ArticlePDF Available

Abstract and Figures

Using social network and proximity-based approaches, this paper analyses the effect of two territorial governance arrangements implemented in the Amazonian region in the framework of ‘The Territories of Citizenship’ programme, which aims to promote interaction and collaboration between the various actors of rural development. In the Amazon region, the great distances between actors made their interactions difficult. Our study reveals, however, that those distances are not insurmountable obstacles to collaboration. Indeed, the measures implemented in the framework of territorial policies promote communication between the different actors of the territories, thus allowing for the development and reinforcement of organized proximity.
Content may be subject to copyright.
Territorial governance and proximity dynamics. The case of two
public policy arrangements in the Brazilian Amazon
Etienne Polge, André Torre
UMR SAD-APT, INRA - AgroParisTech, Université Paris-Saclay, 16, rue Claude Bernard, 75231 Paris Cedex 05,
France (e-mail:,
Received: 3 March 2017 / Accepted: 6 June 2017
Abstract. Using social network and proximity-based approaches, this paper analyses the effect
of two territorial governance arrangements implemented in the Amazonian region in the frame-
work of The Territories of Citizenshipprogramme, which aims to promote interaction and
collaboration between the various actors of rural development. In the Amazon region, the great
distances between actors made their interactions difcult. Our study reveals, however, that those
distances are not insurmountable obstacles to collaboration. Indeed, the measures implemented
in the framework of territorial policies promote communication between the different actors of
the territories, thus allowing for the development and reinforcement of organized proximity.
JEL classication: D85, R58, O38, D7
Key words: Territorial governance, institutional arrangement, social networks, proximity,
Amazon, Brazil
1 Introduction
Changes in the implementation of public policy, adopted in a growing number of sectors and
territories, have contributed to the emergence of notions such as territorial governance or
participatory democracy, reecting the desire for a better integration and participation of local
stakeholders in development policies. Given the intersectoral nature of those policies and the
fact that they involve the participation of local actors, there is now a need to develop governance
mechanisms that respect the principles of subsidiarity and ensure the integration of local actors
representatives in decision-making processes (Torre and Traversac 2011). However, while these
mechanisms appear to be appropriate means of ensuring the sustainable development of
territories and of coping with the challenges associated with globalization, the benets to be
drawn from those mechanisms clearly depend on the capacity of the actors of the territory to
co-ordinate their action (Storper 1997).
Regional science approaches that address the issue of local interactions consider that space
plays an essential role in co-ordination processes. Geographical proximity, dened as the
physical distance between two entities, weighted by the cost in time and money of covering that
© 2017 The Author(s). Papers in Regional Science © 2017 RSAI
Papers in Regional Science, Volume •• Number •• •• 2017.
distance (Torre and Rallet 2005), can reinforce interactions between actors and generate positive
externalities, such as a reduction of transaction costs (Scott 1986), the development of innova-
tions (Camagni 1995) or the production of non-relocatable know-how (Colletis-Wahl et al.
2008). However, activating the benets of this proximity depends on the social relations devel-
oped between the actors. Thus, the spatial interactions between actors can be analysed through
a dialectic between geographical proximity and organized proximity (Torre 2008).
A number of studies, using a proximity-based approach, have examined how actors interact
in the same cluster (Ter Wal and Boschma 2011), in the same economic sector (Balland et al.
2013), in research activities (Hoekman and Frenken 2013) or in situations of land use conict
(Magsi and Torre 2014). However, little empirical work has been devoted to analysing the
proximity dynamics at stake in the institutional arrangements, when they are implemented as
part of territorial governance processes. Yet, those arrangements are intended to facilitate,
through the construction of organized proximity relations, the development of interactions
between actors who did not previously collaborate spontaneously with one another, but whose
joint action can prove important in promoting local development. Co-ordination between those
actors can then facilitate the implementation of public policies and the denition of development
strategies tailored to each territory, including the provision of support to clusters, for example,
or strategies of specialization in a particular sector.
The desire or the need to give more consideration to the opinions or wishes of the popula-
tions as well as their will to participate in local actions has led policy-makers to propose new
instruments of governance of local territories (Lascoumes and Le Galès 2007). The territorial
public policy for rural areas implemented in this spirit in Brazil since 2003, has focused on
inter-communal spaces with low human development indicators. The objective was to generate
a process of social and spatial rebalancing, by providing support to family farming and rural
workersorganizations (Bonnal and Maluf 2009), so as to enable them to interact with public
institutions and to participate in public actions for local development (Piraux et al. 2013). This
policy led, in 2008, to the creation of an ambitious programme, the Territories of Citizenship,
managed locally by collegiates for territorial development or codeters. These arrangements
comprise representatives of civil and of public institutions of each municipality, in equal
proportions and aim to promote the economic development of territories, to facilitate the
populations access to federal programmes, and to help reduce poverty (Cazella et al.2013).
In particular, codeters are in charge of allocating funding for local initiatives (technical
support, structuring production chains, training, production) and, more generally, must ensure
the proper implementation of public policies, such as that of electrication, education,
healthcare and development programmes. Their mission is also to promote collaboration
between the various actors of the territories, whether they be actors of civil society, public
institutions or residents of the different municipalities. To this end, they are provided with
funding to organize regular plenary assemblies (including organizing and covering the cost of
transport, meals, accommodation, etc.), during which decisions are made concerning the
organization of joint projects.
The objective of this paper is to analyse the interactions and in particular, the proximity
dynamics, that condition collaboration among the actors involved in governance arrangements,
in order to be able to evaluate their capacity to generate collective dynamics. Our work is based
on the study of two contrasted territories beneting from the Territories of Citizenship
programme and located in the Brazilian eastern Amazon: the Baixo Amazonas and Nordeste
Paraense. Implementing public policies suited to the particular needs of each region is crucial
for the future of the territories of the state of Pará. Indeed, territorial governance schemes are
confronted with major challenges in the Amazon, because of the immensity of the region as well
as the cognitive distances between various groups of actors with different perceptions and
interpretation of various external phenomena.
E. Polge, A. Torre2
Papers in Regional Science, Volume •• Number •• •• 2017.
These challenges raise the question of the capacity of the local actors to collaborate to set in
motion a dynamic process of territorial development. In this context, we propose to analyse the
mechanisms through which the different groups of actors decide to take part in territorial
governance arrangements and what effects the latter has in different contexts. This eld of
inquiry proves particularly favourable to testing the framework of analysis of proximities,
inasmuch as the emergence of new forms of organized proximities is hindered by the great
geographical and cognitive distances separating the different actors. We seek to establish
whether this initial low level of proximity between the actors which a priori limits collabora-
tion between actors can be overcome through regular communication between them and
effective territorial engineering support. Moreover, we hypothesize that the existence of
relations other than strictly professional ones facilitates the establishment of new forms of
organized proximity.
We present the analytical framework of our study in Section 2 and the methods in the
Section 3. Section 4 is devoted to analysing the characteristics of the collaboration networks
of the actors involved, networks which have formed within the arrangements we have studied.
More specically, we examine whether new collaborative relationships have developed between
the two main categories of actors identied (civil society and public institutions), as well as
between geographically distant actors. In Section 5, we look at the elements that play a key role
in the formation of these networks. Finally, in the discussion, we compare the systems of
interactions of the two programmes and their respective evolutions.
2 Analytical framework
2.1 Social networks, proximity dynamics and territorial governance
Our objective is to analyse how the actors collaborate with one another, within the framework of
territorial governance arrangements; territorial governance understood as a set of processes and
mechanisms through which different categories of stakeholders or actors (productive, associa-
tive, individuals, representatives of public institutions, etc.) contribute to processes sometimes
negotiated, sometimes conictual of development of joint projects for the future development
of territories (Torre and Traversac 2011). Focusing on the local level enables us to study the
actorsbehaviour and to situate them within their economic and social environment.
In this study focusing on inter-individual interactions, we consider that economic relations
are embedded in social relations (Grossetti 2008). Abundant literature, particularly on network
approaches, has been produced since the works of White (1992) and his student Granovetter
(1985) were published. Their research consisted of examining in depth how social exchanges
and collective action develop within economic organizations, and of empirically analysing
social interactions and their characteristics (Favereau and Lazega 2002). Analysing interactions
makes it possible to study the structure of the social networks and to analyse the positioning of
the actors within the territorial governance structures or within the groups they belong to, for ex-
ample (Carrington et al. 2005).
The spatial dimension being considered an essential variable in the processes of
co-ordination and collaboration, we wished to incorporate it in our analysis, and to elaborate
it on the contributions of regional science. Among other things, regional science helps to
understand and analyse local economic externalities, rst highlighted by Alfred Marshall to
explain the economic performance of industrial districts, and subsequently discussed further
by different authors, including Porter (1998) in his cluster theory. In light of the doubts over
the robustness of the cluster model, some researchers wanted to better understand the
mechanisms that result from geographical proximity between local actors. They raised the
Territorial governance and proximity dynamics 3
Papers in Regional Science, Volume •• Number •• •• 2017.
question of whether geographical proximity could or could not facilitate interactions between
actors and contribute to the denition of shared projects and representations. They found that
other forms of, non-geographic, proximity may be necessary to benet from local positive
externalities (Boschma 2005; Torre and Rallet 2005).
It is only recently that studies on social networks have started to take space into account. The
importance of these networks in the functioning of regional clusters was highlighted in the
1980s (Piore and Sabel 1984; Scott 1986; Camagni 1991) in studies, which, on this basis,
empirically analysed the functioning of clusters in order to understand the conditions under
which local externalities occur as well as their effects. Beyond the mere density of market
exchanges, the researchers have focused on the social linkages that allow for processes of
economic development, through the characterization of social capital (Callois and Aubert
2007; Molina-Morales et al. 2013) or through the analysis of knowledge exchange network
evolutions and rmsabsorptive capacities (Giuliani and Bell 2005; Graf 2010, Brenner et al.
2011). Following from those works, more literature has been produced, examining the networks
most likely to support innovation and collective action and therefore to strengthen regional
clusters, especially in the rural world (Spielman et al. 2010; Crespo et al. 2012; Chiffoleau
and Touzard 2013; Compagnone and Hellec 2015).
Building on the analysis of proximity dynamics at work between the actors participating in
programmes of territorial governance and rural development, our pragmatic approach enables us
to examine more thoroughly the effects of the geographical, economic and social dimensions of
interactions on collaborations. In a sense, it is a continuation of the researches which, using
social network analysis, have tested the impact of geographical proximity on the development
of clusters (Weterings and Boschma 2009). We have chosen to use a simple framework, one that
clearly identies two main types of non-contradictory proximity that can lead actors to interact:
geographical proximity and organized proximity (in the line of Torre and Rallet 2005). This ap-
proach is well suited for addressing our questions because geographical proximity considered
as the physical distance separating individuals, weighted by their representations and practices
is essential to analysing the actions of people situated in a given spatial context and is inherent to
their relational context. The desire to move closer to other actors or places is an essential driving
force behind economic development. It may take the form of co-location, but may also occur on
a temporary or transitional basis. Temporary geographical proximity (TGP) offers individuals
the possibility to full a need for face to face interaction by travelling from one location to
another (Torre 2008).
Two logics inherent to organized proximity, likely to facilitate these interactions, have been
identied. These are the logics of similarity and belonging (Torre and Rallet 2005):
1. The logic of similarity corresponds to mental and cognitive adherence to common categories.
It can facilitate interactions between people who did not know one another before, but share
similar references. This concept is close to that of homophily or of adherence to common
categories, used in the sociology of networks.
2. The logic of belonging refers to the interactions between two or more actors; interactions
facilitated by their belonging to the same organization or to the same network, which shares
a certain number of rules and behavioural routines.
Also our approach, based on the dialectic between geographical proximity and organized
proximity, enables us to study the mechanisms of interaction and collaboration between actors
given these spatial and socio-economic dimensions (Torre 2008). Thus, we shall analyse how
these forms of proximity combine, not in the formation of a market or in the development of
innovations stemming from the knowledge economy (Bathelt and Glückler 2011), but in
construction of territories. The construction of various forms of proximity can be facilitated by
E. Polge, A. Torre4
Papers in Regional Science, Volume •• Number •• •• 2017.
ad hoc institutional arrangements, which constitute a dynamic architecture of governance and
must serve to both stimulate and support local development dynamics (Gilly and Wallet 2001).
2.2 Case studies
Our analysis is situated in the context of the Territories of Citizenshipprogramme in Brazil.
Through this voluntarist programme, a number of institutional arrangements vested with
important functions have been implemented with a view to promoting territorial development.
In Brazil, following the end of the dictatorial regime in 1984, civil society emancipated and
structured itself, with growing demands for participation in public decision-making. From 1996,
the federal government, led by the Brazilian Social Democracy Party (PSDB) at that time,
responded to those demands by making the implementation of the various federal programmes
at municipal level conditional on the creation of participatory municipal sector councils,
including councils for rural development (Schneider et al. 2004). From 2003, the limitations
of these councils, faced with the weakness of the municipal civil society organizations and
the persistence of patronage relations (Delgado and Leite 2011), prompted the new government,
led by the Workers Party (PT) close to the social movements, to carry out its actions at the level
of inter-municipal territories characterized by a low Human Development Index and by a large
family-scale farming population.
This process of territorialization of public action in the eld of rural development occurred
in three key stages (Figure 1):
1. Between 2003 and 2007, the CIAT (or Committees for the Implementation of Territorial
Actions) were implemented at inter-municipal level;
2. The 20082011 period was marked by the transformation of the CIATs into collegiates for
territorial development (codeter), which are responsible for co-ordinating the federal
governments actions in the framework of the Territories of Citizenshipprogramme;
3. The years 2012 to 2014 corresponded to the period of Paralysis, during which funding for
the programme was discontinued.
We have examined the implementation of the programme Territories of Citizenshipin the
Pará state (Figure 2), which has achieved some success in terms of participation and of the
Fig. 1. Government history and territorial governance arrangement
Territorial governance and proximity dynamics 5
Papers in Regional Science, Volume •• Number •• •• 2017.
actions implemented (Piraux et al. 2013), despite the low population density of the state,
the weakness of its institutions and the environmental challenges they face. In this state, the
emancipation of civil society materialized in the takeover of agricultural unions (formerly in
the hands of local leaders) by the rural workers, then expressing social demands. From 2003
onwards, the trade unions of rural workers, which constitute the electoral base of the PT, took
control of the CIATs. But the states government, headed by an opposition party, did not
support the territorial approach initiated by the federal government, accused of reducing the
power of mayors. The implementation of the Territory of Citizenshipprogramme in 2008
coincided with a period of political concordance between the federal government and that of Pará
state, both led by the PT. The public institutions of the state then became involved in the codeter.
Finally, the period of paralysis of the programme, from 2012 to 2014, was caused by the freezing
of federal funds and by the election of the opposition party at the head of the Pará government.
We have chosen to focus our study on two contrasting territories, in which the arrangements
were particularly active but whose congurations could foreshadow the development of two
Fig. 2. Map of the study territories
E. Polge, A. Torre6
Papers in Regional Science, Volume •• Number •• •• 2017.
different modes of co-ordination. The Baixo Amazonas region (BAM) is an immense territory
(equivalent to half the area of France, see Figure 2), with low population density. The regional
capital, Santarém, was rst colonized in the seventeenthth century by Portuguese explorers
travelling by waterways. Following the dictatorship, social movements organized themselves
into different regional co-ordination structures (rural workerstrade unions, shermen associa-
tions, associations for women, training associations, etc.), gradually constructing a shared
representation of their territory and becoming a major political force. Smaller in size and more
densely populated, the Nordeste Paraense (NP) is close to Belém, the capital of the state of Pará.
The area was colonized more recently (in the 1960s) by populations travelling mostly by road.
The social movement organized at a more municipal level and the actions carried out at the
territorial level were supported mainly by extension services.
3 Methods
In order to analyse the social networks and the factors likely to determine the collaborative
processes between actors within the two institutional arrangements, we began by studying the
different mechanisms of interaction between the actors who participate in them. Examining
social networks helps us to quantify the different types of relationships that exist between the
actors most involved in governance arrangements, their evolution and their inuence on the
logics of action.
From our initial interviews, we found that only a small number of people had participated on
an ongoing basis in the meetings of the territorial arrangements. This nding is in line with stud-
ies carried out at the level of the country (Delgado and Grisa 2014) or of the state of Pará (Piraux
et al. 2013), which reveal a high turnover of participants. They have prompted us to focus on the
actors who are genuinely involved in the arrangements, that is to say, whose participation was
signicant and whose reasoning could have been inuenced by the implementation of those
arrangements. On this basis, we have developed a method for collecting the relational data
necessary for analysing the entire networks, which can provide us with relevant information
on their structure and characteristics.
An important difculty is to dene the boundaries of the population to be studied, which can
be determined using several various tactics (Laumann et al. 1989) and at different stages of the
research. In our case, access to the attendance sheets of the various plenary assembly meetings
held between 2003 and 2014 enabled us to dene a subpopulation immediately following the
exploratory interviews. On the basis of those attendance records, we established exhaustive lists
of the names of individuals who attended at least one meeting (over 300 in each territory), and
then ranked them according to attendance numbers, which then enabled us to identify, in each of
the two arrangements, a rst subpopulation of some 30 people who participated in more than
three plenary meetings. A rst series of interviews with three or four key actors involved in each
arrangement, who were then asked to test the lists thus established and to evaluate the
participation of the other actors, enabled us to dene these subpopulations more precisely and
to reduce them to 25 actors whose presence and participation were perceived as signicant
(voicing of ones opinion, investing oneself in the implementation of actions, etc.). This can
be likened to reputational approach dened by Laumann et al. (1989). It is from these two
sub-populations that we have carried out the complete network analyses.
In each territory, we conducted interviews consisting partly of semi-directive questions of an
ethnographic nature and partly of sociometric questions. The information thus collected enabled
us to reconstruct the entire social network, such as it existed at three distinct stages in the
governance arrangement trajectory, and according to the nature of the relationships identied.
We asked each of the respondents to tell us about the background of their involvement and
Territorial governance and proximity dynamics 7
Papers in Regional Science, Volume •• Number •• •• 2017.
how they came to participate in the programme. They were then invited to evaluate their
relationships with the other interviewed participants during each of the three identied stages
of operation of the programmes: CIAT (implementation period), Codeter (full operation),
Paralysis (freezing of funding for the operation of the programme).
The sociometric questions, for each period, were related to the frequency of communication,
coded from 0 to 4 (no communication, daily, weekly, monthly, annual), face-to-face meetings
on the one hand (discussion of more than 10 minutes) and long distance communication via
ICT (telephone or messaging) on the other. The interviewees were also asked to evaluate the
absence (0) or existence (1) of collaborations, dened as joint work (excluding collective
meetings) carried out with another actor in order to realize a project. We also asked the
respondents whether they had relations based on friendship with other actors (extra-professional
exchanges, leisure, etc.) or political relations (exchanges during partisan meetings), coded 0, 1
or 2 (none, few, many). But we did not examine those relations longitudinally, as the rst
interviews showed that the relationships between the different actors did not change much.
Family ties (including cousinhood relations) or religious ties (meetings at religious events), were
not included in the quantitative analysis of networks, because they were too few to be signicant,
but were taken into consideration in the qualitative analysis.
The actors were interviewed on the basis of this questionnaire, but they were also asked to
provide additional information of a qualitative nature on the reasons for their choice. Because
of the sometimes difcult conditions in which the interviews were conducted, we were not able
to administer this questionnaire to all the respondents. In Nordeste Paraense, we administered the
questionnaire to 12 people, only qualitative questions were asked to six other actors questions
and seven were not surveyed. In the Baixo Amazonas, we administered the questionnaire to 17
actors, only qualitative questions were asked to three other actors and ve were not surveyed.
We were able to evaluate the relationship between the actors who were not administered the
questionnaire thanks to the qualitative information collected (this option was chosen in view of
the fact that most active members of the programme knew one another and that cross-sectioning
the qualitative data enabled us to infer relational information). Finally, we chose to categorize the
actors into two main groups: actors of civil society (CS) associations/co-operatives,
companies/NGOs, trade unions or actors from public institutions (PI) technicians from public
enterprises and public authorities at the three administrative levels (municipal government,
government of the state of Pará and federal government).
It should be noted that the interactions between these two groups were historically limited,
largely because of the huge cognitive distances separating the different actors: for a long time,
public institutionstechnicians and political decision-makers supported a process of rural
development based on productivist agriculture for export markets. A model contrasting with that
supported by the civil society actors involved in the programme, who advocated a model based
on family farming.
To consider the role of spatial interactions between actors (Boschma et al. 2014), we have
used proximity theory as our analytical framework, following an approach based on the dialectic
between geographical and organized proximity proposed by Torre and Rallet (2005). We
distinguish permanent geographical proximity (PGP), which implies a co-location of the actors,
from temporary geographical proximity (TGP), which implies a need for the actors to travel
substantial distances in order to meet others.
In order to measure TGP, we constructed an indicator that weights distance with the time
needed to cover the space separating the actors and the means of transport used in order to reach
their destination. The indicator shows values ranging from 0 (when the actors are in geograph-
ical proximity to each other, which corresponds to a situation of PGP) to 6 (for the largest
distance separating the actors (Table 1). In the NP, the score of the indicator varies according
to the time needed by an actor to travel by bus (a well-developed transport system) to meet
E. Polge, A. Torre8
Papers in Regional Science, Volume •• Number •• •• 2017.
another. In the BAM, the large variety of transport option prompted us to consider not only the
time necessary to reach a destination but also the cost and arduousness of the trip. TGP scores
can add up when an actor has to use several transport methods to reach his destination. The
analysis grid used in our surveys is summarized in Table 2, which lists the variables of
geographical proximity and organized proximity.
In order to analyse how collaborative relationships develop between actors, we studied the
correlations between the different variables: collaborations, on the one hand, face-to-face or
distance communication using ICTs, temporary geographical proximity (TGP), relations of
friendship and political relations on the other. All the networks were symmetrized (non-directed
links), by keeping only the highest score when two actors evaluated their relationship differently.
We then conducted a longitudinal study of the mean scores of the variables (equivalent to
the density for the symmetric networks), before analysing the interdependencies between the
different variables, in order to identify those that lead the actors to collaborate. For this purpose,
we have applied the quadratic assignment procedure (QAP), using the social network analysis
software, Ucinet (Borgatti et al. 2002), which makes it possible to test whether the association
between two networks is statistically signicant. More precisely, we apply the logistic regression
quadratic assignment procedure (LR-QAP), which consists of a multiple logistic regression of a
dependent variable (here the collaborative network) on independent variables (here TGP,
networks of friendships and networks of political relationships, and the group). These variables
can be social networks (collaborations, friendships, political relationships, face-to-face, ICT) or
present other characteristics (TGP, group). The successive introduction of the independent
variables makes it possible to nely assess their relative inuence on the dependent variable
while comparing the results of the two arrangements, for the periods identied, enables us to
evaluate the inuence of contexts and trajectories on the networks thus studied.
Table 1. Indicators of geographical proximity
Score Time and means of transport
0 Same municipality (PGP)
1 1 to 2.5 hours of transport by bus or high speed motor boat
2 2.5 to 4 hours of transport by bus or high speed motor boat
3 4 to 7 hours of transport by bus, high speed motor boat, or a one night boat trip
4 15 to 24 hours of transport by boat or one hour by plane
5 24 to 30 hours by boat
6 Results of the combination of transport means (added scores)
Table 2. Survey analysis grid
Variables Operationalization Indicator amplitude
Collaboration Collective work, other than collective meetings 1
Geographical proximity TGP Time necessary to travel to the location of
another active member
Face to face Frequency of face to face meetings lasting
more than 10 minutes
Organized proximity ICT Frequency of telephone or email exchanges 4
Friendship Extra professional relationships 2
Politic Exchanges during caucus meetings 2
Status Representatives of civil society or
public institutions
Territorial governance and proximity dynamics 9
Papers in Regional Science, Volume •• Number •• •• 2017.
4 Analysis of collaborative networks
The main objective of the territorial governance arrangement put in place by the public authorities
in the territories of citizenship is to increase the amount of collaboration between actors in order
to promote new joint actions for development. One of its missions is to encourage collaborations,
which would not develop spontaneously in its absence.
To test the impact and the possible success of the system, we rst analyse data sets on entire
networks (Figure 3), from which we could conduct multiplex and longitudinal analyses.
Then we examine their progression according to the two potentially limiting factors we have
identied: (i) the cognitive distances between the actors of civil society on the one hand and of
public institutions on the other, and (ii) the geographical distances between the different actors,
whatever groups they belong to.
4.1 Progression of the collaborations in the two governance structures
Figure 4, which shows how the collaborative networks developed over the three periods we
have identied, reveals a strong underlying tendency: an increase in the amount of collaboration
during the codeter period and a decrease following the paralysis. Thus, we can immediately
conclude that the public governance arrangement contributed to building collaborative relationships,
which have subsequently suffered from the programme being put on the back burner.
There are, however, some differences in the evolution of the relationships according to
territories. In the Baixo Amazonas (BAM), the collaborative relationships strengthened during
the codeter period, whereas they weakened during the period of paralysis. In the Nordeste
Paraense (NP) on the other hand, collaborations developed during the CIAT but increased
slightly but less signicantly during the codeter period, until they unraveled almost entirely
during the period of paralysis in the system. On the whole, and in the two territories studied,
the networks of collaborations grew denser following the implementation of the Territories
of Citizenshipprogramme (the codeter period) but considerably weakened following the
paralysis of the system.
4.2 Collaborations between the actors of civil society and the public institutions
A key mission of the governance arrangements we have studied is to facilitate the collaborations
between the actors of civil society (associations/co-operatives, companies/NGOs, trade unions)
Fig. 3. Networks of collaboration between the active members of the governance arrangement during the Codeter period
E. Polge, A. Torre10
Papers in Regional Science, Volume •• Number •• •• 2017.
on the one hand and public institutions on the other hand (extension services and public
authorities at the 3 administrative levels: municipal governments, Pará state government and
the federal government). To analyse their process of development, we measured the number of
collaborations between both groups as well as within each group, for the three periods identied
(Figure 5), in order to assess changes in the weight of similarity (or homophily) effects, which
cause actors to preferably form relationships with people who resemble them.
The results converge with the more general ones obtained previously. For the BAM, there is
little collaboration between the two groups of actors, as well as within the public institutions (PI)
group during the CIAT period, while there appears to be more collaboration within the Civil
Society (CS). The increase in density of collaboration networks during the codeter period, as
highlighted above, is mainly due to the increasing collaboration within the PI group, but also
between the participants of the two groups. During the paralysis period, collaboration decreased
signicantly within PIs but remains higher than the collaboration within the CS, while the
Fig. 4. Collaboration network changes in the Baixo Amazonas and the Nordeste Paraense
Fig. 5. Collaboration density within civil society (CS) and public institution (IP), and between the two groups
Territorial governance and proximity dynamics 11
Papers in Regional Science, Volume •• Number •• •• 2017.
density of collaborations between CS and PI returned to its initial level. For the NP region, the
measures shows that internal collaboration within each group and collaboration between the two
groups are slightly higher than those observed in the BAM region, during the CIAT period. All
densities somewhat increase during the codeter phase, especially the density of collaboration be-
tween the two groups. During the last period the collaborative relationships strongly weakened,
though to a lesser extent in the case of collaboration within PIs.
In sum, collaboration evolves in a similar way in the two territories, although with a different
intensity. We also note that the implementation of the codeter had a positive effect, above all, on
collaboration between the two groups, and therefore helped to open up the relationships between
people belonging to different social worlds. However, this positive inuence waned when the
programme came to a halt.
4.3 Collaboration between geographically distant actors
The second important mission of the programme is to encourage geographically distant actors to
collaborate. In order to measure the effect of the arrangement on collaboration, we conducted a
test of correlation between Temporary Geographical Proximity and collaborations (Table 3),
applying the QAP procedure, to assess whether the collaborative relationships between distant
actors formed with the implementation of the codeter and were maintained during the paralysis
of the system. The βTGP coefcient reects the inuence of the TGP variable on the
collaborations: the more negative the coefcient, the more difcult it is for co-operative relationships
to develop between distant actors (Figure 6); the closer it is to 0, the weaker the effect of distance on
Results for the BAM territory seem counter-intuitive at rst reading. First of all, we observe
that the negative inuence of distance on collaborations, which is signicant during the CIAT
period, becomes slightly more marked during the codeter period: thus, collaboration occurs
Table 3. βTGP coefcient of correlation between collaboration and TGP
Period CIAT Codeter Paralysis
Territory BAM 0.262*** 0.284** 0.084
NP 0.439** 0.027 0.356***
Notes: N = 25, standard deviation: ***p<0.001; **p<0.01; *p<0.05.
Fig. 6. Graphic representation of changes in the βTGP coefcients
E. Polge, A. Torre12
Papers in Regional Science, Volume •• Number •• •• 2017.
primarily between actors who are geographically close to one another, which seems to be
explained by the large size of the territory. But, subsequently, during the period of paralysis,
the actors began to collaborate with one another in the same way whether they are distant or
close, although one could have expected the collaboration between distant actors to decrease
due to the fall in resources displacements. At this level of analysis, it is impossible to identify
the factors that lead distant actors to collaborate more; further investigation is required.
In the NP, on the other hand, our initial intuition is conrmed: distance negatively impacts
collaboration in the rst period. But its inuence is no longer signicant during the codeter
period, which seems to indicate the advantage of policies providing support for travel expenses
and temporary geographical proximity. The inuence of distance becomes signicant again
during the paralysis period, but less markedly than in the initial period, following the reduction
in subsidies for travel.
We have seen that the implementation of the programme had a relatively similar overall
effect in both territories in terms of collaboration. However, some differences can be observed:
there was a sharp rise in intergroup collaboration during the codeter period in the BAM territory,
but a much less signicant increase in the NP territory; the rise in collaboration between distant
actors occurred during the period of paralysis in the BAM territory during the codeter period
in the NP.
These results are, in part, related to political and institutional changes in the two territories
(Figure 1). However, further investigation is necessary in order to better understand the contrast
effects of the programme, especially at the spatial level. Indeed, factors relating to the modes of
interaction and the logics of belonging may also explain why the mechanisms of co-ordination
between the actors participating in the programme, evolved differently. Let us now examine
them in detail.
5 Analysis of the construction of collaborations in the territorial governance
Now that we have described and analysed the collaboration networks of the two territories and
their process of development, particularly in terms of collaboration between actors of different
groups and geographically distant from one another, we shall attempt to gain a deeper
understanding of how these collaborative relationships developed or waned. To this end, we
analyse the impact of the various proximity variables identied above (Table 2) face-to-face
and remote communications via ICT, TGP, friendship and political ties on the level of
collaboration between actors.
We rst attempt to understand the development of co-ordination between actors participating
in the programmes, by analysing the factors that cause their active members to collaborate.
Tables 4 and 5 present the coefcients γof correlation between the collaborations (dependent
variables) on the one hand, and the independent variables friendship, politics, face-to-face, PGT
and group, on the other hand calculated by applying the LR-QAP statistical procedure
described above. For simplicitys sake, we shall give each coefcient the same name as that of
the independent variable involved. The higher the coefcient, the greater the likelihood of the
independent variable having a positive effect on collaboration.
In order to determine the precise measurements necessary to understand the mechanisms
that lead actors to collaborate, we tested different combinations of independent variables and
constructed four explanatory models. Model 1 includes the two independent variables, which
reect the two key missions of the programme: facilitating collaborations despite differences
in group and the geographical distance separating the actors (TGP). In model 2, we tested the
inuence of the different modes of communication by including the independent variables
Territorial governance and proximity dynamics 13
Papers in Regional Science, Volume •• Number •• •• 2017.
Table 4. Coefcient γof correlation between collaboration and the independent variables, in the BAM governance arrangement
Period CIAT Codeter Paralysis
Model 1 2 3 4 1 2 3 4 1 2 3 4
TGP 0.410** 0.253 0.455*** 0.370* 0.285** 0.112 0.24* 0.064 0.410*** 0.246 0.201* 0.229
Face to face 1.752*** 2.026*** 1.196*** 1.291*** 3.345*** 3.406***
ICT 0.727*** 0.288 0.941*** 0.578*** 0.408 0.396
Friendship 2.509*** 1.974** 2.111*** 0.841* 0.769* 0.674
Politic 0.38 0.313 1.112*** 0.883*** 0.911*** 0.172
Group 0.352 0.266 0.165 0.058 0.141 0.05 0.244 0.01 0.352 0.404 1.003** 0.385
Notes: N = 25, for standard deviation: ***p<0.001, **p<0.01, *p<0.05.
Table 5. Coefcients γof correlation between collaborations and the independent variables, in the NP governance arrangement
Period CIAT Codeter Paralysis
Model 123412341234
TGP 0.511** 0.042 0.381 0.05 0.048 0.002 0.149 0.052 0.929*** 0.406 0.815*** 0.438
Face to face 1.534*** 1.545*** 1.288*** 1.22*** 1.196*** 1.169***
ICT 0.670*** 0.537*** 0.743*** 0.664*** 0.770*** 0.821***
Friendship 0.928*** 0.473* 1.034*** 0.252 0.77*** 0.162
Politic 0.508** 0.082 0.3 0.042 0.092 0.291
Group 0.224 0.282 0.474* 0.395 0.238 0.072 0.043 0.129 0.587* 0.481 0.279 0.623
Notes: N = 25, for standard deviation: ***p<0.001, **p<0.01, *p<0.05.
E. Polge, A. Torre14
Papers in Regional Science, Volume •• Number •• •• 2017.
face-to-face and ICT. In model 3 we tested the inuence of networks of friendship and political
ties. Model 4 includes all the variables, which enables us to assess which factors have the
highest correlation with collaboration.
5.1 Construction of collaborative relationships in the programme implemented in the Baixo
Amazonas region
Let us rst analyse the correlations between the levels of collaboration and the various explana-
tory variables in the BAM arrangement (Table 4). The coefcients group (representative of Civil
Society or of Public Institutions) and TGP often appear to be insignicant (p >0.05) or to have
less explanatory value (low coefcients) than the variables modes of communication (face to face
and ICT variables) or networks of friendship and political ties.
Let us now examine the face-to-face and ICT coefcients in Model 2 (Figure 7) as well as the
Friendship and Political ties coefcients in model 3 (Figure 8) so as to account for changes in
them in the three periods considered. We observe that the face-to-face coefcient, which is high
in the CIAT period, decreases during the codeter period, despite an increase in communication
and collaboration. It then rises strongly in the period of paralysis, surpassing its initial level.
On the other hand, the ICT coefcient, which changes in the opposite direction, is only
signicant during the Codeter period. The friendship coefcient, which is high during the CIAT
period, decreases during the Codeter period, whereas that of the political ties, initially
insignicant, increases markedly thereafter. Following the paralysis of the programme, the two
coefcients decrease but that of the Political ties reaches a higher level than that of the friendship,
which drops sharply. It is, therefore, the political ties, which rest above all on face-to-face
interactions, which seem to contribute the most to maintaining collaboration between the active
members of the system, despite its paralysis. These results must be seen in relation to the fact that
the actors begin to collaborate in an undifferentiated manner whether they are distant from or
close to each other (see subsection 3.3).
Examining the correlations between the different variables reveals the strong but decreasing
inuence of the frequency of face-to-face interactions on collaborations, which decrease in
favour of ICT-based communication, during the codeter period. At the same time, the inuence
of the network of political ties on the development of collaborative relationships increases, while
that of the friendship network decreases. With change in the political circumstances, in which the
Fig. 7. BAM - graphic representation of changes in the coefcients γof model 2
Territorial governance and proximity dynamics 15
Papers in Regional Science, Volume •• Number •• •• 2017.
same political party (Workers Party) took the reins of both the governments of the Pará state and
of the federal state, the programme seems to reinforce a particular political group, in which actors
continue to collaborate regardless of the distance between them and the type of institution they
belong to, thus limiting the openness of the system. Thus, the collaborative relations that survive
are organized and conducted during meetings between members close to the political party in
place (strong correlation between collaboration and frequency of face to face interactions, on
the one hand, and collaboration and political ties on the other). This phenomenon, which leads
to a monopolization of the resources obtained through by the programme (in particular funding),
raises the question of the appropriateness of the territorial governance system, and in particular of
the power and credit distribution mechanisms.
Thus, the apparent paradox of maintaining collaborations despite distance and the withdrawal
of the funding for travel expenses (see subsection 3.3.) can be explained by the existence of
strong ties between actors of the same political tendency, who continue to meet frequently and
collaborate despite the geographical distance between them. The difculty caused by distance
is overcome, here, by the existence of ties of organized proximity built during the previous stages
(the logic of belonging) and resulting from the history of the local institutional actors (the logic of
5.2 Construction of collaborative relations in the system implemented in the Nordeste
Paraense region
Let us, nally, analyse the correlations between the collaborations and the various explanatory
variables in the arrangements implemented in the NP region (Table 5). As in the BAM, the
group and TGP coefcients are often insignicant and less explanatory than those related to
the modes of communication or to the inuence of the networks (friendship ties, political ties).
Let us examine the face-to-face interactions and ICT coefcients of model 2 (Figure 9), as
well as the friendship ties and political ties coefcients of model 3 (Figure 10), in order to account
for how they change over the three periods. We nd that collaboration within the arrangement
continues to take place primarily between people who communicate face-to-face (coefcient
remains high) on a regular basis, but that this correlation weakens over time, while the correlation
with ICT-based communication increases. Furthermore, the friendship ties coefcient increases
during the codeter period, while that of the network of political ties decreases steadily. With
Fig. 8. BAM - graphic representation of changes in the coefcients γof model 3
E. Polge, A. Torre16
Papers in Regional Science, Volume •• Number •• •• 2017.
the paralysis of the system, the role of friendship ties primarily based on face-to-face
interactions in collaboration decreases but remains important, while that of political ties
become negligible.
Examining the correlations between networks reveals that collaborations, less dependent on
the frequency of face-to-face communication and political ties during the codeter period, are
more strongly correlated with friendship ties and ICT-based interactions. During the paralysis
of the system, correlation with friendship and political ties and face to face communication
weakens, while the correlation with ICT increases. Thus, collaborative relationships are main-
tained at a more local level between members of public institutions partly through friendship
ties and ICT-based communication, which allow for the continuation of the projects initiated
through the territorial governance arrangement whereas collaboration with distant actors
weakens. The dynamics of organized proximity, thus reinforced, developed more between actors
linked by a logic of similarity (due to their group as representatives of public institutions),
activated in particular by a project of support to the technicians of the territory, which endowed
them with a leading role in the arrangement.
Fig. 9. NP - graphic representation of changes in the coefcients γof model 2
Fig. 10. NP - graphic representation of changes in the coefcients γof model 3
Territorial governance and proximity dynamics 17
Papers in Regional Science, Volume •• Number •• •• 2017.
Analysing the networks of actors has enabled us to assess, in an empirical and quantitative
manner, how the actors co-ordinate their actions in both territories, and to draw some general
conclusions about how the implementation of the public governance programme of the
territories of citizenshiphas facilitated collaboration between the actors. It has also helped
us to identify the factors that inuence collaborations and promote (geographical or organized)
proximity relations between local actors, as well as to reveal certain differences between the two
territories, which we shall now attempt to explain.
Our study shows that when public policies truly promote territorial development, the actors
succeed in overcoming the initial divide between public institutions actors and the civil society
actors and in collaborating, despite the geographical distances between them. These collabora-
tions develop more between actors who have formed relations of organized proximity, through
political or friendship ties. However, those collaborative relationships can prove fragile and
wane in the absence of support teams and can turn into relations from which other actors are
excluded (political relations in the BAM, or between technicians and close actors in the NP).
We also note that the use of ICTs, which played an important role in collaboration during the
codeter period, became essential in enabling the technicians of the public institutions to continue
to work together, even when, as a result of the paralysis of the system, the latter could no longer
meet face-to-face on a regular basis.
Our analysis of the two territorial governance arrangements highlights the existence of
strong collaborations between the actors of civil society and the representatives of public
institutions mostly technicians during the most active phase of the territorial development
programme. Those actors have a high capacity to collaborate in the BAM and the collaboration
between members of civil society was maintained thanks to political meetings held on a regular
basis, despite the paralysis of the system. This situation has revealed the importance of
temporary geographical proximity and face to face relationships for the survival of collabora-
tions in a time of crisis, but also of the logics of belonging of individuals with the same political
belief system, on which those meetings depend. In the NP, on the contrary, the unions have
tended to rely on the technicians to develop local projects launched through the territorial
governance arrangement but were soon unable to continue collaborating from a distance.
Despite the immensity of the BAM region, the Territories of Citizenshipprogramme was
able to rally a large number of people thanks to the involvement of the regional co-ordination
teams of social movements, who identify with the territory, and to the funds provided for
organizing meetings and for travel expenses. Nevertheless, the involvement of the technicians
was too limited in time for the projects to be carried out and completed. In the NP, on the other
hand, although the distances between actors were smaller, the volunteer involvement of a few
personalities was not enough to create and especially to consolidate relations based on organized
proximity. Thus, it appears that although a number of activities and one-off projects have been
carried out within the framework of the programme, the current delimitation of the territory does
not really correspond to a collective identity.
6 Conclusion
The research presented in this paper has aimed to explore the question of collaborations within
institutional arrangements of territorial governance. It was based on a study of the implementa-
tion of a public policy intended to reinforce development, in two territories of public action
situated in the Brazilian Amazon. Building on a theoretical and methodological framework
for analysing how the actors of institutional arrangements of territorial governance co-ordinate
E. Polge, A. Torre18
Papers in Regional Science, Volume •• Number •• •• 2017.
their actions, we were able to provide some insights into how those arrangements have evolved
and into how local actors appropriate them.
Our work shows that the development of collaborative projects is strongly dependent on
co-ordination support structures. While collaborative relationships are weaker between public
institutions and civil society on the one hand, and between geographically distant actors, on
the other, our social networks analysis reveals that collaborations within a territory are not only
correlated with the frequency of face-to-face interactions, although the latter play an essential
role. Indeed collaborations can also develop through long distance communication, via ICT,
once the actors know each other and are used to working together. Other factors play an
important role in collaborations as well: co-ordination between members of the same group,
but also the existence of relations based on the logic of belonging between members of the same
friendship or political networks. However, although, following the discontinuation of funding in
the BAM region, the relations based on a logic of belonging helped to maintain collaborations,
they can also have an excluding effect and cause the actors to form relationships exclusively
based on a logic of similarity on socio-economic attributes, for example.
These dynamics of interaction are particularly apparent in our case study but they can be
extended to the situation of other regions or local systems with similar context. As a matter of
fact, our analysis does provide some insights into the logics of co-ordination at work in
institutional arrangements of territorial governance. Especially, we have been able to put some
lights on the relations between distant actors and the way they behave by means of ICT, and to
show the modus operandi of the two logics of belonging and similarity of the organized
proximity relations. We also show that the implementation of institutional arrangements
facilitate the collaboration between actors, be they separated by long distance or belonging to dif-
ferent institutional or cognitive worlds. Our work also provides some evidence that continuous
institutional support must be afforded by the state or public authorities in order to avoid exclusion
Moreover, it contributes to the growing literature on interaction dynamics in clusters,
enlarging the focus to a territorial level and operates the junction with another important
thematic of the regional studies which is the territorial governance arrangement studies.
However, our study has some limitations. First, the method used for selecting the sub-
populations restricted us to choosing among the active members of the programme. Analysing
the networks of all the participants would have provided explanatory data on the mechanisms
of exclusion of certain groups of actors. A second limitation is related to the collection of
relational data. Due to the local conditions, we were not able to administer the questionnaires
to all the actors in the identied subpopulations, which forced us to infer, from more qualitative
data, the existence of a number of relations and to abandon the directed nature of the linkages,
reducing the scope of structural analysis. Finally, the longitudinal network analysis was carried
out a posteriori, creating biases related to the difculty of the actors to remember perfectly these
past relationships, even though they can identify with accuracy the main three phases on the
basis of which we have structured our research.
Balland P-A, Suire R, Vicente J (2013) Structural and geographical patterns of knowledge networks in emerging
technological standards: Evidence from the European GNSS industry. Economics of Innovation and New
Technology 22: 4772
Bathelt H, Glückler J (2011) The relational economy: Geographies of knowing and learning. Oxford University Press,
Brenner T, Cantner U, Fornahl D, Fromhold-Eisebith M, Werker C (2011) Regional innovation systems, clusters, and
knowledge networking. Papers in Regional Science 90: 243249
Territorial governance and proximity dynamics 19
Papers in Regional Science, Volume •• Number •• •• 2017.
Bonnal P, Maluf RS (2009) Políticas de desenvolvimento territorial e multifuncionalidade da agricultura familiar no
Brasil. Política & Sociedade 8: 211250
Borgatti SP, Martin GE, Linton CF (2002) Ucinet for Windows: Software for social network analysis. Analytic
Technologies, Harvard, MA
Boschma R (2005) Proximity and innovation: A critical assessment. Regional Studies 39: 6174
Boschma R, Balland P-A, de Vaan M (2014) The formation of economic networks: A proximity approach. In: Torre A,
Wallet F (eds) Regional development and proximity relations. Edward Elgar, Cheltenham
Callois JM, Aubert F (2007) Towards indicators of social capital for regional development issues: The case of French
rural areas. Regional Studies 41: 809821
Camagni R (1991) Innovation networks. John Wiley & Sons, Hoboken, NJ
Camagni R (1995) The concept of innovative milieu and its relevance for public policies in European lagging regions.
Papers in Regional Science 74: 317340.
Carrington PJ, Scott J, Wasserman S (2005) Models and methods in social network analysis. Cambridge University
Press, Cambridge
Cazella AA, Zimmermann S, Leite SP (2013) A gestão do programa Territórios da Cidadania no Brasil: Análise das
ações e políticas implementadas. Perspectivas Rurales Nueva Época 11: 71107
Chiffoleau Y, Touzard J-M (2013) Understanding local agri-food systems through advice network analysis. Agriculture
and Human Values 31: 1932
Compagnone C, Hellec F (2015) Farmersprofessional dialogue networks and dynamics of change: The case of ICP and
no-tillage adoption in Burgundy (France). Rural Sociology 80: 248273
Colletis-Wahl K, Corpataux J, Crevoisier O, Kebir L, Pecqueur B, Peyrache-Gadeau V (2008) The territorial economy: A
general approach in order to understand and deal with globalization. In: Querejeta MJA, Landart CI, Wilson JR (eds)
Networks, governance and economic development: Bridging disciplinary frontiers. Edward Elgar, Cheltenham
Crespo J, Requier Desjardins D, Vicente J (2012) Why can collective action fail in local Agri-food systems? A social
network analysis of cheese producers in Aculco, Mexico. Food Policy 46: 165177
Dekker D, Krackhardt D, Snijders TA (2007) Sensitivity of MRQAP tests to collinearity and autocorrelation conditions.
Psychometrika 72: 563581
Delgado NG, Grisa C (2014) Políticas de desenvolvimento territorial e pobreza rural no Brasil: Análise das
institucionalidades e da governança. Estudos Sociedade e Agricultura 22: 132163
Delgado NG, Leite SP (2011) Políticas de desenvolvimento territorial no meio rural brasileiro: Novas institucionalidades
e protagonismo dos atores. Revista Dados 54: 431473
Gilly JP, Wallet F (2001) Forms of proximity, local governance and the dynamics of local economic spaces: The case of
industrial conversion processes. International Journal of Urban and Regional Research 25: 553570
Giuliani E, Bell M (2005) The micro-determinants of meso-level learning and innovation: Evidence from a Chilean wine
cluster. Research Policy 34: 4768
Granovetter M (1985) Economic action and social structure: The problem of embeddedness. American Journal of
Sociology 91: 481510
Graf H (2010) Gatekeepers in regional networks of innovators. Cambridge Journal of Economics 35: 173198
Grossetti M (2008) Proximities and embedding effects. European Planning Studies 16: 629642
Hoekman J, Frenken K (2013) Proximity and Stratication in European scientic research collaboration networks: A
policy perspective. In: Scherngell T (ed) The geography of networks and R&D collaborations. Springer
International Publishing, New York
Lascoumes P, Le Galès P (2007) Introduction: Understanding public policy through its instruments. From the nature of
instruments to the sociology of public policy instrumentation. Governance 20: 121
Laumann EO, Marsden PV, Prensky D (1989) The boundary specication problem in network analysis. In: Freeman LC,
White DC, Romney AK (eds) Research methods in social network analysis. Transaction Publishers, New
Brunswick, NJ
Favereau O, Lazega E (2002) Conventions and structures in economic organization: Markets, networks, and hierar-
chies. Edward Elgar Publishing, Cheltenham
Magsi H, Torre A (2014) Proximity analysis of inefcient practices and socio-spatial negligence: Evidence, evaluations
and recommendations drawn from the construction of Chotiari reservoir in Pakistan. Land Use Policy 36: 567576
Maillat D (1995) Territorial dynamic, innovative milieus and regional policy. Entrepreneurship & Regional
Development 7: 157165
Molina-Morales XF, Capó-Vicedo J, Martínez-Fernández TM, Expósito-Langa M (2013) Social capital in industrial dis-
tricts: Inuence of the strength of ties and density of the network on the sense of belonging to the district. Papers in
Regional Science 92: 773789
Piore MJ, Sabel CF (1984) The second industrial divide: Possibilities for prosperity. Basic books, New York
Piraux M, de Assis WS, da Cruz Rodrigues V, Silva NNM, Alves JW (2013) Um olhar sobre os Colegiados dos
Territórios da Cidadania no Pará. Novos Cadernos NAEA 16: 101124
Porter ME (1998) Clusters and the new economics of competition. Harvard Business Review 76: 7791
E. Polge, A. Torre20
Papers in Regional Science, Volume •• Number •• •• 2017.
Spielman DJ, Davis K, Negash M, Ayele G (2010) Rural innovation systems and networks: Findings from a study of
Ethiopian smallholders. Agriculture and Human Values 28: 195212
Scott AJ (1986) Industrial organization and location: Division of labor, the rm, and spatial process. Economic
Geography 62: 215231
Schneider S, Mattei L, Cazella A (2004) Histórico, caracterização e dinâmica recente do PRONAF. In: Schneider S,
Kunrath Silva M, Moruzzi Marques PE (eds) Políticas públicas e participação social no Brasil rural. UFRGS
editora, Porto Alegre
Storper M (1997) The regional world: Territorial development in a global economy. Guilford Press, New York
Ter Wal AL, Boschma R (2011) Co-evolution of rms, industries and networks in space. Regional Studies 45: 919933
Torre A (2008) On the role played by temporary geographical proximity in knowledge transmission. Regional Studies
42: 869889
Torre A, Rallet A (2005) Proximity and localization. Regional Studies 39: 4759
Torre A, Traversac JB (2011) Territorial governance. Local development, rural areas and agrofood systems. Springer
Verlag, New York
Weterings A, Boschma R (2009) Does spatial proximity to customers matter for innovative performance? Evidence from
the Dutch software sector. Research Policy 38: 746755
White HC (1992) Identity and control: A structural theory of social action. Princeton University Press, Princeton, NJ
Territorial governance and proximity dynamics 21
Papers in Regional Science, Volume •• Number •• •• 2017.
... Recently mobilized in the example of AD in Cavigny (Niang et al., 2021), the social network tool is here coupled with an analysis of proximities (geographical and organized) to provide a better understanding of the governance associated with the structuring of productive and social interactions. Our approach is in line with some of the work conducted on innovation dynamics in the dairy sector (Torre et al., 2019;Pachoud et al., 2019), the evaluation of farmers' collective agroecological practices (Houdart et al., 2011), local governance and rural development arrangements (Polge and Torre, 2017), and the importance of governance arrangements for the implementation of circular economy strategies in territories (Jambou, 2018). Eventually, we want to test the hypothesis that the mobilization of various types of proximities could help to build a strong local network between local producers, customers, authorities and association of local residents. ...
... In order to understand the modalities of the emergence of collective action and the dynamics of social and economic ties that are created and strengthened or unravelled over time (Lazega et al., 2015), we then proceeded to study the coordination structure. We analyzed the proximity relations between actors and their evolution (Torre et al., 2019;Polge and Torre, 2017). This made it possible to interpret the results obtained in terms of social networks and, in particular, to specify the nature of the interactions and provide a better understanding of the actors' structuring and governance mechanisms at work in the AD process. ...
Anaerobic digestion has recently gained interest in contributing to territorial strategy regarding the deployment of the circular economy and energy transition. Most projects bring together multiple actors from a wide variety of backgrounds. The article analyzes the evolution of synergies and cooperative behaviors between local stakeholders over the period 2010–2020 in an anaerobic digestion cluster in France. The study draws on social network analysis and proximity theory, which have recently been used for analyzing regional innovation systems, local clusters, territorial governance, and rural development. We reveal that local stakeholders develop dense relational networks that vary and evolve throughout the project. Different groups exist and behave in a semi-autonomous manner. All the actors are located in close geographical proximity. Still, their links in terms of organized proximities are related to various types of relations, resulting from cognitive resemblances or common origins. This explains the persistence and resilience of local relationships and how they maintain a collaborative dynamic over time.
... Recently mobilised in the example of AD in Cavigny (Niang et al., 2021), the social network tool was here coupled with an analysis of proximities (geographical and organized), providing a better understanding of the governance associated with the structuring of productive and social interactions. Our approach is in line with some of the work conducted on innovation dynamics in the dairy sector (Torre et al., 2019;Pachoud et al., 2019), the evaluation of collective agroecological practices of farmers (Houdart et al., 2011), local governance and rural development arrangements (Polge & Torre, 2017) and the importance of governance arrangements for the implementation of circular economy strategies in territories (Jambou, 2018). Eventually, we want to test the hypothesis that the mobilization of various types of proximities could help build a robust local network among local producers, customers, authorities and associations of local residents. ...
... In order to understand the modalities of the emergence of collective action and the dynamics of social and economic ties that are created and strengthened or unravelled over time (Lazega et al., 2015), we then proceeded to study the coordination structure. We analyzed the proximity relations between actors and their evolution (Torre et al., 2019;Polge & Torre, 2017). This made it possible to interpret the results obtained in terms of social networks and, in particular, to specify the nature of the interactions and provide a better understanding of the actors' structuring and governance mechanisms at work in the AD process. ...
Full-text available
Anaerobic digestion has recently gained interest in contributing to territorial strategy regarding the deployment of the circular economy and energy transition. Most projects bring together multiple actors from a wide variety of backgrounds. The article analyzes the evolution of synergies and cooperative behaviors between local stakeholders over the period 2010-2020 in an anaerobic digestion cluster in France. The study draws on social network analysis and proximity theory, which have recently been used for analyzing regional innovation systems, local clusters, territorial governance, and rural development. We reveal that local stakeholders develop dense relational networks that vary and evolve throughout the project. Different groups exist and behave in a semi-autonomous manner. All the actors are located in close geographical proximity. Still, their links in terms of organized proximities are related to various types of relations, resulting from cognitive resemblances or common origins. This explains the persistence and resilience of local relationships and how they maintain a collaborative dynamic over time.
... Four million people live on the banks of the basin, including important minorities such as indigenous people, riverbank dwellers and quilombolas (Guerreiro, 2017;Leal, 2016;Peixoto, 2009). focusing specifically on the Amazon (Polge & Torre, 2018). We therefore believe that our study may contribute to the advancement of knowledge on an important but less well-represented region. ...
... Moreover, studies dealing with territorial governance have developed to a remarkable extent in Europe, where they emerged (e.g., Balz & Zonneveld, 2015;Wassenhoven, 2008;Zanon, 2011). However, in our research, we found few studies outside this developed, urbanized and relatively homogenous region (e.g., Polge & Torre, 2018;Torres-Salcido & Sanz-Cañada, 2018). Our article portrays a case that occurred in a poorly developed, sparsely populated region in an emerging country. ...
Sustainable development has become a major topic as the impacts of human actions on the environment have increasingly dramatic consequences. This article describes and analyses a case of organizing for sustainable development that involved multiple social actors working in a participatory way. The research focuses on a project that aimed to define guidelines for carrying out large-scale engineering and construction ventures in the Amazon. It was conducted by a Brazilian research center and an international development agency. The field research included interviews with project leaders, extensive document analysis and nonparticipant observation. As a theoretical lens, we used the concept of territorial governance, from which we created propositions on how interventions should work. Based on these propositions and additional literature on social-ecological systems, an evaluation of the project is presented. We found the elements that were vital to the short-term success of the project, i.e., the skills of the change agents, the reputation of their organizations, and the social technology employed. We also found that less attention to public policies might weaken its long-term prospects. This work advances the knowledge on alternative modes of organizing for sustainability. Additionally, the paper emphasizes the concept of territorial governance, which offers opportunities for engaged scholars interested in tackling grand challenges.
... Cognitive proximity is commonly defined as the similarities in the ways actors perceive, interpret, understand, and evaluate the world (Knoben and Oerlemans, 2006). Some authors speak in this regard about the "logic of similarity" when actors share common knowledge, or cultural or religious values (Pachoud et al., 2020;Pachoud et al., 2019;Polge and Torre, 2018;Torre et al., 2019a). It has been shown that cognitive proximity can both enable and constrain learning. ...
CONTEXT Social networks play an important role in the diffusion of knowledge, and farmers draw on their personal networks to enhance their adaptive capacity to shocks. Different forms of proximity have been long recognized as important factors in knowledge and information exchanges. However, the specific roles and their interactions in agricultural knowledge and innovation systems (AKISs) are still far from clear. In this study, we investigate the underlying forces that drive tie formation within the knowledge-sharing networks of banana farmers in four different villages in Rwanda. OBJECTIVE Our study has three objectives: First, we discuss the importance of various types of proximities in AKIS research. Second, we empirically contribute to how different forms of proximity influence the way knowledge diffuses in formal and informal networks by studying a plant disease's management. Finally, we discuss our findings' relevance for targeted interventions to help rural communities transition to greater resilience. METHODS We review different proximity concepts and adapt them for use within an AKIS context. We then apply this framework to assess the proximity effects on the advice-seeking networks of banana farmers in four purposefully chosen villages in Rwanda. We used a structured questionnaire to collect social network information about the management of banana Xanthomonas wilt (BXW), from all banana growers (N = 491) in these four villages. We distinguished the informal advice networks among farmers from the official government extension system—the formal advice network. We employed exponential random-graph models to assess the determinants of the networks we observed, especially geographical, cognitive and social proximity indices. RESULTS AND CONCLUSIONS We found that geographical proximity significantly affects knowledge exchange within larger villages' informal advice networks; but not in smaller villages, where both cognitive and social proximities play substantial roles. We argue that farmers are socially closer in smaller communities where geographical distance does not matter, and that geographical distance only starts to matter after a certain community size threshold is reached. SIGNIFICANCE We provide solid empirical evidence to help plan targeted interventions toward greater resilience for rural communities. We argue that properly integrating informal social networks can result in more effective knowledge exchange within AKISs, enhancing their resilience.
... On the other hand, this author has conducted studies on territorial governance focused on two Latin American countries: Brazil and Chile. The research focused on Brazil was carried out within the framework of the "Territories of Citizenship" program, whose main challenge was that great distances hindered interaction between stakeholders, so it was sought to implement specific measures that promote and strengthen communication (Polge and Torre, 2017). Concerning Chile's case, the country's context and socio-political projects, territorialization mechanisms, and territorial intervention tools are examined (Gallardo, Lukas, Stamm, and Torre, 2019). ...
Full-text available
Territorial governance has been given different definitions across the academic spectrum; however, it is considered a unit of analysis, an essential subject of study for f regional economic development strategies. Not in vain has it become a field of interest for social scientists in the last few years. The purpose of this research is to analyze the body of literature developed around territorial governance, identifying its growth over time, the countries more academically interest in it, the most relevant authors, fundamental characteristics of territorial governance as described per cluster, recurring terms, and future research lines. To meet this study's objective, we applied a bibliometric analysis to the literature, for which 276 documents retrieved from the Web of Science were examined. The results suggest that the field of territorial governance has grown during the last six years, and Brazil, a Latin American country, is among the five regions with the highest production on the subject. It is observed that the topics of most significant interest have revolved around how territorial governance is related to the development of the territory and territorial policy. Future research should be aimed at integrating water resources management and land use planning in order to promote a sustainable territorial future, i.e., territorial sustainability.
CONTEXT The social network analysis of farmers who have adopted agroecological practices give the possibility to identify which actors are involved in the agroecological transition and are influencing the resilience at territorial level. OBJECTIVE To understand the dynamics of these interactions, we built and tested an analytical framework inspired by quantified narrations approach adapted to agricultural context. We combined social sequences analysis (identification of common phases within individual trajectories and typology of sequences) and relational chains analysis that is a specific approach within social networks analysis focusing on the mode of access to resources. METHOD We applied our analytical framework to study the modes of access to resources mobilized by farmers to adopt agroecological practices in the Limagne plain of the Puy-de-Dôme county in France . We conducted 31 face-to-face interviews with 22 farmers in organic agriculture and 9 farmers in conservation agriculture. RESULTS AND CONCLUSIONS The results show that our approach gives the possibility to identify a large range of actors beyond the commonly pre-identified actors and to analyse their specific roles depending on the phase of the transition. The farmers in conservation agriculture mobilize mainly interpersonal relationships prior to the adoption of practices and have little support at the time of the implementation of conservation agriculture practices while the organic farmers rely more on farmers' groups and on formalized arrangements with support organizations and downstream actors. SIGNIFICANCE The framework should be a useful way to identify in all agricultural systems the actors effectively implicated in the agroecological transition, their differentiated roles, and the support needed to improve the transition.
Full-text available
L’économie circulaire se présente comme un nouveau modèle économique permettant de faire face aux défis actuels du système économique linéaire. Aujourd’hui reprise dans plusieurs pays comme un levier d’évolution des pratiques et des modèles de développement économique, ses démarches sont de plus en plus expérimentées dans les territoires, dans un contexte de transition socioécologique et énergétique. L’objectif de la thèse est de déterminer dans quelle mesure l’économie circulaire peut constituer, par son caractère innovant, une opportunité pour les territoires de mettre en oeuvre des processus de développement territorial. Le cadre d’analyse mobilise différentes théories et méthodes quantitatives et qualitatives dans le but de mieux comprendre les implications de l’ancrage territorial et l’importance des dispositifs de gouvernance pour la mise en place des stratégies d'économie circulaire sur les territoires. Les résultats montrent une croissance locale de l’économie circulaire, en phase avec les enjeux de plus en plus importants en termes de politiques publiques en faveur de l’emploi et de compétitivité de l’économie. Ils indiquent que le défi actuel n’est pas seulement une question d’innovations technologiques autour de l’efficience des ressources et de la création de valeur, mais aussi d’hybridation des actions et de capacité à faire coopérer des parties prenantes variées pour la mise en oeuvre d’externalités territoriales positives. L’exemple de la méthanisation permet de mettre en évidence le rôle du contexte local dans la capacité à générer ces activités nouvelles qui, par leurs vertus économiques et sociales contribuent à créer de nouveaux liens et des processus durables, grâce à l’activation ou au renforcement des proximités et de leurs potentiels.
Full-text available
This thesis deals with the development of statistical methods for the analysis of collections of interaction networks through three original contributions. Interaction networks are a natural way to represent in graph form the exchanges or relationships existing between a set of nodes representing species or individuals. Considering collections of networks allows to study heterogeneous systems, composed of several kinds of interactions involving different types of nodes. When the different networks of the collection are linked by a hierarchical relationship, we speak of multilevel networks. The stochastic block model has proven its relevance to model the heterogeneity of the behavior of nodes in a single network. Extensions to collections of networks and to multilevel networks are proposed. They allow to obtain a clustering of the nodes of the networks according to their role in the ecosystem or social system, and to summarize the structure of the system at the mesoscopic scale through a small number of parameters. The inference of these models is complex and variational methods are adapted for this purpose. Model selection methods are also used to determine the dependence between levels for multilevel networks and the similarity between structures for collections of networks.A last part of this thesis proposes a new method to study the robustness of ecological interaction networks. Each network is modeled by a probabilistic model whose parameters represent the network structure. This allows to make the link between the structure of the ecosystem and its robustness, but also to compare the robustness of a collection of networks and to correct the robustness of a network whose sampling would be incomplete.The developed methods are implemented in R packages and applied on data from social sciences and ecology.
Full-text available
In this paper, we are interested in industrial and territorial ecology (ITE), whose aim is to optimize the management of material and energy flows between local economic players by drawing inspiration from the cyclical nature of natural ecosystems. The organizational elements, specifically the forms of coordination between actors, appear to be central in the setting out of these processes. This is why methodological devices promise to respond to the chronic difficulty of implementing local inter-firm relations conducive to cooperation. The work presented here, based on social network analysis, aims to determine their validity through three case studies. First, we examine the need to consider the spatial dimension of ITE approaches to understand the conditions for the emergence of inter-firm cooperation and sustainable development, and we present the methodological elements of our work. Then, we proceed to the case studies and identify inter-firm relations and study their evolution over time. We conclude with an assessment of the devices studied, the intermediary role of facilitators, and the difficulty of perpetuating these types of cooperative relations, which raises serious questions about the modalities of the implementation of sustainable territorial development processes.
Full-text available
In his seminal work, Mark Granovetter (1973) challenged sociologists to test sociometric hypotheses regarding collective action in communitarian settings. In this article, we tested the two main hypotheses which consider social cohesion in communitarian urban settings-these being firstly cohesion by weak ties and secondly cohesion by multiplex ties. We studied the elite leaders of two slum communities of Belo Horizonte (Brazil). Three social processes were examined as multiplex interactions: recognized status, exchange of useful information and collaboration. Our findings reveal, on the one hand, that multiplexity is associated with the frequency of ties and, on the other, that reciprocity and shared domains of performance fuel such strong multiplexity. If we assume that elite connections conform to a high order structure, our findings, in contrast to previously well-established hypotheses, reveal a segmented social order in which multiplexity does not mean the overlapping of social circles. On the contrary, multiplexed social exchanges are restricted to specialized domains.
Full-text available
No Brasil, as politicas territoriais e, em particular, os Territorios da Cidadania, representam inovacoes institucionais interessantes. A avaliacao e o melhoramento delas necessitam de uma analise dos mecanismos diferenciados da acao publica, e suas consequencias concretas sobre as formas de territorios resultantes. A partir do exemplo do estado do Para, a relacao entre a diversidade de dinâmicas territoriais e o funcionamento diferenciado dos colegiados dos Territorios da Cidadania e analisada, a partir do nivel de participacao, da eficiencia deles como espaco de governanca e da implementacao de projetos de investimento. Os problemas especificos e os aspectos positivos dos colegiados sao analisados e explicados. Perspectivas metodologicas e de desenvolvimentos sao sugeridas para melhorar o programa.
How are firms, networks of firms, and production systems organized and how does this organization vary from place to place? What are the new geographies emerging from the need to create, access, and share knowledge, and sustain competitiveness? In what ways are local clusters and global exchange relations intertwined and co-constituted? What are the impacts of global changes in technology, demand, and competition on the organization of production, and how do these effects vary between communities, regions, and nations? This book synthesizes theories from across the social sciences with empirical research and case studies in order to answer these questions and to demonstrate how people and firms organize economic action and interaction across local, national, and global flows of knowledge and innovation. It is structured in four clear parts. The first part looks at foundations of relational thinking. The next part is about relational clusters of knowledge. The third part looks at knowledge circulation across territories. The final part considers whether there is a relational economic policy. The book employs a relational framework, which recognizes values, interpretative frameworks, and decision-making practices as subject to the contextuality of the social institutions that characterize the relationships between the human agents.
The notion of proximity is increasing in popularity in economic and geographic literature, and is now commonly used by scholars in regional science and spatial economics. Few academic works, however, have explored the link between regional development and proximity relations. This comprehensive book redresses the balance with its assessment of the role of, and obstacles caused by, proximity relations in regional development processes.
Public policy instrumentation and its choice of tools and modes of operation are treated either as a kind of evidence (governing means making regulations, taxing, entering into contracts, communicating, etc.) or as if the questions it raises (the properties of instruments, justifications for choosing them, their applicability, etc.) are part of a rationality of methods without any autonomous meaning. This paper aims to explain the significance of a political sociology approach to public policy instruments in accounting for processes of public policy change: (1) public policy instrumentation is a major issue in public policy, since it reveals a (fairly explicit) theorization of the relationship between the governing and the governed: every instrument constitutes a condensed form of knowledge about social control and ways of exercising it; and (2) instruments at work are not neutral devices: they produce specific effects, independently of the objective pursued (the aims ascribed to them), which structure public policy according to their own logic.
Two dimensions of economic specificity are at the centre of this book's inquiry: knowledge as the basis of technology, and the human relations that are essential to many types of economic coordination. Both are simutaneously necessary to the operation of a normal, competitive adjustment process and yet they incessantly create and recreate imperfect competition and economic differentiation. Through the course of this book, the author proposes a compelling new theory of how regions have sustained their economic viability in the era of multinational corporations. Unlike traditional approaches, which analyse economic systems in terms of their mechanics (inputs, outputs, technology etc), this work views them as systems for coordinating human actions and relationships. Reconceptualising the role of learning, technology, and local institutions in development, the author illuminatesthe key role of regional ecoinomics as building blocks of the increasingly connected world. The book is divided into 5 parts. The first focuses on regional development and regions as relational assets. The chapters in the second section discusses the evolution of economic regions and territorial development. Part Three looks at innovation, technology, product development and territories. The fourth part is concerned with economic development globalisation, and territorial specificity, ad deals with the dynamics of globalisation, investment flows, international trade and the concept of 'technology districts'. In this section Chapter Nine discussesthe nature and role of the city in a globalising capitalist economy. The final section considers regional institutions and economic policy, with the conclusion summarising issues of technology, firm strategies and territorial order.
The paper opens with a description of the division of labor within the firm. The argument then passes on to the question of the vertical disintegration and integration of production and its crucial relations to (a) economies and diseconomies of scope and (b) the costs of intra- and inter-firm transactional activity. An attempt is made to synthesize these issues by providing a unified description of the organization of industry and the theory of the firm. The implications of this synthesis for location theory and spatial analysis generally are described. Two specific geographical problems are addressed, namely, (a) the origins and dynamics of growth centers and (b) restructuring and the multiestablishment firm. The paper closes with a few brief allusions to a prospective research agenda for economic geographers.
In the research reported here, we studied three cereal-growing areas in the French region of Burgundy, concentrating on the dynamics of change with regard to farmers' use of pesticides and new soil management practices on the one hand and on farmers' professional dialogue networks on the other. Our aim in this article is to show the link between the network types and the dynamics of change in their members' behavior. Three types of coalitional networks appeared. In the “mainly bonding” network, the roles of innovator and early adopter were the most highly shared among farmers. In the “mainly bridging” network, these roles were occupied by a single person holding a central position within the network. In the “bonding and bridging” structure, the roles of innovator and early adopter were held by different individuals. We further observed that the farmers developing integrated crop protection techniques occupied different positions from those implementing complete no-tillage, which is not promoted by technical advisory services. A rivalry thus appears to emerge between these two orientations in which the farmers are engaged within the network.