SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

253

Christos J. Paraskevopoulos
Social Capital, Learning and EU Regional
Policy Networks: Evidence from Greece*
THIS ARTICLE DISCUSSES THE IMPORTANCE OF SOCIAL LEARNING FOR

EU public policy-making in general and regional policy in particular.

The first section analyses the implications of the learning process
for EU regional policy and examines its social and institutional
prerequisites. Section two introduces the concepts of social capital
and institutional networks as components crucial for the learning
process and socialization function. The third section, based on the
analysis of the role of social learning, delineates the multi-level system
of governance in EU regional policy. The fourth section presents
empirical evidence from Greek regions on the role of social learning
in the implementation of EU regional policy (Structural Funds)
programmes. Finally, the last section draws conclusions on the role
of social learning in EU regional policy and lessons from the Greek
experience.
The notion of learning has emerged in a wide range of social
sciences as a crucial conceptual tool for explaining adaptation and
change of system parameters at both the micro and macro levels. In
political science, learning, as an explanatory variable for major
changes (paradigm shifts) in the policy-making process, has become
a crucial concept for analysing the statesociety relations and hence
in contemporary theories of the state.1 Thus within the theoretical
framework of integration in Europe learning focuses on the

*
An earlier version of this article was presented as a paper to the 26th ECPR
Workshop on the Institutional Analyses of European Integration, University of
Warwick, 2328 March 1998.
1
Peter A. Hall, Policy Paradigms, Social Learning and the State: The Case of
Economic Policy-making in Britain, Comparative Politics, 25 (1993), pp. 27596.

254

GOVERNMENT AND OPPOSITION

complexities of the process of domestic preference formation,2 which

have been, to a significant extent, overlooked by proponents of both
rational choice institutionalism and liberal intergovernmentalism.3
It is considered as a function of adaptable systems of governance
under uncertainty,4 and in that sense it implies the process by which
actors acquire new interests and identities and form their preferences
through interaction with broader institutional contexts/norms.5
Through these structureactor interactions, actors adapt their
behaviour to meet the changes in the environment. As a predominantly interactive process, learning is crucially dependent on the
way in which the system of institutional interactions is shaped, on
the adequacy of information and communication f lows, and on the
presence of forums for dialogue among the actors. The latter enable
institutional actors, by their joint involvement in the processes of
learning by doing and learning by past successes and failures, to
become adaptable, rather than adapted to changing conditions.6
Learning in this environment is a function of past policy attempts

2
Jeremy Richardson, Actor Based Models of National and EU Policy-Making, University
of Essex, Dept of Government, Discussion paper No. 103 (1996); B. Kohler-Koch,
Catching up with Change: The Transformation of Governance in the European Union,
Journal of European Public Policy, 3:3 (1996), pp. 35980; Jeffrey Checkel, Social
Construction, Institutional Analysis and the Study of European Integration, paper presented
at the 26 ECPR-Joint Sessions of Workshops, University of Warwick, 2328 March
(1998); and C. J. Paraskevopoulos, Interpreting Convergence in the European Union:
Patterns of Collective Action, Social Learning and Europeanization among Greek Regions,
London, Macmillan, forthcoming, 2001.
3
Gary Marks, An Actor-Centred Approach to Multi-Level Governance, Regional
and Federal Studies, 6:2 (1996), pp. 2138.
4
Richardson attempts to introduce the notion of epistemic communities,
originally conceptualized in the field of international relations (see below P. Haas,
1992), within the EU policy-making. Given that the concept of epistemic communities
refers to the uncertainty of international actors and thus points to the role of knowledge
and information f lows in facilitating cooperative relations, it is consonant with the
learning approach to regional integration in Europe. See Richardson, 1996, op. cit.;
Peter Haas, Introduction: epistemic communities and international policy
coordination, International Organization (Special Issue) 46:1 (1992), pp. 135; Checkel,
1998, op. cit.; and B. Kohler-Koch, 1996 op. cit.
5
J. Checkel, 1998, op. cit.
6
Shari O. Garmise, Institutional Networks and Industrial Restructuring: Local
Initiatives toward the Textile Industry in Nottingham and Prato, Unpublished PhD Thesis,
LSE, 1995.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

255

(and the involved actors interpretation of their successes and

failures), of the capacities of institutions to design new activities,
and of the changing ideas and shifting alliances and balance of
power among the actors.
In structural policy the learning process has implications for the
organizational structure of the regional politico-economic system.
On the one hand it requires that the organizations involved are
sufficiently f lexible to make the appropriate structural adjustments
to exploit the benefits of learning. On the other, the learning process
is crucially dependent on experts who specialize in specific policy
areas.7 Because this combination of f lexibility and specialization is
best achieved in networked organizations, the network paradigm
constitutes the appropriate organizational form for the learning
process.8 Finally, since learning is a process of waking up and
catching up9 and therefore usually undermines the stability of
relations between the participating actors, informal social norms
and conventions play a crucial role in providing the glue that cements
and re-stabilizes the relations among the involved actors.
Within the EU structural policy environment, learning is viewed
as significantly affected by the intended and unintended consequences10 of the redistribution of power and resources between the
supranational, national and subnational levels of government within
the EU, and subsequently by the degree of decentralization of the
administrative structure of the state, which allows stimulation of
synergistic effects among the local actors. However, as existing
7

See P. Hall, 1993, op. cit.; and J. Checkel, 1998, op. cit.
Michael Storper, The Resurgence of Regional Economies, Ten Years Later: The
Region as a Nexus of Untraded Interdependencies, European Urban and Regional
Studies, 2:3 (1995), pp. 191221.
9
Charles Sabel, Learning by Monitoring: The Institutions of Economic
Development, in J. J. Smelser and Richard Swedberg (eds), The Handbook of Economic
Sociology, Russell Sage Foundation, Princeton, Princeton University Press, 1994.
10
The impact of the 1988 reforms of the Structural Funds on the redistribution
of power between the levels of government, by strengthening the role of the subnational
level and establishing direct linkages between supranational, national and subnational
authorities through their role in managing and monitoring Operational Programmes
(OPs) of the Community Support Frameworks (CSFs) is a characteristic case of
unintended consequences resulting from institutional or policy reforms at the EU
level. Moreover, the core of these reforms remained almost unchanged even after the
1993 reform put forward after the negotiations over the 199499 Structural Funds
programme.
8

256

GOVERNMENT AND OPPOSITION

evidence shows, intra- and inter-state differentiation at the level of

institutional learning and adaptation is not always associated with
the state structure, and intergovernmental relations constitute a
rather dynamic system which cannot be simply reduced to a symptom
of the state structure.11 Hence, certain capacities for collective action
that facilitate the shaping of the interactions and the process of
coalition-building among key social and economic actors at the
regional and local levels are selected as the most important prerequisites for social learning and adaptation.12 This point has been
reinforced by results of recent research, showing that the strength
of associational culture and regional identity is the underlying factor
of regional mobilization at the European level, rather than a
funding/resource focusing logic. Moreover, associational culture and
regional identity constitute the crucial qualitative variables that have
sustained the institutional performance of strong European regions,
such as the Four Motors for Europe13 interregional partnership or
EmiliaRomagna.14 Although these regions are not usually recipients
of EU funds, they provide strong evidence about the role of collective
action in facilitating institutional and social learning, policy
innovation and successful adaptation of the regional governance
structures to the global and European environment. In that sense,
both formal and crucial informal institutional arrangements play a
decisive role in determining the learning capacity of regional and
local systems of governance and subsequently the degree of their
11
K. Klausen and M. Goldsmith, Conclusion: Local Government and the European
Union, in M. Goldsmith and K. Klausen (eds), European Integration and Local
Government, Cheltenham, Edward Elgar, 1997.
12
Charlie Jeffery, Sub-National Authorities and European Integration: Moving Beyond
the Nation-State?, paper presented at the Fifth Biennial International Conference of
the European Community Studies Association, Seattle, USA, 29 May1 June, 1997;
C. J. Paraskevopoulos, Social Capital and the Public/Private Divide in Greek Regions,
West European Politics, 21:2 (1998), pp. 15376; C. J. Paraskevopoulos, Social Capital,
Institutional Learning and European Regional Policy: evidence from Greece, Regional
and Federal Studies, 8:3 (1998), pp. 3164.
13
The Four Motors for Europe partnership comprises Baden-Wrttemberg, the
initiator of the project, Catalonia, Lombardy and Rhne-Alpes.
14
Philip Cooke and Kevin Morgan, The Associational Economy: Firms, Regions and
Innovation, Oxford, Oxford University Press, 1998; Shari O. Garmise, Economic
Development Strategies in Emilia-Romagna, in Martin Rhodes (ed.), The Regions and
the New Europe: Patterns in Core and Periphery Development, Manchester, Manchester
University Press, 1995.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

257

mobilization at the European level. Further, the increasing importance of capacities for collective action in the learning and adaptation
processes is underlined by the emergence of the network paradigm
as an operational element of the institutional infrastructure at all
stages of policy-making in the EU.15 Thus, institutional networks
and social norms are introduced within the new governance agenda
of the EU studies,16 as conceptual tools for facilitating the learning
and adaptation processes through their capacity for resolving
collective action problems. Networks play a key role in the process
by structuring institutional interactions, while social norms provide
stable rules and procedures that facilitate exchange and f low of
information and also reduce uncertainty.
As is obvious, the learning approach, while accepting the crucial
role of strategic interactions between actors as determinants of policy
outcomes, embodies path dependence and unintended consequences as important parameters of institutional and policy change
processes and social norms through the socialization function as
regulators of the uncertainty that surrounds both policy and
institutional change. In that sense, social learning is a complicated
process and hence it should be distinguished from simple learning
where actors acquire new information, alter strategies but pursue
given, fixed interests. Subsequently, the dynamic model of social
learning incorporates strategic interactions between actors, unintended consequences and social norms.
SOCI AL CA PI TAL, INSTI TUTIONAL THICK NESS A ND TH E
SOCIALIZATION FUNCTION

Although the emergence of the network metaphor in the study of

policy-making in the EU was initially conceived of as a ref lection
15
B. Kohler-Koch, 1996, op. cit.; Patrick Kenis and Volker Schneider, Policy
Networks and Policy Analysis: Scrutinizing a New Analytical Toolbox, in Bernd
Marin and Renate Mayntz (eds), Policy Networks: Empirical Evidence and Theoretical
Considerations, Boulder, Colo., Westview Press, 1991; Adrienne Windhoff-Heritier,
Policy Network Analysis: A Tool for Comparative Political Research, in Hans Keman
(ed.), Comparative Politics: New Directions in Theory and Method, Amsterdam, VU
University Press, 1993; J. Peterson, Decision-making in the European Union: Towards
a Framework for Analysis, Journal of European Public Policy, 2:1 (1995), pp. 6993.
16
Simon Hix, The Study of the European Union II: The New Governance Agenda
and its Rival, Journal of European Public Policy, 5:1 (1998), pp. 3865.

258

GOVERNMENT AND OPPOSITION

of the necessity for mapping the exchange relations among the

actors, the real added-value of the network analysis is linked to its
capacity for capturing the system of institutional interactions. This
is because networks are based on relations or linkages and therefore
are associated with the broader social or organizational system which
involves many other actors.17 This key contribution of the network
paradigm is related to the re-conceptualization of power within
networks, and to its implications for the processes of exchange and
interdependence. In particular, since power within networks is the
ability to attain higher levels of collective action18 and given the
high degree of resource interdependencies among the actors within
inter-organizational networks, power relations within networks are
mainly based on the process of exchange.19 The exchange process,
in turn, involves a variety of resources (money, information, knowledge) and constitutes one way to achieve collective action among
the actors. Thus, institutional networks are defined as systems of
interactions involving both public and private institutional actors,
which are ranged together around a certain policy domain or territory and hence bounded by it.20 In that sense, however, the network
paradigm overrides the rational choice, new institutionalist
approach in a constructive way, on the one hand by taking into
account and mapping the rationality-based exchange process, and
on the other, by capturing the main features of the broad institutional environment in terms of institutional interactions and interdependencies among the actors. In doing so, the network metaphor
becomes an important component of the socialization function,
which is based on the assumption that exchange relations do not
depend on the availability of resources but on actors perceptions
about their value and usefulness; thus the shaping of the exchange
process is profoundly inf luenced by the broad social context.
In this respect, the emergence of the network paradigm is viewed
as consistent with economic sociologys criticism of the under17
David Knoke and James H. Kuklinski, Network Analysis, California, Sage
Publications, 1982.
18
L. Metcalfe, Designing Precarious Partnerships, in P. C. Nystrom and W. H.
Starbuck (eds), Handbook of Organizational Design, Vol. 1, Oxford, Oxford University
Press, 1981.
19
Bernd Marin, Introduction, in B. Marin (ed.), Governance and Generalized
Exchange: Self-Organizing Policy Networks in Action, Frankfurt, Campus Verlag, 1990.
20
C. J. Paraskevopoulos, 2001, op. cit.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

259

socialized character of rational choice new institutionalism and

particularly the functional-neoclassical explanation of the origin of
social institutions. This is particularly evident in Granovetters
embeddedness argument, which emphasizes the social character
of human action and stresses the role of networks in generating
trust and in establishing expectations and norms. 21 Thus, the
embeddedness approach points to the inf luence of social and
cultural contexts (norms, conventions and rules of behaviour) upon
rational or purposive action and views social and institutional
structure as interactive systems, and networks as an institutional
model between markets and hierarchies, which in turn plays an
important role in the socialization function.
Within this theoretical framework, social capital has emerged as
the second important component of the socialization function. It is
theoretically based on a limited-rationality model, similar in many
respects to Axelrods evolutionary approach to norms,22 and refers
to features of social organization, such as trust, norms and networks
that can improve the efficiency of society by facilitating coordinated
action.23 Trust constitutes the first important form of social capital.
It is linked to the volatility and hence uncertainty of modern
institutional settings and is seen as a crucial conceptual mechanism
to resolve this uncertainty by shaping the relations between partners
and facilitating collective action. Social trust in modern complex
settings, however, can arise from two related forms of social capital:
norms of reciprocity and networks of civic engagement. Social
norms transfer the right to control an action from the actor to others
because that action has externalities, that is consequences (positive
or negative) for others. Norms arise when an action has similar
externalities for a set of others, yet markets in the rights of control
of the action cannot easily be established, and no single actor can
profitably engage in an exchange to gain rights of control.24 The
most important norm is generalized reciprocity, which is based on
21
Mark Granovetter, Economic Action and Social Structure: The Problem of
Embeddedness, American Journal of Sociology, 91:3 (1985), pp. 481510.
22
R. Axelrod, The Complexity of Cooperation: Agent-Based Models of Competition and
Collaboration, Princeton, Princeton University Press, 1997, pp. 4068.
23
Robert D. Putnam, with R. Leonardi and R. Nanetti, Making Democracy Work:
Civic Traditions in Modern Italy, Princeton, Princeton University Press, 1993.
24
James Coleman, Foundations of Social Theor y, Cambridge, Mass., Harvard
University Press, 1990, Chs. 8, 12.

260

GOVERNMENT AND OPPOSITION

a continuing relationship of exchange, involving mutual expectations

that a benefit granted now should be repaid in the future. 25
Communities in which this norm is followed can more efficiently
restrain opportunism and resolve problems of collective action by
reconciling solidarity and self interest. Generalized reciprocity, in
turn, is associated with dense networks of social exchange,26 through
which the core relationships between reciprocity, reputation and
trust are developed in a mutually reinforcing way.27 Thus norms,
and hence social capital, are sustained by socialization and by
sanctions. These norms facilitate the stability of intra-network
relations and hence the inbuilt capacity of institutional networks to
learn and adapt to changing circumstances.
Therefore, social capital, being initially itself a by-product of the
exchange process, is transformed into a public good, that is a
resource for action available to individual actors. In that sense, it is
viewed as the appropriate conceptual tool for introducing social
structure into the rational action paradigm and hence for bridging
the gap between rational or purposive action and social structure28
and for facilitating the socialization function. Although Colemans29
definition of social capital as a set of inherent in the social organization social-structural resources that constitute capital assets for
the individual implies that it refers to individual actors (persons),
it has been acknowledged as a crucial factor for facilitating collective
action among corporate actors as well: because purposive organizations can be actors just as persons can, relations among corporate
actors can constitute social capital for them as well. Hence, voluntary
cooperation is easier in institutional settings that have inherited a
25

James Coleman, 1990, op. cit.

It should be noted that generalized reciprocity as a form of social capital
constitutes the most important prerequisite for the process of political exchange. Given
that the norm is rooted in the complexities of social exchange in the broad sense, it
is considered as a crucial function in which the process of political exchange is
embedded. Thus, in the field of regional policy, generalized reciprocity, by sustaining
the process of political exchange among the actors at the regional level, is viewed as
a precondition for network creation and institution-building (see Cooke and Morgan,
1998, op. cit.)
27
Elinor Ostrom, A Behavioral Approach to the Rational Choice Theory of
Collective Action, American Political Science Review, 92:1 (1998), pp. 122.
28
Elinor Ostrom, Self-organization and Social Capital, Industrial and Corporate
Change, 4:1 (1995), pp. 13159.
29
J. Coleman, 1990, op. cit.
26

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

261

substantial stock of social capital, and the pursuit of collective goods

is not seen as in contradiction to the pursuit of maximizing individual wealth. It is in that sense that both Granovetters embeddedness argument and Colemans theory of collective action are
seen as integral parts of the behavioural revolution in political
science, which focuses on the development of a new theoretical
orientation that views social structure and institutional performance
as interdependent concepts.
To sum up, social capital and institutional networks constitute
important components of the socialization function, the former by
inf luencing actors preferences and identities, and the latter by
structuring the exchange process. Subsequently, social capital
constitutes a semi-independent variable (in the sense that it depends
on the exchange process) that, by affecting the formation of actors
preferences, facilitates the stability of intra-network relations and
hence the learning and adaptation processes within institutional
networks, which, in turn, function as an intervening variable between
actors preferences and policy outcomes. As crucial components of
the learning process, social capital and institutional networks are
introduced as the appropriate concepts for softening the rational
choice institutionalist assumptions about institutional thinness.
Hence, they may be seen as important conceptual tools for bridging
the gap between the rational choice and the historical/sociological
institutionalist approaches. The latter, without denying the rational
and purposive character of human behaviour, emphasize path
dependence and unintended consequences as features of institutional development,30 and the role of cultural norms and social
appropriateness in affecting individual action.31 Thus, they adopt a
rather thick interpretation of institutions, which are defined
30

Richard Rose, Inheritance Before Choice in Public Policy, Journal of Theoretical

Politics, 2:3 (1990), pp. 26391; Kathleen Thelen and Sven Steinmo, Historical
Institutionalism in Comparative Politics, in Sven Steinmo, Kathleen Thelen and Frank
Longstreth (eds), Structuring Politics: Historical Institutionalism in Comparative Analysis,
Cambridge, Cambridge University Press, 1992; and Paul Pierson, Increasing Returns,
Path Dependence and the Study of Politics, Jean Monnet Chair Papers 44, Florence,
European University Institute (RSC).
31
James March and Johan Olsen, Rediscovering Institutions: The Organizational Basis
of Politics, New York, Free Press, 1989; Paul DiMaggio and Walter Powell, Introduction,
in Powell and DiMaggio (eds), The New Institutionalism in Organizational Analysis,
Chicago, University of Chicago Press, 1991.

262

GOVERNMENT AND OPPOSITION

as the formal and informal procedures, routines, norms and conventions embedded in the organizational structure of the polity
or political economy; and they are viewed, primarily by the sociological and secondarily by the historical school, as independent
variables, which affect actors perceptions about their interests and
identities.
In the field of EU policy-making, social capital and institutional
networks play an important role in building forms of collective
governance at the supranational, national and subnational levels:
the former as a by-product of the interactions among the actors,
and the latter as a tool for structuring actors preferences. Thus the
planning and monitoring procedures of the Operational Programmes (OPs) of the Community Support Frameworks (CSFs) as well
as of Community Initiatives (CIs) involve dense policy networks,
both vertical and horizontal. The former link actors at different
levels (supranational, national, subnational) of government, while
the latter ref lect partnerships among local actors involved in
Structural Funds programmes. Social capital, on the other hand, is
widely recognized as an intrinsic element of the institutional
infrastructure that sustains political and technological innovation,
competitiveness and hence convergence of European regions, such
as BadenWrttemberg, EmiliaRomagna, or Wales.32 Yet, the
arising crucial issue is related to the role of history and path dependence logic in the creation of social capital and hence in the enhancing of the learning process. The intrinsic element in the
institutional learning evolutionist approach does not contradict the
path dependence analysis, since the function of learning to
cooperate33 should be considered rather as an evolutionary process
and, in that sense, consonant with historical institutionalism.34
However, it should be distinguished from the deterministic interpretations of history, since it is based on the process of making
collective action a rational choice. Additionally, the notions of civic
32

See Cooke and Morgan, 1998, op. cit.

Charles Sabel, Studied Trust: Building New Forms of Cooperation in a Volatile
Economy, in R. Swedberg (ed.), Explorations in Economic Sociology, New York, Russell
Sage Foundation, 1993.
34
R. Rose, 1990, op. cit.; Paul Pierson, The Path to European Integration: A
Historical Institutionalist Analysis, Comparative Political Studies, 29:2 (1996), pp. 123
63; Simon Bulmer, New Institutionalism and the Governance of the Single European
Market, Journal of European Public Policy, 5:3 (1998), pp. 36586.
33

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

263

engagement and strong civil society, based on the presence of

social capital, constitute intrinsic elements of Western culture,
which cannot be confined within the dualism of the rationalitybased models of markets and hierarchies. 35 In this respect,
Sabels optimism,36 based on the notion of studied trust seems to
be relevant. It points to the bottom-up process for the creation of
social capital and hence redefines the role of public policy in
encouraging initiatives, rather than imposing collective action and
coordination.

STRUCTURING MULTI-LEV EL GOV ERNA NCE IN EUROPEA N

REGIONAL POLICY

On the above considerations, the theoretical model of a learning institutional infrastructure in the multi-level system of
governance of European structural policy is based on a regional
or local functional network. Such networks, which tend to have
a horizontal rather than a vertical structure and facilitate collective action by combining governmental and sectoral local
actors and hence shaping the public/private relations and incorporating multiple-type resources, are considered as providing
the appropriate organizational structure for learning and
adaptation.37 Where there is a lack of this type of network, as
in the Objective 1 regions of the smaller and more centralized
member states, the adopted development strategies are usually
driven by the central state administrative structures and therefore, irrespective of their effectiveness and efficiency, inhibit
35
Martha Finnemore, Norms, Culture, and the world Politics: Insights from
Sociologys Institutionalism, International Organization, 50:2 (1996), pp. 32547.
36
C. Sabels notion of studied trust constitutes his rather optimistic answer to the
question. Studied trust refers to a kind of consensus and the associated forms of
economic transactions that theoretically result from associative or cooperative or
autopoietic that is self-creating reflexive systems. These are systems in which the
logic governing the development of each of the elements is constantly reshaped by
the development of all the others: the parts ref lect the whole and vice versa. Sabels
optimistic view on the creation of trust is based on the hypothesis that trust is a
constitutive hence in principle extensive feature of social life. See Sabel, 1993,
op. cit.
37
C. J. Paraskevopoulos, 2001, op. cit.

264

GOVERNMENT AND OPPOSITION

rather than facilitate the bottom-up learning and adaptation

process.38
On the other hand, however, European regional policy is considered to be a fundamental change in the system parameters that
represents simultaneously a threat to the pre-existing institutional
arrangements and an opportunity for institution-building and
network-creation, especially in the less-favoured Objective 1 regions.
In particular, by challenging embedded structures and well-established interests at the regional and local levels, Structural Funds
programmes cause instability in the intra-network relations, which
in turn leads to resistance to change on the part of some organizations for which change means loss of security and power. In that
sense, the EU programmes open up the process of institutional
restructuring, especially in regions with poor institutional infrastructure, by weakening the position of firmly-established interests.
This process facilitates the building of new institutions, because
under specific conditions, which are relevant to the Objective 1
regions, the process of institution-building presupposes the redundancy of the old institutional infrastructure.39
Under these circumstances, the presence of multiple and collective
leadership roles within the network can prevent the de-stabilization
of the intra-network relations, while, simultaneously, the collective
response to the crisis can effectively moderate the local repercussions
of the changes. Moreover, because of the waking up and catching
up40 character of the learning process which destabilizes the relations between the transacting institutional actors, an effective leadership functions as moderator of the tensions between learning and
monitoring of the power relations that the changes may engender.
Defined in this way, leadership requires the decentralization of
power and responsibility and a high level of involvement of the
participants. On the other hand, trust provides the glue that cements
and re-stabilizes the relations between the actors and enhances the
capacity of institutional networks to adapt to changing conditions.41
38

Objective 1 regions are the less developed (GDP below 75 per cent of the
Community average). In the case of Greece, Ireland and Portugal the entire country
qualifies as an Objective 1 region. R. Leonardi, Convergence, Cohesion and Integration
in the European Union, London, Macmillan Press, 1995.
39
See M. Storper, 1995, op. cit.
40
C. Sabel, 1994, op. cit.
41
C. J. Paraskevopoulos, 1998, 2001, op. cit.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

265

Figure 1 illustrates the way in which a learning institutional network

at the regional level, based on the processes of political exchange
and socialization, can facilitate the learning and adaptation capacity
of the region within the multi-level governance structure of
European regional policy. The diagram shows the way in which
the intra-regional interactions should be shaped. There is a group
of actors (intra-regional functional network) linked by strong ties,
while some of them (3,4,5) are loosely connected with other
peripheral actors. This structure of local interactions ref lects both
Figure 1
Institutional Thickness and Learning in Multi-Level Governance Networks

the need for local leadership involving public and private actors
and the importance of linkages with other peripheral local or nonlocal actors for access to new resources (information, knowledge,
new ideas). Furthermore, the Figure shows the distribution of
power among the actors based on the process of exchange, whereby
an actors power is strengthened by his/her ability to combine

266

GOVERNMENT AND OPPOSITION

external, as well as internal linkages and, therefore, access to

multiple resources. Hence, the distribution of power within the
network plays a key role in shaping the intra-regional interactions
and achieving collective action. Finally, the role of social capital is
to facilitate collective action among the actors through the socialization process, which involves their active involvement in the
provision of public goods and services that support the local
productive system.
In conclusion, social capital and institutional networks are
identified as key components of the learning, adaptation and
Europeanization functions of subnational governments, by facilitating collective action among the actors through the processes of
political exchange and socialization. Subsequently, exchange and
socialization constitute procedural components of the learning
process. Thus, social capital, learning institutional infrastructure
and adaptation are seen as interdependent concepts within European
structural policy.

PATTERNS OF SOCIAL LEARNING AMONG GREEK REGIONS

Recent research on Greek regions (Northern and Southern Aegean

Islands) provides empirical evidence on the crucial role played by
the structure of regional and local institutional networks in facilitating or inhibiting the adaptation of the systems of governance in
Objective 1 regions to the European environment. Two reasons make
the case of Greece particularly relevant for an assessment of the
impact of social capital and learning on regional policy in the Europe
of 2000: first, Greece is characterized by a centralized state structure
and a weak civil society;42 and secondly, these features are considered
major impediments to the learning, adaptation and Europeanization
processes at the regional level, by undermining capacities for
collective action.
The stimulus for this research has derived from the politicoeconomic differentiation in the developmental path of the regions
42
C. J. Paraskevopoulos, 1998, 2000, op. cit.; Nicos P. Mouzelis, Greece in the
Twenty-first Century: Institutions and Political Culture, in Dimitri Constas and
Theofanis G. Stavrou (eds), Greece Prepares for the Twenty-first Century, Washington,
DC, Johns Hopkins University Press and Woodrow Wilson Center Press, 1995.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

267

Nomenclature of Statistical Territorial Units (NUTS 2) concerned (Northern and Southern Aegean Islands). Beyond the
repercussions of the multi-fragmentation of space, such as the fragmentation of cultural, political and economic patterns among the
islands (namely each island has its own characteristics and milieu),
there are important common features that differentiate the profile
of each region. The most challenging facets of this differentiation
refer to their prosperity and to their history. The Southern Aegean
Islands (SAI) region consists of two island-complexes and prefectures,
the Cyclades and the Dodecanese. Although there are significant
intra-regional (among the islands) differences in the rate of development, the SAI is one of the most converging regions of the country
and also has a good ranking among European regions (NUTS 2).43
The economic and administrative centre of the region is shared
among the most developed islands (Rhodes, Kos, Mykonos and
Santorini). Conversely, the Northern Aegean Islands (NAI) region,
which consists of three big islands (Lesbos, Chios and Samos), each
of which, along with some smaller islands, constitutes its own prefecture, lags behind within Greece as well as at the European level.44
The main qualitative difference in economic development
between the SAI and the NAI regions is the adjustment of the
economic structure of the former towards the development of the
service (tertiary) sector of the economy, emphasizing tourism in
particular, whereas the latter has continued to rely on the traditional
(for each island) productive sectors (agriculture with an emphasis
on olive oil for Lesbos, shipping/maritime industry for Chios and
agriculture with some small-scale tourist development for Samos),
demonstrating, in general, an incapacity for adapting to the
changing environment. Even though the economic performance of
the SAI region does not at all mean that the huge tourist industry
has resolved the development problem of those islands what
43
The regions three year (198991) GDP average in PPS (Purchasing Power
Standard) per habitant was 52.2 compared with the countrys average of 48.1 while
its unemployment rate was low in 1993 (3.6%) when compared with the countrys
average of 7.8%. See CEC, Competitiveness and Cohesion: Trends in the Regions, Fifth
Periodic Report on the Social and Economic Situation and Development of the Regions
in the Community, Brussels, 1994.
44
Indicative figures: three-year average in PPS per inhabitant is just 35.2 compared
with the 48.1 country average, while its 1993 unemployment rate is one of the highest
in Greece, 9.0%. See CEC, 1994, ibid.

268

GOVERNMENT AND OPPOSITION

remains to be seen is whether the tourist sector in these islands will

be capable of adapting towards more f lexible forms of leisure the
existing gap in the levels of development may be attributed to this
difference in the adaptation process.
On the other hand the Dodecanese, arguably the more prosperous
of the two island complexes of the SAI, was incorporated into Greece
only in 1947, being until then under Italian rule, and some of the
islands in the Cyclades complex (Syros) have had strong traditions
of trade and cultural relations with Western Europe,45 while the
NAI followed the path of other Greek territories, being under
Ottoman rule till the beginning of the twentieth century.
To identify the interactions between the external shock caused
by the implementation of the Structural Funds programmes and
the existing institutional infrastructure, we have carried out network
analysis46 and tried to identify the presence of social capital. The
network analysis is based on density/thickness, centrality and
structural equivalence measurements for both the processes of
general exchange and the implementation of the Structural Funds
programmes in the two regions. Density measurement refers to the
degree of connectedness of the entire network, whereby zero
indicates no connections between any actor and one means that all
actors are linked to one another. Centrality measurements reveal
actors involvement in network relations and demonstrate the
structure horizontal or vertical of the networks. Finally,
structural equivalence reveals the network structure by categorizing
actors according to their common structural positions. The presence
of social capital has been identified by data on membership in
voluntary-community organizations and qualitative analysis by
fieldwork research, because of the lack of suff icient financial
resources required for a mass survey.
The first important outcome from the inter-regional comparison
of both the density and centralization measures is that it is not the
regions but the prefectures that should be used as the appropriate
45

V. Kardasis, Syros: Crossroads of Eastern Mediterranean (18321857), Athens,

Cultural Foundation of the National Bank of Greece, 1987.
46
The network analysis is based on extensive fieldwork involving around 70 semistructured in-depth interviews with representatives of the most prominent institutional
actors in the regions concerned, and the UCINET IV Version 1.00, has been used
(S. P. Borgatti, M. G. Everett and L. C. Freeman, Columbia, Analystic Technologies,
1992).

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

269

Table 1
Density Measures for General Exchange and Policy Networks
Networks of General Exchange
Dodecanese Prefecture
Cyclades Prefecture
Southern Aegean Islands
Lesbos Prefecture
Chios Prefecture
Samos Prefecture
Northern Aegean Islands

0.727
0.545
0.367
0.418
0.528
0.595
0.237

Policy Networks
0.800
0.636
0.414
0.564
0.611
0.667
0.277

level of analysis. In the SAI region the density of the general exchange
networks is just 0.367, while the situation is slightly better (0.414) at
the policy network level. In the NAI region the situation is even
worse with the density measures ranging from 0.233 to 0.277
respectively. The general exchange and policy network centrality
measures, on the other hand, are extremely high in both regions
(83.55 per cent and 79.22 per cent in NAI and 70 per cent and 64.74
per cent for SAI respectively). What the low density rates and the
high centrality measures indicate is that at the regional level, because,
among other reasons, of the fragmentation of space, the networks
are highly centralized around the Regional Secretariat, while there
are no intra-regional networks but only those within each prefecture.
This fragmentation points initially to a highly unbounded local
governance system, given its dependence on the central state for
resources and decision-taking. What should be stressed about the interregional outlook is the different specific weight of the University of
the Aegean between the two regions. Because of its location (its basic
departments are located on the NA islands), it constitutes a more
important actor for the NAI, rather than for the SAI.
The second important feature of the institutional infrastructure,
which arises from the intra-regional (at the prefectural level)
comparisons, is the gap between, on the one hand, the Dodecanese
prefecture and on the other, initially, all the other prefectures. What
the structural equivalence of actors in the Dodecanese general
exchange network reveals is that, beyond the public actors Regions
General Secretariat and Prefecture Council which are completely
connected to all other actors and can constitute the leadership of
the network, there is a second block consisting of both public
Rhodes and Kos City Councils, Association of Municipalities and

270

GOVERNMENT AND OPPOSITION

Table 2
Centrality Measures of General Exchange and Policy Networks
Networks of General Exchange
(percentage)
Dodecanese Prefecture
Cyclades Prefecture
Southern Aegean Islands
Lesbos Prefecture
Chios Prefecture
Samos Prefecture
Northern Aegean Islands

33.33
54.55
70.00
71.11
60.71
53.33
83.55

Policy networks
(percentage)
24.44
43.64
64.74
53.33
50.00
46.67
79.22

private interest organizations: the Chamber of Commerce, Rhodes

and Kos Hotel Owners Associations and the Tourist Agents
Association. Even though they are not completely connected, with
the exception of the two marginal actors of block three, namely the
university and the Dodecanese Development Agency (DDA), they
have a good rate of linkages within the network. The marginal
character of the university and the DDA is owing to the scattering
of the university departments among the Aegean Islands in the
former, and to the specific role of the latter as an organization
created primarily for the management of Structural Funds
programmes (Community Initiatives).
This thick (Table 1) and ideally centralized (Table 2) institutional
infrastructure in the Dodecanese ref lects its strong for the Greek
case tradition of institution-building and synergistic networks
among public and private actors since its incorporation into the
Greek state in the early 1950s. Since then the city councils in the
Dodecanese have proved particularly competent in comparison with
their counterparts in other regions. 47 First, as far as financial
resources are concerned, the council tax that has been imposed
(4 and 2 per cent on the value of imports and exports respectively)
constitutes an important financial resource for all the councils of
the prefecture and points to a comparatively bounded system of
local governance. Secondly, the city councils have undertaken
47

P. Getimis, Development Issues and Local Regulation: The Case of the

Dodecanese Prefecture, in K. Psychopedis and P. Getimis (eds), Regulation of Local
Problems [in Greek], Athens, Foundation for Mediterranean Studies, 1989.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

271

pioneering initiatives in creating infrastructure and providing social

services (the Rhodes council created the first municipal transport
company in Greece RODA). Thirdly, a horizontally-structured
mini-network, comprising the Hotel Owners Association, the Tourist
Agents, the Chamber of Commerce and the Association of Municipalities and Communities, financed by a special council tax and
focusing on tourist promotion, has been created around the Rhodes
Organization for Tourist Promotion. Finally, the Dodecanese Cooperative Bank, which was founded on a Chambers initiative in
association with the Rhodes City Council, constitutes a sort of
rotating credit association and its branches network includes both
the Rhodes and Kos islands.
The structural equivalence of actors in the policy network reveals
the following features. First, with the exception of the university,
almost all other actors are connected to each other. Secondly,
because of the centralized administrative structure of the state (the
state plays the key role even in the implementation of the Community Initiatives and the Pilot Projects), public actors (Regional
Secretariat, Prefecture Council, Cities Councils) provide the
leadership of the policy network. Thirdly, the upgraded status of
the Dodecanese Development Agency (DDA), which has been
established by the Local Association of Municipalities and the
Prefecture Council and focuses on the creation of networks around
specific Community Initiatives (Leader, Interreg, Valoren etc.),
indicates its successful involvement in the Leader initiative. Fourthly,
the Chamber, the most historic and prominent private-interest
organization and the initiator of almost all the forums for information, dialogue and communication in the prefecture, has been
involved in several programmes and initiatives of European regional
policy and participated in important trans-regional networks. Finally,
most of the other local actors have been involved either in
implementing specific programmes of the CSF, or in monitoring
the implementation process (monitoring committees).
The structural equivalence of the actors in both the general
exchange and policy networks in the Cyclades reveals a similar
situation to that of the Dodecanese. Under the leading public
actors (Regional Secretariat, Prefecture Council and Association of
Municipalities), there is a group of public and private-interest
organizations (Mykonos and Ermoupolis City Councils, the Chamber
of Commerce, Hotel Owners Association and tourist agents) which,

272

GOVERNMENT AND OPPOSITION

although less connected within the network, contribute to its cohesive and horizontal character. This is not revealed in the density
and centralization measures (Tables 1 and 2) mainly because of the
marginal presence of the university. The upgraded status of the
Cyclades Development Agency (established by an initiative of
the Prefecture Council and the Association of Municipalities) and
the Ermoupolis Development Agency (established by the Ermoupolis
City Council) in the policy network points to their involvement in
the Leader and Urban Community Initiatives respectively.
In general the predominant feature of network analysis in the
SAI region is that, despite the centralized structure of the state and
subsequently the crucial role of the Regional Secretariat, the
implementation of Structural Funds programmes is a public rather
than state-driven process with a considerable presence of privateinterest organizations and public/private synergies. This relatively
good institutional structure should be attributed to the pre-existing
institutional capacity at the local level rather than to specific
characteristics of the role of the state in the region.
The situation in the prefectures of NAI is completely different.
The actors in the general exchange networks are loosely connected
to each other, which reveals the lack of local synergies, while the
position is significantly improved only in the policy networks. What
the network analysis demonstrates is that because of the lack of
institutional thickness, the main actor is the Regional Secretariat,
whose relatively good administrative structure, in stability and
continuity, provides some explanation for the comparatively effective
response to the turbulence caused by the implementation of the
IMP and CSF. The Lesbos Chamber of Commerce and the Mastic
Producers Association of Chios are the only relatively active privateinterest organizations with some involvement in synergies, but
strictly in the process of the implementation of Structural Funds
programmes. Indeed, the former has been involved, along with the
Association of Municipalities and the Farmers Association, in the
creation of the Lesbos Local Development Agency focused on the
management of the Leader Initiative, while the latter has been
involved in the implementation of the Specific Aegean Islands
Programme, which was an integral part of the first CSF (198993).
Even though the Leader I Initiative was focused on the reorientation
of the agricultural sector in Lesbos (almost exclusively oriented
towards olive oil production), the results achieved were poor.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

273

The institutional infrastructure in Lesbos and Chios illustrates

the unboundedness of decision-taking, while, simultaneously,
providing an explanation for the lack of adaptability of the productive system to changing external conditions. The upgraded role of
the university in Lesbos (locus of two main departments) and Chios
(locus of one important department) does not correspond to an
education-services based model of development. In general, even
though the debate on the need for adaptability of the productive
system towards a combination of a f lexible small-scale tourist
development supported by the primary sector is just under way in
these islands, and despite the plethora of development agencies
involved, the real outcome of the implementation of the Community
Initiatives in the NAI is poor.
Given the lack of inter-regional networks with the exception
of the Chambers of Commerce, which have developed a network for the entire Aegean as a response to an initiative of the
Turkish Chambers across the Asia Minor coast what the analysis
thus far suggests is that there is a differentiation in institutional
capacity first, between the Dodecanese and the other prefectures,
and in a second phase between the SAI as a whole and the NAI
region, which can to a significant extent explain the differentiation
in adaptability.

SOCIAL CAPITAL IN THE AEGEAN ISLANDS

Even though in R. Ingleharts48 classification of European countries,

as regards mutual trust Greece appears to be above all parts of Italy,
research on social trust and civic engagement is completely
overlooked in Greece. Thus it has been difficult to identify clearcut differences in the levels of social capital. The collection of the
data in Table 3 has been facilitated to some extent by a research
project on voluntary organizations which was carried out in Greece.49
48
R. Inglehart, The Renaissance of Political Culture, American Political Science
Review, 82:4 (1988), pp. 120330.
49
The VOLMED research project is financed by the EU Commission (DG V) and
focuses on registering the voluntary organizations in the Mediterranean countries.
The research for Greece has been undertaken by the Panteion University of Social
Sciences (Dept. of Social Statistics); coordinator: Associate Prof. Ms Stasinopoulou.

274

GOVERNMENT AND OPPOSITION

What the data for the two regions demonstrate is an almost clear
differentiation in all the categories of voluntarist participation
between the SAI and NAI regions. Membership in professional
associations and unions is not listed in the Table, because that would
create confusion, given that membership is often compulsory. In
the NAI there is a huge number of members in agricultural
associations (20,000) and a smaller participation in trade and other
associations (12,000). Conversely, in the SAI there is a clear
orientation towards trade and services (35,000), and a smaller
number in agricultural associations (10,000).

Table 3
Membership in Voluntary Organizations 1996
(percentage of population by category)
Cultural

Athletic

Health Care Env/ment Other

Total

Chios Pref.
4.9

0.15

0.9

0.8

0.45

7.2

1.1

0.18

0.15

0.1

1.53

0.1

1.3

0.15

0.25

0.3

0.17

3.02

1.15

3.15

0.45

0.65

9.9

0.75

2.5

0.3

0.45

6.8

0.9

2.8

0.35

0.5

7.95

Lesbos Pr.
Samos Pr.
1.2
Northern Aegean Isl.
2.15
Cyclades Pref.
4.5
Dod/nese Pref.
2.8
Southern Aegean Isl.
3.4

Source: VOLMED Research Project (1996), authors research; elaborated

by the author.

Looking at the qualitative features of the data the following points

must be emphasized. First, we must note the very important presence
of health-care organizations in the Dodecanese and the Cyclades
(Syros). In Syros an almost complete network for home-care services
for disabled people has already been fully established. In the
Dodecanese the Association of Persons with Special Needs with
2,000 members operates under city council schemes, through which
it has been involved in the implementation of several Structural
Funds programmes. During the period of the first CSF (199094)

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

275

it had undertaken the implementation of the Horizon initiative in

the entire prefecture. Secondly, the important differences that evolve
from cultural traditions among the islands should be emphasized:
a) most of the cultural organizations (Lyceums of Greek Women,
theatre groups etc.) in the SAI and, particularly, in the Dodecanese
were established during the last quarter of the nineteenth century
and demonstrate a preference for Western cultural forms; b) in some
of the other Aegean islands, (i.e Syros and Lesbos50) mostly because
of the presence of a civic class and subsequently of a local leadership
with strong Western cultural orientations during the nineteenth
century, Western cultural forms were evident in the past, but they
have gradually disappeared following the collapse of the civic class
on these islands.
Thus an interpretation of the existing differences in attitudes,
mentality and orientation, which requires each regions qualitative
features to be raised, should take into account each islands
specificities. In that sense the first remarkable feature is the collapse
in Lesbos of the old civic class without any group to replace it.
Hence, in Lesbos neither leadership nor networks exist. Chios, on
the other hand, despite the presence of the shipping-owner class,
demonstrates similar problems of lack of leadership in all aspects
of public life.
The second catalytic cultural factor that has been identified is
the predominance of the Communist Party of Greece (the Stalinist
one51) in both the Lesbos and Samos prefectures. Lesbos and Samos
are considered to have been the strongholds of this Communist
Party almost since its foundation in the 1930s and definitely after
the civil war. The combination of the collapse of Asia Minor and
the subsequent wave of refugees with the strong presence of the
Communist Party has created a victimhood attitude and mentality
and a particular type of conspiratorial approach to events. The
answer of one interviewee to the question what would better
describe the region honesty or corruption?, was characteristic.
He said: neither, only misery.
50
Siphnaeou Evridike, Lesbos: Economic and Social History (18401912) [in Greek],
Athens, Trochalia, 1996.
51
In Greece, since 1968, there have been two communist parties: one reformist
and Euro-communist that has more or less followed the trajectory of the Italian PCI
and currently participates in the Coalition of the Left, and the hard-core more powerful
party, which was well-disposed towards the former Soviet Union.

276

GOVERNMENT AND OPPOSITION

CONCLUSION

This article, based on evidence from EU structural policy, has

attempted to show the role of the learning capacity in facilitating
the processes of both institutional and policy adaptation within the
multi-level system of governance of the European Union. Social
capital and institutional networks are identified as crucial conceptual
tools for the learning and adaptation processes, by facilitating
collective action among the actors through the processes of exchange
and socialization. In particular, while the main contribution of the
network paradigm is the mapping of the exchange process among
the actors, because of its capacity for capturing the system of
institutional interactions and interdependencies and therefore the
specifics of the broad structural environment, it is viewed as an
important component of the socialization function as well. On the
other hand, although social capital has been primarily conceived of
as a by-product of the exchange process, through its transformation
into a public good/resource available to individual actors, it
inf luences the process of preference formation, thus constituting a
crucial tool for the socialization function. In that sense, the concept
of social capital plays a key role as a semi-independent variable,
which, being based on the exchange process, affects crucially the
shaping of actors preferences and interests and subsequently the
system of interactions. Institutional networks, on the other hand,
function as an intervening variable between actors preferences and
policy outcomes, thus structuring the exchange process.
The empirical evidence from Greece (planning and implementation of Structural Funds programmes MOPs, CSFs) vindicates
the role of social capital and institutional networks as important
components of the institutional infrastructure that facilitate the
process of social learning, in public policy-making in general and
in regional policy in particular. They affect crucially the levels of
effectiveness and efficiency in policy-making. None the less, even
in cases where these qualitative features are missing and consequently the local institutional infrastructure is poor, the Europeanization process constitutes a positive external shock that initiates
the processes of institution-building and social capital formation.
This is achieved through the transformation of the local governance
structure (local networks), by challenging the balance of power and
hence established interests and patterns of behaviour of the actors.

SOCIAL CAPITAL, LEARNING AND EU REGIONAL POLICY

277

This function of the Europeanization process is aptly illustrated by

the case of the Northern Aegean region.

278

GOVERNMENT AND OPPOSITION

Instructions for Authors

Articles submitted to Government and Opposition are refereed,
normally by members of the Editorial and Advisory Boards.
The journal also commissions some of its articles. Authors are
paid a fee of 100 on publication.
In general the length of articles should be limited to 7,000
words. An article may be divided into sections by the use of
side-headings in capital letters, not by Roman numerals. Two
copies should be submitted, typed in double spacing on one
side only of A4 paper. In addition, a copy on disc may be sent
(preferably Word 97). A short biographical note on a separate
sheet is desirable.
Quotations of more than five lines should be single-spaced
but not indented. Numbers between 20 and 999,999 (and percentages) are given in figures, one to nineteen and millions in
words. Abbreviations are avoided except in footnotes (per cent,
not %), but i.e. and e.g. are acceptable. Single quotation marks
are used (double for quotations within quotations). Words to
be printed in italics should be underlined. Spelling: in general
the ize form is preferred to ise.
Footnotes
N.B. We do not use the Harvard style of citation and
do not publish bibliographies. It is preferable for footnotes to
be on a separate sheet at the end of the article, rather than
on each page. References in footnotes to books should include
author, title (underlined), place of publication, name of
publisher, date, pp. (in that order). (In subsequent references
to the same work, ibid. and op. cit. can be used.) References
to articles should include author, title (in single quotation
marks), title of periodical (underlined), number of volume and
issue, year of publication, pp. (in that order).