contradictions
. For some observers, youthful farmers are more likely to
participate, for others, older participants are the keener subscribers. For others,
financially buoyant farms are more likely to participate, while for others it is the
lly more marginal farmers that turn to agri
environmental measures as a
specific form of income support (see Morris and Potter, 1995; Wilson, 1996; 1997a;
Lobley and Potter, 1998; and Wilson and Hart, 2000, for detailed investigations of
these factors).

Second, few of the identified factors are
, but are frequently part of a
complex social and/or economic cognitive system. To focus simply upon observable
farm/farmer characteristics is to ignore the context within which these characteristics

developed and sustained. Third, the problem of scheme

makes the use
of structural indicators in creating an operational methodology for assessing agri
environmental scheme effectiveness problematic (see also Section 2.4 above).
Further, there

is the issue of appropriate levels of
. To what degree can
one draw comparable and generalisable information from studies of farmer
participation when that participation, and indeed farming as an activity, is so
embedded in a whole array of
national and sub
national social, technological,
political, cultural and territorial factors that ultimately make up the very diversity and
multifunctionality of rural Europe? Ultimately, while understanding farmers’ reasons
for participating (or not) in a
environmental schemes may be of value in targeting
and designing schemes to achieve greater rates of take
up, it ultimately “offers little in
the way of accessible, comparative, operational and quantitative indicators for the
evaluation of scheme

(Wilson and Hart, 2000, p. 2182).

Another level of investigation focuses on the impact of schemes upon participants.
Critical to any evaluation of the sustainability of policy objectives is the degree to
which they lead to long
term changes in pra
ctices and/or attitudes

changes that
potentially stretch beyond shorter periods of compensation and financial support or
encouragement (Wilson, 1996). The drive towards more sustainable forms of
agriculture should not be founded in a short
term funding p
olicy, but in a strategic and
jointly negotiated shift in, or re
emphasis of, values on the part of producers,
consumers and regulators. For, as Lowe
et al.

(1999) point out:


Seemingly similar schemes might, in fact, have very different objectives within different national or
regional contexts. A classic example is grassland protection which

varies from being a clearly
ecological measure designed to protect highly sensitive zones (e.g. along rivers), to being a general
funding measure for extensive husbandry methods in high landscape value areas.


“Agricultural policy reforms in general, and agri
environment measures in partic
should enhance farmers’ attitudes towards the environment and clarify their
understanding of their environmental responsibilities. It might reasonably be expected
that there would already be discernible changes in farmers’ attitudes, and even in farm
cultures, from participation in agri
environment schemes, even where the environmental
consequences could still not be gauged” (p. 271)

This, perhaps, is the ultimate challenge for evaluation strategies of public policy,
going beyond the quantitative
realm of performance indicators to identify cognitive
and institutional shifts and responses among actors, organisations and juridical
regimes (Muller, 1998). The methodology for applying such an approach to AEP
remains, however, still largely undeveloped
despite exhortations to go in this direction
et al.
, 1998). The research from which this paper is drawn sought to address
this issue in two ways, first, by examining changing farmer attitudes towards agri
environmental preoccupations as an indicator

of scheme effectiveness (Buller, 1999;
Wilson and Hart, 2001) and, second, by considering the shifting role of farmers,
agricultural extension services and territorial management bodies to the provision of
environmental advice and the local construct
ion of agri
environmental schemes
(Buller and Lenormand, 1999; Buller and Brives, 1999).

What these various analyses have in common is the manner in which they have gone
beyond traditional quantifiable indicators. While each has, to a greater or lesser
xtent, revealed shifts in farmer attitudes and in institutional and process responses to
the requirements of environmental policy integration, particularly as schemes have
matured and local consultation exercises have multiplied, specific indicators of suc
change have proved more difficult to construct. Shifts in attitudes are highly
dependent on the attitudes of farmers before entering the scheme, on the types of
scheme proposed to farmers and upon the degree to which farmers are involved in the
ion and implementing of individual schemes alongside extension services,
countryside management structures and environmental interests (Wilson, 1996;
Buller, 1999; Wilson and Hart, 2001). Using such behavioural approaches, scheme
success can therefore be d
efined in three ways: in terms of the willingness of farmers
to adopt the management prescriptions, in which case extensive schemes involving
few changes to management practices might still be considered ‘successful’ as
farmers accept them more readily; in

terms of the degree to which schemes change
farmers attitudes towards the agricultural environment, in which case different types
of schemes emerge that involve more or less profound changes to farm management
(e.g. conversion to organic methods); and, fi
nally, in terms of the re
alignment of
local interests, stakeholders and institutional actors into new forms of partnership
based upon local and territorially specific agri
environmental management
frameworks (Buller and Lenormand, 1999; Buller and Brives,


4. Conclusions

It is clear from the above analysis that while indicators have a central role to play in
the evaluation of AEP, that role is essentially limited. Either that, or the notion of
what constitutes an indicator needs to be
examined in light of conceptual caveats
mentioned in the first part of the paper. Thus, our analysis suggests that the
assumption that agri
environmental indicators may act as a vehicle for summarising,


or simplifying, and communicating information about

policy parameters that are of
importance to agri
tal decision
makers needs to be questioned

confirming Moxey
et al
.’s (1998) and Lowe
et al
.’s (1999) in
itial fears about the
fallacy of uncritically adopting indicator
based models in
policy evaluation. We,
need to question whether indicators can provide managerially significant
information about patterns or changes in the state of the environment and human
activities as suggested in much of the literature based on posit
ivist notions of human
environment interaction (e.g. Eurostat, 1997; OECD, 1997,1998; CEC, 1998;
Jesinghaus, 1999). Our discussion on measuring the ‘success’ of AEP has particularly
highlighted the simplistic assumptions about the
of human
vironmental action and resulting environmental change that increasingly appear to
guide many policy evaluations, to the neglect of more complex indicators where
immediate quantitative data may not be available (especially actor and procedural
indicators; s
ee above).

Ultimately, what is required is a shift
away from simple
information provision
the realm of public policy
. The limits of indicators in evaluating
environment policy derive, in part, from th
e precipitate manner in which policy
makers and others turned to them in the mid
1990s as a means of justifying and
legitimising an emerging, fast
growing and poorly monitored financial sector of the
CAP. Yet, while researchers were being asked to supply a
ppropriate methodologies
for evaluating individual agri
environmental schemes, the central question over the
last few years has been that of the integration of environmental considerations, not
only into on
farm management practices but into the very archi
tecture of the CAP.
Thus AEP constitutes, in many EU states

a significant part of the so
pillar’ of the post
Agenda 2000 CAP,
even though this is
still a poor cousin to the
‘first pillar’ of commodity support
. The
that the EU
Rural Development

as the mainstay of the ‘second pillar’

bring about are, in a certain
manner, more revealing of the advances made over the last te
n years in the integration
of environmental concerns into agricultural policy (
et al.
, 2001) than any set of
quantifiable agri
environmental indicators.

This paper has identified different types of indicator which we have regrouped into
‘levels’. These have different functions within the policy evaluation process.
‘Uptake’ or ‘output’ indicators are politically useful in that they provide a ready
assessment of scheme ‘success’ defined in the most simplistic of ways, although they
may shed

little light on reasons for the implementation gap. ‘Outcome’ indicators,
meanwhile, show the effects of AEP, revealing whether or not individual schemes are
delivering their stated objectives. These indicators may be particularly useful to
to address possible inconsistencies between policies developed at the
‘top’ and implementation failure at the grassroots level (cf. Winter, 1990; Adriaanse,
1993; Brouwer and Crabtree, 1999a). ‘Actor’ and ‘procedural’ indicators go beyond
this to display,
first, the extent to which these objectives are the correct and most
relevant ones and, second, the overall position of AEP within a broader socio
and institutional agenda. Yet, as AEP moves on to become a component of a wider
rural development a
genda, the use of output and outcome indicators will, we believe,
become less important, as alternative methodologies of evaluation currently employed
for integrated rural development schemes, LEADER territories, and the EU structural
funds, will gradually

take their place.



The authors wish to acknowledge financial support from DGVI (European
Commission) for empirical aspects of this study as part of an EU
wide transnational
project (FAIR CT95
274). Many thanks also to all the teams and
people from 10
European countries involved in this project, who made valuable conceptual and
practical contributions during discussions about the utility of the indicator approach in
assessing the effectiveness of European agri
environmental policy.


ACANZ [Agricultural Council of Australia and New Zealand] 1993:
agriculture: tracking the indicators for Australia and New Zealand
. Victoria
(Australia): CSIRO.

Adriaanse, A. 1993:
Environmental policy performance indicators: a study on

development of indicators for environmental policy in the Netherlands
. The
Hague: Ministry of Housing, Physical Planning and the Environment.

Andersen, E.
et al.

1999: Environmental effects of agri
environmental measures
implemented under Regulation 2
078/92. In: Schramek, J.
et al.

implementation and effectiveness of agri
environmental policy after
Regulation 2078/92
. Final Consolidated Report to DG VI of the European
Commission (Contract FAIR CT95 274). Brussels: DGVI (Agriculture) of
Commission of European Communities, pp. 125

Asheim, G.B. 1994: Net national product as an indicator of sustainability.
Scandinavian Journal of Economics
96: 257

Baldock, D. 1997: Implementation of Regulation 2078/92 in Europe.
La Cañada
7: 3

rrett, S. and C. Fudge 1981:
Policy and action: essays on the implementation of
public policy
. London: Methuen.

Billaud, J
et al.

Sociological enquiry into the conditions required for the
success of the supporting environmental measures within the
reform of the
CAP: French report
. Report to the DGXII of the European Commission

BirdLife International 1994:
Implementation of EU Agri
environment Regulation
. Sandy (UK): BirdLife International.

Brotherton, I. 1989: Farmer participa
tion in voluntary land diversion schemes: some
observations from theory.
Journal of Rural Studies
5 (3): 299

Brouwer, F. 1995:
Indicators to monitor agri
environmental policy in the
. The Hague: Agricultural Economics Research Institute.

ouwer, F. 1999: Agri
environmental indicators in the European Union: policy
requirements and data availability. In: Brouwer, F. and B. Crabtree (eds):
Environmental indicators and agricultural policy
. Wallingford: CABI, pp.

Brouwer, F. and B. Crabtr
ee (eds) 1999a:
Environmental indicators and agricultural
. Wallingford: CABI.

Brouwer, F. and B. Crabtree 1999b: Introduction. In: Brouwer, F. and B. Crabtree
Environmental indicators and agricultural policy
. Wallingford: CABI,
pp. 1


wer, F. and S. van Berkum 1996:
CAP and environment in the European Union:
analysis of the effects of the CAP on the environment and assessment of
existing environmental conditions in policy
. Wageningen: Wageningen Pers.

Buller, H. 1999:
Assessing the impl
ementation and effectiveness of agri
environmental policy through farmer participation
. Unpublished paper to the
SAGA Workshop, European Commission Concerted Action, Fondation Ettei
Mettei, Milan.

Buller, H. 2000: The agri
environmental measures. In: Brou
wer, F. and P. Lowe
CAP Regimes and the European Countryside
. Wallingford: CAB
International, pp. 195

Buller, H. and H. Brives 1999: Farmers, actors and the local construction of agri
environmental knowledge. In: Schramek, J.
et al.

and effectiveness of Agri
environmental policy after Regulation 2078/92
Final Consolidated Report to DG VI of the European Commission (Contract
FAIR CT95 274). Brussels: DGVI (Agriculture) of the Commission of
European Communities, pp. 185

Buller, H. and P. Lenormand 1999: Mesures agri
environnementales et territoires.

Revue de l'Economie Méridionale

47 (1
2): 159

Buller, H., Wilson, G. and A. Höll (eds) 2000:
environmental policy in Europe
Aldershot: Ashgate.

Campbell, D.E. 1987:
Resource allocation mechanisms
. Cambridge: CUP.

CEC [Commission o
f the European Communities] 1992:
Council Regulation 2078/92
on the Introduction and Maintenance of Agricultural Production Methods
Compatible with the Requirements of the Preservation of the Environment
and the Management of the Countryside
. Luxembourg: O
ffice for Official
Publications of the European Communities.

CEC [Commission of the European Communities] 1997:
Report from the
Commission to the Council and Parlement on the application of Council
Regulation 2078/92

(COM (97) 620 Final). Brussels: CEC (D
G VI).

CEC [Commission of the European Communities] 1998:
Evaluation of agri
environmental programmes
. Brussels: CEC (DG VI), Working Document

CEC [Commission of the European Communities] 2000:
Indicators for the
Integration of Environmental Co
ncerns into the Common Agricultural
. Luxembourg: Office for Official Publications of the European
Communities (COM[2000] 20 Final).

CNASEA [Centre national pour l’Aménagement des Structures des Exploitations
Agricoles] 1997:
Contribution à l'évaluat
ion des mesures agri
. Paris: CNASEA.

CPRE [Council for the Protection of Rural England] 1995:
Measuring the
. London: CPRE.

Dabbert, S., Kilian, B. and S. Sprenger 1999: Site
specific water
quality indicators in
Germany. In: B
rouwer, F. and B. Crabtree (eds):
Environmental indicators
and agricultural policy
. Wallingford: CABI, pp. 177

Department of the Environment 1996:
Indicators of sustainable development for the
United Kingdom
. London: HMSO.

EEA [European Environmental
Agency] 1998:
Environment in the European Union
at the turn of the century
. Copenhagen: European Environmental Agency.

Environmental Challenge Group 1995:
Environmental measures: indicators for the
UK environment
. London: ECG.


Eurostat 1997:
Indicators of
sustainable development: a pilot study following the
methodology of the United Nations Commission on Sustainable
. Luxembourg: Eurostat.

Farrow, S. and J.A. Krautkraemer 1991: Economic indicators of resource scarcity:
Journal of Environ
mental Economics and Management
21: 190

FoE [Friends of the Earth] 1992:
Environmentally Sensitive Areas
. London: FoE.

Garrod, G.D., Willis, K.G. and C.M. Saunders 1994: The benefits and costs of the
Somerset Levels and Moors ESA.
Journal of Rural Stu
10: 131

Garrod, G.D. and K.G. Willis 1995: Valuing the benefits of the South Downs ESA.
Journal of Agricultural Economics
46: 160

Hammond, A., Adriaanse, A., Rodenburg, E., Bryant, D. and R. Woodward 1995:
Environmental indicators: a systema
tic approach to measuring and
reporting on environmental policy performance in the context of sustainable
. Washington: World Resources Institute.

Hogwood, B. and L. Gunn 1984:
Policy making for the real world
. Oxford: OUP.

House of Commons Agri
culture Committee 1997:
Environmentally Sensitive Areas
and other schemes under the Agri
environment Regulation
. London: HMSO.

IEEP [Institute for European Environmental Policy] 1998:
Assessment of the
environmental impact of certain agricultural measures

London: IEEP.

IFEN [Institut Français de l’Environnement] 1997:
Agriculture et environnement: les
. Orléans: IFEN.

Jesinghaus, J. 1999: Agricultural sector pressure indicators in the European Union. In:
Brouwer, F. and B. Crabtree (eds):
nmental indicators and
agricultural policy
. Wallingford: CABI, pp. 45

Lehmann, B. and H. Schmid 1997: Defining and identifying effectiveness and
efficiency of 2078 measures. In: Institut für ländliche Strukturforschung:
Implementation and effectiveness

of agri
environmental schemes established
under Regulation 2078/92
(First Progress Report). Frankfurt a.M.: IFLS, pp.

Lemery, B., Soulard, C. and B. Degrange 1997:
Les mesures agri
un terrain d'expression de la diversité des co
nceptions de métier en
. Unpublished paper presented to the SFER Conference ‘Mesures
environnementales en Europe’, 2
4 November, Paris.

Lobley, M. and C. Potter 1998: Environmental stewardship in UK agriculture: a
comparison of the Environm
entally Sensitive Area Programme and the
Countryside Stewardship Scheme in south east England.
29 (4):

Lowe, P.D., Ward, N. and C. Potter 1999: Attitudinal and institutional indicators for
sustainable agriculture. In: Brouwer, F. and B. C
rabtree (eds):
indicators and agricultural policy
. Wallingford: CABI, pp. 263

Lowe, P.D., Buller, H. and N. Ward 2001:
Setting the next agenda? British and
French approaches to the Second Pillar of the CAP
. University of
yne: CRE Working Paper N° 54.

MAFF [Ministry of Agriculture, Fisheries and Food] 1993:
Environmentally Sensitive Area Report of Monitoring
. London: HMSO.

MAFF [Ministry of Agriculture, Fisheries and Food] 1994:
Sensitive Areas:
environmental objectives and performance indicators for
ESAs in England
. London: HMSO.


MAFF [Ministry of Agriculture, Fisheries and Food] 1996:
Monitoring Report: South
Downs ESA
. London: HMSO.

MAFF [Ministry of Agriculture, Fisheries and Food] 1997:
onmentally Sensitive
Areas and other schemes under the Agri
environment Regulation
. London:

MAFF [Ministry of Agriculture, Fisheries and Food] 1998:
Development of a set of
indicators for sustainable agriculture in the UK: a consultation document
leased 22.6.1998). London: MAFF.

McKenzie, D.H., Hyatt, D.E. and V.J. McDonald 1992:
Ecological indicators
London: Chapman and Hall.

Ministère de l'Agriculture/ISARA 1998:
Evaluation des measures agri
. Paris: Ministère de l'Agriculture.

Moldan, B. and S. Billharz (eds) 1994:
Sustainability indicators
. Chichester: Wiley.

Morris, C. and C. Potter 1995: Recruiting the new conservationists: farmers’ adoption
of agri
environmental schemes in the UK.
Journal of Rural Studies
11 (1):

y, A., Whitby, M. and P. Lowe 1998:
Environmental indicators for a reformed
CAP: monitoring and evaluating policies in agriculture
. University of
Tyne: Centre for Rural Economy.

Muller, P. 1998:
Politiques publiques et risques collectifs
. Pa
ris: Centre National de la
Recherche Scientifique (Actes du seminaire du programme Risques
Collectifs et Situations de Crise).

National Audit Office 1997:
Protecting Environmentally Sensitive Areas
. London:

Norgaard, R.B. 1990: Economic indicators of

resource scarcity: a critical resource.
Journal of Environmental Economics and Management
19: 19

OECD [Organisation for Economic Cooperation and Development] 1993:
OECD core
set of indicators for environmental performance reviews
. Paris: OECD.

rganisation for Economic Cooperation and Development] 1997:
Environmental indicators for agriculture
. Paris: OECD.

OECD [Organisation for Economic Cooperation and Development] 1998:
sustainable development: environmental indicators
. Paris: OECD.

ate, J.J., Andersen, E., Peco, B. and J. Primdahl 2000: Agri
environmental schemes
and the European agricultural landscapes: the role of indicators as valuing
tools for evaluation.
Landscape Ecology
15: 271

Parris, K. 1999: Environmental indicators fo
r agriculture: overview in OECD
countries. In: Brouwer, F. and B. Crabtree (eds):
Environmental indicators
and agricultural policy
. Wallingford: CABI, pp. 25

Peco, B., Malo, J.E., Oñate, J.J., Suárez, F. and J. Sumpsi 1999: Agri
s for extensive land
use systems in the Iberian Peninsula. In:
Brouwer, F. and B. Crabtree (eds):
Environmental indicators and
agricultural policy
. Wallingford: CABI, pp. 137

Potter, C. 1998:
Against the grain: agri
environment reform in the United St
ates and
the European Union
. Wallingford: CAB International.

Romstad, E. 1999: Theoretical considerations in the development of environmental
indicators. In: Brouwer, F. and B. Crabtree (eds):
Environmental indicators
and agricultural policy
. Wallingford:
CABI, pp. 13

Rose, N. 1991: Governing by numbers: figuring out democracy.
Organization and Society
16: 673


Schramek, J., Biehl, D., Buller, H. and Wilson, G. (eds) 1999:
The implementation
and effectiveness of Agri
environmental polic
y after Regulation 2078/92
Final Consolidated Report to DG VI of the European Commission (Contract
FAIR CT95 274). Brussels: DGVI (Agriculture) of the Commission of
European Communities.

United Nations 1996:
Indicators of sustainable development framewor
k and
. New York: UN Department for Policy Coordination and
Sustainable Development.

Whitby, M. (ed) 1996:
The European environment and CAP reform: policies and
prospects for conservation
. Wallingford: CAB International.

Wilson, G.A. 1996: Far
mer environmental attitudes and ESA participation.
27 (2): 115

Wilson, G.A. 1997a: Factors influencing farmer participation in the Environmentally
Sensitive Areas scheme.
Journal of Environmental Management
50: 67

Wilson, G.A. 1997b: Sele
ctive targeting in Environmentally Sensitive Areas:
implications for farmers and the environment.
Journal of Environmental
Planning and Management
40 (2): 199

Wilson, G. and Hart, K. 2000: Financial imperative or conservation concern? EU
farmers’ moti
vations for participation in voluntary agri
Environment and Planning A
32 (12): 2161

Wilson, G. and Hart, K. 2001: Farmer participation in agri
environmental schemes:
towards conservation
oriented thinking?
Sociologia Ruralis
41 (2): 254

Wilson, G.A. and O.J. Wilson 2001:
German agriculture in transition: society,
policies and environment in a changing Europe
. Basingstoke: Palgrave.

Winter, S. 1990: Integrating implementation research. In: Palumbo, D.J. and D.J.
Calista (e
Implementation and the policy process: opening up the black
. Westport (Connecticut): Greenwood Press.

World Resources Institute 1995:
Environmental indicators: a systematic approach to
measuring and reporting on environmental policy performance in

the context
of sustainable development
. Washington DC: WRI.