Académique Documents
Professionnel Documents
Culture Documents
Publisher
Socit Franaise d'conomie Rurale
(SFER)
Electronic version
URL: http://economierurale.revues.org/406 Printed version
DOI: 10.4000/economierurale.406 Date of publication: 15 dcembre 2008
ISSN: 2105-2581 Number of pages: 23-38
ISSN: 0013-0559
Electronic reference
Ian Hodge et Peter Midmore, Models of Rural Development and Approaches To Analysis Evaluation
And Decision-Making , conomie rurale [En ligne], 307 | septembre-octobre 2008, mis en ligne le 01
septembre 2010, consult le 30 septembre 2016. URL : http://economierurale.revues.org/406 ; DOI :
10.4000/economierurale.406
Tfromhedence-based
recent increase in emphasis on evi-
policy must be applauded
a number of perspectives. It is good
classification built up from the location of
individual households, in an attempt to
uncover the needs of rural areas and com-
from a social point of view because pol- munities (Defra, 2004a)1. In England also
icy-making ought to be more precisely (although not in the other constituent parts
developed and targeted as a result of taking of the United Kingdom), levels of population
research findings into account; likewise, for density and urbanisation differ significantly
academic and other researchers, more atten- in relation to the European norm, so that the
tion to their efforts to understand the mech- classically assumed general equation
anisms and impact of policy intervention between rurality and disadvantage is not
provides an incentive to focus on immedi- valid. There are certainly some specific and
ate and relevant questions. However, in the intractable pockets of poverty and the
specific case of rural development there are socially mixed character of communities, but
some fundamental barriers to analysis and these are hard to identify (Cloke et al.,
evaluation of policy which need to be 1994). In the United Kingdom, responsi-
resolved. The most important of these stems bility for rural policy and rural develop-
from the fact that rural development, while ment has been complicated by the process of
it might reasonably in the past have been political devolution to constituent countries.
viewed in terms of sectoral policy, has The Westminster Government, represented
shifted to a territorial policy, or arguably, by Defra (and previously one of its prede-
further towards a local policy. Long- cessors, the Ministry of Agriculture, Fish-
standing controversies exist regarding the eries and Food), has overall responsibility on
nature, scope and definition of rural territory a European and international level, but in
itself. Different designations provide arbi- territorial terms covers only England. Else-
trarily different results, and those which are where, the devolved administrations carry
based on some kind of threshold such as that out the policy function and there is an
provided by the OECD (less than 150 per- increasing involvement at the regional level
sons per square kilometre) conceal what (Ward et al., 2003).
most commentators agree is a diverse range A further impediment to evaluation is
of socio-economic conditions (Hodge and caused by confusion over terminology. Since
Monk, 2004; Yarwood, 2005). the Agenda 2000 reforms, most of the Euro-
In England, the re-organisation of minis- pean Unions non-commodity European
terial responsibility following on from the Agricultural Guarantee and Guidance Fund
foot-and-mouth disease outbreak resulted (EAGGF) spending has been consolidated
in a Department of Environment, Food and into programmes delivered under the Rural
Rural Affairs (Defra), at least part of which
has a remit based on an uncertain geography: 1. Cf. Page 5.
Development Regulation (RDR)2. Cursory conditions in rural areas, the ways in which
examination reveals that the accompany- these conditions have been conceptualised
ing measures of MacSharrys Common into rural development theories, the politi-
Agricultural Policy (CAP) reform, consol- cal influence of different interest groups,
idated into the Second Pillar, are nar- and the policy approaches that have been
rowly focused on farming and its environ- implemented in practice. There is no clear
mental impact. Bryden (2000)3 has shown linear causality amongst these factors; rather
that less than 10% of planned expenditures we see interactions amongst them in a simul-
under the 2000-06 RDR programmes were taneous process of development. In practice,
on Article 33 measures focused on activ- this has been an evolutionary process, more
ities outside the agricultural sector, and con- a continuum than a set of discrete changes.
sequently a negligible fraction of overall However, we argue that these four models
EAGGF payments. While there are some do capture the characteristics of this more
evident indirect linkages between agriculture gradual change.
and the non-farm rural economy, it is diffi- The paper4 relates these changing and
cult to disentangle the various strands of developing contexts of rural development
EAGGF support, other Structural Funds based on the experience within the United
activity, experimental LEADER funding Kingdom to the implications for policy eval-
and national programmes for rural action uation, using specific examples that show
when attempting to link outcomes to activ- how the development of guidelines by both
ities. There is little evidence of radical the United Kingdom Government (HM
change under the current Rural Develop- Treasury, 2003) and the European Com-
ment Regulation (Dwyer et al., 2007). mission (CEC, 1999) have contributed to
These are the challenges which the paper making this a mechanical, path-dependent
sets out to explore. The United Kingdom activity. It concludes by suggesting that
experienced a relatively early period of although measurement of impacts on rural
industrialisation and urbanisation in the 19th economies, environments and communities
century compared with other European is a necessary component of overall evalu-
countries and a consequent transformation ations, without more discursive and quali-
of rural economies. More recently, in the 20th tative inquiry, it is not, alone, sufficient.
century, there has been a period of signifi- This in turn has implications for the ways in
cant counterurbanisation (Robert and Ran- which rural development decisions are made
dolph, 1983; Champion, 1994) when popu- in practice. Given the relatively early expe-
lations have increased even in relatively rience of these trends within the United
remote rural areas. We set out a series of Kingdom following an early industrialisation
four models of rural development that seek and rural transformation (Grigg, 1982), there
to chart the changes in the predominant may be implications for the ways in which
approaches to rural development over time. rural development is practised in other Euro-
While they differ in their focus and spatial pean countries as their rural areas pass
coverage, we argue that they represent dom- through similar stages of development.
inant characterisations and policy
approaches at particular points in time, and
imply different types of analysis and scales
of policy implementation. The models are 4. An earlier version of this paper was presented at
influenced by changing economic and social the joint Socit Franaise dconomie Rurale and
Agricultural Economics Society Conference on
Rural Development, Paris, 29 March 2006. The
2. EC1257/1999. authors are grateful for the helpful comments made
3. Cf. page 10. by the referees.
RESEARCH
Ian HODGE, Peter MIDMORE
Multisectoral Diversification
tural policy approach. In this model, agri- rural areas which can have less and less
culture represents the major sector in the local economic impact more generally. This
rural economy and its success determines the suggests an alternative, multisectoral
performance of the local economy more approach. The relatively small contribu-
generally. Agricultural decline promotes tion of agriculture to many rural areas
rural depopulation and a decline in rural means inevitably that other economic sec-
service provision. Thus, a policy to stimu- tors have come to play an increasing role in
late agricultural production not only supports the rural economy. Recreation and tourism
domestic food supply, agricultural employ- and more generally the service and indus-
ment and farm incomes, it also deters out- trial sectors have become dominant. With a
migration from rural areas and supports the continuing policy focus on supporting farm
rural economy and service provision more incomes, policy thus began to seek other
generally. However in the mid 20th century, approaches and in the later 1980s farm
a variety of, by now familiar, factors under- diversification became the buzzword in
mined this approach and the general con- policy circles (Newby, 1988). Farmers
sensus about the appropriate policies. The were encouraged to look for alternative
high costs, inefficiency and environmental sources of income by adding value to agri-
impacts of commodity price supports, espe- cultural products, by making use of farm
cially in the context of surpluses of agri- assets, especially land and buildings for
cultural products undermined the approach non-agricultural uses, by undertaking agri-
taken to agricultural protection (Buckwell et cultural work on other farms and by becom-
al., 1982). The changing nature of technol- ing involved in non-agricultural economic
ogy applied in agriculture with increasing activities off the farm. The emphasis on
mechanisation and application of inputs the diversification of the farm business sub-
imported from beyond the local economy sequently broadened to a wider analysis of
reduced the local economic impact of agri- farm households and the potential for pluri-
culture. The combined decline in the sig- activity, drawing on multiple household
nificance of the agricultural sector and the income sources, as a strategy for long term
widespread experience of counterurbanisa- farm household survival (Shucksmith, et
tion has meant that agriculture plays an al., 1989). This challenged the conven-
increasingly less important role in the rural tional view in the United Kingdom, in con-
economy. In the United Kingdom for trast to other European perspectives, that
instance in 2006, agriculture was estimated small farms represented only a temporary
to contribute some 0,5% of total value added phase in the process of agricultural adjust-
at basic prices (Defra, 2007). But there is ment towards an agricultural sector based
substantial regional variation; agricultures on full-time efficient farm businesses.
shares in the English regions, varying in Following this logic, it might be argued
2004 between 0,02% for London, and 0,6% that the conventional view of agriculture as
for the South East, to 1,3% in the East Mid- supporting the rural economy has come to
lands and 1,7% in the South West. be reversed to a situation where it is a suc-
cessful local economy that offers the means
2. A multisectoral approach of support for pluriactive farm households.
Thus, support directed exclusively through While it was recognised that pluriactivity
the agricultural sector faced increasing was not a new phenomenon, it gained an
exchequer costs in terms of dealing with the increased policy relevance. However, as
agricultural surpluses that can result from noted by Gasson (1988) at the time, the
increased production and with the declining goals of rural development might be pur-
relative importance of agriculture within sued more effectively by encouraging
RESEARCH
Ian HODGE, Peter MIDMORE
employment completely unrelated to agri- terms (Lowe and Ward, 2007) but particu-
culture. lar areas continue to suffer from problems of
low wages and underemployment. In areas
3. A territorial approach with low activity rates and high unemploy-
However, even so, such an approach is only ment, it may not matter very much what
partially multisectoral. A truly multisec- sort of economic stimulus is introduced.
toral approach to rural development policy Any sort of new activity can have multiplier
would look more generally and equally at effects that work through to other sectors and
the actual and potential roles for other sec- may in turn promote new opportunities for
tors in rural areas. While located in rural farm diversification, thus supporting the
areas, these will often have no economic farm population. In fact, it will often be
linkages at all with agriculture. The focus easier to create employment opportunities
thus shifts towards a more general analysis through the development of non-land based
of conditions within particular types of area, activities, either by encouraging the move-
or a territorial approach. And in practice, this ment of new economic activity into the area
means a focus on rural areas. Rural areas can or through endogenous growth. The latter
offer attractive locations for the establish- may be seen as more sustainable, although
ment of new economic activity, often asso- the former may be a more feasible alterna-
ciated with the most advanced sectors of a tive in areas where the economy is especially
modern economy, such as in information undeveloped.
technology, and many areas have gained In other areas, economic change is char-
employment from the establishment of new acterised by a rather different pattern of
firms and types of employment (Keeble and development, which we can term the con-
Tyler, 1995; North, 1998). This reflects the temporary model of rural change (Hodge,
generally declining significance of trans- 1997); in contrast to the traditional model
port costs in industrial production, the attrac- that is driven by changes within the agri-
tiveness of living in rural areas and the con- cultural sector. This recognises that a pro-
gestion costs of urban locations. portion of rural areas have a significant
These socio-economic changes in rural comparative advantage leading to economic
areas have been associated with the break- success and population growth or coun-
down of longstanding networks and link- terurbanisation. This embraces a variety of
ages, such as associated with the supply of different processes of varying importance
agricultural inputs and the marketing of across different localities. A major driving
agricultural products. In a context of relative force behind it is the fact that rural areas
agricultural decline the significance and offer attractive environments in which to
penetration of agricultural norms is dimin- live and work, while higher incomes and
ished within the wider community and this improved transport infrastructure reduce
has not been replaced by any alternative the constraints on locational choices. Thus
single dominant perspective. In practice, those working in towns can travel longer dis-
we can recognise rural areas in a variety of tances to work, increasing the level of com-
different circumstances and facing quite muting. But the effect is more widespread
different types of problem. But given the than this; even relatively remote locations
variety of circumstances found in rural areas, have experienced population growth. Earlier
we may then suggest that most generalisa- retirement has freed up older people to live
tions about the character of rural areas in attractive locations away from a place
will be wrong (Hodge and Monk, op. cit.). of work. The increased congestion in urban
Rural areas in the United Kingdom generally areas and improved road and rail networks
have performed relatively well in economic outside them have altered the relative acces-
sibility of different types of locality; the the less remote rural areas. Different con-
less remote rural areas are generally more ditions in rural areas can also be associated
accessible than central urban locations that with different types of problems. We can, for
suffer from traffic congestion. Rural areas example, identify two different sorts of
are also attractive to new forms of employ- problem associated with housing: poor hous-
ment, often based on entrepreneurs choos- ing conditions as represented by over-
ing to establish new businesses in places crowding or lack of facilities, or problems of
where they want to live. Finally, there is access to housing as represented by a high
anecdotal evidence of downshifting, peo- level of housing costs relative to local
ple deciding to opt out of more stressful incomes (Midgley et al., 2003). The differ-
employment to take up a less pressured ent distributions of these indicators are
lifestyle, often in a rural location. These shown in figure 2. Relatively high levels
have different impacts on different groups of of both indicators of disadvantage are found
the population. For instance, those living in rural areas. But problems of housing
in rural areas tend to have higher income affordability tend to be concentrated in the
levels than those in urban areas, while those more affluent south-eastern part of the coun-
working there often have lower levels. try around London, while problems of hous-
Thus rural areas often follow divergent ing conditions tend to be concentrated in the
paths, some in long term decline and others more remote rural areas. Thus they might
experiencing considerable prosperity. Some both be seen as rural problems, but relat-
continue to be characterised by the tradi- ing to very different types of rural areas.
tional rural problems. Even if their popu- These issues suggest some limits to a
lations are not significantly declining, they general territorial approach, especially one
can often have low incomes and activity that distinguishes simply between urban
rates, although those on the lowest incomes and rural areas. Changes in the circum-
are not necessarily engaged in the agricul- stances in rural areas indicate a higher
tural sector. Others with relatively high degree of complexity. There is no single
average incomes experience quite different sector that can be seen as a source of
sorts of problems. While the majority of employment growth across rural areas in
the population is often generally well off and general. Rather, specific opportunities will
can get good access to services, there is a depend on local characteristics, especially
minority which experiences problems that the natural environment, such as landscape,
are in many ways a consequence of the topography or an attractive coastline. It may
affluence of the majority, the fact that house also depend on the presence of employment
prices are high or that, because the majority clusters in nearby urban areas.
do not demand certain services such as pub- Other relationships also seem less
lic transport, they are not provided at all. straightforward. While it may have been
This divergence of experience across assumed that the maintenance of popula-
rural areas is seen in various ways. The tion numbers will provide for the mainte-
higher numbers of people in some areas nance in the provision of local services, this
disguise the incidence of problems. Defra no longer holds (Stockdale, 2004). Under the
(2006) has recently highlighted the distri- traditional model of rural decline, the level
bution of employees who are paid less than of service provision falls with the reduced
two-thirds of the English median wage. demand associated with a declining popu-
Concentration on the proportion of employ- lation and the emphasis in debate has gen-
ees who are low paid highlights the more erally been on the decline in services pro-
remote rural areas, but the absolute numbers vided in rural areas. But in practice many
of low paid employees are often higher in other factors are associated with the level of
RESEARCH
Ian HODGE, Peter MIDMORE
Map 1 - Rank of access to housing, lowest 20% Map 2 - Rank of housing conditions, lowest 20%
and the ease with which it occurs at all will zontal associations, such as land manage-
depend on local institutions and the level of ment co-operatives, while others develop
social capital. A sectoral approach required vertical associations, such as facilitation for
little institutional development at the sub- the implementation of policy. More attention
national level. However, the move towards is needed on the optimal form and level of
a territorial, and especially to a local administrative intervention in the delivery of
approach, involves a much greater degree of rural development policies. This sort of
choice and discretion in the ways in which activity falls between the conventional roles
public resources might be applied. This com- of the public and private sectors, presenting
plexity makes far greater demands on infor- a challenge to analysis that casts the two sec-
mation and local institutional developments tors in clearly separate roles. It introduces
are required in order to handle it. investment in and maintenance of social
Experience with rural development capital as legitimate elements of a rural
schemes to date suggests that they can be development policy.
successful in the development of institu-
tions and social capital, especially as embod-
ied in the organisations that have been devel-
Policy indicators and analysis
oped in order to facilitate the implementation These changes in the nature and pattern of
of the schemes. Valuable initiatives have rural development have profound implica-
been made towards the development of local tions for rural analysis and policy evaluation.
institutional structures through such schemes In the positivist tradition (Weimer, 1998)
as Objective 5b and LEADER albeit in a policy evaluation is undertaken to test the
sporadic and piecemeal way (Ward and efficiency and effectiveness of specific pub-
McNicholas, 1998; Ray, 2000). But such ini- lic actions designed to achieve social welfare
tiatives are very small relative to the total benefits. For evaluation to work, therefore,
volume of support for rural areas that con- policy objectives need to be unambiguously
tinues to be put into rural areas through the stated, and causal mechanisms need to be
Common Agricultural Policy. Local insti- clearly understood. The latter is particu-
tutions have an important role in dealing larly important since other events or
with the increasing complexity of policy processes rather than the policy itself may
implementation by building social capital for affect the outcome. Increasingly, therefore,
dissemination of information, networking and especially in the study of rural devel-
amongst participants and co-ordination of opment, there has been a search for vali-
activities. dating measures, or indicators, which can
A variety of institutional arrangements discriminate whether policy action has been
and networks at the local level are involved, justified.
such as in public sector facilitation, by Such indicators should, according to the
organisations such as local authorities or European Commission (CEC, 2001), cover
National Parks, development, housing and efficiency (economic output in terms of
service provision associations, collective quality and quantity, competitiveness and
supply associations for environmental goods, viability, and institutional efficiency) and
local dedicated environmental funds, or equity (viability of rural communities and
conservation trusts. Some of these are purely the maintenance of a balanced pattern of
in the public sector, such as local govern- development, access to resources, services
ment facilitation. Others are essentially pri- and opportunities, and labour conditions).
vate, non-profit organisations, but gener- Further, to appreciate the range of compre-
ally substantially supported through hension of different parts of the system and
government funding. Some develop hori- the stages at which policies impact, differ-
RESEARCH
Ian HODGE, Peter MIDMORE
ent kinds of indicators are required. Process urban areas. This might reflect either the
indicators focus on policy implementation; cost of registering to claim the benefit where
output indicators provide quantitative meas- it requires personal attendance in a local
urements of effects identified as resulting town, or else where social norms may give
from the policy; outcome indicators assess greater priority to independence and greater
the extent to which policies achieve their social stigma to claiming benefits from the
stated objectives (Moxey et al., 1998). state.
Clearly, public resources for develop- A second type of problem relates to the
ment assistance must be targeted on defined way in which data are generally collected
priorities. But two types of problems are and analysed in compiling indicators of
often encountered in the targeting of rural local economic conditions (Fieldhouse and
development areas (Midgley et al., op. cit.) Tye, 1996). Thus, the smallest statistical
The first results from an urban character- unit within the Population Census is the
isation of local economic problems. While enumeration district, the area covered by a
the approach has now changed, in the United single enumerator. These districts are then
Kingdom deprivation has in the past been aggregated into larger statistical units on
assessed against indicators measuring chil- which the analysis is conducted. In urban
dren in flats, Commonwealth immigrants areas, groups of people with similar socio-
or overcrowded housing. None of these is economic characteristics tend to live in cer-
representative of rural problems. No account tain localities. These are often large enough
was taken of the availability of local serv- to be identified as separate statistical units.
ices, often a particular rural concern. Even However, within rural areas with smaller
an indicator of registered unemployment settlements, the unit will often include the
might be argued to be biased against rural whole settlement and so households with
priorities. In a large labour market, those lower income will tend to be included
who are unemployed can expect that regu- together with those on higher incomes. Thus
lar job search will lead to the identification the mean figure for the rural unit may well
of a suitable employment opportunity. In fail to reveal the presence of a low income
contrast, in a small labour market people population.
who are unemployed may well know that The selection of indicators and the meth-
suitable vacancies are unlikely to occur and ods that are used to analyse and evaluate
so decide to move to another area rather rural development policy are clearly asso-
than remain unemployed within the local ciated with the underlying model of the
area. This suggests that recorded unem- rural development process and its objec-
ployment might be lower because of out- tives. Table 1 suggests the different indica-
migration. Further, it may be that the costs tors and methods that may be associated
of registering as unemployed are higher in with the different rural development models.
a rural area because of the distance to be They also have different implications for
travelled to the employment office and the the sort of information collected and the
potential benefits lower as information might potential policy inferences.
be more readily available by other, personal The sectoral model concentrates on farm
means. Thus we might expect that a rural businesses and the means of raising farm
area with a given level of economic disad- incomes through agricultural production.
vantage would exhibit a lower level of reg- Even where the emphasis has shifted from
istered unemployment. This sort of argument increasing production, there is clearly poten-
might be generalised in that it is possible that tial for development by investing to reduce
the take up of social security benefits is on costs and rationalise farm production struc-
the whole lower in rural areas than it is in tures. The methods of analysis draw partic-
ularly on farm management but the approach remains quantitative and concentrates on
clearly misses both the non-agricultural quantifiable impacts and changes. The local
potentials for agricultural businesses and model gives greater attention to the softer
households as well as the conditions and attributes of development. It seeks to recog-
opportunities in other sectors. The multi- nise the variations in experiences amongst
sectoral approach recognises this wider eco- households and businesses within a partic-
nomic environment and looks more gener- ular local area and the significance of social
ally at indicators of the state of the economy and institutional capital in facilitating col-
as a whole and the interrelationships lective and community development. This
between sectors. However, in practice the indicates the introduction of qualitative
focus tended to remain on farm business research techniques, case studies or dis-
and households. Development is still inter- course analysis, and more deliberative
preted largely in terms of employment and approaches towards decision making. These
so policy evaluation concentrates on the different models and methods have direct
costs of creating new employment oppor- implications for the sorts of information
tunities. This may suggest initiatives to that may be available for policy decisions
attract new firms into the area or to stimu- and hence for decision-making processes
late employment creation from the devel- (table 1).
opment of endogenous resources. The ter-
ritorial model recognises the wider set of 1. Approaches in the United Kingdom
social and environmental determinants of Despite the contextual differences between
human welfare beyond employment and the constituent parts of the United King-
service provision. This suggests a cost-ben- dom, the articulation of policy and the
efit approach that seeks to bring market and framework of evaluation are relatively
non-market values together into a single similar (perhaps because all four admin-
accounting framework. The approach istrations share a common civil service,
Table 1. Indicators and methods in different development contexts
Indicators Indicative methods Implications
Sectoral Farm incomes Farm models Narrow focus misses
Agricultural population Enterprise and significant determinants
commodity studies of rural welfare
Multisectoral Farm household income Household surveys May still be limited
Employment Input-output analysis to agriculture sector
and unemployment Cost per job created Misses social and
Local value added environmental issues
Employment incomes
Territorial Population change Cost-benefit analysis Misses variations in
Proportion of population incomes and welfare
in disadvantage amongst population
Average incomes and specific local
Levels of service provision circumstances
Local Social indicators Case studies Capacity to consider full
Numbers of people in Qualitative analysis range of experiences
particular circumstances Deliberative methods but problems
Individual experiences with quantification
and aggregation
High transactions costs
RESEARCH
Ian HODGE, Peter MIDMORE
and the cultural imprint of is strong). Thus, its definition: Evaluation is the process
for example, in England prior to the out- which objectively judges the actual out-
break of foot-and-mouth disease, there comes, including any unintended side
was a strategic review of the nature and effects, of a policy or group of policies
role of rural economies (PIU, 1999), fol- against the policy objectives, or intended
lowed by a statement of rural policy pub- outcomes, and the resources that are used
lished by the two responsible Westminster in policy delivery.
Ministries (DETR/MAFF, 2000). This The planned evaluation framework con-
established the scope of rural policy, which sists of several streams: improving statistical
covers fair access to rural service provi- resources to establish a baseline for moni-
sion, including housing and transport; toring; using this to assess progress using
business performance in both the farm and the Rural White Paper indicators, and also the
non-farm sectors; rural conservation and rural Public Service Agreement targets set
leisure uses of the countryside; and the for Defra by the Treasury6; rural-proofing
vitality of communities and rural civil the programme-based evaluation other West-
society. Attached to these four priorities minster ministries policies; a study of local
are a series of 15 indicators. For economic rural services; and either a longitudinal study
development, for example, performance of rural households or case studies of a num-
of policy initiatives has been measured ber of rural communities, to examine cross-
from employment activity rates and unem- cutting impacts of policies. When examining
ployment rates in rural areas, the propor- these policies themselves, however, there
tions of market towns that are thriving, are some challenging complexities. The focus
stable or declining (based on service pro- on economic and social regeneration is
vision, business activity and employment), divided into two, sustaining the relative pros-
new business start ups and turnover of perity of the majority of rural territory, and
businesses in rural areas, total income more specific measures to address rural areas
from farming and off farm income, and with economic and social disadvantage. Most
levels of agricultural employment5. This of these consist of rural top-up funding for
suggests a dominance of the multisectoral existing economic development policies
model in policy-thinking. (skills, business support, broadband tech-
The consequences of the foot-and- nologies) delivered through other Ministries
mouth outbreak caused something of a or their agencies, and some minor regula-
paradigm shift (Scott et al., 2004), ini- tory modification of the land use planning
tially in terms of perception of the relative system. Improvement of the economic and
importance of constituent parts of rural environmental performance of farming and
economic activity, but perhaps more fun- food production is argued to be directly rel-
damentally a recognition that the admin- evant to economic regeneration, although
istrative framework of policy delivery and the contribution it can actually make may
evidence base was poorly suited to deliv- be small7. The new paradigm may be seen as
ery of the policy objectives. In addition to mixing the territorial and local models, with
a streamlining and reorganization of rural the more general territorial approach apply-
policy mechanisms, the new Rural Strategy ing across rural areas, but recognising the
(Defra, 2004b) provides a more detailed
and comprehensive approach to policy
6. This is to reduce the gap in productivity between
evaluation, so that much of the introduc-
the least well performing quartile of rural areas and
tory discussion of this paper is reflected in the English median by 2008, demonstrating progress
by 2006, and improve the accessibility of services
5. Ibid., p. 96. for rural people (HM Treasury, 2002, p. 111).
potential for local variations in experiences Much more importance needs to attach to
and the role of case studies and some degree identifying the specifics and spatial distri-
of decentralisation in decision-making. But it butions of problems and their causes; but
may be argued that the approach towards also, it is necessary to reveal the causal
evaluation has not followed through the processes that have the potential to resolve
implications of the changes taking place, and the problems. As has been indicated, this
there are inherent weaknesses in tracing the may well require novel developments in
chain of causality from actions to impacts the civil society of rural areas, but we have
(Basl, 2006). little systematic information on the roles
and impacts of networks and associations in
improving social and economic conditions.
The status of analysis, And we know less about how they may be
evaluation and decision-making successfully established and sustained.
What may seem to be lacking from this Analysis crosses the boundaries between
approach is a revised conceptual framework economics and sociology. Quantitative infor-
that fully recognises the changed and dif- mation is required on economic activities,
ferentiated circumstances of rural locali- but a necessary complement is required in
ties. In the context of a single dominant qualitative analysis of the influence of net-
sector, support for this sector may well have works, trusts or social norms.
trickled down to the population more gen- In principle, a case study approach offers
erally, although even here there may be scope for development of an appropriate eval-
doubts as to the extent to which such support uative strategy for rural policy. Rigorous in
ever did get to those who were most in depth study of carefully selected local areas,
need. Contemporary rural change involves using a mixture of quantitative and qualitative
more complex interactions and interdepen- data, can develop a sense of the interaction
dences in highly diverse contexts, so that as between increasingly diverse mixes of meas-
Saraceno (1999)8 argues, policies should ures in contrasting rural contexts where dif-
make different assumptions about factors ferent factors influence their expression and
influencing economic development and can- impacts, and contribute to understanding of
not be evaluated with the same tools that how and why they operate in the way they do.
have been developed for homogeneous, sin- This centre of interest of multiple case stud-
gle administration, top-down policies. Sta- ies, described as a quintain by Stake (2006)
tistical generalisation based on replicated is of a contemporary phenomenon in a real-
observation of a large number of cases is life context, especially when the boundaries
unhelpful since it has to assume things to be between the phenomenon and context are not
equal, whereas in most cases they are not. clearly evident (Yin, 1994)9. Therefore, much
of the preceding discussion can be related to
7. According to Defras own economic summary, Yins strategic recommendations for case
regional agricultural gross valued added does not study design and implementation. These begin
fully reflect the contribution of agriculture to rural with selection and exploration of the objects
economies... (and) ... presents a number of com-
plexities as neither sectoral nor area based indica-
of study, on the basis of general suppositions
tors currently provide a good basis for capturing the about the impact of policy which require test-
rural economy. Agricultural businesses account for ing. Multiple evidence sources should be
16% of all businesses in rural areas, but they only scrutinized to test rival hypotheses, which
account for 7% of both employees and turnover. might provide alternative explanations. Com-
Employees in rural businesses are more likely to
mon protocols to investigate different expres-
work in the manufacturing (17%), tourism (8%)
and retail (15%) industries.
8. Cf. p., 440. 9. Cf. page 13.
RESEARCH
Ian HODGE, Peter MIDMORE
sions of the phenomenon impart additional occupation, are additional conceptual tools
robustness. Analysis requires assessment of for analysis and investigation.
different patterns in the multiple data sources In practice, though, significant barriers
to refine and rule out competing hypotheses, impede the development and application of
both within individual case studies and such methods for improving understanding
between case studies carried out in different of the evolution, structure, and function of
contexts (see, for example, Coffey and Atkin- rural economies for refinement of policy
son, 1996)10. design. Because case-studies require exam-
The increased complexity and differen- ination of a great many variables, in detail,
tiation of rural development also has impli- in a small number of cases, they are rela-
cations for the ways in which policy deci- tively expensive, and skilled evaluators are
sions may be made. Local diversity implies scarce. There is a risk of becoming over-
that decisions must vary at the local level, whelmed by detail in mixed method evalu-
but an appropriate multi-level governance ations conducted at local level, due to their
system for the administration of rural devel- discursive nature. It is difficult to elaborate
opment undermines the traditional under- local level evaluation that fully reflects the
standing of effective sovereign govern- complexity and diversity of rural areas, and
ments delivering policies and assessing at the same time convey the critical infor-
their impacts. Differences exist in the oper- mation back up to higher levels to permit
ation of the networks of interests which balanced and informed decisions to be taken
have arisen to bridge the lack of coordina- about resource allocation. Generalisation
tion and consistency, overlapping with for- from case studies, especially from cross-
mal government structures and including case comparison where each individual
specialist (and highly effective) interest study has been carried out in a consistent
groups, and informal frameworks embodied manner, is possible, but involves a different
in conventions, each able to inhibit or facil- logic to conventional induction. In eco-
itate the actions of others (Morrison, 2006). nomic analysis, acceptance and consequent
The incidence of these, their effectiveness adoption of case study approaches is far
in addressing disadvantage, their impacts, from widespread (Bitsch, 2000) because
and efficiency in deploying limited they do not allow for the familiar statistical
resources and expertise are all poorly under- generalisations which come from large scale
stood and require investigation. There is a surveys. In contrast, theoretical generalisa-
risk that, rather than opening up opportu- tions deriving from identifying causal
nities to those who are excluded in present dependencies in one context contribute to
circumstances, they reinforce the influence better understanding of different mixes of
of particular interests (see for instance influences in other rural areas. Our ability to
Yarwoods, (2002) analysis of the operation make sense of different studies conducted in
of the rural exceptions policy and Shortall cases selected for varying purposes (of
(2004)). Case study methods can contribute which an increasing number have now been
to understanding of what is analogous to completed: for example, Hart (2003); Lee et
diverse ecosystems of intersecting associ- al. (2005); Midmore et al. (2004) is improv-
ations and organisations, businesses, infra- ing as a result of evolving prescriptions for
structures, and environmental systems rigorous meta-evaluation techniques
(Edwards, 2004). Extending this metaphor, (Cooksy and Caracelli, 2005).
interaction, duplication, and synergy of Responding to these challenges will
rural civil society, and niche creation and require a trade-off between qualitative eval-
uations to support decision-making at a
10. Especially Chapter 6. more local level, closer to the level of pol-
icy implementation, and the need to pass information back up to higher levels to per-
some information on performance back up mit balanced and informed decisions to be
to higher levels in order to permit higher taken about resource allocation across dif-
level resource allocation and financial con- ferent regions and even countries.
trol. Perhaps this is the fundamental chal-
lenge to combine local level evaluation that Thanks
fully reflects the complexity and diversity of The authors are grateful for the helpful com-
rural areas, and yet to convey the critical ments made by the referees.
REFERENCES
RESEARCH
Ian HODGE, Peter MIDMORE
PIU (1999). Rural Economies. London, Sta- community consequences and individual
tionery Office, Performance and Inno- migrant experiences. Sociologia Ruralis,
vation Unit. Vol.44, no 2, p. 167-194.
Ray C. (2000). The EU LEADER pro- Van der Ploeg J., Renting H., Brunori G.,
gramme: Rural development laboratory. Knickel K., Mannion J., Marsden T., de
Sociologia Ruralis, Vol.40, no 2, p. 163- Roest K., Sevilla-Guzmn E., Ventura, F.
171. (2000). Rural development: from prac-
Robert P., Randolph W.-G. (1983). Beyond tices and policies towards theories. Soci-
decentralisation: the evolution of the pop- ologia Ruralis, Vol.40, no 4, p. 391-408.
ulation distribution in England and Wales Ward N., McNicholas K. (1998). Objec-
19611981. Geoforum, Vol.14, n o 1, tive 5b of the structural funds and rural
p. 75-102. development in Britain. Regional Stud-
Saraceno E. (1999). The evaluation of local ies, Vol.32, no 4, p. 369-374.
policy making in Europe: learning from Ward N., Lowe P., Bridges T. (2003). Rural
the LEADER Community Initiative. and regional development: the role of the
Evaluation, Vol.5, no 4, p. 439457. regional development agencies in Eng-
Scott A., Christie M., Midmore P. (2004). land. Regional Studies, Vol.37, no 2,
Impact of the 2001 foot-and-mouth dis- p. 201-214.
ease outbreak in Britain: implications Weimer D.-L. (1998). Policy analysis and
for rural studies. Journal of Rural Stud- evidence: a craft perspective. Policy Stud-
ies, Vol.20, no 1, p. 1-14. ies Journal, 26(1), p. 114-128.
Shortall S. (2004). Time to re-think rural Yarwood, R. (2002). Parish councils, part-
development. EuroChoices, Vol.3, no 2, p. nership and governance: the development
34-39. of exceptions housing in the Malvern
Shucksmith D.-M., Bryden J., Rosenthall P., Hills District, England. Journal of Rural
Short C., Winter D.M. (1989). Pluriac- Studies, Vol.18, no 3, p. 275-291.
tivity, farm structures and rural change. Yarwood R. (2005). Beyond the rural idyll:
Journal of Agricultural Economics, images, countryside change and geog-
Vol.40, no 3, p. 345-360. raphy. Geography, Spring, Vol.90,
Stake R.-E. (2006). Multiple Case Study p. 19-31.
Analysis. New York, The Guilford Press. Yin R.-K. (1994). Case Study Research.
Stockdale A. (2004). Rural out-migration: CA, Sage, Thousand Oaks.