Control and Public Management Performance in Brazil: Challenges for Coordination

This study analyzes the multiplicity of organs and control actions as a distinctive characteristic of the Brazilian control system, and the positive effect of the performance of these organs upon the management of federal programs. As shown in international literature, we can also see in Brazil an expansion of the action of control organs, which have started evaluating performance and, therefore, go beyond mere verification of the legality of administrative actions. Nevertheless, the multiplicity and diversity of these organs have meant the appearance of tensions and conflicts in the relationship between controllers and those controlled. Through the analysis of documents and interviews with managers and the controllers of federal government oversight bodies, we conclude that the process of auditing and evaluation leads to the appearance of cooperative practices that improve the administration’s performance, but which themselves need improving.


Introduction
The control systems of public administration have been gaining importance in many contemporary democracies as a function of the growing demand for more government transparency and accountability.These issues entered the public agenda in Latin America countries in the 1980s and 1990s, boosted by the processes of re-democratization and by the State's experimentation with reforms.
In Brazil, control and auditing institutions have undergone changes since the re-democratization of 1985 and the new Constitution of 1988, which set out processes for institutional renewal and the strengthening of republican practices.Contrary to international experience, however, Brazil is characterized by a multiplicity of control institutions, with varying scope, but which are often superimposed in an inconsistent manner.Another difference in Brazil relative to the international experience is the latent conflict between control agencies and the organs of public administration.The control and auditing activity is still seen by officials as predominantly based on the standards of strict legality, and there are divergences about how to define the performance assessment standards.
What are the impacts of institutionalization and expansion of the control actions of public policies in Brazil today?This is the question that guided our study.Empirical research has shown that, beyond the strengthening of control agencies in general, conveyed by the Brazilian democratization process, changes of the old decentralized system of internal control-dispersed in the various bodies of the Executive and subordinate to each Minister-and its replacement to a new a model that centralized the internal control procedures in a single organ, created with ministerial status, therefore with more importance before the governmental structure, had two contradictory effects.On the one hand, such change increased the efficiency of the control itself.On the other, it had led to conflicts between the various control agencies, among auditors (whom had their activity now carried out effectively) and managers who resisted the control activities, once they seen it as barriers to management.If this double and paradoxical situation-advancement of the control and increasing demands for more government efficiency in delivering public services-is the outcome of the very expansion of the democratic order in the country.What we seek to emphasize in this research is a concrete matter of public administration: how to encourage forms of coordination and avoid conflicts that could hinder or even disrupt the effective government action.
The text is organized as follows: after the description of the research methodology and a brief literature review, we analyze the control system structure in Brazil and its incremental advances.Thus, we discuss the core of our analytical problem, which is to show that the main challenges to coordination and cooperation between auditors and policymakers.The final considerations present some procedures, found during the empirical research, aiming at a better coordination between the actors involved.

Brief Review of Literature
The main recent changes in the organs of control and auditing pointed out by international literature were the broadening of the scope of control and its politicization.The broadening of the control organs' range of actions, which meant the inclusion of evaluations of policy implementation, together the traditional verification of their legality, increased the political sensibility of their activities, that is to say, the political importance of the assessment produced by the audit body for the decision-making organs or positions responsible for government programs (Barzelay, 1997).This process is the result of the adaptation of the form and action instruments of the agencies of control that is external to the new arrangements and criteria for the performance of public administration, which were defined in the administrative reforms of the 1970s and 1980s.The control organs had to adapt to the new defining criteria of public administrative performance in order to supervise it, and these criteria ceased to refer to just legal compliance.They went further and included standards of quality in economic terms (efficiency) and of the effectiveness and impact of public action.This is how the current dual nature of the agencies arose: a control of legality and performance (Pollitt et al., 2008).
In Latin America, mechanisms of transparency and accountability are historically weak due to the origins and characteristics of the political institutions, marked by a strong legacy of patronage and authoritarianism (O´Donnell, 1998).But progress has been made over the last two decades in Latin America (CLAD, 2000;Bresser Pereira, 2004;Pinto & Flisfisch, 2011).In Brazil, we can indicate the reform and strengthening of control organs and the creation of laws for access to information, among others (Speck, 2002;Pessanha, 2003;Olivieri, 2010;Arantes et al., 2010;Loureiro e Teixeira, 2009).

Methodology
The result of this process is a system with various organizations, residing in the three Powers (Executive, Legislative and Judiciary), with the ability to exercise control over public administration in various spheres (political, administrative and judicial).Here also we can observe the same process of broadening control beyond legality to cover performance assessment and the politicization of audits, in the sense of increasing interference in the decision-making processes of public administration.
The result of this process is a system with various organizations, residing in the three Powers (Executive, Legislative and Judiciary), with the ability to exercise control over public administration in various spheres (political, administrative and judicial).Here also we can observe the same process of broadening control beyond legality to cover performance assessment and the politicization of audits, in the sense of increasing interference in the decision-making processes of public administration.
The empirical data examined here was collected in 2011 and 2012 from source documents and in the course of interviews with employees from control institutions and heads of federal government organs under their supervision, using a methodology called "survey of the elite".Its purpose was to capture perceptions regarding the functioning of control institutions using the privileged insight of the system's central players at their different levels (Note 1).
The choice of a survey of the elite method is justified, on the one hand, on the basis of the scarce literature specifically dealing with control from a political and institutional standpoint as well as the way in which it routinely operates.Interviews, to a far greater extent than documents, also enable the transmission of knowledge on how institutions and control mechanisms work, as long as this functioning does not derive exclusively from existing rules, and there are practical dimensions and an organizational culture which may prevail over procedures or determine the ways in which they are interpreted and applied.
The survey of the elite is not by nature a sampling process.Rather, it is representative of the different perceptions of key players of the control system and federal public administration.Its objective, therefore, is qualitative, because it makes it possible to collect evaluations regarding the functioning of the system by central institutional players.Such evaluations, in spite of being partial, are extremely qualified owing to the experience of interviewers and their positions within the institutions.Sixteen in-depth interviews were carried out with professionals chosen in accordance with a combination of criteria relating to institutional position and professional experience in the area.
The choice of APF organs, in its turn, was made taking into consideration the diversity of the areas (purpose and means, in addition to social and infrastructure), the level of structuring of the ministries (oldest and most structured as well as the more recent and least organized of the organs) and institutional diversity (ministries, of direct administration, public companies and autarchies, of indirect administration).
Accordingly, of those subject to control we interviewed managers from five ministries (Education, Health, Social Development, Cities, Planning and President's Chief of Staff) and two indirect administrative entities-INFRAERO (public company managing airport infrastructure) and the National Department of Infrastructure and Transport (DINIT), an autarchy responsible for the operation and maintenance of the transportation system.Among the auditors, those interviewed were technicians from the CGU (Federal Comptroller General), the TCU (Federal Auditors Court), the AGU (Attorney-General of Brazil), the MPF (Public Prosecutor's Office) and the Congressional Budget Commission.

Incremental Advances in Control Institutions and How the Control System Is Structured
Is it adequate to speak of control system in the Brazilian political structure?The adoption of the system concept attributed to organs of internal and external control in Brazil as a whole can be justified not only for reasons of a legal nature-the constitutional text of 1988, in Article 70, defines it as such-but also for analytical reasons.Using the notion of system as an interrelated whole of elements interacting in performing a function, we can state that the control organs in Brazil constitute a system.Although problems of coordination with one another exist, as will be pointed out further on in this text, the various internal and external control organs of the Brazilian federal administration are interrelated in exercising their common activities, working with different means in order to make the various public agencies accountable for their actions and omissions.
From a formal standpoint, the control of the Brazilian Federal Administration (APF) is in charge of different entities, comprising a complex system made up of different organs with discrete powers and spheres of activity.Among these are the National Congress (CN), the external control organ (Federal Auditors Court-TCU), the internal control agency (the Federal Comptroller General-CGU), the organ responsible for defense and legal advice for the Executive Power (the Attorney General's Office-AGU), the Federal Prosecutor General's Office (MPF), and the Judiciary itself.
In other words, the control system is made up of agencies that are both internal and external to the Executive, which reflects the presidential system and the separation of powers.Both the Legislative, as well as the Executive have their own control institutions.The Legislature exercises classical political control over the Executive, with assistance from the TCU.The Executive, in its turn, maintains a system of internal control that is centralized in the Federal Secretariat of Internal Control (SFC) which is currently linked to the CGU.Although control of legality prevails in Brazil, the Federal Constitution of 1988 sets forth that control must also be exercised in relation to legitimacy and economic factors.There is thus a growing understanding that control should equally involve efficiency and the outcomes of government action.
As expected, the re-democratization that began in 1985 and the Constitution of 1988 led to institutional advances in all organs of control in the country, in terms of the qualification and structuring of their actions.While this is still ongoing, it is a slow and gradual process, as recent research indicates.Let us examine in slightly more detail the areas of external and internal control of the federal administration.

The Federal Court of Auditors-TCU-and Its Links to Other Control Institutions
Recent research has shown that not only the Federal Court of Auditors (TCU), but all the country's courts of accounts have in the last two decades undergone processes of organizational modernization and skills training of their technical staff (each of the 27 states of the federation has its own court, and some large municipalities do too).Careers in control were thus created, the transparency of their actions expanded and the sphere of these organs' actions broadened in the implementation of public policies.A large part of them already had information of public interest on their websites, opening themselves up to dialogue with various players in different sectors of civil society (Loureiro, Teixeira e Prado, 2008).
In addition to this, the 1993 law of public tenders (Law 8666/93) allowed any person to submit questions to the court of accounts pertaining to fraud in public procurement, a recourse that exponentially increased the number of cases and even altered the routines of the organ (the courts are obligated to investigate all cases brought by citizens).Congress began to insert provisions in the Budget Guidelines Laws (LDOs) that oblige the TCU to send congressional members information regarding the progress of federal projects every year (this occurred due to work by the Congressional Committee of Inquiry (CPI) on unfinished government projects, which between 1995 and 1996 identified various irregularities in the public works).
Other demands have also put pressure on the country's courts of account.The first refers to the Law of Fiscal Responsibility (LRF) which attributed to them the duty of supervising, through the emission of legal opinions, the limits of federal spending.The second has to do with the need to respond to complaints presented to the ombudsman, an area recently created in almost all courts.With regard to this series of changes and their impacts, it is here worth quoting the comments of one of the TCE's functionaries: We've had to adapt due to changes in society, which now demands more transparency from the government.We've had to adapt due to this relationship with the National Congress and also because of a series of competences that are attributed by infra-constitutional legislation.The press today is impressive: not a day goes by that we don't have to talk to some organ of the press.This also increased the number of demands.(...) The Court therefore adapted, but this isn't visible to the outside population, also because the Court was not accustomed to showing itself, the Court was an institution closed in on itself.
The TCU has in fact restructured its internal organization, as well as the professional requalification of its personnel, adapting them to meet the new demands.It has also undergone a reorientation of its oversight activities, which are now more centered on visits in the field rather than merely documentary analysis.New specialist areas have been created, such as the department dedicated to the analysis in loco of works financed by the federal government.It has also adopted new practices of operational auditing, with international standards, aimed at prioritizing the evaluation of the performance of public management, as well as the supervision of projects.
On the other hand, in 2000 the TCU created its parliamentary advisory service to deal with relations with the Legislature, and began to offer its reports and technical assistance to congressional members and to permanent committees.This preventive action, as well as allowing the improvement of the Court's relationship with the Legislative Branch, which is its main "client", was justified because the demands of the parliamentary committees made to the TCU were considered to be poorly formulated, imprecise and difficult to pass forward.To reconcile the congressional members' demands with its own capability, the TCU began presenting, through its Parliamentary Advisory Service, its "portfolio" of projects, so as to encourage committees to make demands on the supervision that was already being undertaken.In this way, it sought to improve its service to the National Congress and retake supervision over its own work routine.
The definition in the TCU's strategic planning of the goal to strengthen relations and partnerships with Congress is a positive indicator of the close relationship between the two strongest external control organs of Brazilian bureaucracy.Nevertheless, this effort comes up against the relative disinterest among members of Congress relating to questions of control, a behavior that still prevails as a general rule, because few congressional members see this attitude as a politically viable platform.In the view of career bureaucrats in Congress, there are individual initiatives from someone's favoring control, mainly through demands for information, but the Brazilian Legislative Branch in general makes little use of the structure of the TCU to supervise the APF.Despite the relative disinterest on the part of Congress, institutional relations between the two organs have gotten closer, allowing for optimism in terms of the coordination of their actions to foster transparency and responsibility in public administration.
In summary, the relationship of the TCU with Congress has become closer, whether by way of demands from Congress in the area of supervision of public works, or because of efforts by the Court to adapt to the requirements and demands of the parliamentary committees.
Over the last two decades the TCU has also straitened its relations with the Public Prosecutor's Office-MPF.The attorneys have become energetic petitioners of supervision and auditing, as they have a small team of auditors.These closer ties resulted from a TCU initiative.Similar to the work done with the congressional members, the organ presented the attorneys with its "products" (audit and supervision reports).Despite this, difficulties between the two organs still remain and have to do, according to the auditors, with the autonomy of the MPF, together with the technical deficiencies with regard to the functioning of public policies.The creation of the Control Network in 2009 seems to be improving the articulation between the actions of the organs, as the agreement between the TCU, MPF and the Federal Police bears witness.This partnership established a minimum common scope for audits carried out by the three bodies, in such a way that the work of one institution can be used by the others in cases where all are investigating the same project or activity, thus eliminating duplication.
The relations of the TCU with the Federal Comptroller-CGU are of fundamental importance, since rendering accounts-the main input of the TCU's works done by the Comptroller.The contact is broad and constant, in the view of the TCU workers-there is no overlap between the work of the two organs, with the boundaries of each being well defined.Nevertheless, the formal initiatives of inter-institutional cooperation are still recent and the result is still little known (Note 2).

Internal Control: The Emergence of the CGU as a Central Organ for Internal Control
The Federal Comptroller's Office-CGU is a recent creation in the country's institutional scenario, having arisen in 2003 at the beginning of Lula's first presidential mandate.It centralizes the internal control of Brazil's public federal administration and has performed the role of a platform upon which to build Brazil's democratic order, with an area of operation that extends well beyond a mere organ of internal control, and includes other functions, such as combating corruption, monitoring public policies and promoting transparency.As one of its more innovative aspects, it has played an important role in the mobilization of civil society, capacitating it for the exercise of the function of social control of the governing classes (Loureiro et al., 2012).
The creation of the CGU also represents an institutional innovation in the control process of government in Brazilian democracy for various reasons.Firstly, because it reorganized the internal control of the Federal Public Administration (APF), centralizing it into an organ with ministerial status.Until 1994 internal control was fragmented with a department that had little efficiency within each ministry (the Internal Control Secretariats, known as CISETs).This reorganization involved, as well as organizational centralization, an expansion of activities also for monitoring federal public policies carried out in bodies at the sub-national level, through supervision in municipalities with federally financed projects (Olivieri, 2010).
In second place, the creation of the CGU consolidated the expansion of internal control activities for promoting quality administration, thereby going beyond the traditional control of legality.Parallel to this, the CGU became Brazil's "anticorruption agency", responsible for sanctions against administrative impropriety and for the promotion of transparency in the APF and the fostering of ethics and integrity (Corrêa, 2011).In addition, it encouraged the dissemination of internal control to governments at the sub-national level, which are creating their own controllership bodies.
Last, but not least, the CGU began to act as a body that encourages and strengthens social control, in other words, the participation of civil society in controlling public administrations, thereby contributing decisively to the institutionalization of this process through various initiatives, such as the technical training of councilors and the initiative to organize conferences on social control (Loureiro et al., 2012).

Sharing Objectives for Cooperation between Controllers and Those Controlled
Considering that control is not an end in itself, but an instrument for improving the management of public policies, its purpose and modalities should be defined by the Public Administration together with the control bodies, both respecting their constitutional attributes and their operating specificities.In other words, control institutions cannot only be developed in response to their own needs and ignoring the particularities of ministries in terms of improving their management.Sharing goals is of paramount importance, as both supervisors and the supervised need to bear in mind that the function of both is to ensure the legality and efficiency of the use of public funds.
Therefore, overcoming mistrust and the establishment of constructive dialogue resulting in benefits for both depend on investment in agreements based on common objectives and guided by respect for legality and the search for greater efficiency in the management of public policies.
The most decisive factor for the appearance of cooperative practices between controllers and the controlled is the existence, in the controlled bodies, of their own audits and/or internal ability to monitor the implementation of their programs.The more consolidated these practices, the greater the chances of dialogue, with the controllers contributing towards improving management quality.It is thus more likely that conflict and tension in the auditing and supervision processes will arise in organs or ministries created more recently, or those that have had to operate programs at a national level without a decentralized organizational structure.
Moreover, interaction between control and management in these ministries will contribute to the detailing of programs that are normally formulated in a more generic manner, by specifying stages and procedures for their implementation.As will be shown below, this interaction will also foster the designing of intergovernmental and intersector coordination mechanisms, even leading to the definition of performance evaluation criteria.
The absence of rules formalizing the link between management and control can make the collaboration relationship more tense and problematic.In those organs with a longer period of institutionalization, there are signs that the space for dialogue with the supervisors is already more consolidated, in many cases having their own organs for internal control which facilitate dialogue between supervisors and the supervised.These more recent ministries, especially those that deal only, or as a priority with decentralized resources, often view problems pointed out by control bodies in their auditing as obstacles to the development of their actions, and not as suggestions as to how policies or monitoring and evaluation mechanisms could be improved.The direct consequences of this situation are generally audits/supervision that point to a series of problems that reflect only the observations of the supervisor, but find no support in the concerns of the administrator.
The following examples illustrate these formulations.An interviewee in the internal control department in the Social Development Ministry, created in 2003, at the beginning of President Luis Inacio Lula da Silva's first mandate and the central nucleus of his agenda in the social area, relates a positive control experience.Despite criticizing the amount of rework demanded of him, rendering accounts on the same issue for different control organs, he was able to a build a collaboration relationship with the control area: after long discussions with auditors from the TCU, his department was able to reach a consensus over the composition of the Decentralized Management Index (IGD), an indicator to demonstrate the administrative quality of the Program Bolsa Família (Family Allowance -an income transfer program).After long disagreements between his department and the TCU auditors, both decided to establish a dialogue and ended up building an indicator that reflected the expectations of the area supervised and included the concerns of the control body, thereby fostering a greater management capacity for the Ministry itself.In this case, the concerns of the supervisor went beyond mere observance of the legality, which contributed to broadening the vision of the TCU itself on how to evaluate the public policy in question at the time of its audits.
Similar situations have been repeated in other federal government organs, such as in the Transport Ministry, in which the routines involved in contracting public works and of monitoring resources were altered because of suggestions made by the control bodies or by way of dialogue of the latter with the managers.According to another interviewee, the decisions of the TCU to forbid hiring relatives for outsourced contracts produced a good result: they ended once and for all the occurrence of this frequent practice.There was also considerable improvement in engineering work public tenders with the incorporation of the recommendations contained in the TCU accords, which were sent to all personnel by corporate e-mail and made available for consultation by the managers.
Faced with excessively formalized control, also according to the managers, various expenditure takers in the different ministries are fearful of signing projects or authorizing expenditure due to action by the control bodies, which prioritize excessively formal aspects, even in situations in which there is no loss to public coffers or that where motivated by the need to make policy management more efficient.
In fact, a large part of the interviewees point out the absence of a consensual definition over what should be the object of control, in other words, controllers and those controlled have differing views about the purpose of control activities and the way in which they should be carried out.Obviously, no one is suggesting that those subject to control should define the parameters or the instruments of the supervision and auditing.But one would expect that a minimum degree of agreement regarding objectives and the way audits should be carried out could be established, in such a way that those being controlled feel themselves impelled (not just obliged) to collaborate with the supervision and that control does in fact lead to improvements in management.
In short, if control is not an end in itself, but is as necessary as good public management in a democratic order, the dialogue between controllers and managers, and the coordination of their actions are practices of fundamental importance.Not only does the language between them need to be fine-tuned, but also sharing the goal of reaching the greatest efficiency for public policies is indispensable.It is also fundamental that the "internal affairs culture" with its practices of punishment on the part of the auditors and, on the part of the managers, the view that "control interferes with management" must end, These mistaken conceits make finding paths to coordination difficult and prevent mutual learning through dialogue and shared experience.

The Building of Cooperation Mechanisms between Management and Control
Two subjects relative to cooperation between control organs may be highlighted: defining the scope of the control of legality and the creation of institutional mechanisms for generating cooperation.
With regard to the first topic, it appears that today there is a clear consciousness among members of the internal control body that formal controls demand time and do not necessarily produce results in terms of efficiency in public policies.Ultimately, the ideal would be to reduce them or do away with certain procedures where there is a low cost-benefit ratio (cost of control and benefit in terms of the quality of management).In other words, controlling of legality should not be totally abandoned, but it is necessary to assess by how much the registers founded solely on legality produce substantial benefits for the effective performance of public services.
In relation to the second topic, on the other hand, there are certain actions that are aimed at cooperation.The External Organ Monitoring and Service Permanent Committee was created within the TCU with the function of "coordinating, monitoring and dealing with administrative procedures involving the TCU, the CGU and the Chief of Staff to the President of the Republic and other external organs of control" This committee's objective is to render the procedures more efficient for the resolving the more than one hundred cases initiated by the TCU which have paralyzed important public activities, such as large-scale construction work, in addition to lending support to officials being investigated by the control organ.
Indeed, the External Organ Monitoring and Service Permanent Committee has led to a significant reduction in the volume of pending cases in the TCU to bring about greater compliance by the controlled organ in terms of suggestions for external control, or even the scope of negotiated solutions between both.The Committee has also given managers more institutional security for their decisions.
In addition, the Committee has looked for solutions for the specificity of purchases in particular areas of public administration.The adoption of fixed parameters to evaluate the price of the public construction work is very important and has caused a great many conflicts between controllers and the controlled.Presently, the organs of control adopt, as a parameter for evaluating the cost of a project, the table of the National Survey System of Civil Construction Costs and Indexes (SINAPI), jointly produced by the Caixa Econômica Federal (a public bank) and the Brazilian Institute of Geography and Statistics (IBGE).Based on prices featured in the SINAPI, the TCU determines whether the price of a project has been appropriately estimated or whether the invoices have been fraudulently inflated, something which has given rise to questions from sectors in the public administration, such as the aerospace and petroleum industries and road construction and repair.They claim that the SINAPI only includes prices of civil construction material and does not consider the specificities of products used in very specific activities and projects such as those conducted by Petrobras and by Infraero, and for that reason disagree on the irregularities pointed out by the controlling organs.Recognizing the specificity of airport construction work, a committee was formed in conjunction with the Caixa Econômica Federal and the TCU in order to discuss the construction of a SINAPI exclusively for the aircraft sector.The petroleum sector is also proposing the same solution.
If such examples prove that dialogue can assist in overcoming differences, it should be clear, however, that it is not sufficient for the adequate functioning of the system of control and improvement in public administration.It is only with the due institutionalization of arenas of coordination between management and control organs, the establishing of clear and specific criteria of evaluation for the different administration organs, which respect their diversity, that there will be greater affinity between the objectives of control and management.

Coordination between Internal and External Control Organs
Difficulties in coordinating control activities manifest themselves chiefly in relations between internal and external control.Despite enjoying the autonomy to plan its supervision activities, the SFC does not define its activities entirely independently of the TCU.Even being an organ of the Executive, therefore, without an institutional link to the Legislative, or to the Federal Court of Accounts, the Constitution tasks the SFC with supporting external control.This support is provided mainly by means of all managers rendering accounts, which are organized and systemized by the SFC, and forwarded to external control.
The MPF is another important component of the control system.Those interviewed belonging to the TCU and CGU recognize the importance of the MPF in supervising the activity of public agents and the application of resources, as well as seeking legal action in cases in which crimes against public property have been determined.Nevertheless, many point to the difficulty in holding a systematic dialogue with the institution, given that freedom of action by the prosecutors obstructs mechanisms of cooperation with all members of the MPF, which does not, however, rule out institutional cooperation between the organs.
With regard to those controlled, their perception of MPF activity is more partial and conditioned by their restricted contact with the institution: the managers generally consider the MPF to be just one more organ demanding information (obliging the organ to divert working hours from its core activity to deal with its requests, which are all too often repeated by other control organs.In many cases they point to the lack of knowledge of prosecutors of how public policies function and the part played by managers, which leads them to requesting information that is difficult, if not impossible to provide.Here once again, the need arises for coordination between control organs in order to avoid this extra workload for managers, while at the same time ensuring that all the information is provided in the supervisory and/or investigatory processes.
If on the one hand managers within the ministries recognize that the importance of control over legal procedures cannot be neglected-their abandonment would be reckless and would open the doors to all kinds of irregularities-they do, however, criticize the excessive preoccupation of auditors with the formal dimensions.They believe that formalities that do not constitute corruption or a deliberate intent to cause damage to public coffers should not constitute a source of obstacles to the continuity of policies.In these cases, even when the objective of the audits is to preserve legality and increase the efficiency of management, the effect on the latter is practically null, and the risks of seriously affecting the rendering of important services and causing prejudice to the population are very high.

Conclusions
The empirical material analyzed here allows us to point to certain measures to foster coordination, and thereby improve the quality of control and of management: 1) Development of a continuous dialogue between the organs of control between themselves and with society, to clarify the content of supervisory reports and to avoid possible confusion between problems of corruption and failures in management (society should be enlightened as to the difference between the two); 2) Creation of effective institutional coordination mechanisms between organs of control and between controllers and those controlled, by means of the standardization of common procedures and the sharing of information systems.It is important to establish a regular forum for coordination, with executive departments, and which involve controllers and those controlled, with the goal of creating clearer control criteria, balancing the goals of management with those of democratic supervision; 3) Institutional strengthening and the qualification of the workforces of ministries' internal control advisors, by setting up collective structures that combine the ministry's sectorial expertise with expertise in supervision and auditing; 4) Creation of mechanisms to foster integration between the CGU and the ministries so that the results of internal audits provide feedback for the planning process.Control must regain its role of auditor of government programs, helping the ministries improve their public management, and the administration organs need to qualify and gear themselves up for using the results of audits as one more instrument working for public management.
5) Creation of arbitrage chambers between control and management organs.Institutional mechanisms for solution of doubts and questions between managers and auditors arising during inspections and audits are still practically non-existent.In other words there are no means for solving an irregular situation.Suggestions from those interviewed indicate that the control organs could have a legal instrument, similar to the term of conduct adjustment, which would provide an opportunity to correct faults before having to register the irregularity.In the current situation, the supervisor is unable to escape from the position of non-cooperation or just saying "nothing can be done": the controller has to register the irregularity (under pain of shared responsibility), and there is no instrument allowing the manager to correct the irregularity before registering it.What is more, there is a legal problem of how to attribute to the auditor discretionary power over which problems should be considered "summarily" and over which he should and could allow the manager to carry out corrections.
6) The arbitrage chambers could include among their attributes the problem of identifying responsibility for irregularities.As the TCU requires that the CGU indicate those potentially responsible for registered irregularities, but the Controllership Department has no instrument with which to make this identification with precision at this stage of the inspection, it ends up appointing the head of the managing institution or the whole chain of decision-makers, which will probably do someone an injustice.
The development stage of the control system in Brazil allows us to state that it is aligned with recent trends as regards the institutional advance of control bodies and in its direction taken toward promoting democracy and good public service performance.These advances are undeniable in Brazil's recent democracy, but this process is incremental and marked by corrections along the way.
The multiplicity and diversity of control organs and actions are features of our institutional system, directly related to the principle of mutual controls between the Powers inherent to a presidential regime.These features are not in themselves negative attributes, even though they cause difficulties when it comes to coordinating.On the contrary, the diversity of institutions is the result of the demands of checks and balances and, at the same time, one of the mechanisms for achieving a balance between the Powers.
The plurality of control bodies does not constitute a failure of Brazil's system of controls, but it does require the building of forms of coordination between its activities and institutions and the coordination of control practices with the objectives of management.Without these coordinating instruments, institutional fragmentation will