European Integration and Europeanization processes in welfaresystems:

a comparative analysis with typological purposeof Gabriella Punziano

European Integration and Europeanization processes, who have hadtheir engine in political and economic instances, but with directeffect on social field, are the focus of the analysis hereproposed. In facts, starting from the social implications and fromthe effects that the different European welfare regimes have onthem, it will be shown a possible re-arrangement and thematizationof the differences within the various regimes, as well as of theirrelationships and their new characterizations.The starting point of the proposed reasoning is found in a

previous study conducted for my doctoral dissertation: UniqueEuropean Welfare or local net welfares? Decision-making process in the social spherebetween convergence and autonomy (it. Welfare Europeo o welfare locali? I processidecisionali nel sociale tra convergenza ed autonomia, Diogene Edizioni, Napoli,2012). With this study it was intended to deepen the dual thrustbetween Europeanization and decentralization of social policy insome European countries, which are considered as a guidance ofspecific welfare regimes (Esping-Andersen 1996), in order tounderstand at what level of governance the decisions that shapethis area can be attributed. This objective was pursued through acomparative geographic and policy analysis based on a mixed methods approachintended as a merge of standard and non-standard approaches,techniques and tools. Effectively, starting from the cases ofMilan, Naples and Berlin, the above mentioned analysis contemplatesjointly the study of supranational, national and subnationaldirections, but also the study of local dynamics concerning thespread of specific models of implementation, decision of socialpolicies and regimes that these decision generates. This kind ofmethodological structure is growing both in macro perspective(through multivariate and multi-level analysis of Eurostat and the resultingcluster analysis) as well as in the micro perspective (through theanalysis of projects in different local contexts and in particularby applying impact, implementation and comparative network analysis).The conclusion was the realization of a general model ofinterpretation and classification of the changes occurred in thedifferent European welfare regimes (to be tested further more).However, while the analysis of original project involved five

Nations (Italy, Germany, France, Spain and United Kingdom) and tenlocal contexts, two for each Nation (Milan and Naples, Berlin andMunich, Paris and Rouen, Barcelona and Vigo, London and Liverpool),

1

the analysis hereby proposed aims to pass the limits of theprevious study, enlarging the comparative basis of the analyzedNations and starting the statistical survey in other countries andlocal contexts (i.e. Scandinavian or Eastern European Countries).Below, we will present (together with a recovery of the theoreticaland methodological requirements) the main empirical results of theresearch.

1. Theoretical and methodological requirements

The concept of Europeanization, in the middle of numerous studieson the changes in welfare regimes and on the dynamics of governancein the European Union (Lebfried, Pierson, 1995; Le Gales, 2002;Gullién, Palier, 2004; Giuliani, 2004; Ferrera, 2006; Naldini,2007; De Leonardis, 2012; Kazepov, 2009-2010) can be considered ina multiple meaning (acting differently) on the respective operativecontexts on which it has an influence. In the relations between EU institutions and member states, this

concept concerns the European integration process and thedevelopment of multilevel governance aimed to a full convergencetowards a unitary system of government of social policies approvedthroughout the EU and based essentially on the principles ofsubsidiarity and activation (Bifulco, 2005). This acts on thesingle national political systems, which, in turn, respond byshowing different degrees of intensity and adaptation in complyingwith the scenario outlined. If we consider the same concept starting from the local contexts

(to be considered as different centers of branch of the socialgovernance in a decentralized manner, rather than starting from thecore of community interest - the European supranational actor) sowe have the dynamics of implementation and a differential andcontextual process, which are the expression of local autonomy inthe social policies process of governance, undertaken in many waysand addressed to the convergence (Radaelli, 2003).In addition to the concept of Europeanization, we have also the

issue of European integration which, according to Giuliani (2004),refers to something extremely normative, regulatory and proceduraland capture macro dynamics and relations in the system, while withthe concept of Europeanization we refer explicitly to microdynamics contextualized as silent, differential and oftencontradictory process which can be found in the governanceprocesses and in which the actors decide and give a shape to the

2

different emerging structures (Börzel, 2004). The Europeanizationcan be understood as a process through which the European Unionstructures, procedures and policies are specialize and automatethemselves from the national level, reflecting the more generalprocess of institutionalization of the European Union (Mörth, 2003)and of its structures of governance. In facts, theinstitutionalization becomes the hallmark and also the point ofunion between a conception of community integration, withregulatory meaning, and a conception of Europeanization, withprocedural meaning. This link becomes clear especially in thetransition from an hard to a soft law legislation in the field ofsocial policies between the different levels of governance(supranational, national and local). The process of Europeanizationfocuses on strategic convergence towards full EU integration whichis pursued, however, through decentralized mechanisms, subsidiarityand open method of coordination that is aimed to revalue theterritorial dimension of social policy. The wished integration(which is the basis of this process) will be realized essentiallyin two directions: an integration directed to the context (moreoften identified into social and territorial cohesion policy) whichaims to reevaluate it and make it competitive and able to face theglobal challenges, and a kind of integration facing to the person(identified with social inclusion policies), which aims, instead,to guarantee common standards of living and to invest onindividuals for making them an active part of the inclusion andparticipatory process (place and people, Donzelot, 2003). Twodifferent ways of conceiving the process of integration are notalways fully compatible because of the scarcity of resources toinvest in the social as well as the involvement of differentstakeholders that this process can implicate. The factor thatdrives these forces is to be found in the ability of contexts,either national or local, to move towards a full EU integration(unique European welfare system) or diverge from it (local net welfaresystems). Therefore, the EU integration variable becomes thediscriminant in reconsidering (through a comparative study) adifferent typology of welfare regimes built on the gradualintegration, the differential convergence strategy and the recoveryof implementing autonomous spaces: factors that push to a unitarysystem or to multiple local systems. This kind of typology is morefocused on the role played by the European regulatory level asmanagerial and addressing actor of welfare policies (Punziano,2012), far from the classical conceptions based on the ownership ofthe right to provide social intervention, on the more or less

3

pervasive intervention of the National State, spending levels,access requirements, the coverage funding, the recipients or thecriteria for the award of disbursement, as well as on the mix ofactors involved in local governance (Titmuss, 1974; Esping-Anderson, 1990; Ferrera, 1994/2006; Naldini, 2007). That isbecause, in this historical moment, in a generalized way, thewelfare systems was passed from monetary disbursement to theprovision of services, from dependency to activation, from self-exclusion to self-employment, strengthening of capabilities andcompetitive capacity first of individuals and then of the systemthat they make up (Sen, 1993), from a straight and pervasive actionof the National State to the emergence of other social actors(Third Sector, Family, the private social, etc.)1, more oftendefined welfare mix (Ferrera, 2006). This process oftransformation, supported by demographic, social and economicchanges in the member Countries, as well as by the passage from thelaw-making of exclusive competence of the State in the social fieldto an orientated legislative agency entrusted to the bodies of theUnion (also detected in the passage from soft to hard lawlegislation), which led the welfare states to become a consolidatedimension of the solidarity and sharing of risk system (Beck,2000).One wonders, therefore, how the decisions are taken in the social

field and from which point they spread. Thus outlined, thisquestion concerns several levels of analysis and more analyticobjects that require, for their conjunction, an integratedapproach, not necessarily convergent, and which allows toinvestigate the field of the analysis both in macro and in microperspective. This approach is represented by Mixed Methods(Tashakkori, Teddlie, 2003; Teddlie, Yu, 2007; Creswell, 2012); anintegrated approach of methods, techniques and tools aiming to theelaboration of an interpretive tool and cognitive tool that becomethe result, pursuing different strategies both on the standard andnon-standard side. The mechanism method which is essential for theintegration in this kind of mix strategy has been identified in thetranslation of both languages, standard and non-standard, and in acommon code. The result of it is the construction of the generalmodel of classification, the main result of the analysis conducted,in which every result that emerged from the different analyticalstep has been transformed into characteristics and attributes thatsubstantiate the different emerged types and they can livetogether, in this way, without particular ontological and

1 Cfr. Paci, 2003; Ferrera, 2006; Bifulco, 2005; Naldini, 2007; Kazepov, 2009.4

epistemological issues, in a flexible, dynamic and ever changingmulti-methods model, because it integrates the change in the systemunder investigation as a systemic variable. This integratedapproach don’t aspire to bring into question the efficacy ofclassic typology, but intends to offer new opportunities to bringout details unmanageable or undetectable when the realityinvestigated is vast for extension, history and cultural heritage(as in the case of Europe), but even more when what we want is tobreak up this reality and put it in a comparative design. Toapproach this complexity, the pattern of analysis used was dividedinto sequential steps that will be hereby described and that arebuilt one inside the other as a sub-designs within the overall morecomplex design (nested, Crassewell, 2003), which takes the name ofComplex Mixed Methods Design (Punziano, 2012). Postponing thediscussion elsewhere on the mathematical and statistical technicaldetails of the model proposed (Punziano, 2012), we discussedherewith the main objectives and results achieved.

2. The typological axes

Since the initial phase of the study, the first step of analysis(aimed to the emergence of the axes underlying the typology in thegeneral model of classification) was characterized by the recoveryof a policy comparative approach which didn’t replace but integratedgeographical comparisons. These two method characterizations are goingto fit on the delimitation of the typological axes binding to thepossibility of covering the two plans, semantic and spatial, intothe multiple levels reality considered in this study. A semanticcontinuum (vertical axis), defined as the EU Integrational axis andgenerated in the opposition between the two poles of development,social inclusion and territorial cohesion, in a methodologicalapproach centered, precisely, on the policy comparison. A spatialcontinuum (horizontal axis), which is based, instead, on thegeographical comparisons, impregnated by developments and byevolutionary dynamics of welfare systems (Europeanization vs.Decentralization, Convergence vs. local Autonomy) that will resultin a stretching of an opposition between a type welfare that can bedefined as a unitary European and convergent welfare system, and a kind offragmented and local welfare, defined as net local welfare systems. The axesaim, therefore, to extrapolate the effect of the method (comparativepolicy analysis vs. comparative geographical analysis), thecontinuum intersecting (semantic vs. spatial) and the main changes

5

occurring in social policies (Europeanization vs.Decentralization/fragmentation; activation vs. welfarism;convergence vs. local autonomy, center vs. suburbs own ship ofdecision-making process in social policy), assuming, on one hand,a separation in terms of local or social development (cohesion vs.inclusion) and, on the other, the trend towards Europeanization ordecentralization/fragmentation that can be seen like general way ofdoing not necessarily exclusive one with the other; particularforms of merging and mixture that lead to the emergence ofparticular systems that have to be treated as a model and not asabsolute and generalizable classifications. A typology incontinuous evolution that becomes the integrated auspicatedanalytical tools that is able to accommodate different and highlyheterogeneous elements for nature and origin (fig. 1). Although this typology shows in this phase all its theoretical

nature, this step was also tested in the second phase of the studywith the formalizations of a macro model generated on the use ofsocial indicators at territorial level. Whereas the aims of thework, as just mentioned, is to test and complete a new model forthe classification of European welfare systems by introducing newdiscriminant variables, following the discussed argumentations wecan highlight two essential discriminant variables that correspondto the emergent axes: EU Integration and Europeanization. This canbe seen as a polarization between European integration like focusedon legislative dynamics and Europeanization as the mirror ofprocedural factors and practices concerning the integration thatallows us in understanding and classified the change occurred inthe investigated welfare regimes with the category of convergenceand autonomy. Therefore, we will have: - An EU integration axis based on the double-side of cohesion policies,aimed at territorial connoted development and directed to thecontexts (place), and of inclusion policies, aimed at sociallyconnoted development and addressed to the subject (people). It isthe semantic continuum based on the policy comparison.

- An Europeanization axis based on double-side convergence to theEuropean welfare and autonomy of local net welfare (the spatialcontinuum based on the geographical comparison).Crossing the two axes and joining the theoretical and the

empirical plans (test that will discussed in the followingparagraph with the application of principal component analysis onthe Eurostat and OECD indicators – to the delineation of the twocomponents and for the intersecting of the axes – and PLS PathModeling – for the definition of polarity in opposition) what we

6

obtain is typological space with all the connotation for reading init the different ways in which the welfare regimes can be dividedby introducing these two new discriminant variables.

Figure 1: Typological axes originated in the model.

3. Macro Analysis on social indicators: which position for the nations and localcontexts?

The second step of investigation is focused on the multivariateanalysis (Principal Component Analysis – Pearson, 1901, Hotelling,1933) and multi-level (joint analysis of variation levels – Hox,2002) with the formalizations of a macro model generated on the useof social indicators at territorial level. This objective will be pursued by the creation of six indexes

starting from six concepts. Two of them are the main concepts usedto create the typological space and for this reason they weredefined also discriminant variables: EU Integration – a concept ofregulatory nature – and Europeanization – a concept of procedural

7

nature. These two concepts are based on a continuum line with twopolarizations. The first concept, the EU Integration, is used to describe the

opposition between policy data and results concerning integration,with social connotation (Social Inclusion constructed as a secondlevel index), and data concerning policy and result that look atplace development, with territorial connotation but together todirect social impact (Territorial Cohesion constructed at the sameway as a second level index). The second concept is Europeanization and it is focused on the

opposition between indicators that look, on one hand, to theprocesses of convergence (legislative, procedural and social) and,on the other hand, to the processes of empowerment of localcontexts (in expression of local autonomy). So, the polarizationsused to construct the Europeanization index are Convergence Factors(Convergence index as second level index) and Expression ofAutonomy (Autonomy index as second level index). The involved analysis born in a comparative perspective in order

to build synthetic indices of performance of the different contextsinvestigated with the intent to:- Highlight the incisive decision-making of national or sub-national level (multiple regression models and for blocks – PLS-Path Modeling, Edwards Vinzi, 2009 – on disjointed levels beforenational and then regional) in the constitution of socialpolicies in Europe;- Establish a plan for the selection of these contexts on thebasis of assumed differences on the discriminant variables;

- Draw, from these analyzes, the elements for projected space intothe typological obtained space generated crossing the axes orNations (NUTS02) and Regions (NUTS23) (classification based onRebus-PM – Esposito Vinzi et al., 2008 - and Cluster Analysis).The question from which we started was, starting from classic

typology, what/how much has changed in European welfare regimes?The analysis confirmed the composition of the axes identified attheoretical level supporting the hypothesis of a linear function ofintegration and Europeanization respect to the double highlightpolarizations (fig. 2). This configuration is suitable to be takenas the dividing line between welfare systems. What follows is thatthe models classically understood as described by Ferrera (1998)begin to mingle and merge one with the others. The distinction2 Nomenclature of Territorial Statistics Unit, in acronym NUTS (from the Frenchnomenclature des unités territoriales statistique) that identifies the division of the territory of theEuropean Union for statistical purposes (national level of aggregation).3 Cfr.note 3 (level of regional aggregation).

8

point of the models moves rather than on differentiating factors, thatwas fundamental in a time where prevailing a soft law legislationon social field, toward those of convergence, which are relevant ina time when Europe begins to establish and delineate precise andbinding trajectories with respect to social growth. This is clearin the territorial component in which pressures to Europeanizationor to localization, leading to stretching two other trends, on theone hand we have a Unitarian European welfare system, that can also becalled Europeanization welfare, based on the full convergence of nationson a single and integrated model, on the other hand we have thebirth of many fragmented local net-welfare systems like small centers ofdecisional gravitation from which start the impetus for theempowerment of the different contexts, while maintaining balancedthe fundamental objective of convergence in the economic and socialdevelopment results achieved.

Fig. 2: PLS-PM: a) regional level only on Eu integration; b) national level onlyon Eu integration model restrict; c) national level only on Eu integration fullmodel; d) national level only on Europeanization model restrict.

9

a

d)

c)

b)

If we look at the national level, the position of nations is amirror of a convergence driven by a legislative principle ratherthan a spontaneous and dynamical action of the local network.Considering, instead, the regional level, it is not difficult tosee the typical territorial dimension that acts on theimplementation of social policies and shows in the recovery oflarge areas of local autonomy. In the first case, it would seem toprevail the idea of moving towards a Unitarian European Welfare System,which, with the exception of the Scandinavian Countries already nowheavily aimed at the systematic convergence, it is what wasregistered in Germany, France and Great Britain, which has promptedthe selection of these contexts for the development of the micromodel described in the next paragraph. In the second case, however,this trend is fully rejected by the emergence of Local Net WelfareSystems, different within them self and not comparable orattributable with a general model. It is this dynamics that prevailin countries such as Italy and Spain, also chosen for the micromodel since returning from the crisis, or even in countries such asGreece, for the same reason, or countries of Eastern and CentralEurope of recent entry into the Community with them economies andsocial dynamics that are still in trouble with respect to requestsexpress by the supranational actor. Therefore, although it has beenused for the macro model analysis purely quantitative techniques,in the stage of selection of the contexts for the micro analysis,the choice becomes rational and this is made to try to explain thereal differences in the levels of performance achieved by specificcountries and regions that reveal clearly the mix of classicwelfare regimes.In the end, looking at a jointly model that merge national and

regional level (multilevel model) for understand what of these twolevel have most weight in the decision-making process for socialpolicies, what is evident is that however the nations tend toconvergence rather than local contexts, is in these lasts that theadaptive thrust concrete achieve, sometimes not perfectly in linewith national trends. So, the major weight in act for social and inthe social fields is recovered by region, and is in the convergentand divergent performance on these recorded that we can found a

10

response to our opening question. Therefore we have to distinguishbetween an integration of shape and an integration of substance, ofmanagement or of address, and so because the European Union is allin line with the principle of integration, however, it is in itsvarious nations and regions that it assumes different characters.It is emerging the subjectivity of the territories (Kazepov, 2009),their being actors in oneself and the power that the dynamics thatpervade them have to engrave it on the addresses dictated by theactors at higher levels, but from which the territories can’t breakdown because they are the basis of multilevel governance.From the application of PCA and Cluster Analysis another think

become more evident and deserves a little discussion first ofpresenting the projection of group and nations in the typologicalspace: the consolidation of the difference between group both froma legislative and regulatory plan (European integration) then froma procedural point of view (the Europeanization process). Theseemerging difference can be read as different way of convergencesometime driven by the national level (fig. 3, such as in the caseof Germany, Sweden, France), where it don’t pass great differencebetween the national average and the averages of the regions in thenation, and sometimes driven by the local level (fig. 4, such as inthe case of Italy, Spain, Greece and Eastern European Country)where exists a great difference and dispersion between the nationaland the regionals average.

Fig. 3: PCA on European Nation (a. on Oecd dataset – 21 nations; b. on Eusilcdataset – 27 nations).

11

Fig. 3: PCA on European Region (Nut2 level).

12

a

b)

The main result obtained, in the aim of the integration of theresults of each step of analysis in the general model ofclassification, is the provision of nations and regions withrespect to the EU Integration and Europeanization index jointly tothe emersion of the resulting groups (clusters) that this entails,which were projected as features and attributes within thequadrants emerged from the intersection of the axes. This completesthe space of attributes and allows us in delineated four distinctmodes of integration.

Fig. 5: Projection of nations and group of nations in the typological space.

13

The modality of integration for each quadrants and the

arrangement of the groups of nations are outlined and assumedifferential characteristics. The nation's groups take over thegeo-political connotation and lead to the definition of specificway of reading the emerging dimensions. Inclusion and cohesion, forexample, take on the meaning of breathlessness in addressing socialdevelopment, on the one hand, and territorial development, on theother hand, when referring to more poor and undeveloped countries.In contrast, these same dimensions take the connotation of humandevelopment and competitiveness when they are associated withcountries with more stable and developed economies. It isinteresting to see, at this regard, as the Eastern and minor Europeancountries prove clearly split into two groups that recover historicaland cultural heritage and come to form a new continuum that crossquadrants: independence vs. dependence, revenge of their specificagainst cultural orbiting and approval. We are talking aboutcountries like Ireland, Czech Republic, Slovakia, Malta, Cyprus andSlovenia, which, although near to powerful centers of influence(e.g. UK to Ireland or Russia for the Czech Republic that is themost European nations of the boundary Russian countries), however,have asserted their uniqueness and promoted their specificity moreterritorial then social (thinking about tourism competitiveness forMalta and Cyprus for example). The quadrant in which they occur isborn from a cross between a marked tendency to focus on policies ofterritorial cohesion and a strategy based on the recovery of spaces ofautonomy that leaving clear glimpse the membership to the local net

14

Indipendence vs.

Dipendence

welfare systems. The strategy of integration that characterizes thisquadrant is defined Development as a strategy for growth in newforms of competitiveness. Different is the group of Pure EasternEuropean Countries (Latvia, Bulgaria, Romania, Hungary, Estonia,Lithuania and Poland) under the aegis of the Russian thatcharacterized the levels of development of these countries.Distant, socially and economically, from the European model, thesecountries are currently working to reinvent strategy of recovery ofsocial gap strongly felt that puts them in trouble with respect tothe expressed needed of development also on territorial andeconomic side, on which, however, in differential way in each localcontexts, appear to be climbing rapidly. This explains the positionin the quadrant that crosses stretch out towards social inclusionpolicies in conjunction with recovery of spaces of autonomy as wellas in a local net welfare. Here the integration strategies used is theone defined as Emergence of contextual specificity and states ofbackwardness which involve the need for a strong recovery of thesocial gap still persists, precisely according to the culturalheritage which acts with its strong influence. Not surprisingly,however, in a rather cohesive and close way you find theMediterranean countries affected by crisis (Greece, Spain, Italy andPortugal), and then united by a marked slowdown of the system thatleads them to decline in the levels of performance and ranking ofdevelopment. Those countries that characterize a specific quadrant,are placed on the average with respect to the axis of integration,so moving markedly on either side of cohesion or of inclusion,while on the Europeanization axes they stretch, although not in anuniform way (which will be explained in the analysis of the casesof Milan and Naples) towards autonomy and differentiation of localregimes, also in this case identifiable with the local net welfaresystems. At the level of social performance recorded by the analyzedindicators, these countries are in sharp decline, which, ratherthan bringing these countries to focus on integration strategiesaimed at developing competitiveness or on emerging of somespecificity, drives them to become the emblem of the consequencesof the economic crisis and of its impact on the weakening of thesocial protection system, moving to a decisive turning thesecountries that is the need of recovery both economic as socialgaps. Italy and Spain, as countries identified for theinvestigation micro, lead one on the slope of the Emergenceintended as a recovery of social problems greatly inflated by theeconomic crisis (Italy), and the other on the side of Development,understood as focusing on competitiveness, enhancing on the

15

empowerment of territory in a scenario in which was the socialcrisis that has led to the emergence of economic-contextualcontradictions (Spain). There is, then, the group of Social-Democraticand pro-Scandinavian Countries (Austria, Finland, Denmark, Sweden,Luxembourg and the Netherlands) who are currently living a form ofsystemic Stabilization with respect to the welfare that was made of upward trend over time in social performance levels and acorresponding possibility of joint growth both from the point ofview of social and from the point of view of competitiveness andterritorial. The quadrant in which are inserted is establishing astrong integrative strategy defined Stabilization. This strategyreflects nothing more than the achievement of a welfare regimestrongly characterized by peculiarities and distinctive features,which don’t have to recover backwardness but that pushes withdecision towards the reinforcement of its specificity, attractingto whether nations orbiting in decisive mechanisms of convergenceat the European welfare state which strongly advocates thesupranational actor (the United Kingdom is to be drawn). Finally,there is the group of Continental and Liberal European Countries (Belgium,France, Germany and the UK) that point to a tight systemicconfluence, not to the functional Scandinavian model but toward amodel with all European specificity, often identified with what hasbeen called the European Social Model ( Bifulco, 2005) and that inreality the last few years has shown tends to its utopian characterwhen we try to decline it on vastly different contexts for thecontradiction in term that see it focused on principles ofdifferential revaluation, subsidiarity and open method ofcoordination. The quadrant in which these nations fall is formedfrom the intersection of integration that is based on cleardevelopment strategies in the sense of territorial cohesion, provisionof services and increase competitiveness, as well as the dynamicsof convergence understood in the sense of integrative strategy ofConfluence toward regulatory dynamic and procedural practices thatbring directly to the creation of a Unitarian European Welfare System. IfGermany and France are linked without doubt to this quadrant, theUnited Kingdom is released little by this group of countries towhich it belongs since to stretch out with the countries in theside of Stabilization.

16

4. Micro-model on local contexts and projects analysis: spaces of autonomy,legislative inconsistency and social networks such as relational and decisionspaces

The last two steps lead the reflection over the macro level, witha micro-model developed on the some specific cases, passing fromthe context analysis through social indicators to the study of theprojects implemented in local contexts, as well as to the study ofnetworks of actors who work there. The perspective adopted in thisphase is the one of Europeanization and the one of the differentways in which it presents itself (Graziano, 2004), including thepossibility of convergence or legislative inconsistency that thisperspective create in each context. In the original study, the testof the model “mix” was limited to a comparison within localcontext, Milan-Naples (expressions of the difference betweennorthern and southern part of the country) and internationalcontext, Milan-Berlin (generally comparable for dimensions anddevelop). The extended study included, instead, the analyzescarried out in the other nations and their local contexts (Munichcompared to Berlin in Germany; London to Liverpool in the UK;Barcellona to Vigo in Spain; Rouen in France). The main aim of theimplemented study was to complete with features and attributes allquadrants of the emerging typology and to understand if it is inthe micro dynamics that can actually be traced the dimension ofdecision-making and of the address of the different welfaresystems.Taking Graziano (2004), and then using the strategy of

comparative policy analysis, we selected two projects for twodifferent areas of policy, one relating to territorial cohesion(urban regeneration projects with expected social impact), and theother on social inclusion (projects aimed to the integration ofyoung unemployed people), as these represent the two extremes of EUintegration axis. For each project and context, five interviewshave been foreseen: to promoters, financiers, planners, operatorsand technicians involved in the implementation and development ofthe considered projects. With the interviews was intended to:contextualize the projects; undertake an analysis of theimplementation and impact based on the perception of witnesses; thereconstruction of networks of actors who have actually operated anddecided in the investigated local context. The aim was to determinethe effects of Europeanization and localization on the actual localrelational configurations and on the spaces of legislative anddecision-making autonomy in the context that the actors are able to

17

recover. Therefore, if the approach for the variables used in theanalysis of social indicators pointed to the emergence of thedifferences (for the selection of cases) and their classificationwithin the different dimensions of integration (emerged usingcases/countries as places for the measurement), the approach forcases proposed in this third step has been used to recognize thecomplexity of the unique and unrepeatable events and to investigatedevelopments. However, the comparison between policies showed thatthe differences conferred to the individual sphere of policy is notconsidered as a direct impact on the structure of the differentwelfare systems developed, while significantly affect the manner inwhich it is perceived and implemented the integration by thedifferent contexts analyzed. Therefore, much more interesting isthe geographic comparison that leaves out the weight of theterritorial and potential decision expressed by the lower levels ofgovernance.Leaving aside the specificity of each context and each project,

it is interesting to point out the emerging differences: from onehand, considering regulatory constraints, perception ofincongruity, degree of autonomy, impact and implementationprocedures; and, on the other hand, referring to the conformationof networks, dynamics of decision-making, conflict and mediation.For Graziano (2004), the processes of convergence or empowerment

of the local level of government, compared to Europe, stem fromdifferent ways to implement and translate the regulations atdifferent levels of governance, also generating inconsistencies inthe application of legislation. The existence of this space ofincongruity and the perception of it felt by the actors involved inthe implementation of policies in the projects can be functional tothe structure of spaces of action and autonomous/decision-making compared tothe system of general legislative constraints. To detect thisconnection, witnesses has been asked about the normative question(type of regulatory bound; perception of the discrepancy in thereception and application of European regulations on the locallevel; classification of the type of discrepancy - facilitating orretardant in the process of European integration and projectimplementation; consequences with regards to a possible adaptationof policy) and detection of the degree of perceived autonomy(detected in the form of scales from 0 to 10 times to investigatethe perception with respect to the recovery of autonomy space bythe actors involved in planning, expense management, implementationand strategic decisions to achieve goals). The basic hypothesis isthat the increase in perceived autonomy in each phase is directly

18

related to the perception of incongruity application of therelevant legislation and the binding nature that this assumed.Shown in the figures 6, 7, 8, 9 and 10 is the average trends

measured with regards to the degree of autonomy perceived in thedescribed strategic phases, together with an indication of thenature of the bond and the perception inconsistency. In general,the only difference found for the investigated different areas ofpolicy is in the quantity of European funds related to the cohesionpolicy which is higher than the one related to the inclusion in allthe contexts. It makes the legislation obviously more binding andthe degree of autonomy perceived with respect to these projects, ingeneral, for all contexts, lower than those of inclusion. Concerning the differences for geographical comparison and

starting from Italy, in the original model, the level of autonomy(with regards to Milan and Naples) is rather high and it isassociated to a regulatory restriction that goes from taxation to acoordinative bound for the different levels of governance involvedand a perceived inconsistency in the application and translation ofrules (but differential in the contexts): while the inconsistencyin Milan leads to slowdowns and delays (instead of facilitatespaces of autonomy) in the possibility to align the city to thescenario of the Unitarian European welfare System, in Naples, theinconsistency is rather perceived as an advantage in the ability toadopt alternative strategies without reducing the achievement ofcommon goals.In Berlin, however, the level of perceived autonomy is good and

the regulatory restriction is moderate. In the German city, themindset to create a process of incongruity in a conscious way doesnot exist and the main aim is to achieve a full convergence inevery regulatory and practical aspect, because the successfulnormative is perceived as a functional set for the regulatoryadaptation. This arrangement has obviously some implications interms of impact, so if the best results, both on beneficiaries andon territory, will be achieved in Berlin (where the implementationis a procedural standardization and a function of convergence withthe overarching goal to make bureaucratic the whole process,approved and loyal to a standardized application), a good level ofimpacts are recorded also in Milan (where the implementation takesgreater account of the concerns of concordance with the needsemerged locally in comparison with an application of procedures andwith the aim to create a local network. On the other hand, morelimited and relative impacts are achieved in Naples – in whichthere is a differential implementation for the strategies applied,

19

but with regards to the objectives previously set and theoverarching goal to capitalize on powerful, economic and decision-making resources. The processes of integration and Europeanizationare outlined (in the Italian scenario) to be still far from fullconvergence and structural capacity of regulatory adaptation, tiedto a vision of the differential development of social policy on aterritorial basis. Always returning to the comparison intra-national for Germany, Munich back in high levels of perceivedautonomy, together with regulatory constraints and coordinativemanagement and absolute absence of perception of incongruity. Asfor Berlin, also in this context the impacts are very good and theimplementation assumes standardized features aimed at obtaining thebest possible results. The processes of integration andEuropeanization are outlined in the German scenario aimed to theconvergence as fully absorbing European addresses. Very similar to Germany with respect to the characteristics

investigated is the United Kingdom with London and Liverpool, twolocal contexts that are almost completely interchangeable with thecharacteristics observed, showing a marked inconsistency of intra-national comparison. Autonomy levels recorded are extremely high,the regulatory constraint are made of coordinative bound. Theinconsistency, despite the large degree of autonomy, is notperceived and the overarching goal is the respect of localdimension which is totally in line and ready to implement theobjectives and procedures dictated by the supranational actor. Itbecomes essential to reduce slowdown factors, which can engageimplementations heavily bureaucratized, promoting more local initiativeand less Europe, especially with respect to the design and therecovery of funds for local development. The resulting impactsappear to be limits to the territory that is compared to therecipients of the interventions investigated. The processes ofintegration and Europeanization are outlined in the Anglo-Saxonscenario as aimed at the convergence fully absorbing Europeanaddresses but claiming its own specific implementation of localdecision-making and autonomy.The French case is different and is currently represented

exclusively by the context of Rouen as interviews of Paris are inthe process of development. In this context, the observed levels ofautonomy are fairly good, the relevant legislation have strongbound and the inconsistency (rather than show it in its legislationcharacterization) is shown with procedural connotation as obstaclesencountered of administrative and bureaucratic nature in theimplementation of standardized procedures in a context that would

20

require flexible application. This peculiarity is due to the factthat France has a cross-cutting objective of the consolidation ofthe partnership networks, which is able to listen and understanddeeply the requests of the territory following its needs, butwithout detaching the aim of convergence. The impacts achieved aretherefore good, but suffer in the implementation phase, of abreathlessness in adaptation of legislation for specific localneeds and context recall, not fully conjugated with the Europeanimpositions. The processes of integration and Europeanization areoutlined in the French scenario as aimed at regulatory convergenceslowed down from the implementation procedures and from theincreased concern for the local.Finally, in Spain, both in Barcelona and in Vigo, the situation

has not registered a significant gap in terms of intra-nationalcomparison. Here is the lowest degree of perceived autonomy,stringent regulatory constraints, detected perception ofinconsistency legislation on differential basis for the differentlevels of governance involved, detected mainly in the translationsof local autonomy than the national context, and yet not seen asfacilitating factor. The unifying goal is the willingness of thecreation of networks of communication between the various levels ofgovernment that may, in facts, address the issue of thedifferential application between different levels. This leads topositive impacts on average, with a tendency to respect theobjectives set by the European actor (rather than to theprocedures), but showing an inevitable trouble in regulatoryadaptation, due to the actual lack of coordination betweendifferent levels of government involved. The processes ofintegration and Europeanization that are outlined in the Spanishscenario, like the Italian one, are still far from full convergenceand structural capacity to absorb regulatory, tied to a vision ofdevelopment differential on a territorial basis.

Fig. 6-7-8-9-10: autonomy degrees in relation with policy and with referred tolocal and national contexts.

21

The last step of analysis involved the networks analysisarrangements as relational and power configurations that can bemore or less “conflicting”: Policy network made up of actors (as wellas public then private) with qualitatively and quantitativelydifferent resources and operating within a defined space of policy.The process can be understood in the sense of variableconfigurations of links between individual and collective actors(Scott, 1991; Wasserman, Faust, 1998) but also in the sense ofpoles of attraction focusing on socially perceived problems andcapable of attracting variously qualified and interested actors(Milward, Wamsley, 1984). The application of network analysis hasthe aim to understand whether it is in the micro dynamics that thedimensions of decision-making and address for the different welfaresystems can be traced and it would be done by giving to thestructure (or to the actors of the network) the key role of thedynamics of address. In the original study Therefore It was takeninto account socio-centered network representative of theinvestigated projects (respectively one of inclusion and one ofcohesion for each local context). In them the bonds were weighed onthe intensity of the relationship understood as multiplicity(multiplexity) that means the possibility that (within the bondhighlighted) there is more than one type of bond (professional,

22

friendship, parental, institutional). After it, we put in evidencethe conflicting and the mediation actors (economic for materialsresource; emotional and personal) to figure out which position isconfined within the network for these actors and how this can beuseful for the decision-making purposes. By analyzing the levels ofopenness of the network, its density and cohesion, are highlightedthree spheres of strategic importance on which actors operate: thepositional (official and institutional), reputational (subjective)and decision-making plan4.Starting from Italy, in Milan there is the emergence of extended

network, dynamic, inclusive, loose knit, arranged to contain anddiscuss the external and internal conflicts within the constitutedrelational space, equipped with many points of mediation, based onintegration, dynamism, openness, horizontal structure, division byspheres of influence and composed of multiple actors forconstituted areas, with many stakeholders next to the implementers.The tendency is to flow into the local net welfare system whose goal isthe integration in the EU, but whose strategies go to inextricablybind to territorial specificities and to the network of actorsexploiting the spaces of incongruity in the creation of a regionalstrategy fully adapted to the Milan context. The structure of thedetected network, based on a core of primary relationships andbranching of these secondary relationships, suggests that decisionsare created and placed on the network right from the center, alsoidentified as the decision-making sphere, which is to be configuredin such a way, also as a center of power. A capacity ofcentralization and accumulation of functional communication,information and resources typical of the issue network, which bringsthis context to stretch out together both towards social inclusionpolicies that territorial cohesion.

Fig. 11: Network for social inclusion Milan. Fig. 12:Network for territorial cohesion Milan.

4 To the witnesses were asked to indicate the actors involved compared to thethree spheres of matter and their interconnections, indicating the nature of thelink between subjects.

23

In Naples, instead, there are lobbyist network, tight and focused,made of multiple bonds, strong and closely joined, in which theconflict is included in the net and is involved in the informationsharing, decision-making, strategic and operating power ability,providing, however, also figures devoted to mediation. Based onexclusivity, strong powers, cross- memberships in most areas ofpolicy and with decision-making structures at the top rather thanhorizontal connection. In Naples the focus are on policies ofinclusion in a system of local net welfare whose aim is the Europeanintegration, but with strategies inextricably bound to theterritorial peculiarities and to the mixture of interests which areon it. These strategies are aiming to move as much networks ofactors as possible, using the spaces of normative inconsistency aswell as the possibility of management and emergence of the conflictby incorporating it internally and causing it to become an activepart of the decision-making process and the formation of the powerof the local management in the choices on the welfare system. Theactors involved are implementers but also local powers, politicaland institutional actors, with the absence of externalstakeholders.

Fig. 13: Network of social inclusion in Naples. Fig. 14: Network ofterritorial cohesion Naples.

In Berlin there are restrictednetwork, cohesive and structured,whose actors are part of a singleinstitution who is the generalassociation of the third sectorwhich is responsibleoperationally of the implementation of the projects. This network

24

does not require the necessity to be open outside (Milan) or toinclude strong interests in the network formed (Naples). Instead,it has tight knit, thick and made of multiple and strong bonds,structured to contain the conflict, even when it is not revealed,providing always the presence of mediators. Stability, exclusivity,closure and synergy are the benefits of this network that leads tothe structure of real policy community, in which there is an interestin an exchange not merely material but based on a more extensivesharing of the same value system, which aims to the success of theproject as a function of regulatory adaptation. The actors involvedare all implementers and operators. The tendency is towards aregulatory, institutional and community convergence and toward aUnitarian European Welfare Systemstructure more focused on theterritorial impacts andtherefore more dedicated to givespace to cohesion policies.

Fig. 15: Network of social inclusionBerlin. Fig. 16:Network of territorial cohesion Berlin.

In Munich the networks are highly interconnected, tight,cohesive, dense, in which the conflict is not detected but at thesame time it is expected at least one mediator. Closure, littleinclusiveness, stability and internal synergy leading to thetypical structure of the policy community in which always the sameactors are involved for the three spheres considered (decision-making, reputational and positional). The dealers and implementerswhich are in it come from an unique association with the absence ofexternal stakeholders, not recognizing, as in Berlin, the need tobe open to the territory or to include strong powers. The relaxed,marked and regulated atmosphere (stringent regulatory restrictions)leads to mechanical decisions and implementation of social policiesin local contexts, exactly like in standardized protocols, in which

25

prevails the tendency to converge to Unitarian European Welfare Systemand the achievement of territorial cohesion policies.

Fig. 17: Network of social inclusion Munich. Fig. 18: Network ofterritorial cohesion Munich.

Moving to France, in Rouen the network return extended,inclusive, open, dynamic, with no cross-memberships or multiplebonds, structured around a core of primary relationships in thedecision-making sphere from which branch opening secondary actorsand diversified areas in the positional and reputational spheres.The conflict is revealed and broken down on the side of theprocedural and decision-making activity and it requires manymediators to stem it. The relational configuration for the networkof cohesion is really specific, as the same actors conflicting areperceived as the main mediators of the conflict generated by therelational dynamic in the network. Openness, inclusion of thedynamics and the needs of the territory, extension, horizontalstructure divided by spheres of influence, are the characters thatbring a conformation of issue network less stable and formalized butmore elastic and flexible. The actors involved come from differentlevels of government and governance, as well as implementers andoperators, stakeholders, external actors to the specific area of interest, just the desire to recover the soul of a strictly localsocial policy. However, even here, there is still a tendency aimingto the convergence with Unitarian European Welfare System and to theleverage of territorial cohesion policy more focused on thedevelopment of competitiveness and economic growth.

26

Fig. 19: network of social inclusion in Rouen. Fig. 20: Network ofterritorial cohesion in Rouen.

In the UK the situations recorded in London and Liverpool aretotally equivalent. The network are based on small cliques ofactors maximally connected, without a division into spheres ofinfluence, managing the different stages and planningresponsibilities, with the absence of conflict although figures ofmediation are also provided in the networks of London, the figuresdisappearing neatly into those of Liverpool. Closure, littleinclusiveness, narrowness, synergy are the characteristics thatlead to the conformation of professionalized network, consisting ofoperators, implementers and institutional actors in the totalabsence of stakeholders or actors outside of the scope of thepolicy involved. The prevailing trend is towards a convergentsystem of Unitarian European Welfare System based (not as in Germany andFrance) on a strong pressure towards social inclusion policies, asector that is still lacking, and on which these contexts plans tocomplete a global development of individuals and territory.

Fig. 21: Network of social inclusion in London. Fig. 22: Network ofterritorial cohesion in London.

27

Fig. 23: Network of social inclusion in Liverpool. Fig. 24: Network ofterritorial cohesion in Liverpool.

The last national context analyzed is Spain with Barcelona andVigo which are maximally interconnected network, tight, cohesive,with crossed affiliations and strong internal coordination with thepresence of multiple bonds that, approaching the professionalrelationship, leads to the presence of the parental relationship.Therefore, these networks are inclusive for operational levels, butclosed out. In fact, closure, exclusivity and internal coordinationboth institutional and politic lead to a vertex structure ofrelationship closer to combined network with cross-memberships. Theconflict is in the perception of interviewed witnesses, but,nevertheless, it is not specified, or embodied in specificlocations or actors. Mediation is, however, expected to be bothinternal and external to the network, and it is also importantthat, in the networks of Vigo, the actor is responsible for thechecks to be mentioned as influential but has no connection withthe actors of the network. Finally, few actors are involved in thenetwork, including implementers, operators, administrative,institutional and representatives of powerful interests, while theyare totally absent stakeholders and external actors. For the

28

structural conformation, the networks of Barcelona are very similarto those in Germany, while those of Vigo are very close to thoserecorded in the Anglo-Saxon context. However, the factor thatdifferentiate these networks is the nature of the ties that bindsthe subjects (multiple and extremely strong ties), but especiallythe fact that they are always the same subjects, regardless to theaim of the policy in question, which is aiming to maneuverdecision, implementation and maintain power in territorialdevelopment of social policy. Even if the used network appears tobe so strongly associated in each area of policy, the localdynamics leading to the emergence of a regime closer to the local netwelfare, mostly aimed to the territorial cohesion policy, in whichthe context, limited by law, but stimulated procedurally, becomesthe responsible entity of the broad welfare, rather than delegatesupranational actor.

Fig. 25: Network of social inclusion in Barcelona. Fig. 26: network ofterritorial cohesion in Barcelona.

Fig. 27: Network of social inclusion in Vigo. Fig. 28: Network ofterritorial cohesion in Vigo.

29

The different relational patterns detected shows that it is notthe structure of relationships to influence the field (think to thestructural similarities between Naples, Berlin and Barcelona, orbetween London and Vigo, or even between Milan and Rouen), but arethe actors involved in it, with their roles and positions, to givelife to different implementations.

5. Conclusion: the general model of classification

The results presented represent an evolving typology, born from apilot study, currently completed and tested in its specific parts.The conclusions that have been reached led the characteristics ofproject implementation, the conformations and the propensity toparticular types of networks, to be screened in the space ofattributes built in macro analysis stage. This is to complete thenecessary details for the proposed general model of classificationand clearly demarcate the systems emerged and the directions ofintegration that they create in comparison with the first twoquadrants analyzed. A new polarity is revealed: issue networks vs.policy community (Marin, Mayntz, 1991) and on regards to this thereare local contexts and the features of network in it. For thisreason, in the quadrant of convergence together with the tendencythat leads to a common European welfare and policies which areaiming to territorial cohesion and to economical competition we cansee also the presence of network intended as policy community, closedtight, cohesive and highly structured, embodied in the Germanscontexts.At the center of this new polarization and therefore also at the

center of the space of attributes, there are French contexts,midway between the strong cohesion and the evolution of the network

30

towards the issue networks intended as open and dynamics network,inclusive and extensive fully embodied in the extreme polaritycontexts such as Milan. We are therefore in the quadrant ofEmergence that intersects the local net welfare systems and propensityfor local social inclusion policies. An exception in this quadrant is theNeapolitan context, showing that the more we move away from thecenter the more extreme the local net welfare systems aimed at localcharacterizations may show quite unexpected. In this context, thenetwork back to being closed, not inclusive and based on lobbyistdynamic, a clear expression of the fact that it is moving more andmore towards convergence by reference to the implementation ofsocial policy as a dimension purely contextual and local. ThisItalian bifurcation is not trivial, since it shows theorganizational and territorial management and differentiatedsystems of local net welfare systems.In the quadrant that intersects trends towards local net welfare

systems and propensity to social cohesion policies, the quadrant ofDevelopment, we find the Spanish contexts, which are to becharacterized by having locally developed networks and stronglycharacterized by the presence of the cross members in more areas ofpolicy, which are, therefore, limited, highly cohesive, selectedand interconnected. Also in this case we have a strong tendency toautonomy, marking the distance from the path to convergence, asthey are the same local contexts, already burdened by economic andsocial crisis, not to lend itself to systematic and standardizedimplementations of social policy, and so because they are the placeof foothold and deep knowledge of the territory on which it isimplemented.In the last quadrant, that of Stabilization, consisting of the

intersection between regimes aimed at the convergence to the Europeanwelfare and propensity towards social inclusion policies, the strategy ofthe network that emerged is the one of professionalized networks,extremely narrow actors, without strong ties, but with functionalobjective, purpose networks that tend to maximize the local impactof social policy conceiving the implementation process as a processprofessionalized and highly specialized, i.e. involving the "minds"(managers and institutional actors) rather than the "arms"(operators and implementers) in the process of localimplementation of social policy.This typology reveals its character as an integrated tool that

includes the results of different analytical methods to make theminto an overall system. This typology can be understood as a toolfor reading the changes taking place in the differences and in the

31

trends that pervade the current welfare systems. It offersinterpretive categories and insights, which are useful to expandand retrofit additional elements, the same typological tool ofinterpretation generated, as in continuous evolution and ready topick up the change and make it a systemic integral variable.

Fig. 29: General Model of Classification.

32

BibliographyAnderson J. E. (1984), Public Policy Making: An Introduction, Houghton Mifflin, Boston MA. Bifulco L. (2005), Come cambiano le politiche sociali europee, in Id. (a cura di), Le politiche sociali. Temi e

prospettive, Carocci, Roma, pp. 13-36. Bifulco L., de Leonardis O. (2006), Integrazione tra le politiche come opportunità poltica, in Donolo C. (acura di), Il futuro delle politiche pubbliche, Mondadori, Milano, pp. 31-58.

Börzel T. A., Risse T. (2004), One Size Fits All! EU Policies for the Promotion of Human Rights, Democracy, and theRule of Law, in paper presented for the Workshop on Democracy Promotion organized by the Centerfor Development, Democracy, and the Rule of Law, Stanford University, October 4-5, 2004.

Capano G., Giuliani M. (1996), Dizionario di Politiche Pubbliche, NIS, La nuova Italia Scientifica,Roma.

Cremaschi M. (2006), Europeizzazione ed innovazione nelle politiche del territorio, in Donolo C. (a cura di), Ilfuturo delle politiche pubbliche, Mondadori, Milano, pp. 205-218.

Cresswell J.W. (2003), Research Design: Qualitative, Quantitative and Mixed Method Approaches, Sage, London.ID. (2012), Qualitative Inquiry and Reaearch Design: Choosing Among Five Approaches, Sage, London.De Leonardis O. (1998), In un diverso welfare. Sogni ed incubi, Feltrinelli, Milano.ID. (2003), Le nuove politiche sociali, in Bifulco L. (a cura di), Le politiche sociali. Temi e prospettive,

Carocci, Roma.Dye T. R. (1972), Understanding Public Policy, Prentice-Hall, Englewood Cliffs. Donzelot J. (2003), Faire société, Seuil, Paris.Edward O., Laumann F., Pappi U. (1976), Networks of collective action: a perspective on community influence

systems, New York, Academic Press. Esping-Andersen G. (1985), Politics against Market. The social democratic road to power, Princeston UniversityPress, Princeston.

ID. (1990), The three words of welfare capitalism, Polity Press, Princeton. ID. (1996), After the Golden Age? State Dilemmas in a Global Economy, in G. Esping-Andersen G. (ed.),

Welfare State in Transition, Sage, London. ID. (2002), Why we need a new welfare state, Oxford University Press, Oxford. ID. (2005), Il welfare state senza il lavoro. L’ascesa del familismo nelle politiche sociali dell’Europa continentale, in«Stato e Mercato» n.45, Il Mulino, Bologna.

ID. (2005), Le nuove sfide per le politiche sociali nel XXI secolo, in «Stato e Mercato» n.74, Il Mulino,Bologna.

Fargion V., Morlino L., Profeti S. (a cura di, 2006), Europeizzazione e rappresentanza territoriale, IlMulino, Bologna.

Ferrera M. (1984), Welfare State in Italia. Sviluppo e crisi in prospettiva comparata, Il Mulino, Bologna. ID. (1993), Modelli di solidarietà. Politica e riforme sociali nelle democrazie, Il Mulino, Bologna. ID. (1995), The Rice and Fall of Democratic Universalism. Health Care Reform in Italy, 1978-1994, in «Journal of Health

Politics, Policy and Law», vol. 20, n. 3, pp. 275-302. ID. (1996), The “Southern Model” of Welfare in Social Europe, in «Journal of European Social Policy», n.6, pp. 17-37.

ID. (1998), Le trappole del welfare, Il Mulino, Bologna. ID. (2006), Le politiche sociali. L’Italia in prospettiva comparata, Il Mulino, Bologna. ID. (2008), Trent’anni dopo. Il welfare state europeo tra crisi e trasformazione, in «Stato e Mercato», n. 81,pp. 341-375.

Flora P., Heidenheimer A. (eds, 1981), The development of Welfare State in Europe and America, NewBrunswick Transaction Press, New Brunswick.

Giuliani M. (2004), Europeizzazione come istituzionalizzazione: questioni definitorie e di metodo, URGE WorkingPaper 4/2004, in “Rivista italiana di politiche pubbliche”, 1/2004, pp. 141-161. 

Graziano P. (2004), Europeizzazione e politiche pubbliche italiane, Il Mulino, Bologna.Gullièn A., Palier B. (2004), Does Europe Metter? Accession to EU and Social Policy Developments in Recent and

New Member States, in “Journal of European Social Policy”, 14/3, pp. 203-209.Hox J. (2002), Multilevel Analysis. Techniques and Applications, Lawerance Erlbaum Associates, London.Jenkins W.I. (1978), Policy Analysis: A Political and Organizational Perspective, Martin Robertson, London. Jenson J. (1998), Mapping Social Cohesion: The State of Canadian Research, Canadian Policy Research Networks, CPRN

Study, n° F/03Kazepov Y. (a cura di, 2009), La dimensione territoriale delle politiche sociali in Italia, Carocci, Roma.ID. (a cura di, 2010), Rescaling social policies: towards multilevel governance in Europe, Ashgate. Lasswell H. D. (1959), Power and Society, Yale University Press, New Haven. Le Gales P. (2002), European Cities. Social Conflicts and Governance, Oxford University Press, Oxford.Leibfried S., Pierson P. (eds.) (1995), European Social Policy. Between Fragmentation and Integration, The

Brookings Institution, Washington.

33

Naldini M. (2007), Le politiche sociali in Europa. Trasformazioni dei bisogni e risposte di policy, Carocci, Roma.Paci M. (1997), Welfare State, Ediesse, Roma. ID. (2004), Le ragioni per un nuovo assetto di welfare in Europa, in «La Rivista delle Politiche Sociali»,n.1/2004, pp. 333-373.

ID. (2007), Nuovi lavori, nuovo welfare, II ed., Il Mulino, Bologna. ID. (2008), Welfare, solidarietà sociale e coesione della società nazionale, in «Stato e Mercato», n. 82, pp. 3-29.

ID. (a cura di, 2008), Welfare locale e democrazia partecipativa, Il Mulino, Bologna. Palier B. (2000), Does Europa Matter? Europèanisation et rèforme des politiques socials des pays de l’Union

Europèenne, in «Politique Europèenne», n. 2, pp. 7-28.Punziano G. (2012), Mixed Methods as mirror of Social Research, in (a cura di) C. Cipolla, A. de Lillo,

E. Ruspini,Il sociologo, le sirene e le pratiche di integrazione, Franco Angeli, Milano.ID. (2012), Welfare europeo o welfare locali? I processi decisionali nel sociale tra convergenza ed autonomia, DiogeneEdizioni, Napoli.

Radaelli C. M. (2003), The Europeanization of Public Policy, in K. Featherstone and C. M. Radaelli(eds), The Politics of Europeanization, Oxford, Oxford University Press.

Scott J. (1997), L’analisi delle reti sociali, ed. it. Enrica Amaturo (a cura di), NIS, Roma.Sen A. (1993), Capability and Well-Being, in Nussbaum M. C., Sen A., (eds.): The Quality of Life.Clarendon Press, Oxford, pp. 30-53.

Tashakkori A., Teddlie C. (2003), Handbook of Mixed Methods in Social and Behavioral Research, Sage,London.

Teddlie C., Yu F. (2007), Mixed methods sampling: A typology with examples, Journal of Mixed MethodsResearch, n. 1, pp. 77-100.

Titmuss R. (1974), Social Policy, Allen &Unwin, Londra.Wasserman S., Faust K. (1998), Social Network Analysis. Methods and applications, Cambridge University

Press, Cambridge.

34

Appendix 1: Selected projects for the inter and intra national comparison:

35