Intersectorality and social networks: the implementation of programs for homeless people in São Paulo

Abstract

Apart from the normative and prescriptive defense of intersectorality, the conditions related to the implementation of intersectoral programs are still little explored. This article aims to fill this gap, using an analytical-methodological strategy based on studies on implementation and the methodology of social network analysis (SNA). Based on primary data collected through in-depth interviews and analyzed using SNA, this article compares two programs for homeless people: the Oficina Boracea program, from its beginning in 2002 to 2016, and the program De Braços Abertos (DBA), from 2013 when it started until its end, in 2016. The results indicate that intersectorality is more than a well-designed management model. It is the product of everyday interactions and coordination strategies built with the participation of different levels of bureaucracy and state and non-state actors, gathered in different implementation arrangements.

Keywords:
intersectorality; policy integration; implementation; coordination; social network analysis; homeless people

Resumo

Para além da defesa normativa e prescritiva da intersetorialidade, sabe-se pouco sobre os condicionantes de sua implementação. O objetivo deste artigo é suprir essa lacuna, usando uma estratégia analítico-metodológica baseada nos estudos acerca de implementação e metodologia da análise de redes sociais (ARS). Com base em dados primários coletados em entrevistas aprofundadas e analisados por meio da ARS, o texto compara duas ações intersetoriais destinadas à população em situação de rua: o projeto Oficina Boracea, desde seu início, em 2002, até 2016, e o programa De Braços Abertos (DBA), a partir de sua formulação, em 2013, até seu término, em 2016. Os resultados indicam que, mais do que um modelo de gestão bem-formulado, a intersetorialidade é um produto de interações cotidianas e estratégias de coordenação construídas entre diferentes escalões da burocracia e entre atores estatais e não estatais, em distintos arranjos de implementação.

Palavras-chave:
intersetorialidade; implementação; coordenação; análise de redes sociais; população em situação de rua

Resumen

Con excepción de la defensa normativa y prescriptiva de la intersectorialidad, poco se sabe sobre las condiciones para su implementación. El objetivo de este artículo es llenar ese vacío, utilizando una estrategia analítico-metodológica basada en estudios de implementación y en la metodología de análisis de redes sociales (ARS). A partir de datos primarios recopilados a través de entrevistas en profundidad y analizados mediante ARS, este artículo compara dos proyectos intersectoriales para personas sin hogar: el proyecto Oficina Boracea, desde su inicio en 2002 hasta 2016, y el programa De Braços Abertos (DBA), desde su formulación en 2013 hasta su finalización en 2016. Los resultados indican que, más de un modelo de gestión bien formulado, la intersectorialidad es el producto de interacciones cotidianas y estrategias de coordinación construidas entre diferentes niveles de burocracia y entre actores estatales y no estatales, en diferentes arreglos de implementación.

Palabras clave:
intersectorialidad; implementación; coordinación; análisis de redes sociales; población sin hogar

1. INTRODUCTION

This article analyzes the factors that explain the conformation and implementation of public programs’ intersectoral arrangements. Intersectorality is often defended as the best way to respond to complex social problems (Inojosa, 2001; Junqueira, 2000). However, national and international analyses of the factors influencing intersectorality are still incipient, usually adopting normative and prescriptive approaches (Candel & Biesbroek, 2016; Cunill-Grau, 2014; Tosun & Lang, 2017; Shankardass, Solar, Murphy, Greaves, & O’Campo, 2012).

This study contributes to filling this gap by analyzing the implementation of two intersectoral programs in Brazil. The research considers both the connections established through formal instances and the role of personal and sometimes informal relationship networks in the development of coordination strategies between government sectors and state and non-state actors. We used social network analysis (SNA), which stands out as a promising methodology for an analytical and procedural approach to intersectorality - as advocated by Candel and Biesbroek (2016) - and for observing the implementation of intersectorality as a chain of interactions (Lotta, 2018; Winter, 2006).’

This research follows Marques (2000), going beyond the metaphorical use of networks and employing them as an analytical and methodological strategy to identify formal and informal connection patterns developed in intersectoral programs. The systematic mapping of de facto relationships leads to moving beyond the usual analysis of formal interactions regulated by coordination committees - as observed by Cunill-Grau (2014). In addition, such an approach allows identifying coordination strategies and arrangements unforeseen in documents recording these programs.

Thus, even though this research does not corroborate Junqueira’s (2000) normative conception, we share the author’s point of view regarding the relevance of SNA in intersectoral relation studies that emphasize the role of personal relationship networks in the construction of coordination strategies (Junqueira, 2000; O’Toole, 2010). The use of SNA as a methodological strategy allows identifying possible asymmetric relationships in interactions among government sectors that are formally at equivalent levels. Such asymmetries can be an obstacle to the construction and institutionalization of intersectoral arrangements.

As noted in studies addressing the challenges of building coordination, it is not possible to assume that the organizational hierarchy grants sufficient authority to ensure coordination, just as it is not always possible to infer that sectors of the same hierarchical level develop symmetrical relationships (Gontijo, 2012; O’Toole, 2010).

In Brazil, some studies address the challenges of implementing social policies with intersectoral arrangements at the national (Bichir, Oliveira & Canato, 2016; Lotta, Galvão & Favareto, 2016; Silva, 2013) and municipal levels (Bronzo, 2007; Veiga & Carneiro, 2005, 2014). However, the national studies on intersectionality are marked by a polysemy and low theoretical density (Akerman, Moyses, Rezende & Rocha, 2014), either as a research category or category for policy evaluation. From an empirical point of view, these authors argue that intersectoral strategies are “experiments” of certain administrations rather than a government praxis, which is a situation that reinforces the importance of this article.

Internationally, several authors recognize that public policies are increasingly organized in complex, cross-sectional, or intersectoral arrangements (Candel & Biesbroek, 2016; Cunill-Grau, 2014; Hupe, 2014; Sabatier, 2007; Tosun & Lang, 2017). However, typological approaches are still dominant in the specific discussion on intersectoral programs and policies. They seek to classify “ideal types” of interaction between government agencies or arrangements involving partnerships with civil society organizations (CSO) - as observed in work by Mendonça, Medeiros, and Araújo (2019). Also, Shankardass et al. (2012) carried out a systematic review of the literature on intersectorality in the health area, noting a gap in the international debate regarding how to build complex multi-sectoral arrangements going beyond the prescription of ideal types of interorganizational relationships.

This article focuses on the importance of empirical analysis on building cooperation and coordination strategies among different actors and organizations, based on authors such as O’Toole (2010) and Gontijo (2012). The study considers “implementation” as a set of formal and informal decisions and interactions (Lotta, 2018; Winter, 2006) and intends to contribute to analytical approaches to intersectorality. Authors such as Candel and Biesbroek (2016) are adopted as references. They highlight the integration of policies as a process with contradictions, different strategies, and rhythms of coordination among sectors and possibilities for reversal and emphasize that intersectoral arrangements often derive from incremental constructions - exactly as found in the empirical cases presented here. Finally, even without explicitly mentioning it, the basis of Candel and Biesbroek’s (2016) arguments is observed in work by Lindblom (1979), author of the concept of “muddling through.”

In this analytical perspective, generalizing categories such as state policy versus government policy are not very useful. It is essential to understand how and why certain programs created in specific political situations are continued or not. In line with classic (Lindblom, 1979; Lowi, 1972) and contemporary (Marques, 2013; Menicucci, 2018) discussions on the relationship between politics (political dynamics) and policies (public policy choices), it appears that political disputes around the construction of intersectoral arrangements are always present. They also include the point of view of paradigms and perspectives of transformation (Hall, 1993), not only as an exogenous dimension linked to a simplistic view of party government theory, or as something that interferes with certain ideal models of integration.

As for the empirical part of this research, we selected two intersectoral programs designed to serve the homeless population in the Brazilian city of São Paulo: the Oficina Boracea program, which started in 2003 and is still in operation; and the program De Braços Abertos (DBA), in operation from 2014 to 2016. The cases were selected because of the diversity of implementation strategies. They allow a comprehensive exploration of the many ways of implementing intersectorality, particularly regarding the dimension of relationships established.

The DBA program sought to promote the psychosocial rehabilitation of socially vulnerable people, abusers of psychoactive substances, whereas the Oficina Boracea program aimed to improve the standards of care for the homeless population. The analysis was divided into two moments t₀ - 2002-2004 and t₁ - 2007-2016. The two programs were established under the administration of the Workers Party (PT) - Oficina Boracea under Mayor Marta Suplicy’s administration (2001-2004) and DBA under Mayor Fernando Haddad’s administration (2013-2016). Although started by leaders from the same political party, the programs occupied different centrality in the municipal agenda, which affected the outcome regarding their (dis)continuity.

This article is divided into six sections, including this introduction. The second section discusses the analytical lenses that guided the work, emphasizing the literature on implementation that addresses relational dynamics. The third section presents the methodology adopted, especially ways of using SNA, while the fourth section brings the main characteristics of the two programs and the analysis of their networks. Finally, the fifth section discusses the main results in light of the literature, followed by the final considerations that portray the study’s main analytical contributions.

2. IMPLEMENTATION, INTERSECTORALITY, AND SOCIAL NETWORKS

This section presents the main analytical lenses that guided the research. It does not aim to offer an exhaustive review of the literature on implementation - found, for example, in Hupe (2014), Lotta (2019), and Saetren (2014) - on intersectorality (Akerman et al., 2014; Candel & Biesbroek, 2016; Carmo & Guizardi, 2017; Cunill-Grau, 2014; Shankardass et al., 2012), or on the coordination role of networks of relationships (Gontijo, 2012; Marques, 2000; O’Toole, 2010).

Intersectorality is sometimes presented as a panacea to address complex problems. In contrast, this article argues that it is necessary to overcome prescriptive approaches, such as those present in Inojosa (2001) and Junqueira (2000), and analyze how institutional and relational implementation arrangements were developed, as stated by Pires and Gomide (2018). According to these authors, an implementation arrangement is defined based on rules, interactions, and roles played by social actors and the bureaucracy that, together, produce different government actions:

The arrangement is the “place” in which governmental bureaucracies’ decisions and actions connect with those [...] of political and social actors, leading to policy dilemmas and challenges or lessons learned and innovation (Pires & Gomide, 2018, p. 29, our translation).

According to Pires (2016), intersectorality is something to be produced. It involves collective action problems and different forms of interaction among government sectors and CSOs. This perspective is that of part of international literature that highlights processes of building intersectoral arrangements with different levels of integration and without any guarantee of “success” or “good performance” (Candel & Biesbroek, 2016), considering the inherent complexity of implementation of policies in inter-organizational contexts (O’Toole, 2010). Even internationally, approaches based on typologies abound, such as Cunill-Grau (2014) and Tosun and Lang (2017). There is also a lack of studies that systematically analyze how complex multi-sector arrangements are produced (Shankardass et al., 2012).

This analytical approach of mapping the implementation arrangements put into practice is in line with the classic teachings highlighted in thoughts defended recently by many authors (Hupe, 2014; Lotta, 2019; Saetren, 2014): implementation as a complex chain of decisions, involving a multiplicity of state and non-state actors in policy governance networks. From an empirical and analytical point of view, it is worth stressing the contingency of the analyzed processes. Pires and Gomide (2018, p. 30, our translation) state it is about seriously addressing the “indeterminate nature of implementation processes,” in contrast to “prescriptive-formal approaches.”

Although the literature has emphasized the relative independence of the flows of problems and solutions (Kingdon, 1984), the intersectoral approach often arises from the perception of complex problems that require integral analysis (Costa & Bronzo, 2012; Inojosa, 2001). Particularly in Latin America, this interpretation emerges in the context of the new public management, based on the perception of inadequate and hierarchical sectoral structures of the state to face new themes and social demands (Bronzo, 2007; Cunill-Grau, 2014).

In Brazil, intersectorality comes into the debate about the state reform seeking efficiency, efficacy, and effectiveness beyond the direct state action (Akerman et al., 2014). However, even though it has been a much-discussed topic since the 1990s through the normative defense of integrating knowledge and sectors (Inojosa, 2001; Junqueira, 2000), little is known about the challenges for its implementation (Carmo & Guizardi, 2017).

Lessons learned from the international (Cunill-Grau, 2005, 2014; Shankardass et al., 2012) and national (Akerman et al., 2014) literature demonstrate that it is more fruitful - given the variability of intersectoral arrangements and coordination strategies - to empirically consider how these arrangements have been constructed within the scope of public management, instead of defining ideal models of intersectorality. Along with Akerman et al. (2014), Candel and Bisbroek (2016), and Shankardass et al. (2012), this research is critically positioned to totalizing or holistic biases of intersectorality, which disregard the relevance of sectoral approaches or partial integrations between sectors and actors.

Thus, the first aspect of analysis that guided the empirical research was how public problems are defined among actors integrating intersectoral arrangements, involving learning processes (Hall, 1993) and persuasion, including which implementation instruments would be more suitable (Sabatier, 2007). Cognitive dimensions, values, ​​and perceptions about the problem to be faced can vary greatly among sectors, affecting both the construction of a theme as a problem and the definition of methods to face it. These dimensions make the formulation and implementation of intersectoral actions more complex (Bronzo, 2007).

Understanding which actors stand out in defining the problem and building the intersectoral arrangement is one of the analytical aspects present in the recent debate on intersectorality (Akerman et al., 2014; Shankardass et al., 2012). Junqueira (2000) shows that this integration of different knowledge to solve a common problem occurs through networks of relationships. Depending on how the knowledge and relationships are put together, there may be lasting consequences: “The network can result in an intersectoral or even cross-sectoral knowledge, which transcends intersectoral relationships in the construction of new knowledge, new paradigms” (Junqueira, 2000, p. 40, our translation).

The second aspect refers to modes of coordination of intersectoral programs. Cunill-Grau (2014) argues that intersectorality implies a higher degree of integration among sectors than coordination. However, intersectoral arrangements may or may not be coordinated or occur through different strategies. Candel and Biesbroek (2016), criticizing the “stage” perspective of intersectorality typologies, mention that the full integration of government sectors is not always possible or desirable, and softer interaction strategies - with or without formal coordination - should be considered.

The construction of new coordination structures and mechanisms is a challenge, as demonstrated by the systematic review of articles on intersectorality in health carried out by Shankardass et al. (2012). Thus, formal coordination instances, such as management and inter-secretariat committees, are not always central to sectorial relationships. These instances can politically legitimize agendas or be only symbolic, i.e., without interfering in the daily policy implementation. They are certainly not “sufficient to guarantee the effectiveness of the strategy” (Veiga & Bronzo, 2014, p. 612, our translation). The coordination of the actors involved can occur in a less institutionalized way, through relationships woven in the daily implementation, as seen in the literature (Candel & Biesbroek, 2016; Junqueira, 2000) and in specific studies on implementation and coordination (Lotta, 2019; O’Toole, 2010; Winter, 2006).

A third analytical aspect refers to decision-making processes involving decision-makers in the high-level bureaucracy to actors operating at the lowest level of bureaucracy, interacting with the non-state street-level bureaucracy represented by civil society organizations (CSOs) (Brodkin, 2011). The rationale is that political decisions made by central authorities are relevant to initiating intersectoral integration processes (Cunill-Grau, 2014). However, intersectorality can also be built through daily management, when actors from different areas and hierarchical levels feel the need to resort to other sectors to seek solutions to problems.

Thus, it is possible to have non-institutionalized arrangements based on the network of relationships established in the policies day-to-day (Pires & Gomide, 2018). Arrangements with less formal integration or formed at the lower levels of bureaucracy should not be underestimated. They may be the most appropriate and viable for solving certain problems (Candel & Biesbroek, 2016).

The fourth aspect is the implementation arrangements, which involve institutional and relational dynamics among actors linked to different state bureaucracies and actors from civil society (Pires & Gomide, 2018). For Lotta (2014, 2018), it is necessary to broaden the understanding of the factors that influence the construction of practices and the interaction of implementing agents executing processes. The author recognizes the importance of two central dimensions: relational factors, which concern the networks and the processes of interaction of bureaucrats with actors from different hierarchical levels, users, and other professionals; and institutional factors, i.e., the rules and institutions that determine the construction of practices (Lotta, 2014).

Three hierarchical levels for analyzing interconnection patterns among bureaucrats stand out. First, the high-level bureaucracy is central in formulating cross-sector policies and guidelines and building relationships with peers in other sectors, which can be essential to the effectiveness of cross-sector agendas. Second, the mid-level bureaucrats (Lotta, Pires & Oliveira, 2014) connect formulators and implementers, exercising essential technical-political roles. Finally, at the third level, street-level bureaucrats are responsible for building networks with peers or actors from other sectors to solve everyday problems. In the case of the programs studied, the governance arrangements also involve CSOs associated with the services, which were included in the mapping of the networks.

The networks studied in this article are close to what Massardier (2007) calls “project networks,” mobilized to achieve specific and existing objectives only during project execution (Gaudin, 1995, as cited in Massardier, 2007). Rhodes and Marsh (1992, as cited in Massardier, 2007) call this an issue network, which is limited to exposed issues and problems. As for types of networks, the so-called epistemic communities must be considered (Haas, 1992). They bring together specialists from a particular field who share ideas and beliefs based on knowledge. As discussed below, specialists played an important role in building the implementation arrangement for one of the analyzed programs.

3. METHODOLOGY

In addition to reviewing the relevant literature and analyzing documents, legislation, and regulations on the two selected programs, this study article collected relational data through in-depth interviews with managers and implementers. The analytical assumptions presented in the previous section and social network analysis (SNA) were used. This methodological approach is based on the importance of connection patterns between units of interest, helping to identify relational structures that organize different social phenomena and allow the systematic mapping of informal networks and connections among different actors (Wasserman & Faust, 1994).

SNA is the most suitable method to measure relationship and interaction patterns, in addition to the metaphorical use of the idea of ​​connection, as recognized by important scholars in the field of implementation studies (Winter, 2006). The article aimed to map the implementation arrangements of intersectoral programs based on high expectations of interaction. Thus, the method allowed the identification of connections beyond the formal manifestations of institutional support in the analyzed programs’ official design.

Forty semi-structured interviews were carried out, collecting data from the two programs. The first interviewees were chosen based on the identification, in official documents, of actors involved in the conception and implementation of the programs, and, at each interview, new relevant actors were identified. The first part of the interview focused on deepening knowledge about the cases and encouraging the interviewee to bring other relevant actors, with questions that worked as a name generator: “Who are the most important people/organizations involved in decision-making processes and in implementing this program?”

In the second part of the interview, a list of names composed of previously identified and new actors was cited so that the interviewee could recommend up to three names related to each actor on the list, with the option of not mentioning anyone. Thus, each actor was responsible for indicating relationships of other actors, and not just their own. For inclusion in the analysis, names had to be cited in at least three interviews. Therefore, the network boundary was delimited based on the snowball method (Hanneman, 2001). Interviews ceased after saturation when no new names were mentioned.

The objective was to map the entire network of each program and not self-centered networks on relevant actors. In entire networks, the possibility of mutual recognition of all participants was assumed, and the veracity of the ties mentioned was guaranteed by the triangulation of information in the various interview rounds (Hanneman, 2001). Analytically, the option was not to restrict the mapping of networks to formally involved actors and institutions, considering that governance arrangements could be more diverse and broader than initially foreseen, as Marques (2000) observed in a study on urban policy networks.

The initial hypothesis was the relevance of different levels of bureaucracy in the phase of implementation of programs. Therefore, it would not be useful to restrict the analysis to the highest level. In addition, the mapped state actors were classified into three bureaucratic levels. The actors’ attributes were considered in mapping each program’s networks, along with information on ties to specific government or civil society bodies. It was then possible to measure the centrality and cohesion derived from positions in the network combined with information from the formal occupation of positions (“positional” power), in dialogue with the so-called “relational power” derived from the daily and informally woven web of relationships (Marques, 2000).

Gephi software was used to statistically and graphically analyze the collected data- through sociograms - supporting the analysis and conclusions of this research. As discussed in the results, measures of centrality (Wasserman & Faust, 1994) were essential to test which actors were coordinators between government sectors.

4. THE PROGRAMS AND THEIR SOCIAL NETWORKS

This section presents the cases and analyzes the networks of the two programs, which sought to map both the relationships among state actors, in the direct administration, and between them and external actors (CSOs, specialists, among others). All sociograms have the actors, represented as dots, and their ties, represented by dashes. The thickness of the lines indicates the frequency of citation of a given relationship - the higher, the more recurrent the mention of the relationship. However, this dimension is not covered in this article. Some attributes of the actors that make up the network are considered, emphasizing institutional affiliation and the levels of bureaucracy in the case of state actors.

4.1. Oficina Boracea program

The Oficina Boracea program began in 2003 to offer comprehensive care to the homeless population. The program was conceived by the Municipal Secretariat for Social Welfare (SAS) of São Paulo, created by the manager responsible for the secretariat, who had relevant experience in the area. She was a professor of the social service undergraduate program at the Pontifical Catholic University of São Paulo (PUC-SP). She had served in the high-level bureaucracy of the city government as Secretary of Regional Administrations (1989-1990), Secretary of Social Welfare (2002-2004), and City Councilor ​​for three consecutive terms (1993-2004).

Prior to the consolidation of the Brazilian National Social Welfare System (Suas), Oficina Boracea was recognized as a pioneering initiative in promoting a new model of care. One of its differentials was welcoming recyclable material collectors with flexible working hours and space for carts and animals. The program had a recyclable material collection center, a restaurant-school, an industrial-scale laundry room to serve other hostels, and a bank agency through which collectors could receive payments. It also promoted leisure activities, digital inclusion, and professional training.

The program was formulated in 2002, with the creation of an inter-secretariat workgroup involving the secretariats of social welfare, health, public works, supply, culture, labor, sports, environment, and regional administration (called sub-prefeituras). Civil society organizations and other community representatives were invited to contribute to the program’s design. The group met during the formulation period, and its main role was to offer legitimacy to the proposal for comprehensive care, inside and outside the government. The secretary managing SAS played a crucial role in ensuring that the program was central to the municipal agenda, enabling networking among municipal secretariats, CSOs, and other actors.

Agreements were made with seven CSOs that already served the homeless population. Each of them participated in managing the facilities where the program was developed and was responsible for providing the services and its team of professionals.

The program’s coordination team involved a manager and exclusively designated advisors. The coordination of the service was sectoral, encompassing only the SAS and the CSOs that entered into agreements with the secretariat. According to interviewees, decision-making processes at that time were more centralized at the high-level bureaucracy. However, CSOs had decision-making autonomy over the services’ daily operations.

The connection with other secretariats resulted in different degrees of involvement. Secretariats of areas such as culture, labor, sports, services and works, supply, and social communication had specific roles in the interaction with Oficinas Boracea program. The most daily and effective connections were with the Secretariat of Health (SMS), which can be explained by the tie between specialists who shared similar values ​​regarding the approach to the homeless population, highlighting the importance of this aspect in certain epistemic communities (Haas, 1992).

The SAS maintained a partnership with a mid-level sanitary doctor - regional health coordinator - who had academic and clinical experience serving the homeless population, enabling health services to this public. This networking between SMS and SAS sought to guarantee primary and mental health care through family health agents.

Oficina Boracea operated under the original guidelines from mid-2003 to the end of 2004 (t₀). With the change of municipal government in 2005, Boracea became a SAS service exclusively. The program’s cost explains the discontinuity of ties with other secretariats and the offer of other services. In 2007, a new CSO took over the program’s administration and identified a more significant demand for health services to serve the public, made up mainly of the elderly and convalescents (t₁). The CSO conducted a negotiation process with SAS and SMS, obtaining the right to use plots of land to install an outpatient medical unit (AMA) and, later, a basic health unit (BHU). This new configuration allowed for greater networking among professionals from SAS and SMS, especially street-level bureaucrats.

4.2. Oficina Boracea’s social networks

The comparative analysis between the two moments of the Oficina Boracea program (t₀and t₁) allows us to identify the changes mentioned above concerning the reduction of ties between sectors. The networks compared represent different actors; it is not just a modification of the pattern of existing relationships but a complete change in the network’s structure.

The initial network of Oficina Boracea (Figure 1, Network t₀) presents actors from different institutions and government sectors: SAS; Municipal Urbanization Company (Emurb); Municipal Secretariat of Sports, Leisure, and Recreation (Seme); Municipal Secretariat for Development, Labor, and Solidarity (SDTS); Municipal Secretariat for Communication and Social Information (SMCIS); Municipal Secretariat of Supply (Semab); Municipal Housing Secretariat (Sehab); Municipal Secretariat of Health (SMS) and Secretariat of Finance (SF). The network also brought actors from the Inter-American Development Bank (IDB), from the university PUC-SP, and business people and local community (identified, respectively, as “empresários” and “comunidade local” in Figure 1).

However, the relationships between the affiliated CSOs and SAS actors stand out, which account for almost half of the relationships. There were five different CSOs (identified as “OSCs” in the sociogram in Figures 1 and 2) directly contracted with SAS in the network. They were not distinguished because, at that time, they played similar roles in the program implementation and similar patterns of relationship with SAS.

Figure 1
Oficina Boracea’s Sociogram (t0, 2002-2005)

In t₀ Oficina Boracea was centered on social assistance when CSOs and SAS were working closely. Although there were 14 other actors/institutions, these actors appeared as more peripheral in the network. Part of the network nodes located on the left of the figure (SDTS, Emurb, BID, and SF) only connected with two SAS actors. According to interviewees, these institutions were relevant to define and formulate the program, but did not participate in its implementation.

Among the actors mentioned in the network at t₀, 15% were at the high-level bureaucracy, and 46% were at the mid-level bureaucracy. Only 5% of actors were considered street-level. A possible explanation for this configuration is the study’s data collection method. As it is a matter of rebuilding a network from almost 15 years ago, it is reasonable to assume that the interviewees remembered more actors in strategic and decision-making positions. The other actors (34%) were classified as “others” as they had some connection with the program but were not central to its implementation - among them were actors from the IDB, the community, and local entrepreneurs. The actor with the highest number of relationships was the person in the position of Secretary of SAS.

When contrasting this initial moment with 2016 (t₁, Figure 2), it is clear that the network has become less heterogeneous. The reduction of institutions can be explained by the change in the municipal government and the consequent change in the program’s position on the city’s agenda. Although the participation of some agencies was peripheral in t₀, there was a greater diversity of institutions.

Figure 2
Oficina Boracea’s sociogram (t1, 2007-2016)

The Oficina Boracea’s network in t₁ was made up of a single CSO in partnership with the SAS (identified as SMADS in the sociogram in Figure 2), which had almost half of the network’s ties (45.95%), by the SAS itself and by the CSOs that entered agreements with the SMS. The dimension of health at Boracea t₁ stood out, presenting a significant diversity of nodes.

Regarding the role of the different levels of bureaucracy, the Oficina Boracea program’s network in t₁ had only 5% of actors in high-level bureaucracy, all of them from SAS. Most were mid-level bureaucrats (51%), supervisors and coordinators from SAS and SMS, and the CSO’s managers. Implementing agents represented 32% of the network and were employees of OSCs and SMS. The CSO’s employees who did not work directly with the project were called “others” (11%), including the organization’s president and its administrative personnel.

We can also compare the general characteristics of the program’s network at t₀ and t₁ (Table 1), normalizing the data to take into account the different sizes. The comparison shows that the number of actors (nodes) in the network remained similar, while there was an increase in density, calculated from the proportion of existing ties and the total possible ties between the actors in a given network. This increase indicates a greater connection between the actors at time t₁, also evidenced by the increase in the average degree, calculated by the average number of direct ties of each actor. The difference for the weighted average degree is that it considers the repetition of dyads - the intensity of the relationship between two actors. The diameter of the networks, represented by the longest path between two actors, remained the same.

Table 1
General data of the Oficina Boracea’s network t0 and t1

Data	Boracea t0	Boracea t1
Nodes	42	37
Total ties	102	117
Average degree	4.857	6.324
Weighted average degree	5.667	9.514
Diameter	5	5
Density	0.118	0.176

• Source: Elaborated by the authors based on data collected from interviews and the Gephi.

4.3. De Braços Abertos program

The De Braços Abertos (DBA) (Open Arms) program was established by Decree 55067 in April 2014. The program aimed to promote the psychosocial rehabilitation of socially vulnerable people, abusers of psychoactive substances, adopting the harm reduction paradigm. The program initially focused on the territory known as Cracolândia, in the region called Luz, in São Paulo’s downtown, involving secretariats of five different areas: social welfare, human rights, health, urban security, and labor. The DBA program was also operated with the support of civil society organizations (CSOs) that worked through operational agreements with these secretariats. The emergence of the DBA on the agenda was stimulated by connecting the program with an initiative of the Brazilian Federal government to combat crack (Crack, é Possível Vencer, or “crack, it is possible to quit”).

In 2013, due to the connection with the federal program, the city created the municipal executive group (GEM), coordinated by the SAS and composed of 13 other municipal secretariats, two secretariats of the state government, and civil society. The GEM aimed to plan and implement actions related to the Intersectoral Policy Plan on Crack, Alcohol, and Other Drugs.

The GEM worked as a space to define and consolidate concepts that would guide the program. It was a locus open to the participation of those who already worked in the territory, who could help elaborate a diagnostic of the situation and the services that should be offered. The expectation of involvement of a large number of sectors did not materialize. Each secretariat’s role was defined incrementally, adapting existing regulations to ensure attention to future beneficiaries. In 2014, with the beginning of the program’s implementation, the committee stopped meeting.

The operationalization of the DBA was constantly agreed upon with different levels and sectors involved, covering the supply of food, accommodation, and work, in addition to health care and inclusion in the social assistance network. The participants joined the program after visiting a provisory facility set up in the territory, which offered services related to hygiene and a place for socializing.

According to the interviewees, the conformation of the coordination and implementation arrangement was gradual and changed over time. As it was a project with great centrality in the governmental agenda, the DBA mobilized, from the beginning, the city’s high-level bureaucracy, creating a management committee in which the person responsible for the secretariats and their assistants participated. During the program’s implementation, this managerial group reduced the frequency of meetings, distributing tasks to a coordinator due to the strengthening of the role of mid-level bureaucrats.

After a year of implementation, Ordinance 1752, of November 19, 2015, defined that the DBA would have a coordinator and representatives appointed by the secretariats, responding to the demands of the actors involved and making official the participation of mid-level bureaucrats who already had relationships with other agencies of the local government. The main role of the coordinator was to establish relationships among all levels and sectors, centralizing demands for institutional representation and reduce communication noise.

Local coordination and meetings held periodically contributed to the alignment between guidelines from high-level bureaucracy and implementing agents - information about the program’s day-to-day reached the decision-making bodies. At the DBA, the territory had a centrality shared by all hierarchical levels, and actors at the mid-level bureaucracy valued the opportunity to experience street-level activities.

There were disputes between the city hall and the state government at the high-level bureaucracy since both instances of government had programs for the same population and territory but operating under very different guidelines. However, the field research revealed coordination between street-level bureaucrats operating the different programs, who, to some extent, shared common goals.

4.4. De Braços Abertos’ social networks

Figure 3 represents the DBA program’s network. Not all actors and institutions present in the network formally participated in the program arrangement, but they were recognized as relevant by others, so they were included in the sociogram. Compared with Oficina Boracea program’s networks, the DBA’s had a greater number of participating agencies and institutions, making it relatively more complex and diversified.

Interviewees mentioned nineteen institutions someway related to the project. Among municipal departments and agencies, in descending order of ties are SMS, SAS (identified as SMADS in the sociogram in Figure 3), Municipal Secretariat for Human Rights and Citizenship (SMDHC), Metropolitan Civil Guard (GCM), Municipal Secretariat for Labor and Entrepreneurship (SDTE), Municipal Secretariat for Urban Security (SMSU), Municipal Secretariat of Government (SGM). Most of the secretariats operated in partnership with CSOs, which also appeared in the sociogram: Instituto de Atenção Básica e Avançada à Saúde (Iabas) (Institute of Basic and Advanced Health Care) and Associação Saúde da Família (ASF) (Family Health Association) - CSOs working in partnership with the SMS; Sociedade Amiga e Esportiva do Jardim Copacabana (Saec) (Sports Association of Jardim Copacabana) - organization partner of the SAS; Adesaf - organization partner of SDTE; and Projeto Oficinas and Casa Rodante - CSOs partners of the SMDHC. Actors identified as specialists (Espec.) - related to academia and directly and indirectly participating in the project through consultancy and research - and the CSO É de Lei - an organization that works with harm reduction in Cracolândia - represented a large percentage of ties in the network, totaling almost 20%.

Other actors observed in the sociogram were those who were part of the program Recomeço - a program of the state government working with the same population. Although the DBA and Recomeço had different conceptions and forms of working with users, the interviewees mentioned cooperation relationships among agents implementing the programs. Two actors belonging to the federal government were mentioned, one from the Ministry of Justice (Min. Justiça) and the other from the Ministry of Health (Min. Saúde). Two users were mentioned by interviewees, tied to an actor from the organization É de Lei.

Figure 3
DBA program’s sociogram

Table 2 identifies the low density of the network, lower than the previous ones. This can be explained by the existence of groups of peripheral actors, poorly connected, and the tendency that the larger the network, the lower the density.

Table 2
General data of the DBA program’s network

Nodes	82
Ties	80
Average degree	4.39
Weighted average degree	6.707
Diameter	6
Density	0.054

• Source: Elaborated by the authors based on data collected from interviews and the Gephi software.

For the DBA program, bringing the secretariats and associated CSOs together helped to analyze the intersectoral relationships and the percentage of ties in each sector. When considering that CSO workers are the secretariats’ implementing agents, the sociogram below (Figure 4) indicates that the SMS continued to have the highest percentage of ties (25.61%), in addition to the greater participation of SMDHC (12.2 %) and SMADS (10.98%). These were the most relevant secretariats in the program. GCM merged with SMSU, to which it was institutionally subordinate. The ties of all departments together represented almost 70% of the ties in the network, a relevant difference in relation to the Oficina Boracea program t₀ and t₁.

Figure 4
DBA’s sociogram (Secretaries + CSO)

When aggregating actors according to levels of bureaucracy (Figure 5), the ratio of high-level bureaucrats (BAE) was 14.86%; the mid-level (BME), 39.19%; and street-level (BNR), 18.92%. Those defined as “espec.” were the specialists. In Figure 5, the actors tied to academia were related to the organization where they worked in the territory. The organization was also considered an actor and identified as “espec.” for its extensive experience in harm reduction.

The sociogram in Figure 5 corroborates the literature: the mid-level bureaucracy plays a role in the ties between the high-level and the street-level bureaucracy, transforming decisions and guidelines into practice. As noted, no relationship was directly established between high-level and street-level bureaucracy. Mid-level actors also had a strong tie with specialists and organizations operating in the territory (i.e., they were permeable to the knowledge brought by these actors).

Figure 5
DBA program’s network, organized per level of bureaucracy

In the DBA network, it was not possible to identify a predominant profile of professionals. However, by focusing only on professionals related to the program’s characteristics, the network was predominantly composed of psychologists, physicians, and anthropologists. Of the 80 actors (nodes) in the DBA’s network, 19 already had professional experience or training in areas related to drugs, mental health, and harm reduction.

5. DISCUSSION OF RESULTS

The first aspect of the analysis was how public problems are defined among actors integrating intersectoral arrangements. The importance of this aspect was reinforced throughout the empirical research by the evidence that the tie between actors was only effective when there was some degree of shared views on the problem and how to approach it. This process often occurred incrementally, permeated by mutual and successive adjustments between participants, in line with Lindblom (1979), and through persuasion and learning, aspects cited by cognitive approaches to policies, such as in the studies by Hall (1993) and Sabatier (2007).

In line with the analytical model proposed by Shankardass et al. (2012), we observed that identifying which government sector leads the intersectoral construction and this sector’s relative power in relation to others makes a difference to the sustainability of the arrangement through time.

Thus, at the time t₀ of Boracea, the manager responsible for SAS played a central role in convincing and mobilizing actors from other government sectors. However, this conviction did not withstand the changes in government. The importance of the interaction between SMS and SAS was recognized only after a few years.

In the DBA program, many actors shared a view about the importance of treatment based on the harm reduction paradigm. In this sense, exchanges with specialists on the subject were essential (Hall, 1993). However, given the complexity of the theme/problem, actors reported a lack of clarity regarding the program’s main goal during implementation, which imposed difficulties in coordination among the different areas.

Another issue about problem definition was that, in the same territory, the municipal and state governments had programs with almost conflicting approaches and treatment methods. They started from quite different views on the problem and possible solutions, which again called into question the most simplistic views of public policies such as “applied problem solving” (Howlett, Ramesh, and Perl, 2013) as if problems were objectively “ready” in the world.

This research aimed to advance the analysis beyond formal coordination instances and mechanisms, highlighted in the literature on intersectorality (Cunill-Grau, 2014). Based on an initial hypothesis and empirical analysis, it was observed that the intersectoral programs’ modes of coordination can be consolidated gradually through different factors, not always relying on formal regulation. Formal arrangements for intersectoral coordination are often essential for the political legitimacy of high-level bureaucracy’s agendas, but they are not necessarily mobilized in the day-to-day policy implementation, as identified by the literature on intersectorality (Candel & Biesbroek, 2016), and in studies on building coordination strategies (Gontijo, 2012; O’Toole, 2010). Also, we had the opportunity to observe this dynamic regarding formal arrangements in previous research works (Bichir & Canato, 2019)

On the other hand, informal networking and coordination are often unintentionally constituted and can be essential to implementing intersectoral initiatives. It was also observed that, at times, civil society actors play a central role in coordinating public programs, in line with the findings reported by Brodkin (2011) and in the direction of the implementation arrangements discussed by Pires and Gomide (2018).

Paradigmatic disputes observed in the high-level bureaucracy did not necessarily interfere with informal cooperation strategies among street-level bureaucrats in the programs’ day-to-day. This phenomenon could be analyzed thanks to in-depth interviews with actors from different levels, as recommended in the international literature (Shankardass et al., 2012, p. 25). Thus, the methodological decision to map the programs’ entire networks beyond the limits formally defined proved to be the right one.

In Oficina Boracea program t₀, the formal instance created during the program’s formulation brought together different secretariats, business people, and representatives of the homeless population, which was important for negotiating specific issues regarding the facilities where the program’s initiative took place and represented the expectation of involvement from various sectors. However, this instance was ineffective in building lasting partnerships. After the formulation moment, the coordination and supervision of Boracea t₀ involved only actors from the SAS and partner CSOs.

In the case of the facilities where Oficina Boracea was conducted at t₁, no formal instance of intersectoral coordination was created. However, SAS supervisors held meetings to connect the program’s services with the SMS, adopting an operational problem-solving nature. The relationship between social welfare and health, at the institutional level, was perceived as asymmetric in terms of capabilities and resources, as health held more power in this regard. However, in the daily services, interactions were collaborative between mid-level and street-level bureaucracies of the different secretariats. Although there was no formal coordinator, the incentive for cooperation occurred through the encouragement of exchange and the perception of actors’ interdependence.

The DBA program had many intersectoral instances, some formalized and top-down appointed - such as the first executive group (GEM), which was formally instituted and was in charge of convening the secretariats to participate. Afterward, the management committee was created, which had political relevance and was consolidated as a decision-making space accessed by high-level bureaucrats. With the consolidation of the program and its governance arrangement, the high-level managers of each secretariat started weekly meetings, creating an instance later formalized as a coordination group. This instance and the role of coordinator emerged from the demand and the daily activities of these actors. Slowly, the decision-making process was transferred from the high-level to the mid-level bureaucracy, facilitating the flow of information and coordination.

Finally, another instance was created to coordinate the implementing agents. An instrument was developed that helped coordinate: the single registry of beneficiaries - with information on social welfare, health, and labor - updated by street-level agents, which helped systematize and share information among the secretariats involved.

Concerning decision-making processes, the literature on intersectorality considers it relevant to know which sectors make decisions, whether they are shared, and in which stages of public policy or program these decisions are made (Costa & Bronzo, 2012; Cunill-Grau, 2014; Shankardass et al., 2012).

In Oficina Boracea program t₀, the decision-making process was centralized at the high-level bureaucracy, involving mid-level actors, and decisions were sometimes imposed on the CSO. In Boracea t₁, the CSO had the autonomy to make decisions on some issues within the scope and resources provided in the agreement with secretariats or when managing to operate relying on other funding sources. Other decisions needed to pass through the secretariat’s mid-level bureaucracy. However, the centrality of the CSO in this service and in social welfare as a whole gave it bargaining and negotiation power in the relationship with different government sectors.

The DBA program showed a gradual shift of the centrality of decision-making from the high-level to the mid-level bureaucracy. This change, initially not formalized, was accompanied by the subsequent development of norms and regulations for new instances.

The implementation arrangements dimension, understood as portrayed in the most recent literature, highlights decision chains and relationships between state and non-state actors in the process of translating intentions into practices (Hupe, 2014; Lotta, 2019; Pires & Gomide, 2018; Winter, 2006). In Boracea t₀, the intersectoral and implementation arrangements showed prominence of the tie between SAS and CSOs, an effective relationship between SAS and SMS, and specific relationships with other sectors. In Boracea t₁ there was an evident tie between implementing agents of CSOs that worked in partnership with SAS and SMS, and these agents had difficulties accessing the high-level bureaucracy and influencing structural political decisions.

In the DBA, the intersectoral arrangement was built gradually and involved the intense ties of five government sectors and CSOs. There was a well-defined flow of information and coordination in the three levels involved, both horizontally - among actors at the same level from different sectors - and vertically - among actors in different levels of bureaucracy.

In line with Candel and Biesbroek (2016), the implementation arrangement of the DBA program was consolidated throughout the implementation process, with the interaction of several actors that gradually built the program’s coordination. There was an intense integration among sectors, as all phases of the program were somehow shared. Another determinant of intersectorality was the DBA’s visibility in the municipal agenda, which, from the beginning, involved the mayor and the highest levels of government. One respondent, a high-level bureaucrat of the SMS, considered that the DBA had some essential characteristics to make intersectoral relationships possible: “you have a defined territory, a very clearly delimited vulnerable population and a government, the mayor, as a protagonist, calling everyone. So there must be a territory, a defined population, and a management instance” (Canato, 2017).

In the cases analyzed, the daily presence of CSOs, aiming to create connections between users and bureaucrats involved in the implementation, was highlighted by the interviewees as a facilitating aspect of intersectorality. The interviewees considered that it would be more challenging to ensure that municipal civil servants were intensely involved with the day-to-day life of the territory and with highly vulnerable beneficiaries.

6. FINAL CONSIDERATIONS

This article analyzed the factors that explain the conformation and implementation of intersectoral arrangements without normatively assuming an ideal type of integration. In line with the analytical debate on policy integration and following assumptions observed in the literature on the implementation of arrangements and analysis of social networks, the study highlighted the relevance of formal and informal relations between state and non-state actors and among actors in different levels of bureaucracy, built during daily life. Formal arrangements are relevant to ensure political legitimacy and authority to programs, but they are only the starting point. Thus, it is important to analyze the dynamics of the relationships created among actors in the day-to-day of implementing public policies and programs.

The article corroborates the literature on intersectionality, which views totalizing perspectives with skepticism (Akerman et al., 2014) and highlights the importance of partial and incremental constructions (Candel & Biesbroek, 2016). The study stresses the relevance of incremental dynamics and relationships of coordination and communication among government sectors, between these and CSOs, and among the levels of bureaucracy, without assuming that program governance networks are necessarily horizontal.

It was also highlighted that political processes and decision-making dynamics that occur in daily life and at the level of implementation could have a decisive weight in the institutionalization of intersectoral arrangements. In addition to decisions made by high-level bureaucrats - legitimized by political mandates and more formalized intersectoral relations - adaptations, adjustments, and lessons learned at the street-level in the daily implementation, can help to redo decision-making processes and create new coordination structures, as long as these “bottom-up” flows are institutionally valued. Thus, incentive structures and coordination arrangements for intersectoral programs must be aware of what happens at the local level, with particular attention to intragovernmental communication strategies, as pointed out by Cunill-Grau (2014).

The main contribution of this study was to include in the analysis of the process of implementing and defining intersectoral arrangements, the use of a specific methodology (SNA) to evidence the conformation of networks and modes of interaction among actors from different areas of government, different hierarchical levels, and among CSOs workers. The relational dynamic mapped is not restricted to horizontal relationships. It also includes vertical relationships involving levels of bureaucracy, which can be decisive for the programs’ success.

It is a mistake to assume the horizontality in relations among different government sectors only because of the lack of formal hierarchy. In each case analyzed, the municipal secretariats had different levels of negotiating capacity, which facilitated or jeopardized building and consolidating relationships with other sectors.

A future research agenda could further and systematically investigate the role of CSOs in building networks and bridges with other sectors. In most social policies, CSOs workers are implementing agents, so it is relevant to analyze more cases in which these organizations eventually build their own intersectoral networks in different patterns of regulation and coordination, determining the process of public policy implementation.

• [Translated version] Note: All quotes in English translated by this article’s translator

REFERÊNCIAS

• Akerman, M; Sá, R; Moyses, S; Rezende, R; & Rocha, D. (2014). Intersetorialidade? IntersetorialidadeS!Ciência & Saúde Coletiva, 19(11), 4291-4300.
• Bichir, R. M; Brettas, G. H; & Canato, P. C. (2017). Multi-level governance in federal contexts: the social assistance policy in the city of São Paulo. Brazilian Political Science Review, 11, 1-28.
• Bichir, R. M; & Canato, P. C. (2019). Solucionando problemas complexos? Desafios da implementação de políticas intersetoriais. In R. Pires(Org.), Implementando desigualdades: reprodução de desigualdades na implementação de políticas públicas (pp. 243-266). Brasília, DF: Ipea.
• Bichir, R. M; Haddad, A. E; Hoyler, T; Lotta, G. S; Canato, P. C; & Marques, E. C. L. (2018). Early childhood in the city of São Paulo: the implementation of São Paulo Carinhosa Program in Glicério. Cadernos de Gestão Pública, 24, 1-23.
• Bichir, R. M; Oliveira, M. C; & Canato, P. C. (2016). Para além da transferência de renda? Limites e possibilidades na articulação intersetorial de políticas sociais. Brasília.Cadernos de Estudos, 26, 81-102.
• Brodkin, E. (2011). Policy work: street-level organizations under new managerialism. Journal of Public Administration Research and Theory, 21(2), 253-277.
• Bronzo, C. (2007). Intersetorialidade como princípio e prática nas políticas públicas: reflexões a partir do tema do enfrentamento da pobreza. InAnais do 20º Cogresso del CLAD sobre la Reforma del Estado y Modernización de La Administración Pública, Caracas, Venezuela.
• Canato, P. C. (2017). Intersetorialidade e redes sociais: uma análise da implementação de projetos para população em situação de rua em São Paulo (Dissertação de Mestrado). Universidade de São Paulo, São Paulo, SP.
• Candel, J. L; & Biesbroek, R. (2016). Toward a processual understanding of policy integration. Policy Sciences, 49, 211-231.
• Carmo, M; & Guizardi, F. (2017). Desafios da intersetorialidade nas políticas públicas de saúde e assistência social: uma revisão do estado da arte. Physis - Revista de Saúde Coletiva, 27(4), 1265-1286.
• Costa, B. L. D; & Bronzo, C. (2012). Intersetorialidade no enfrentamento da pobreza: o papel da implementação e da gestão. In C. A. P. Faria (Org.), Implementação de políticas públicas: teoria e prática Belo Horizonte, MG: PUC Minas.
• Cunill-Grau, N. (2005). La intersectorialidad en el gobierno y gestión de la política social. In Anais do 10º Congresso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Santiago, Chile.
• Cunill-Grau, N. (2014). La intersectorialidad en las nuevas políticas sociales: un acercamiento analítico-conceptual. Gestión y Política Pública, 23, 5-46.
• Gontijo, J. G. L. (2012). Coordenação, cooperação, e políticas públicas: organizando percepções e conceitos sobre um tema caro à implementação. In C. A. P. Faria (Org.), Implementação de políticas públicas: teoria e prática Belo Horizonte, MG: PUC Minas .
• Haas, P. M. (1992). Epistemic communities and international policy coordination. International Organization, 46(1), 1-35.
• Hall, P. (1993). Policy paradigms, social learning and the state: the case of economic policymaking in Britain. Comparative Politics, 25(3), 275-96.
• Hanneman, R. (2001). Introduction to social network methods Berkeley, California: University of California.
• Howlett, M; Ramesh, M; & Perl, A. (2013). Política pública, seus ciclos e subsistemas: uma abordagem integradora Rio de Janeiro, RJ: Elsevier.
• Hupe, P. (2014). What happens on the ground: persistent issues in implementation research. Public Policy and Administration, 29(2), 164-182.
• Inojosa, R. M. (2001). Sinergia em políticas e serviços públicos: desenvolvimento social com intersetorialidade. Cadernos Fundap, 22, 102-110.
• Junqueira, L. (2000). Intersetorialidade, transetorialidade e redes sociais na saúde.Revista da Administração Pública, 34(6), 35-45.
• Kingdon, J. (1984). Agendas, alternatives, and public policies Boston, Massachusetts: Little, Brown and Company.
• Lindblom, C. (1979). Still muddling, not yet through. Public Administration Review, 39, 517-526.
• Lipsky, M. (2010). Street-level bureaucracy: dilemmas of the individual in public service New York, NY: Russell Sage Foundation.
• Lotta, G. (2014). Agentes de implementação: uma forma de análise de políticas públicas. Cadernos Gestão Pública e Cidadania, 19(65).
• Lotta, G. (2018). Burocracia, redes sociais e interação: uma análise da implementação de políticas públicas. Revista de Sociologia e Política, 26(66), 145-173.
• Lotta, G. (2019). A política pública como ela é: contribuições dos estudos sobre implementação para a análise de políticas públicas. In G. Lotta(Org.), Teoria e análises sobre implantação de políticas públicas no Brasil Brasília, DF: Enap.
• Lotta, G; Galvão, M; & Favareto, A. (2016). Análise do Programa Mais Médicos à luz dos arranjos institucionais: intersetorialidade, relações federativas, participação social e territorialidade. Ciência & Saúde Coletiva, 21(9), 2761-2772.
• Lotta, G; Pires, R; & Oliveira, V. (2014). Burocratas de médio escalão: novos olhares sobre velhos atores da produção de políticas públicas. Revista do Serviço Público, 65 (4), 463-492.
• Lowi, T. (1972). Four systems of policy, politics, and choice. Public Administration Review, 32(4), 298-310.
• Marques, E. (2000). Estado e redes sociais: permeabilidade e coesão nas políticas urbanas no Rio de Janeiro Rio de Janeiro, RJ: Revan/Fapesp.
• Marques, E. (2013). As políticas públicas na ciência política. In E. Marques, & C. Faria (Orgs.), A política pública como campo multidisciplinar São Paulo, SP: Unesp.
• Massardier, G. (2006). Redes de política pública. In E. Saravia, & E. Ferrarezi(Orgs.), Políticas públicas(Coletânea, Vol. 2, pp. 167-186). Brasília, DF: Enap .
• Mendonça, P. M; Medeiros, A; & Araújo, E. (2019). Models for government-nonprofits partnerships: a comparative analysis of policies for aids, social assistance and culture in Brazil. Revista da Administração Pública, 53(5), 802-820.
• Menicucci, T. (2018). Perspectivas teóricas e metodológicas na análise de políticas públicas: usos e abordagens no Brasil. Revista Política Hoje, 27(1), 42-55.
• O’Toole, L Jr;. (2010). Relações interorganizacionais no processo de implementação. In G. Peters & J. Pierre (Orgs.), Administração pública: coletânea São Paulo, SP: Unesp .
• Pires, R. C. P. (2012). Estilos de implementação e resultados de políticas públicas: fiscais do trabalho e o cumprimento da lei trabalhista no Brasil. In C. A. P. Faria (Org.), Implementação de políticas públicas: teoria e prática Belo Horizonte, MG: PUC Minas .
• Pires, R. C. P. (2016). Arranjos institucionais para implementação de políticas e ações governamentais em direitos humanos. In A. L. M. Delgado, C. I. Gatto, M. S. Reis, & P. A. Alves(Orgs.). Gestão de políticas públicas de direitos humanos Brasília, DF: Enap .
• Pires, R; & Gomide, A. (2018). Governança e capacidades estatais a partir da abordagem dos arranjos e instrumentos de políticas públicas. Boletim de Análise Político-Institucional, 19, 25-32.
• Sabatier, P. (2007). The need for better theories. In P. Sabatier(Ed.), Theories of the policy process (2a ed.). New York, NY: Westview Press.
• Saetren, H. (2014). Implementing the third-generation research paradigm in policy implementation research: an empirical assessment. Public Policy and Administration, 29(2), 84-105.
• Shankardass, K; Solar, O; Murphy, K. Greaves, L; & O’Campo, P. (2012). A scoping review of intersectoral action for health equity involving governments.International Journal of Public Health,57, 25-33.
• Silva, L. (2013). A construção federal da intersetorialidade na política de desenvolvimento social brasileira: o caso do Programa Bolsa Família (Dissertação de Mestrado). Universidade de São Carlos, São Carlos, SP.
• Tosun, J; & Lang, A. (2017). Policy integration: mapping the different concepts. Policy Studies, 38(6), 553-570.
• Veiga, L; & Bronzo, C. (2014). Estratégias intersetoriais de gestão municipal de serviços de proteção social: a experiência de Belo Horizonte. Revista de Administração Pública, 48(3), 595.
• Veiga, L; & Carneiro, C. (2005). Desafios, possibilidades e limites de programas locais de combate à pobreza e à exclusão: as iniciativas das cidades de São Paulo de Belo Horizonte. In Anais do 10º Congresso Internacional Del Clad sobre Reforma del Estado y Modernización de la Administración Pública, Santiago, Chile.
• Wasserman, S; & Faust, K. (1994). Social network analysis: methods and applications New York, NY: Cambridge University Press.
• Winter, S. (2006). Handbook of Public Policy Londres, UK: Sage.

Publication Dates

• Publication in this collection
  27 Sept 2021
• Date of issue
  Jul-Aug 2021

History

• Received
  23 Aug 2020
• Accepted
  05 May 2021