Relationship between tax avoidance and managerial ability: the moderating role of risk management

ABSTRACT

The aim of this study was to assess the moderating role of corporate risk management in the relationship between managerial ability and tax avoidance. Previous studies have focused on the variation in the level of tax avoidance among companies through personal characteristics of managers, such as managerial ability, with mixed results. This study questions whether corporate risk management can explain these mixed results by influencing the relationship between tax avoidance and managerial ability, which provides advancement in the understanding of the determinants of tax avoidance. The study highlights the relevance of the risk management process in defining tolerance and appetite for tax risk, facilitating the capitalization on opportunities to reduce tax expenses with the consequent increase in organizational performance. It shows that the interaction between managerial ability and risk management has different effects on tax avoidance practices that generate permanent differences compared to practices that generate temporary differences, which can be useful for managers, auditors, investors, regulators, and other stakeholders interested in corporate tax avoidance. This study applied the regression analysis technique estimated by the Ordinary Least Squares (OLS) method on a sample of non-financial Brazilian listed companies for the period from 2016 to 2021. Tax avoidance was measured by the Total Book-Tax Differences (Total BTD) and its components, Temporary BTD and Permanent BTD. Managerial ability was measured according to Demerjian et al. (2012), and risk management, according to the Enterprise Risk Management Index (ERMI), proposed by Gordon et al. (2009). The results suggest that risk management improves internal and external communication and provides managers with information to make more efficient tax decisions. They also highlight that, in companies with better risk management processes, managers with higher managerial ability choose to engage only in more aggressive tax strategies, which result in permanent differences.

Keywords:
tax avoidance; managerial ability; risk management

RESUMO

O objetivo deste estudo foi avaliar o papel moderador da gestão de riscos corporativos na relação entre habilidade gerencial e agressividade tributária. Estudos anteriores se concentraram na dispersão no nível de agressividade tributária entre as empresas por meio de características pessoais dos gestores, tais como a habilidade gerencial, com resultados mistos. Este estudo questiona se a gestão de riscos corporativos pode explicar esses resultados mistos, influenciando a relação entre a agressividade tributária e a habilidade gerencial, o que oferece um avanço no conhecimento dos determinantes da agressividade tributária. O estudo evidencia a relevância do processo de gestão de riscos na delimitação da tolerância e do apetite ao risco fiscal, favorecendo o aproveitamento de oportunidades de redução da despesa tributária com o consequente aumento do desempenho organizacional. Ele demonstra que a interação entre a habilidade gerencial e a gestão de riscos exerce, sobre práticas de agressividade tributária que geram diferenças permanentes, efeitos distintos dos que exerce sobre práticas que geram diferenças temporárias, o que pode ser útil para gestores, auditores, investidores, reguladores e demais interessados na agressividade tributária corporativa. Este estudo aplicou a técnica de análise de regressão estimada pelo método dos Mínimos Quadrados Ordinários (MQO) sobre uma amostra de empresas brasileiras não financeiras listadas, no período de 2016 a 2021. Mensurou-se a agressividade tributária pela Total Book-Tax Differences (BTD Total) e por suas componentes, BTD Temporária e BTD Permanente. A habilidade gerencial foi mensurada conforme Demerjian et al. (2012), e a gestão de riscos, conforme o Enterprise Risk Management Index (ERMI), de Gordon et al. (2009). Os resultados sugerem que a gestão de riscos melhora a comunicação interna e externa e oferece aos gestores informações para tomar decisões tributárias mais eficientes. Evidenciam também que, em empresas com melhores processos de gestão de riscos, os gerentes com maior habilidade gerencial decidem engajar-se apenas em estratégias tributárias mais agressivas, as quais resultam em diferenças permanentes.

Palavras-chave:
agressividade tributária; habilidade gerencial; gestão de riscos

1 INTRODUCTION

The significant volume of taxes required by the State, across all three levels of government, creates incentives for companies to adopt strategies aimed at reducing their tax expenses. This is because taxes decrease the cash flows available for investments and dividend payments, making the reduction of tax expenses something desired by shareholders (Chen et al., 2010).

Hanlon and Heitzman (2010) use the term tax avoidance, translated as “agressividade tributária” in the Brazilian national literature, to refer to business activities and strategies aimed at reducing corporate taxes. These activities to reduce tax expenses are subject to different degrees of uncertainty, mainly due to difficulties in interpreting tax regulations.

Activities related to tax reduction can be positioned on a continuum ranging from legality to illegality, between which there is a broad gray area characterized by varying degrees of uncertainty and risk associated with the firm’s tax position (Lietz, 2013). The concept developed by Hanlon and Heitzman (2010), widely adopted in the literature, considers this entire spectrum, thus covering everything from clearly legal activities (tax avoidance) at one edge to those with a high probability of being understood as illegal by tax authorities (tax evasion) at the other.

Martinez (2017) recognizes that the terms “tax evasion” and “tax planning” are widely used, but often in an undefined manner. Nonetheless, although the author clearly places tax planning within the realm of legality, in concrete cases, the boundary between legal and illegal will always depend on the interpretation of tax legislation and the actions taken by the company. Given the difficulty of empirically distinguishing between tax avoidance and tax evasion (Wang et al., 2020), this study adopts the broad conception of Hanlon and Heitzman (2010) and considers tax avoidance as the result of any activities that reduce corporate tax expenses, regardless of the point on the continuum where the company’s tax position falls.

Previous studies have addressed the influence of managers’ personal characteristics and corporate governance mechanisms on tax avoidance (Wang et al., 2020; Wahab et al., 2017; Amri et al., 2023). This study differs from the others by focusing on a specific personal characteristic, managerial ability, and a specific governance mechanism, corporate risk management.

Managerial ability is a personal characteristic of managers that can increase the level of tax avoidance, since managers with higher ability can more efficiently transform the company’s resources into results through the reduction of tax expenses (Koester et al., 2017). However, Francis et al. (2022) introduced divergent results, suggesting that managerial ability decreases tax avoidance. These authors attributed their results to a greater aversion to the reputational costs of tax avoidance on the part of managers with higher ability.

Managerial ability affects the economic performance of the company and decisions regarding accounting estimates and disclosures (Demerjian et al., 2012, 2013; Huang et al., 2017; Lunardi et al., 2022; Moura et al., 2019). Managers with higher ability have a better understanding of the firm’s operating environment, manage the business more effectively, and take advantage of investment opportunities with a view to maximizing the company’s value (Demerjian et al., 2012; Lunardi et al., 2022). Studies tend to relate managerial style and ability with a greater understanding of financial and accounting issues, from which managers can adopt strategies that impact the level of tax expense (Dyreng et al., 2010; Francis et al., 2022; Koester et al., 2017).

This study contributes to the understanding of the phenomenon by proposing that corporate risk management moderates the relationship between managerial ability and tax avoidance. The results are expected to help to explain the discrepancies introduced in previous studies. By conditioning the adoption of more aggressive tax strategies on the risk limit deemed acceptable by shareholders (Beasley et al., 2021), risk management can influence tax avoidance, particularly through its interaction with managerial ability.

The adoption of tax planning is particularly controversial in Brazil, where it is only occasionally recognized as valid, after years of discussions in administrative courts and/or in the Judiciary Branch, given the difficulties in interpreting the legislation and tax-related facts (Martinez, 2017). In light of the uncertainties associated with tax decisions, managers may be compelled to make decisions different from those they would make in another scenario.

In a context of high uncertainty, the process related to Enterprise Risk Management (ERM) becomes relevant, as it allows the company to identify and measure, in an integrated manner, a wide range of risks (Gordon et al., 2009; Hoyt & Liebenberg, 2011; Naseem et al., 2020), including tax risks. By addressing the risks to which the company is subjected in a holistic way, ERM focuses attention on the interrelationships among operational, financial, regulatory, and tax risks (Eastman et al., 2024).

ERM considers not only risks but also opportunities. According to the Committee of Sponsoring Organizations of the Treadway Commission (COSO, 2004, 2017), risk is associated with the likelihood that an event will occur and negatively impact the achievement of goals, while opportunity is associated with the likelihood that an event will occur and positively influence the achievement of goals.

Regarding tax avoidance, business decisions generate risks associated with the possibility of detection and penalties by the tax authority, with negative impacts on the reputation of the firm and its managers. Simultaneously, they generate opportunities related to tax saving and, consequently, to the increase in economic performance and firm value.

ERM can assist in decision-making by providing procedures and techniques capable of identifying and assessing the risks and opportunities associated with each tax planning strategy introduced to managers. The assessment of risks and opportunities establishes the optimal level of tax avoidance specific to each company (Kovermann & Velte, 2019). Additionally, managers with greater managerial ability can make better use of the information generated by risk management.

Given this context, it is expected that the risk management process will be able to influence the relationship between managerial ability and the level of tax avoidance, which leads to the following research question: what is the moderating effect of corporate risk management on the relationship between managerial ability and tax avoidance? Thus, the aim of the study was to assess this effect.

The research was operationalized using the multiple linear regression technique, with coefficients estimated by the Ordinary Least Squares (OLS) method. In order to measure tax avoidance, three metrics already well-established in the literature were used, derived from the Book-Tax Differences (BTD) proxy: Total BTD, Temporary BTD, and Permanent BTD (Hanlon & Heitzman, 2010; Dunbar et al., 2010; Fonseca & Costa, 2017). The measurement of managerial ability was carried out based on the score proposed by Demerjian et al. (2012), and risk management was calculated according to the Enterprise Risk Management Index (ERMI), proposed by Gordon et al. (2009).

The results suggest that managerial ability is positively associated with tax avoidance, and that risk management mitigates this relationship; however, this is only the case for Total BTD and Temporary BTD. It was observed that managers with greater managerial ability, who also have better risk management processes, are more likely to retain only those tax strategies that result in permanent differences between the accounting base and the tax base, rather than those that generate temporary differences, when assessing the risk-benefit relationship of tax strategies.

The study contributes to the literature by highlighting the role of the interaction between managerial ability and risk management as a determinant of the level of tax avoidance. Additionally, it contributes by showing that this interaction has distinct effects on practices that generate permanent differences compared to those that generate temporary differences.

By clarifying the influence of corporate risk management on the ability of managers with higher ability to act on tax strategies, the study provides a relevant practical contribution. In this regard, managers, auditors, investors, regulators, and other stakeholders can focus their attention on corporate governance mechanisms in order to identify those that tend to have a lesser or greater influence on the level of tax avoidance. It also reveals evidence that corporate risk management has the potential to lead managers to follow the tax strategies established by corporate governance and aligned with the risk appetite preferred by the Board of Directors to the detriment of the individual interests of managers with higher ability.

2 BACKGROUND AND RESEARCH HYPOTHESES

2.1 Tax Avoidance

Tax avoidance consists of the behavior of avoiding or reducing corporate tax expenses and encompasses various practices aimed at this reduction. The concept involves from more conservative practices, where the legitimate organization of the company’s business reduces tax incidence, to more aggressive behaviors, which may be considered illegal and entail penalties (Martinez, 2017; Wang et al., 2020), forming what is referred to as the continuum of tax avoidance (Hanlon & Heitzman, 2010; Lietz, 2013). Between these two edges, there is a wide gray area, where the risk associated with the firm’s tax position increases as this position approaches the more aggressive edge.

In turn, the concept of tax planning involves only tax strategies considered legal, being a subset of tax avoidance (Lietz, 2013; Blouin, 2014). The tax planning strategies closest to the aggressive edge of the continuum are classified as abusive tax planning and are often based on weak legal arguments (Martinez, 2017; Lisowsky et al., 2013). As a rule, the recognition of the legality of abusive tax planning is uncertain, which increases the tax risk for companies (Martinez, 2017).

The literature has analyzed tax avoidance from the agency perspective, where risk-averse managers must determine a level of tax avoidance that meets the interests of risk-neutral shareholders (Kovermann & Velte, 2019; Wang et al., 2020; Wilde & Wilson, 2018). Specifically, Wilde and Wilson (2018) proposed a framework where the level of tax avoidance arises from the comparison between the benefits and the costs associated with it. As this decision is usually made by managers (Dyreng et al., 2010; Kovermann & Velte, 2019), the literature has focused on analyzing the influence of their personal traits on tax avoidance, such as military experience, narcissism, political orientation, and managerial ability, among others (Wilde & Wilson, 2018).

In the long run, aggressive tax strategies can result in undesirable non-tax costs, such as cash outflows to pay potential fines, high reputational costs, and increased cost of capital (Wilde & Wilson, 2018; Beasley et al., 2021). Managers compare the potential benefits of reducing their tax burden with the costs associated with this behavior, in order to choose an optimal level of tax avoidance, compatible with the shareholders’ risk appetite and tolerance (Kovermann & Velte, 2019).

Accordingly, Cook et al. (2017) showed that investors increase the cost of capital when companies exhibit extreme values of tax avoidance, whether very high or very low. This reinforces the understanding that, despite strong incentives to reduce taxes, there is an optimal level of tax avoidance, determined, among other factors, by tax and non-tax costs inherent to the adopted strategies (Dyreng et al., 2010; Kovermann & Velte, 2019; Wilde & Wilson, 2018).

In addition to the personal characteristics of managers, the literature has analyzed various organizational factors as determinants of tax avoidance, such as corporate strategy (Higgins et al., 2015), ownership structure (family and non-family firms) (Chen et al., 2010), executive compensation (Armstrong et al., 2012; Desai & Dharmapala, 2006), financial distress (Edwards et al., 2013), and internal control mechanisms (Richardson et al., 2013), including risk management (Masri et al., 2019; Eastman et al., 2024). More recently, the literature has associated tax avoidance with corporate social responsibility (Kovermann & Velte, 2021), digital transformation (Zhou et al., 2022), financial development (Allam et al., 2024), and tax risk (Drake et al., 2019; Brühne & Schanz, 2022; Saragih & Ali, 2023a).

Regarding tax risk, Martinez (2017) argues that it is a construct positively correlated with tax avoidance. Since different strategies entail different levels of tax risk, this study addressed separately strategies that generate temporary differences between accounting profit and taxable income, on one hand, and strategies that generate permanent differences, on the other.

As a rule, temporary differences only entail the deferral of the payment of Income Tax (IT) and the Social Contribution on Net Profit, since the reduction of tax incidence in the present is achieved at the expense of its increase in future periods. The situations that can generate temporary differences arise from estimates that alter accounting profit and generally do not change taxable income in the same period, such as depreciation rates different from those allowed for tax purposes, estimated losses on doubtful accounts, impairment adjustments, among others. In this study, temporary differences were captured as Temporary Book-Tax Differences (TBDT).

Conversely, permanent differences arise from situations that change accounting profit without affecting taxable income, or that change taxable income without affecting accounting profit. In both cases, there is no expectation of reversal in subsequent periods. Examples of the first type of situation include the non-deductibility of fines for non-tax infractions and expenses with royalties above the limit allowed by tax legislation, among others. As examples of the second type of situation, there are the tax amortization of goodwill, exclusions for investment grants, adjustments for transfer pricing, the distribution of interest on equity, and the taxation of the results of foreign subsidiaries, among others.

In general, the practices that constitute this second type of permanent differences are considered tax planning strategies that have high uncertainty and risk, being closer to the more aggressive edge of the continuum of tax avoidance (Castro, 2010; Fonseca & Costa, 2017; Martinez, 2017). In this study, permanent differences were captured by the Permanent Book-Tax Differences (PBTD).

2.2 Managerial Ability

Studies have highlighted that the manager is the one who defines the level of corporate tax avoidance (Desai & Dharmapala, 2006; Dyreng et al., 2010; Li et al., 2016; Francis et al., 2022). When making decisions with tax implications, the manager is subject to a series of contingencies, among which are those related to the characteristics of the firm itself, such as size and capital structure, among others.

As organizational factors do not fully explain the dispersion observed in levels of tax avoidance, an additional explanation can be found in the personal characteristics of managers (Dyreng et al., 2010). Managers with higher managerial ability produce higher quality accounting reports, exhibit greater persistence of earnings and accruals, and have fewer errors in estimating losses from doubtful accounts (Demerjian et al., 2013), increase the informativeness of earnings (Baik et al., 2020), and demonstrate greater accounting conservatism (Haider et al., 2021).

Koester et al. (2017) found a positive relationship between managerial ability and tax avoidance. Based on the managerial ability proxy developed by Demerjian et al. (2012), Koester et al. (2017) showed that the Cash ETR (Cash Effective Tax Rate) is reduced by 3.15% when comparing the lowest quintile with the highest quintile of managerial ability. The reduction in the effective tax rate is indicative of greater tax avoidance.

Koester et al. (2017) argue that the positive relationship between managerial ability and tax avoidance can be explained by three main factors. First, managers with higher ability have a greater understanding of the business and the company’s operational environment, which enables them to identify the best tax planning opportunities.

Secondly, managers with higher ability are efficient at reducing costs and emphasize cutting expenses that do not harm the company’s operations, such as tax expenses. In contrast, reducing the cost of raw materials, for example, may be associated with a decrease in the quality of these inputs and, consequently, the final product. Thirdly, managers with higher ability recognize the importance of redirecting resources saved from taxes towards investments that have the potential to generate positive returns for the company.

Conversely, Francis et al. (2022) documented a negative relationship between managerial ability and tax avoidance. In commenting on the divergent result of the study by Koester et al. (2017), Francis et al. (2022) underline that the previous study used a measure that captures tax avoidance in general, while their study used more specific measures of tax aggressiveness or abusive tax planning (Martinez, 2017).

As a rule, strategies classified as abusive aim at reducing taxes through operations that have weak legal support and, therefore, carry a higher risk of being challenged by the tax authority (Lisowsky et al., 2013). In this sense, the phenomenon studied by Francis et al. (2022) diverges from the broader definition of Hanlon and Heitzman (2010), being limited to behaviors with higher tax risk, closer to evasion than to tax avoidance.

The study by Francis et al. (2022) highlights that managers with higher ability are more recognized in the labor market and have more to lose if their reputation is damaged in the event that illegal corporate activities come to light. In addition, the authors underline that, considering the significant direct and indirect costs associated with tax planning, managers with higher ability are able to generate revenue and create value for shareholders through other investments, where the tax aspect is not determinative of the associated return.

For this reason, Francis et al. (2022) suggest that managers with greater managerial ability, although they do not miss opportunities to engage in activities that result in lower tax burdens if these opportunities are associated with low risk, prefer to avoid more risky tax reduction activities and strategies. The difficulty generally lies in the correct identification and assessment of the tax risk associated with planning.

Subsequent studies that analyzed the relationship between managerial ability and tax avoidance continued to introduce mixed results. Based on a sample of companies listed on the Indonesia Stock Exchange, Saragih and Ali (2023b) concluded that companies with more managers with higher ability have lower tax risk and higher long-term tax avoidance, which is consistent with the results of Koester et al. (2017). Conversely, the results of Eigenstuhler et al. (2023) suggest that managerial ability positively influences accounting-tax compliance, which means it negatively impacts tax avoidance, aligning with the results of Francis et al. (2022).

Consistent with the aim of broadly assessing the influence of managerial ability on tax avoidance, rather than on the most aggressive end of the continuum, we align with the studies of Koester et al. (2017) and Saragih and Ali (2023b), and establish the following research hypothesis:

H₁: Managerial ability is positively related to tax avoidance.

2.3 Corporate Risk Management

The high uncertainty related to tax planning raises questions about the importance of the risk management process as a determinant of the level of tax avoidance. The risk management process helps senior managers to define the level of uncertainty they are prepared to face while seeking to increase firm value (Beasley et al., 2021). ERM aligns risk appetite with strategy by developing mechanisms to efficiently manage risks (COSO, 2004).

In this regard, it is estimated that the risk management process contributes to the reduction of the company’s tax expense, since this reduction is commonly sought by the shareholders and, therefore, is part of the strategy (Wang et al., 2020). Specifically, it is expected that the risk management process will lead to an increase in the level of tax avoidance, due to the improvement of the company’s internal and external communication systems (Eastman et al., 2024). Internal communication, which is one of the components of risk management (COSO, 2004), enhances communication among the various departments of the company, enabling the tax department to obtain all the information and resources necessary to assess scenarios and make the most efficient tax decisions.

In turn, external communication, which is one of the objectives of risk management (COSO, 2004), implies greater reliability of financial reports, which favors tax avoidance. The greater reliability of financial reports offsets the reduction in transparency caused by more aggressive tax planning (Eastman et al., 2024), which decreases the costs associated with more aggressive tax positions.

The reduction in transparency has been observed in companies that engage in corporate restructuring operations, where tax reduction is generally a significant component. Such companies tend to provide limited information regarding the restructuring in their financial statements (Nakayama & Salotti, 2014; Fogaça et al., 2020), which impairs investor assessment and negatively affects the firm’s value. This situation may influence the manager to decide not to carry out the restructuring or to accomplish it while recognizing a smaller tax saving. An efficient risk management process can enable the manager to complete the transaction without compromising disclosure (Eastman et al., 2024), encouraging them to achieve the maximum possible tax savings.

However, such understanding has not yet been consolidated in the literature. Masri et al. (2019) analyzed the moderating effect of tax risk management on the relationship between the practice of international tax planning and tax avoidance in companies from Indonesia and Malaysia. Different from those of Eastman et al. (2024), the results suggest that tax risk management weakens the positive relationship between international practices and tax avoidance, indicating that risk management can be a tool to limit tax avoidance.

It is expected that the tax decisions of managers with higher ability will be influenced by the risk management process, insofar as risk management is capable of providing the manager with relevant information for tax decision-making. The complexity of more aggressive tax strategies requires an efficient risk management process. Managers who have higher ability to transform corporate resources into results may make inefficient tax decisions if they do not have reliable information about the risks and opportunities associated with tax planning. Given this context, the following research hypothesis is established:

H₂: Corporate risk management strengthens the positive relationship between managerial ability and tax avoidance.

3 METHODOLOGICAL PROCEDURES

3.1 Population and Sample

The study population consisted of non-financial Brazilian companies listed on the Brasil, Bolsa, and Balcão (B3). Financial companies were not considered due to their accounting and tax peculiarities (Eigenstuhler et al., 2023). The unbalanced sample, formed from data collected from the Refinitiv Eikon® database and the Reference Forms available on the B3 website, consisted of companies that provided all the necessary information for the calculation of the variables for the period from 2016 to 2021, totaling 247 companies and 1,261 observations.

The start of the analysis period in 2016 is justified by the end of the Transitional Tax Regime (TTR) as of 2015, as determined by Law nº 12,973/2014. The TTR established that the calculation of federal taxes, including income taxes, should be carried out based on the accounting criteria in effect prior to the convergence with international accounting standards. Since the models use data from the previous year, it was also necessary to exclude the year 2015. The end of the analysis period in 2021 is justified by the unavailability of later data at the time of collection.

3.2 Constructs

In order to meet the aim of the research, the constructs listed in Table 1 were operationalized, which are discussed in greater detail below.

Table 1
Constructs

Variable		Definition	Authors
DEPENDENT VARIABLES
BTD	Total book-tax differences	$\frac{{EBT}_{it}-\frac{{CurrentIT}_{it}}{34\%}}{{Finaltotalassets}_{it}}$	Dunbar et al. (2010); Hanlon and Heitzman (2010); Fonseca and Costa (2017)
TBTD	Temporary book-tax differences	$\frac{\frac{{DeferredIT}_{it}}{34\%}}{{Finaltotalassets}_{it}}$	Dunbar et al. (2010); Hanlon and Heitzman (2010); Fonseca and Costa (2017)
PBTD	Permanent book-tax differences	$BTD-TBTD$	Dunbar et al. (2010); Hanlon and Heitzman (2010); Fonseca and Costa (2017)
INDEPENDENT VARIABLES
MA	Managerial Ability	According to Equations (1) and (2)	Demerjian (2012, 2013); Koester et al. (2017); Lunardi et al. (2022)
H_ERMI	Higher Enterprise Risk Management Index	1, 𝑖𝑓 𝑡ℎ𝑒 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛 𝑖𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑙𝑎𝑠𝑡 𝑞𝑢𝑖𝑛𝑡𝑖𝑙𝑒 𝑜𝑓 𝐸𝑅𝑀𝐼, 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑎𝑐𝑐𝑜𝑟𝑑𝑖𝑛𝑔 𝑡𝑜 𝐸𝑞. 3 0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒	Gordon et al. (2009); Naseem et al. (2019); Jacomossi et al. (2019)
CONTROL VARIABLES
S	Size	$\ln ({Totalassets}_{it})$	Chen et al. (2010); Armstrong et al. (2012)
LEV	Leverage	$\frac{{Longtermdebt}_{it}}{{Totalassets}_{i,t-1}}$	Armstrong et al. (2012)
ΔNRG	Net revenue growth	$\frac{{Netrevenue}_{it}-{Netrevenue}_{i,t-1}}{{Netrevenue}_{i,t-1}}$	Fonseca and Costa (2017)

• Notes: EBT = earnings before income tax (IT) and social contribution on net profit (SCNP); current and deferred IT = current and deferred IT and SCNP expenses, as applicable.
• Source: Elaborated by the authors.

In order to measure tax avoidance, three variations of BTD were used, which consist of the difference between accounting profit and taxable income, estimated from current tax expense and the combined nominal rate of IT and SCNP. It is considered that tax avoidance will be greater as the difference between accounting profit and taxable income increases. The Total BTD and its components were used: Temporary BTD (TBTD) and Permanent BTD (PBTD). Tax expense is understood as the expense with income tax and Social Contribution on Net Profit.

The differences that reflect situations where the timing of revenue and/or expense recognition differs between accounting standards and tax regulations generate Temporary Book-Tax Differences (BTDs). For these, it is expected that, at some future period, the difference will be reversed. Conversely, Permanent BTDs arise from revenues and/or expenses that will or will not form part of the calculation base for IT and SCNP definitively, without any differences to be accounted for future taxation or deduction.

According to Hanlon and Heitzman (2010), there is no metric of tax avoidance appropriate for all research contexts. In this study, the use of BTDs is justified by the possibility of using their permanent and temporary components, which allowed for a comparative analysis of the effects of tax strategies with different levels of risk, since permanent (temporary) differences are more (less) associated with the most aggressive edge of the continuum of tax avoidance (Castro, 2010; Fonseca & Costa, 2017; Martinez, 2017).

In order to measure managerial ability, the metric developed by Demerjian et al. (2012, 2013) was used. Given the perception that managers with higher ability are able to make the company transform its resources into results more efficiently, Demerjian et al. (2012) developed and validated a proxy for managerial ability, defined as the residual of the regression of institutional efficiency against efficiency drivers attributable to firm characteristics. It is assumed that institutional efficiency arises from both factors related to the corporation and the manager's ability. Thus, the portion of efficiency that cannot be explained by the firm’s characteristics, such as size, age, market share, among others, must arise from managerial ability.

Following Demerjian (2012), the proxy for managerial ability (MA) was calculated in two steps. The first step consisted of estimating the firm efficiency through the data envelopment analysis (DEA) technique, which compares the inputs used by management with the revenues (REV) earned in a given period. The inputs considered are: cost of goods sold (COGS), selling, general & administrative expenses (SG&A), plant, property, and equipment (PPE), operating leases (OpsLease), research and development expenses (R&D), acquired goodwill, and other intangible assets (OtherIntan). The firm efficiency (FE) assumes values between 0 and 1 and indicates the degree to which the firm is efficient. The calculation consisted of solving the optimization problem represented by Equation (1), using the method called constant returns to scale (CRS) (Charnes et al., 1978).

$${max}_{\upsilon }\theta =\frac{REV}{{\upsilon }_{1}COGS+{\upsilon }_{2}SG\&A+{\upsilon }_{3}PPE+{\upsilon }_{4}OpsLease+{\upsilon }_{5}R\&D+{\upsilon }_{6}Goodwill+{\upsilon }_{7}OtherIntan}$$ (1)

The optimization finds the coefficient vector (υ_n) that determines the firm efficiency (FE), which is attributable to the firm’s intrinsic characteristics and the manager. Given the same institutional resources, a manager with higher ability will be able to obtain revenues in a higher amount than those obtained by a manager with lower ability. Accordingly, the second step for calculating MA consisted of identifying the portion of efficiency not explained by the firm’s characteristics, consistent with the residual of the Tobit regression represented by Equation (2).

$${FE}_{it}=\alpha +{\beta }_1{TA}_{it}+{\beta }_2{MS}_{it}+{\beta }_3{FCFI}_{it}+{\beta }_4{A}_{it}+{\beta }_5{BSC}_{it}+{\beta }_6{FCI}_{it}+\sum {industryfixedeffects}_t+{\epsilon }_{it}$$ (2)

Where: FE = firm efficiency, calculated according to Equation (1); TA = natural logarithm of total assets at the end of the tax year; MS = market share, which is the proportion of the firm’s revenues relative to the total revenue of its economic sector; FCFI = free cash flow indicator, dummy variable that equals 1 if the firm’s free cash flow is positive, or 0 otherwise; A = natural logarithm of the firm’s age; BSC = concentration indicator in the business segment, which equals the proportion of the revenue of the company’s main business segment relative to its total revenue; FCI = foreign currency indicator, dummy variable that equals 1 if the firm has foreign exchange-related revenues or expenses, or 0 otherwise; ε = residual (MA).

In order to measure the effectiveness of the risk management process, a dichotomous variable based on the Enterprise Risk Management Index (ERMI) developed by Gordon et al. (2009) was used. The ERMI is based on the notion that corporate risk management consists of a process that identifies potential events capable of affecting the achievement of four categories of organizational goals: strategic, operational, communication, and compliance (COSO, 2004).

Although the ERMI by Gordon et al. (2009) uses two indicators for each goal, it is possible to calculate the variable with only one indicator for each goal (Jacomossi et al., 2019; Naseem et al., 2020). Due to the difficulty in obtaining the data necessary to calculate all the indicators, it was decided to perform the calculation with only one indicator per goal. Accordingly, the ERMI was determined according to Equation (3).

$${ERMI}_{it}=\sum _{k=1}^1{Strategy}_{it}+\sum _{k=1}^1{Operation}_{it}+\sum _{k=1}^1{Reporting}_{it}+\sum _{k=1}^1{Compliance}_{it}$$ (3)

The strategy refers to the way the company positions itself in the market against the competition in search of a competitive advantage. The greater the competitive advantage achieved, the lower the risk of failure in accomplishing the business strategy, which indicates the quality of risk management. The strategy indicator is calculated according to Equation (4).

$${Strategy}_{it}=\frac{{REV}_{it}-{\mu }_{{REV}_t}}{{\sigma }_{{REV}_t}}$$ (4)

Where: REV = net sales revenue; μ_REV = average net sales revenue of the economic sector in the period; σ_REV = standard deviation of net sales revenue in the same sector and period.

The operational indicator is related to the company’s productivity and can be measured by the input-output ratio in operational processes. Higher productivity can reduce the risk of operational failure, serving as a good indicator of risk management effectiveness. The operational indicator is measured according to Equation (5).

$${Operation}_{it}=\frac{{Netsalesrevenue}_{it}}{{Totalassets}_{it}}$$ (5)

The reporting indicator aims to measure the quality of the financial reports issued by the company. In general, it is believed that higher reliability of financial reports reduces the overall risk of failures. A measure of accounting report quality is derived from the relationship between discretionary (abnormal) accruals and normal accruals, since, as a rule, the higher the discretionary accruals, the worse the quality of accounting information. The reporting quality indicator uses the absolute values of the accruals, according to Equation (6).

$${Reporting}_{it}=\frac{\left|{Normalaccruals}_{it}\right|}{\left|{Normalaccruals}_{it}\right|+\left|{Abnormalaccruals}_{it}\right|}$$ (6)

Following Gordon et al. (2009), abnormal accruals were calculated according to the Jones (1991) model. Thus, abnormal accruals consist of the residual from the regression of total accruals against changes in sales and plant, property and equipment, as represented by the regression in Equation (7).

$${TAC}_{it}={\alpha }_0+{\alpha }_1\left(\frac{1}{{TA}_{i,t-1}}\right)+{\alpha }_2\left(\frac{{∆NRG}_{it}}{{TA}_{i,t-1}}\right)+{\alpha }_3\left(\frac{{PPE}_{it}}{{TA}_{i,t-1}}\right)+{\epsilon }_{it}$$ (7)

Where: TAC = total accruals calculated according to Equation (5); TA = total assets; ΔNRG = net revenue growth between period t-1 and period t; PPE = plant, property, and equipment.

According to Boina and Macedo (2018), the total accruals used as the dependent variable in Equation (4) can be calculated using either the Balance Sheet method or the Cash Flow Statement (CFS) method. In this study, the total accruals were calculated using the CFS method, as displayed in Equation (8).

$${TA}_{it}={NI}_{it}-{FCO}_{it}$$ (8)

Where: NI = net income for the period; FCO = cash flow from operations for the period.

Finally, the coefficients obtained in the regression of Equation (7) were used to estimate the normal and abnormal accruals, according to Equations (9) and (10).

$${Normalaccruals}_{it}={\alpha }_1\left(\frac{1}{{TA}_{i,t-1}}\right)+{\alpha }_2\left(\frac{{∆NRG}_{it}}{{TA}_{i,t-1}}\right)+{\alpha }_3\left(\frac{{PPE}_{it}}{{TA}_{i,t-1}}\right)$$ (9)

$${Abormalaccruals}_{it}={TA}_{it}-{Normalaccruals}_{it}$$ (10)

The last indicator used for the calculation of ERMI refers to compliance with laws and regulations. Greater compliance decreases the overall risk of failure. According to Gordon et al. (2009), audit fees are expected to be positively related to corporate compliance, in such a way that the compliance indicator was calculated as the ratio between audit fees and total assets, according to Equation (11). The dummy variable H_ERMI (Higher ERMI) was created to indicate the observations where the ERMI is in the top quintile of the distribution within each year, thus signaling the companies that demonstrated more effective risk management processes.

$${Compliance}_{it}=\frac{{Auditfees}_{it}}{{TA}_{it}}$$ (11)

Finally, control variables related to firm size (S), leverage (LEV), and net revenue growth (ΔNRG) were operationalized in line with previous studies (Koester et al., 2017; Saragih & Ali, 2023b). Larger companies have greater incentives and a higher capacity to influence the political process; therefore, they are expected to exhibit higher tax avoidance (Chen et al., 2010; Wang et al., 2020). More leveraged companies and those with higher sales growth tend to manage their taxable income more intensively (Fonseca & Costa, 2017). Except for S, which consists of the natural logarithm of total assets, all control variables were scaled by total assets.

3.3 Empirical Model

The empirical model used to test the two research hypotheses is the regression represented by Equation (12), which was adapted from the model used by Koester et al. (2017). The adaptation consisted of the inclusion of H_ERMI and its interaction with managerial ability. The coefficients were estimated using the OLS method, with robust standard errors clustered by firm, due to heteroscedasticity of the residuals, as well as year and industry fixed effects. All continuous variables were winsorized at 1%. For the endogeneity test, a 2SLS regression was carried out, using the average managerial ability of the sector as an instrument.

$$TA={\beta }_0+{\beta }_1{MA}_{it}+{\beta }_2{H\_ERMI}_{it}+{\beta }_3{MA}_{it}{*H\_ERMI}_{it}+{\beta }_4{S}_{it}+{\beta }_5{LEV}_{it}+{\beta }_6{\mathrm{\Delta }NRG}_{it}+\sum Yearfixedeffects+\sum Industryfixedeffects+{\epsilon }_{it}$$ (12)

Where: TA = tax avoidance, measured by BTD, TBTD, or PBTD, as applicable.

4 RESULTS

Initially, the descriptive statistics of the sample are displayed (Table 2, Panels A and B), followed by the Pearson correlation matrix (Table 3). Subsequently, the results of the operationalized regressions are displayed (Table 4).

Table 2
Descriptive statistics

Panel A. Continuous variables (N = 1.261)
Variable	Average	SD	P25	Median	P75
BTD	-0.0169	0.1304	-0.0392	0.0075	0.0494
TBTD	-0.0042	0.0576	-0.0160	0.0000	0.0091
PBTD	-0.0122	0.1355	-0.0180	0.0134	0.0407
MA	-0.0031	0.1128	-0.0756	-0.0216	0.0555
ERMI	1.0158	1.0803	0.3206	0.8341	1.387
S	21.8237	1.9484	20.5635	21.9253	23.1624
LEV	0.2451	0.3184	0.0567	0.1936	0.3189
ΔNRG	0.1715	0.5769	-0.0299	0.0863	0.2412
Panel B. Dummy variable
Variable	0	1	Total
H_ERMI	1.010 (80,1%)	251 (19.9%)	1.261 (100%)

• Note: The variables are described in Table 1.
• Source: Elaborated by the authors.

The results in Table 2 indicate that TBTD has a standard deviation of 0.0576, which is lower than that of the other two variations of BTD (0.1304 and 0.1355). The lower variability of TBTD suggests that tax practices focused on deferring taxes are more limited than those related to permanent differences, indicating that managers have fewer opportunities or less interest in manipulating deferred tax expense compared to current tax expense.

Table 3 indicates positive and significant correlations between Total BTD and its Temporary (0.20) and Permanent (0.87) components. Conversely, it indicates a significant negative correlation between Temporary BTD and Permanent BTD (-0.27), which suggests that, when making decisions about the level of corporate tax avoidance, managers may consider permanent and temporary differences as substitutive tax management practices.

Table 3
Pearson correlation matrix

Variables	BTD	TBTD	PBTD	MA	ERMI	S	LEV	ΔNRG
BTD	1
TBTD	0.20***	1
PBTD	0.87***	-0.27***	1
MA	0.11***	0.07**	0.09***	1
ERMI	0.18***	0.02	0.17***	0.05*	1
S	0.24***	0.03	0.21***	0.02	0.49***	1
LEV	-0.18***	0.00	-0.18***	0.06**	0.02	0.06**	1
ΔNRG	0.15***	0.06**	0.13***	0.12***	-0.01	-0.05*	0.00	1

• Note: The variables are described in Table 1.
• *, **, and *** indicate statistical significance at the levels of 10%, 5%, and 1%, respectively.
• Source: Elaborated by the authors.

Regarding Table 3, the positive and significant correlations between managerial ability, risk management, and the three variations of BTD are also highlighted, except for the correlation between TBTD and risk management, which was not significant. These results provide preliminary evidence supporting H₁, which predicts a positive relationship between managerial ability and tax avoidance.

Table 4 displays the relationships tested in this study. The models demonstrate adequate overall significance, considering the p-value associated with the F statistic. There are no multicollinearity issues, as the VIF values are within the expected standards. Nonetheless, it is observed that the model with TBDT has low explanatory power, as it showed a low R² and a negative Adjusted R². The Adjusted R² is calculated from R², which is penalized due to the number of independent variables in the model, which can result in a negative value, as in this case.

Table 4
Relationship between managerial ability and tax avoidance, moderated by risk management

Dependent variable:	BTD		TBTD		PBTD
	Coefficient	T stat.	Coefficient	T stat.	Coefficient	T stat.
Constant	-0.3181***	-3.45	0.0128	0.41	-0.3025***	-3.10
MA	0.1731***	3.40	0.0499***	3.07	0.1383***	2.72
H_ERMI	0.0058	0.67	0.0040	0.76	0.0057	0.54
MA*H_ERMI	-0.1322**	-2.37	-0.0598**	-2.24	-0.0921	-1.62
S	0.0135***	3.66	-0.0009	-0.67	0.0128***	3.23
LEV	-0.0806***	-4.08	-0.0026	-0.53	-0.0785***	-3.23
ΔNRG	0.0286**	2.41	0.0044	1.24	0.0259**	2.21
Robust standard errors	Yes		Yes		Yes
Industry fixed effects	Yes		Yes		Yes
Year fixed effects	Yes		Yes		Yes
F statistic (p-value)	5.41 (0.0000)		2,32 (0.0015)		4,25 (0.0000)
R2	16.87%		3.61%		13.64%
Adjusted R²	10.11%		-4.23%		6.62%
Maximum VIF	1.51		1.51		1.51
DW	1.32		1.83		1.38
N	1.261		1.261		1.261

• Notes: The variables are described in Table 1.
• R² represents the coefficient of determination; VIF represents the Variance Inflation Factor; DW represents Durbin-Watson; and N represents the number of firm-year observations. In order to minimize the influence of outliers, all continuous variables were winsorized at 1%.
• The 2SLS model was estimated for endogeneity, and the results showed that the instrumental variable “average MA by sector and year” had a positive and significant effect at the level of 1%.
• ** and *** denote statistical significance at the levels of 5% and 1%, respectively.
• Standard errors were clustered by firm.
• Source: Elaborated by the authors.

The results in Table 4 show that managerial ability is positively related to tax avoidance when considering the three variations of BTD tested in this study, which implies that hypothesis H₁ is not rejected. This finding aligns with the results of Koester et al. (2017), suggesting that managers with higher ability are able to identify the best tax planning opportunities, reducing both current and deferred tax expenses. The results show that managers with higher ability are better able to deal with the uncertainties inherent in the Brazilian tax system (Martinez, 2017), in such a way that the personal characteristics of managers, and not just the characteristics of the firm, are relevant for determining the level of corporate tax expenditure (Desai & Darmaphala, 2006; Dyreng et al., 2010).

Regarding the economic analysis, an increase of one standard deviation in MA (0.11) is associated with a 187% increase (0.11*0.17/0.01) in Total BTD, compared to the average. Similarly, an increase of one standard deviation in MA (0.11) impacts a 110% increase (0.11*0.049/0.004) in TBTD. In addition, an increase of one standard deviation in MA (0.11) is associated with a 143% increase (0.11*0.13/0.01) in PBTD. From an economic point of view, a more significant effect of MA on the Total BTD stands out, which allows for inferring that, disregarding the influence of ERM, the managers with higher ability do not have a preference between practices that generate permanent differences or practices that generate temporary differences.

However, the situation changes when risk management is taken into consideration. This is because Table 4 shows that risk management has a moderating effect on the relationship between managerial ability and the level of tax avoidance. Contrary to expectations, this effect is negative, since the regressions with BTD and TBTD show negative and significant coefficients for the interaction between MA and H_EMRI. The interaction coefficient was not significant in the regression with Permanent BTD.

In the regression with Total BTD, considering that the sum of the coefficients of MA and MA*H_ERMI is not statistically different from zero (0.1731 - 0.1322 = 0.0409; F-test: 1.47; p < 0.2267), it is observed that the moderating effect is quite relevant, as it cancels out the effect of managerial ability on tax aggressiveness. In the regression with the TBTD, the sum of the coefficients of MA and MA*H_ERMI is also not statistically different from zero (0.0499 - 0.0598 = -0.0101; F-test: 0.21; p < 0.6461), making it possible to state that, in companies with better risk management, managerial ability is not associated with higher tax avoidance that generates temporary differences. These results are confirmed by the 2SLS regression (not tabulated), since the instruments proved to be sufficiently exogenous and appropriate for the performed test.

Together, the results determine the rejection of hypothesis H₂, which predicted that risk management would intensify the positive relationship between managerial ability and tax avoidance. The negative and significant coefficients of the interaction between MA and H_ERMI in the regressions with BTD (-0.1322) and with TBTD (-0.0598) can be explained by the role of risk management in organizing internal communication (Eastman et al., 2024). Consistent with those of Masri et al. (2019), the results suggest that, in companies with higher ERMI, managers with higher ability have more information about the opportunities and the risks associated with tax avoidance.

Accordingly, in the case of strategies that generate temporary differences, the risk management process influences the assessment of managers with higher ability, as they understand that the risks of these strategies outweigh their benefits, in such a way that the positive effect of managerial ability on the level of tax avoidance is nullified. However, in the case of strategies that generate permanent differences, the effect is not significant. This suggests that managers with higher ability already have sufficient clarity that the benefits outweigh the risks associated with strategies that generate permanent differences, regardless of the risk management process. By highlighting not only the risks but also the opportunities to increase corporate results through the reduction of tax expenses, risk management processes more comprehensively fulfill their purpose, as outlined in COSO (2004).

The results of Table 4 suggest that, in companies with poorer risk management processes (low ERMI), the presence of managers with higher ability is associated with higher temporary and permanent differences, since, when the dummy H_ERMI assumes the value zero, the effect of MA on TBDT is positive and significant, and this effect is always positive and significant for PBTD, regardless of H_ERMI.

Conversely, in companies with better risk management processes, the presence of managers with higher ability is only associated with higher PBTD, since H_ERMI nullifies the effect of MA on TBTD and on Total BTD (as the sum of the coefficients of MA*H_ERMI and MA is not statistically different from zero). The reasons for this difference may be associated with the fact that practices that generate temporary differences offer limited benefits compared to the risks associated with them. Given the same level of risk, the benefits associated with permanent differences are greater than the benefits arising from temporary differences.

The risk management process improves internal communication (Eastman et al., 2024), providing managers with the necessary information to properly align risk appetite with strategy (COSO, 2004). The results in Table 4 suggest that risk management informs managers with higher ability that tax deferral practices do not favor corporate strategy, insofar as the assumed risk is considered greater than the benefit associated with merely deferring the tax payment.

The same does not happen with strategies that permanently reduce tax expenses, which generate benefits consisting of an effective reduction in the organization’s tax burden. In this case, risk management moderation was not significant, which suggests that managers with higher ability already have sufficient information to conclude that the benefits outweigh the costs of tax strategies that generate permanent differences, regardless of the risk management process.

The results in Table 4 suggest that, in companies with better risk management processes (H_ERMI = 1), managerial ability is positively associated only with tax avoidance that generates permanent differences, without relationship between managerial ability and strategies that generate temporary differences. Previous studies have shown that managers prefer permanent strategies, as these strategies effectively reduce tax expense and increase reported profit, whereas temporary strategies do not provide such a reduction, since they decrease current expense and increase deferred expense by the same amount (Dunbar et al., 2010).

Conversely, in companies with low ERMI, managers with higher ability retain practices that generate both permanent and temporary differences, exerting effort and assuming risks in activities with lower potential benefit.

Average tests (not tabulated) corroborate this conclusion, as they indicate that the average PBTD is significantly higher for companies with higher risk management indexes (H_ERMI = 1) than for the others and that the average TBTD is not different between the two groups. This shows that companies with better risk management processes engage more intensively in strategies that generate permanent differences than other companies.

In light of the foregoing, it is observed that, although hypothesis H₂ was rejected, the comparison between the behavior of Temporary BTD and Permanent BTD aligns with the concept that corporate risk management favors the achievement of organizational goals (Eastman et al., 2024), insofar as it directs tax planning efforts towards a more consistent (since definitive) reduction of corporate tax burden. Considering the negative correlation between Temporary BTD and Permanent BTD, as highlighted in Table 3, it can be stated that, in companies with better risk management processes, managers with higher ability adopt permanent practices instead of temporary ones.

The results of this study indicate that, by using information from risk management processes to assess the risks and benefits associated with tax practices, managers with higher ability choose to engage only in higher-risk tax strategies, as permanent differences are considered strategies that attract higher tax risk (Castro, 2010; Fonseca & Costa, 2017; Martinez, 2017). Thus, it is assumed that, in the assessment of the managers with higher ability, the risk of having such practices questioned by tax authorities is low, at least compared to the potential benefits. In light of this, the question arises: what could be the possible reasons for this assessment?

The literature has some suggestions, which are generally associated with the limitation of legal consequences arising from the detection of tax offenses in the Brazilian context, such as the successive granting of special payment plans with substantial discounts on fines and interest (Paes, 2014) and the delay in the resolution of tax disputes, which leads companies to use tax non-compliance as a form of financing (Plutarco, 2012). Knowledge in the field would benefit from new research that investigates more deeply the relationship of these practices with managers’ risk perception.

5 CONCLUSION

This study aimed to assess the moderating effect of corporate risk management on the relationship between managerial ability and tax avoidance, using a sample of 1,261 observations, consisting of non-financial Brazilian companies listed on the stock exchange. Tax avoidance was measured by Total BTD and its components, Temporary BTD and Permanent BTD, which allowed for examining the differences between strategies that aim merely to defer tax payments and those that aim to achieve a permanent reduction in the corporate tax burden.

In general, the results indicate that in companies with better risk management processes, managers with higher levels of managerial ability choose to engage only in more aggressive tax planning, which generates permanent differences. In these companies, managerial ability is not associated with tax planning strategies that generate only the deferral of taxes. In companies with poorer risk management processes, managers with higher levels of managerial ability choose to engage in tax planning strategies that generate both temporary and permanent differences.

These findings are consistent with the expected role of corporate risk management processes (Eastman et al., 2024; COSO, 2004), as they highlight their relevance in defining risk appetite and facilitate the exploitation of opportunities to reduce tax expenses, thus increasing organizational performance.

Regarding managerial ability, this study advances beyond the findings of Koester et al. (2017), Francis et al. (2022), Saragih and Ali (2023b), and Eigenstuhler et al. (2023) by showing that the corporate risk management process moderates its relationship with tax avoidance. In general, the findings of this study align with those of Koester et al. (2017) and Saragih and Ali (2023b), as they indicate that managerial ability is positively related to tax avoidance.

The divergence with the results of Francis et al. (2022), who found a negative relationship between managerial ability and tax avoidance positioned at the most aggressive edge of Lietz’s (2013) continuum, presumably associated with managers’ reputational costs, can be explained by the specificity of the Brazilian context, where the legal consequences of detecting tax offenses by tax authorities are mitigated, for example, by successive granting of special payment plans and by delays in resolving tax disputes (Paes, 2014; Plutarco, 2012). It is expected that new studies may clarify the relationship between these factors specific to the Brazilian institutional reality and the perception of corporate risk.

The study contributes to the literature on tax avoidance by showing that there are significant differences between temporary and permanent strategies. The results show that, in a context where managers with higher ability benefit from better risk management processes, there is greater clarity in assessing the opportunities and risks associated with tax avoidance, which leads companies to favor the adoption of permanent strategies.

Despite efforts to ensure the reliability and validity of this research, it is not without limitations. One of them is the data lag, since the last year analyzed was 2021, which was the latest year available at the time of data collection. Nonetheless, no subsequent events are anticipated that could have altered the studied relationships, which mitigates this limitation. In any case, the research analyzed more recent data than previous studies on managerial ability and tax avoidance. The most recent studies, namely those by Francis et al. (2022), Eigenstuhler et al. (2023), and Saragih and Ali (2023b), used data up to 2009, 2019, and 2018, respectively.

Another limitation refers to the adequacy of the used metrics. Regarding BTD, it is already established in the literature that it can capture both tax avoidance and earnings management (Fonseca & Costa, 2017). Martinez (2017) suggests that this limitation can be minimized by including controls for earnings management in regressions that use BTD.

In this study, this recommendation was addressed through the inclusion of the ERMI in the used model, as this construct contains a component that captures discretionary accruals. Regarding managerial ability, the managerial ability score introduced by Demerjian et al. (2012) was used, which has been previously validated and used in subsequent studies. Nevertheless, other metrics can be used, which is expected to be done in future research.

REFERENCES

• Allam, A., Abou-El-Sood, H., Elmarzouky, M., & Yamen, A. (2024). Financial development and tax evasion: International evidence from OECD and non-OECD countries. Journal of International Accounting, Auditing and Taxation, 57, 100653. https://doi.org/10.1016/j.intaccaudtax.2024.100653
  » https://doi.org/10.1016/j.intaccaudtax.2024.100653
• Amri, K., Ben Mrad Douagi, F. W., & Guedrib, M. (2023). The impact of internal and external corporate governance mechanisms on tax aggressiveness: Evidence from Tunisia. Journal of Accounting in Emerging Economies, 13(1), 43-68. https://doi.org/10.1108/JAEE-01-2021-0019
  » https://doi.org/10.1108/JAEE-01-2021-0019
• Armstrong, C. S., Blouin, J. L., & Larcker, D. F. (2012). The incentives for tax planning. Journal of accounting and economics, 53(1-2), 391-411. https://doi.org/10.1016/j.jacceco.2011.04.001
  » https://doi.org/10.1016/j.jacceco.2011.04.001
• Baik, B., Choi, S., & Farber, D. B. (2020). Managerial ability and income smoothing. The Accounting Review, 95(4), 22. https://doi.org/10.2308/accr-52600
  » https://doi.org/10.2308/accr-52600
• Beasley, M. S., Goldman, N. C., Lewellen, C. M., & McAllister, M. (2021). Board risk oversight and corporate tax-planning practices. Journal of Management Accounting Research, 33(1), 7-32. https://doi.org/10.2308/JMAR-19-056
  » https://doi.org/10.2308/JMAR-19-056
• Blouin, J. (2014). Defining and measuring tax planning aggressiveness. National Tax Journal, 67(4), 875-899. https://doi.org/10.17310/ntj.2014.4.06
  » https://doi.org/10.17310/ntj.2014.4.06
• Boina, T. M., & Macedo, M. A. D. S. (2018). Capacidade preditiva de accruals antes e após as IFRS no mercado acionário brasileiro. Revista Contabilidade & Finanças, 29, 375-389. https://doi.org/10.1590/1808-057x201806300
  » https://doi.org/10.1590/1808-057x201806300
• Brühne, A. I., & Schanz, D. (2022). Defining and managing corporate tax risk: perceptions of tax risk experts. Contemporary Accounting Research, 39(4), 2861-2902. https://doi.org/10.1111/1911-3846.12785
  » https://doi.org/10.1111/1911-3846.12785
• Castro, L. F. M. (Org.) (2010). Planejamento tributário: análise de casos. MP Editora.
• Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision making units. European Journal of Operational Research, 2(6), 429-444. https://doi.org/10.1016/0377-2217(78)90138-8
  » https://doi.org/10.1016/0377-2217(78)90138-8
• Chen, S., Chen, X., Cheng, Q., & Shevlin, T. (2010). Are family firms more tax aggressive than non-family firms? Journal of Financial Economics, 95, 41-61. https://doi.org/10.1016/j.jfineco.2009.02.003
  » https://doi.org/10.1016/j.jfineco.2009.02.003
• Cook, K. A., Moser, W. J., & Omer, T. C. (2017). Tax avoidance and ex ante cost of capital. Journal of Business Finance & Accounting, 44(7-8), 1109-1136. https://doi.org/10.1111/jbfa.12258
  » https://doi.org/10.1111/jbfa.12258
• COSO (2004). Committee of Sponsoring Organizations of the Treadway Commission. Integrated framework. Executive summary.
• COSO (2017). Committee of Sponsoring Organizations of the Treadway Commission. Integrating with strategy and performance. Executive summary. https://www.coso.org/SitePages/Home.aspx
  » https://www.coso.org/SitePages/Home.aspx
• Demerjian, P., Lev, B., & McVay, S. (2012). Quantifying managerial ability: A new measure and validity tests. Management Science, 58(7), 1229-1248. https://doi.org/10.1287/mnsc.1110.1487
  » https://doi.org/10.1287/mnsc.1110.1487
• Demerjian, P. R., Lev, B., Lewis, M. F., & McVay, S. E. (2013). Managerial ability and earnings quality. The Accounting Review, 88(2), 463-498. https://doi.org/10.2308/accr-50318
  » https://doi.org/10.2308/accr-50318
• Desai, M. A., & Dharmapala, D. (2006). Corporate tax avoidance and high-powered incentives. Journal of Financial Economics, 79(1), 145-179. https://doi.org/10.1016/j.jfineco.2005.02.002
  » https://doi.org/10.1016/j.jfineco.2005.02.002
• Drake, K. D., Lusch, S. J., & Stekelberg, J. (2019). Does tax risk affect investor valuation of tax avoidance? Journal of Accounting, Auditing & Finance, 34(1), 151-176. https://doi.org/10.1177/0148558X17692674
  » https://doi.org/10.1177/0148558X17692674
• Dunbar, A., Higgins, D. M., Phillips, J. D., & Plesko, G. A. (2010). What do measures of tax aggressiveness measure? Proceedings. Annual Conference on Taxation and Minutes of the Annual Meeting of the National Tax Association, 103, 18-26. https://www.jstor.org/stable/prancotamamnta.103.18
  » https://www.jstor.org/stable/prancotamamnta.103.18
• Dyreng, S. D., Hanlon, M., & Maydew, E. L. (2010). The effects of executives on corporate tax avoidance. The Accounting Review, 85(4), 1163-1189. https://doi.org/10.2308/accr.2010.85.4.1163
  » https://doi.org/10.2308/accr.2010.85.4.1163
• Eastman, E., Ehinger, A., & Xu, J. (2024). Enterprise Risk Management and Corporate Tax Planning. Journal of Risk and Insurance, 91(3), 529-566. https://doi.org/10.1111/jori.12469
  » https://doi.org/10.1111/jori.12469
• Edwards, A. S., Schwab, C. M., & Shevlin, T. (2013). Financial Constraints and the Incentive for Tax Planning. 2013 American Taxation Association Midyear Meeting. https://ssrn.com/abstract=2216875
  » https://ssrn.com/abstract=2216875
• Eigenstuhler, D. P., dal Magro, C. B., & Mazzioni, S. (2023). Habilidade gerencial e a conformidade contábil-fiscal: estudo cross-country. Revista Catarinense da Ciência Contábil, 22, 1-18. https://doi.org/10.16930/2237-766220233341
  » https://doi.org/10.16930/2237-766220233341
• Fogaça, F. M., Souza, B. S., de Oliveira, W. S., & Silva, J. R. (2020). Nível da divulgação dos ativos identificáveis reconhecidos em uma combinação de negócios. Revista Liceu On-Line, 10(1), 96-112. https://liceu.fecap.br/LICEU_ON-LINE/article/view/1829
  » https://liceu.fecap.br/LICEU_ON-LINE/article/view/1829
• Fonseca, K. B. C., & Costa, P. S. (2017). Fatores determinantes das book-tax differences. Revista de Contabilidade e Organizações, 11(29), 17-29. https://doi.org/10.11606/rco.v11i29.122331
  » https://doi.org/10.11606/rco.v11i29.122331
• Francis, B. B., Sun, X., Weng, C. H., & Wu, Q. (2022). Managerial ability and tax aggressiveness. China Accounting and Finance Review, 24(1), 53-75. https://doi.org/10.1108/CAFR-02-2022-0002
  » https://doi.org/10.1108/CAFR-02-2022-0002
• Gordon, L. A.; Loeb, M. P.; Tseng, C-Y. (2009). Enterprise risk management and firm performance: A contingency perspective. Journal of Accounting and Public Policy, 28(4), 301-327. https://doi.org/10.1016/j.jaccpubpol.2009.06.006
  » https://doi.org/10.1016/j.jaccpubpol.2009.06.006
• Hanlon, M., & Heitzman, S. (2010). A review of tax research. Journal of Accounting and Economics, 50(2-3), 127-178. https://doi.org/10.1016/j.jaccpubpol.2009.06.006
  » https://doi.org/10.1016/j.jaccpubpol.2009.06.006
• Haider, I., Singh, H., & Sultana, N. (2021). Managerial ability and accounting conservatism. Journal of Contemporary Accounting & Economics, 17(1), 100242. https://doi.org/10.1016/j.jcae.2020.100242
  » https://doi.org/10.1016/j.jcae.2020.100242
• Higgins, D., Omer, T. C., & Phillips, J. D. (2015). The influence of a firm's business strategy on its tax aggressiveness. Contemporary Accounting Research, 32(2), 674-702. https://doi.org/10.1111/1911-3846.12087
  » https://doi.org/10.1111/1911-3846.12087
• Hoyt, R.; Liebenberg, A. (2011). The value of enterprise risk management. Journal of Risk and Insurance, 78(4), 795-822. https://doi.org/10.1111/j.1539-6975.2011.01413.x
  » https://doi.org/10.1111/j.1539-6975.2011.01413.x
• Huang, H., Sun, L., & Zhang, J. (2017). Environmental uncertainty and tax avoidance. Advances in taxation, 24, 83-124. https://doi.org/10.1108/S1058-749720170000024002
  » https://doi.org/10.1108/S1058-749720170000024002
• Jacomossi, F. A., Lunardi, M. A., & Silva, M. Z. (2019). Enterprise risk management e o desempenho empresarial: uma perspectiva contingencial. Revista Mineira de Contabilidade, 20(3), 45-58. https://doi.org/10.21714/2446-9114RMC2019v20n3t04
  » https://doi.org/10.21714/2446-9114RMC2019v20n3t04
• Jones, J. J. (1991). Earnings management during import relief investigations. Journal of Accounting Research, 29(2), 193-228. https://www.jstor.org/stable/2491047
  » https://www.jstor.org/stable/2491047
• Koester, A., Shevlin, T., & Wangerin, D. (2017). The role of managerial ability in corporate tax avoidance. Management Science, 63(10), 3285-3310. https://doi.org/10.1287/mnsc.2016.2510
  » https://doi.org/10.1287/mnsc.2016.2510
• Kovermann, J., & Velte, P. (2019). The impact of corporate governance on corporate tax avoidance - A literature review. Journal of International Accounting, Auditing and Taxation, 36, 100270. https://doi.org/10.1016/j.intaccaudtax.2019.100270
  » https://doi.org/10.1016/j.intaccaudtax.2019.100270
• Kovermann, J., & Velte, P. (2021). CSR and tax avoidance: A review of empirical research. Corporate Ownership and Control, 18(2), 20-39. https://doi.org/10.22495/cocv18i2art2
  » https://doi.org/10.22495/cocv18i2art2
• Li, Y., Luo, Y., Wang, J., & Foo, C. T. (2016). A theory of managerial tax aggression: Evidence from China, 2008-2013 (9702 observations). Chinese Management Studies, 10(1), 12-40. https://doi.org/10.1108/CMS-01-2016-0001
  » https://doi.org/10.1108/CMS-01-2016-0001
• Lietz, G. M. (2013). Tax Avoidance vs. Tax Aggressiveness: A Unifying Conceptual Framework. https://dx.doi.org/10.2139/ssrn.2363828
  » https://dx.doi.org/10.2139/ssrn.2363828
• Lisowsky, P., Robinson, L., & Schmidt, A. (2013). Do publicly disclosed tax reserves tell us about privately disclosed tax shelter activity? Journal of Accounting Research, 51(3), 583-629. https://doi.org/10.1111/joar.12003
  » https://doi.org/10.1111/joar.12003
• Lunardi, M. A., Ferrari, A., & Klann, R. C. (2022). Habilidade gerencial e gerenciamento de resultados contábeis. Revista Contemporânea de Contabilidade, 19(51), 53-72. https://doi.org/10.5007/2175-8069.2022.e79278
  » https://doi.org/10.5007/2175-8069.2022.e79278
• Martinez, A. L. (2017). Agressividade tributária: um survey da literatura. Revista de Educação e Pesquisa em Contabilidade, 11(0), 106-124. https://doi.org/10.17524/repec.v11i0.1724
  » https://doi.org/10.17524/repec.v11i0.1724
• Masri, I., Syakhroza, A., Wardhani, R., & Samingun. (2019). The role of tax risk management in international tax avoidance practices: evidence from Indonesia and Malaysia. International Journal of Trade and Global Markets, 12(3-4), 311-322. https://doi.org/10.1504/IJTGM.2019.101561
  » https://doi.org/10.1504/IJTGM.2019.101561
• Moura, G. D. de, Fank, D. R. B., Mazzioni, S., Angonese, R., & Silva, G. (2019). Habilidade gerencial e perdas do valor recuperável do goodwill: análise em companhias abertas listadas na B3. Revista de Educação e Pesquisa em Contabilidade, 13(2), 197-218. https://doi.org/10.17524/repec.v13i2.2002
  » https://doi.org/10.17524/repec.v13i2.2002
• Nakayama, W. K., & Salotti, B. M. (2014). Fatores determinantes do nível de divulgação de informações sobre combinações de negócios com a entrada em vigor do pronunciamento técnico CPC 15. Revista Contabilidade & Finanças, 25, 267-280. https://doi.org/10.1590/1808-057x201411260
  » https://doi.org/10.1590/1808-057x201411260
• Naseem, T., Shahzad, F., Asim, G. A., Rehman, I. U., & Nawaz, F. (2020). Corporate social responsibility engagement and firm performance in Asia Pacific: The role of enterprise risk management. Corporate Social Responsibility and Environmental Management, 27(2), 501-513. https://doi.org/10.1002/csr.1815
  » https://doi.org/10.1002/csr.1815
• Paes, N. L. (2014). Os efeitos dos parcelamentos sobre a arrecadação tributária. Estudos Econômicos (São Paulo), 44, 323-350. https://doi.org/10.1590/S0101-41612014000200004
  » https://doi.org/10.1590/S0101-41612014000200004
• Plutarco, H. M. (2012). A sonegação e a litigância tributária como forma de financiamento. Economic Analysis of Law Review, 3(1), 122-147. https://doi.org/10.18836/2178-0587/ealr.v3n1p122-147
  » https://doi.org/10.18836/2178-0587/ealr.v3n1p122-147
• Richardson, G., Taylor, G., & Lanis, R. (2013). The impact of board of director oversight characteristics on corporate tax aggressiveness: An empirical analysis. Journal of Accounting and Public Policy, 32(3), 68-88. https://doi.org/10.1016/j.jaccpubpol.2013.02.004
  » https://doi.org/10.1016/j.jaccpubpol.2013.02.004
• Saragih, A. H., & Ali, S. (2023a). Corporate tax risk: a literature review and future research directions. Management Review Quarterly, 73(2), 527-577. https://doi.org/10.1007/s11301-021-00251-8
  » https://doi.org/10.1007/s11301-021-00251-8
• Saragih, A. H., & Ali, S. (2023b). The impact of managerial ability on corporate tax risk and long-run tax avoidance: empirical evidence from a developing country. Corporate Governance, 23(5), 1117-1144. https://doi.org/10.1108/CG-08-2022-0346
  » https://doi.org/10.1108/CG-08-2022-0346
• Wahab, E. A. A., Ariff, A. M., Marzuki, M. M., & Sanusi, Z. M. (2017). Political connections, corporate governance, and tax aggressiveness in Malaysia. Asian Review of Accounting, 25(3), 424-451. https://doi.org/10.1108/ARA-05-2016-0053
  » https://doi.org/10.1108/ARA-05-2016-0053
• Wang, F., Xu, S., Sun, J., & Cullinan, C. P. (2020). Corporate Tax Avoidance: a Literature Review and Research Agenda. Journal of Economic Surveys, 34(4), 793-811. https://doi.org/10.1111/joes.12347
  » https://doi.org/10.1111/joes.12347
• Wilde, J. H., & Wilson, R. J. (2018). Perspectives on corporate tax planning: Observations from the past decade. The Journal of the American Taxation Association, 40(2), 63-81. https://doi.org/10.2308/atax-51993
  » https://doi.org/10.2308/atax-51993
• Zhou, S., Zhou, P., & Ji, H. (2022). Can digital transformation alleviate corporate tax stickiness: The mediation effect of tax avoidance. Technological Forecasting and Social Change, 184, 122028. https://doi.org/10.1016/j.techfore.2022.122028
  » https://doi.org/10.1016/j.techfore.2022.122028

• DATA AVAILABILITY STATEMENT
  The entire dataset supporting the results of this study is available upon request to the authors.
• FUNDING
  The authors would like to thank the Santa Catarina State Research and Innovation Support Foundation (FAPESC), Call Notice 19/2024, for its financial support in carrying out this research.
• This is a bilingual text. This article was originally written in Portuguese, published under the DOI https://doi.org/10.1590/1808-057x20252200.pt.
• Paper presented at the XXVI SemeAd, São Paulo, SP, Brazil, November 2023.

Edited by

• Editor-in-Chief:
  Andson Braga de Aguiar

• Associate Editor:
  Cláudio de Araújo Wanderley

Data availability

The entire dataset supporting the results of this study is available upon request to the authors.

Publication Dates

• Publication in this collection
  19 Jan 2026
• Date of issue
  2025

History

• Received
  19 Aug 2024
• Reviewed
  27 Aug 2024
• Accepted
  11 Apr 2025