Agency problems in the public sector: the role of mediators between central administration of city hall and executive bodies

Abstract

This article aims to identify the effects of mid-level bureaucrat’s allocation in the intermediation of relations between the main authority of a city’s executive branch and the city’s secretariat, considering the principal-agent relationship in public administration. The study interviewed analysts responsible for goal and project management who mediate the agency relationship between the Chief of Staff and the executive bodies implementing projects put forward by Rio de Janeiro’s City Hall, in order to examine the effects of the analysts’ intermediation in terms of reducing agency problems. The agency problems referred to in the article are due to public administration’s hierarchical structure, resulting from the differences in motivation and goals of the actors, information asymmetries, risk propensity and different planning horizon. Results show this intermediation is positive to reduce agency problems, impacting in terms of difference in motivation and information asymmetries.

Keywords:
agency problem; city management; information asymmetry; public administration; mid-level bureaucrats.

Resumo

Este trabalho visa identificar o efeito da alocação de indivíduos, definidos na literatura como burocratas de médio escalão, na intermediação das relações entre a direção superior do Poder Executivo e as secretarias de Estado na relação principal-agente no âmbito da administração pública. Com base em entrevistas em profundidade com analistas de gerenciamento de projetos e metas, intermediadores da relação de agência entre a Casa Civil e unidades executoras de projetos da Prefeitura do Rio de Janeiro, foram examinados os efeitos de sua intermediação sobre os problemas de agência na estrutura hierárquica da administração pública decorrentes de diferenças de motivações e objetivos entre as partes, assimetrias de informação, propensões ao risco e distintos horizontes de planejamento. Os resultados mostraram os benefícios dessa intermediação para reduzir os problemas de agência, com efeitos sobre as diferenças de motivações e assimetrias de informação.

Palavras-chave:
problema de agência; gestão municipal; assimetria de informações; administração pública, burocrata de médio escalão.

Resumen

Este estudio tiene como objetivo identificar el efecto de intermediación desempeñada por funcionarios especializados, los burócratas del escalón intermedio, en la relación entre la dirección de la rama ejecutiva y las secretarías de Estado, considerando la relación principal-agente en la administración pública. Basado en entrevistas con analistas de proyectos y objetivos de gestión, los intermediarios de la relación entre la Casa Civil y unidades ejecutoras de proyectos de la municipalidad de Rio de Janeiro, se examinaron los efectos de su intermediación en la reducción de las diferencias en las motivaciones y objetivos entre las partes, las asimetrías de información y diferentes propensiones a riesgo y horizontes de planificación, con evidencias de efectos positivos sobre los dos primeros.

Palabras clave:
problema de agencia; gestión municipal; asimetría de información; administración pública; burócratas del escalón intermedio.

1. Introduction

This study analyzes the public sector as an integrated set of secretariats, departments and a diversity of public organizations that carry out different contractual arrangements, where one party delegates activities (principal) and another one executes them (agent). This point of view based on contracts, portrays problems and costs involved in these relations, which are discussed by the agency theory and other theoretical frameworks originated in the New Institutional Economics (NIE).

These theories consider that agency relations are omnipresent in society, in the economy, and in various sciences (Eisenhardt, 1989; Przeworski, 2003). Schillemans (2013) adds that, in the last decades, a wide range of reforms have been modeled, implicitly or explicitly, around the agency theory in several countries. For Przeworski (2003), the issue of the correct configuration of contracts between principal and agent is crucial because the performance of the economy, in a holistic way, depends on the design of the institutions regulating this relationship.

Agency theory deals with the so called ‘agency problem’, i.e., problems related to the separation between ownership and management, and it has represented the dominant theoretical perspective applied to studies on corporate governance (Daily, Dalton and Cannella, 2003). Agency theory has also been used in countless other sectors, especially in the public sector and its internal and external relations, such as in outsourcing processes and in public-private partnerships. The theory’s basic premise is that if both parties in a principal-agent relationship seek to maximize utility function, the agent will not always act in the best interest of the principal. The principal may limit interests that are divergent from their own, by establishing appropriate incentives for the agent and incurring monitoring costs in order to limit deviant activities of the agent (Jensen and Meckling, 1976).

In general, the principal-agent problem refers to issues such as information and incentive schemes. The agency problems can originate not only in the differences of motivation and goals between principal and agent, but also in information asymmetry, parties’ risk preferences and planning horizon (Eisenhardt, 1989).

When it comes to the state, broadly speaking, or the state apparatus more specifically, there are several agency relations, as discussed by Przeworski (2003), exemplified by the tacit or explicit contracts between citizens and politicians, the state and the economic system, and politicians and bureaucracy. The latter is the relationship at the base of this article.

Although more obvious in indirect administration, agency problems can also become serious in direct administration, particularly in political contexts characterized by coalition mechanisms. As defined by Abranches (1988:21), the coalitional presidentialism that characterizes the Brazilian model, “in addition to combining proportionality, multiparty and ‘imperial presidentialism’, organizes the executive branch on the basis of large coalitions”. This produces a high fragmentation of the ministerial control by the parties, weakening the command in the executive branch and extending the conditions for rent-seeking by the coalition parties or by the bureaucracy itself.

This research investigates the recent experience of the City Hall of Rio de Janeiro in dealing with these problems, seeking to understand how the introduction of an individual (a public servant) as intermediary of the relationship between the central power and administration bodies, influences the reduction of agency problems between the parties. The study analyzed Results Agreements signed annually between the Project Management Office of the Municipal Chief of Staff Secretary, characterized as ‘principal’, and the departments and entities of the direct and indirect municipal administration related to city’s secretariats, characterized as ‘agents’. Intermediating this principal-agent relationship is the Projects and Goals Management Analyst (PGMA), a position established by Law 5.595/2013 and whose main tasks are related to the monitoring of the routines of the city hall’s departments and entities, as well as the progress of the goals established at the Results Agreement.

Thus, the main objective of this article is to identify how the PGMA affects the agency problems in the relationship between the Projects Office and the departments and entities (executive bodies) implementing strategic projects and goals, based on the perception of these actors regarding the result of their own actions in minimizing these problems.

This study also seeks to contribute to the literature on the Mid-Level Bureaucrat (MLB) (Cavalcante and Lotta, 2015; Oliveira and Abrucio, 2011; Luz, Gobbo and Lavarda, 2013), an actor considered to be essential for the management of integrative processes (Lavarda, Canet-Giner and Peris-Bonet, 2010), such as those studied here, and for organizational performance itself (Currie and Procter, 2005).

Understanding the role of the MLB may be an attempt to relate the performance of politicians and top management, where policies are constructed, to the practice of street level bureaucracy, where policies are implemented. It is during the implementation that the policies receive new interpretation based on the “occupational and professional ideology” of bureaucrats, and are transformed based on the daily operation and on contact with citizens (Lipsky, 2010). In addition, to understand the role of the mid-level bureaucracy is an attempt to revisit its importance in response to the criticism received from scholars associated with the New Public Management, who adopt principles of efficiency in order to reduce hierarchical levels and eliminate intermediate functions (Chen, Bermand and Wang, 2017).

2. The agency problem in the public sector

The economic aspect of organizations has been the subject of many academic studies, starting with the seminal article by Coase (1937), who stimulated the diffusion of the concepts of the NIE. This theoretical framework is based on the study of the organization (firm) in terms of contracts (formal or informal), signed between parties (Nilakant and Rao, 1994). According to Fama (1980) and Jensen and Meckling (1976), organizations can be understood as an environment in which a complex network of contracts between principals and agents is observed, a nexus of contracts in which ownership becomes irrelevant.

Establishing contracts where a certain level of authority is delegated to make decisions about implementing an activity by one party on behalf of another is the foundation of the agency theory. The contract is the instrument through which the principal charges the agent to act on his behalf, and this relationship established in contract is called agency relationship (Schillemans, 2013). Problems and costs to minimize them, surge because of the agency relationship, both in private administration and in public administration (Attila, 2012; Yesilkagit, 2004).

Agency theory falls within the field of New Institutional Economics (NIE), which encompasses a set of theories that study organizations based on complex forms of contracting, and understand that economic relations may occur outside the pattern of traditional market transactions and this is not a reason to say they are market failures (Nilakant and Rao, 1994). Ménard and Shirley (2014) argue, however, that the focus of NIE is quite broad, since it examines, at the micro-analytical level, the organizations and their impacts on public policies. For these authors, NIE promotes dynamic rather than static explanations for economic developments, as well as being receptive to interdisciplinary approaches, and providing space for case studies and other non-mathematical methodologies.

Jensen and Meckling (1976) define the agency relationship as a contract in which one party (the agent) is charged with performing a particular activity on behalf of another party (the principal). The theory emphasizes the search for the contract model that best governs the agreement between the principal and the agent — both considered to be acting according to rational behavior and motivated by self-interests — and allows minimizing the costs of this contract (Downs, 1964; Eisenhardt, 1989).

The biggest challenge of the principal is to ensure that the agent will fulfill their interest. The advantages of a contract based on specific behavior on the part of the agent should be compared with those of a contract aimed at delivering predefined results. In a context where all information is accessible and the principal is able to observe the agent’s activities, Eisenhardt (1989) argues that the most efficient contract model is based on the behavior of the agent; otherwise, the contract must be based on the delivery of results.

Eisenhardt (1989) argues that the contract is only an illustrative metaphor of the principal- agent relationship. This argument is corroborated by Przeworski (2003), who says that the agreement between the parties may be explicit or implicit and the agency relationship is omnipresent in society, manifesting itself between managers and employees, owners and managers, and even in the public sector, among citizens and politicians, politicians and bureaucracy. Buchanan (1975) adopts the view that the state, represented by democratically elected politicians, is the agent of citizens’ demands, who hold the title of ‘principal’. According to the authors, it is on behalf of the citizens that politicians and bureaucrats make collective decisions and establish public policies.

The good performance of organizations, governments and the economy, in general, is directly related to adequate structure of agent-principal relationships (Przeworski, 2003). Improving the configuration of contracts between principal and agent means improving assumptions regarding information, people and organizations, as described by Eisenhardt (1989).

The agent has information about requirements and results of their activities that are unknown and often unobservable by the principal, which constitute a source of several agency problems (Attila, 2012). Thus, it is an assumption of the agency theory that there is information asymmetry within organizations, which creates conditions for systematic problems regarding the credibility of information exchanged between principal and agent.

For Schillemans (2013), the core of the theory is precisely the incompatibility of interests between self-centered agents and principals — individuals oriented to serve their interests and maximize their own utility function. The principal is more risk-tolerant than the agent, since they are able to diversify their investment portfolio, making them indifferent to unsystematic risk (the risk connected to a specific project or investment). The agent is adverse to risks as they have difficulty to diversify their portfolio as their remuneration and status are linked to their performance regarding a specific contract (Lane, Cannella and Lubatkin, 1998). In addition, due to the limited rationality of individuals — characterized by the limited access to information, cognitive and time limitations, complexity (costly decision-making process, preventing the capacity to anticipate), and by the risk and uncertainty — the possibilities of managerial opportunism are multiplied. Examples of this opportunism are the manipulation of information asymmetry in order to take over the flow of profits (or benefits), and the problems of adverse selection and moral hazard (Fiani, 2002).

The problems arising from an agency relationship impose (agency) costs for the principal. The first of these problems points to the existence of divergent goals between agent and principal, based on the assumption that both parties seek to maximize their primary utility function from the contractual relationship; therefore, there are good reasons to believe that the agent will not always act based on the interests of the principal (Lane, Cannella and Lubatkin, 1998). The asymmetries of information between the principal and the agent create conditions for the latter to report incomplete or biased information, voluntarily or not. According to Williamson (1987), opportunistic action is always malicious and involves lying, stealing, cheating, and subtle forms of fraud. This opportunism, in agency theory, manifests itself as both ex-ante and ex-post in relation to the time the contract is valid, resulting in problems of adverse selection and moral hazard, respectively.

Different risk preferences become problematic when the principal and agent need to share the same risks under the contract (Nilakant and Rao, 1994), although they have different predispositions to systematic and non-systematic risk. Finally, the problem of agency linked to the incompatibility of planning horizons is due to the transitory relationship that the agent has with the organization, while for the principal, in general, the link presents a more permanent character. Guimarães dos Santos (2015) argues that often the agent needs to allocate efforts in actions whose effects occur in the short or long term, and that their preference is usually for the actions that bring results in the short term, characterizing what he calls ‘behavior of managerial myopia’.

Jensen and Meckling (1976), argue that there are three types of agency costs: monitoring costs, bonding costs and residual losses. The role of monitoring costs is not exhausted by the direct measurement and observation of the agent. It includes the principal’s efforts to control the agent’s actions through budget constraints, compensation policies, operating rules, audits, hiring and training for senior managers.

In the public sector, agency relationships are often dealt with from the standpoint of public choice theory, which considers that it is up to the state, represented by democratically chosen politicians, to promote public policies that serve the interests of civil society (Lemieux, 2015). Silva (1996) asserts that it is possible to identify, in this context, an agency relationship where the voter places himself as principal and the State as agent. There is no reason to believe that the behavior of individuals regarding public decisions will be fundamentally different from the behavior observed in particular choices. In addition, considering that individuals seek to maximize the satisfaction of their own interests, it is possible to say that in democratic regimes, government policies will reflect irrationalities of ordinary people (Buchanan and Tullock, 1962; Viscusi and Gayer, 2015).

From the point of view of public choice, the state is not a dominant entity, but rather the pro­duct of political exchange, which exists in order to serve society (Buchanan, 1975). This reasoning resembles the concept of agency relationship of Jensen and Meckling (1976), considering that civil society can be seen as the principal and the State as the agent. Decisions taken by bureaucrats consist of public choices, which deal with the definition of which public goods are to be produced and in what quantities (Silva, 1996).

Using agency theory to study relations within the state, Przeworski (2003) analyzes three types of agency relationships that must be considered: between the state and private economic agents; between citizens and politicians; and between politicians and the public apparatus, that is, bureaucracy. This latter relationship inevitably implies the delegation — from the state to the bureaucracy — of the services that the government must render to the citizens, taking into account that delegation of an activity is the basic condition for the occurrence of the agency relationship (Eisenhardt, 1989). For Yesilkagit (2004) a fundamental tension then occurs between delegation of tasks, political control and accountability to civil society.

Gains, John and Stoker (2008) claim that there is a constant battle between politicians (principal) and bureaucrats (agent) supported by the fact that agents have information and technical knowledge about the activities they carry out, which are not easily shared with the principal. Politicians, however, are able to increase or reduce the discretion of the bureaucracy through the definition of rules. In addition, bureaucrats, despite their interest in maintaining their discretion and autonomy, have no alternative but to work with the public apparatus the way it is legally designed.

Przeworski (2003) points out that bureaucrats do not necessarily have the same goals as politicians or citizens, which is one of the agency problems highlighted by several authors: the misalignment of goals between principal and agent. This, according to Gains, John and Stoker (2008), may cause the bureaucrat to adopt public policies of their own preference over those of politicians interest.

Yesilkagit (2004) suggests two modalities of control that politicians can adopt regarding bureaucrats’ actions, one being ex-ante and one ex-post. As principals, politicians can exercise ex-ante control over bureaucracy by defining the laws, rules, and design of the organization, limiting the bureaucrats’ options of action and seeking to ensure that possible alternatives converge with their interests. Ex-post control can take place through sanctions - budget cuts, changing managers or review of the organization’s tasks and assignments — or monitoring — ombudsman’s offices, inquiries, investigations, and media and civil society oversight.

A problematic issue, emphasized by Bevan and Hood (2006), in the relationship between the high political management of public organizations and managers of public policies is related to gaming in the pursuit of fixed goals. The authors define gaming as subversive actions that lead to the achievement of the goal, in purely numerical terms, but not to the effective fulfillment of the aim to which the goal was proposed. The difficulty for politicians to monitor the efforts made by the bureaucracy is related to moral hazard and the agency conflict resulting from the asymmetries of information between agent and principal after establishing the contract (Fiani, 2002; Sung, 2005).

It should be noted, however, that the modeling of relationships between the political system and the bureaucracy could become more complex when bureaucracy is analyzed in an inhomogeneous way (separating the decision-making structures of the political elite at the top of the command chain; the street level bureaucrats responsible for policy implementation; and the intermediary bureaucrat, MLB). MLB is an important element in the activation of informational flow, minimization of agency problems between the central body and the secretariats, and in the implementation of the projects and strategies (Barton and Ambrosini, 2013).

3. Methodology

In order to evaluate the effect of introducing an individual as a mediator of the relationship between central power and executive bodies of public administration in order to reduce agency problems, a case study was conducted at the City Hall of Rio de Janeiro, observing the Projects and Goals Management Analyst (PGMA), public servant responsible for this mediation.

The research first conducted document analysis based on materials presented and minutes registering meetings conducted to structure and monitor the Results Agreements signed between the Project Office and the executive bodies implementing projects, referring to the years 2014, 2015 and 2016. The analysis studied the documents’ final versions, as well as their first drafts, in order to identify information and other perspectives not presented or discussed between representatives of the Project Office and the executive bodies.

Interviews with PGMA were conducted, using a script (presented in the appendix), elaborated around the four sets of typical agency problems described in the literature: a) differences in motivation and goals between principal and agent; b) the occurrence of informational asymmetries; c) different risk propensities; and d) different planning horizons.

Data collection was performed through semi-structured interviews with those employed as PGMA, completing 10 interviews — which was the amount considered the saturation point (Thiry-Cherques, 2009). The literature on techniques used in qualitative research indicates that although identifying the saturation point encompasses a considerable level of subjectivity, this limit is reached in the twelfth interview and it is already possible to highlight the key concepts during the sixth interview (Charmaz, 2006; Guest et al., 2006; Mason, 2010). Charmaz (2006) adds that the appropriate amount of interviews depends on the scope of the study as well as on the complexity of the subject matter. The interviews for this research, observed saturation by the reduction of the percentage of new proposals or ideas on the topic throughout the conversations, which occurred more drastically from the seventh to the eighth interview. Thus, it was decided to continue with interviews until the tenth, when the main concepts of each conversation were listed and it was possible to determine the saturation point had been reached.

In order to select the 10 interviewees from the population of 50 PGMA, the research considered that an adequate sample should seek the representativeness of the phenomenon. Thus, the sample was gathered considering elements such as plurality of education backgrounds, different sectors of activity (contemplating end-areas/mean-areas), age, time of professional activity, and professional activity as public servant. This heterogeneity in the characteristics of the interviewees allowed a greater variety in the responses. It is important to highlight the fact that one of the authors of this study occupies the position of PGMA, which naturally brings facilities and limitations to data collection, requiring additional care in the analysis and discussion of the results. The material collected in the interviews was transcribed, organized into pre-defined formal structures, and codified in order to promote a more objective analysis of the four agency conflicts that constitute the focus of the research.

4. City Hall and the career of projects and goals management analyst (PGMA)

In 2009, the City Hall adopted efficient and result-oriented management practices and methods (PCRJ, 2012). The Municipal High Performance Management System (SMGAD) comprises the development of long-term vision; strategic plan; detailing, organization and control of projects and strategic goals; elaboration of results agreements; and follow-up of all these initiatives (Rio de Janeiro, 2013a). Strategic initiatives, projects and targets are broken down annually into Results Agreements, with a pre-agreed bonus for delivering the items settled in the agreements.

In this context, Law 5.595/2013 created the career of PGMA, considered an integral part of the SMGAD. This employee acts with departments and entities of the Municipal Administration, following the planning and implementation of projects and goals, reporting their status to the Projects Office and indicating, when appropriate, deviations and inaccuracies in the progress of the strategic initiatives. PGMA is responsible for updating information and offering progress reports on the strategic initiatives, subsidizing the follow-up of the Projects Office, which in turn reports to the Mayor. Based on literature about MLB, the PGMA can be seen as “a fundamental link between the top and the base, between decisions and rules and their practical applications, between the multiple actors and their understandings for the feasibility of joint actions to implement governmental initiatives” (Pires, 2015:179).

The hiring process through which 50 PGMA were contracted in 2013 was not restrictive regarding the education of the professionals. The process required the applicants to have superior education, independent the field of study. This shows that it was not important for the position to have in-depth knowledge on a particular topic, but it was intended to bring together generalist professionals with a good notion of project management and follow-up of goals and indicators, regardless of the department in which they would be allocated. The knowledge of the PGMA, therefore, is not the same, in terms of technical refinement, as employees of the executive bodies implementing policies. However, daily living with these employees and total immersion in the context of the executive body enables PGMA to make a much more credible reading of the reality at the implementation level than the individuals who work in the Projects Office.

Based on the Results Agreements, the Projects Office is considered the principal, and the executive bodies the agents. As in Przeworski’s (2003) view, this is an example of an agency relationship between politicians — represented by the Projects Office — and bureaucracy — represented by executive bodies of the public administration.

5. Results and analysis

The results of the interviews are grouped in the categories related to the agency problem, as identified in the theoretical framework. They show how the PGMA perceive their role and deal with conflicts in the relationship between the Projects Office and the executive bodies. The results are presented followed by an analysis based on the theoretical framework.

A. Differences in motivation and goals

Often, managers at the Projects Office and managers working at the executive bodies deal with a specific subject differently, resulting in activities not included in the strategic plan and in its unfolding in Results Agreement. When this happens, the commitment of the managers implementing the policy to the strategic goals and projects is negatively impacted. Some reports from the executive bodies showed interests significantly aligned with those of the Projects Office, but respondents point out that such alignment is temporary and vulnerable to factors outside the contractual relationship. In these situations, an initial alignment occurs when the Results Agreement of a department is drafted, when the contract is signed. However, when managers face parallel agendas or budget cuts, they end up distancing themselves from what had previously been agreed.

The degree of this convergence of goals varies among employees of the same executive body. As pointed out by respondent E6, “this identification is stronger for employees working close to the Secretary’s office. Lower-level employees often do not even know the strategic plan and have questions about the Results Agreement”. In addition, it was reported that the managers of the Projects Office are unaware of several technical aspects related to the operation of the executive bodies and therefore it is difficult to align motivation and goals between technical-operational bodies and the Projects Office.

In order to approximate the interests of the parties involved in the Results Agreement, the PGMA reported to seek that the executive bodies’ managers understand the goals of the Projects Office; at the same time, they take to the Projects Office the interests of the executive bodies’ managers. PGMA are in a position that allows them to put forward the point of view of both parties clarifying arguments or justifications and helping to establish a reasonable and balanced Results Agreement. As mentioned by respondent E7, “because they belong to the Projects Office, the PGMAs obviously know the objectives, and as they work at the level of implementation, they are sensitive to the interests of the executive department/entity. So, it is possible to make everything clear”.

Although the PGMA is an employee of the Projects Office, some have stated that they act as a representative of the executive body when in dialogue with the Projects Office. One PGMA mentioned, “the day-to-day work of the PGMA is to try to pass on to the Projects Office a little of the Secretariat’s vision regarding the strategic projects and initiatives” (E8). For the PGMA, the erosion in the relationship with the executive bodies’ employees can be harmful or even jeopardize the execution of their work.

This result corroborates the exercise of PGMA upwards relationship roles, defending alternative solutions and synthetizing information, as proposed by Floyd and Wooldridge (1992), as well as the mere supervision of the execution of the goals. Chen, Bermand and Wang (2017), in a survey with public servants in Taiwan, also corroborated the evidence of both roles as well as the predominance of the first in line with the manifestations of the respondents of this study.

The problem of incompatibility of goals between principal and agent was identified in the relationship between the Projects Office and the bodies implementing the goals and the strategic projects. Considering that the contract analyzed in this research was the annual Results Agreement, which includes the goals for each body, it is understood that the goals of the principal (Projects Office) are listed there. The contract, according to the agency theory, must be balanced in order to simultaneously maximize the principal’s wealth and the agent’s utility (Nilakant and Rao, 1994). When it comes to the environment of City Hall, it does not make sense to imagine that the Projects Office expects to obtain wealth through the Results Agreement.

The divergence between the interests of the principal and the agent is evidenced by the fact that not all the activities performed by the executive bodies are contemplated in the Results Agreement. One of the interviewees points out that “the current strategic plan only addresses a portion of the activities of the Secretariat. Thus, employees see their interests represented only in part”. The agent has, therefore, a series of interests that are not among those of the principal, in a potential misalignment. The PGMA allocated to the executive bodies can be perceived as an attempt of the Projects Office to closely monitor the implementation of the policies.

The results show that when managers’ and executive bodies’ goals are not aligned, the employees of the executive bodies’ are discouraged from fulfilling what has been agreed, ratifying the theoretical premise of self-centered individuals who seek to serve their own interests (Viscusi and Gayer, 2015). This research, however, identified cases where there is an alignment of interests between the parties, those are cases where the projects are contemplated in the strategic plan in force and consequently in the Results Agreement signed with the Projects Office.

As a mechanism for reducing the agency problem, the PGMA uses dialogue with both parties, not other incentive structures as proposed by Jensen and Meckling (1976). Because the analyst is physically working in the executive body, they are aware of the routine affairs of the agent and observe clearly the agent’s goals. In parallel, the PGMA is well positioned to expose to the agent the principal’s interests. The effort to mitigate the problem of different motivation and goals is evident in the words of respondent E7 mentioned above (“the PGMA, because they belong to the Projects Office, they obviously know the objectives, and as they work at the level of the implementation, they are sensitive to the interests of the executive department/entity. So, it is possible to make everything clear”) and E8 (“the day-to-day work of the PGMA is to try to pass on to the Projects Office a little of the Secretariat’s vision regarding the strategic projects and initiatives”). Obtaining the agent’s trust, the PGMA seeks to force an approximation of their vision to that of the principal, therefore reducing the problem of goal misalignment. In line with what Bresser-Pereira (2009) and Eisenhardt (1989) claim to be at the heart of the agency theory, the PGMA provides a more reasonable Results Agreement for both the Projects Office and the executive body.

However, the intermediary is not always able to maintain their neutrality throughout the process, since the PGMA is afraid to risk the trust gained from executive bodies’ managers. When this occurs, the performance of the PGMA does not promote any impact on the issue of goal difference between principal and agent.

B. Information asymmetry

Prior to the creation of the position of PGMA, Projects Office did not have the effective means to obtain information on the performance history and executive capacity of the executive bodies: “before the creation of the position of PGMA, the Project Office was not able to obtain virtually any information about the Secretariat” (E6). Currently, the PGMA identifies resources — in terms of budget, technology or personnel — that are wasted or underused. According to the interviewees, this means that the executive bodies are more able to deliver than what they claim to be. However, the PGMA does not have enough tools or technical knowledge about the operation of the body where they are allocated, and they are unable to properly assess the dimension of this unused capacity or to measure the impact of these underused resources in the overall result of the executive body operation.

Eventually there is “a strong culture of resistance from the Secretariat, and its staff hides the potential of their resources from the PGMA” (E8). In these situations, the PGMA can only have a superficial view of the true capabilities of the executive body. In cases where the PGMA has good interpersonal relationships with the staff, access to available processes and resources is facilitated.

For the elaboration of the strategic plan and of the goals to build the Results Agreement, the interviewees affirm to bring as much information as possible to the Projects Office so that “in the next Results Agreement the goals are not surreal, or far from the possibilities of the executive body” (E1). Interestingly, interviewees mentioned the existence of “cultural resistance” of some bodies that do not allow the PGMA to know about the processes within the secretariat/department, preventing access to relevant information about the body’s previous performance.

Considering there are exceptions, the perception is that managers of executive bodies, on a number of occasions, manipulate available resources so that the goals agreed are met without requiring all the effort planned with the Projects Office. Often the bodies do not act according to the Results Agreement due to political factors, budget cuts or changes in management. “For the Secretariat to achieve goals, they must find a way to perform tasks regardless of how they do it, just to guarantee that bonus” (E2).

There are PGMAs who understand that it is easier for the executive bodies to make an intentionally distorted interpretation of goals when these are essentially qualitative. In these cases, managers do not have great difficulties to “circumvent what has been agreed” (E5), not fulfilling their activities under the terms provided. However, there have also been reports from executive bodies where the PGMAs do not identify attempts to camouflage actions or employ fewer resources and efforts than the expected.

When the executive body does not show good acceptance of the PGMA, not allowing them to observe the actions of the managers and preventing access to information about the situation of the projects, the PGMA’s task of monitoring the goals becomes superficial and irregular. This makes it difficult to identify eventual deviations in the actions of the managers or position the Projects Office regarding the status of the goals. In cases where the PGMA has a good interpersonal relationship with employees, or gains managers’ trust, they are able to obtain a wide range of information and to follow several key indicators for the strategic projects.

It is important to highlight that before informing the Projects Office about the monitoring that has been conducted, the PGMA clarify to the managers of the executive body what information will be sent, the analyses executed and expected implications. “The idea is that no information should be left behind and all is passed on to the Projects Office, but the body should always be aware of the report. The trust of the executive body is important because it enables our work. That is, without their trust, our work is practically unfeasible. We need the executive bodies’ staff to trust us” (E2).

Political issues can also affect the exchange of information with the Projects Office. Respondents stated that, when they identify that the Projects Office (because of political reasons) may choose not to sanction a body that failed to accomplish an agreement, the PGMA might decide not to report the failure to the Projects Office. The rationale in such cases is that, as no sanctions are applied in anyway, reporting may create an unnecessary tension in the interpersonal relationship with the staff of the executive body. This behavior is not, however, a rule among the PGMAs. Some of the respondents categorically affirm that they do not neglect information, sending to the Projects Office all data about the executive body that can impact the progress of the strategic goals.

The analysis of these results showed the possibility of adverse selection before the creation of the position of the PGMA, given the lack of mechanisms by the Projects Office to gauge information about the history and capacity of the executive bodies.

The ex-ante opportunist action of the agent, giving incomplete information to the principal prior to setting the goals of the Results Agreement — which characterize adverse selection (Williamson, 1987) — can be recognized in this study. The resistance of the executive body to be transparent with the Projects Office regarding its real capacity reflects the concept of slack or inefficiency subject to the discretionary power of the manager, as discussed by Moe (1984). The cost to the principal to gain access to this information hidden by the agent is considerably high. In the case studied here, the cost for the Projects Office is hiring and allocating the PGMAs.

The results suggest that the presence of an intermediate in the agent’s work environment helps to access the information about the existing resources. It was identified that the executive bodies have budget, technology and personnel that were underused or idle, a fact that comes to the knowledge of the Projects Office through the action of the PGMA. According to one of the interviewees, “sometimes the culture is very resistant, they (the employees of the executive bodies) say that there is not enough people, but you can see that there are a lot of idle people who could be doing some work” (E5). The PGMA acts to reduce the problem of adverse selection by sending to the Projects Office management reports with information on the performance history and availability of executive body’s resources at the time of drawing up the Results Agreement. Respondent E2 points to this fact stating, “I have agreed with the staff of the Secretariat some routines of data sharing, such that they send it to me on a weekly or monthly basis and then I consolidate everything into a managerial report”. It should be noted, however, that the PGMA does not have mechanisms to evaluate any information omitted by the agent.

Just as there is an opportunistic action by the agent before the contract, there is also the action that follows the contract, defined as a moral hazard (Moe, 1984; Sung, 2005). The results point to the frequent occurrence of this agency problem in the context analyzed in a very similar way to that described by Eisenhardt (1989): the executive bodies do not put the effort as agreed with the principal and avoid coping with their responsibilities as contracted.

The research has shown that the executive bodies may act differently to what was agreed with the Projects Office because of political factors, budget cuts or changes in management. Regardless of the reasons, the results allow to consider that the bodies seek to deliver the expected result by the Projects Office, albeit by means other than those established in the Results Agreement, in order to guarantee the monetary bonus. Respondent E3 supports this statement when reporting that “there is manipulation of resources, so that the goals are achieved even if the service for citizens is not consistently provided”. This corroborates the work by Bevan and Hood (2006) regarding the practice of gaming — achievement of goals without achieving the actual final objective proposed at the core of the goal — in public organizations.

However, the capacity to reduce the problem of moral hazard can be influenced by the interpersonal relationship that PGMA has with the body’s employees. The present research indicated that PGMAs with better interpersonal relationships within the workplace of the executive bodies are better positioned to observe the activities and, consequently, to work to reduce the problem of moral hazard. For interviewee E2, the mitigation of the moral hazard problem is feasible, because the PGMA “manages to build a good relationship with the Secretariat staff, so they are very transparent with me and do not try to hide information”. In contrast, when the PGMA is isolated from the processes within the executive body and is not able to develop a good relationship with the employees, the impact of their action on the problem of moral hazard is reduced.

C. Risk propensity

The general perception among the PGMAs interviewed is that the executive bodies are considerably more concerned with achieving the goals of the Results Agreement than the manager at the Projects Office. Two reasons have been identified for this difference in the risk perception associated with the Results Agreement. The first is related to the image or status of the executive body in the context of City Hall as a whole, which may be damaged if the Results Agreement is not fulfilled. The second is the bonus attached to the fulfillment of this agreement, which can reach up to two salaries for each employee of the executive body, which is a great motivating element to achieve goals. As stated by respondent E2, “the executive body has much to lose if it does not meet its goals. The Projects Office, not so much”.

The fact that some bodies do not entirely comply with the Results Agreement can be positive for the Projects Office, since this would evidence that the goals set are bold and therefore only those that exceed the expectations are entitled for bonuses. On the other hand, it would be demoralizing for the Projects Office if the executive bodies did not achieve their goals in a massive way, which would represent an inability in the elaboration of the Results Agreement. However, for those interviewed, this is unlikely to happen. The concern of the Projects Office regarding the achievement of the goals established in the Results Agreement is directly related to the alignment between the provisions of this agreement and the promises made by the Mayor during the political campaign. When these two factors are not strongly related, the involvement of the Projects Office in efforts to achieve the goals tends to be reduced.

Regarding the difference between how executive bodies view the achievement of goals in the results agreement in terms of associated risks, in comparison to how it is viewed by the Projects Office, the perception of the PGMA interviewed is that there is nothing to be done about it. It was stated that “for the executive body the risk is much higher, and this cannot be changed” (E10).

The agency conflict related to the different risk propensity in the case studied here reflects the findings described by Lane, Cannella and Lubatkin (1998). For the authors, non-systemic risks — those related to a specific project or contract — pose no threat to the principal, only to the agent whose portfolio is difficult to diversify. The incompatibility of risk propensity becomes an agency problem since principal and agent are forced, via contractual arrangement, to share the same risks (Nilakant and Rao, 1994).

The results showed that executive bodies have greater concerns about meeting the goals of the Project Office, corroborating the theory. It is possible to draw a parallel between the literature and the reasons that would justify the agent’s greater concern: preserving the status of the body regarding City Hall and the possibility of receiving a significant monetary bonus. In the first case, the agent considers their reputation and the way in which other actors evaluate their performance along the interests of the principal, characterizing an external control mechanism (Saltaji, 2013). In the second, the references (Guimarães dos Santos, 2015; Misangyi and Acharya, 2014) bring the need for establishing mechanisms to reward the agent according to the general performance of the organization, which consists of an internal control system.

However, contrary to the theory, managers at the Projects Office can eventually understand failure in achieving a certain number of goals in a positive way. The rationale is that, to establish a coherent meritocratic policy, only the best performance must be rewarded, which means that failure is somehow needed to promote merit. In addition, it was identified that the principal in relation to the executive bodies may demonstrate more interest in the fulfillment of certain items of the Agreement for eminently political reasons.

As for the issue of the different risk propensities, the results of the research indicate that there are no mechanisms for the PGMA to act to reduce or influence this situation. The understanding is that each party views the risk associated with the Results Agreement in a specific way, and that there are no actions from PGMA able to change this reality. Respondents themselves acknowledge that, as stated by E5 “the Projects Office is not disturbed if the Secretariat does not achieve the goals in the Results Agreement” and by E2 as mentioned before “the executive body has much to lose if it does not meet its goals. The Projects Office, not so much”.

D. Different planning horizon

Regarding the time horizon for which the Projects Office works, respondents say that it is intrinsically connected to the Mayor’s term. The understanding is that the Projects Office is very political and its logic operates according to the issues of the current mandate, the ‘political timing’. The Projects Office therefore seeks to ensure that the Results Agreements signed with the executive bodies reflect the goals of the Mayor.

As for the executive bodies, however, the interviews revealed two widely divergent views. One respondent claims that the Results Agreement is simply a way to get a bonus at the end of the year. These managers, therefore, prefer that the goals deal with circumscribed issues in the short term, which can be easier to predict. However, all other PGMAs suggest that the executive body proposes its goals and projects based on the long-term planning. Respondents understand that “the executive body cares very much about the future impact of projects on citizens’ lives. Often, managers choose to neglect the short-term because of broader goals” (E3). It was emphasized that public servants with regular contracts generally adopt more focused planning in the long term, in projects that may lead to better delivery of services to the citizen or saving resources for City Hall.

A difference was observed in the temporal horizon between principal and agent, but different from the one described in theory. Theory suggests that the principal tends to consider a longer time horizon because of their link with the organization, whereas in general the agent has a transitory relationship and, thus, focuses on the short term (Guimarães dos Santos, 2015). In the context analyzed here, the Projects Office shows a planning horizon restricted to the term of office of the acting Mayor, so that the period considered by the principal is typically four years, or eight in case of re-election.

The executive bodies, however, operate for a longer time horizon than expected by the Projects Office, contrary to what has been suggested by the theoretical framework. This is because the executive bodies are massively composed of public servants, whose condition of employment stability provides safety regardless of political or electoral issues. Interviewee E10 understands that “perhaps because of the stability of the public service, all servants end up thinking more up the road. Because of politics, those in the City Hall focuses more on the short term”.

No means were identified for PGMA to impact agency conflict related to the mismatch of planning horizons. The research points to the maintenance of the principal’s vision connected to the Mayor’s mandate, as well as the agent’s vision towards long-term planning.

6. Conclusion

This study aimed at identifying how the performance of an intermediary, the PGMA, considered a typical representative of the Mid-Level Bureaucracy, affects the problems of agency within the relationship between the central power, represented by the Projects Office, and the executive bodies of a municipality, based on the perception of these actors.

The four agency problems addressed in this research were a) differences in motivation and goals; b) information asymmetries; c) different risk propensity; and d) different planning horizons. These agency problems were identified and discussed in the relationship between the Projects Office and government bodies implementing projects and strategic goals, which corroborates the argument of Jensen and Meckling (1976) that agency problems occur in all organizations and cooperative efforts (companies, universities, public agencies, unions, etc.). The results suggested that these agency problems do not always manifest in the exact form described in the literature or as predicted in theory.

It was identified that, physically present in the executive bodies, the PGMA develops sensitivity on the operational, budgetary and procedural aspects of the government agency. This, in turn, allows the dialogue and establishes trust with managers, promoting the alignment of goals between the principal and the agent.

The research demonstrates that, through the preparation of reports, management opinions and the continuous monitoring of executive bodies, the PGMA informs the Projects Office about the progress of the strategic goals and projects, providing the main data on the agent’s performance and status on the fulfillment of the terms agreed in the contract. The possibilities of hidden actions and evasion of information by the agent are then significantly reduced.

Results indicate that the PGMA, as an intermediary of the agency relationship, effectively contributes to reduce agency problems related to differences in motivation and goals and information asymmetries, including adverse selection and moral hazard. When it comes to the conflicts related to different risk propensities and different planning horizons, no impact from PGMA’s action was observed.

The contribution observed regarding the PGMA’s role may present additional challenges. As observed by Lotta and Cavalcante (2015: 297), mid-level bureaucrats (MLB) get personally involved with policies and programs in which they are working, and they act to institutionalize them and to “safeguard their memory”. However, the possibilities to frustrate expectations and motivation of MLB are many, given the intensity of the changes of goals, interruption of plans and policies, alternation of command and instability of political coalitions (Ring and Perry, 1985). These elements are possibly more intense in the context of the Brazilian public administration, with a political system based on a presidential coalition (Abranches, 1988). In addition to weakening this link in the chain of public policies implementation, this instability can generate a force of resistance or even incite the emergence of a cynical attitude and low involvement of these agents (Barton and Ambrosini, 2013), which may aggravate the agency problem.

Future research could advance in identifying these potential setbacks, but also in understanding the impact of interpersonal relationships on agency problems. The results of this study have indica­ted that agency relationships may have their conflicts attenuated or aggravated by the nature of the relationship between individuals acting as intermediaries and the agent, which confirms there are important elements that deserve further research.

Referências

• ABRANCHES, Sergio. Presidencialismo de coalizão: o dilema institucional brasileiro. Dados: Revista de Ciências Sociais, v. 31, n. 1, p. 3-55, 1988.
• ATTILA, Gyorgy. Agency problems in the public sector. Annals of the University of Oradea (Economic Science Series), v. 21, n. 1, p. 708-712, 2012.
• BARTON, Lisa C.; AMBROSINI, Veronique. The moderating effect of organizational change cynicism on middle manager strategy commitment. International Journal of Human Resource Management, v. 24, n. 4, p. 721-746, 2013.
• BEVAN, Gwyn; HOOD, Christopher. What’s measured is what matters: targets and gaming in the English public health care system. Public Administration, v. 84, n. 3, p. 517-538, 2006.
• BRESSER-PEREIRA, Luiz Carlos. Construindo o estado republicano: democracia e reforma da gestão pública. Rio de Janeiro: Editora FGV, 2009.
• BUCHANAN, James M. The limits of liberty: between anarchy and Leviathan. Indianapolis: Liberty Fund, 1975.
• BUCHANAN, James M.; TULLOCK, Gordon. The calculus of consent: logical foundations of constitutional democracy. Indianapolis: Liberty Fund, 1962.
• CAVALCANTE, Pedro; LOTTA, Gabriela(Org.). Burocracia de médio escalão: perfil, trajetória e atuação. Brasília: Enap, 2015.
• CHARMAZ, Kathy. Constructing grounded theory: a practical guide through qualitative analysis. Londres: Sage Publications, 2006.
• CHEN, Chun-An; BERMAN, Evan M.; WANG, Chun-Yuan. Middle managers’ upward roles in the public sector. Administration & Society, v. 49, n. 5, p. 700-729, 2017.
• COASE, Ronald H. The nature of the firm. Economica, v. 4, n. 16, p. 386-405, 1937.
• CURRIE, Graeme; PROCTER, Stephen J. The antecedents of middle managers’ strategic contribution: the case of a professional bureaucracy. Journal of Management Studies, v. 42, n. 7, p. 1325-1356, nov. 2005.
• DAILY, Catherine M.; DALTON, Dan R.; CANNELLA, Albert A. Corporate governance: decades of dialogue and data. Academy of Management Review, v. 28, n. 3, p. 371-382, 2003.
• DOWNS, Anthony. Inside bureaucracy Rand Corporation, 1964. Disponível em: <Disponível em: http://www.rand.org/content/dam/rand/pubs/papers/2008/P2963.pdf >. Acesso em: 14 dez. 2015.
  » http://www.rand.org/content/dam/rand/pubs/papers/2008/P2963.pdf
• EISENHARDT, Kathleen M. Agency theory: an assessment and review. Academy of Management Review, v. 14, n. 1, p. 57-74, 1989.
• FAMA, Eugene F. Agency problems and the theory of the firm. The Journal of Political Economy, v. 88, n. 2, p. 288-307, 1980.
• FIANI, Ronaldo. Teoria dos custos de transação. In: DAVID, Kupfler; HASENCLEVER, Lia (Org.). Economia industrial: fundamentos teóricos e práticos no Brasil. Rio de Janeiro: Elsevier, 2002. p. 267-286.
• FLOYD, Steven W.; WOOLDRIDGE, Bill. Middle management involvement in strategy and its association with strategic type: a research note. Strategic Management Journal, v. 13, p. 153-167, 1992.
• GAINS, Francesca; JOHN, Peter; STOKER, Gerry. When do bureaucrats prefer strong political principals? Institutional reform and bureaucratic preferences in English local government. British Journal of Politics and International Relations, v. 10, n. 4, p. 649-665, 2008.
• GUEST, Greg; BUNCE, Arwen; JOHNSON, Laura. How many interviews are enough? An experiment with data saturation and variability. Field Methods, v. 18, n. 1, p. 59-82, 2006.
• GUIMARÃES DOS SANTOS, Luis P. Comparing the use of forward-looking and contemporary performance measurement to formulate incentive contracts in the presence of the horizon problem: an experimental analysis. Revista Contabilidade e Finanças, v. 26, n. 68, p. 195-207, 2015.
• JENSEN, Michael C.; MECKLING, William H. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, v. 3, p. 305-360, 1976.
• LANE, Peter J.; CANNELLA, Albert A.; LUBATKIN, Michael H. Agency problems as antecedents to unrelated mergers and diversification: Amihud and Lev reconsidered. Strategic Management Journal, v. 19, p. 555-578, 1998.
• LAVARDA, Rosalia A. B.; CANET-GINER, Mária Teresa; PERIS-BONET, Fernando J. Como gerentes de nível médio contribuem para o processo de formação de estratégia: conexão entre processos estratégicos e práticas estratégicas. Revista de Administração de Empresas, v. 50, n. 4, p. 358-370, 2010.
• LEMIEUX, Pierre. The state and public choice. The Independent Review, v. 20, n. 1, p. 23-31, 2015.
• LIPSKY, Michael. Street-level bureaucracy Nova York: Russell Sage Foundation, 2010.
• LUZ, Cláudio L. M.; GOBBO, Gustavo; LAVARDA, Rosalia A. B. A participação do gerente de nível intermediário no processo de implementação da estratégia considerando os estudos de Floyd & Wooldridge (1992). Revista Iberoamericana de Estratégia, v. 12, n. 4, p. 123-149, 2013.
• MASON, Mark. Sample size and saturation in PhD studies using qualitative interviews. Forum: Qualitative Social Research, v. 11, n. 3, art. 8, 2010. Disponível em <Disponível em http://www.qualitative-research.net/index.php/fqs/article/view/1428/3028 >. Acesso em:4 maio 2017.
  » http://www.qualitative-research.net/index.php/fqs/article/view/1428/3028
• MÉNARD, Claude; SHIRLEY, Mary M. The future of new institutional economics: from early intuitions to a new paradigm? Journal of Institutional Economics, v. 10, n. 4, p. 541-565, 2014.
• MISANGYI, Vilmos F.; ACHARYA, Abhijith G. Substitutes or complements? A configurational examination of corporate governance mechanisms. Academy of Management Journal, v. 57, n. 6, p. 1681-1705, 2014.
• MOE, Terry M. The new economics of organization. American Journal of Political Science, v. 28, n. 4, p. 739-777, 1984.
• NILAKANT, V.; RAO, Hayagreeva. Agency theory and uncertainty in organizations: an evaluation. Organization Studies, v. 15, n. 5, p. 649, 672, 1994.
• OLIVEIRA, Vanessa E.; ABRUCIO, Fernanda L. Entre a política e a burocracia: a importância dos burocratas de nível médio para a produção de políticas públicas em saúde e educação. In: ENCONTRO ANUAL DA ANPOCS, 35., 2011, Caxambu 2011. Anais..
• PCRJ. Prefeitura da Cidade do Rio de Janeiro. Gestão de alto desempenho 2012. Disponível em: <Disponível em: http://www.conselhodacidade.com/v3/pdf/ RioGestaoAltoDesempenho.pdf >. Acesso em: 20 mar. 2015.
  » http://www.conselhodacidade.com/v3/pdf/ RioGestaoAltoDesempenho.pdf
• PIRES, Roberto R. C. Por dentro do PAC: dos arranjos formais às interações e práticas dos seus operadores. In: CAVALCANTE, Pedro L. C.; LOTTA, Gabriela S. (Org.). Burocracia de médio escalão: perfil, trajetória e atuação. Brasília: Enap, 2015. p. 177-222.
• PRZEWORSKI, Adam. Sobre o desenho do Estado: uma perspectiva agente × principal. In: BRESSER-PEREIRA, Luiz Carlos (Org.). Reforma do Estado e administração pública gerencial 5. ed. Rio de Janeiro: FGV, 2003. p. 39-73.
• RING, Peter S.; PERRY, James L. Strategic management in public and private organizations: implications of distinctive contexts and constraints. Academy of Management Review, v. 10, n. 2, p. 276-286, 1985.
• RIO DE JANEIRO. _{^{Edital SMA no 172 de 26 de julho de 2013}} 2013. Regulamenta o Concurso Público para Provimento no Cargo de Analista de Gerenciamento de Projetos e Metas, do Quadro Permanente de Pessoal da Administração Direta.
• SALTAJI, Issam MF. Corporate governance and agency theory: how to control agency costs. Internal Auditing and Risk Management, v. 8, n. 4, p. 47-60, 2013.
• SANCHEZ, Oscar A. Os controles internos da administração pública: a e-governança e a construção de controles no Governo do Estado de São Paulo. Tese (doutorado) - Pós-graduação em Ciência Política, Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2005.
• SCHILLEMANS, Thomas. Moving beyond the clash of interests on stewardship theory and the relationships between central government departments and public agencies. Public Management Review, v. 15, n. 4, p. 541-562, 2013.
• SILVA, Marcos F. G. Políticas de governo e planejamento estratégico como problemas de escolha pública II. Revista de Administração de Empresas, v. 36, n. 4, p. 38-50, 1996.
• SUNG, Jaeyoung. Optimal contracts under adverse selection and moral hazard: a continuous-time approach. The Review of Financial Studies, v. 18, n. 3, p. 1021-1073, 2005.
• THIRY-CHERQUES, Hermano R. Saturação em pesquisa qualitativa: estimativa empírica de dimensionamento. Revista Brasileira de Pesquisas de Marketing, Opinião e Mídia, n. 3, p. 20-27, 2009.
• VERGARA, Sylvia. C. Projetos e relatórios de pesquisa em administração 2. ed. São Paulo: Atlas, 1998.
• VISCUSI, W. Kip; GAYER, Ted. Behavioral public choice: the behavioral paradox of government policy. Harvard Journal of Law and Public Policy, v. 38, n. 3, p. 973-1007, 2015.
• WILLIAMSON, Oliver E. The economic institutions of capitalism Nova York: The Free Press, 1987.
• YESILKAGIT, Kutsal. The design of public agencies: overcoming agency costs and commitment problems. Public Administration and Development, v. 24, p. 119-127, 2004.

• 5
  {Translated version} Note: All quotes in English translated by this article’s translator.

Appendix: Interview script

Source: Elaborated
by the authors.

Publication Dates

• Publication in this collection
  Jul-Aug 2017

History

• Received
  01 Nov 2016
• Accepted
  29 June 2017