Lack of institutional work and organizational change failure in Brazilian local governments

Abstract

This article proposes a theoretical framework to explain the failure of the compulsory adoption of new practices in the financial management cycle in local governments. The empirical analysis applies the framework to the control systems implementation in Brazilian municipalities. The findings of a qualitative comparative analysis applied to seven local governments show that despite human capabilities and information systems being a necessary condition, they are insufficient to deeply implement such internal control practices. Institutional capital is a condition sine qua non for the adoption of new practices on financial management and control in Brazilian local governments. The adoption of such reforms depends on an active network of powerful actors involved in institutional work supporting the changes.

Keywords:
internal control; municipalities; institutional work; institutional capital; reforms

Resumo

Este artigo propõe um modelo para explicar falhas na adoção compulsória de novas práticas no ciclo de gestão financeira em municípios. Empiricamente, aplica-se o modelo à adoção de sistemas de controle interno (SCI) em municípios brasileiros. As evidências da análise qualitativa comparada (qualitative comparative analysis - QCA) em sete prefeituras mostram que, apesar de recursos humanos e de sistemas serem condição necessária, eles não são suficientes para a implementação das práticas de controle interno. O chamado capital institucional é uma condição sine qua non para a adoção de novas práticas de administração e controle financeiro em governos locais no Brasil. A adoção dessas reformas depende da atuação de uma rede de “atores com poder” que realizam trabalho institucional em prol das mudanças.

Palavras-chave:
controle interno; municípios; trabalho institucional; capital institucional; reformas

Resumen

Este artículo propone un modelo para explicar fallas en la adopción obligatoria de nuevas prácticas en el ciclo de gestión financiera en municipios. Empíricamente, se aplica el modelo a la adopción de los sistemas de control interno en municipios brasileños. Las evidencias del análisis comparativo cualitativo (qualitative comparative analysis - QCA) en siete municipalidades demuestran que la existencia de recursos humanos y de sistemas son una condición necesaria, pero no suficiente para la implementación de las prácticas del control interno. El capital institucional es una condición sine qua non para la adopción de nuevas prácticas de gestión y control financiero en gobiernos locales en Brasil. La adopción de esas reformas depende del apoyo de varios “actores con poder” que realizan trabajo institucional en pro de las reformas.

Palabras clave:
control interno; municipios; trabajo institucional; capital institucional; reformas.

1. INTRODUCTION

Public sector reforms fail when they are not implemented and do not create or change practices in government. The mandatory adoption of a new law-enforced practice across the country depends both on resources for process review, system suitability and training, and on overcoming resistance from various stakeholders (Andrews, 2013; Scott, 2014).

Resistance emerges when new practices associated with the reform conflict with individuals’ motivations and with how they understand, work and act in the organizations changing their practices (Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011). For organizational institutionalism, reflexive actors in leadership positions (institutional entrepreneurs) overcome such resistance working together with other individuals to clarify, persuade and coordinate support for new practices (Battilana, Leca, & Boxenbaum, 2009). Thus, these leaders are essential in the process of theorization of reform (Kraatz, 2009). Their activities are forms of institutional work (Lawrence & Suddaby, 2006) that are fundamental to the diffusion and institutionalization of new practices. As the concepts brought into the new practice challenge the current practices, greater is the necessity to undertake deep forms of institutional work, based on symbols and relations between actors (Hampel, Lawrence, & Tracey, 2017) to optimize institutional capital (Oliver, 1997).

This article proposes a model to explain the failures in reforms based on the compulsory adoption of new practices applied to local governments in Brazil. Due to the strong mayor model adopted in the country (Heinelt & Hlepas, 2006), where the leader is a mayor elected for a mandate and to whom political-administrative powers are invested, local governments are the ideal scenario to discuss the institutional and relational effect of “powerful actors” increasing institutional capital to adopt new practices of control and financial administration. In general, the literature attributes the non-effectiveness of organizational changes in Brazilian governments to the lack of resources or low coercion from the public sector audit organizations, named Courts of Accounts (Azevedo & Aquino, 2016; C. F. Cruz, Ferreira, Silva, & Macedo, 2012; Souza, 2016). This study proposes that, if the analysis does not take into consideration the institutional capital, the understanding of the failures and limitations regarding the implementation of reforms is restricted.

The model introduced by this research was applied to the adoption of internal control systems in Brazilian municipalities, which is compulsory since the provisions of Law 4320/1964. Although this demand has been reinforced legally and normatively in the last decades, these practices are not homogeneous throughout the country. They have also not contributed to the increase in governments’ transparency or performance control (Araújo, Libonati, Miranda, & Ramos, 2016; M. C. M. T. Cruz, Silva, & Spinelli, 2016).

A qualitative comparative analysis (QCA) was performed in a small sample design, with within-case analysis. The analysis was based on participant observation, followed by two rounds of interviews in six municipalities in the state of Minas Gerais and triangulation with secondary data.

The article presents in the next sections the theoretical model and the methodology, followed by its application in the adoption of internal control systems in the municipalities. Then, the academic and practical implications for public administration are discussed, followed by the final considerations.

2. THEORETICAL MODEL

Expanding administrative and control activities among other organizational processes, demands material resources. The lack of resources such as capacity building or access to systems poses limits to the implementation of these practices (Browne, 2010). However, organizations such as municipalities may reduce resources available for specific purposes when the administrative and control activities are not aligned with the interests of the organization’s leaders. Thus, even when resources are available the adoption of new practices would not occur (Oliver, 1997).

When an organizational change is required, the organization’s response pattern will depend on the internal dynamics of powerful actors who have the competence and authority to address or influence the issue (Lawrence, 2008; Washington, Boal, & Davis, 2008). The mayor is an example of a powerful actor in municipalities. Both these powerful actors and the other individuals in the organization will present favorable, indifferent or contrary opinions to the adoption of new practices. Such opinions depend on their motivations, understanding the changes, and projection of consequences (Lawrence & Suddaby, 2006).

External actors supporting new practices, such as the Court of Accounts (public sector audit organizations in Brazil), will expect that the organization complies with the regulation, based on their mandatory nature. As the organization and the powerful actors operating in it need to legitimize themselves in the organizational field (Meyer & Rowan, 1977; Suchman, 1995), there will be observed responses ranging from compliance to decoupling (Oliver, 1991).

The need to be legitimized emerges when the organization and the powerful actors operating in it are motivated to meet expectations from (and monitored by) powerful external actors who can threaten their positions - external sources of legitimation. Likewise, individuals with less power within the organization will seek support in the opinion of the internal powerful actors to justify their choices and actions, seeking internal sources of legitimation (Johnson, Dowd, & Ridgeway, 2006).

External sources of legitimation, such as Courts of Accounts, have legitimacy in society because they are seen as guardians of probity and legality. Thus, mayors, their appointed cabinet members (hereafter secretaries), and internal auditors will meet or pretend to meet the expectations of these external sources. When these actors and their accounts are considered as “approved” by the Court of Accounts, or “in agreement” with the policies of the Office of the Comptroller General of Brazil (CGU), the mayor and their staff of secretaries ‘borrow’ that legitimacy.

On the other hand, these powerful actors may be sources of internal legitimation. The opinion of the mayor about an initiative may be used to justify the decisions of secretaries, department directors, or civil servants. The same can be attributed to any actor who is recognized as an authority on a particular issue, either because of their specialized knowledge or formal competence (Johnson et al., 2006). Therefore, powerful actors can support the process of internalizing the expectations of external sources of legitimation to the organization, when the actors agree with the ideas and opinions expressed by the sources. Otherwise, the powerful actors become a legitimizing alternative for the other individuals inside the organization, who can then choose the source to justify their opinions.

The actors’ position will depend on how the new practices contradict the values ​​and the “way things are way things are,” which are implicit in the organizational routines. There is plenty of evidence of the resistance that emerges when a new institutional logic attempts to be deployed in a field dominated by conflicting logics (Greenwood et al., 2011; Thornton & Ocasio, 2008). The institutional logic that dominates an organization is an abstract structure that leads the cognition of individuals as a practical guide to action (Thornton & Ocasio, 2008). As a logic of action, it carries a set of values, meanings, and beliefs that are shared by all and taken for granted, and which are reflected in processes design, task routines and prioritization, how decisions are made, and in other interactions between individuals within the organization.

If the values and meanings carried out by the new practices diverge from pre-established dominant values, the organization and its powerful actors may have difficulty to understand the new logic and on accepting the associated values and may resist to change (Maitlis & Christianson, 2014). In these cases, these actors’ understanding must be expanded by theorization in favor of the new practices, clarifying their motivation and the “how to do.” A priori, external sources of legitimation would theorize the practices focusing on the powerful actors within organizations, who would help to implement the desired routines, via recurrent relational institutional work (Hampel et al., 2017).

Theorization, therefore, is a form of what has been called “institutional work” (Lawrence & Suddaby, 2006), which is the individuals’ effort to change the values, beliefs, and practices that make up standard social behavior. Actions that shape institutional work (be it symbolic, relational or material) include persuasion, negotiation of support, development of institutional frames and carriers (Lawrence & Suddaby, 2006). The symbolic and relational institutional work to support and promote new practices is deliberately conducted by “institutional entrepreneurs” (Battilana et al., 2009; Maguire, Hardy, & Lawrence, 2004). Because of their higher reflexivity, these actors can understand the changes and their consequences, as well as lead and influence the teams.

Therefore, when powerful actors are favorable to the new practices in a reform, they work (in a symbolic, relational or material way) as its institutional entrepreneurs. As they are an internal source of legitimation and use their power to influence teams with less power to change their position, the resistance is likely to be overcome by the combination of clarifying the reform and by coercion. However, if powerful actors show indifference, any other individuals who decide to support change, and engage in the institutional work, will have to seek outside support (Hampel et al., 2017). In this scenario, it will be unlikely that institutional work will be conducted by powerless individuals.

Support for change coming from powerful actors, internal or external to the organization, is a type of institutional capital (Oliver, 1997) complementary to resources capital (assets, strategy and technological capacity). Any source of legitimation to a reform can be used by individuals to justify their choices in line with the new practice. Likewise, the absence of such capital makes it difficult for teams alone to change if the practices contradict the organizational status quo defended by the dominant logic. Even when resources are available, they will be wasted since the practices will not be implemented (Oliver, 1997). On the other hand, if there is institutional capital, the scarcity of resources may be temporary since powerful actors can raise the necessary resources (Oliver, 1997).

It is possible to develop and accumulate institutional capital in favor of a new logic through the of external legitimation of new practices. The reallocation and combination of sources of institutional capital is a form of institutional work. The initial goal can be achieved depending on the degree of reflexivity of the actors performing the work, and unexpected consequences may be observed. For example, the continued theorization of the Court of Accounts in favor of fiscal limits accumulates institutional capital supporting the compliance to the Fiscal Responsibility Law. At same time a looser interpretation from the Courts of Accounts regarding the calculation of limits on costs with personnel expenditures, increases uncertainty and reduces this capital. Thus, internal institutional capital is accumulated, as (i) it receives support from powerful actors that serve as a source of external legitimation, and (ii) the ideas, values, concepts, and procedures of a new practice are implemented, accepted and settled within the organization. Figure 1 summarizes the discussion of this model.

Figure 1
Theoretical model about the response pattern in the adoption of government reforms

The success of a compulsory reform under the strong mayor model, such as the Brazilian one, is subject to the following challenge: in every electoral cycle new mayors and secretaries will take positions in the local administration with specific opinions regarding each of the reforms in progress. If they have an opposing or indifferent opinion, the institutional capital will be insufficient to mobilize the civil servants, either for the initial adoption or to sustain an incipient adoption process. The lack of support may be limited only to the absence of theorization - for example, the issue does not receive attention in the government’s narrative, and space in the agenda - or may be in the form of veiled boycotts and cut off resources.

Alternatively, if the process of adopting the reform is already underway and there is institutional capital distributed among several supporters around the organization, together with other institutional entrepreneurs who are close to external sources of legitimation, it would be costly for the mayor and their team to boycott the work in progress. In that case, they would be isolated regarding their contrary position, and the process would be protected against short-term opportunistic actions.

3. METHODOLOGY

The proposed model was applied to empirically observe the local government’s response pattern in the implementation of its internal control systems, and the conditions in which more favorable responses to such systems emerge. The research used the qualitative comparative analysis (QCA), commonly adopted in sociology and political science (Rihoux & Marx, 2013). Despite some criticisms (e.g., Lucas & Szatrowski, 2014), it is suitable for small samples, because of the use of deterministic causal logic. The QCA identifies and validates which causal conditions are “necessary,” “sufficient,” and explain an outcome (Mahoney, 2000; Thomann & Maggetti, 2017). The choice of relevant cases was conducted according to the theory that supports the variation of causal conditions and outcomes between the cases (Mahoney & Goertz, 2006).

The empirical analysis started with a participant observation carried out by one of the authors in a city hall (case 4). The choice for this particular local administration was based on the convenience of access. The initial case suggested the possible factors inhibiting the development of internal control systems at the local level. Referring to the notion of “institutional work” (Lawrence & Suddaby, 2006) a theoretical model was developed to explain the perpetuation of a ceremonial internal control. As expected in the QCA, the following cases were selected and compared based on the previous theory (Thomann & Maggetti, 2017). Thus, the other six cases were chosen from municipalities under the jurisdiction of the same Court of Accounts, in the state of Minas Gerais. The choice of cases in the same state allowed to control the effect of the same Court of Accounts, considering that there is a variation of performance and coercion from one court to another (Lino & Aquino, 2018). Also, the choice was based on theoretical variation in the internal control response pattern and the presence of resources capital. The QCA analysis calibrates the cases into non-diffuse sets on a nominal scale (0-1) (crisp-set QCA in Rihoux & Marx, 2013), according to the causal conditions (resources capital and institutional capital) and the outcome (posture adopted in internal control). Nominal scales are adequate when logic is deterministic (Mahoney, 2000). Ordinal scales would bring complexity and subjectivity in the observation of resource and institutional capital through interviews. The main advantage of crisp-set QCA is the within-case analysis via interviews (Rohlfing & Schneider, 2016). The two-round interviews increase internal validity by tracing alternative explanations (process tracing) to the “local government’s posture regarding internal control.” Such a procedure is possible and recommended to research designs adopting small samples (Thomann & Maggetti, 2017).

The first round of telephone interviews with those in charge of internal controls of municipalities aimed to collect information on their performance and the resources considered relevant. Noting that the resources held by the internal control departments were not sufficient for effective action, a new round of interviews was conducted, with a protocol focused on institutional capital. The new protocol was built after returning to the previous literature, a necessary practice in the process of theory-building - “back-and-forth between prior knowledge and cases,” as stated by Thomann and Maggetti (2017, p. 10). There were 16 interviews, with an average duration of 45 minutes. The interviews were not recorded to reduce the bias in the responses by the interviewees who might be concerned about how the statements could be used. They were, however, registered by the researchers in their notes, without endanger the purpose of the study.

The analysis and discussions present the main characteristics of the compared cases (Box 3). The lines “team resources,” “information resources,” “mayor’s support,” “controller’s support,” and “frequent contact with the court” are the causal contexts of the QCA, and “posture” is the outcome affected by the causal contexts. The causal context “presence of legal mandate” present in all cases was disregarded by parsimony. In cases where one of the components of the causal context was absent, but the others present, “1(-)” was attributed to point out the absence of that component.

The Boolean Truth Table of the QCA (Table 1) and the treatment of the limited diversity and logical reminders are presented in the analysis. Finally, the study relates the interaction dynamics of the three actors with the mechanism of causality (Rohlfing & Schneider, 2016) between causal contexts and the posture of the internal control system (ICS).

4. The application to the dynamic of the internal control in the Brazilian municipalities

The creation and maintenance of internal control systems (ICS) in all spheres of government in Brazil is required since Law 4320/1964, reinforced by the Federal Constitution of 1988, by the Fiscal Responsibility Law and by regulations of the Courts of Accounts. Despite initiatives of training and raising awareness such as events in the Courts of Accounts, manuals from the National Council for Internal Control (Conaci) and having the Office of the Comptroller General (CGU) as benchmark, local governments have made little progress in adopting internal control (Araújo et al., 2016; M. C. M. T. Cruz et al., 2016).

In the implementation of an ICS in local governments, the mayor and the chief controller are “powerful actors.” As is well documented in studies on Brazilian local governments, the dominant logic driving these actors is the budgetary logic (Azevedo, 2016), focusing on legality, whether in budget execution or in the procedures to contract materials, people and public infrastructure works. Also, local governments tend to function based on clientelistic relations (Nunes, 2010). In this institutional logic, which is “contrary to internal control,” there is little space for political actors and civil servants operating in the several city hall departments to accept interference and control from the ICS teams.

While a logic opposing ICS would consider the interruption of a process by internal control teams for an inquiry as an inconvenience, the favorable logic argues that it is inadequate and insecure the completion of a process without such inquiry.

The ICS’s posture was classified based on two typical responses by Oliver (1991): compliance or decoupling. An ICS that has reached active posture via compliance can exercise its mandate to validate the government’s accounts and reject the execution of acts that violate the legality. The response does not depend on the scope of control activities (such as accounting policies, budget execution, fiscal goals, government acts) and the amplitude (local governmental agencies audited by the ICS) of the audit work, which tends to grow with increasing resources and demand. Box 1 shows the main differences between the ICS’s posture based on the two typical responses.

Box 1
Differences between the characteristics of two ICS’s posture

In the ceremonial posture the internal control report included in the annual financial statement required by some Courts of Accounts, and the signature of the person responsible for the budget execution report (required by the Fiscal Responsibility Law) are considered as formalities. Thus, the ICS does not use these reports to oversight the civil servants and elected politicians.

The natural opinion of the ICS is aligned with the institutional logic contrary to control and tends to a ceremonial posture because of the resistance in accepting the logic of control. The resistance can be covert or exposed, as when the associations of municipalities appealed to the Brazilian Superior Court of Justice questioning the jurisdiction of the Office of the Comptroller General’s control over municipalities (Loureiro, Abrucio, Olivieri, & Teixeira, 2012). If the ICS is created only to comply with a provision from the Court of Accounts, with no real interest in making it operative, naturally follows the ceremonial posture.

For the emergence of another pattern of response, institutional work must be done to push local management toward a logic that promote control. In general, the Court of Accounts, the primary external source of legitimation who favors internal control and has a high coercive power, is the one responsible for the theorization of new practices; as Arantes, Loureiro, Couto, and Teixeira (2010) say, “what the civil servants really fear is the audit from the Court of Accounts.” Such a role may vary in intensity in the way the Courts advocate for the implementation of internal controls and in the way they monitor and punish nonconformities (Lino, 2015; Lino & Aquino, 2018; Piccoli, Balestrin, & Rover, 2015). The City Councils, as a legislative branch, could also act by fostering a favorable logic to control. In practice, however, due to political coalitions and bargaining there is little independence of the legislative in relation to the executive at the municipal level (Couto & Abrucio, 1995).

The powerful actors in the organization would use the external sources of legitimation to theorize the new practices, at the same time to support, justify and value their position of influence. If internal sources of legitimation (including leaders and established practices, routines, and values) indicate that the ICS active posture is desired, the amplitude and scope of control activities have the potential to unfold. That is, the available resources capital - such as (i) clearly defined ICS mandate to act; (ii) technically qualified audit and supervision teams; (iii) adequate technology (software and hardware); and (iv) access to audit material (data and information) - would be converted into process improvement, greater frequency and scope of audit work, supervision in new government agencies and departments, and questioning authorities for improper acts. Precisely the institutional capital mentioned in Oliver (1997) is the necessary condition for an active ICS.

Any actor doing institutional work, favorable or opposing to internal control, whether able to lead significant changes, will follow the behavior of these three actors (mayor, controller and the Court of Accounts). The interaction between the attitude of these three actors (opposing or favorable to the ICS) and the dominant institutional logic sets up four interaction dynamics and two expected postures of the ICS presented in Box 2.

Box 2 Institutional pressures, interaction dynamics and posture adopted by the ICS

1. Dynamic A — The active ICS: The three actors are aligned with a logic “favorable to control.” The controller is legitimated both by the mayor and the Court of Accounts. There is widespread support from civil servants, even if they are immersed in a remaining logic that is “opposing to control.” The controller seeks the mayor’s support to change the status quo. There is institutional capital supporting the internal control teams. In these cases, either resources are already available, and the ICS is continuously developing in scope and amplitude, or resources are not available, and institutional entrepreneurs (mayor and controller) are seeking to bring resources to support the ICS. The expected posture is an active ICS, in which even with a natural shortage of resources there would be institutional work to generate non-ceremonial posture.

2. Dynamic B — The ICS constrained: The mayor opposes internal control; the chief controller counts only on the support of the Court of Accounts. The mayor and their secretaries are immersed in the logic “opposing to control.” Although the internal auditing team wants to act and have some resources available, they encounter political barriers (explicit or not) that impose a limit on their performance. The controller may seek to lead change in favor of control, but as subordinate to the mayor, the initiatives may be interrupted. Because the ICS is recognized as “something undesirable,” existing resources are not fully utilized, and the expected response is a ceremonial ICS, with no incentives to institutionalization.

3. Dynamic C - The ICS under pressure: If the teams and the supervision of the chief controller are aligned to the dominant logic “opposing to control,” and lack the motivation to take on the responsibilities and proactivity that the logic of control demands, they will prefer a ceremonial posture. However, the mayor may have the interest of developing internal control by following the Court of Accounts and may use their authority to change the controller position. If this is not possible, there would be constant tension with the mayor and wasted resources. The expected response is decoupling.

4. Dynamic D - The protected ICS: Mayor and secretaries opposing the control, in an organization where the dominant logic is also “opposed to control,” create a condition of isolation in relation to the Court of Accounts. Thus, the whole organization is protected from the monitoring efforts of the Court of Accounts and definitively the outcome would be a ceremonial posture. The ICS would look for legitimation from the court by fulfilling the minimum legal requirements (e.g., production and submission of a report with minimal content, legalistic auditing and with no interference in the day-to-day dynamics of city hall). Internally, it would be legitimized by the opinion of the mayor who does not accept an active control system.

In B, C and D dynamics the response is decoupling, distinguished by the reasons why the ICS controller and team do not act. Respectively, because they “do not have the support,” “resist/do not want to,” and “do not need to.” To change the response from decoupling to compliance, an intense institutional work is necessary, and abundant resources do not imply in the optimized application. However, when there are not enough resources, institutional capital could be applied to obtain them. The role of the mayor is fundamental. Even if teams want to change their posture, they ultimately depend on political authority (Loureiro, Olivieri, & Martes, 2010). If the mayor does not legitimize the expansion of the scope and amplitude of the ICS, other actors in that government such as secretaries and directors of municipal agencies can stop the internal control following the mayor.

Box 3
Causal conditions and outcome of the internal control in the analyzied cases cases.

5. ANALYSIS OF THE PATTERNS OBSERVED IN THE COMPARED MUNICIPALITIES

5.1 Analysis of sufficiency and necessity

Table 1 shows the association between the resource capital (team and information) and institutional capital (support of the mayor and controller, and frequency of interaction with the Court of Accounts) with the consequent posture of the internal control (active or ceremonial). This study rejects the usual explanation that the existence of resources would be the leading cause for the occurrence of an active control system. Capital resource conditions are necessary, but not sufficient. While cases 4 and 5 validate that the existence of teams and access to information are not sufficient conditions, cases 6 and 7 validate the necessary condition. The presence of institutional capital in cases 1 and 2 boosts the use of team resources and access to information. Case 3 would represent a temporary situation of non-compliance, where the work to create institutional capital and mobilize resources for the ICS would be underway.

Table 1
Truth table of the QCA

Lines	Conditions (Causal Contexts)					Outcome	Cases observed via interviews
	Resource team	Resource information	Support from mayor	Support from controller	Suport from Court of Accounts	Posture
1	1	1	1	1	1	1	1
1.a	1	1(-)	1	1	1	1(-)	2
2	1	0	1	1	0	0	3
3	1	1	0	0	0	δ	-
3.a	1(-)	1(-)	0	0	0	0	4
3.b	1	1(-)	0	0	0	0	5
4	0	0	0	0	0	0	6 e 7
5	1	0	1	1	1	α	-
6	0	1	1	1	1	α	-
7	0	0	1	1	1	α	-
8	0	0	1	0	1	α	-
9	1	1	0	1	1	δ	-
10	1	1	0	1	0	δ	-

• Source: Elaborated by the authors. Notes: The lines of the truth table of the QCA are possible combinations between the causal conditions. Lines “1.a”, “3.a” and “3.b” are the lines “1” and “3” detailed, respectively. The table has 32 lines. Lines 5 to 10 are logical reminders, combinations that must be analyzed in relation to the limited diversity. Lines 11-32, omitted here, are logical reminders of potential situations of lack of resources. We did not consider, per parsimony, the lines that do not contribute to the analysis (Schneider and Wagemann, 2013). The following reminders are discussed: α — unstable situation, the mayor usually changes the controller and hire teams or decides to limit the access to the information systems; δ — possible situations, but without the mayor’s support, the probability is that changes will not occur.

The causal logic of the QCA, typical of the set theory, seeks the sufficient and necessary conditions for the occurrence of the phenomenon. External validity is threatened with a lack of cases for potential situations (combination of causal contexts and outcomes), or limited diversity (Rihoux & Marx, 2013). First, there were no cases of presence of institutional capital and absence of resources capital (lines 5, 6 and 7 of Table 1). In these cases, the potential outcome is a temporary situation of non-compliance (α) but could be solved with the mobilization of resources by the actors who support the practice, in a similar way to case 3. Second, there were no observed cases in which only the controller acts on a logic opposing the control. Again, it would be a temporary situation, in which the mayor should replace the controller (line 8). Third, there were no observed cases without mayor’s support and with the support from the controller (lines 9 and 10), with a potential ceremonial outcome (δ), despite attempts by the chief controller to change the situation. These unobserved situations do not invalidate the proposition that material resources are necessary, but they are not sufficient.

5.2 Moderating factors of dynamics

The natural response observed in ICS is decoupling, as shown in Box 3. Process tracing analysis highlights the mechanisms that link causal contexts with outcomes (Rohlfing & Schneider, 2016). Specifically, we analyzed how civil servants, mayor, controller and Court of Accounts interact to the point of transforming institutional capital and use resources for compliance or decoupling.

The mechanism presented in the proposed model can be interpreted as follows: given the existence of resources, the mayor and controller will seek legitimation with the Court of Accounts (public sector audit organization) and they will do institutional work to change the ICS posture and strengthen the favorable logic of control, which will develop an active posture (dynamics A).

When there are resources, institutional capital expands or reduces their use. Case 1 shows institutional capital expanding the use of resources, and in cases 4 and 5 (ceremonial), the mayor’s opposing position leads him or her to reduce institutional capital and the use of available resources. In cases 4 and 5 the mayor is distant and has little interaction with the ICS. Particularly in case 4, the mayor appointed the controller on the bases of political bargaining, and they do not meet even though they occupy next door offices. The mayor does not respond to requests from the ICS team. An emblematic event in case 4 shows the mayor interfering in the composition of the control teams. The internal auditor (accountant of the ICS), well-known by their outstanding technical competence, was changed by the auditee (an accountant in the finance department), and in the exchange of functions, the ICS lost the capacity to audit the municipality’s secretary of finance.

In this event, the symbolism of “least preference” is strong. Red flag reports issued by the internal auditors that were ignored by the controller, with the excuse that they ‘did not want to disturb the mayor’ (who did not receive them), decreased the auditors’ motivation, evidenced by the participant observation. Institutional capital that was already low (basically, is restricted to the formal mandate in the law that established the ICS and to the internal auditor’s motivation alone) is even more reduced. In addition to the example of loss of team resources, case 4 also suggests that the internal auditors’ teams under a non-tenured job position, or in which the civil servants have not been hired specifically to fill an auditor job position, are susceptible to the mayor’s discretionary decision to reduce or replace audit teams.

On the other hand, in case 1 (active) the institutional capital favorable to the ICS is present in the auditors’ reputation, in the design of the local regulation that protects the ICS, in the shared understanding that the ICS must be respected, and in the constant dissemination and support on the idea of the ICS’ relevance and benefits.

Local regulation, in addition to creating the ICS (as in other cases), also establishes job positions in a specific career of internal control, which are well remunerated. The calls for application to fill such positions are, therefore, the subject of great competition. This recruitment process protects teams from political interference and attracts skilled professionals. There is a recursive relationship between trained, well paid, and autonomous teams, and the respect and fear for their technical capacity attached to their status in the organization. The ICS can be considered an asset for the local government in this case. In 2015, the city hall celebrated the 10-years of the creation of the ICS, and disseminated the advantages the control practices brought to the local administration. The ICS team also acts reinforcing the relevance of control function (another example of institutional theorization), offering courses about control to the civil servants operating in other areas, internally consolidating and legitimating a position of management support. One future mayor would need to spend significant political capital internally to the administration to go against the ICS.

Finally, the absence of resources is a consequence of the low institutional capital, insufficient for the development of the ICS. In these cases, even if all the ICSs have been created by municipal laws, considering the existing low institutional capital and lack of support from the mayor and the controller, the ICSs will have insufficient resources. In cases 6 and 7 (ceremonial) there is no personnel allocated to control functions, nor access to necessary information. In these two cases, the ICS is a normative fiction.

In case 3 (ceremonial) the ICS finds resistance from other secretaries and civil servants, even counting on mayor’s support. The suggestions for improvement of management processes and requests posed by the ICS are not “taken seriously.” Low reputational capital is related to the inability of the mayor to play the role of institutional entrepreneur, either because of the lack of knowledge on the subject or because of the little capacity to mobilize their power relations in the local administration to promote change.

Case 2 tends to be more active, but still has restrictions. It has a smaller team than case 1, although it presents access to information and acts in the validation of the accounts (which gives it the status of active), it has a reduced scope since it does not include auditing the municipalities nor conducts accounting audit. As in case 1, case 2 has frequent interaction with external sources of legitimation such as the Court of Accounts, the Office of the Comptroller General, the Public Prosecutor’s Office, and the Brazilian National Treasury. The presence of auditors of the Court of Accounts is not rare in events promoted by the mayor of case 1, events that often count on the support from the School of Tax Administration (ESAF) (a federal agency dedicated to disseminate best practices on public finance and to coordinate national debates on technologies for tax collection). This closeness ends up protecting the ICS: if a new mayor wishes to hinder the ICS performance, they will have a political cost in reducing the scope and the amplitude of its work.

6. CONCLUDING REMARKS

The analysis pointed out that resources are necessary, but not sufficient for the adoption of new practices proposed in compulsory reforms in local governments. Our reflections observe three aspects:

1. First, that powerful actors must work to continuously accumulate and preserve institutional capital to support new practices for a sufficient time so such practices will become embedded in the organization.

2. Second, external and internal sources of legitimation operate as fuel for relational institutional work promoting the accumulation of institutional capital.

3. Third, institutional capital favorable to new practices helps institutional work throughout the change process, but it also protects the already implemented stages of reform from attacks by actors wishing its disruption.

Some implications of these findings for the Brazilian public sector are worth mentioning. The Courts of Accounts and the Office of the Comptroller General of Brazil (CGU) are the primary promoters of internal control and other numerous reforms in the public financial management cycle. When considering the institutional capital and their strategies, these two stakeholders could intentionally strengthen relations among: (i) individuals to be engaged in the new institutional logic pro-control, (ii) the powerful actors, and (iii) the sources of legitimation.

First, the use of management indices or indices of compliance with norms - such as the Escala Brasil Transparente (transparency index) or Índice de Efetividade da Gestão Municipal (municipal management efficiency index) - should cover beyond the resources capital. It should proxy the institutional capital and the mechanisms in place which allow the accumulation of institutional capital (such as the above-mentioned example of a specific call for application to fill job positions on control-related functions in local governments). These indices, when they detect resources capital based on self-declaration, are often limited and in effect encourage a ceremonial ICS. Second, the organizations measuring such indices should work together with the mayors given their importance in the process. Mayors could be involved in the process of dissemination by the external source of legitimation, leading activities of training, lectures and producing materials to publicize successful cases. Third, it is crucial to invest in the creation of symbolic material such as prizes for good initiatives and success cases of reform implementation, which would block disruption by new elected mayors. Fourth, the development of other external sources of legitimation, and the involvement of these actors in the process of relational institutional work, such as the National Council for Internal Control (Conaci) in the case of internal control, which would support civil servants working on ICS, who sometimes find themselves without direct support to carry out their jobs.

Even though local governments are currently at different levels regarding the development of reforms, and the regulatory pressure of the external sources of legitimation is different throughout Brazil, increasing institutional capital may favor the resources already allocated in the various departments of internal control in the municipalities turning them from ceremonial to active ICS.

REFERÊNCIAS

• Andrews, M. (2013). The limits of institutional reform in development New York, NY: Cambridge University Press.
• Arantes, R. B., Loureiro, M. R., Couto, C., & Teixeira, M. A. C.(2010). Controles democráticos sobre a administração pública no Brasil: Legislativo, tribunais de contas, Judiciário e Ministério Público. In F. L., Abrucio, M. R, Loureiro, & R. S.Pacheco (Orgs.), Burocracia e política no Brasil contemporâneo (pp. 109-147). Rio de Janeiro, RJ: Ed. FGV.
• Araújo, D. J. C., Libonati, J. J., Miranda, L. C., & Ramos, F. S. (2016). Unidades de controle interno dos municípios brasileiros: análise sob a ótica do Coso II. Revista Universo Contábil, 55(81), 39-58.
• Azevedo, R. R. (2016). Resistências ao planejamento de resultados e a lógica orçamentaria dominante em municípios (Tese de Doutorado). Universidade de São Paulo, Ribeirão Preto, SP.
• Azevedo, R. R., & Aquino, A. C. B. (2016). O planejamento em municípios de pequeno porte em São Paulo. Revista de Contabilidade e Organizações, 10(26), 63-76.
• Battilana, J., Leca, B., & Boxenbaum, E. (2009). How actors change institutions: towards a theory of institutional entrepreneurship. The Academy of Management Annals, 3(1), 65-107.
• Browne, E. (2010). Reforming budget systems: a practical guide New York, NY: United States Agency for International Development.
• Constituição Constituição da República Federativa do Brasil, de 5 de outubro de 1988 (1988). Brasília, DF.
• Couto, C. G., & Abrucio, F. L. (1995). Governando a cidade? A força e a fraqueza da Câmara Municipal. São Paulo em Perspectiva, 9(2), 57-65.
• Cruz, C. F., Ferreira, A. C. S., Silva, L. M., & Macedo, M. Á. S. (2012). Transparência da gestão pública municipal: um estudo a partir dos portais eletrônicos dos maiores municípios brasileiros. Revista de Administração Pública, 46(1), 153-176.
• Cruz, M. C. M. T., Silva, T. A. B., & Spinelli, M. V. (2016). O papel das controladorias locais no cumprimento da Lei de Acesso à Informação pelos municípios brasileiros. Cadernos EBAPE.BR, 14(3), 721-743.
• Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional complexity and organizational responses. The Academy of Management Annals, 5(1), 317-371.
• Hampel, C., Lawrence, T. B., & Tracey, P. (2017). Institutional work: taking stock and making it matter. In R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin-Andersson (Eds.), The SAGE handbook of organizational institutionalism (2a ed., pp. 558-590). London, England: SAGE.
• Heinelt, H., & Hlepas, N.-K. (2006). Typologies of local government systems. In H. Bäck, H. Heinelt, & A. Magnier (Eds.), The European mayor(pp. 21-42). Wiesbaden, Germany: VS Verlag.
• Johnson, C., Dowd, T. J., & Ridgeway, C. L. (2006). Legitimacy as a social process. Annual Review of Sociology, 32(1), 53-78.
• Kraatz, M. S. (2009). Leadership as institutional work: a bridge to the other side. In T. B. Lawrence, R. Suddaby, & B. Leca (Eds.), Institutional work: actors and agency in institutional studies of organizations (pp. 59-91). Cambridge, England: Cambridge University Press.
• Lawrence, T. B. (2008). Power, institutions and organizations. In R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin-Andersson (Eds.), The SAGE handbook of organizational institutionalism (pp. 170-198). Thousand Oaks, CA: SAGE.
• Lawrence, T. B, & Suddaby, R. (2006). Institutions and institutional work. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (2a ed., pp. 215-254). London, England: SAGE .
• Lei Lei n. 4.320, de 17 de março de 1964 (1964). Estatui Normas Gerais de Direito Financeiro para elaboração e contrôle dos orçamentos e balanços da União, dos Estados, dos Municípios e do Distrito Federal. Brasília, DF.
• Lino, A. F. (2015). Reforma da contabilidade pública e os tribunais de contas (Dissertação de Mestrado). Universidade de São Paulo, Ribeirão Preto, SP.
• Lino, A. F., & Aquino, A. C. B. (2018). Diversidade dos tribunais de contas na auditoria de governos locais. Revista Contabilidade & Finanças, 29(76), 26-40.
• Loureiro, M. R ., Abrucio, F. L., Olivieri, C., & Teixeira, M. A. C. (2012). Do controle interno ao controle social: a múltipla atuação da CGU na democracia brasileira. Cadernos Gestão Pública e Cidadania, 17(60), 54-67.
• Loureiro, M. R ., Olivieri, C., & Martes, A. C. (2010). Burocratas, partidos e grupos de interesse: o debate sobre política e burocracia no Brasil. In, M. R. Loureiro, F. L., Abrucio , & R. S. Pacheco (Orgs.), Burocracia e política no Brasil: desafios para o Estado democrático no século XXI (pp. 73-108). Rio de Janeiro, RJ: Ed. FGV .
• Lucas, S. R., & Szatrowski, A. (2014). Qualitative comparative analysis in critical perspective. Sociological Methodology, 44(1), 1-79.
• Maguire, S., Hardy, C., &Lawrence, T. B. (2004). Institutional entrepreneurship in emerging fields: HIV/Aids treatment advocacy in Canada. The Academy of Management Journal, 47(5), 657-679.
• Mahoney, J. (2000). Strategies of causal inference in Small-N analysis. Sociological Methods & Research, 28(4), 387-324.
• Mahoney, J., & Goertz, G. (2006). A tale of two cultures: contrasting quantitative and qualitative research. Political Analysis, 14(3), 227-249.
• Maitlis, S., & Christianson, M. (2014). Sensemaking in organizations: taking stock and moving forward. Academy of Management Annals, 8(1), 57-125.
• Meyer; J. W., & Rowan, B. (1977). Institutionalized organizations: formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340-363.
• Nunes, E. O. (2010). A gramática política do Brasil. Clientelismo, corporativismo e insulamento burocrático. Rio de Janeiro, RJ: Garamond.
• Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16(1), 145-179.
• Oliver, C. (1997). Sustainable competitive advantage: combining institutional and resource-based views. Strategic Management Journal, 18(9), 697-713.
• Piccoli, M. R., Balestrin, R. S., & Rover, A. (2015). O controle interno municipal sob a ótica do controle externo: estudo de caso em três municípios da região do meio-oeste de Santa Catarina. Revista de Contabilidade da UFBA, 9(2), 72-90.
• Rihoux, B., & Marx, A. (2013). QCA 25 years after “the comparative method”: mapping, challenges, and innovations - mini-symposium. Political Research Quarterly, 66(1), 167-235.
• Rohlfing, I., & Schneider, C. Q. (2016). A unifying framework for causal analysis in set-theoretic multimethod research. Sociological Methods & Research, 47(1), 37-63.
• Schneider, C. Q., & Wagemann, C. (2013). Doing justice to logical remainders in QCA: moving beyond the standard analysis. Political Research Quarterly, 66(1), 211-220.
• Scott, W. R. (2014). Institutions and organizations. Ideas, interests, and identities(4a ed.). Thousand Oaks, CA: SAGE .
• Souza, K. M. (2016). Balanço patrimonial e demonstração das variações patrimoniais: uma análise de conformidade nos municípios brasileiros. Contabilometria: Brazilian Journal of Quantitative Methods Applied to Accounting, 3(1), 1-13.
• Suchman, M. C. (1995). Managing legitimacy: strategic and institutional approaches. Academy of Management Review 20(3), 571-610.
• Thomann, E., & Maggetti, M. (2017). Designing research with qualitative comparative analysis (QCA). Sociological Methods & Research, 66, 1-31.
• Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood , C. Oliver , R. Suddaby , & K. Sahlin-Andersson (Eds.), The SAGE handbook of organizational institutionalism (pp. 99-129). Thousand Oaks, CA: SAGE.
• Washington, M., Boal, K. B., & Davis, J. N. (2008). Institutional leadership: past, present, and future. In R. Greenwood , C. Oliver , R. Suddaby , & K. Sahlin-Andersson (Eds.), The SAGE handbook of organizational institutionalism (pp. 721-735). Thousand Oaks, CA: SAGE.

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

Publication Dates

• Publication in this collection
  25 Apr 2019
• Date of issue
  Mar-Apr 2019

History

• Received
  29 Dec 2017
• Accepted
  03 Aug 2018