93rd EAAE Seminar “Impacts of Decoupling and Cross-Compliance on Agriculture in
the New EU Member States”, Prague, 22-23 September 2006
Decoupling and Cross-Compliance as Concepts and Instruments in Agricultural
Kenneth J. Thomson1, TOP-MARD Project2
Multifunctionality in agriculture has been defined and analysed in a number of ways,
most notably by the OECD and in a WTO context, and has been promoted directly
and indirectly via various EU policy instruments in pursuit of the “European Model of
Agriculture”. This paper first reviews the concept(s) of multifunctionality, and then
proposes a 20*15 “Policy-Function Matrix” as a framework for analysing this concept
in more empirical terms. The Matrix forces consideration of the ways in which
different policy areas, such as components of the CAP, the Structural Funds,
biodiversity policy, etc., influence - positively or negatively - the delivery of
beneficial “functions” by affecting the extent and nature of “activities” which use
farm resources, including farm household labour as well as land etc. This framework
is then applied to a number of Case Study Areas within the TOP-MARD project, and
the “completions” of the Matrix for these Areas provide a cross-section of initial
judgements about the practical significance of the multifunctionality concept across
several Member States of the EU.
A second part of the paper considers the special features of decoupling and cross-
compliance within the CAP in the context of agricultural multifunctionality, in both
the “old” and the New Member States. Some limitations and inefficiencies of these
particular instruments are noted, and their prospects during the 2007-2013 period are
Emeritus Professor, Department of Geography, University of Aberdeen, St Marys, Elphinstone Road,
Aberdeen AB24 3UF, Scotland UK. Tel.: +44(0)1224 274138. E-mail: email@example.com.
2 This publication derives from the EU-funded project on ‘Towards a Policy Model of Multifunctional
Agriculture and Rural Development’ (TOP-MARD), with collaborating partners: the University of
Highlands and Islands, UK (co-ordinator); the Agricultural University of Athens, Greece; the
Institute for Rural Development Research, Germany; the Federal Institute for Less-Favoured and
Mountainous Areas, Austria; the Autonomous University of Barcelona, Spain; the Rural Economy
and Research Centre, Teagasc, Ireland; the University of Rome, Italy; the Nordic Centre for Spatial
Development, Sweden; the Norwegian Agricultural Economics Research Institute, Norway; the
University of Ljubljana, Slovenia; the Budapest University of Economic Sciences and Public
Administration, Hungary; and the University of Aberdeen, UK.
Decoupling and Cross-Compliance as Concepts and Instruments in Agricultural
Kenneth J. Thomson
Definitions of Multifunctionality
According to FAO (1999): “There are no internationally agreed definitions of the
multifunctional character of agriculture. However, … there exist several
internationally agreed references to the term”, including the “Agenda 21” of the
United Nations Conference on Environment and Development in Rio de Janeiro in
1992, and Commitment 3 of the 1996 Rome Declaration on World Food Security and
the World Food Summit Plan of Action. The concept has also been the source of some
discussion and tension in WTO circles, especially as regards concerns in the U.S. and
elsewhere that the EU has tried to use it to justify trade-distorting subsidies to
agriculture (Garzon, 2005).
Following Shumway (1984), Blandford and Boisvert (2002) suggest that “the simplest
view of multifunctionality is one in which two or more outputs are technically
interdependent”. While this originally involved only simple physical outputs
(“commodities”: see below) produced for sale, such as milk and beef, a broader
approach includes both marketed services such as farmhouse bed-and-breakfast, and
unsold “non-commodity” outputs such as landscape.
The major source of reference as regards the definition of multifunctionality is OECD
(2001), which specified (Box 1.1) that "multifunctionality is ... an activity-oriented
concept that refers to specific properties of the production process and its multiple
outputs". This represents a still broader approach, since “the production process”
could include (say) the treatment of livestock, labour practices, or land ownership,
which is not reflected in actual outputs unless differentiation is introduced by organic
According to the "working definition" of the OECD (2001, p. 7), the "key elements"
of (agricultural) multifunctionality are: "(i) the existence of multiple commodity and
non-commodity outputs that are jointly produced by agriculture; and (ii) the fact that
some of the non-commodity outputs exhibit the characteristics of externalities or
public goods". A “non-commodity” is a product which is not usually traded, usually
because it is an externality, i.e. it cannot be sold and bought on the market (and so is
“external” to the market). OECD (2001, Box 1.4 and Annex 1) also analyses in some
detail (not reproduced here) associated terms such as “jointness in production”, “by-
products”, “side-effects”, and “multiple outputs”3.
In English, “commodity” has two meanings, not always strictly distinguished or defined. In one (non-
specialist) sense, a “raw” commodity is an unprocessed (or lightly processed) farm product, such
as grain, milk or timber. In a more economic sense, a commodity is a product, i.e. a good (see
below) or service - or possibly even a resource or asset, such as land and labour (Vatn et al., 2002)
- that is supplied/provided, usually in sufficient quantities and with sufficient homogeneity
(substitutability) to enable a price to be established in market trading. OECD (2001, Box 1.4) uses
Mathematically, the joint production of two or more outputs from one activity (e.g. a
land use) or set of resources raises some difficulties. A standard mathematical
function such as y = f(x1, x2, …) is single-valued; multiple values can only be simply
derived from x-values for fixed-proportion (linear) cases. Multiple outputs y1, y2,
…can be handled by “product-substitution” functions such as g(y1, y2,…, yn) = 0,
where g( ) expresses the (not necessarily linear) trade-offs between the different
outputs. Optimality, of course, requires some overall welfare function, or at least
some lexicographic ordering.
Other definitions of multifunctionality, explicit or implied, and associated comments,
appear from time to time in the literature, e.g. “… there is still considerable confusion
among World Trade Organisation (WTO) member states about what is really meant
by the term ‘NTCs’ [non-trade concerns] or its synonym ‘multifunctionality’”
(Guyomard and Le Bris, 2004). From a New Institutional Economics (NIE)
perspective, Hagedorn (2005) has provided a paper on "the role of integrating
institutions for multifunctionality", with eight "interpretations" of the concept, and
distinguishing technical and institutional jointness. Freshwater (2005) has suggested
that: “multifunctionality is largely an urban-fringe issue for agriculture. Most of the
public good outputs of farming, whether positive or negative, require the presence of
non-farmers to be significant”; this re-focusses attention (see OECD quotation above)
on how important the non-commodity outputs of agriculture should be, compared to
its commodity outputs.
Furthermore, the multifunctionality concept can be, and has been, broadened from the
already wide range of public and private “goods and services” to outputs which may
be considered more “soft” if no less important, such as “social cohesion”, rural
culture, and a sense of entrepreneurship amongst rural residents. A TOP-MARD
colleague (Spissoy, 2006) has suggested that local farming can enhance “agro-
(il)literacy” i.e. understanding of the food chain by urban residents. And recently, a
German Minister of Agriculture (Seehofer, 2006) has pointed to “the agricultural
sector’s special responsibility towards society which can be seen in, for example, the
above-average proportion of disabled persons working in this sector. It is this ‘lived
multifunctionality’ that will make European agriculture a source of important
economic, scientific and political impetus in Europe in the future as well.”
OECD (2001, Box 1.2) notes that multifunctionality is not specific to agriculture; for
example, forestry is another widespread rural land-using activity that has long been
considered as a source of “non-timber products” as well as “softer” benefits. Other
industries might be considered in a similar light, e.g. the visual and environmental
aspects of buildings, or the cultural implications of public and private service
provision. However, it seems that land-extensive sectors, and agriculture in particular,
almost inevitably involve the mixture of multiple outputs implied by the term, and,
given the contention over the sector and its policies, thus attract particular attention.
“commodity output” in the second sense defined above, and considers this term as preferable to
“food output” since farms may produce fibre, energy, etc.
Economic Characteristics of Multifunctionality
OECD (2001, Box 1.3) also considers whether multifunctionality is a characteristic
of an economic activity or is an objective of society and therefore policy. The first or
“positive” approach is adopted in its report. However, as noted by OECD, this does
not exclude a discussion of the second or “normative” approach, which is implied by
the 1998 OECD Agricultural Ministerial Communiqué (which stimulated the 2001
study), and by EU declarations such as the “European Model of Agriculture”.
The “opponents” of multifunctionality (e.g. Rude, 2001) argue that some “functions”,
e.g. food security and viability of rural areas, are not external effects associated with
agricultural production. This approach may have led more recent OECD literature on
multifunctionality to deal mostly (or solely) with environmental effects. However,
both sides agree at least that agriculture generates both positive and negative
The homogeneity or substitutability (and hence what is considered an economic
“commodity”) mentioned above is a matter of context and judgement. For example,
litres of milk may differ in microbiological quality (and hence producer price) but are
generally regarded as equivalent; a battlefield may be locally unique, but from the
tourist point of view is comparable with other cultural sites; an eagle may be
ecologically rare, but to most people is simply another interesting species.
OECD (2001, Box 1.4) considers “market(ed)” and “non-market(ed)” outputs as
alternative terms to “commodity” and “non-commodity” outputs, but point out that
the policy issue is partly to decide which outputs can, in fact, be traded or not
(consider carbon, or organic products). Several recent CAP developments offer
farmers, but do not entitle them to, a range of payments for the supply of various
services, or the avoidance of potentially damaging operations. Hence, classifying a
product per se as “marketed” or otherwise pre-judges or confuses matters.
The term “function” in “multifunctionality” is therefore taken to mean the production
of a specific output (commodity or non-commodity). That is, there is a one-to-one
correspondence between each “function” and each type of output. Of course, since
outputs may be considered at different levels of aggregation (“the environment”,
“wildlife”, “birds”, etc.), the level of such functions must be similarly decided.
Most of the above discussion assumes commodities and non-commodities to be
“goods” (i.e. physical products) or services which are supplied to the market or
provided otherwise. The underlying assumption is that they are valued by at least
some people who are prepared to pay to “consume” them: they are “goods” in a value
(not just a physical) sense, as in the standard English meaning of generally useful or
beneficial5. However, as mentioned above, agriculture also produces negative effects
or “bads” such as pollution or atmospheric carbon6.
Thanks are due to Dimitrios Psaltopoulos for this point.
However, different people have different valuations, depending on their tastes, incomes or economic
situation; consider moorland (beautiful or ugly), landscape (familiar or otherwise), or eagles
(fascinating, or dangerous to livestock).
Some standard farm products, such as tobacco or alcohol, may be considered as “bad”. However, the
existence of a legal market with a positive price means that these are economic “goods” in a social
In some approaches, these “bad” or unwanted products are simply included in the
analysis symmetrically, i.e. as negatives. However, this gives rise to two problems. A
terminological difficulty, or confusion, derives from the (normal) English use of
“goods” as physical products (alongside services) whereas “bads” can be physical or
otherwise (e.g. ugly landscape). The other, economic, problem is that policy measures
to address public goods and public bads (see below), such as taxes and subsidies, or
regulations, are unlikely in practice be symmetrical in their administration or effects.
Indeed, most fiscal and regulatory measures (carbon trading is an exception) deal only
with one side or the other, in encouraging/discouraging or enforcing/prohibiting the
“good” or “bad” respectively. If both “goods” and “bads” must be considered
together, OECD (2001, Box 1.4) suggests that the term “effects” is used (though no
“cause” is directly implied).
A Policy-Function Matrix for Multifunctionality7
The main aim of the TOP-MARD study is to “develop the concept of
multifunctionality as a rural development policy instrument”
(http://www.policyweb.uhi.ac.uk/topmard). The study has therefore involved the
consideration and review of a number of relevant policy areas (chosen from an EU
perspective, rather than a local, regional or national one), and the specification of a
number of “multiple functions” of agriculture defined broadly, i.e. as including the
activities of farm household members, and possibly also (non-farm) forestry, which is
important for many EU rural regions. The study also involves 11 Case Study Areas
(CSAs), each roughly a NUTS3 region, and selected by a project Partner in each of 11
In order to analyse more precisely the concept of multifunctionality and its
relationship to policy, a matrix framework was adopted, as a tool with which to
consider the effects of various policies on the selected functions. Table 1 shows this
“Policy-Function Matrix” as it has been developed to date.
The columns of the Matrix represent various policy areas, i.e. collections of public
measures or instruments, whether expenditures (“subsidies” or “payments”), or non-
expenditures, e.g. border tariffs, regulations. Some column labels have been
generalised beyond CAP terminology, so as not to exclude Norway, which is a TOP-
MARD Partner. The following comments on each policy area have been written to
clarify their usage for TOP-MARD purposes:
Market Support comprises import barriers, export subsidies, domestic market
“intervention” purchasing, geographical indicator regulations, etc. for
agricultural products. In the EU, this is mostly “the old CAP”, now a
diminishing component of Pillar 1 expenditure but still important in OECD
sense. The problems of classifying illegal products such as opium and child labour are not
This concept arose from the author’s conversations with Prof. Demetrios Psaltopoulos, of the
Department of Economics, University of Patras, Greece. He therefore shares (at least) the credit
for the idea, but bears no blame for this author’s use of it.
Farm Income Support includes partly or mainly decoupled payments such as those
made on the basis of crop or grass area, breeding livestock numbers, etc. In the
EU, these include the new Single Farm Payment (or Single Area Payment in 8
of the 10 New Member States), i.e. the greater part of CAP Pillar 1. Cross-
compliance requirements should be included here.
Agricultural Development policy includes subsidies for farm development and
modernisation, and for agricultural product processing and marketing, mainly
for capital (i.e. grants, interest subsidies) but also for training, information,
advice, etc. Tax exemptions for agriculture, especially for land and farm
capital investment, might be included. In the EU, Axis 1 of the new RDR
covers some of this.
Agri-Environment policy includes specific payments for agricultural management
activities promoting environmental protection and enhancement, including
Less Favoured Area payments. In the EU, Axis 2 of the new RDR covers most
Biodiversity policy includes measures for wildlife or wildlife habitat more general
than specifically or mainly farmland.
Other (i.e. Non-Agricultural) Environment policy includes the Directives for Birds,
Habitat, Water, etc.
Structures policy includes the EU Cohesion Fund and Norwegian regional measures.
LEADER etc. comprises locally based schemes to promote socio-economic
Land Use policies include “development control”, “spatial planning”, etc.
Transport policy includes measures for road, rail, water and air transport, including
e.g. fuel taxation.
Energy policy includes general measures but also e.g. specific arrangements for
taxation of diesel for agricultural use, and subsidies for biofuel crops.
Housing, Forestry and Other policy columns have been included for completeness,
but have not yet been much utilised within TOP-MARD. One strong candidate
is policy for tourism.
The rows of the Matrix are the “functions” of agricultural multifunctionality, i.e. the
output of physical goods and/or non-physical services, whether “private” or “public”
or in-between (e.g. “club goods”).
Farm products means physical farm “commodities”, whether conventional or non-
conventional, e.g. organic, crocodiles. At the request of some TOP-MARD
Partners, separate rows have been inserted (lower in the Matrix) for Food
Security (presumably at national/EU level) and Food Quality, although these
may overlap somewhat.
Processed Products include those processed for drink etc., e.g. of grapes (for wine),
wool, cotton, tobacco, timber, straw (for heating).
Tourist Accommodation means bed-and-breakfast, camping or similar services in or
near the farmhouse (on the farm), but may include providing occasional meals
in that accommodation.
Tourist Activities means marketed on-farm facilities such as a café/restaurant, fishing,
“pick-your-own” fruit, etc.
Landscape means the visual aspect of the countryside, e.g. maintaining openness,
field features, etc.
Water includes flooding (control or worsening), aquifers (extraction or
replenishment), purification (or pollution).
Soil includes erosion, avalanches, landslips, etc.
Air includes smells.
Climate includes emission of greenhouse gases (GHGs) or carbon (C) sequestration.
Culture includes the preservation and “marketing” (not necessarily for payment) of
archaeological or historical remains, local customs (e.g. farming festivals), etc.
Entrepreneurship means business self-confidence, competence, innovation, etc.
Social Cohesion means contacts and trust between local people.
Food Security overlaps somewhat with (quantity of) Farm Products.
Food Quality overlaps somewhat with(quality of) Farm Products.
Animal Welfare may be taken to include Animal Health (a public good amongst
farmers, at least).
The first two of the three final “functions”, i.e. the provision of Employment, Incomes
and Wealth have been inserted at the request of some TOP-MARD Partners (and the
third by the author of this paper, for the sake of economic completeness). There is
debate whether these are “commodities” and/or “non-commodities” which are
“supplied” by agriculture, or whether they are higher-level concepts, e.g. policy
objectives. Ollikainen & Lankowski (2005) consider agricultural employment as a
measure of "rural viability".
Finally, each entry in a cell of the Matrix should indicate how a particular policy area
affects a specific “output” (either of farmers in their use of land etc., or of farm
household members in their pluriactvity or diversification activities) as the production
of a “function”. So far in TOP-MARD, the actual cell entries are only symbolic
indicators, such as a plus or minus sign, sometimes with an indication of relative
significance (e.g. two pluses, or the use of a larger font) and/or descriptive text in an
endnote, which may describe the measure(s) in more detail, or explain the time
pattern. A cell can contain more than one entry - perhaps for different policy
components, or different parts of the Case Study Area. It is not necessary to insert a
cell entry in all cells: in fact, only the important influences (direct and indirect) should
In most cases, quantity or quality (or both) of the function can be affected by the
policy. Moreover, policy may be considered to reduce the provision of a “bad” (e.g.
pollution) rather than the better (i.e. greater, more timely, cleaner, etc.) provision of a
“good”. In “completing” the Matrix for a Case Study Area (or otherwise), the key (but
not always simple!) question is: what would happen if the relevant policy measure
were removed (and not replaced by an equivalent national or other measure, and not
changing other policies). It is the policy influence that is wanted, not the absolute
quantity or quality of the function. There are many difficult issues of policy
effectiveness (e.g. per Euro) and efficiency, but these have not so far been considered
The entries in the cells of Table 1 summarise the “completions” of the Policy-
Function Matrix received from all TOP-MARD Partners for their respective Case
Study Areas. Each supplied short textual descriptions and explanations (not recorded
here) of their entries. Since not all Partners used exactly the same conventions as
regards positive, negative, ambiguous or “significant” effects (of Policy X on
Function Y), and in one or two cases a slightly different version of the Matrix was
used, some interpretation of the submitted material has been necessary. Moreover, not
all Partners chose to utilise the three “extra“ functions of Employment, Incomes and
Wealth (see above); therefore the relative number of entries for these rows is not very
meaningful. However, although some judgemental differences from CSA to CSA (i.e.
country to country) are to be expected, these factors are not thought to have altered
the general meaning of the aggregated material. The clustering of entries suggests the
For policies (columns):
Agricultural policies of all types (market support, direct payments, structural aid
and agri-environmental schemes) have major effects on almost all the identified
“functions“, though negative or ambiguous ones in some cases.
As is to be expected, biodiversity and other general (non-agricultural)
environmental policies have positive effects for natural resources such as water,
soil, air and wildlife.
General (non-agricultural) structural policies such as Regional Fund measures and
LEADER schemes have generally positive effects on several agricultural
functions, with LEADER cited in many cases.
Energy policy is considered to have a positive effect on farm production
(unprocessed and processed) and on the (global) climate.
Forestry policy (considered generally, i.e. not only for farm forestry) has effects
on farm production (directly or indirectly, depending on whether timber is a
„farm product“, or the product of a neighbouring industry offering jobs and
perhaps competition for land), and also on landscape and wildlife; these latter
effects may be negative for landscape.
For functions (rows):
The supply of farm products - unprocessed and processed - is affected (generally
positively) by a number of policies, not all „agricultural“. Food security is
enhanced by agricultural (and transport) policy. Food quality (actual or perceived)
is generally considered to be improved by several policies. Animal welfare
(including health) is also affected by several policy areas, generally positively.
Tourist facilities are promoted by both agricultural (farm) development policies
and LEADER-type schemes.
Landscape and natural resources (water, soil, air, wildlife) are affected, both
positively and negatively, by most policies; this is the main area of conflicting
„Soft“ functions such as encouraging entrepreneurship and social cohesion are
promoted by several types of policy, but other policies may have uncertain or
It is likely that these initial entries will be modified during further “completion
rounds” and discussion, by clarifying mutual understanding of the concepts involved,
by suggesting further policy-function effects, and/or by raising or lowering the
significance of such effects. If so, one or more revised versions will be published in
Some countries - notably Italy (I) and Eire (J) - have several of their entries in cells
unoccupied by other countries (and probably the reverse, although this is more
difficult to measure). This may be due to differing conceptions of the matrix by the
experts responsible for these entries, or to the special situation in those countries
and/or their case study areas.
Policy Scenarios for TOP-MARD
The TOP-MARD “Description of Work” specifies that the study should “… analyse
the consequences of various policy scenarios”. Work has only begun to address this
requirement, and this section sets out some initial thinking.
One consideration is the time horizon to be adopted: relevant future dates include:
2006: final agreement on Financial Perspective (budget allocations), e.g. for CAP
Pillar 2 and structural funds, for forthcoming EU programming period
2007: start of forthcoming EU budget/programming period
2008: possible signing of WTO agreement, and/or start of implementation period
2009: decisions following the Mid-Term Review of the CAP scheduled for 2008
2010 middle of Kyoto Protocol commitment period (2008-2012)
2010/2011: next decennial population census date (for which official projections
available?); also end of some EU Action Plan periods, e.g. Transport.
2013: budget and other decisions taken for 2014+ programming period
2020/2021: next but one decennial population census (for which official
projections may be available?); and possibly also 2030/2031 etc.
The effects of policies to be covered by TOP-MARD are very widespread. Some of
these effects may be short-term (e.g. falls in farm incomes if direct payments are
reduced); others will be gradual and much longer-term, e.g. farm structural
adjustment, local economic development, environmental effects. Moreover,
successive policy changes (e.g. of the last two or three CAP reforms, and of different
structural measures in the last two or three programming periods, and future changes)
have overlapping as well as extended effects, making analysis difficult. However,
perhaps at least 3 years should be allowed for rural (economic) development measures
to take effect, i.e. for new enterprises to become established, or to fail; this suggests a
time horizon of say 2009/2010 for structural measures to date, and 2015/2016 if the
effects of measures taken in 2007-2013 are to be captured. A similar time horizon
might be considered for other techno-economic policy measures, e.g. for biodiesel.
Environmental effects may take much longer to have their full effects, e.g. up to 100
years for afforestation, even if action “on the ground” (e.g. planting) has already taken
place. However, excluding climate change (as non-local), perhaps 10 years is a
reasonable time period, suggesting say 2023 for forthcoming measures to take effect.
Social effects of policy changes may be quite short-term, e.g. LEADER-type
encouragement to community groups to be formed, thus creating social capital, but
also medium- and long-term, e.g. continued emigration of young people from remote
areas, and eventual collapse of some community structures.
All in all, a time horizon for TOP-MARD policy scenarios (whether “gradual” or
“sudden”) of 2013 has been proposed, but with model simulation to be continued until
at least (say) 2033, i.e. 20 years beyond the policy changes being modelled, or 30
years from the base year of 2001.
Of course, future policy change is not the only influence on the variables included in
POMMARD; others include past policy changes, e.g. the “final” effects of recent
Structural Fund measures, future market changes (e.g. in relative price of food,
energy, transport), future social changes (e.g. more households, falling birth rates) and
future environmental change (e.g. climatic). Ignoring the possibility of an economic
or environmental collapse, either local or general, most of these changes can be
considered long-term steady pressures with resultant trends in POMMARD output
variables. However, some (e.g. an oil price shock) might be considered as a once-for-
all event, to which policy might or might not react. In any case, it has to be
determined whether POMMARD (and/or TOP-MARD scenarios) seeks to include
these non-policy phenomena, or whether a simpler “comparative-static/dynamic-
effect” or “shock/evolution” approach to modelling is to be taken, i.e. “once-for-all”
changes (mostly/all policy changes) are imposed on the modelled system, whose
medium/long-term dynamic effects are then simulated and studied, largely in terms of
Within the policy area, scenario specification might be carried using the structure of
the policy function matrix, i.e. explicit specifications for each of Market or Direct
Income Support, Development (Structural) Policies, Environmental Policy,
Transport/Energy Policies, etc. However, this would make for a very complicated set
of specifications. Instead, it is proposed that a smaller number of policy change
dimensions be used, e.g.:
A) CAP “reduction” (lower total CAP budget, tariff protection, etc.)
B) Further CAP reform (modulation of finance from Pillar 1 to Pillar 2)
C) More (or less) total structural fund spending, either as a result of overall EU
budgetary decisions, or by switching from (or to) large-scale and/or urban-
focussed targets to smaller-scale rural targets
D) Switching of structural fund spending from EU-15 to New Member States
E) More intensive/expensive environmental policy intervention
F) More intensive energy/transport policy action.
Each of these dimensions actually contains a multitude of sub-dimensions (e.g. which
Pillar 2 Axes are in B) above? Does F contain biocrop subsidies as well as fuel
taxes?). However, rather than introduce more dimensions, it seems better to limit the
number of possibilities, which already total 26 = 64 possible policy scenarios if each
dimension contains one alternative to the baseline policy. Thus the following “joint”
Policy Scenarios are proposed (for “2013” and afterwards):
I. Baseline: CAP as at (approximately) 2001; structural and other policies as
II. Current (2006): as I, but CAP as reformed in 2005/06 by 2003 decisions,
especially as regards decoupling and cross-compliance (see below)
III. Further CAP Reform: modulation of (say) 20% of Pillar 1 funds to Pillar 2;
structural policy as (to be) agreed in 2006; possibly implementation of a
(weak) WTO Doha Agreement after 2008
IV. Major CAP etc. Reform: modulation of (say) 50% of Pillar 1 funds to Pillar 2;
further structural policy reforms beyond those (to be) agreed in 2006;
implementation of a (weak) WTO Doha Agreement after 2008
V. Axis 2 (or pro-Agricultural Multifunctionality?) Reform: as III, but no (more?)
funds for EAFRD Axes 1 and 3, all extra funds going to farmers and
landowners for environmental purposes, land-based tourism, etc.
VI. Axis 3 (or pro-Rural?) Reform: as III, but no (more?) funds for EAFRD Axes
1 and 2, all extra funds going to individuals (trainees, entrepreneurs) and/or
communities for non-agricultural (i.e. “rural”) development purposes.
All the above are “plausible” scenarios, rather than “extreme” ones which might have
the attraction of greater simplicity in specification and results (impacts). However,
TOP-MARD is intended as a policy-oriented project rather than a purely academic
Decoupling and Cross-Compliance for Agricultural Multifunctionality
Given the above background, this section of the paper considers the special features of
decoupling and cross-compliance within the CAP in the context of agricultural
multifunctionality, in both the “old” and the New Member States. The development of
“decoupling” of agricultural support in the EU and elsewhere has been widely
discussed in the official and academic literature (e.g. OECD, 2006; Swinbank and
Tranter, 2005). However, its exact nature (e.g. the degree of decoupling, purposes),
and its direct and indirect effects (e.g. on production, investment and wildlife), are
still hotly debated. Both the mixed situation (i.e. partial decoupling, and country-to-
country differences) and theoretical uncertainties (should the decoupled direct
payments be treated as pure lump sums paid annually to households indefinitely, or as
perhaps temporary transfers to farm businesses enmeshed in developing agri-rural
policy?) present modellers with difficult issues when trying to simulate the likely
effects of recent reforms of the CAP.
Cross-compliance, i.e. the need to meet technical requirements in terms of ensuring
“good agricultural and environmental conditions” (GAECs) as a condition for
receiving the Single Farm Payment (SFP), has been somewhat less subject to analysis.
This is possibly because the CAP requirements so far specified by Member States in
terms of land management etc. are not difficult for most farmers, and the degree of
policing, including the risk and level of payment loss for lack of compliance, is
At least initially, cross-compliance in the New Member States (NMSs) which apply
the Single Area Payment Scheme (SAPS) rather than SFPs8 has been required only
for permanent pasture (European Commission, undated), which however is a type of
land use subject to great variation of definition. Hofhanzl and Postulka (2004) have
summarised the development and implementation of GAECs in the NMSs, and report
that varied approaches have been adopted, though with emphasis on agricultural
aspects, compared to the EU-15. They report that it has often proved difficult in the
Slovenia and Malta implemented SFPs directly, and so are subject to full cross-compliance.
NMSs to deal with the problems of land abandonment and with the protection of
features of high nature and landscape value.
All agricultural schemes of course carry some conditionality. In fact, “cross-
compliance” may be simply a convenient but temporary term for the conditions
attaching to eligibility for SFPs, which have been introduced with no clear rationale.
As (or if) the environmental or other (e.g. animal health and welfare) conditions are
made more strict over time, SFPs may become more like an environmental scheme
transfer, with the amounts (rates of payment) becoming more and more recognisable
as a reward (i.e. a payment, and not a subsidy) for “multifunctional” services provided
to the general population.
In different parts of the UK, various new schemes are making this clearer, with SFP
recognised as “Tier 1” (basic legal requirements being Tier 0), and higher-level
payments (e.g. the Land Management Contracts in Scotland) as Tiers 2 and 3 9.
However, this approach may not be adopted in all other member states, especially
those in the south and east of the EU-25, where the social, rather than the
environmental, rationale for SFPs (or the SAPS) is more widely assumed. Moreover,
the political forces attempting to maintain this form of income support are very
strong, and are not directed at clarifying, e.g. by differentiation, the more precise
“value for money” aspects of the new CAP arrangements.
When the difficult concept of multifunctionality is added to those of decoupling and
cross-compliance, the problems of analysis and quantification are multiplied10,
although Dwyer and Guyomard (2006) have recently presented a brave attempt at a
“framework” for such work. Most other papers attempting quantitative analysis in this
area appear to have selected only a few aspects of multifunctionality, e.g. Brunstad et
al. (2005), who investigate the efficiency of current policies in Nordic countries in
terms of measures of landscape preservation and national food security.
The 2007-2012 EU financial planning period will see the adoption and
implementation of new Rural Development Programmes by all Member States of the
EU-25 (about to become the EU-27). Many of these Programmes explicitly or
Agricultural multifunctionality continues to stimulate controversy and discussion in
both official and academic circles, and, despite the recent suspension of the WTO
An interesting situation has arisen in Scotland, where SFP payments to individual farmers (or at least
to farm businesses) have been made available to the public (e.g. the media) under Freedom of
Information legislation. However, payments made to farmers under agri-environmental schemes
are not being so released, apparently on grounds of commercial confidentiality attaching to all
public contracts. Since such schemes include (for example) the creation and improvement of
public paths (access to the countryside), the apparent beneficiaries (the general public) and their
representatives (e.g. local council and tourist agents, walking clubs) are finding it difficult to locate
the recreational improvements purchased with their tax contributions!
For example, the “key question” raised above for the PFM analysis - i.e. the with/without
comparison - is not appropriate, nor can the “before/after decoupling” point be confidently
Doha Round, this seems likely to continue, especially within the EU where the task of
reforming the CAP continues. In order to maintain progress, detailed and clear-headed
assessment of the effects of recent reforms is essential, so as to identify which
components are beneficial, and efficiently and effectively administered
(implemented), thus putting the necessary political pressure on those parties opposed
to further reform, or attempting to utilise the reform process in less efficient ways.
The broad-based nature of the multifunctionality concept may be its main attraction as
a political term, but this should not be allowed to allow the evasion of specific
identification and measurement of the various “functions” involved, even if this
cannot be pressed to a final conclusion, due to e.g. the problems of quantifying e.g.
“cohesion”, or of arriving at aggregate measures such “quality of life” indicators.
Within the TOP-MARD project, a Policy-Function Matrix has been developed as a
tool for clarifying and identifying - and perhaps ultimately for quantifying and
aggregating - the effects of various EU policy areas on a number of agricultural
functions, the latter considered as the outputs of the activities of farmers and other
farm household members. The use of this tool has already suggested some
judgemental conclusions that may be confirmed in further work, both from model
construction and policy scenario analysis, and from more detailed and focussed
measurement work in each case study area. The ultimate use of the tool to reach
aggregate conclusions at an EU level remains to be explored.
As major features of the most recent CAP reform, decoupling and cross-compliance
are of intrinsic interest, as well as being potentially important for the development of
the multifunctional nature, or the European Model, of agriculture. In this paper, it is
argued that decoupling has not yet operated widely or long enough to discern its
effects on farm production or on non-production functions of agriculture, although
some form of extensification, and thus lower production on the same or smaller EU
farmland area, seems inevitable.
Cross-compliance is seen as a so far weak, if widespread, form of a payments scheme
to farmers for public-good services. In this interpretation, further development of this
instrument would convert it into a more conventional - if not everywhere acceptable -
scheme, or schemes, for the delivery of specified services, such as the protection and
enhancement of natural resources. However, disagreement over the scope of such
services - specifically, whether they include the sheer survival of farm businesses and
“ways of life”, and/or greater connection of farming with the rest of the EU
community - make such development uncertain, and probably slow. In that case, the
proponents of further and faster CAP reform will face a strategic choice between
focussing on this aspect of the CAP’s “market and income support” Pillar 1,
emphasising the “rural development” Pillar 2 (including the role of modulation in
switching funds between the two Pillars), or urging the general reduction of the CAP
per se, in favour of broader “territorial” socio-environmental measures open to all-
Blandford D. and Boisvert R. (2002) Multifunctional Agriculture and Domestic/
International Policy Choice, Estey Centre Journal of International Law and Trade
Brunstad R. J., Gaasland I. and Vårdal E. (2005) Multifunctionality of Agriculture: an
inquiry into the Complementarity between Landscape Preservation and Food
Security, Eur. Rev. agric. Econ., 32(4), 469-488.
Dwyer J. and Guyomard H. (2006) International Trade, Agricultural Policy Reform
and the Multifunctionality of European Agriculture, paper presented to ENARPRI
conference on Trade Agreements and EU Agriculture, Brussels, June.
European Commission (undated) Cross-compliance in CEECs.
FAO (1999) Outcome of the Conference on the Multifunctional Character of
Agriculture and Land (Maastricht, Netherlands, September 1999), Rome.
Freshwater, D. (2005) Rural and Agricultural Intersections and Analytical
Opportunities, comments for OECD Workshop, Bratislava, 24-26 October, and
published in Coherence of Agricultural and Rural Development Policies, OECD,
Garzon, I. (2005) Multifunctionality in the European Union: Is There Substance
behind the Discourse’s Smoke?, seminar paper, University of California at
Guyomard, H. and Le Bris, K. (2004) Multilateral Agricultural Negotiations and
Multifunctionality: Some Research Issues, ENARPRI Policy Brief no. 4 February.
Hagedorn K. (2005) The Role of Integrating Institutions for Multifunctionality, paper
presented to EAAE Congress, Copenhagen. Humboldt University, Germany.
Hofhanzl A. and Postulka Z. (2004) Cross-compliance in CEECs, summary of
presentations and discussions at seminar ‘Cross-compliance in CEECs’, Prague.
Ollikainen M. and Lankowski J. (2005) Multifunctional Agriculture: the Effects of
Non-Public Goods on Socially Optimal Policies, Discussion Paper no. 1, MTT,
Helsinki (version also presented to EAAE Congress, Copenhagen)
Organisation for Economic Cooperation and Development (2001) Multifunctionality:
Towards an Analytical Framework. OECD, Paris.
Organisation for Economic Cooperation and Development (2006) The Challenge of
Decoupling Agricultural Support. OECD website “Reform Options and Analysis”,
Rude J. (2001) Multifunctionality: An Examination of the Issues and Remedies, in
Globalization and Agricultural Trade Policy (ed. Michelmann, Rude, Stabler, and
Storey), Lyne Rienner, Boulder.
Seehofer, H. (2006) European Agriculture in the Wake of Globalisation: What are the
Prospects for Diversity?, guest editorial in EuroChoices, 5(2), 6-11.
Shumway, C., Pope, R. and Nash E. (1984) Allocable Fixed Inputs and Jointness in
Agricultural Production: Implications for Economic Modeling, American Journal
of Agricultural Economics. 66: 72-78.
Spissoy, A. (2006) Comments on the Policy Matrix, TOP-MARD study paper, June.
Swinbank, A. and Tranter, R. (2005) Decoupling EU Farm Support: Does the New
Single Payment Scheme Fit within the Green Box? Estey Centre Journal of
International Law and Trade Policy, 6(1), 47-61.
Vatn, A., Kvakkestad, A. and Rørstad, P. C. (2002) Policies for Multifunctional
Agriculture: the Trade-off between Transaction Costs and Precision, Report no.
23, Department of Economics and Social Sciences, Agricultural University of
Table 1: Policy-Function Matrix: Case Study Area Completions
GDA GAE GA±
EJIS ±GA DJS
Farm productsix NTH
-IS -T -J
STU H N
Processed GAE JSH
JISU G JINT GSU EJ G±J
productsx N U
Tourist Accom.xi INH
JT GT J GS
DA- GEJ GDA G±D ±JIS-
G±J ±D±J -NT -J
-G- G- ±GJ GDJI
Waterxiv D-J-I D±J H TH
IS JIS G J T -J G
-GE- ±GE GAI -
I ISHU G
-D -D GE
-GE H I SHU G
±G- G±D ±GJ GDA AJN GE±J
Wildlife D-J J U JH
Climatexvii -N E H H SU GE
Culturexviii J-N- ±J GEH J I JINH G J J
Entrepreneurship GAJI GEJI GAE
GJH -E I
xix H STH JHU
Social GJS- GEJ
GSH GS J J ±J ±JN
Cohesionxx NTH HU
Food Securityxxi JSN AN
GEJS GDA GJN
Food Qualityxxii G-I-J GJ
THU JN U
Animal -G-- AEJS
JISN JT -N SNU
Welfarexxiii JIU HU
Fire Reduction E
Employmentxxiv EJI E-J -JI A IT A D EN
I I I IT A ±J
Wealth JIN AIN I I I
NB: G = Greece, D = Germany, A = Austria, E = Spain, J = Eire, I = Italy, S = Sweden, N = Norway, T =
Slovenia, H = Hungary, U = UK (Scotland); signs (-, ±, ?) refer to letter immediately following.
* Other Functions: Fire Reduction and Manure Absorption (Spain), Recreation (Germany, Austria)
i Market Support comprises import barriers, export subsidies, domestic market “intervention”
purchasing, geographical indicator regulations, etc. for agricultural products. In the EU, this is
mostly “the old CAP”, now a diminishing component of Pillar 1 expenditure but still important in
ii Farm Income Support includes partly or mainly decoupled payments such as those made on the basis
of crop or grass area, breeding livestock numbers, etc. In the EU, these include the new Single Farm
Payment (or Single Area Payment in 8 of the 10 New Member States), i.e. the greater part of CAP
Pillar 1. Cross-compliance requirements should be included here.
iii “Agricultural Development” policy includes subsidies for farm development and modernisation, and
for agricultural product processing and marketing, mainly for capital (i.e. grants, interest subsidies)
but also for training, information, advice, etc. Tax exemptions for agriculture, especially for land and
farm capital investment, might be included. In the EU, Axis 1 of the new RDR covers some of this.
iv Agri-Environment policy includes specific payments for agricultural management activities
promoting environmental protection and enhancement, including Less Favoured Area payments. In
the EU, Axis 2 of the new RDR covers most of this.
v Biodiversity policy includes measures for wildlife or wildlife habitat more general than specifically
or mainly farmland
vi Non-Agricultural Environment policy includes the Directives for Birds, Habitat, Water, etc.
vii “Structures” policy includes the EU Cohesion Fund and Norwegian regional measures
viii “Land Use” policies include “development control”, “spatial planning”, etc.
ix “Farm products” means physical farm “commodities”, whether conventional or non-conventional,
e.g. organic, crocodiles. On request, separate rows have been inserted for “Food Security”
(presumably at national/EU level) and “Food Quality”, although these may overlap somewhat.
x Including processing for drink etc., e.g. of grapes (for wine), wool, cotton, tobacco, timber, straw (for
xi “Tourist Accommodation” means bed-and-breakfast, camping or similar services in or near the
farmhouse (on the farm), but may include providing occasional meals in that accommodation.
xii “Tourist Activities” means marketed on-farm facilities such as a café/restaurant, fishing, “pick-
your-own” fruit, etc.
xiii “Landscape” means the visual aspect of the countryside, e.g. maintaining openness, field features,
xiv “Water” includes flooding (control or worsening), aquifers (extraction or replenishment),
purification (or pollution),
xv “Soil” includes erosion, avalanches, landslips, etc.
xvi “Air” includes smells.
xvii “Climate” includes emission of greenhouse gases (GHGs) or sequestration of carbon (C).
xviii “Culture” includes the preservation and “marketing” (not necessarily for payment) of
archaeological or historical remains, local customs (e.g. agricultural festivals), etc.
xix “Entrepreneurship” means business self-confidence, competence, innovation, etc.
xx Social Cohesion” means contacts and trust between local people.
xxi “Food Security” overlaps somewhat with (quantity of) “Farm Products”
xxii “Food Quality” overlaps somewhat with(quality of) “Farm Products”
xxiii “Animal Welfare” may be taken to include Animal Health (a public good amongst farmers, at
xxiv These three “functions” have been inserted at the suggestion of a PGA participant. They may
represent higher-level policy goals rather than the simpler “agricultural functions” in the rows above,
and so have been italicised. “Employment” should be taken to include all local (Case Study Area)
employment, and similarly for “Incomes” and “Wealth” (e.g. rises in land values or property prices).