Collaborative governance outcomes and obstacles: Evidence from Portuguese armed forces

Abstract The shortage of resources, which threats the accomplishment of the Armed Forces’ Missions, has urged them to increase their Branches’ integration to improve their efficiency as a whole. However, previous research pinpointed several issues in the Portuguese Armed Forces’ (PAF) governance that have threatened the intended integration’s efficacy. Therefore, we analyse PAF’s governance model at the Military Strategic Planning (MSP) level to understand its suitability to PAF’s current and future needs. We extend theoretical understanding of collaborative governance in defence, using a mixed method approach through a case study analysis, conducting structured and semi-structured interviews (15 interviewees), a survey (97 answers), and several document analyses. Participants’ sample includes military and civilian personnel working on MSP or financial resources management. The results highlight that PAF’s governance model is mostly a classic Public Administration model that is not sufficiently suited to answer its integrative needs. This study identifies the main obstacles of Defence Military Branches’ integration and provide theoretical propositions to frame collaborative governance in Public Administration. It also shows evidence to back up the importance of collaborative governance models, supported on interdependent networks, and argue the impact that non-rational factors (such as organizational culture) can have in public entities’ governance.


PUBLIC INTEREST STATEMENT
We analyse Portuguese Armed Forces (PAF)'s governance model at the Military Strategic Planning (MSP) level to understand its suitability to PAF's current and future needs, since the shortage of resources, which threats the accomplishment of the Armed Forces' Missions, has urged them to increase their Branches' integration to improve their efficiency as a whole. We present a case study analysis, conducting structured and semi-structured interviews (15 interviewees), a survey (97 answers), and several document analyses. Participants' sample includes military and civilian personnel working on MSP or financial resources management. This study identifies the main obstacles of Defence Military Branches' integration and provide theoretical propositions to frame collaborative governance in Public Administration. It also shows evidence to back up the importance of collaborative governance models, supported on interdependent networks, and argue the impact that non-rational factors (such as organizational culture) can have in public entities' governance. three different sources (. . .) converge" (p. 161). Therefore, this study argues an overview of the PAF's governance model, revealing its potentialities and vulnerabilities, identifying what is difficulting the intended integration, and indicating future challenges regarding the evolution of its governance system.
The results suggest that, although bureaucratically, there is an intention to adopt more recent network collaboration models, but the current characteristics of the Portuguese Armed Forces' governance model are mostly features of a classical public administration model. Besides, the model's excessive bureaucracy, and forceful hierarchy, the detached organizational culture within each Military Branch, the lack of some cross methodologies among them, and their competition for the available resources compromise Portuguese Armed Forces' integrative needs. These features make it difficult to obtain efficiency gains and compromises its ability to respond effectively to its missions, exposing a governance model that is not yet sufficiently adequate to meet its needs.

Literature review
Challenges currently faced by governments require quick, flexible, and inclusive responses given from collaborative networked structures that also consider the efficiency and efficacy of those decisions (Agranoff & McGuire, 2003;Klijn & Koppenjan, 2016;Osborne et al., 2013;Pedersen & Johannsen, 2018;Provan & Kenis, 2007;Torfing & Triantafillou, 2013). This demand required a broader approach to governmental action, outdating the hierarchy-focused Public Administration theory, and the performance-focused New Public Management theory. Thus, Osborne (2006) has argued a new governance approach centred on collaboration, interdependence, and trust: the New Public Governance.
Governance is a broader concept than management since it also includes planning and decisionmaking processes (Ansell & Gash, 2007). Due to the complexity of the political and social system, this new paradigm argues that the efficacy of public services' provision depends on the collaboration and trust among inter-organizational networks' members (Haveri, 2006;Lindsay et al., 2014;Osborne, 2006;Osborne et al., 2013). Furthermore, as the Public Sector needs to do "more with less," several authors highlight the importance of improving collaboration and partnerships' formation between governments and other public, private, and non-profit actors. This idea shaped the concept of Collaborative Governance (Entwistle & Martin, 2005;Farrell & Goodman, 2013;Lowndes & Skelcher, 1998;Scholl & Scholl, 2014).
This theory crosses the boundaries of public institutions, government levels, and public, private, and civil domains to effectively and efficiently achieve a public purpose that would not be possible otherwise (Emerson et al., 2012). This paradigm highlights participants' interdependency, focusing on resolving their conflicts, and improving their shared trust and legitimacy. These features allow them to integrate more relevant knowledge into the decision-making process (Alford & Hughes, 2008;Bryson et al., 2006;Emerson et al., 2009;Entwistle & Martin, 2005;Farrell & Goodman, 2013;Skelcher et al., 2005).
Despite the success of this type of relationships (Agranoff & McGuire, 2003;Meier & O'Toole, 2003;Provan & Milward, 1995), Bryson et al. (2006) and Huxham et al. (2000) state that collaborative relationships require great effort and that sometimes they do not meet the conditions to create the necessary level of trust. Therefore, the literature highlights four requirements to ensure the success of these relationships: (1) the development of trust among partners, which brings closer together the needs, interests, values, and constraints of the collaborative partners; (2) the existence of common goals and a shared understanding of the complexity and diversity of their relationships with one another; (3) the existence of collaborative incentives that encourage different stakeholders to share their knowledge; (4) the presence of specific leadership skills (Escortell et al., 2020) such as negotiation (through consensus-building) and conflicts' resolution abilities (Agranoff & McGuire, 2003;Alford & Hughes, 2008;Ansell & Gash, 2007;Bryson et al., 2006;Emerson et al., 2012;Haveri, 2006;Huxham & Vangen, 2013;Kickert & Koppenjan, 1997;Koppenjan & Klijn, 2004;Leach & Sabatier, 2005;Tonelli et al., 2018) As a result of this complexity, Emerson et al. (2012) developed a model that tries to explain the collaborative dynamics among organizations. This model was called "Collaborative Governance Regime," and it includes six components: the System Context, the Drivers, the Collaborative Dynamics, the Collaborative Action, the Impacts, and the Adaptation. The System Context encompasses political, socioeconomic, cultural, environmental, and resource variables that influence this system's various entities. This context might contain elements (Drivers) that encourage those entities to start a Collaborative Dynamics, which includes the Principled Engagement, Shared Motivation, and the Capacity for Joint Action. Principled Engagement is the interaction between different organizations to solve common problems and create value. It involves basic principles such as open, inclusive, fair, and civilized communication and all stakeholders' representation. This interaction triggers the development of Shared Motivation, the interpersonal and relational aspect of collaborative relationships, which consists of a sequential cycle of four elements: mutual trust, mutual understanding, internal legitimacy, and shared commitment. Principled Engagement and Shared Motivation together settle the conditions that allow Capacity for Joint Action to develop. This model component focuses on developing capacities that can generate results that each participant could not achieve separately.
It is also important to note that governments' economic and social challenges go beyond their traditional bureaucratic structures' response capabilities in this dynamic, uncertain, and complex geopolitical environment. These challenges require timely and flexible responses and a better ability to predict, adjust, and innovate. According to several authors, this requires the integrated use of Information and Communication Technologies (ICT) in a collaborative network that allows the collection and processing of all information relevant to decision-making (Farrell & Goodman, 2013;Gaulė, 2014;McGuire, 2006;Scholl & Scholl, 2014;Šiugždinienė et al., 2017).
Empirical evidence suggests that the most appropriate governance model depends on the task's nature and context, and the available technologies and resources (Alford & Hughes, 2008). We can observe this contingency in several empirical cases. Håkenstad and Larsen (2012) compared the governance and planning models used in the Ministries of Defence of seven European countries (Norway, Sweden, Finland, Denmark, The Netherlands, The United Kingdom and France). They concluded that they were more different than similar in their participants, chronologies, perspectives, structures, and products. The study's main conclusion is that each country's unique traditions and customs shape their planning models' governance. Concerning Norwegian governance, their response to COVID-19 pandemics was based on a top-down collaborative decision-making style, involving political, administrative, and expert levels. High mutual trust relations, consensusbuilding, and open argumentation and feedback dictated its success (Christensen & Laegreid, 2020). Regarding the governance models themselves, Šiugždinienė et al. (2017) identified that the Lithuanian government has a traditional governance model, making their decisions centrally. Although there are formal interaction platforms that allow collaboration between institutions, their success depends on developing and maintaining personal relationships (Šiugždinienė et al., 2017). The literature also points hybrid governance models that have features of several approaches. Iacovino et al. (2017), for example, found that in the Tuscany region of Italy, the three primary forms of governance coexist. Hence, PA dominates in top managers' experience, NPM prevails in policymakers and the expectations of top managers, and NPG can be found in laws and documents. To justify this disparity, the authors argue that there might be resistance to change due to the bureaucratic nature of Italy's culture.
Extant literature has been putting to the test the theories and concepts that seek to explain managers' choices rationally. As such, new explanations take into account the culture, beliefs, and values of organizations (Denzau & North, 2000;Kooiman & Jentoft, 2009;Leach & Sabatier, 2005;Weare et al., 2014). Kooiman and Jentoft (2009), for example, claim that values, norms, and principles guide decision-makers implicitly, shaping the way they think, judge, and act. Leach and Sabatier (2005) also admit that an organizational community's cultural norms and other characteristics can predispose individuals to trust and behave reliably with others. Besides, Weare et al. (2014) concluded that despite common goals that could attract organizations, the collaboration between them might end due to differences in their visions and individual behaviour. These results are supported by Denzau and North (2000), who argue that these differences result from each organization's culture, where people share the same mental model. Thus, Kooiman and Jentoft (2009) claim that the success of organizations' interactions depends on explicitly stating their sets of values, identifying differences of opinion and individual skills, and explore potential areas of consensus-building. These authors explain that the only way to this is by designing an interactive learning process where organizations learn from each other in a continuous and shared process.
While recent literature urges higher Public Administration collaboration to face today's geopolitical uncertainties and demands, some obstacles might arise during these collaboration dynamics. Hence, our research aims to analyse what obstacles may hamper a real Collaborative Governance Regime in the Military Branches.

Methodology
Among all Public Administration sectors, we argue that military Branches had the highest incentive to collaborate with each other. Therefore, we conducted a case study regarding the Portuguese Armed Forces' governance model in their Military Strategic Planning. We analysed primary non-confidential documents connected to the Portuguese military planning to understand its various phases, structure organization, and political guidelines during the first stage of this investigation. Secondly, we conducted 15 interviews with "privileged witnesses" (Quivy & Campenhoudt, 2003) using the snowball method to choose the respondents in which each interviewer recommends the next most relevant person to be interviewed (Fylan, 2005;Leech, 2002;Yin, 2016). The interview guide is adapted from Emerson et al. (2012). All military personnel interviewed are currently working or have worked in at least one of the strategic planning phases and were coded with the letter E followed by a random number. The interviewees belong to the following military Branches: Navy (3), Army, (6), and Air Force (6). The interviews took place between May and June 2020.
Finally, we ran a questionnaire survey to understand military personnel's perceptions about the Portuguese Armed Forces' current governance model and its possible future evolution. This survey was addressed to military and civilian members of the three Branches who currently participate or have already participated directly in any of the military planning phases (61) or perform financial resource management functions (36), therefore getting 97 responses. This questionnaire was separated into two parts. First we measured respondents' perceptions about collaborative governance concepts based on recent literature- Klijn et al. (2010) and Šiugždinienė et al.'s (2017) questionnaires and Ansell and Gash (2007), Bryson et al. (2015), Emerson et al. (2012), and Kooiman and Jentoft (2009). Secondly, we assessed respondents' perceptions about the need to implement military planning measures foreseen in the Plan's Grand Options for 2020-2023 (The Presidency of the Council of Ministers, 2019)-the government's strategic document stating their primary objectives for their mandate. We measured each item using a Likert even scale (six points) to mitigate the impact of common response bias (Boone & Boone, 2012). The results were explored qualitatively, but leveraging the number of participants in this study. Before analysing the results, we tested the data's internal consistency using Cronbach's Alpha and obtained 0.939 for the data set (Gliem & Gliem, 2003). We coded with the letter Q, followed by a random number, the comments collected on the answers to the survey.
We use Emerson et al.'s (2012) Collaborative Governance Regime model to frame our data exploration. We chose this model due to its relevance and its adjustment to exploratory studies, given its precise descriptive nature.

Case study: Portuguese armed forces' strategic military planning
Portugal is currently facing an uncertain and complex environment with unpredictable threats, which increases its military alliances' responsibilities. Furthermore, there is also a growing concern with efficiency due to resource shortage. 1 Therefore, it is urgent to create joint capabilities and establish network relationships that allow higher collaboration and integration between the Branches (Colom, 2017;Delgado, 2017;Favinha, 2016;Spiegeleire, 2011;Stojković et al., 2016;Tagarev, 2009;Yue & Henshaw, 2009). Tagarev (personal communication,24 March 2020) admits that currently, "The general trend is towards increase cooperation (often ad hoc), coordination and collaboration [between actors within a country's security sector]." This trend creates the need to make a systemic approach to the defence sector, making its organizations specialize in developing unique capabilities (Spiegeleire, 2011). This approach creates synergies from a good governance perspective, guaranteeing the highest security and an efficient, transparent, and responsible use of limited public resources (Tagarev, 2009).
To understand military governance, we have to understand the military planning framework first. After Cold War, the global geopolitical framework has become more complex, unpredictable, and dynamic than before, preventing States from clearly identifying their threats (Christianson, 2016;J. Correia, 2019;Doughty et al., 2015;Ribeiro, 2019;Stojković et al., 2016). Therefore, it became necessary to use a planning model that could identify the capabilities that States must have to respond flexibly to this environment's uncertainty. Thus, the Capabilities-Based Planning emerged (Davis, 2002;Davis et al., 1998; International Centre for Defence and Security (ICDS), 2008; Rumsfeld, 2001;Spiegeleire, 2011;Walker, 2005).
A "capability" is the set of resources and skills required to accomplish certain tasks under specific circumstances and following pre-defined performance standards, generating the necessary effects to achieve the desired goals (J. Correia, 2019; Department of Defense, 2019; Stojković et al., 2016;Tagarev, 2009).
Since its conception, Capabilities-Based Planning is the most chosen planning process by States (Stojković et al., 2016). However, concerning the Spanish government's planning model, Colom (2017) points out the lack of metrics and analysis methodologies capable of evaluating planning progress, making it hard to apply lessons learned in future planning processes. On the other hand, Young (2019) argues that the Armed Forces' military commanders in central and eastern Europe do not have enough budgets or power to ensure adequate force readiness. The author concludes that the decentralization of ministries' decision-making power to military commanders along with higher budget allocations are sine qua non conditions for creating military forces capable of responding to their requests. As for Australia, Young (2006) encourages joint capabilities development and the implementation of a joint command centre that would connect the military and the civilian security forces.
In response to the geopolitical environment's uncertainty, Portugal decided to adopt the Capabilities-Based Planning in 2011, using the Military Programming Law (Parliament, 2015) as its main instrument for managing and controlling national military capabilities. However, several authors have already pinpointed financing issues regarding this investment instrument, which might threaten PAF's ability to re-equip itself and consequently undermining the accomplishment of its missions (Delgado, 2017;Mata, 2013;Audit Office, 2017;Tendeiro, 2017, as cited in Da Costa, 2018. Accordingly, Mata (2013) also argues that capabilities sustainability has not received sufficient funding. The author also states that the Capabilities-Based Planning lacked political commitment as the Strategic Concept for National Defence's objectives were too broadly defined. The author concludes by suggesting that the Military Programming Law is a mere plan of intentions often disrespected, urging the need to make choices and be realistic. Together with these criticisms, there are also other critics related to the weak integration of processes between the key players of National Defence (Central Organs and Services of the Ministry for National Defence, Armed Forces General Staff (AFGS), Navy, Army, and Air Force). Delgado (2017) and Martins (2015) argue that each of the key players individually interprets MSP's central documents. According to Delgado (2017), this hardens strategies' standardization and creates efforts' redundancy to achieve common goals. Thus, to overcome these difficulties, the author suggests the cross standardization of the strategic management instruments among them. The author also explains that the Branches have different risk assessment algorithms used to prioritize different military capabilities, creating incoherent and questionable investment proposals among them. The author concludes that resource constraints result in interaction's weaknesses between key players, namely lack of consensus, competition for resources, and lack of standardization in prioritizing capabilities (Delgado, 2017). Given the urgency to increase resources allocation efficacy and efficiency, several authors and core planning documents call for greater integration between military Branches, pointing out the importance of building joint military capabilities and creating a joint and integrated National Defence's planning (Delgado, 2017;Martins, 2015;Mata, 2013;Rodrigues, 2015;Strategic Concept for National Defence 2013;General Staff Council, 2014). However, Rodrigues (2015) argues that current joint capabilities result from the mere merging of each Branch's resources, instead of resulting from their planning and development from scratch, suggesting a reassessment of this process. In this regard, Madeira (2009) also pointed out the Branches' need to develop inter-agency cooperation capabilities that allow National Defence to interact with other State departments and civil entities. Besides, just as Colom (2017) identified in the Spanish governance model, Audit Office's (2017) audit report about the Portuguese 2015 Military Programming Law also alerted that there were no performance indicators to evaluate capabilities implementation. Finally, in Pires (1999) study about PAF's governance reform, one of his interviewees-Commander Moreira-highlighted also the importance of working on the military organizational culture, stating that in addition to rethinking the organization, we have to face mentalities.

Collaborative Outcomes
PAF's intended outcomes are at stake, namely its needs to get higher efficiency while ensuring an adequate ability to respond effectively to its missions. In part, this is because there are not enough financial-and human (Audit Office, 2017; E1; E5 2 )-resources to meet the Branches' needs and the ambition level defined. The Military Programming Law (Parliament, 2019) insufficient funding illustrates this scarcity (E1; E2; E3; E5; E6; E7; E9; E13). This underfunding generates capability gaps, keeps military capabilities' operational qualifications only at the minimum levels required, and slows the assembly of new capabilities (E2; E6; E7). Thus, the PAF needs to develop solutions that increase efficiency gains, take advantage of economies of scale, and ensure adequate readiness, interoperability, and responsiveness to national needs (E1; E10; E13; E15; Ministry of National Defense, 2020)

Collaborative actions
Due to this shortage of resources, the importance of obtaining efficiency gains has increased over the last decade (E5; E8; E13; E14; E15; Q1). Several interviewees point out that the best solutions to achieve PAF's intended outcomes include strengthening integration, synergies, and cooperation mechanisms among the Branches (E1; E3; E7; E8; E10; E15; Q4). There is even a great political pressure (E1) to deepen the Branches' integration, articulation, and sharing, namely through the reinforcement of the PAF's joint command-suggestion also made by Young (2006) to the Australian model-and the establishment of a higher number of joint and integrated forces and capabilities. This political pressure is evident in the MSP's primary documents (General Staff of Armed Forces, 2018; Military Strategic Planning Division, 2020; Ministry of National Defense, 2020; The Presidency of the Council of Ministers, 2013a, 2013b, 2019) and a criterion for prioritizing capabilities that benefits the ones that can engage in joint operations (E1; E2; E3).
In fact, according to the Organic Law for the Organization of the Armed Forces of 2014, the Branches' primary mission is to participate, in an integrated manner, in the Republic's military defence. Besides this legal imperative, nowadays, the military also gives higher importance to the Multi-Domain-land, sea, air, space, and cyberspace-, which requires interoperability to become a capabilities' building elementary condition. This interoperability ensures that capabilities allow a modular and flexible performance in all domains, increasing its integration according to PAF's collective purposes (E4; 2020 Ministerial Directive for Military Defence Planning). This interoperability is already lightly included in the current Military Programming Law (E4) since this 2019 version already introduces the concern with the creation of a coherent and integrated vision for national defence and highlights the role of interoperability, flexibility, and adaptability within the Military Forces Arrangement. Developing capabilities and joint and integrated processes promote institutional and operational synergies, economies of scale, and efficiency during capabilities' acquisition (E3; E15). There are already some joint and integrated capabilities and skills created from structures such as the Armed Forces' Hospital, the Military University Institute, the Armed Forces Institute for Social Care, the Integrated Management System of National Defence, and the Cyber Defence Centre (E1; E4; E13; E14; E15). However, E7 claims that each Branch has redundant infrastructures and services that should be integrated into AFGS. Due to Branches' singularities, some interviewees are afraid that operational integrations might disrupt operational productivity. Therefore, they suggest that the fields with greatest integration potential (E7) for efficiency gains are administrative, financial, logistical, and support areas like ICT (E1; E8; E10).
However, E4 believes that PAF still has a long way to go since its joint capabilities are still very few. This interviewee claims that there are only some joint procedures and some joint fields mostly in the support area, and that everything else turns out to be a branch-like structure. Although in theory, Branches' final product is joint and approved by the Chiefs of Staff Council, this branch-like structure can be seen during their strategic planning since the Branches continue to conduct their priorities' definition processes independently from each other (E9). This situation happens because all Branches need to re-equip themselves, making them compete for the available financial resources (E9). Thus, only joint military capabilities are included in Joint Force Proposals and are susceptible to joint debate and decision-making. This proceeding means that each Branch contributes to the strategic objectives established only in the right measure (E1).
Besides this gap in obtaining adequate integration among the Branches, there are also other output problems in this collaboration process. First, there are no procurement cells that allow Branches' knowledge integration, hindering resources' marshalling output (E6). Second, regarding new management practices, there are no lessons learned systems to provide learning opportunities among the Branches (E7; E10; E11). Third, there is also a problem regarding policy enacting outputs in civil-military relations since the primary strategic military documents are outdated (E4; E8; E12). Finally, concerning implementation supervision, three years after Audit Office's (2017) audit report, there are still no Key Performance Indicators to monitor capability assembly, check for disparities, prepare forecasts, and make decision-makers accountable (E5). However, a tool that regularly measures military capabilities is currently under development (E10). Therefore, PAF is still a long way from the inter-agency principle, where its Branches operate together in the Multi-Domain with high interoperability (E4; E11; E13). Thus, E4 argues that the political, economic, and social areas must be integrated to obtain a solid Holistic Command, as recommended in the 2020 Ministerial Directive for Military Defence Planning. Hence, the 2020 AFGS Strategic Military Planning Manual encourages political, military and business sectors' coordination by deepening the relationships between the PAF, the Technological and Industrial Base of Defence, and the National Scientific and Technological System. This coordination enhances innovation and economies of scale that allow PAF to close the identified gaps efficiently (2020 Ministerial Directive for Military Defence Planning; AFGS Strategic Military Planning Manual). Thus, political power should implement a profound restructuring in PAF's organization (Q5), conveying the message that the best way to defend Portugal is through greater collaboration (E13).

Collaborative dynamics
4.1.3.1. Principled engagement. According to Emerson et al. (2012), Principled Engagement encourages civilized debate, open and inclusive communication, the inclusion of all relevant stakeholders, and decision-making based on all participants' perspectives and knowledge. This inter-Branches interaction occurs during formal meetings led by AFGS and seminars and exchanges between the Branches where each participant's problems and needs are shared (E10; E13), contributing to harmonizing their priorities (E3). However, this inclusion only occurs at the joint capabilities' level (E1) since each Branch is responsible for its specific genetic strategy and has the autonomy to manage its programs and define the priority of its investments (E9). Therefore, each Branch defines its own equipment's operational requirements to be acquired. Nevertheless, the other Branches can also contribute to this task, particularly when they also benefit from this equipment (E1; E2).
As investment choices depend on financial resources, E4 reinforces the importance of considering all costs during decision-making processes. Together with mathematical factors, E3 argues that algorithms used to make decisions should also consider decision-makers' perceptions and knowledge acquired by experience, previous analysis of the past, and future forecasts. On one hand, the adoption by the three Branches of the priority definition model defined by the North Atlantic Treaty Organization (NATO) standardized this process-which did not happen in previous cycles (E3). On the other hand, there is still no methodology for analysing and solving problems during decision-making processes (E7), nor a standard methodology for identifying needs and jointly defining capability requirements (E14). According to E14, this last flaw is Directorate-General of Resources of National Defence's most significant problem during the integration of Military Programming Law's investment proposals from each Branch. Hence, E14 believes it is essential to find an analytical tool that enables these processes' integration.
The Branches also have several platforms where alternatives and decisions are listed and discussed, namely: the Chiefs of Staff Council; governance boards; debating groups; strategy development groups where different specialists from each Branch participate; and ad hoc platforms created during specific projects (E1; E6; E10; E13). During decision-making, the Branches pay special attention to the basic principle of strategic planning: balancing available resources, necessary capabilities, and international commitments (E4). Thus, decision-making seeks the most effective solution from a Bes-Value-for-Money perspective (E3).

Shared motivation.
Building trusting relationships is essential to create a collaborative environment among organizations and developing a proper strategic plan (E6; E8; E11; E12; E13; E14). These relationships facilitate coordination between members of the different bodies and accelerate the decision-making process (E6), allowing the creation of synergies that increase operational productivity and enhance efficiency gains (E8). E13 even argues that the development of trusting relationships will soon be the primary driver of significant efficiency gains-laying aside the prevailing hierarchy and bureaucracy. (E13). The joint education given to the military personnel of the three Branches at the Military University Institute (E7; Q4), processes' transparency (E4), and the reduction of bureaucracy (E15) all contribute to the development of these relationships. However, the high turnover of positions, the lack of communication between the political, strategic, and operational levels, and the lack of transparency of the criteria used during political decisionmaking can threaten these relationships (E7). Therefore, it is necessary to reinforce a clear language, providing greater transparency and confidence to the processes (E4) and mutual trust (E12), giving greater importance to the construction of this type of relationship (E13). Regarding Shared Commitment, E3 states that governance works well because each employer is not just working to himself, to which E4 adds that the current Military Programming Law sought to obtain more integrated capabilities for the collective purposes. This collective commitment also contributes to AFGS and Ministry for National Defence's existence, where soldiers from the three Branches work for the same purpose (E14). However, the rivalry that is still felt among the Branches is one of the main factors that hinder the creation of a collective commitment and the integration of their strategies (E13; E15).

Capacity for joint action.
Inter-Branches institutional arrangements comprehend structures and groups that aid Branches' collaboration. First, E3 highlights the role of the Defence Planning Assistance Group as a transparent mechanism that allows the meeting, discussion, understanding, and harmony of those participating in the MSP, enabling the achievement of PAF's strategic objectives. Second, to face the lack of standard rules in decision making, E1 emphasizes the Military Strategic Planning Division's role in harmonizing strategic planning between the Branches. Finally, E7 enhances the role of the Directorate of Weaponry and Equipment Services (located in the Directorate-General of Resources of National Defence) in harmonizing its different needs while considering the determined political guidelines.
Regarding procedural arrangements, PAF has been receiving new missions related to public interest activities coordinated with civil entities such as the Civil Protection, the National Republican Guard, town halls, and the Institute for the Conservation of Nature and Forests (E3; E4; E6; E15; Plan's Grand Options for 2020-2023). This growing military participation in civilian missions and the concern of acquiring dual-use capabilities-which can be used by military and civilian entities (E1; E4)-means that PAF's planning is not exclusively military (E15). However, the 2014 Military Strategic Concept focuses only on the PAF, 2020 Ministerial Directive for Military Defence Planning is mainly limited to military missions, and 2013 Strategic Concept for National Defence is not adequate nor up to date (E4; E8; E15). For this reason, it is necessary to introduce the characterization of what is happening in civil society in MSP's doctrine (E15) by developing a Strategic Concept of Security and Defence that models the interaction between PAF and the civil entities with whom they work (E8).
It is also essential to notice that even though current legal documents promote integrated collaborative efforts, this is not entirely happening in practice. That is why E15 states that it is necessary to give a clear architectural and governmental expression to all these concepts by turning theory into practice.
During military planning, information and knowledge sharing between the Branches are crucial to rigorously perform the assigned functions and increase PAF's integration (E15). Therefore, using the know-how of a large number of stakeholders through, for example, the establishment of lessons learned and the dissemination of good practices makes it possible to diminish bureaucracy and make the development of military capabilities more agile, promoting efficiency gains (E7; E10; E11). These suggestions confirm the importance of implementing interactive learning systems, as recommended by Kooiman and Jentoft (2009). On the one hand, the joint education of the three Branches at the Military University Institute promotes the creation of a similar base of values and the establishment of a contact network among them, improving inter-organizational communication flow. On the other hand, military institutions' intrinsic hierarchy can limit this fluidity (E7). Accordingly, several interviewees argue that the lack of communication and promptly information sharing between the Branches is what hampers their skills and resources' sharing the most (E5; E6; E7; E13; E15). Contributing to this obstacle is the slowness of this governmental system's hierarchical/bureaucratic circuits, often only countered with ad hoc solutions (E7; E13). To overcome this challenge, E5 and E6 suggest creating an integrative platform with all MSP's key players that allows cross and strategic communication to assist political and military decision-makers in making their decisions.
Thus, E6 suggests creating procurement cells-a concept that involves buying goods and services for the government (Hagan, 2009)-with more staff, financial, and ICT resources than Directorate-General of Resources of National Defence. These procurement cells would work as an extensive archive of integrated knowledge, with operational, logistical, legal, and financial knowledge of weapons systems. Currently, PAF already use technological and integrated management platforms such as the Enterprise Project Management and the Integrated Management System of National Defence. Together, they serve as managing tools for assembling capabilities and consult financial execution rates of these capabilities' investment projects (E7). Despite these platforms, the Branches also use different non-integrated information systems to manage everyday activities (e.g., managing items, weapon systems and equipment, and sustaining military capabilities). Besides, there are no cross and standard tools for communicating and organizing work (e.g., document management and project management software), which can hinder PAF's governance function (E7). To overcome these vulnerabilities, E7 suggests integrating information, communication, and work management systems, arguing that this would improve PAF's governance and promote efficiency gains. Therefore, the Branches must interact more frequently, investing in mechanisms that allow the intervention and the cross and strategic communication of all entities at the different levels (political, strategic, and operational) (E5; E7; E15).
E13 claims that the scarcity of human and financial resources felt is an incentive for PAF's integration. Indeed, several interviewees claim that sharing resources may promote efficiency and efficacy gains, which is why they recommend the reinforcement of joint and integrated capabilities and processes among the Branches (E3; E4; E7; E8; E10; E11; E15; Q2; Q4; Q5). Currently, there are already some structures that allow the sharing of resources among the Branches, namely the Armed Forces' Hospital, the Military University Institute, the Armed Forces Institute for Social Care, the Integrated Management System of National Defence, and the Cyber Defence Centre (E1; E4; E7; E13; E14; E15). However, due to each Branch's "genetics", E14 admits that, although there are joint capabilities among the Branches, he claims that there is no sharing of resources between them in the operational area. Here, the Branches only support the one that has the Command and Control and provision logistics in each mission.
On one hand, there are currently several factors that promote resources and skills' sharing between the Branches, namely: the growing number of joint and combined operations within international organizations to which Portugal belongs; the political will to invest in a military structure based on joint capabilities and commands and values' sharing; policies standardization and motivation to cooperate provided by the joint education given at Military University Institute; the integrated management of a single financing instrument (the Military Programming Law); and the existence of centralized management platforms (the Enterprise Project Management and the Integrated Management System of National Defence) that allow financial supervision in AFGS and Central Organs and Services of the Ministry for National Defence (E7; E13; E15). On the other hand, the factors that most hinder the Branches' sharing are: the rivalry that still exists among them and that leads them to isolate themselves (E9); the robust hierarchical and bureaucratic component of the PAF's governance model, encouraging their segmentation (E7; E8); and people who delay integration policies because they disagree with them (E14). Hence several participants stress the need for the Branches to collaborate more with each other, sharing more strengths, resources, and know-how, and create more joint and integrated processes (E7; E10; E11; E15; Q2; Q5). Besides, PAF must also integrate strategies and share resources with civil entities to avoid ad hoc planning of activities and resources (E8; E15) and improve its efficiency (E13). These suggestions are in line with Spiegeleire (2011) andTagarev's (2009) concerns about taking a systemic approach in the security sector to create synergies. As Madeira (2009) has already suggested, E4 and E8 claim that Portugal needs to reinforce its self-interaction by deepening the relationships between its several ministries (E4; E8). In short, the Branches should integrate doctrines, training, and procedures and combine resources to acquire greater efficacy and efficiency for the National Defence and the country as a whole (E13). This integration is only possible with the combination of efforts and wills at the political, strategic, and operational level (E7).

System context
4.1.4.1. Policy frameworks. The constitution, laws, and other national regulations reinforce the hierarchical feature of PAF's governance and planning models (E10; E11; E13). This hierarchy is essential in the military community (E5; E10; E11; E13; E14; E15), as it would become dysfunctional without it (E10). This hierarchy contributes to an effective top-down governance (E5), providing job roles' and responsibility levels' definitions for the MSP key players (E7). It also promotes order in priorities' establishment and decision-making processes (E10). For this reason, E5 suggests reinforcing the powers of the AFGS to streamline and coordinate the military organization, increasing governance's strength and efficiency. However, this hierarchy also encourages the Branches' detachment (E7), challenging cross mechanisms' implementation, collaboration with each other, and decentralized decision-making (E7; E8). Given the speed, unpredictability, and volatility with which the geopolitical environment can change (2020 Ministerial Directive for Military Defence Planning) and the Portuguese Democratic State's constitutional limitation, E10 recommends the readjustment of its hierarchical structures, processes, communication platforms, and collaboration skills.
Politics plays a significant role in this strategic planning model, influencing the strategy to be followed, the national ambition level, and the Portuguese's alignment with the international organizations they belong to (E1; E3; E4; E7; E9; E10; E11; E14). Thus, it is up to the political bodies to define the MSP's political guidelines (E1), outline strategic priorities, and make decisions about equipment to be purchased (weighted by the opinions of each Branch) (E3; E6). Therefore, as Young (2019) concluded, the military does not have discretionary power in strategic decisionmaking. Despite this, during Military Programming Law's management by the Directorate-General of Resources of National Defence, E7 claims that they try to conciliate the political guidelines with the Branches' needs. Likewise, E7 and E12 believe that it is necessary to bring political officers and military leaders together and contribute to an informed political authority.
Besides this model's hierarchy, its bureaucracy is also seen as a core factor of PAF's governance model, guiding the Branches' strategic planning and working as a compliance mechanism (E1; E3; E4; E5; E10; E13; E14; E15). In fact, the military Branches only create Joint Forces Investment Proposals if the upstream strategic documents-the 2014 Military Strategic Concept and the 2014 Military Forces Arrangement (General Staff Council, 2014; The Presidency of the Council of Ministers, 2013b)-have already specified joint capabilities (E1). Furthermore, E5 argues that although the development of trusting relationships is important, it can only happen as is strictly necessary and under the defined bureaucracy. Also, some interviewees admit that it is important to have core documents (E10; E15) to guide the planning process (E3). These documents guarantee that everything is done as it is written (E15), promoting the processes' clarification and easiness (E4). However, some interviewees highlight the lack of some standardized guiding documents across MSP's key players-e.g., regarding the management of projects and processes that bring several Branches into play (E7)-and the outdating of existing manuals-particularly the 2013 Strategic Concept for National Defence and the 2014 Military Strategic Concept (General Staff Council, 2014; The Presidency of the Council of Ministers, 2013b; E4; E7; E8; E13).
Another gap in Portugal's public administration is the practice of a "preventive" bureaucracy, which focuses too much on ensuring that the processes contain all guarantees, licenses, and verifications before proceeding (E4). This practice turns hierarchical and bureaucratic circuits too "heavy" and slow (E2; E5; E13), creating a high number of steps that block the processes immensely, inhibit several initiatives (E4), and do not have enough flexibility to guarantee the missions' accomplishment in a timely and efficient manner (E5; E13). Therefore, despite the increased concern with efficiency, sometimes both bureaucracy and budget execution and control rules make the legality principle override this concern (E10; E13). Along with the "preventive" bureaucracy, MSP's several levels of decision where the process can go before reaching the Minister (E7)from the Branches, passing through the Armed Forces General Staff, until the Central Organs and Services of the Ministry for National Defence-also contribute to this legislative "embarrassment" (E10). Thus, when there finally exists financial feasibility to proceed with the processes, it is too late and the year is over (E4), restricting PAF's ability to respond to its national threats (E11). To lessen bureaucracy's impacts on the processes' delay, some interviewees suggest investing in a cross procedural evaluation, sharing good practices, establishing a lessons-learned mechanism, and developing trusting relationships that allow processes' streamline through informal channels (E4; E6; E7; E10; E11; E15). Finally, E4 advocates using a "sequential" bureaucracy to endorse the importance of supervision, verification, control, inspection, and accountability mechanisms, which end up not being used in the current model due to the excessive number of guarantees granted a priori to MSP's processes.
4.1.4.2. Organizational culture. The three Branches' shared military genesis promotes values and resources' sharing between them (E10) since their cultural "DNA" brings them closer (E5). However, some interviewees believe that it is essential not to forget that the nature of each Branch's military capabilities differentiates them (e.g., land, air, sea) (E10). These differences are called by some interviewees the Branches' "genetics" (E1; E10; E14), which seems to be related to the idea of organizational culture as proposed by Denzau and North (2000). For some interviewees, these singularities are not only impossible to overcome (E3) but are also conflicting with the idea of improving PAF's efficacy and efficiency since there is a need for each Branch to have redundant resources to fulfil its specific missions (E10). For this reason, E1, E8, and E10 argue that the Branches should seek efficiency gains through the integration of non-operational areas such as logistics, financial and information technology. Regarding the MSP, E3 considers that the Branches do not have to understand each other since they only have to identify their needs. The politicians are the ones who make the decisions and are held accountable for them before the Portuguese society (E6; E11). E3 adds that he does not consider it necessary to rank the Branches' priorities since, in a scenario where the highest priority projects belong to the same Branch, there would not be enough funding to assemble other Branches' capabilities. E12 describes previous opinions as mental barriers of the prevailing military "parochialism", exposing an organizational culture that hinders Branches' integration. In fact, E13 states that the differences in organizational cultures are one of the main hampering factors of the Branches' integration, setting them apart. For this reason, several interviewees consider that these "barriers" urge a change in the Branches' organizational cultures (E13) by eradicating stigmas (E10), changing mentalities (E14), eliminating the Branches' corporate spirit (Q3) and phasing out what E12 describes as the unreasonable search for alleged, but false, corporate superiority. Thus, E14 alerts that Branche's need to start looking at the "forest" and not just for the "tree," focusing on the capability-creation process as a common whole and not just centred on their processes. During this process, it is also essential that PAF's leaders strive to preserve what is specific and unique in each of the institutions. This rewiring will make it possible to eliminate corporate superiority and reinforce a truly unbiassed purpose for serving Portugal and the Portuguese society in the best way possible (E12). E12 concludes that once the Branches overcome those mental barriers, they will notice that the sharing difficulties between them do not exist. E13 summarizes these measures by stating that more than defending a uniform, it is necessary to defend the country and proclaim that Branches are stronger together (E13).

Discussion of theoretical propositions
PAF's governance model has features from several types of government models. First, this governance model's prevailing hierarchy and bureaucracy and the concentration of discretionary power in the political sector seem to suggest a predominantly classic model of PA (Osborne, 2006). Second, the growing concern with efficiency gains and other performance indicators suggests NPM's traits (Hood, 1991;Osborne, 2006). Finally, the importance given by interviewees and MSP's essential documents, to the expansion of Branches' collaboration, to the development of trusting relationships, and to the establishment of network relationships with other ministries and civil entities suggest an intention to bring PAF's governance model closer to NPG and Collaborative Governance models (Emerson et al., 2012;Lindsay et al., 2014;Osborne et al., 2013).
However, the results suggest that bureaucracy and budget execution and control rules often place the principle of legality (PA) over efficiency (NPM). Also, the Branches' apparent corporatism, the reduced inter-ministerial interaction, and the vulnerabilities pointed out concerning existing joint capabilities, sharing of resources, knowledge, and skills, network communications, and process integration show a model that is still far from being considered NPG or Collaborative Governance. Like the Tuscany region, there seems to be a discrepancy between the bureaucratic intention to increase Branches' integration (NPG and Collaborative Governance) for efficiency reasons (NPM), and the hierarchy, bureaucracy and political influence predominance (PA) found in practice in the governance model (Iacovino et al., 2017;Osborne, 2006). Thus, even though MSP's essential documents show clear signs of NPM, NPG and Collaborative Governance influences, in practice, PAF's governance model continues to be a predominantly classic PA model.
Consequently, although the present governance model has several potentialities, the results also expose several features that threaten the attainment of PAF's needs. First, the corporatist hierarchy and organizational culture encourage Branches' detachment, hampering their integration. Second, excessive bureaucracy slows down processes, decreases communication fluidity, and places the legality principle over efficiency, threatening PAF's ability to respond effectively to its missions. Third, the lack of common methodologies and tools among the Branches that support joint decision-making is Directorate-General of Resources of National Defence's most significant problem during the integration of their MSP's investment proposals. Finally, Branches' competition for available resources hinders the establishment of a collaborative environment within PAF, threatening their efficacy, efficiency, and effectiveness. Therefore, although the results suggest several potentialities of PAF's governance model regarding its strategic planning, several vulnerabilities seem to hinder Branches' integration and, consequently, more significant efficiency gains. We extend Emerson et al. (2012)'s Collaborative Governance model, introducing several propositions to its obstacles and outcomes: Proposition 1: The participants' culture in a collaborative relationship influences their collaboration.
Regarding the System Context, results suggest that the culture related to the Branches' interaction is the element that influences them the most. In fact, even though their organizational culture brings them closer together-which is also supported by the survey's response to the statement that measured Close Values (average agreement score in 6-points likert scale survey [µ] = 4.85)-the differences in their organizational cultures hinder their integration (especially at the operational level) and the sharing of their resources among them. For this reason, as proposed by Commander Moreira in Pires (1999) work, several interviewees recommend facing the Branches' corporatist mentality by convincing the whole military organization that their collaboration efforts will strengthen them. These results support the idea that decision-makers' mental model shapes how they think, judge, and act, as advocated by Kooiman and Jentoft (2009). What has been exposed suggests that the Branches' individualistic cultural mentalities negatively influence their collaboration and resource sharing. These results seem to validate the impact that non-rational factors (such as organizational culture) can have on public entities' governance, supporting the conclusions of Denzau and North (2000), Kooiman and Jentoft (2009), Leach and Sabatier (2005), and Weare et al. (2014).

Proposition 2:
The integrated use of Information and Communication Technologies (ICT) in a collaborative network allows the improvement of governance models' responsiveness.
Within the Capacity for Joint Action, interviewees' opinions suggest a direct connection between Branches' knowledge sharing and the governmental system's efficacy and efficiency, which matches the thoughts of Collaborative Governance literature's main authors. However, the slowness of hierarchical and bureaucratic circuits hinders their sharing ability. To overcome this vulnerability, the interviewees suggest integrating a more significant number of information systems since there is still room for improving government system's efficiency and responsiveness. The direct association found between ICT's integration and the improvement of PAF's responsiveness supports Meijer and Bolívar's arguments (2015, as cited in Šiugždinienė et al., 2017).
Regarding Principled Engagement, despite MSP's several interaction platforms, this connection only happens concerning joint capabilities. The autonomy to outline strategies and define priorities in isolation seems to reinforce organizational individualism, revealing a structure with an insufficient strategic commitment between the Branches. Also, adopting a unique methodology for defining priorities fills in Delgado's (2017) identified gap. However, the lack of some cross standard methodologies that could simplify joint decision-making leads some interviewees to suggest its adoption to promote Branches' integration. These ideas validate the direct correlation between the standardization of Branches' strategic management instruments and their integration capacity, as pointed out by Delgado (2017).
Proposition 3: Increased trust, mutual understanding, internal legitimacy, or commitment promotes collaboration among military Branches.
According to some interviewees, all Branches work for the same purpose, which is supported by the positive survey results measuring Shared Commitment (µ = 4.04). However, the existing rivalry-backed by the survey data measuring Competition (µ = 5.07)-makes it challenging to integrate Branches' strategies. This challenge is even more clear if we consider the reduced average obtained (µ = 2.7) in the questionnaire's affirmation measuring Branches' willingness to collaborate to achieve strategic objectives even if it compromises their individual objectives. This idea suggests that the Branches place their individual objectives above their strategic ones, revealing an insufficient collective commitment that supports the existence of a corporatist spirit within PAF's organizational culture. The results seem to advocate that the development of trusting relationships among the Branches and their commitment to follow a shared path are directly related to their collaborative ability, which matches Collaborative Governance literature's beliefs.
Regarding Shared Motivation, the development of trusting relationships is considered indispensable during collaboration. Trust allows MSP's processes to be informally streamlined-as also verified by Šiugždinienė et al. (2017) in the Lithuanian reality-facilitates coordination and accelerates decision-making, which promotes efficiency gains. These results confirm the association pointed out by Šiugždinienė et al. (2017) between the development of interpersonal relationships and the increased success of institutions' collaboration. Accordingly, survey results included the Trust in the factor that most explained Branches' motivation to collaborate. One aspect that contributes to developing trusting relationships is the joint education given at Military University Institute to the three Branches' military personnel. This statement endorses Leach and Sabatier (2005) idea that institutions' cultural norms (reinforced during joint education) may improve trust levels among them, encouraging the creation of what Denzau and North (2000, p. 24) called a "common mental model".

Proposition 4: Greater integration of the Branches improves PAF's resource allocation efficiency.
Military Programming Law's insufficient funding increases gaps in existing capabilities, challenges the assembly of new capabilities, and keeps military capabilities' operational qualifications only at the minimum level required, threatening PAF's ability to respond to its designated missions. This correlation between Military Programming Law's underfunding and its impact on PAF's preparation level supports Mata (2013) and Young's (2019) opinions. This demonstration shows that, apart from efficiency and efficacy needs, respondents and MSP's documents are also concerned with PAF's ability to respond to its assigned missions, that is, its effectiveness. Also, as all Branches need to re-equip themselves, capabilities' underfunding encourages them to compete with each other for the available resources-which is supported both by the interviewees and the high agreement level in the survey's measure of Competition (µ = 5.07).
This association corroborates Delgado's (2017) belief that resource scarcity encourages the Branches to compete with one other for their allocation. To face this financial restriction, the interviewees claim that PAF must urgently seek efficiency gains.
Thus, in order to respond to the PAF's needs of efficacy, efficiency, and effectiveness, the most recent strategic documents and several participants in this study, suggest the increased integration of the Branches as the leading solution, matching what was already recommended by Delgado (2017), Martins (2015), and Mata (2013). This integration involves strengthening their collaboration mechanisms and increasing their resources, skills, and knowledge sharing, focusing on assembling a more significant number of joint and integrated capabilities, processes, structures, and commands and integrating doctrines, procedures, and ICT systems. The questionnaire's results show that all statements belonging to the Cooperation and Coordination factor got a mean higher than 5, which shows respondents' will to increase Branches' collaboration. These statements measure agreement degrees concerning some measures of the Plan's Grand Options for 2020-2023 regarding the increase of joint capabilities and equipment and the reinforcement of sharing mechanisms. According to the interviewees, these mechanisms include developing cross standard methodologies, implementing cross ICT systems, updating doctrine, and centralizing support areas. These results and opinions can indicate a possible future evolution of the governance system, given the recognized need to deepen interinstitutional sharing and interaction mechanisms.

Conclusion
This study analyses PAF's governance suitability given its need to get higher efficiency while ensuring an adequate ability to respond effectively to its missions. Based on a mixed method (interviews and survey) analysis of PAF's model, it is possible to conclude that it is not sufficiently adequate to meet PAF's integrative needs.
Thus, even though growing concerns about the Branches' efficiency and integration indicate NPM, NPG, and Collaborative Governance influences, the prevailing hierarchy, bureaucracy, and political influence in PAF's governance model suggest that it is closer to a PA model. However, MSP's documents and this study's participants seem to recognize the importance of adopting integrative and collaborative characteristics of NPG and Collaborative Governance, which may indicate the direction to which PAF's governance system is evolving. Despite the several potentialities found, some of PAF's governance characteristics seem to compromise Branches' integration, namely the predominant hierarchy and bureaucracy, the prevailing corporate mentality, the lack of some cross standard analytical methodologies to support joint decision-making, and their competitive behaviour for available resources. Hence, this study's participants encourage a greater integration among the Branches by creating a higher number of joint capabilities and processes, standardizing more methodologies and doctrines, increasing the communication between them, and sharing more resources and knowledge.
This study highlights the primary obstacles that hamper collaboration's intended outcomes in the military division of the Public Sector. The identified issues help us draw political and managerial implications for practitioners. At the political level, we can conclude that regulating collaboration dynamics is not enough. Public administrators also need to come up with ways of implementing what has been written. At the managerial level, the results suggest that a competitive mentality among leaders may result in sub-optimal investment decisions which encourages managers to improve collaborative relations. For scholars, this study also contributes as empirical evidence to support the relevance of governance models that encourage collaboration in interdependent networks and endorse the impact that non-rational factors (such as organizational culture) can have in public entities' governance.