CHALLENGES IN COMPARATIVE STUDIES OF PUBLIC SECURITY FUNDING IN BRAZILIAN STATES

ABSTRACT

This article examines public security funding in five Brazilian states: São Paulo, Rio de Janeiro, Ceará, Pará, and Rio Grande do Sul. The study presents the budget structures of these states and aims to understand variations in public security spending, as well as the feasibility of comparing such information across the states. The results reveal that public security expenditures are structured differently, reflecting unequal and complex governance systems where similar actors, notably the state police forces, hold distinct significance in each state. Furthermore, in addition to the disparities between states, the analyzed budgets vary within a state over the years. Consequently, comparing public security expenditures in Brazil becomes challenging, highlighting the need for standardized accounting practices among states and greater transparency in monitoring relevant actions to combat violence. Lastly, the findings suggest the potential for comparing the correlation between state expenditures and their outcomes, analyzing efficiency and effectiveness.

Keywords:
public security; financing; governance; comparison; effectiveness

RESUMO

Este artigo busca investigar o financiamento da segurança pública em cinco unidades federativas do Brasil: São Paulo, Rio de Janeiro, Ceará, Pará e Rio Grande do Sul. O objetivo principal é investigar a variação do gasto com segurança pública nesses estados, apresentando sua estrutura orçamentária, buscando responder se é possível, ou não, compará-la. O resultado mostra que os estados organizam suas despesas de segurança pública de maneiras muito distintas, reflexo de estruturas de governança muito diferentes e complexas e que envolvem pesos diferentes nos estados para atores similares, como as polícias militares. As diferenças nos orçamentos estaduais são verificadas não apenas na comparação entre os estados, mas também no mesmo território ao longo dos anos. Esse retrato contábil torna muito difícil a comparação de custos da segurança pública no Brasil, indicando a necessidade de padronização contábil entre os entes e maior transparência para acompanhamento de ações relevantes no combate à violência. Dessa forma, será possível comparar, com mais exatidão, os gastos realizados pelos estados com seus resultados e, consequentemente, sua eficácia e efetividade.

Palavras-chave:
segurança pública; financiamento; governança; comparação; efetividade

RESUMEN

Este artículo busca investigar el financiamiento de la seguridad pública en cinco unidades federativas de Brasil: São Paulo, Río de Janeiro, Ceará, Pará y Rio Grande do Sul. El objetivo principal es investigar la variación del gasto en seguridad pública en estos estados, presentando su estructura presupuestaria, buscando responder si es posible, o no, compararlo. El resultado muestra que los estados organizan sus gastos de seguridad pública de manera muy desigual, reflejando estructuras de gobernanza muy diferentes y complejas que involucran distintos pesos en los estados para actores similares, como la policía militar. Las diferencias en los presupuestos estatales se verifican no solo en la comparación entre los estados, sino también en el mismo territorio a lo largo de los años. Este panorama contable hace que sea muy difícil comparar los costos de la seguridad pública en Brasil, lo que indica la necesidad de estandarización contable entre las entidades y una mayor transparencia para monitorear las acciones relevantes para combatir la violencia. De esta manera, será posible comparar, con mayor precisión, los gastos realizados por los estados con sus resultados y, en consecuencia, su eficiencia y efectividad.

Palabras clave:
seguridad pública; financiamiento; gobernanza; comparación; efectividad

EXECUTIVE SUMMARY

It is not sufficient to merely know the amount of expenditure or investment the government has allocated to a particular public policy to understand and evaluate its impact. It is essential to assess the outcome of the actions enabled by this investment and to consider whether the same results could have been achieved more efficiently. This article delves into the security policies in Brazil, acknowledging the complexity of this topic due to the absence of regulation defining what constitutes public security expenditure, as well as the lack of a general rule for earmarking revenues.

In the institutional architecture of public security in Brazil, state governments primarily manage the investigative police (known as “Policia Civil”), the law enforcement department (state police, known as “Policia Militar”, the forensic department, and the fire brigades. State governments account for approximately 80% of the country’s public security expenses (Fórum Brasileiro de Segurança Pública, 2022). This is a significant amount compared to other spending areas under the state government’s responsibility. A study by the Solidarity Research Network (Rede de Pesquisa Solidária, 2022) revealed that security ranks as the fifth highest expenditure of Brazilian states when examining budgets since 2019, following social security, education, health, and social contributions. The challenges in financing security policies have recently multiplied with the approval of Complementary Law 194 on June 23, 2022, which caps the charge for state taxes similar to VAT (in Brazil, “ICMS”) on fuels at 17% (Peres et al., 2022). An analysis of state revenues shows a high dependence on indirect taxation, especially ICMS, as well as the insufficient redistributive capacity of the State Participation Fund (FPE) (federal transfers). This economic and tax heterogeneity results in inequality in the provision of public services by different Brazilian states. Given this structure, and considering that a significant portion of its tax resources is constitutionally earmarked to health and education, any tax reduction, such as the ICMS, represents a substantial loss of resources for sectoral policies, particularly those without constitutional protection, such as public security.

This study examines five Brazilian states - São Paulo (SP), Rio de Janeiro (RJ), Ceará (CE), Pará (PA), and Rio Grande do Sul (RS) - to enhance our understanding of the disparities in the structure of expenditures and the policies implemented in public security. These states exhibit varying behaviors in terms of revenue generation.

An analysis of spending by expense groups over the last decade reveals that most of the public security budget is allocated to police salaries and other current expenses, leaving a minimal margin for any form of investment. Therefore, when examining the proportion of expenditure committed based on the state agency (department, secretary, budgetary unit linked to public security) responsible for the expense, it becomes evident that the state police (responsible for law enforcement) is the unit with the highest expenditure. This is in line with the unit’s size and structure, given its role in overt policing. However, these expenses vary significantly among Brazilian states.

The necessity for a debate on governance in public security arises in this context. As discussed by Costa (2015, 2023), public security departments in Brazil exhibit varying governance capabilities, which are understood as the ability of these institutions to formulate and coordinate public policies in their respective fields. These differences are most apparent in terms of the departments’ organizational structure, personnel, staff training, and their ability to coordinate with other entities both within and outside the state.

Findings by Costa (2015) identify two elements that are directly related to the variations in spending highlighted in this study: the organizational structure of the departments and the profile of their human resources. In various states, the secretariats exhibit diverse organizational structures pertaining to their responsibilities in areas such as the fire brigades, prison system, civil defense, social defense, justice, and forensics. Additionally, associated structures, such as police offices and ombudsman offices, fall under their management.

There are instances, for example, where all of these structures might be managed by the state secretary of public security, while in other cases, they are distributed among various agencies or may not even exist. As for the profile of human resources, states have the autonomy to implement governance models that establish diverse methods of hiring and structuring security careers, which will undoubtedly impact expenses.

The international literature points to at least four theoretical models (nodal governance, anchored pluralism, responsive regulation, and network theory) related to governance in the provision of public security. These models help us understand the dynamics between the actors involved (we will detail them later in this article). From our point of view, the multiplicity of active actors directly interferes with how security policies are offered, which, in turn, directly impacts the budget preparation and execution.

We begin with the understanding that different governance arrangements will result in distinct budgets. This assumption complicates the measurement and comparison of the amount spent and the allocation of expenses in public security across different territories. With this in mind, the importance of standardizing public security spending is underscored. This is not to diminish the autonomy of the states, but rather to translate spending into a ‘common language’ that allows for comparison and assessment of effectiveness.

This article is structured into five sections in addition to this executive summary. The subsequent section discusses the problem-situation concerning financing public security policies in Brazilian states. Section three arranges the principal ideas and theoretical concepts related to public security governance. Section four details the methodological framework of this research, which is then analyzed in section five. Lastly, the sixth section presents the main conclusions and recommendations.

PROBLEM-SITUATION

An analysis of the states’ revenues in recent years aids in understanding the challenges of financing public security, as they are responsible for most funding for policies in this area, one of the most sensitive of public management. According to the Brazilian Constitution, the states and the Federal District manage and maintain the police forces (investigative and law enforcement), accounting for 80.9% of the total security expenses, representing 1.6% of the national GDP. In 2021 alone, approximately BRL 105 billion was committed to security by the federal government, states/Federal District, and municipalities, a substantial amount. For comparison, the total public spending on health is around 3% of GDP.

The contention between federal, state, and local governments over the Brazilian tax burden has been a persistent theme in the country’s history. This issue has permeated the evolution of Brazil’s federal constitutions from the period preceding the Proclamation of the Republic (Lopreato, 2018) to the most recent constitution promulgated in 1988. This latest constitution marked a new phase of adjustment among the federation units during the period known as re-democratization. During this time, elected leaders at the state level were empowered and pursued political, tax, and fiscal decentralization.

The new constitution allowed these transformations throughout the 1990s. At the same time, the country went through a monetary stabilization program and faced a significant economic crisis that fragilized the economy of the states, which lost space in the correlation of political forces. In this context, the federal government used constitutional provisions to redesign intergovernmental relations (Almeida, 2005; Arretche, 2005; Peres & Santos, 2020). This movement was marked by increasing the federal tax burden with social contributions not shared with states and municipalities, reducing the pace of financial decentralization, and penalizing the states in particular.

Significant vertical tax inequality forms the backdrop of this dispute among different levels of government. Historically, Brazil has experienced a situation of horizontal inequality, with economic development concentrated in certain states in the South and Southeast and a greater dependence on federal transfers in the states of the North and Northeast. It’s worth noting that although the Southeast Region has been the main collector of ICMS for decades, this trend shifted from the 1970s and 1980s, with a decrease in this proportion accompanied by an increase in the North, Northeast, and Central-West regions.

It is not accurate to assume that there was a convergence in the collection of this tax among the regions, but it is important to realize that this dynamic altered the federative dispute and the fiscal war between states. The change in ICMS collection in the Northeast, Central-West, and North in the last decade was significant and led to great improvement. However, the discrepancy in the absolute value of the collection in the Southeast in relation to the other regions is still outstanding, and it is not compensated by federal transfers via State Participation Fund (FPE) (Afonso, 2009, 2016; Lopreato, 2020; Rezende, 2010).

As Peres and Santos (2021) analyzed, this relative growth of other regions compared to the Southeast has led to a deconcentration of revenues across the national territory. This shift was largely due to the regulatory changes that the ICMS underwent, as Afonso (2016) noted. These changes resulted in an increase in the collection by the tax authorities of these states, concurrent with a reduction in both production and collection in the Southeast. However, these regulatory changes were insufficient to ensure revenue equalization between states.

Thus, despite the existence of fiscal federalism mechanisms, such as the FPE, which aim to balance these differences in economic potential, this calculation model has never resulted in fiscal equalization (Rezende, 2010).

While the FPE does provide some compensation to states with lower tax revenues, it is unable to equalize revenues, thus maintaining a scenario of vastly differing economic potentials for the provision of public policies. Another important point to note regarding state revenues is that a significant portion of the collected volume is earmarked for education and health, following the 1988 Constitution provision that mandates the allocation of 25% of tax revenues for education and 12% for health at the state level.

While the constitution does not specify a certain percentage of revenues to be earmarked for public security, as it does for education and health, states make a significant fiscal effort to fund these services. Some of them allocate more than 10% of their current revenues to security expenses, as revealed by the results of this study.

Given this complex tax scenario for financing public security by the states, this article aims to highlight the importance of spending on public security in the states. It does so by demonstrating the percentage of total budgets spent on security to understand how these expenses are allocated. The guiding research questions are: i) How do public security expenditures vary among Brazilian states?; ii) Is it possible to compare the volume of public security spending by Brazilian states, or are there distinct elements that differentiate the states?; and iii) How can we compare the effectiveness of different actions in combating violence if we cannot distinguish these spending patterns?

Before addressing these questions, it is crucial to explore the theoretical concepts of governance used in the context of public security and the relevant elements in different governance patterns.

THEORETICAL FOUNDATION: THE DEBATE ON GOVERNANCE IN PUBLIC SECURITY

The debate surrounding governance in the provision of public security services has been a focal point following global changes since the 1990s, particularly with the end of the Cold War. Democracies have increasingly diversified the provision of policing across various agencies and agents, a task that was long considered the exclusive domain of the sovereign state (Loader, 2000).

In this context, governance can be defined as the coordination of political actors who may not necessarily have hierarchical relationships with each other but are interdependent. Their challenge lies in establishing practices and consensus capable of planning, articulating, and implementing public actions (Costa, 2023). In the realm of public security, the state continues to be the fundamental structure for coordinating and supervising the policing of public institutions. However, private policing, transnational police arrangements, security technology companies, and even organized crime are increasingly taking the lead as part of a growingly plural network that influences the implementation of public policies.

This new moment seems to distance us from the Weberian concept of a state monopoly on legitimate violence, where the police would be the force capable of guaranteeing and reinforcing the state’s sovereignty over a territory. Today, the regulation of social life is far more complex and based on a diversity of arrangements and actors, implying that the accountability and coordination of the system need modernization.

Against this backdrop, many authors have argued that the complexity of the current context necessitates new forms of system accountability and coordination, as the state’s traditional intraorganizational arrangements no longer adequately respond to the new institutional arrangement (Loader, 2000). At least four theoretical security governance models have been discussed in international literature: nodal governance, anchored pluralism, responsive regulation, and network theory. In all cases, the premise is a network of relationships between different actors in the provision of security services, but each model significantly differs in terms of the hierarchy of relationships among the actors and the role of the state.

The nodal governance model, which gained prominence following the work by Shearing (2005), presents the perspective of security as a service provided by polycentric networks, with the state not playing a major role in the coordination or regulatory process. Johnston and Shearing (2003) argue that security governance requires a new morphology, contradicting the traditional view that the state should take the lead. The normative justification for this is not the hollowing out of the state as a coordinating entity, but rather the strengthening of its capabilities to efficiently provide public services from a neoliberal perspective (Boutellier & Steden, 2011).

The network involved in providing public security would consist of a series of nodes, represented by institutions equipped with a set of technologies, mentalities, and resources. These nodes mobilize the knowledge and capacity of their members to influence a given scenario. From this perspective, less hierarchical relationships would facilitate interaction between different actors, enabling them to develop solutions for each problem they encounter, thereby enhancing their creative potential

Critics of this model emphasize that each of these actors has their own objectives and mentalities, which may not necessarily align with legal norms and could potentially deviate from them. For instance, how can one ensure that profit does not supersede the interests of the rule of law in the actions of private actors? From this viewpoint, it is unclear who is formulating and implementing the policy and what legal and symbolic tools are available to each of the actors in the network. Consequently, it is not possible to guarantee that the network operates through conflict mediation and consensus-building, i.e., in a democratic manner (Stenning, 2000).

The theoretical model of anchored pluralism is constructed in contrast to the nodal governance model, which removes the state as the leading and regulating entity in the relationships established within the security organizational arrangement. Anchored pluralism is deeply concerned with the democratic perspective. It raises the following question: In the absence of a regulatory state, who is responsible for monitoring safety nets and protecting the rights of the most vulnerable communities? Which entity or actor would serve as a last resort in defending public order?

This perspective was developed by Loader and Walker (2006) to advocate for the necessity of the state’s presence in security governance, utilizing the network perspective. Although they initially propose this concept in a normative sense, Loader and Walker also present valuable analytical insights to identify and evaluate the distinctive anchor points that states have in guiding plural organizational networks (Boutellier & Steden, 2011). According to the authors, the state remains a fundamental actor in the formation of social identities, as well as the only entity capable of precisely allocating collective resources. It is a legitimate actor in regulating security networks and creating deliberation channels so that police actions are shaped based on democratic participation. This approach enhances the population’s perception of the state’s legitimacy as a regulatory sphere, and the state becomes pivotal in this relational network among state agencies, the market, and civil society.

The responsive regulation model, advocated by John Braithwaite (2002), underscores the limitations of a transaction between the state and private companies. It argues that monetary influence will corrupt regulation unless one or more actors serve as a third party in the regulatory game.

Responsive regulation involves listening to multiple stakeholders and making a deliberative and flexible (responsive) choice from a range of regulatory strategies that can be conceptually organized into a pyramid. At the base of the pyramid, less coercive, less interventionist, and cheaper strategies are used more frequently. The premise is to try the least intrusive strategy in the pyramid until it fails. Then, its weaknesses can be addressed by the strengths of the strategies higher up the pyramid, which have a coercive effect and would motivate cooperation.

This implies that the effectiveness of order in security depends on the credible ability to escalate toward punitive and coercive measures. They can operate at their full capacity only when local security policies are systematically established and backed by a robust state in the background of nodal arrangements (Braithwaite, 2002).

From this theoretical perspective, governance arrangements function as an incremental “regulatory pyramid” with numerous preventive bodies (such as social services, youth assistance, and medical assistance) and fewer repressive and punitive institutions (like police and criminal justice). This pyramid comprises a hierarchy of regulatory tools designed to induce compliance, ranging from persuasion, stimuli, and incentives to sanctions (Boutellier & Steden, 2011).

In close dialogue with the theory of anchored pluralism, network governance theory posits that the state is the only entity capable of establishing binding guidelines to ensure democratic governance at the local level. Network governance emphasizes the democratic aspect that governance must uphold and alerts to the risks of rights violations in the absence of an entity protecting minority rights (Bogason & Musso, 2006; Stenning, 2009). The emergence of local security networks can represent both promises and threats to accountability mechanisms. Thus, while a less hierarchical model may seem positive, the constitutional authority manifested in the hierarchy continues to be a productive structure for relational networks to function democratically.

In the Brazilian context, regardless of the concept defended as ideal, many actors are involved and directly influence the delivery of security policies. Far from being a state monopoly, even if its centrality in the model is recognized, it is crucial to consider the emergence of new actors and expectations, such as the growing influence of private companies, civil society, and organized crime, represented by factions and militias.

Without ignoring this reality, our analysis focuses on the governance capacity of state actors in public security. The aim is to understand how the different actors at the state level, which make up the majority of the public security forces (including security departments, investigative and law enforcement police, forensic departments, fire brigades, and the prison system), interact and compete for resources.

METHODOLOGY

This research involved collecting state revenue and expenditure data from various sources. For the initial analytical layer, which covers the 26 states and the Federal District, we gathered information through the Brazilian Public Sector Accounting and Tax Information System (Siconfi) databases from the Ministry of Finance. We extracted data from 2013 to 2020 and compiled it into a single Microsoft Excel file with value adjustments made using the Broad National Consumer Price Index (IPCA) for January 2021.

We began by conducting a comprehensive analysis of budget revenues, which allowed us to understand the macroeconomic behavior of each state, considering taxes such as ICMS and other revenues transferred by the federal government. In this case, we also collected data on expenditures and expenses by function, focusing on detailing the public security function and its subfunctions to understand the patterns and allocation options in this sectoral policy for each Brazilian state. Two points must be highlighted here: 1) due to Siconfi standardization, only four subfunctions (general administration, policing, civil defense, and information and intelligence) of public security are identified, so any other subfunction is grouped into the “other subfunctions” category; 2) the data do not include intra-budgetary expenses (expenses that occur when entities that are part of the budget purchase materials, goods, and services, pay taxes, fees, and contributions, for instance, with the recipient of the resources also being an entity that is part of that same budget, the same level of government - federal, state, or local).

After collecting nationwide data from Siconfi, we continued by collecting specific information related to the revenues and expenses of the states selected for this study: CE, PA, RJ, RS, and SP. Here, we grouped and standardized budget codes. We also added a new field to the existing database, which provided a detailed revenue classification, allowing us to view the codes for the following taxes: ICMS, IPVA, and ITCD. Similarly, it was possible to view the State Participation Fund (FPE) quota codes and the codes referring to transfers, contributions, and quotas in the areas of social assistance, education, and health, respectively.

To detail expenses, we accessed the transparency portals of the selected states. We extracted data from 2012 to 2020 into a single database in Microsoft Excel files, with respective inflation adjustments (using the IPCA index), making all amounts represent the January 2021 values. To standardize and summarize terms and classifications, enabling temporal comparison between the selected states, we proceeded with the creation of three new fields in the database: i) summary of economic category and group of nature of expense; ii) summary of function and subfunction; iii) summary of departments, agencies, and budgetary units.

From the compilation of this information, we standardized the nomenclature of agencies, departments, and budgetary units, according to each one’s scope and activities, with the following classification: “prison administration,” “fire brigade and civil defense,” “Transit Department - Detran/DER,” “investigative police,” “state police,” “forensic department,” “Secretary of Public Security,” “other security and prison administration agencies.”

RESULTS AND ANALYSIS

The analysis of the states’ revenues (CE, PA, RJ, SP, and RS) reveals some interesting information regarding the behavior of tax collection and revenues from specific funds. It is essential to highlight three points about the revenues of the five states analyzed: i) the importance of ICMS as a source of revenue; ii) the relevance of the FPE, but its failure to guarantee equity of revenue between states; iii) the importance of sectoral education and health funds, for which there are constitutional binding rules.

Figure 1 presents a comparison of these revenues. Different behaviors can be seen between the states. While in SP and RS, the ICMS corresponds to 65.6% and 57.3% of the revenue collected, CE, PA, and RJ have an ICMS collection of less than 45% of total revenue. The FPE has a significant volume only in the states of CE and PA, corresponding to around 20% of revenues. In RJ, the item “other current revenues and transfers” is responsible for 48.3% of all revenue, a category in which oil royalties are recorded.

Figure 1
Comparison of the main items of current revenues of selected states (2020)

States are very dependent on indirect taxes, especially the ICMS, a consumption tax similar to the VAT, with the addition that the FPE is very relevant for CE and PA. With this structure, a large part of tax resources are linked to health and education. Other resources from federal transfers are allocated to specific funds for this area, such as Fundeb and transfers from the Ministry of Health for the national health system (SUS). Thus, most of the states’ current revenue is previously allocated to education and health. Even so, the amounts allocated to public security have been relevant, as shown in Figure 2. RJ is the state that, over time, allocates the most funds from its budget to the area, which largely reflects the permanent crisis that the state is experiencing in public security. Also noteworthy is the significant growth in expenditure that CE allocates to the public security function.

Figure 2
Percentage of expenses committed to public security related to total revenues

Figure 2 shows a fluctuation in the weight of these expenses in relation to revenues over time. However, in each state, there is an average level of expenditure that varies from 5.5% in RS to around 13% in RJ and CE. This level is quite high if we consider that, in addition to constitutional obligations with health and education, states have mandatory and significant expenses with debt charges, social security, and maintenance of the public sector, for example. Therefore, Figure 3 elucidates the importance of public security expenses in the states when comparing them with other priority spending areas.

Figure 3
Percentage of expenses committed in the five larger functions in relation to the total expenses committed in 2019

In CE and PA, for example, the areas of education, health, and public security have almost the same weight. In RJ, the weight of public security is greater than that of health and education. In SP, the weight of public security is slightly lower than health. Only in RS does this expense have a smaller proportion, but it is still quite relevant.

Another way to study public security expenses is to understand who is responsible for the expenditure. A preliminary analysis of the state expenditure base was carried out to identify the main bodies in proportion to committed expenditure, departments, and budgetary units linked to the public security function. Additionally, we conducted a preliminary analysis of each state’s organizational structure (organizational chart) to understand the main bodies, departments, and budgetary units linked to the public security function.

For the five states analyzed, the most considerable portion of spending on the public security function is carried out by the organizational structures of the military police. However, these expenses vary between states, ranging from 34.9% of total public security expenses in RS to 62.9% in SP. The same analysis can be made in relation to the amount allocated to the civil police. While in PA, 22.9% of committed expenses were allocated to the investigative police, in SP, this percentage was only 13.8%.

This classification by agency underscores the differences between the organizational structures of public security and the structuring of expenses in each state analyzed. As previously discussed, differences in governance are reflected in budget differences, making it challenging to determine how much is spent and how it is spent on public security in each territory. What seems evident, considering Table 1, is that even if we acknowledge that the implementation of public security policy and its governance involves various public actors - investigative police, state police, forensic departments, fire brigades, Detran, prison administration, and secretaries of public security - the state police appear to be the stronger agency and the one with more governance capacity from a budgetary perspective.

Table 1
Proportion of expenses committed in the area of public security, per agency - 2020

AGENCY	CE	PA	RJ	RS	SP
Prison Administration	17,1%	5,7%	9,0%	17,4%	13,3%
Fire Brigade and Civil Defense	5,5%	8,4%	15,9%	6,8%	0,4%
Transit Department - Detran / DER	1,5%	13,1%	6,6%	12,4%	0,0%
Other Security or Prison Administration Agencies	0,5%	0,2%	0,2%	0,0%	0,4%
Investigative Police	15,5%	22,9%	17,7%	18,1%	13,8%
State Police	54,0%	42,1%	50,6%	34,9%	62,9%
Forensic Department	2,6%	3,9%	0,0%	2,9%	2,4%
Secretary of Public Security	3,4%	3,6%	0,0%	7,4%	6,8%
Total	100,0%	100,0%	100,0%	100,0%	100,0%

Source: Elaborated by the authors based on data from state DBEs (Selected States).

On the one hand, this occurs because the state police are, on average, three times larger than the investigative police, considering the size of the force, which would naturally necessitate a larger share of the budget for them. However, this prominence of the state police is explained precisely by their strong influence in the political world, so the solutions related to which policies should be prioritized almost naturally involve the reinforcement of overt policing. In three of the five states analyzed (CE, RJ, and SP), the amount allocated for state police is more than 50% of the budget. The low value allocated to investigative police in most states is another element that draws attention. In SP, for example, only 13.8% of the budget is directed to these departments, which explains the institution’s 50% reduction in the last decade (FBSP, 2022). In PA, on the other hand, the investigative police budget accounts for 22% of total expenses, indicating a certain priority given to the agency.

Another important element to highlight from a governance perspective is the existence of the state secretary of public security or a similar agency in the state. This agency usually assumes a superior position in the hierarchy of the police forces. The most common model in Brazil is the existence of a secretary who serves as the head of the two police forces (investigative and state) and acts as an intermediary between them and the governor. In RJ, however, since 2019, the secretariat has been abolished so that the heads of the police forces have the status of secretary, reporting directly to the governor. This model offers even more autonomy to police forces, as the work of this agency normally requires some type of coordination between them. In the absence of this entity, and considering the fragmented model of Brazilian policing, it seems unlikely there will be daily coordinated work between these actors.

States make different allocation decisions - and have the autonomy to do so - when executing their expenses, opting to use various subfunctions and agencies. This finding directly implies that there are clear limitations in comparing the amount spent on the public security function since very different expenses between entities can be aggregated in this function. In addition to the different institutions aggregated in this function presented previously, the graph by subfunctions below also qualifies the variation in state spending choices and states’ allocative options.

This detailed view allows us to see how expenses for very different purposes are classified in the same large area called public security. In 2020, for example, the State of RJ classified spending on air transportation and road transportation as spending on public security, which did not occur in any other state in the series analyzed. The states of RJ, RS, and SP spend on hospital and outpatient care (most likely from the police, given that they have their own hospitals) and understand this as spending on public security, while in CE and PA, there are no expenditure records in this sub-area.

The use of two subfunctions, general administration and policing, is noteworthy. Although they appear in all states analyzed, the weight in each one is quite different. In the states of CE and RJ, for example, the general administration subfunction occupies almost all public security spending, while in SP, it represents only 4%. On the other hand, policing represents 53% of security expenses in RS and 38% in SP, while CE, PA, and RJ represent only 4%, 6%, and 9%, respectively. This does not mean that these last entities almost do not spend on policing, which is very unlikely, but rather that they interpret differently how the classification of such expenditure should be carried out - an understanding that is divergent in RS and SP, where these sub-functions are more expressive.

These different interpretations arise because, although functional classification is mandatory for all entities with standardized categories, the legislation does not define what exactly should be classified into which subfunction and function. This decision is made by whoever manages the budget at the local level. This finding underscores the importance of noting that despite information from nationally consolidated bases, with minimal standardization, comparing data is a challenge, as the understanding of expenditure classification is not uniform across all entities. These aspects, primarily the data and information in Table 1 and Figure 4, corroborate the discussion about how different governances, with multiple actors, imply different public policies. These policies, in turn, employ different budgetary strategies, thus introducing difficulties in the analysis and comparison between states. In this regard, we emphasize that the state still remains the key structure in the governance of actors in the security area. However, it is evident how the different state actors interact in the competition for budgetary resources, as observed from the expenditure commitment by the agencies in Table 1 and the provision of public security policies in the functional budget classifications shown in Figure 4.

Figure 4
Composition of expenses with public security in 2020, per sub-function

CONSIDERATIONS AND RECOMMENDATIONS OF PUBLIC POLICIES

Analyzing the public budget is crucial for understanding governments’ allocation decisions. The implementation of public policies - and politics itself - depends on tax and budgetary decisions. This article aims to demonstrate that public security is a priority political function in Brazilian states, with an expenditure exceeding BRL 100 billion, according to data from the latest Brazilian Yearbook of Public Security (2022). States finance 80% of the total spent in this sphere.

However, this expenditure varies significantly among states. Firstly, the economic potential, reflected in the level of tax collection and types of taxes, diverges greatly between regions, revealing territorial inequality. Furthermore, the organization of spending also differs considerably between states and even within the same state over time.

Thus, based on a sample study of five states (SP, RJ, PA, CE, and RS), we analyzed public security expenses in detail, crossing different budget classifications. As a result, we found that from the institutional organization of public security, which implies, among other things, a command arrangement over the military and civil police, to the structuring of funding expenses, there are substantial distinctions in spending among the states. The differences in governance, pointed out in Costa (2015, 2023) and also discussed based on theoretical models, are reflected in budget differences. This makes it challenging for us to know how much is spent and how it is spent on security in these states, complicating the comparison of expenses and the evaluation of their efficiency or effectiveness.

However, this analysis was only possible based on a breakdown of the expense bases of the sample of five states. Despite facilitating the global consolidation of expenses, national databases do not allow for this specific understanding. Therefore, we believe it is essential to consolidate more detailed data to better understand what is happening throughout the national territory regarding public security expenses.

Without this, coordinating the action of subnational entities in implementing their public security policy, and even at an informational level of production and publication of budgetary data, becomes difficult. While we have seen progress in the standardization of reporting and consistency of consolidated revenue data from subnational entities, it is clear that control over expenses is still fragile. Therefore, (re)formulating strategies for inducing accounting standardization among entities is a relevant and necessary possibility if we want to carry out comparisons that allow us to properly understand the expenditures made by different states and evaluate them in terms of their results.

ACKNOWLEDGEMENTS

We express our gratitude to the Brazilian Forum of Public Security and the Lafer Institute for their invaluable research support and funding.

Reviewer

Peer review report available in this link

• Evaluated through a double-anonymized peer review.

REFERÊNCIAS

• Afonso, J. R. R. (2009). FPE: Rateio sem custeio. Observatório da Jurisdição Constitucional, 1(3), 1-9. https://www.portaldeperiodicos.idp.edu.br/observatorio/article/view/386
  » https://www.portaldeperiodicos.idp.edu.br/observatorio/article/view/386
• Afonso, J. R. R. (2016). Federalismo fiscal brasileiro: Uma visão atualizada. Caderno Virtual, 1(34), 1-24. https://www.portaldeperiodicos.idp.edu.br/cadernovirtual/article/view/2727
  » https://www.portaldeperiodicos.idp.edu.br/cadernovirtual/article/view/2727
• Almeida, M. H. T. (2005). Recentralizando a federação? Revista de Sociologia e Política [online], 24, 29-40. https://doi.org/10.1590/S0104-44782005000100004
  » https://doi.org/10.1590/S0104-44782005000100004
• Arretche, M. (2005). Quem taxa e quem gasta: A barganha federativa na federação brasileira. Revista de Sociologia e Política [online], 24, 69-85. https://doi.org/10.1590/S010444782005000100006
  » https://doi.org/10.1590/S010444782005000100006
• Braithwaite, J. (2002). Rewards and regulation. Journal of Law and Society, 29(1), 12-26. https://doi.org/10.1111/1467-6478.00209
  » https://doi.org/10.1111/1467-6478.00209
• Bogason, P., & Musso, J. A. (2006). The democratic prospects of network governance. American Review of Public Administration, 36(1), 3-18. https://doi.org/10.1177/0275074005282581
  » https://doi.org/10.1177/0275074005282581
• Boutellier, H., & Steden, R. (2011). Governing nodal governance: The “anchoring” of local security networks. In A. Crawford (Ed.), International and Comparative Criminal Justice and Urban Governance: Convergence and Divergence in Global, National and Local Settings (pp. 461-482). Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511974953.018
  » https://doi.org/10.1017/CBO9780511974953.018
• Costa, A. T. M. (2015). Estado, governança e segurança pública no Brasil: Uma análise das secretarias estaduais de Segurança Pública. Dilemas - Revista de Estudos de Conflito e Controle Social, 8(4), 607-632. https://www.redalyc.org/pdf/5638/563865503001.pdf
  » https://www.redalyc.org/pdf/5638/563865503001.pdf
• Costa, A. T. M. (2023). Das políticas de segurança pública às políticas públicas de segurança. In: A. T. M. Costa (Org.), Segurança pública, redes e governança (pp. 22-28). Editora Universidade de Brasília.
• Fórum Brasileiro de Segurança Pública. (2022). Anuário Brasileiro de Segurança Pública, ano 16.
• Johnston, L., & Shearing, C. (2003). Governing security: Explorations in policing and justice Routledge.
• Loader, I. (2000). Plural policing and democratic governance. Social & Legal Studies, 9(3), 323-345. https://doi.org/10.1177/096466390000900301
  » https://doi.org/10.1177/096466390000900301
• Loader, I., & Walker, N. (2006). Necessary virtues: The legitimate place of the state in the production of security. In J. Wood & B. Dupont (Eds.), Democracy, society and the governance of security (pp. 165-195). Cambridge University Press.
• Lopreato, F. L. C. (2018). Governos estaduais: O retorno à debilidade financeira (Texto para discussão n. 338). Instituto de Economia, Unicamp. https://www.economia.unicamp.br/noticias/governos-estaduais-o-retorno-a-debilidade-financeira
  » https://www.economia.unicamp.br/noticias/governos-estaduais-o-retorno-a-debilidade-financeira
• Lopreato, F. L. C. (2020). Federalismo brasileiro: Origem, evolução e desafios. Economia e Sociedade [online], 31(1), 1-41. https://doi.org/10.1590/1982-3533.2022v31n1art01
  » https://doi.org/10.1590/1982-3533.2022v31n1art01
• Peres, U. D., Bueno, S., & Nascimento, T. (2022). Desafios no financiamento da segurança pública: Recursos estagnados e reforma do ICMS. 16º Anuário Brasileiro de Segurança Pública Fórum Brasileiro de Segurança Pública. https://forumseguranca.org.br/wp-content/uploads/2022/07/16-anuario-2022-desafios-no-financiamento-da-seguranca-publicarecursos-estagnados-e-reforma-do-icms.pdf
  » https://forumseguranca.org.br/wp-content/uploads/2022/07/16-anuario-2022-desafios-no-financiamento-da-seguranca-publicarecursos-estagnados-e-reforma-do-icms.pdf
• Peres, U. D., & Santos, F. P. (2020). Gasto público e desigualdade social: O orçamento do governo federal brasileiro entre 1995 e 2016. Revista Brasileira de Ciências Sociais [online], 35(103), 1-23. https://doi.org/10.1590/3510307/2020
  » https://doi.org/10.1590/3510307/2020
• Peres, U. D., & Santos, F. P. dos. (2021). Descoordenação e desigualdades federativas no Brasil com a Covid-19: Análise da situação fiscal dos estados em 2020. Cadernos Gestão Pública e Cidadania, 26(85), 1-23. https://doi.org/10.12660/cgpc.v26n85.83742
  » https://doi.org/10.12660/cgpc.v26n85.83742
• Rede de Pesquisa Solidária. (2022). Boletim No. 38 de 28 fevereiro de 2022.
• Rezende, F. (2010). Federalismo fiscal: Em busca de um novo modelo. In R. P. Oliveira & W. Santana (Org.), Educação e federalismo no Brasil: Combater as desigualdades, garantir a diversidade (Cap. 1, pp. 71-88). Unesco.
• Shearing, C. D. (2005). Nodal security. Police Quarterly, 8(1), 57-63. https://doi.org/10.1177/1098611104267327
  » https://doi.org/10.1177/1098611104267327
• Stenning, P. C. (2000). Powers and accountability of private police. European Journal on Criminal Policy and Research, 8(3), 325-352. https://doi.org/10.1023/A:1008729129953
  » https://doi.org/10.1023/A:1008729129953
• Stenning, P. C. (2009). Governance and accountability in a plural policing environment: The story so far. Policing, 3(1), 22-33. https://doi.org/10.1093/police/pan080
  » https://doi.org/10.1093/police/pan080

Edited by

Associate Editor: Fernando de Souza Coelho

Publication Dates

• Publication in this collection
  04 Mar 2024
• Date of issue
  2024

History

• Received
  07 Oct 2022
• Accepted
  20 Oct 2023