Models of Rural Development and Approaches To Analysis

Économie ruraleNuméro 307 (Novembre-décembre 2008)Nouvelles frontières entre les politiques rurales en Europe

...............................................................................................................................................................................................................................................................................................

Ian Hodge et Peter Midmore

Models of Rural Development andApproaches To Analysis Evaluation AndDecision-Making...............................................................................................................................................................................................................................................................................................

AvertissementLe contenu de ce site relève de la législation française sur la propriété intellectuelle et est la propriété exclusive del'éditeur.Les œuvres figurant sur ce site peuvent être consultées et reproduites sur un support papier ou numérique sousréserve qu'elles soient strictement réservées à un usage soit personnel, soit scientifique ou pédagogique excluanttoute exploitation commerciale. La reproduction devra obligatoirement mentionner l'éditeur, le nom de la revue,l'auteur et la référence du document.Toute autre reproduction est interdite sauf accord préalable de l'éditeur, en dehors des cas prévus par la législationen vigueur en France.

Revues.org est un portail de revues en sciences humaines et sociales développé par le CLEO, Centre pour l'éditionélectronique ouverte (CNRS, EHESS, UP, UAPV).

...............................................................................................................................................................................................................................................................................................

Référence électroniqueIan Hodge et Peter Midmore, « Models of Rural Development and Approaches To Analysis Evaluation And Decision-Making », Économie rurale [En ligne], 307 | Novembre-décembre 2008, mis en ligne le 01 septembre 2010. URL :http://economierurale.revues.org/index406.htmlDOI : en cours d'attribution

Éditeur : Société Française d'Économie ruralehttp://economierurale.revues.orghttp://www.revues.org

Document accessible en ligne à l'adresse suivante : http://economierurale.revues.org/index406.htmlCe document est le fac-similé de l'édition papier.© Tous droits réservés

ÉCONOMIE RURALE 307/SEPTEMBRE-OCTOBRE 2008 • 23

Introduction

The recent increase in emphasis on evi-dence-based policy must be applauded

from a number of perspectives. It is goodfrom a social point of view because pol-icy-making ought to be more preciselydeveloped and targeted as a result of takingresearch findings into account; likewise, foracademic and other researchers, more atten-tion to their efforts to understand the mech-anisms and impact of policy interventionprovides an incentive to focus on immedi-ate and relevant questions. However, in thespecific case of rural development there aresome fundamental barriers to analysis andevaluation of policy which need to beresolved. The most important of these stemsfrom the fact that rural development, whileit might reasonably in the past have beenviewed in terms of sectoral policy, hasshifted to a territorial policy, or arguably,further towards a “local” policy. Long-standing controversies exist regarding thenature, scope and definition of rural territoryitself. Different designations provide arbi-trarily different results, and those which arebased on some kind of threshold such as thatprovided by the OECD (less than 150 per-sons per square kilometre) conceal whatmost commentators agree is a diverse rangeof socio-economic conditions (Hodge andMonk, 2004; Yarwood, 2005).

In England, the re-organisation of minis-terial responsibility following on from thefoot-and-mouth disease outbreak resultedin a Department of Environment, Food andRural Affairs (Defra), at least part of whichhas a remit based on an uncertain geography:

consequently, it sought an entirely new def-inition, based on an “underlying settlementclassification” built up from the location ofindividual households, in an attempt touncover the “needs of rural areas and com-munities” (Defra, 2004a)1. In England also(although not in the other constituent partsof the United Kingdom), levels of populationdensity and urbanisation differ significantlyin relation to the European norm, so that theclassically assumed general equationbetween rurality and disadvantage is notvalid. There are certainly some specific andintractable pockets of poverty and thesocially mixed character of communities, butthese are hard to identify (Cloke et al.,1994). In the United Kingdom, responsi-bility for rural policy and rural develop-ment has been complicated by the process ofpolitical devolution to constituent countries.The Westminster Government, representedby Defra (and previously one of its prede-cessors, the Ministry of Agriculture, Fish-eries and Food), has overall responsibility ona European and international level, but interritorial terms covers only England. Else-where, the devolved administrations carryout the policy function and there is anincreasing involvement at the regional level(Ward et al., 2003).

A further impediment to evaluation iscaused by confusion over terminology. Sincethe Agenda 2000 reforms, most of the Euro-pean Union’s non-commodity EuropeanAgricultural Guarantee and Guidance Fund(EAGGF) spending has been consolidatedinto programmes delivered under the Rural

Models of Rural Development and ApproachesTo Analysis Evaluation And Decision-MakingIan HODGE and Peter MIDMORE • Department of Land Economy, University of Cambridgeand School of Management and Business, Aberystwyth University

1. Cf. Page 5.

307 COMPLET 7/11/08 9:18 Page 23

conditions in rural areas, the ways in whichthese conditions have been conceptualisedinto rural development theories, the politi-cal influence of different interest groups,and the policy approaches that have beenimplemented in practice. There is no clearlinear causality amongst these factors; ratherwe see interactions amongst them in a simul-taneous process of development. In practice,this has been an evolutionary process, morea continuum than a set of discrete changes.However, we argue that these four modelsdo capture the characteristics of this moregradual change.

The paper4 relates these changing anddeveloping contexts of rural developmentbased on the experience within the UnitedKingdom to the implications for policy eval-uation, using specific examples that showhow the development of guidelines by boththe United Kingdom Government (HMTreasury, 2003) and the European Com-mission (CEC, 1999) have contributed tomaking this a mechanical, path-dependentactivity. It concludes by suggesting thatalthough measurement of impacts on ruraleconomies, environments and communitiesis a necessary component of overall evalu-ations, without more discursive and quali-tative inquiry, it is not, alone, sufficient.This in turn has implications for the ways inwhich rural development decisions are madein practice. Given the relatively early expe-rience of these trends within the UnitedKingdom following an early industrialisationand rural transformation (Grigg, 1982), theremay be implications for the ways in whichrural development is practised in other Euro-pean countries as their rural areas passthrough similar stages of development.

Development Regulation (RDR)2. Cursoryexamination reveals that the accompany-ing measures of MacSharry’s CommonAgricultural Policy (CAP) reform, consol-idated into the “Second Pillar”, are nar-rowly focused on farming and its environ-mental impact. Bryden (2000)3 has shownthat less than 10% of planned expendituresunder the 2000-06 RDR programmes wereon “Article 33” measures focused on activ-ities outside the agricultural sector, and con-sequently a negligible fraction of overallEAGGF payments. While there are someevident indirect linkages between agricultureand the non-farm rural economy, it is diffi-cult to disentangle the various strands ofEAGGF support, other Structural Fundsactivity, experimental LEADER fundingand national programmes for rural actionwhen attempting to link outcomes to activ-ities. There is little evidence of radicalchange under the current Rural Develop-ment Regulation (Dwyer et al., 2007).

These are the challenges which the papersets out to explore. The United Kingdomexperienced a relatively early period ofindustrialisation and urbanisation in the 19th

century compared with other Europeancountries and a consequent transformationof rural economies. More recently, in the 20th

century, there has been a period of signifi-cant counterurbanisation (Robert and Ran-dolph, 1983; Champion, 1994) when popu-lations have increased even in relativelyremote rural areas. We set out a series offour models of rural development that seekto chart the changes in the predominantapproaches to rural development over time.While they differ in their focus and spatialcoverage, we argue that they represent dom-inant characterisations and policyapproaches at particular points in time, andimply different types of analysis and scalesof policy implementation. The models areinfluenced by changing economic and social

24 • ÉCONOMIE RURALE 307/SEPTEMBRE-OCTOBRE 2008

Models of rural development and approaches to analysis

2. EC1257/1999.3. Cf. page 10.

4. An earlier version of this paper was presented atthe joint Société Française d’Économie Rurale andAgricultural Economics Society Conference onRural Development, Paris, 29 March 2006. Theauthors are grateful for the helpful comments madeby the referees.

307 COMPLET 7/11/08 9:18 Page 24

may still be on farming, but there is encour-agement for agricultural diversification.The territorial approach recognises thewider interactions within the rural economyand the importance of social and environ-mental as well as economic issues. Finally,the differentiation between rural areas andthe variation in individual circumstanceswithin areas promotes a search for actionsthat recognise the specificity of solutions atmost local levels. These changes havereflected both forces fundamentally asso-ciated with national economic change andother factors more governed by local cir-cumstances. And they have major impli-cations for the methodologies that are rel-evant for the analysis of rural problemsand the evaluation of policies.

1. A sectoral approachIn the period following the Second WorldWar there were overwhelming prioritiesthat dictated the approaches taken to agri-cultural policy. These were driven by a needto ensure domestic food security and thecentral role of agriculture in rural economiesas reflected, for instance, in the analysisand conclusions of the Scott Report (Com-mittee on Land Utilization in Rural Areas,1942). This placed support for the agricul-tural sector at the centre and promised ameans of meeting a variety of objectivesfor food security, rural development, farmincomes and environmental protectionsimultaneously through a single agricul-

The transformationof rural development

There has been considerable recent discus-sion of the changes that are taking place inrural development both in terms of thenature of the changes underway within ruraleconomies and in terms of the approachesadopted towards rural policy. The predom-inant characterisation is of a single change,commonly from an approach focussed fun-damentally on the agricultural sector towardsone focussed on rural territories and morediversified economic activity (Van der Ploeget al., 2000; Léon, 2005; OECD, 2006).However, we argue that there has been amore steady process of economic and socialchange in rural areas over a longer period oftime.

The figure 1 sets out the basic argu-ments, illustrating the four predominantmodels of rural development. The imme-diate post-war model centred on the agri-cultural sector. Increasing food productionwas a first priority and other objectives,such as enhancing rural employment andservices, were seen as following directlyfrom the production support given to theagricultural sector. But through time theapproach has changed, shifting to multi-sectoral, territorial and local approaches.The multisectoral policy recognises thelimits to agricultural production supportand sees agriculture as one of several eco-nomic sectors through which the develop-ment objectives can be attained. The focus


RESEARCHIan HODGE, Peter MIDMORE

Figure 1. The evolution of rural development policies

General policy orientation Predominant models Policy implementationof rural development

Agricultural Policy Sectoral Commodity support

Multisectoral Diversification

Territorial Rural development

Rural Policy Local Local community developmentSource: the authors

307 COMPLET 7/11/08 9:18 Page 25

rural areas which can have less and lesslocal economic impact more generally. Thissuggests an alternative, multisectoralapproach. The relatively small contribu-tion of agriculture to many rural areasmeans inevitably that other economic sec-tors have come to play an increasing role inthe rural economy. Recreation and tourismand more generally the service and indus-trial sectors have become dominant. With acontinuing policy focus on supporting farmincomes, policy thus began to seek otherapproaches and in the later 1980s farm“diversification” became the “buzzword inpolicy circles” (Newby, 1988). Farmerswere encouraged to look for alternativesources of income by adding value to agri-cultural products, by making use of farmassets, especially land and buildings fornon-agricultural uses, by undertaking agri-cultural work on other farms and by becom-ing involved in non-agricultural economicactivities off the farm. The emphasis onthe diversification of the farm business sub-sequently broadened to a wider analysis offarm households and the potential for pluri-activity, drawing on multiple householdincome sources, as a strategy for long termfarm household survival (Shucksmith, etal., 1989). This challenged the conven-tional view in the United Kingdom, in con-trast to other European perspectives, thatsmall farms represented only a temporaryphase in the process of agricultural adjust-ment towards an agricultural sector basedon full-time “efficient” farm businesses.Following this logic, it might be arguedthat the conventional view of agriculture assupporting the rural economy has come tobe reversed to a situation where it is a suc-cessful local economy that offers the meansof support for pluriactive farm households.While it was recognised that pluriactivitywas not a new phenomenon, it gained anincreased policy relevance. However, asnoted by Gasson (1988) at the time, thegoals of rural development might be pur-sued more effectively by encouraging

tural policy approach. In this model, agri-culture represents the major sector in therural economy and its success determines theperformance of the local economy moregenerally. Agricultural decline promotesrural depopulation and a decline in ruralservice provision. Thus, a policy to stimu-late agricultural production not only supportsdomestic food supply, agricultural employ-ment and farm incomes, it also deters out-migration from rural areas and supports therural economy and service provision moregenerally. However in the mid 20th century,a variety of, by now familiar, factors under-mined this approach and the general con-sensus about the appropriate policies. Thehigh costs, inefficiency and environmentalimpacts of commodity price supports, espe-cially in the context of surpluses of agri-cultural products undermined the approachtaken to agricultural protection (Buckwell etal., 1982). The changing nature of technol-ogy applied in agriculture with increasingmechanisation and application of inputsimported from beyond the local economyreduced the local economic impact of agri-culture. The combined decline in the sig-nificance of the agricultural sector and thewidespread experience of counterurbanisa-tion has meant that agriculture plays anincreasingly less important role in the ruraleconomy. In the United Kingdom forinstance in 2006, agriculture was estimatedto contribute some 0,5% of total value addedat basic prices (Defra, 2007). But there issubstantial regional variation; agriculture’sshares in the English regions, varying in2004 between 0,02% for London, and 0,6%for the South East, to 1,3% in the East Mid-lands and 1,7% in the South West.

2. A multisectoral approachThus, support directed exclusively throughthe agricultural sector faced increasingexchequer costs in terms of dealing with theagricultural surpluses that can result fromincreased production and with the decliningrelative importance of agriculture within



307 COMPLET 7/11/08 9:18 Page 26

terms (Lowe and Ward, 2007) but particu-lar areas continue to suffer from problems oflow wages and underemployment. In areaswith low activity rates and high unemploy-ment, it may not matter very much whatsort of economic stimulus is introduced.Any sort of new activity can have multipliereffects that work through to other sectors andmay in turn promote new opportunities forfarm diversification, thus supporting thefarm population. In fact, it will often beeasier to create employment opportunitiesthrough the development of non-land basedactivities, either by encouraging the move-ment of new economic activity into the areaor through endogenous growth. The lattermay be seen as more sustainable, althoughthe former may be a more feasible alterna-tive in areas where the economy is especiallyundeveloped.

In other areas, economic change is char-acterised by a rather different pattern ofdevelopment, which we can term the “con-temporary” model of rural change (Hodge,1997); in contrast to the traditional modelthat is driven by changes within the agri-cultural sector. This recognises that a pro-portion of rural areas have a significantcomparative advantage leading to economicsuccess and population growth or coun-terurbanisation. This embraces a variety ofdifferent processes of varying importanceacross different localities. A major drivingforce behind it is the fact that rural areasoffer attractive environments in which tolive and work, while higher incomes andimproved transport infrastructure reducethe constraints on locational choices. Thusthose working in towns can travel longer dis-tances to work, increasing the level of com-muting. But the effect is more widespreadthan this; even relatively remote locationshave experienced population growth. Earlierretirement has freed up older people to livein attractive locations away from a placeof work. The increased congestion in urbanareas and improved road and rail networksoutside them have altered the relative acces-

employment completely unrelated to agri-culture.

3. A territorial approachHowever, even so, such an approach is onlypartially “multisectoral”. A truly multisec-toral approach to rural development policywould look more generally and equally atthe actual and potential roles for other sec-tors in rural areas. While located in ruralareas, these will often have no economiclinkages at all with agriculture. The focusthus shifts towards a more general analysisof conditions within particular types of area,or a territorial approach. And in practice, thismeans a focus on rural areas. Rural areas canoffer attractive locations for the establish-ment of new economic activity, often asso-ciated with the most advanced sectors of amodern economy, such as in informationtechnology, and many areas have gainedemployment from the establishment of newfirms and types of employment (Keeble andTyler, 1995; North, 1998). This reflects thegenerally declining significance of trans-port costs in industrial production, the attrac-tiveness of living in rural areas and the con-gestion costs of urban locations.

These socio-economic changes in ruralareas have been associated with the break-down of longstanding networks and link-ages, such as associated with the supply ofagricultural inputs and the marketing ofagricultural products. In a context of relativeagricultural decline the significance andpenetration of agricultural norms is dimin-ished within the wider community and thishas not been replaced by any alternativesingle dominant perspective. In practice,we can recognise rural areas in a variety ofdifferent circumstances and facing quitedifferent types of problem. But given thevariety of circumstances found in rural areas,we may then suggest that most generalisa-tions about the character of “rural” areaswill be wrong (Hodge and Monk, op. cit.).Rural areas in the United Kingdom generallyhave performed relatively well in economic



307 COMPLET 7/11/08 9:18 Page 27

the less remote rural areas. Different con-ditions in rural areas can also be associatedwith different types of problems. We can, forexample, identify two different sorts ofproblem associated with housing: poor hous-ing conditions as represented by over-crowding or lack of facilities, or problems ofaccess to housing as represented by a highlevel of housing costs relative to localincomes (Midgley et al., 2003). The differ-ent distributions of these indicators areshown in figure 2. Relatively high levelsof both indicators of disadvantage are foundin rural areas. But problems of housingaffordability tend to be concentrated in themore affluent south-eastern part of the coun-try around London, while problems of hous-ing conditions tend to be concentrated in themore remote rural areas. Thus they mightboth be seen as “rural” problems, but relat-ing to very different types of rural areas.

These issues suggest some limits to ageneral territorial approach, especially onethat distinguishes simply between urbanand rural areas. Changes in the circum-stances in rural areas indicate a higherdegree of complexity. There is no singlesector that can be seen as a source ofemployment growth across rural areas ingeneral. Rather, specific opportunities willdepend on local characteristics, especiallythe natural environment, such as landscape,topography or an attractive coastline. It mayalso depend on the presence of employmentclusters in nearby urban areas.

Other relationships also seem lessstraightforward. While it may have beenassumed that the maintenance of popula-tion numbers will provide for the mainte-nance in the provision of local services, thisno longer holds (Stockdale, 2004). Under the‘traditional’ model of rural decline, the levelof service provision falls with the reduceddemand associated with a declining popu-lation and the emphasis in debate has gen-erally been on the decline in services pro-vided in rural areas. But in practice manyother factors are associated with the level of

sibility of different types of locality; theless remote rural areas are generally moreaccessible than central urban locations thatsuffer from traffic congestion. Rural areasare also attractive to new forms of employ-ment, often based on entrepreneurs choos-ing to establish new businesses in placeswhere they want to live. Finally, there isanecdotal evidence of “downshifting”, peo-ple deciding to opt out of more stressfulemployment to take up a less pressuredlifestyle, often in a rural location. Thesehave different impacts on different groups ofthe population. For instance, those livingin rural areas tend to have higher incomelevels than those in urban areas, while thoseworking there often have lower levels.

Thus rural areas often follow divergentpaths, some in long term decline and othersexperiencing considerable prosperity. Somecontinue to be characterised by the ‘tradi-tional’ rural problems. Even if their popu-lations are not significantly declining, theycan often have low incomes and activityrates, although those on the lowest incomesare not necessarily engaged in the agricul-tural sector. Others with relatively highaverage incomes experience quite differentsorts of problems. While the majority ofthe population is often generally well off andcan get good access to services, there is aminority which experiences problems thatare in many ways a consequence of theaffluence of the majority, the fact that houseprices are high or that, because the majoritydo not demand certain services such as pub-lic transport, they are not provided at all.

This divergence of experience acrossrural areas is seen in various ways. Thehigher numbers of people in some areasdisguise the incidence of problems. Defra(2006) has recently highlighted the distri-bution of employees who are paid less thantwo-thirds of the English median wage.Concentration on the proportion of employ-ees who are low paid highlights the moreremote rural areas, but the absolute numbersof low paid employees are often higher in



307 COMPLET 7/11/08 9:18 Page 28

approach. In principle, resources need tobe directed towards particular problems atthe individual household or business level.This is clearly an impossible task for a cen-tral or federal government and indicates therequirement for decentralisation of deci-sion-making. But it may still not be feasiblefor a regional government and may demandan even more localised approach.

What is required is some mechanism forconnecting the objectives and resources thatare given for development policy at thenational level to the problems and prioritiesthat apply at the individual level. This isessentially a problem of information. Thecomplexity of the problems and the diminu-tion of traditional agricultural relationshipshave increased the attention given to therole of social capital and networks in thedelivery of rural development (Lee et al.,2005). There needs to be a system wherebylocal circumstances can be assessed againstnational priorities and information dissemi-nated to individual households and busi-nesses on the opportunities and resourcesthat can be made available in support of theobjectives. This will not occur at a single step

service provision relating to both supplyand demand. Economies of size and cen-tralisation in the supply of services,increased personal mobility, privatisationof service providers and altered patterns ofdemand have also led to major changes inthe way in which services are delivered.

The position is also complex when lookedat from the perspective of particular indi-viduals. An analysis of labour markets tendsto assume that the presence of unemploy-ment is a consequence of a lack of employ-ment opportunities within the local labourmarket, with the obvious policy implica-tion that the solution will lie in employ-ment creation. However, there is a variety offactors that can prevent individual access toemployment beyond a crude lack of vacantjobs (Hodge et al., 2002). These can includelack of transport, lack of childcare facilitiesor a mismatch between the types of jobsavailable and the skills of those withoutwork.

4. A local approachA response to these sorts of factors may beto adopt a “local” or even an “individual”



Figure 2. Access to housing and housing condition indicators in England

Source: Hodge and Monk (2004)

Map 1 - Rank of access to housing, lowest 20% Map 2 - Rank of housing conditions, lowest 20%

307 COMPLET 7/11/08 9:18 Page 29

zontal associations, such as land manage-ment co-operatives, while others developvertical associations, such as facilitation forthe implementation of policy. More attentionis needed on the optimal form and level ofadministrative intervention in the delivery ofrural development policies. This sort ofactivity falls between the conventional rolesof the public and private sectors, presentinga challenge to analysis that casts the two sec-tors in clearly separate roles. It introducesinvestment in and maintenance of socialcapital as legitimate elements of a ruraldevelopment policy.

Policy indicators and analysis

These changes in the nature and pattern ofrural development have profound implica-tions for rural analysis and policy evaluation.In the positivist tradition (Weimer, 1998)policy evaluation is undertaken to test theefficiency and effectiveness of specific pub-lic actions designed to achieve social welfarebenefits. For evaluation to work, therefore,policy objectives need to be unambiguouslystated, and causal mechanisms need to beclearly understood. The latter is particu-larly important since other events orprocesses rather than the policy itself mayaffect the outcome. Increasingly, therefore,and especially in the study of rural devel-opment, there has been a search for vali-dating measures, or indicators, which candiscriminate whether policy action has beenjustified.

Such indicators should, according to theEuropean Commission (CEC, 2001), coverefficiency (economic output in terms ofquality and quantity, competitiveness andviability, and institutional efficiency) andequity (viability of rural communities andthe maintenance of a balanced pattern ofdevelopment, access to resources, servicesand opportunities, and labour conditions).Further, to appreciate the range of compre-hension of different parts of the system andthe stages at which policies impact, differ-

and the ease with which it occurs at all willdepend on local institutions and the level ofsocial capital. A sectoral approach requiredlittle institutional development at the sub-national level. However, the move towardsa territorial, and especially to a localapproach, involves a much greater degree ofchoice and discretion in the ways in whichpublic resources might be applied. This com-plexity makes far greater demands on infor-mation and local institutional developmentsare required in order to handle it.

Experience with rural developmentschemes to date suggests that they can besuccessful in the development of institu-tions and social capital, especially as embod-ied in the organisations that have been devel-oped in order to facilitate the implementationof the schemes. Valuable initiatives havebeen made towards the development of localinstitutional structures through such schemesas Objective 5b and LEADER albeit in asporadic and piecemeal way (Ward andMcNicholas, 1998; Ray, 2000). But such ini-tiatives are very small relative to the totalvolume of support for rural areas that con-tinues to be put into rural areas through theCommon Agricultural Policy. Local insti-tutions have an important role in dealingwith the increasing complexity of policyimplementation by building social capital fordissemination of information, networkingamongst participants and co-ordination ofactivities.

A variety of institutional arrangementsand networks at the local level are involved,such as in public sector facilitation, byorganisations such as local authorities orNational Parks, development, housing andservice provision associations, collectivesupply associations for environmental goods,local dedicated environmental funds, orconservation trusts. Some of these are purelyin the public sector, such as local govern-ment facilitation. Others are essentially pri-vate, non-profit organisations, but gener-ally substantially supported throughgovernment funding. Some develop hori-



307 COMPLET 7/11/08 9:18 Page 30

urban areas. This might reflect either thecost of registering to claim the benefit whereit requires personal attendance in a localtown, or else where social norms may givegreater priority to independence and greatersocial stigma to claiming benefits from thestate.

A second type of problem relates to theway in which data are generally collectedand analysed in compiling indicators oflocal economic conditions (Fieldhouse andTye, 1996). Thus, the smallest statisticalunit within the Population Census is theenumeration district, the area covered by asingle enumerator. These districts are thenaggregated into larger statistical units onwhich the analysis is conducted. In urbanareas, groups of people with similar socio-economic characteristics tend to live in cer-tain localities. These are often large enoughto be identified as separate statistical units.However, within rural areas with smallersettlements, the unit will often include thewhole settlement and so households withlower income will tend to be includedtogether with those on higher incomes. Thusthe mean figure for the rural unit may wellfail to reveal the presence of a low incomepopulation.

The selection of indicators and the meth-ods that are used to analyse and evaluaterural development policy are clearly asso-ciated with the underlying model of therural development process and its objec-tives. Table 1 suggests the different indica-tors and methods that may be associatedwith the different rural development models.They also have different implications forthe sort of information collected and thepotential policy inferences.

The sectoral model concentrates on farmbusinesses and the means of raising farmincomes through agricultural production.Even where the emphasis has shifted fromincreasing production, there is clearly poten-tial for development by investing to reducecosts and rationalise farm production struc-tures. The methods of analysis draw partic-

ent kinds of indicators are required. Processindicators focus on policy implementation;output indicators provide quantitative meas-urements of effects identified as resultingfrom the policy; outcome indicators assessthe extent to which policies achieve theirstated objectives (Moxey et al., 1998).

Clearly, public resources for develop-ment assistance must be targeted on definedpriorities. But two types of problems areoften encountered in the targeting of ruraldevelopment areas (Midgley et al., op. cit.)The first results from an ‘urban’ character-isation of local economic problems. Whilethe approach has now changed, in the UnitedKingdom deprivation has in the past beenassessed against indicators measuring chil-dren in flats, Commonwealth immigrantsor overcrowded housing. None of these isrepresentative of rural problems. No accountwas taken of the availability of local serv-ices, often a particular rural concern. Evenan indicator of registered unemploymentmight be argued to be biased against ruralpriorities. In a large labour market, thosewho are unemployed can expect that regu-lar job search will lead to the identificationof a suitable employment opportunity. Incontrast, in a small labour market peoplewho are unemployed may well know thatsuitable vacancies are unlikely to occur andso decide to move to another area ratherthan remain unemployed within the localarea. This suggests that recorded unem-ployment might be lower because of out-migration. Further, it may be that the costsof registering as unemployed are higher ina rural area because of the distance to betravelled to the employment office and thepotential benefits lower as information mightbe more readily available by other, personalmeans. Thus we might expect that a ruralarea with a given level of economic disad-vantage would exhibit a lower level of reg-istered unemployment. This sort of argumentmight be generalised in that it is possible thatthe take up of social security benefits is onthe whole lower in rural areas than it is in



307 COMPLET 7/11/08 9:18 Page 31

remains quantitative and concentrates onquantifiable impacts and changes. The localmodel gives greater attention to the “softer”attributes of development. It seeks to recog-nise the variations in experiences amongsthouseholds and businesses within a partic-ular local area and the significance of socialand institutional capital in facilitating col-lective and community development. Thisindicates the introduction of qualitativeresearch techniques, case studies or dis-course analysis, and more deliberativeapproaches towards decision making. Thesedifferent models and methods have directimplications for the sorts of informationthat may be available for policy decisionsand hence for decision-making processes(table 1).

1. Approaches in the United KingdomDespite the contextual differences betweenthe constituent parts of the United King-dom, the articulation of policy and theframework of evaluation are relativelysimilar (perhaps because all four admin-istrations share a common civil service,

ularly on farm management but the approachclearly misses both the non-agriculturalpotentials for agricultural businesses andhouseholds as well as the conditions andopportunities in other sectors. The multi-sectoral approach recognises this wider eco-nomic environment and looks more gener-ally at indicators of the state of the economyas a whole and the interrelationshipsbetween sectors. However, in practice thefocus tended to remain on farm businessand households. Development is still inter-preted largely in terms of employment andso policy evaluation concentrates on thecosts of creating new employment oppor-tunities. This may suggest initiatives toattract new firms into the area or to stimu-late employment creation from the devel-opment of endogenous resources. The ter-ritorial model recognises the wider set ofsocial and environmental determinants ofhuman welfare beyond employment andservice provision. This suggests a cost-ben-efit approach that seeks to bring market andnon-market values together into a singleaccounting framework. The approach



Table 1. Indicators and methods in different development contexts

Indicators Indicative methods Implications

Sectoral Farm incomes Farm models Narrow focus missesAgricultural population Enterprise and significant determinants

commodity studies of rural welfare

Multisectoral Farm household income Household surveys May still be limitedEmployment Input-output analysis to agriculture sectorand unemployment Cost per job created Misses social andLocal value added environmental issuesEmployment incomes

Territorial Population change Cost-benefit analysis Misses variations inProportion of population incomes and welfarein disadvantage amongst populationAverage incomes and specific localLevels of service provision circumstances

Local Social indicators Case studies Capacity to consider fullNumbers of people in Qualitative analysis range of experiencesparticular circumstances Deliberative methods but problemsIndividual experiences with quantification

and aggregationHigh transactions costs

307 COMPLET 7/11/08 9:18 Page 32

its definition: “Evaluation is the processwhich objectively judges the actual out-comes, including any unintended sideeffects, of a policy or group of policiesagainst the policy objectives, or intendedoutcomes, and the resources that are usedin policy delivery”.

The planned evaluation framework con-sists of several streams: improving statisticalresources to establish a baseline for moni-toring; using this to assess progress usingthe Rural White Paper indicators, and also therural “Public Service Agreement” targets setfor Defra by the Treasury6; rural-proofingthe programme-based evaluation other West-minster ministries’ policies; a study of localrural services; and either a longitudinal studyof rural households or case studies of a num-ber of rural communities, to examine cross-cutting impacts of policies. When examiningthese policies themselves, however, thereare some challenging complexities. The focuson economic and social regeneration isdivided into two, sustaining the relative pros-perity of the majority of rural territory, andmore specific measures to address rural areaswith economic and social disadvantage. Mostof these consist of rural top-up funding forexisting economic development policies(skills, business support, broadband tech-nologies) delivered through other Ministriesor their agencies, and some minor regula-tory modification of the land use planningsystem. Improvement of the economic andenvironmental performance of farming andfood production is argued to be directly rel-evant to economic regeneration, althoughthe contribution it can actually make maybe small7. The new paradigm may be seen asmixing the territorial and local models, withthe more general territorial approach apply-ing across rural areas, but recognising the

and the cultural imprint of is strong). Thus,for example, in England prior to the out-break of foot-and-mouth disease, therewas a strategic review of the nature androle of rural economies (PIU, 1999), fol-lowed by a statement of rural policy pub-lished by the two responsible WestminsterMinistries (DETR/MAFF, 2000). Thisestablished the scope of rural policy, whichcovers fair access to rural service provi-sion, including housing and transport;business performance in both the farm andnon-farm sectors; rural conservation andleisure uses of the countryside; and thevitality of communities and rural civilsociety. Attached to these four prioritiesare a series of 15 indicators. For economicdevelopment, for example, performanceof policy initiatives has been measuredfrom employment activity rates and unem-ployment rates in rural areas, the propor-tions of market towns that are thriving,stable or declining (based on service pro-vision, business activity and employment),new business start ups and turnover ofbusinesses in rural areas, total incomefrom farming and off farm income, andlevels of agricultural employment5. Thissuggests a dominance of the multisectoralmodel in policy-thinking.

The consequences of the foot-and-mouth outbreak caused something of aparadigm shift (Scott et al., 2004), ini-tially in terms of perception of the relativeimportance of constituent parts of ruraleconomic activity, but perhaps more fun-damentally a recognition that the admin-istrative framework of policy delivery andevidence base was poorly suited to deliv-ery of the policy objectives. In addition toa streamlining and reorganization of ruralpolicy mechanisms, the new Rural Strategy(Defra, 2004b) provides a more detailedand comprehensive approach to policyevaluation, so that much of the introduc-tory discussion of this paper is reflected in



5. Ibid., p. 96.

6. This is to “reduce the gap in productivity betweenthe least well performing quartile of rural areas andthe English median by 2008, demonstrating progressby 2006, and improve the accessibility of servicesfor rural people” (HM Treasury, 2002, p. 111).

307 COMPLET 7/11/08 9:18 Page 33

Much more importance needs to attach toidentifying the specifics and spatial distri-butions of problems and their causes; butalso, it is necessary to reveal the causalprocesses that have the potential to resolvethe problems. As has been indicated, thismay well require novel developments inthe civil society of rural areas, but we havelittle systematic information on the rolesand impacts of networks and associations inimproving social and economic conditions.And we know less about how they may besuccessfully established and sustained.Analysis crosses the boundaries betweeneconomics and sociology. Quantitative infor-mation is required on economic activities,but a necessary complement is required inqualitative analysis of the influence of net-works, trusts or social norms.

In principle, a case study approach offersscope for development of an appropriate eval-uative strategy for rural policy. Rigorous indepth study of carefully selected local areas,using a mixture of quantitative and qualitativedata, can develop a sense of the interactionbetween increasingly diverse mixes of meas-ures in contrasting rural contexts where dif-ferent factors influence their expression andimpacts, and contribute to understanding ofhow and why they operate in the way they do.This centre of interest of multiple case stud-ies, described as a “quintain” by Stake (2006)is of a “contemporary phenomenon in a real-life context, especially when the boundariesbetween the phenomenon and context are notclearly evident” (Yin, 1994)9. Therefore, muchof the preceding discussion can be related toYin’s strategic recommendations for casestudy design and implementation. These beginwith selection and exploration of the objectsof study, on the basis of general suppositionsabout the impact of policy which require test-ing. Multiple evidence sources should bescrutinized to test rival hypotheses, whichmight provide alternative explanations. Com-mon protocols to investigate different expres-

potential for local variations in experiencesand the role of case studies and some degreeof decentralisation in decision-making. But itmay be argued that the approach towardsevaluation has not followed through theimplications of the changes taking place, andthere are inherent weaknesses in tracing thechain of causality from actions to impacts(Baslé, 2006).

The status of analysis,evaluation and decision-making

What may seem to be lacking from thisapproach is a revised conceptual frameworkthat fully recognises the changed and dif-ferentiated circumstances of rural locali-ties. In the context of a single dominantsector, support for this sector may well havetrickled down to the population more gen-erally, although even here there may bedoubts as to the extent to which such supportever did get to those who were most inneed. Contemporary rural change involvesmore complex interactions and interdepen-dences in highly diverse contexts, so that asSaraceno (1999)8 argues, policies should“make different assumptions about factorsinfluencing economic development and can-not be evaluated with the same tools thathave been developed for homogeneous, sin-gle administration, top-down policies.” Sta-tistical generalisation based on replicatedobservation of a large number of cases isunhelpful since it has to assume things to beequal, whereas in most cases they are not.



7. According to Defra’s own economic summary,regional agricultural gross valued added does notfully reflect the contribution of agriculture to ruraleconomies... (and) ... presents a number of com-plexities as neither sectoral nor area based indica-tors currently provide a good basis for capturing therural economy. Agricultural businesses account for16% of all businesses in rural areas, but they onlyaccount for 7% of both employees and turnover.Employees in rural businesses are more likely towork in the manufacturing (17%), tourism (8%)and retail (15%) industries.8. Cf. p., 440. 9. Cf. page 13.

307 COMPLET 7/11/08 9:18 Page 34

occupation, are additional conceptual toolsfor analysis and investigation.

In practice, though, significant barriersimpede the development and application ofsuch methods for improving understandingof the evolution, structure, and function ofrural economies for refinement of policydesign. Because case-studies require exam-ination of a great many variables, in detail,in a small number of cases, they are rela-tively expensive, and skilled evaluators arescarce. There is a risk of becoming over-whelmed by detail in mixed method evalu-ations conducted at local level, due to theirdiscursive nature. It is difficult to elaboratelocal level evaluation that fully reflects thecomplexity and diversity of rural areas, andat the same time convey the critical infor-mation back up to higher levels to permitbalanced and informed decisions to be takenabout resource allocation. Generalisationfrom case studies, especially from cross-case comparison where each individualstudy has been carried out in a consistentmanner, is possible, but involves a differentlogic to conventional induction. In eco-nomic analysis, acceptance and consequentadoption of case study approaches is farfrom widespread (Bitsch, 2000) becausethey do not allow for the familiar statisticalgeneralisations which come from large scalesurveys. In contrast, theoretical generalisa-tions deriving from identifying causaldependencies in one context contribute tobetter understanding of different mixes ofinfluences in other rural areas. Our ability tomake sense of different studies conducted incases selected for varying purposes (ofwhich an increasing number have now beencompleted: for example, Hart (2003); Lee etal. (2005); Midmore et al. (2004) is improv-ing as a result of evolving prescriptions forrigorous meta-evaluation techniques(Cooksy and Caracelli, 2005).

Responding to these challenges willrequire a trade-off between qualitative eval-uations to support decision-making at amore local level, closer to the level of pol-

sions of the phenomenon impart additionalrobustness. Analysis requires assessment ofdifferent patterns in the multiple data sourcesto refine and rule out competing hypotheses,both within individual case studies andbetween case studies carried out in differentcontexts (see, for example, Coffey and Atkin-son, 1996)10.

The increased complexity and differen-tiation of rural development also has impli-cations for the ways in which policy deci-sions may be made. Local diversity impliesthat decisions must vary at the local level,but an appropriate multi-level governancesystem for the administration of rural devel-opment undermines the traditional under-standing of effective sovereign govern-ments delivering policies and assessingtheir impacts. Differences exist in the oper-ation of the networks of interests whichhave arisen to bridge the lack of coordina-tion and consistency, overlapping with for-mal government structures and includingspecialist (and highly effective) interestgroups, and informal frameworks embodiedin conventions, each able to inhibit or facil-itate the actions of others (Morrison, 2006).The incidence of these, their effectivenessin addressing disadvantage, their impacts,and efficiency in deploying limitedresources and expertise are all poorly under-stood and require investigation. There is arisk that, rather than opening up opportu-nities to those who are excluded in presentcircumstances, they reinforce the influenceof particular interests (see for instanceYarwood’s, (2002) analysis of the operationof the rural exceptions policy and Shortall(2004)). Case study methods can contributeto understanding of what is analogous todiverse ecosystems of intersecting associ-ations and organisations, businesses, infra-structures, and environmental systems(Edwards, 2004). Extending this metaphor,interaction, duplication, and synergy ofrural civil society, and niche creation and



10. Especially Chapter 6.

307 COMPLET 7/11/08 9:18 Page 35

information back up to higher levels to per-mit balanced and informed decisions to betaken about resource allocation across dif-ferent regions and even countries.

ThanksThe authors are grateful for the helpful com-ments made by the referees.

icy implementation, and the need to passsome information on performance back upto higher levels in order to permit higherlevel resource allocation and financial con-trol. Perhaps this is the fundamental chal-lenge to combine local level evaluation thatfully reflects the complexity and diversity ofrural areas, and yet to convey the critical



Baslé M. (2006). Strengths and weaknessesof European Union policy evaluationmethods: ex-post evaluation of Objec-tive 2, 1994–99. Regional Studies, Vo.40,no 2, p. 225–235.

Bitsch V. (2000). Agricultural economics andqualitative research: incompatible para-digms? Forum Qualitative Sozialforschung(On-line Journal), Vo.1, no 1. Availableat: http://www.qualitative-research.net/fqs-texte/1-00/1-00bitsch-e.htm.

Bryden J. (2000). Is there a “New RuralPolicy”? Paper presented at EuropeanRural Policy at the Crossroads, ArkletonCentre for Rural Development Research,University of Aberdeen.

Buckwell A.-E., Harvey D.-R., ThomsonK.-J. and Parton K.-A. (1982). The Costsof the Common Agricultural Policy. Lon-don, Croom Helm.

Commission of the European Communities(1999). MEANS collection: evaluatingsocio-economic programmes. Luxem-bourg, Office for Official Publicationsof the European Communities, 6 Vol-umes.

Commission of the European Communities(2001). A framework for indicators for theeconomic and social dimensions of sus-tainable agriculture and rural develop-ment. Agriculture Directorate General,Brussels, available at http://europa.eu.int/comm/agr icul ture /publ i / repor ts /sustain/index_en.pdf.

Champion A. (1994). Population changeand migration in Britain since 1981: evi-

dence for continuing deconcentration,Environment and Planning A, Vol.26,no10, p. 1501–1520.

Cloke P., Milbourne P., Thomas C. (1994).Lifestyles in rural England. Salisbury,Research Report, Rural DevelopmentCommission, Vol. 18.

Coffey A., Atkinson P. (1996). MakingSense of Qualitative Data. ThousandOaks, CA: Sage.

Committee on Land Utilization in RuralAreas (1942). London, Report of theCommittee on Land Utilization in RuralAreas. Cmd 6378, HMSO.

Cooksy L.-J., Caracelli V.-J. (2005). Quality,context, and use: issues in achieving thegoals of metaevaluation. American Jour-nal of Evaluation, Vol. 26, no1, p. 31-42.

Defra (2004a). Social and economic changeand diversity in rural England. DefraPublications, London Birkbeck College,University of London, A report by theRural Evidence Research Centre.

Defra (2004b). Rural Strategy 2004. Lon-don, Stationery Office.

Defra (2006). Rural Development Pro-gramme for England 2007-2013. Lon-don, Department for Environment, Foodand Rural Affairs, Consultation.

Defra (2007). Agriculture in the UnitedKingdom. London, Department for Envi-ronment, Food and Rural Affairs.

DETR/MAFF (2000). Our Countryside: thefuture. A fair deal for rural England.London, Cm 4909, Stationery Office.

Dwyer J., Ward N., Lowe P., Baldock D.

REFERENCES

307 COMPLET 7/11/08 9:18 Page 36

capital and identities in European ruraldevelopment. Sociologia Ruralis, Vol.45,no 4, p. 269-283.

Leon Y. (2005). Rural development inEurope: a research frontier for agricul-tural economists. European Review ofAgricultural Economics, Vol.32, no 3,p. 301-317.

Lowe P., Ward N. (2007). Sustainable ruraleconomies: some lessons from the Eng-lish experience. Sustainable Develop-ment, Vol.15, no 5, p. 307-317.

Midgley J., Hodge I., Monk S. (2000).Developing Indicators of Rural Disad-vantage: Stage II. University of Cam-bridge, Department of Land Economy,Final report to the Countryside Agency.

Midgley J., Hodge I., Monk S. (2003). Pat-terns and concentrations of disadvantagein England: A rural-urban perspective.Urban Studies, Vol.40, no 8, p. 1427-1454.

Midmore P., Foster C., Schermer M. (2004).Organic Producer Initiatives and RuralDevelopment: Four European Case Stud-ies. University of Wales Aberystwyth,School of Management and Business,Organic Marketing and Rural Develop-ment Series, Vol. 3.

Morrison T.-H. (2006). Pursuing rural sus-tainability at the regional level: key les-sons from the literature on institutions,integration, and the environment. Journalof Planning Literature, Vol.21, no 2, p.143-152.

Moxey A., Whitby M., Lowe P. (1998).Agri-environmental indicators: issues andchoices. Land Use Policy, Vol.15, no 4, p.265-269.

Newby H. (1988). The Countryside in Ques-tion. London, Hutchinson.

North D. (1998). Rural industrialization. InB. Ilbery (Ed.), “The Geography of RuralChange”, Harlow, Longman, Chapter 8.

OECD (2006). The New Rural Paradigm:Policies and Governance. Paris, Organ-isation for Economic Co-operation andDevelopment.

(2007). European rural developmentunder the Common Agricultural Policy’s“Second Pillar”: Institutional conser-vatism and innovation. Regional Stud-ies, Vol.41, no7, p. 873-887.

Edwards M. (2004). Civil Society. Cam-bridge, Polity Press.

Fieldhouse E.-A., Tye R. (1996). Deprivedpeople or deprived places? Exploring theecological fallacy in studies of deprivationwith the samples of Anonymised Records.Environment and Planning A, Vol.28,no 2, p. 237-259.

Gasson R. (1988). Farm diversification andrural development. Journal of Agricul-tural Economics, Vol. 39, no 2, p. 175-182.

Grigg D. (1982). The Dynamics of Agri-cultural Change. London, Hutchinson.

Hart K. (2003). Decentralized developmentin the European Union. Progress inDevelopment Studies, Vol.3, no 1, p.59–64.

HM Treasury (2002). Opportunity and secu-rity for all: investing in an enterprising,fairer Britain. London, Stationery Office,Public Spending Plans 2003-2006, Cm5570.

HM Treasury (2003). The green book:appraisal and evaluation in central gov-ernment. London, Stationery Office.

Hodge I. (1997). The integration of the ruraleconomy. Built Environment, Vol.23,no 3, p. 192-200.

Hodge I., Monk S. (2004). The economicdiversity of rural England: stylised falla-cies and uncertain evidence. Journal ofRural Studies, Vol.20, no 3, p. 263-272.

Hodge I., Dunn J., Monk S., Fitzgerald M.(2002). Barriers to participation in resid-ual rural labour markets. Work, Employ-ment and Society, Vol.16, no 3, p. 455-474.

Keeble D., Tyler, P. (1995). Enterprisingbehaviour and the Urban-Rural Shift.Urban Studies, Vol.32, no 6, p. 975-997.

Lee J., Árnason A., Nightingale A., Shuck-smith D.-M. (2005). Networking: social



307 COMPLET 7/11/08 9:18 Page 37

community consequences and individualmigrant experiences. Sociologia Ruralis,Vol.44, no 2, p. 167-194.

Van der Ploeg J., Renting H., Brunori G.,Knickel K., Mannion J., Marsden T., deRoest K., Sevilla-Guzmán E., Ventura, F.(2000). Rural development: from prac-tices and policies towards theories. Soci-ologia Ruralis, Vol.40, no 4, p. 391-408.

Ward N., McNicholas K. (1998). Objec-tive 5b of the structural funds and ruraldevelopment in Britain. Regional Stud-ies, Vol.32, no 4, p. 369-374.

Ward N., Lowe P., Bridges T. (2003). Ruraland regional development: the role of theregional development agencies in Eng-land. Regional Studies, Vol.37, no 2,p. 201-214.

Weimer D.-L. (1998). Policy analysis andevidence: a craft perspective. Policy Stud-ies Journal, 26(1), p. 114-128.

Yarwood, R. (2002). Parish councils, part-nership and governance: the developmentof “exceptions” housing in the MalvernHills District, England. Journal of RuralStudies, Vol.18, no 3, p. 275-291.

Yarwood R. (2005). Beyond the rural idyll:images, countryside change and geog-raphy. Geography, Spring, Vol.90,p. 19-31.

Yin R.-K. (1994). Case Study Research.CA, Sage, Thousand Oaks.

PIU (1999). Rural Economies. London, Sta-tionery Office, Performance and Inno-vation Unit.

Ray C. (2000). The EU LEADER pro-gramme: Rural development laboratory.Sociologia Ruralis, Vol.40, no 2, p. 163-171.

Robert P., Randolph W.-G. (1983). Beyonddecentralisation: the evolution of the pop-ulation distribution in England and Wales1961–1981. Geoforum, Vol.14, no 1,p. 75-102.

Saraceno E. (1999). The evaluation of localpolicy making in Europe: learning fromthe LEADER Community Initiative.Evaluation, Vol.5, no 4, p. 439–457.

Scott A., Christie M., Midmore P. (2004).Impact of the 2001 foot-and-mouth dis-ease outbreak in Britain: implicationsfor rural studies. Journal of Rural Stud-ies, Vol.20, no 1, p. 1-14.

Shortall S. (2004). Time to re-think ruraldevelopment. EuroChoices, Vol.3, no 2, p.34-39.

Shucksmith D.-M., Bryden J., Rosenthall P.,Short C., Winter D.M. (1989). Pluriac-tivity, farm structures and rural change.Journal of Agricultural Economics,Vol.40, no 3, p. 345-360.

Stake R.-E. (2006). Multiple Case StudyAnalysis. New York, The Guilford Press.

Stockdale A. (2004). Rural out-migration:



307 COMPLET 7/11/08 9:18 Page 38

Documents

Models of Rural Development and Approaches To Analysis