Publisher's note: This version of the publication was mistakenly typeset and published. The correct version can be found at http://doi.org/10.36745/ijca.317.

## 1. Introduction

Organizations of the justice system, such as the Public Prosecutor, Public Attorney, Advocacy, and Public Defender Office are responsible for initiating the action of the judiciary and are essential for the functioning of justice. The Brazilian Federal Public Defender’s Office is the agency responsible for:

• providing free legal assistance to the most vulnerable citizens among the population,
• promoting human rights, and judicial and extrajudicial defense.1

This agency gained relevance in Brazil through amendments to the Constitution, particularly No. 74 (2013), and No. 80 (2014). These amendments assigned functional and administrative autonomy to the Brazilian Federal Public Defender’s Office, including the capacity to initiate a budget proposal and to forward bills to Congress, and established the need for expansion of the Federal Public Defender’s Office. In this context, governance of the office has acquired greater importance. Despite this, few studies upon governance of that organ could be found in the literature review.2 To fill in this gap, this study identifies and discusses standards of good governance in the Federal Public Defender’s Office, as well as relationships between its dimensions and variables.

This study contributes to the advancement of the literature on the administration of justice, which has a greater focus on courts and neglects other organizations of the justice system. It also proposes a governance scale (in the form of a questionnaire) that can be useful both to researchers and practitioners to assess public defense bodies management. In addition, underlying the discussion on governance, this article reflects on the relationships between access to justice performance, and other dimensions of governance.

## 2. Public Governance

The concept of governance comes from several theoretical approaches including the Agency theory, Transaction Costs theory, and the Stakeholder approach. Similarly, the multilevel governance and policy networks approaches provide support for understanding public governance. From this perspective, governance refers to the management of stakeholders’ relationships as a means to solve conflicts that could threaten to divert the mutual gains of stakeholders.3 Another view of the concept of governance is related to a way of governing, moving from a hierarchical model to a more cooperative form, which can include several state and non-state stakeholders. These two meanings of governance deals with stakeholder relationships and the mechanisms used to manage such relationships. However, while the first concept is more related to the scope of private relations, the latter is within the scope of the state, where public governance is exercised.

Public governance involves actions related to the decision-making process, the functioning of government institutions and systems, and the relationship between state and society.4 It is a more cooperative way of governing, including a network of social stakeholders, in order to strike a balance between public power and civil society. This concept prompts reflection on the roles of state and society in public decisions, pointing to the importance of ensuring that stakeholders have the conditions for participation in the decision-making process.5

There is a principal-agent relationship, in which state organizations (agents) exercise the powers delegated to them to achieve the society’s (principal) interests. That is, the purpose of public administration is the achievement of society’s interests. However, the lack of resources needed to supply quality public services to all citizens has led to a shift in the role of the state. Instead of being a provider of public services, the state has become a coordinator, bringing together public and private resources at low direct costs to the public budget. The state becomes less self-sufficient and more inclined to act through networks of public and private stakeholders.6

Civil society organizations and the private sector can be seen as alternatives to the state bureaucracy and are important in the delivery of basic services to citizens. Thus, there are many parts to be considered in the process of modern public governance.7 The state is the central stakeholder of a system influencing public policies. The success of these policies depends on the abilities of public managers to negotiate with several public and private stakeholders. The domination of public policies by networks is a characteristic feature of the governance model in the public sector.8 The growing importance of this type of arrangement for modern governance emphasizes the highly interactive nature of political processes.9

Political networks are patterns of interdependent social relations between stakeholders with their own goals and strategies. Their interactions result from the sharing of interests.10 Contemporary public governance arrangements include mechanisms for the participation and cooperation of public, semi-public, and private stakeholders in the process of public policy decision-making. The interaction between state and society occurs with the aim of producing consensus.11

The solution for social problems is most effective when it involves the various stakeholders. Groups that are linked to the cause the problem, those affected by the problem and those who wish to solve it should be involved in defining the problem and addressing its solution.12 Public policies tend to be most successful when there is a combination of resources from both public and private sectors, as well as the continuous process of negotiation between stakeholders from these two sectors.13 Public governance is related to the collective solution of problems, in which the state plays an essential role of intervention and legal ratification, and other stakeholders interact in a cooperative way to address the solution of problems.14

The various forms of public governance depend on structural and institutional conditions to emerge. First, it is important that power is efficiently distributed in society in a non-fragmented way. Second, there is a need for strong and democratically legitimate public authorities capable of representing the interests of the main social groups. Third, that the authorities have sufficient resources and capacity to implement the chosen solutions. Fourth, it is desirable that civil society is strong, functionally differentiated and well organized; composed of citizens with the same rights under the law and endowed with guarantees and fundamental rights to exercise their citizenship.15 Finally, the dispersion of power and social control depend on the existence and guarantee of citizens’ rights, with special attention to transparency and the disclosure of information about public bodies. Democracy is the key input to better governance.16

These structural conditions define principles of good governance that are fundamental to the development of this new form of governing, based on the relationship between multiple stakeholders. The change from a hierarchical state based command and control relationships to a state characterized by complex patterns of contingencies and dependence on external stakeholders depends on the trajectory of institutional relations specific to each national context.17 Good governance of public organizations involves the identification and definition of responsibility, effective understanding of stakeholder relations and interests related to resource control and division of results,18 as well as the effectiveness of public policies, the legitimacy of procedures and their control by citizens.19

### 2.1. Governance in the Public Defender’s Offices

The Public Defender’s Office is a permanent instrument of the democratic regime. This agency is responsible for legal guidance, promotion of human rights, and defense of individual and collective rights of socially vulnerable people. There are 28 Public Defenders’ Offices in Brazil; one at the Federal level (object of this study) and 27 at the State level. There is a division of powers between these agencies. The Federal Office acts at the national level before the Federal Justice, High Courts and administrative bodies of the Union. The State’s Offices provide legal assistance in administrative instances and in all jurisdictional levels of the States and may act before the High Courts in case of judicial appeal.

The Federal Public Defender’s Office is chaired by the General Public Defender, who is responsible for administrative decisions. The General Public Defender also leads the Superior Council, a board composed by nine public defenders. This board is responsible for the normative decisions within the Federal Public Defender’s Office.20 Currently, this Office has 71 physical locations around the country, from which it serves more than 600,000 customers annually.21

In the justice system, the concept of governance has been associated mainly with dimensions such as independence, accountability, access to justice, structure and strategic resources.22 In the specific case of the Public Defenders’ Office, governance is related to similar dimensions; organizational resources, strategy, stakeholders, organizational structure, control, accountability and access.

Organizational resources include the various types of resources controlled by the organization to enable the design and implementation of strategies. An important point in the study of governance is the efficient allocation of resources,23 which is fundamental for compliance with decisions and achievement of objectives.24 In fact, the central point related to the autonomy and functioning of organizations of justice is precisely the ability to provide resources.25 The ability to implement plans and strategies is also related to governance. Building organizational strategy with defined objectives and priorities, and the creation of coalitions with stakeholders; ensures coherence in public policies.26 The Federal Public Defender’s Office is seen as an agency with high level of autonomy, but low levels of capacity.27

Stakeholder interaction – in an intentional, conscious and deliberate way – is fundamental to good governance.28 The nature of the relationship between the organization and its stakeholders is important in the choice of strategy. Stakeholders interested in the performance of organizations can influence and can be influenced by the strategy, depending on the possession of critical resources.29 These stakeholders organize themselves into social networks that can influence the delivery of public services.30 These social networks involve stakeholders of different types, public and private, individual and organizational, that act in the most varied federative levels, with different scopes of action.31 The broad representation of stakeholders at the highest level of judicial agencies is a fundamental element of governance for justice organizations.32

Organizational structure is another dimension linked to the concept of governance. It should be noted that the organization itself can be considered a governance structure,33 whose design influences the balance between the power of employees and managers, and also contributes to the control of the principal over the agents.34 It also highlights the autonomy of bureaucracy as relevant in this context, since highly institutionalized political systems have bureaucracies with a high degree of autonomy. However, an appropriate degree of autonomy does not imply the isolation of bureaucracy from society or decision-making process that is against the public interest.35

Mechanisms of control aim to mitigate possible opportunistic behaviour of agents, avoiding risks and inhibiting opportunistic actions.36 In the public sector, internal measures of administrative control could help to reduce abuses and inefficiencies.37 This dimension is linked to accountability, which covers issues related to transparency, answerability and enforcement imposed on managers, so as to allow control of the principal over the agent. Answerability encompasses two dimensions; one informational linked to transparency, relating to the agent’s reporting on their actions. The other explanatory, related to providing the agent’s justification regarding the motivation of their actions. Enforcement, in turn, is linked to the reward of good and punishment of bad behaviour.38 In justice organizations, there is a need for a higher level of transparency and accountability to increase efficiency and performance39 and to compensate its independence.40

Access to justice is another essential dimension of good governance of the Public Defender. The reason this organization exists is to enable access of the most vulnerable citizens to justice. Governance mechanisms are geared towards producing results that benefit stakeholders in a balanced way.41 A measure related to the capacity to deliver public services is important, but it should not be used as a single measure of governance,42 but combined with other measures. This dimension is linked to an increase in the provision of legal assistance services through the expansion of the public defender system.43

## 3. Methodology

Governance measures tend to shape the way policymakers see the world as they set standards of behaviour for organizations.44 Such measures can be based on secondary data and on stakeholder perceptions. There are authors who question the accuracy of governance measures based on perceptions.45 This research utilised a methodology based on perception; as the concept is still incipient for Public Defenders’ Offices consequently there is insufficient data for the construction of a measure based only on secondary data.

The identification of dimensions of good governance in the Brazilian Public Defenders’ Office involved interviews with managers of that agency and content analysis of those interviews. Between May and June 2015, fourteen (14) defenders and civil servants of the highest rank of that agency were interviewed, based on a semi-structured script (Appendix 1). The number of interviews was defined by theoretical saturation.46 The contents of the interviews were transcribed and analysed through categorical content analysis, following the guidelines suggested by Bardin.47 The governance dimensions for analysis were defined a priori based on the literature and included: Organizational Resources, Stakeholders, Control, Access to Justice, Strategy, Accountability, and Structure.

The researchers observed the occurrence of the dimensions in the interviews, their frequency and order of appearance, as well as the association between them. Considering the frequency of dimensions in the interviews, the total number of occurrence of the themes, and the order of occurrence, it was possible to group the dimensions in terms of their relative importance according to the interviewees perceptions. The most important group included the Organizational Resources, Stakeholders and Control dimensions. The intermediate group consisted of Access and Strategy. And the group of least relative importance included Accountability and Structure.

The Control and Access dimension was observed at least once in all interviews. Organizational Resources and Stakeholders dimensions appeared in eleven and ten interviews, respectively. On the other hand, some dimensions, such as Accountability and Structure, occurred infrequently, appearing in only a few interviews. It was also confirmed that the subjects that occurred more frequently in the interviews also contained a greater number of units of records. That is, in addition to being remembered by the interviewees, they were stated several times in each interview.

Subsequently to these interviews, a questionnaire was developed. It was composed of items that represented concrete operations and behaviours, through which the dimensions of good governance are expressed in the Public Defenders’ Office. The items were written according to the following criteria:

1. each item was to be expressed in only one behaviour;
2. the sentences had to be clear and intelligible to the target population; and
3. the item must be consistent with the attribute to be measured.48

In addition, researchers endeavoured to ensure that the set of all items related to an attribute covered its entire dimension, as Pasquali,49 Fowler50 and Gisselquist51 suggest. An initial version of the questionnaire with 84 items was built. As the theoretical, semantic and empirical validation steps were accomplished, the number of items was reduced; the final version of the questionnaire contained 32 items. The theoretical and semantic validation stages was intended to identify whether the questionnaire represents the governance construct. For this purpose, an analysis was carried out to assess the behavioural adequacy of the attributes, the relationship between items and attributes, their relevance to the theoretical context, and the clarity of the language used. This stage was carried out with nine experts in the field of public administration.

The second validation procedure involved the semantic evaluation of the questionnaire with the target population. Twelve individuals were selected from all strata of the target population and a pilot questionnaire was applied. The understanding of the items was observed, and possible doubts of interpretation were discussed. The number of participants was defined during the process; observing the criterion of confirmation of the theoretical saturation point.52 The time taken to respond to the questionnaire and the difficulties in its completion were also scrutinised.

Proceeding the validation procedure, the questionnaire was sent to all public defenders and administrative staff of the Federal Public Defender’s Office between February and March 2016, through the Google Drive Forms tool. The questionnaire was adequately answered. There were no records with the same score for all items; indicating that the form was filled in seriously. Only one of the records contained no data, which shows that Google Drive has recorded all submissions, even one in which there was not a single response.

The database with the responses to 393 questionnaires was analysed using the statistical software R (free). The main socio-demographic characteristics of the sample coincide with the Federal Public Defender’s Office global workforce profile, that is prevalence of young male adults.53 There is a dominance of males in the age group of 25 to 34 years. Breaking down the responses – 46% were public defenders and 54% were technical workers; 54.3% were male and 45.7% female.

In 15 questionnaires, 31 or fewer of the 55 items (56.0% of the questionnaire) were complete; the pattern indicated that they had been abandoned before completion. After these records have been excluded, the questionnaire with the lowest number of items answered had 34 answers. However, the respondent filled out all the fields of the socio-demographic information and even left a commentary on the questionnaire. There was no evidence that the participants of the survey left the questionnaire before they had considered the questions they had left unanswered. The presence of outliers were also identified. The method used for this purpose was the Mahalanobis distance, which can detect unusual patterns of response.54 Five outliers were found and were excluded from the sample.

In the end, there were 370 questionnaires answered, which corresponds to a ratio of 6.7 respondents per item. This number exceeds the sample standard stated by Pasquali,55 who suggests 5 to 10 individuals per item and at least 200 respondents. Tabachnick and Fidell56 suggest samples of at least 300 participants for analysis of discrimination. Field, Miles, and Field57 clarify that a sample of more than 300 subjects will probably provide a stable factorial solution.

In this database, 41 items had non-response rates of less than 5%. The remaining 14 items had between 5.1% and 9.4% non-responses. The items with the highest percentage of non-responses belong to the Control dimension. For the definition of correlation and covariance matrices, necessary for the analysis, the pairwise method of treatment of missing data was used, in which the data is excluded only from the calculations involving the missing variable.

## 4. Results and Discussion

The variables were not distributed normally or homogeneously. However, the violation of these data distribution requirements does not lead to erroneous statistical conclusions, especially if the sample size exceeds five individuals per item, when the distribution of the means tends to normal.58 In order to avoid any deviation, the extraction method used in the exploratory factor analysis was the Principal Axis Factoring; best suited for samples with non-normal distribution.59 The Exploratory Factor Analysis and Principal Component Analysis techniques were used to identify groups of variables. These techniques have three main uses: to understand the structure of a set of variables, to build a questionnaire to measure a latent variable, and reduce a set of data to a more manageable size while retaining as much information as possible.60

The Kaiser-Meyer-Olkin (KMO) test was used to measure the suitability of the sample for these statistical analyses, which indicated a very adequate adjustment. That is, the measurement of sample adequacy (MSA) was 0.96.61 The Bartlett sphericity test was performed on the correlation matrix to assess whether that matrix differs from identity. A significant test indicates that the correlation matrix is not equal to the identity matrix, with significant relationships between the variables. The Barlett sphericity test is highly significant for this data set (chisq = 11052.35, and p-value = 0). Thus, exploratory factorial analysis is appropriate.62 Considering the KMO and the Bartlett test for the correlation matrix, the data is adequate for the analyses.63 Both analyses were performed with “oblimin” oblique rotation, which maximizes the relationship between the variables and the factors. This rotation technique is based on the assumption that the factors are correlated.64

Principal component analysis indicated that a four factor framework could be considered sufficient. In addition, the dataset used is valid for the model being tested. These four components explain 41% of the variance of the model. The exploratory factor analysis also indicated the existence of four factors in a well-adjusted model, which corroborates the results of the principal components analysis. These four factors explained 39% of the variance of the model.

The similarity between the models estimated through principal components analysis and exploratory factor analysis is evidence that the analysed variables are reliable.65 This similarity is such that 25 items were shared between the models, which are grouped in a similar way in each factor. The dimensions of major importance to the managers participating in the interviews, namely Stakeholders, Control, Organizational Resources, and Access to Justice, were the same factors that emerged from the statistical analyses. This result confirms the use of a four factor model that emerged from the results of principal components and exploratory factor analysis.

The seven dimensions initially identified were regrouped into four factors, in a more parsimonious model. The items of the questionnaire were grouped so that the dimension “Organizational Resources” remained cohesive, constituting the factor Strategic Resources. The dimension “Control” merged with the dimension “Accountability”, forming the factor Control and Accountability. In fact, the concept of accountability is strongly related to transparency, culpability and enforcement, aimed at the control of public managers.66 Thus, it was to be expected that characteristics of these dimensions would be grouped into a single factor.

The dimension “Stakeholders” was divided. The items that dealt directly with the users of the legal aid service were grouped with the items of the dimension “Access”, constituting the factor Access to Justice. The other items of the dimension “Stakeholders”, related to the engagement of the stakeholders in the Public Defenders’ Office internal processes and decisions, were grouped into their own factor, called Social Participation.

The dimensions “Structure” and “Strategy” had few items emerging in the statistical analyses. Since the analysis based on the variation of the items, so that items with common variation tend to be classified in the same factor. The items related to these dimensions have little common variation among them, since the population studied belongs only to the Federal Public Defender’s Office.

This framework is both similar to and different from the one proposed by Akutsu et al.67 The study of those authors suggested the dimensions of judicial governance would be institutional environment, performance, access to justice, accountability, independence, resources and structure, and governance practices. For the Public Defenders’ Office, access to justice becomes a preponderant dimension since it reflects the mission of that institution. Social participation is also a significant factor because it is an organization that deals with the most vulnerable public and needs to be attentive to the problems and desires of social leaders. This factor was not considered in the Akutsu et al68 study, perhaps because it is understood to be somewhat conflicting with judicial independence. Control and accountability, as well as strategic resources, find their correspondents in the Akutsu et al69 model, accountability and resources and structure, respectively.

The questionnaire with the best statistical and theoretical coherence is composed of 32 items, grouped into four factors. Factor 1 – Control and Accountability – consists of ten items. Factor 2 – Social Participation – is composed of six items. Factor 3 – Strategic Resources – contains eight items. Factor 4 – Access to Justice also has eight items. Table 1 shows the factors, items linked to them, their factorial loads, and the medians of the responses. The responses were made on a seven-point agreement scale, where “1” represents total disagreement and “7” total agreement.

Table 1

Factors, items, factorial loads of the Public Defenders’ Governance Scale, and median responses.

Factor Item Factor Load Responses Median

Control and Accountability 32 – The Public Defenders’ Office carries out orientation campaigns with its members in order to prevent functional faults. 0.41 2
33 – The Public Defenders’ Office uses mechanisms to control the actions of its managers. 0.50 4
34 – The internal control bodies of the Public Defenders’ Office carry out periodic audits. 0.51 4
35 – The Public Defenders’ Office uses the recommendations of internal control bodies in the decision-making process. 0.57 4
37 – The organs of the higher level of the Public Defenders’ Office exert control over each other. 0.40 4
39 – The Public Defenders’ Office establishes mechanisms that guarantee the autonomy of internal control bodies. 0.47 4
41 – The Public Defenders’ Office establishes mechanisms to guarantee the answerability of its public agents. 0.56 5
42 – The Public Defenders’ Office punishes the public agents that act with impropriety. 0.61 4
43 – Public Defenders’ Office managers are held accountable for their actions. 0.70 4
44 – The Public Defenders’ Office applies, in its daily life, the Rule of Access to Information. 0.51 5

Social Participation 13 – The bodies of the higher level of the Public Defenders’ Office adopt formal mechanisms of social participation for decision-making. 0.52 2
14 – The Public Defenders’ Office considers the opinions of groups of society in its decisions. 0.40 4
15 – The Public Defenders’ Office allows the participation of diverse stakeholders (defenders, servants and citizens) in its decision-making process. 0.53 2
17 – The Public Defenders’ Office conducts public hearings to address the needs of the population. 0.36 3
19 – Public Defenders’ Office allows the participation of all stakeholders (defenders, servants, and citizens) in the internal processes for the choice of leaders. 0.71 1
21 – Public Defenders’ Office establishes mechanisms of social participation in the election of its leadership. 0.72 1

Strategic Resources 1 – Public Defenders’ Office conducts training activities for the public officials responsible for the services provided to citizens. 0.51 3
2 – The Public Defenders’ Office provides opportunities to update defenders’ knowledge. 0.47 5
3 – Public Defenders’ Office trains its public agents of the finalistic area in several fields of knowledge. 0.48 4
4 – The Public Defenders’ Office agencies maintain multidisciplinary teams for citizen services. 0.42 3
5 – The Public Defenders’ Office conducts leadership training. 0.50 2
6 – The Public Defenders’ Office provides the necessary working conditions for public agents to carry out their tasks. 0.49 3
8 – Public Defenders’ Office’s budget resources have increased over the last three years. 0.43 5
9 – Public officials use Public Defenders’ Office action plans to guide their activities. 0.43 4

Access to justice 16 – Public Defenders’ Office establishes channels of communication with society. 0.59 4
18 – Public Defenders’ Office conducts educational campaigns to inform citizens about the services that are provided. 0.60 4
22 – The Public Defenders’ Office works together with other public agencies to increase access to vulnerable populations. 0.66 5
23 – The Public Defenders’ Office works together with society’s organizations to increase access to justice. 0.67 4
24 – The Public Defenders’ Office contacts other public bodies through administrative channels with the aim of solving the problems of its public customers. 0.42 6
49 – The Public Defenders’ Office takes action to serve vulnerable populations in hard-to-reach locations. 0.62 5
50 – The Public Defenders’ Office conducts roaming customer service campaigns. 0.51 6
55 – Public Defenders’ Office acts to bring the defender closer to society. 0.49 4

Source: Research data.

For most of the items of the Control and Accountability factor, the respondents appear to be indifferent. The exceptions are items 32 – “The Public Defenders’ Office carries out orientation campaigns with its members in order to prevent functional faults” where the median respondent stated that they did not agree, and 41 – “The Public Defenders’ Office establishes mechanisms to guarantee the accountability of its public agents” and 44 – “The Public Defenders’ Office applies, in its daily life, the Rule of Access to Information”, where the median respondent said they agreed. The distributions of the responses related to the items of the Control and Accountability factor are shown in Figure 1.

Figure 1

Control and Accountability.

Source: Research data.

These results indicate that public defenders, managers and technicians do not perceive the structures and mechanisms of control as components of the Public Defenders’ Office governance. Accountability was neither a dimension highlighted by the subjects in the high ranking employees of the Public Defenders’ Office at the time of the interviews. This result is even more remarkable when considering that the Office is an autonomous agency, slightly permeable to democratic control mechanisms. The establishment of control measures is essential to supervise and prevent the abuses and inefficiencies of government, and is an essential tool to improve the management of public policies.70 It is worth mentioning that the Public Defender’s Office is considered an agency with a high degree of autonomy and low levels of capacity.71

For the Social Participation factor, the median opinion of the research participants was disagreement with all item. Their perception was that there is little or no social participation in decision-making processes. Items number 19 – “The Public Defenders’ Office allows the participation of all stakeholders (defenders, servants, and citizens) in its internal electoral processes” and 21 – “The Public Defenders’ Office establishes mechanisms of social participation in the election of its leadership”, in particular, elicited almost total disagreement. The distributions of the answers to the items of the Social Participation factor are shown in Figure 2. The mode of these answers corresponds to “totally disagree” in all cases.

Figure 2

Social Participation.

Source: Research data.

In fact, appointments to the highest positions in the Federal Public Defender’s Office take place through an election in which only the defenders participate. In the case of the General Public Defender (the Minister), the Federal Deputy Public Defender (the deputy Minister), and the General Corregidor, the President of the Republic appoints from closed lists built by the choice of the defenders themselves. Only the nominee for General Defender is submitted to oral test before approval by the Senate. In the case of the Superior Council, the normative body of the Public Defenders’ Office, there is not even the participation of the President in the appointment.72 Therefore, it is perceived that the Public Defenders’ Office is insulated from democratic controls. To enhance social participation, two good governance principles must be balanced: guaranteed independence,73 and for this independence to be matched with greater levels of transparency and accountability and broad representation of stakeholders in the supervisory board.74 This is not perceived to be the case in the Public Defenders’ Office. Maintaining an appropriate degree of autonomy does not mean isolating the bureaucracy from society or making decisions that are not in accordance with the public interest.75

Regarding the Strategic Resources factor, there is a tendency for respondents to disagree with some aspects of the questionnaire items. Respondents agree that the resources of the Public Defenders’ Office have increased in recent years, and that has provided an opportunity for defenders to update their knowledge. However, there is a perception that the necessary working conditions for public agents to carry out their activities are not available, nor are actions taken to train leaders or public agents who provided care for citizens. There is also an indifference to the use of action plans to steer activities. That is, the median responses indicate that the respondents do not perceive the strategic orientation of the agency. Figure 3 shows the distributions of the answers related to the items of the Strategic Resources factor.

Figure 3

Strategic Resources.

Source: Research data.

As stated in Figure 3, respondents agree that the budget has increased in recent years. The results illustrate there is a tendency to pay more attention to financial resources than to human resources. For good governance, it is not enough to accumulate resources, but they must be allocated efficiently. It is important to increase the level of education and professionalization of public servants to increase the capacity of government agencies.76

Regarding to the Access to Justice factor, the opinion of the research participants is in broad agreement with the statements in the items. The median answers do not disagree with any item related to this factor. Items 24 – “The Public Defenders’ Office contacts other public agencies through administrative channels with the aim of solving the problems of its public customers” and 50 – “The Public Defenders’ Office conducts roaming customer service campaigns”, both elicited a strong positive response from the respondents, who consider that the agency is making an effort to fulfil its institutional mission, either by its own means or through cooperation with other organs. Figure 4 shows the distribution of answers to each item of this factor.

Figure 4

Source: Research data.

The mission of the Public Defenders’ Office is to provide access to justice for vulnerable citizens. With this objective, the agency can: (a) meet citizens who seek the agency or (b) work in cooperation with other public agencies or civil society entities. In both cases, coordinated action is required to extend the coverage of the Public Defenders’ Office. This can be achieved through partnerships with other public agencies, political networks that involve non-governmental stakeholders alongside public agencies at different levels of the federal government are important to harmonize efforts and resources, as well developing more sophisticated strategies for the management and allocation of these resources.77 The Public Defenders’ Office relies on stakeholders to access and use the resources they need to extend their activities.78 This Office needs to cooperate with agencies responsible for controlling budgetary resources in order to increase its capacity.

## 5. Conclusions

This article objective was to identify good governance standards in the Brazilian Federal Public Defender’s Office, as well as the relationships between their dimensions and variables. The following dimensions were initially identified: Organizational Resources, Stakeholders, Control, Access, Strategy, Accountability and Structure. These dimensions were then rearranged into four factors: Strategic Resources, Control and Accountability, Access to Justice and Social Participation. As the study of the governance structure of the Public Defenders’ Office was examined in depth; the analysis indicated that the dimensions Structure and Strategy are relatively less perceived by the managers and technicians of the Public Defenders’ Office. Therefore, the 32 items in the questionnaire used to explain the construct are not directly related to the dimensions Structure and Strategy.

The results reveal that the governance structure adopted in the Public Defenders’ Office is still fragile. Public defenders, managers and technicians are indifferent to the attributes of Control and Accountability; demonstrating that they do not perceive structures and mechanisms of control to be components of good governance. These stakeholders perceive little or no social participation in decision-making processes. Considering Strategic Resources, respondents agree that resources have increased, although they realize that the necessary working conditions are not available to carry out their activities, and the actions to train leaders and agents in the main activity are inadequate. They could perceive an effort by the agency to promote access to justice, either by its own means or through collaboration with other public agencies or civil society entities.

In addition to these results, another contribution of this study is in the building and testing of a scale that can be used to evaluate governance on other public defenders’ offices. This scale is an appropriate tool to evaluate the perceptions of technicians and managers of such agencies, and to diagnose shortcomings in the decision making process, providing important information and data for the formulation and improvement of public policies and management necessary for good governance of public defenders’ offices. This questionnaire also complements and advances knowledge about the functioning of the Brazilian Justice System, as well as providing a foundation for future studies.

It is important to examine the principles of good governance in a context in which the state is charged to engage with society and produce quality public services that meet the needs of the population. The information and knowledge produced by this study can help the development of more efficient strategies in the management and allocation of public resources, and consequently more effective provision of legal aid services.

This research is limited in the sense that it is an initial attempt to fulfil a research gap upon governance adopted in the Public Defenders’ Offices. Although relevant for future research, the results are preliminary and more for guidance than for reaching firm conclusions that practitioners can use. The sample was not defined probabilistically; making generalizations insecure.

Further research is recommended, as well as replication in other public defenders’ offices in Brazil and in other countries, which will allow the construction of a scale that is valid for this sort of agency in general. The cross-sectional nature of this research can also be considered a research limitation. The use of a longitudinal perspective would make it possible to see the evolution of the concept of good governance in Public Defenders’ Offices, as well as providing more empirical evidence to improve the accuracy of the questionnaire built. Furthermore, the results are in keeping with the bias of Public Defenders’ Office managers and technicians. The present study did not examine the quality of the legal aid service from the users’ perspective, and future research that seeks to understand citizens’ perceptions of governance in the Public Defenders’ Office is also recommended.