Top management’s financial experience, backgrounds in regulatory and non-regulatory financial institutions, and corporate tax aggressiveness Open Access

An emerging stream of literature grounded in upper echelons theory (Hambrick and Mason, 1984; Zhang et al., 2022) emphasizes that top executives' backgrounds shape corporate strategies. Managerial attributes operate as cognitive filters through which strategic choices are interpreted and executed. Among these attributes, prior career experience, especially experience formed early in professional life, critically determines managerial preferences and decision styles (Reger et al., 1994; Barr et al., 1992; Gibbons and Waldman, 2004). Building on this perspective, recent studies examine how demographic and psychological traits of top management team (TMT) members affect corporate tax planning. Prior studies have documented that executive gender diversity (Francis et al., 2014), educational attainment (Jbir et al., 2021), executive overconfidence (Kubick and Lockhart, 2017), and age or compensation profiles (Huang et al., 2018) may significantly affect corporate taxation. Consistent with the view that experiential learning fosters durable belief systems shaping behavior across contexts (Caspi et al., 2005; Roberts et al., 2003), we argue that financial work experience functions as a form of imprinting. Once internalized, professional logics acquired in financial institutions affect how executives perceive compliance, evaluate regulatory risk, and deploy tax-planning tools—effects that persist over careers and materially influence firm tax policies. Our study, therefore, investigates whether, and through which channels, top executives' financial work experience affects corporate tax avoidance, with particular attention to the institutional context in which that experience was acquired—namely, regulatory-oriented versus non-regulatory-oriented financial institutions.

Several mechanisms may link managerial financial experience to tax behavior. First, executives with such backgrounds possess technical expertise in applying financial reporting standards and tax rules (Li et al., 2023; Rhodes and Russomanno, 2021), potentially allowing them to exploit the divergence between book income and taxable income. This capability may enable them to reduce tax liabilities through methods that preserve reported earnings, such as deferrals, deductions, and the selective use of deferred tax assets and liabilities. Second, their cognitive framing, shaped by early exposure to financial arbitrage (Custódio and Metzger, 2014), risk-adjusted financial return optimization (Custodio et al., 2025), and complex deal structuring (Bodnaruk and Simonov 2015), cultivates a mindset that is not only comfortable with regulatory ambiguity but also oriented toward identifying and exploiting “flexible” areas in the tax code, potentially encouraging more aggressive tax planning, especially where enforcement is opaque. Third, financially experienced executives are more adept at embedding tax strategies into broader corporate financial policies, using instruments such as transfer pricing, intercompany financing, and hybrid securities to engineer integrated solutions that serve both tax efficiency and capital-market objectives. Together, the mechanisms suggest that financial expertise is not merely a technical skill set but a strategic orientation that shapes how firms perceive, navigate, and respond to tax regulation. Accordingly, financial expertise functions as both a human-capital asset and a behavioral lens through which tax aggressiveness is operationalized.

To further unpack how financial expertise translates into tax planning behavior, it is essential to distinguish the institutional contexts in which such expertise is acquired. Financial institutions where top executives gain their formative experience vary widely in goals, culture, and professional norms (Custódio and Metzger, 2014), which may result in fundamentally different cognitive frames, compliance postures, and strategic inclinations. We then classify managerial financial experience into two broad categories—non-regulatory-oriented and regulatory-oriented organizations—based on whether the organizations are primarily market-profit-driven or compliance-focused.

Top executives with experience in non-regulatory-oriented financial institutions, such as insurance companies, investment and commercial banks, hedge funds, and asset management entities, are often trained in environments that prioritize risk-taking and performance-driven outcomes. These institutions foster a culture where maximizing returns and identifying value-enhancing opportunities are central goals (Stowell, 2010). As Bolton et al. (2015) show, such settings often incentivize short-term financial performance through compensation structures that promote aggressive risk-taking and reward regulatory arbitrage. Consequently, executives emerging from these environments tend to develop a skill set centered on navigating complex financial instruments, optimizing resource allocation, and identifying cost-reduction strategies, including those related to tax planning. Their exposure to sophisticated financial instruments, advanced modeling techniques, and dynamic market environments may equip them with the ability to devise creative tax-saving strategies. These executives may often view tax avoidance as a legitimate tool for enhancing shareholder value (Desai and Dharmapala, 2006), leveraging their understanding of intricate tax codes and legal frameworks to exploit available loopholes while remaining within the boundaries of legality. Moreover, their results-oriented training emphasizes delivering measurable financial outcomes, which may incentivize the pursuit of aggressive tax minimization strategies to boost short-term profitability. However, this approach may also lead to a greater tolerance for risk when pursuing tax strategies, as these executives are accustomed to operating in high-stakes environments where calculated risks are integral to achieving success. This perspective may result in a more aggressive stance toward tax planning, prioritizing short-term gains over potential long-term reputational or regulatory risks.

By bringing financial expertise and mindset to their roles in corporate management, these executives shape tax planning practices in ways that reflect the strategic and performance-driven principles they internalized in non-regulatory financial institutions. Consequently, firms led by such top executives may often engage in more aggressive tax avoidance strategies, leveraging their leaders' deep financial acumen and risk management capabilities.

In contrast, top executives with experience in regulatory-oriented financial institutions (i.e. stock exchanges, policy/central banks, or regulatory-related commissions) are likely to approach tax planning with a compliance-oriented mindset. Their professional background emphasizes adherence to regulatory standards, transparency, and accountability (Li et al., 2022), fostering a strong awareness of the reputational and legal risks linked to aggressive tax strategies. Training in such institutions often promotes balancing corporate objectives with broader policy and societal goals, encouraging risk-averse and ethically grounded decision-making. Rather than pursuing aggressive tax avoidance, these executives may favor transparent, compliant tax planning that safeguards the firm's reputation, maintains regulatory goodwill, and supports long-term governance objectives. This orientation contributes to organizational credibility and resilience in increasingly scrutinized regulatory environments.

We examine this question in the Chinese institutional setting, which presents a particularly fertile ground for investigating the role of managerial discretion in tax planning. Despite decades of tax reform in China, enforcement remains highly fragmented across regions. Local variations in legal capacity, bureaucratic efficiency, and regulatory professionalism, well-documented by the NERI Index of Marketization, have resulted in a patchwork of enforcement standards ^[1]. In many provinces, particularly in less-developed areas, limited administrative resources and regulatory ambiguity allow informal practices to substitute for formal oversight. As a result, firms operate under conditions of partial enforcement, where tax rules are inconsistently applied and compliance pressure is low (Tang, 2020). This institutional backdrop significantly elevates the discretionary power of top executives in tax planning. When formal rules are unclear or weakly enforced, managerial judgment becomes the primary determinant of how aggressively firms pursue tax minimization. In such settings, the personal expertise and cognitive orientation of executives take on heightened importance. Financially trained executives (particularly those from investment and commercial banks or asset management firms) possess both the technical knowledge and behavioral inclination to exploit regulatory gaps. Their ability to structure complex transactions, interpret ambiguous rules, and integrate tax strategies into broader financial policies enables them to exercise discretion more effectively than their peers. Thus, China's regulatory heterogeneity and uneven enforcement create an ideal empirical setting to observe how managerial financial experience shapes the intensity and quality of corporate tax avoidance.

Analyzing a large panel of Chinese listed firms, we find that firms led by top executives with financial expertise tend to exhibit lower effective tax rates (ETRs) and larger book–tax differences (BTDs). Economically, firms led by financially experienced top executives exhibit, on average, an ETR approximately 4.6% points lower than that of firms without such executives. Additionally, a one standard deviation increase in the proportion of financially experienced executives is linked to a 23.85% increase in the BTD. Notably, our channel analysis shows that this effect is driven by executives with experience in non-regulatory-oriented financial institutions, while those with backgrounds in regulatory-oriented institutions exhibit no significant impact on tax avoidance. Our cross-sectional analyses show that institutional ownership strengthens the negative association between financial expertise and ETR, implying that institutional investors may tolerate or even support tax minimization when implemented by top executives with financial experience, especially those with experience in non-regulatory-oriented financial institutions. We also find that this link becomes stronger following regulatory enforcement actions, suggesting a pressure-induced response whereby financially skilled executives, particularly those with experience in non-regulatory financial institutions, intensify tax planning efforts to restore firm performance. Finally, we find that tax avoidance is more positively related to firm value when undertaken by financially expert executives.

Our study contributes to the literature in several ways. First, it deepens our understanding of how managerial financial expertise shapes tax outcomes by distinguishing between top executives' financial work experience in regulatory versus non-regulatory institutions, thereby offering a novel perspective on the heterogeneity of financial experience and its implications. While prior literature has largely treated financial expertise as a uniform construct (Li et al., 2023; Lusardi and Mitchell, 2014; Custódio and Metzger, 2014; Kalelkar and Khan, 2016), our study reveals that the institutional setting in which financial skills are developed fundamentally shapes executives' normative outlook, risk tolerance, and compliance orientation. This insight not only refines our understanding of functional background effects but also underscores the importance of contextualizing managerial expertise in terms of its origins. We extend the literature on managerial financial expertise by introducing new empirical evidence on its influence over corporate tax aggressiveness, an important strategic domain that has been largely overlooked. While prior research has explored the influence of financially experienced managers on internal control quality (Oradi et al., 2020), secondary management buyouts growth performance (Jelic et al., 2019), earnings manipulation (Gounopoulos and Pham, 2018), financial reporting conservatism (Bamber et al., 2010; Custódio and Metzger, 2014), corporate social responsibility (Li et al., 2022), and investment efficiency (Li et al., 2023), the tax planning dimension has received limited attention. Our paper addresses this gap by showing that the institutional context in which financial expertise is acquired matters: financial experience from market-oriented, profit-driven institutions tends to foster more aggressive tax behavior, whereas experience gained in compliance-centered regulatory settings does not. This distinction contributes to a more refined understanding of financial sophistication within the upper echelons framework.

Second, we contribute to the research on the determinants of tax aggressiveness by offering the first empirical investigation into whether TMTs' prior work experience in regulatory versus non-regulatory financial organizations influences firms' ETRs and BTDs. While prior studies have explored how firm fundamentals, non-financial disclosure quality (Lanis and Richardson, 2012), board characteristics (Richardson et al., 2013; Wen et al., 2020), multinational companies (Otusanya, 2011), chief financial officer (CFO) social networks (Fang et al., 2025), managerial military experience (Law and Mills, 2017), corporate governance structures (Chen et al., 2010), and firms' e-commerce (Argilés-Bosch et al., 2021) influence corporation taxation, we distinguish between top executives' work experience acquired in regulatory-oriented versus non-regulatory-oriented financial institutions ^[2]. This differentiation is crucial, as it captures the heterogeneity in managerial cognition, compliance attitudes, and risk-taking tendencies that stem from formative professional environments. By incorporating this refined classification, our study advances the understanding of how individual-level attributes, beyond general financial expertise, can explain variation in corporate tax planning strategies.

Third, this study contributes to the literature on the determinants of firm value (Hasan et al., 2021; Elamer et al., 2024; Xu et al., 2020) by uncovering a novel channel through which top executives' financial expertise enhances shareholder value—namely, through the strategic implementation of tax avoidance. While prior research documents the interplay between tax planning behaviors and executives' demographic or functional backgrounds (Baghdadi et al., 2022; Chyz, 2013), it remains unclear whether such an interplay translates into value creation. Our analysis addresses this gap by showing that financially expert top managers not only engage more intensively in tax avoidance but do so in ways that are more effective in generating long-term firm value. These findings underscore the importance of managerial quality in shaping not just tax outcomes, but also the broader financial outcomes of the firm.

Upper echelons theory (Hambrick and Mason, 1984; Hambrick, 2007) posits that top executives play a central role in shaping strategic choices and organizational outcomes, as their decisions are filtered through personal interpretations of the firm's environment. These interpretations are, in turn, shaped by observable attributes such as career trajectories, functional expertise, and educational backgrounds, as well as less tangible factors including values, cognitive styles, and personality traits. Two core assumptions underpin the theory: first, that strategic choices reflect the bounded rationality and selective perceptions of top executives; and second, that these perceptions are influenced by “administrative givens,” a set of deeply ingrained characteristics accumulated through prior professional and personal experiences. Within this framework, the composition and characteristics of the TMT become critical determinants of firm behavior and performance. Empirical studies have shown that TMT members rely on their domain-specific experience and functional training to interpret strategic challenges, assess risk, and formulate responses (Benmelech and Frydman, 2015). Accordingly, the presence of financial expertise within the TMT may provide a distinctive cognitive and analytical foundation that can shape a firm's approach to financial decision-making (Li et al., 2023), including tax-related strategies.

Financial work experience can equip top executives with advanced technical knowledge in financial reporting, regulatory compliance, risk management, and capital allocation (Bodnaruk and Simonov 2015). This expertise does not merely enhance technical proficiency; it also introduces a cognitive framing through which executives view tax planning not as a routine compliance task but as a strategic tool for resource optimization and firm value maximization. Executives with financial backgrounds are typically trained to operate under conditions of uncertainty and complexity (Li et al., 2023), allowing them to navigate regulatory asymmetries, structure transactions efficiently, and align tax strategies with broader corporate goals. In this regard, financial experience functions as both a knowledge base and a decision-making lens (Custódio and Metzger, 2014), enabling executives to integrate tax planning into the firm's overall financial architecture. Their ability to evaluate and implement tax-minimizing strategies, while maintaining or even enhancing reported earnings, positions them as key agents in affecting tax aggressiveness.

To deepen this framework, we draw on the institutional logics perspective (Thornton and Ocasio, 1999; Greenwood et al., 2011), which explains how the professional environments in which executives are embedded shape their cognitive schema, behavioral preferences, and decision heuristics. Different institutional fields, such as regulatory-oriented financial institutions versus non-regulatory institutions, impart distinct logics of appropriateness that executives internalize.

We further integrate the concept of organizational imprinting (Marquis and Tilcsik, 2013), which suggests that early-career experiences in distinct institutional settings leave lasting cognitive and behavioral marks on individuals. These imprints persist over time, shaping how executives interpret strategic issues and respond to decision-making challenges. In this context, executives trained in regulatory institutions may carry forward norms emphasizing compliance and caution, while those from non-regulatory financial sectors are more likely to retain performance-driven logics that legitimize aggressive financial strategies, including tax planning.

China's institutional environment varies significantly across regions, with differences in legal enforcement, market development, and administrative capacity influencing corporate tax behavior. The NERI Marketization Index documents substantial provincial variation in legal institutions and government-market relations (Cai and Zhang, 2011; Ma et al., 2019). Inland areas typically face weaker enforcement and greater regulatory leniency (Marquis and Qian, 2013), allowing informal practices to substitute for formal regulation. In such settings, financially experienced executives—particularly those from banking and investment sectors—are adept at exploiting institutional voids to design complex tax strategies. While tax reforms in 1994 and 2008 unified statutory rates and streamlined rules, weak enforcement persists due to limited administrative capacity and inconsistent application (Tang, 2020). The 2024 amendment to China's Company Law, which requires limited liability firms to fully contribute their registered capital within five years ^[3], may increase scrutiny over capital reductions and interest deductions, raising compliance pressures and tax burdens for some firms. Financially trained executives thus play a critical role in navigating regulatory ambiguity and aligning tax strategies with reporting goals.

The variation in corporate tax avoidance has drawn growing interest in accounting and finance research, as it reflects not only regulatory arbitrage and financial engineering, but also strategic choices shaped by managerial preferences and capabilities. As noted earlier, observable traits, such as industry background and functional expertise, affect how executives assess risk, weigh trade-offs, and execute complex strategies.

Among the many dimensions of managerial heterogeneity, financial work experience, defined as prior employment in financial institutions such as commercial banks, investment banks, fund management firms, regulatory commissions, or other finance-oriented entities (Li et al., 2022), may be particularly relevant for tax-related decisions. Executives with such backgrounds possess expertise in financial reporting, risk-return analysis, and regulatory navigation. Trained to enhance firm value through sophisticated planning, they are more adept at exploiting complexities and ambiguities in the tax code. Several theoretical mechanisms explain how financially experienced executives may shape corporate tax behavior:

1. Technical competence and tax planning expertise

Taxable income originates from financial accounting earnings but is modified by country-specific rules that account for both temporary and permanent discrepancies between financial reporting standards and tax regulations ^[4]. Executives with financial experience possess specialized knowledge of financial reporting standards, tax regulations, and transaction structuring (Maletta et al., 1999), equipping them with a unique skill set to manage these differences in a way that minimizes tax liability while preserving reported earnings.

Such expertise is especially critical in the context of complex tax planning, where the effective implementation of strategies such as income deferral, accelerated depreciation, transfer pricing, and the use of hybrid financial instruments hinges on a nuanced understanding of both accounting treatments and tax law (Barrios and Gallemore, 2024). Financially experienced executives are more adept at identifying and exploiting these opportunities not only because of their technical fluency but also due to their prior involvement in the design, approval, or oversight of such tax-motivated transactions (Li et al., 2023). Importantly, this competence enables them to deliberately engineer divergences between book income and taxable income, an essential feature of aggressive yet legal tax planning. For example, through the strategic recognition of deferred tax assets and liabilities or the permanent reinvestment of foreign earnings, these executives can effectively reduce current tax obligations without significantly affecting financial reporting metrics. Consequently, firms led by financially experienced executives may be more likely to exhibit lower ETRs and larger BTDs.

1. Cognitive framing and attitude toward risk

Financial work experience plays a critical role in shaping how top executives approach strategic decision-making under uncertainty. Professionals trained in financial institutions, such as investment banks, commercial banks, trust, and asset management firms, are routinely exposed to high-stakes environments where success hinges on value maximization, sophisticated arbitrage, and the efficient deployment of capital. These formative experiences cultivate a transaction-oriented, financially optimized decision-making style that is particularly attuned to identifying and exploiting regulatory, accounting, or structural asymmetries (Li et al., 2023).

Executives with such backgrounds may be more likely to frame tax planning as an integral component of firm value creation, rather than as a compliance or administrative function. Given their familiarity with financial instruments, structuring, and regulatory arbitrage, these executives are better equipped, both cognitively and technically, to evaluate and execute tax strategies that reduce statutory tax liabilities while preserving or even enhancing reported book income (Bonner et al., 1992). This mindset enhances their receptiveness to aggressive yet legally defensible tax positions, including those that rely on transfer pricing arrangements, hybrid entities, and intercompany financing structures.

Moreover, financial experience often fosters a greater tolerance for risk and ambiguity, as decision-making in financial institutions frequently involves navigating complex regulatory landscapes and optimizing under imperfect information (Graham et al., 2013). Such conditions parallel the nature of corporate tax planning, which often entails interpreting evolving tax codes, anticipating enforcement thresholds, and balancing financial and reputational outcomes. As a result, these executives may be more inclined to authorize or support tax avoidance strategies that introduce material BTDs or contribute to lower ETRs.

1. Strategic integration of tax and financial reporting

Following Hanlon and Heitzman (2010), firms can engage in coordinated tax and financial reporting strategies that exploit the flexibility in accounting standards (GAAP or IFRS) without necessarily triggering tax liabilities. Financially trained executives are likely to understand how to simultaneously manage book income upward (to satisfy capital markets) and taxable income downward (to minimize tax burdens), thus creating significant and persistent BTDs. These executives may also structure transactions (i.e. intercompany loans, off-balance-sheet vehicles, transfer pricing arrangements) that generate tax savings without materially affecting reported earnings.

Moreover, top executives with financial backgrounds are more likely to understand how tax planning interacts with broader financial strategy, including earnings management and cash flow optimization, allowing them to embed tax-saving mechanisms within firm operations in a manner that is both effective and difficult to detect ex ante (Armstrong et al., 2012). In this sense, financial work experience does not merely facilitate compliance with tax rules; it actively enables the design of strategic tax positions that are optimized for both regulatory visibility and shareholder value. Taken together, we propose the following hypothesis:

While financial expertise broadly enhances an executive's ability to manage complex financial and tax decisions, the institutional context in which this expertise is acquired may shape how that expertise is applied. Building on the imprinting (Marquis and Tilcsik, 2013) and the institutional logics perspectives (Thornton and Ocasio, 1999), we argue that heterogeneity in the source of financial experience—specifically, whether it was acquired in regulatory-oriented versus non-regulatory-oriented financial institutions—affects how top executives perceive and engage in tax planning behavior.

Top executives with backgrounds in non-regulatory financial institutions, such as commercial banks, investment banks, securities firms, and fund management companies, typically operate in competitive, profit-driven environments where value maximization and transaction-level optimization are prioritized (Li et al., 2022). Exposure to complex financial instruments, cross-border arbitrage, and earnings management practices fosters a strategic mindset that treats tax planning as an essential lever for enhancing firm value (Bartram et al., 2011). Such executives are likely to view tax avoidance not as regulatory evasion but as financial engineering within acceptable legal boundaries. Their cognitive framing encourages proactive opportunity-seeking behavior, leading to more aggressive tax strategies such as income deferrals, hybrid structuring, and transfer pricing adjustments.

In contrast, executives whose financial work experience stems from regulatory-oriented authorities, including regulatory commissions, policy banks, and stock exchanges, are immersed in institutional environments that emphasize compliance, risk mitigation, system stability, and public accountability (Li et al., 2022). These settings tend to cultivate a rule-abiding, risk-averse orientation, where adherence to regulatory frameworks takes precedence over transactional innovation or private value extraction. Consequently, executives from regulatory backgrounds are likely to internalize a more conservative and compliance-focused approach to tax decisions. Even if technically capable of designing aggressive tax strategies, they may be less motivated to pursue such actions due to professional norms emphasizing institutional integrity and legal conformity. Thus, although financial expertise generally enhances the technical capacity for tax planning, the institutional imprinting from prior work experience conditions whether that expertise is channeled into aggressive tax behavior. Thus:

Our firm-level dataset comprises all publicly listed companies on the Shanghai Stock Exchange (SSE) and the Shenzhen Stock Exchange (SZSE) over the period 2008–2018. The sample selection proceeds as follows. We begin with the universe of Chinese A-share firms listed on the SSE and SZSE, as these two exchanges constitute the core of China's capital market and account for most publicly listed companies in the country. A-share firms are subject to uniform financial reporting standards, regulatory oversight, and disclosure requirements governed by the China Securities Regulatory Commission (CSRC), ensuring consistency and comparability across firms. We then collect data on executives' financial experience, firm-level financials, stock market performance, variables necessary for computing corporate tax aggressiveness, as well as governance and board characteristics from the China Stock Market and Accounting Research (CSMAR) database (Zhang et al., 2023; Chen et al., 2022a; Chen et al., 2021). Specifically, data on top executives' career backgrounds are obtained from the CSMAR Personal Characteristics of Directors, Supervisors and Senior Management and Positions of Directors, Supervisors and Senior Management sub-databases. We begin our sample period in 2009, as CSMAR's executive characteristics database began systematically collecting executives' functional backgrounds and employment histories in 2008, with broader data availability and improved completeness from 2009 onward. Starting from this year ensures greater consistency and accuracy in constructing key executive-level variables.

Financial variables are sourced from the China Listed Firms Research Series sub-database. Stock market information is drawn from the CSMAR Stock Market Trading sub-database. Information on board structure and governance attributes is collected from the CSMAR Corporate Governance sub-database. To ensure data quality and consistency, we exclude firms designated as Special Treatment (ST) and firms operating in the financial sector, identified by CSRC industry codes J66, J67, J68, and J69. We also restrict the sample to observations with non-missing values for managerial financial experience, tax aggressiveness proxies, and all relevant control variables. Our final sample for the regression models incorporating deferred tax expenses (i.e. ETR1 specifications) consists of 18,253 firm-year observations across 2,879 unique firms. For models excluding deferred tax components (i.e. ETR2 specifications), we obtain a slightly larger sample comprising 21,912 firm-year observations across 3,186 unique listed firms.

We employ several proxy measures, including ETRs and the BTD. Specifically, ETRs are calculated as total tax expense divided by pre-tax accounting income (Lanis et al., 2017; Hanlon and Slemrod, 2009). ETRs serve as an important indicator of a firm's ability to minimize its tax burden relative to its reported earnings and facilitate comparisons of tax avoidance behavior across firms. Firms engaging in greater tax aggressiveness typically report lower ETRs by reducing their total tax expenses (Rego, 2003). Our first proxy for tax aggressiveness (ETR1) is defined as total book income tax expense minus deferred income tax expense, scaled by pre-tax accounting income (Omer et al., 1993). The second proxy (ETR2) is measured as total tax expense divided by pre-tax accounting income (Lanis et al., 2017; Chen et al., 2010; Dyreng et al., 2008). Following prior research (Lanis et al., 2017; Chen et al., 2010), we truncate ETR values to the range of 0–1 to address extreme observations. In both measures, lower ETR1 and ETR2 values correspond to higher levels of corporate tax aggressiveness.

In addition, we employ the BTD as an alternative proxy for tax aggressiveness, recognizing that firms engaging in more aggressive tax strategies often exhibit larger discrepancies between book income and taxable income. Unlike ETRs, the BTD captures tax avoidance activities that involve deferral strategies. Our third measure, BTD1, is calculated as pre-tax accounting income minus estimated taxable income, scaled by total assets (Hanlon and Heitzman, 2010; Lanis et al., 2017). Estimated taxable income is derived by dividing total tax expense by the firm's statutory tax rate. Next, to account for the potential influence of earnings management on the book-tax gap, we adopt a fourth measure, BTD2, following Desai and Dharmapala (2006). BTD2 is derived as the residual from a regression of the BTD on total accruals (an important determinant of earnings management) thereby isolating the component of the gap more directly attributable to tax aggressiveness. It is calculated as the sum of the regression residual and the estimated constant term. In both cases, higher values of BTD1 and BTD2 indicate a greater extent of tax avoidance activities.

In our baseline econometric model testing the first hypothesis, we introduce three proxies for managerial financial experience: Financial_D, Financial_Ratio, and Financial_CEO. For the second hypothesis, we employ RegulatoryFinancial, NonregulatoryFinancial, RegulatoryFinancial_Ratio, and NonregulatoryFinancial_Ratio to further distinguish the influence of different types of financial experience. The CSMAR database defines financial experience—or a career background in finance—as work experience in stock exchanges, regulatory-related commissions, policy banks, insurance companies, commercial banks, investment banks, fund management companies, securities depository and clearing organizations, futures and trust, and investment management entities ^[5].

First, consistent with Wiengarten et al. (2017), Li et al. (2024), and Li and Liu (2025), we conceptualize the TMT as including top executives who play a central role in making key strategic decisions and overseeing critical investment and financial functions. This includes a firm's chief executive officer (CEO), executive chairperson, CFO, and managing director/manager ^[6]. These top executives may directly or indirectly influence a firm's tax aggressiveness through strategic financial planning, risk management, and resource allocation decisions, particularly those related to tax planning, leverage, and investment structuring. Financial_D is a binary indicator equal to one if a firm has at least one top executive with financial experience, and zero otherwise. To further look into the extent to which financially expert executives influence corporate tax planning, we introduce a continuous variable, Financial_Ratio, defined as the proportion of top managers with prior financial experience relative to the total number of top managers. This captures the intensity of financial expertise within the TMT and reflects the degree to which a higher concentration of such expertise may influence strategic decisions related to tax planning.

Next, we examine the effect of CEO financial experience, as CEOs typically exert greater influence over firms' reporting, disclosure policies, and strategic decision-making than other members of the TMT. Early formative career experiences can shape executives' belief systems and managerial styles, which in turn influence their behavior in key leadership roles such as the CEO position. In the context of tax planning, CEOs with a financial background are more likely to possess advanced knowledge of financial instruments and tax regulations. This expertise may enable them to identify and implement more sophisticated or aggressive tax planning approaches. Thus, we introduce an indicator variable, Financial_CEO, which equals one if the CEO of a firm has a functional background in finance and zero otherwise.

RegulatoryFinancial is an indicator variable equal to one if any top executive has prior financial work experience in regulatory-oriented institutions, such as regulatory commissions, policy banks, or stock exchanges, and zero otherwise. NonregulatoryFinancial is defined as an indicator variable equal to one if any top executive previously worked in non-regulatory financial institutions, including insurance companies, investment banks, commercial banks, fund management companies, securities depositories and clearing organizations, futures and trust entities, and investment management companies. RegulatoryFinancial_Ratio measures the share of top executives who previously worked in regulatory-focused financial institutions, expressed as a fraction of the firm's total TMT. Similarly, NonregulatoryFinancial_Ratio measures the proportion of top executives with experience in non-regulatory financial institutions.

To empirically examine Hypothesis 1, we estimate the following ordinary least squares (OLS) model with year, industry, and province fixed effects, in line with prior work (Custódio and Metzger, 2014; Chyz, 2013; Chen et al., 2022b):

Consistent with Hypothesis 1, we expect the coefficients on Financial_D, Financial_Ratio, and Financial_CEO to be significantly negative when the explained variable is either ETR1 or ETR2. When the dependent variables are BTD1 and BTD2, the coefficients on the key independent variables are expected to be positive.

Turning to the test of Hypothesis 2, we estimate the following model, classifying financial expert top executives based on their prior functional backgrounds into those with regulatory-oriented and non-regulatory-oriented financial experience:

To account for firm-level heterogeneity in tax avoidance behavior, we control for a comprehensive set of variables informed by prior literature (Gupta and Newberry, 1997; Derashid and Zhang, 2003; Cook et al., 2017; Lanis and Richardson, 2012). First, we include a binary indicator for state ownership (SOE), which equals one if the firm's ultimate controlling shareholder is the central or local government, or a related government agency, and zero otherwise. The expected effect of SOE status on tax avoidance is theoretically ambiguous. On one hand, state ownership may provide firms with political connections that can be leveraged to secure tax benefits, thereby encouraging more aggressive tax planning. On the other hand, SOE executives often face political performance incentives, in which higher tax contributions enhance promotion prospects or reputational standing. As such, managers in SOEs may overpay taxes to signal loyalty or competence to political authorities, potentially leading to higher ETRs. We also control for firm size (Size), measured as the natural logarithm of total assets. Larger firms tend to be subject to greater regulatory scrutiny and public attention, which may deter aggressive tax avoidance strategies and result in higher ETRs (Cook et al., 2017). Firm age (Age), defined as the natural logarithm of the number of years since the firm's initial public offering, is included to capture organizational maturity. Older firms may have more established governance structures and reputational concerns that reduce the incentives or opportunities for tax avoidance. Consistent with Gupta and Newberry (1997) and Lanis and Richardson (2015), we include asset tangibility, leverage, and R&D intensity. Tangibility, measured as the ratio of tangible assets to total assets, is positively associated with tax avoidance due to the availability of depreciation-related tax deductions. Lev, defined as total liabilities divided by total assets, is expected to reduce ETRs given the tax deductibility of interest expenses. RD intensity, calculated as R&D expenditures scaled by total assets, is similarly expected to be negatively related to ETRs, as such expenditures are often tax-deductible.

Firm profitability is controlled for using ROA, which equals EBIT relative to total assets. Firm growth (Growth) is captured by the year-over-year percentage change in total revenues, which may signal internal information asymmetries or investment opportunities that influence tax behavior. Given the inconsistent findings in prior studies (Gupta and Newberry, 1997; Lanis and Richardson, 2012; Derashid and Zhang, 2003), we remain agnostic about the signs of ROA and growth.

We also incorporate corporate governance variables. CEO Duality is a dummy variable set to one when the CEO simultaneously holds the position of board chair, a structure that may reduce the effectiveness of board oversight and increase managerial latitude in tax-related decisions. To capture board monitoring strength, we include Board Independence—measured by the ratio of independent directors—and Board Gender Diversity (Female), defined as the percentage of female board members. Prior research suggests that greater independence and gender diversity may improve governance and reduce the likelihood of aggressive tax strategies (Lanis and Richardson, 2012).

Following Cook et al. (2017), we include a Loss indicator variable that equals one if the firm reported a net loss in the prior year. Firms with tax loss carryforwards can offset current taxable income, resulting in lower current-year ETRs. Earnings quality is proxied by TotalAccruals, defined as net income minus operating cash flow scaled by total assets, as poor earnings quality has been linked to higher levels of tax avoidance. Finally, we control for other TMT characteristics. TMT_Age is defined as the natural log of the mean age across all members of the TMT. Managerial age may influence attitudes toward risk and compliance. We also include the proportion of TMT members with foreign work or educational experience (TMT_Foreign), which may affect exposure to international norms and institutional environments, and the percentage of TMT members with academic backgrounds (TMT_Academic), as academic experience may shape ethical orientations (Li and Liu, 2025) and reputational concerns related to tax strategies.

Year fixed effects (⁠$FEyear$⁠), industry fixed effects (⁠$FEindustry$⁠), and province fixed effects (⁠$FEprovince$⁠) are included to control for unobserved heterogeneity across time, industries, and geographic regions. All continuous variables are trimmed at the top and bottom 1% to reduce the impact of extreme values and ensure robustness against outliers. Except for Growth and Loss, all independent variables are lagged by one year to reduce potential endogeneity concerns. Standard errors are two-way clustered at the firm and year levels to account for heteroskedasticity and intra-cluster correlation (Petersen, 2009). Detailed variable definitions can be found in Table A1 of the Appendix section.

Table 1 presents the year-by-year distribution of the regression samples used for the four primary outcome variables (ETR1, ETR2, BTD1, and BTD2) alongside the corresponding prevalence of top executives with financial experience. In Panel A, the proportion of firms with at least one financially experienced executive (Financial_D) increased from 25.56% in 2009 to 31.68% in 2018, while the average share of financial experts within the TMT (Financial_Ratio) rose from 5.56% to 6.04% over the same period. Notably, after 2012, Financial_D increased from 23.77% to 31.68% in 2018, and Financial_Ratio rose from 4.46% to 6.04%, suggesting a more pronounced accumulation of financial expertise in recent years. During this same period, ETR1 declined from 20.07% to 18.02%, and ETR2 fell from 17.73% to 13.92% (Panel B). These concurrent trends provide descriptive evidence of a potential negative association between the growing presence of financially expert executives and corporate tax burdens.

Table 2 presents the summary statistics for the main variables used in the empirical analysis. The mean of ETR1 is 19.6%, with substantial cross-sectional variation (SD = 0.197), indicating heterogeneity in tax burdens across firms. A similar pattern is observed for ETR2, which averages 16.5%. The two alternative BTD measures, BTD1 and BTD2, also exhibit considerable dispersion around small positive means (0.004 and 0.003, respectively), consistent with prior literature using residual-based proxies to capture abnormal tax behavior (Desai and Dharmapala, 2006). Regarding managerial characteristics, approximately 25% of Chinese listed firms in the sample have at least one top executive with financial work experience (Financial_D), while the average proportion of financial experts on the TMT (Financial_Ratio) is 5.0%. Approximately 5.5% of sample firms are headed by Financial_CEO. In terms of firm-level controls, the average leverage ratio (Lev) is 44.5%, and about 41.5% of the sample is composed of state-owned enterprises (SOE). Board independence averages 37.1%, while CEO duality is observed in 24.5% of firms. These characteristics are consistent with McGuinness et al. (2017) and Li et al. (2022).

To test Hypothesis 1, we estimate Eq. (1) and report the empirical results in Table 3. In Panel A, we employ four distinct tax avoidance proxies as dependent variables—ETR1, ETR2, BTD1, and BTD2—to ensure robustness across both ETR and BTD measures. Columns (1) and (2) show that Financial_D and Financial_Ratio are both negatively and highly significantly associated with ETR1. Specifically, in Column (1), the coefficient on Financial_D is −0.009 (t = −2.813), while the mean value of ETR1 reported in Table 2 is 0.196. This implies that, on average, firms led by financially experienced top executives have an ETR approximately 4.6% lower (i.e. −0.009/0.196) than firms without such executives. In Column (2), the coefficient on Financial_Ratio is significantly negative (−0.031, t = −2.290). Columns (3) and (4) reveal similar patterns using ETR2, which includes deferred taxes. Both proxies for financial experience remain negative and statistically significant, reinforcing the conclusion that financial expertise is systematically associated with lower tax burdens and lending support to our central hypothesis.

Columns (5) through (8) of Panel A report regression estimates using two alternative proxies for tax aggressiveness: BTD1 and BTD2, which capture the BTD and its discretionary component, respectively. In Columns (5) and (6), where the explained variable is BTD1, the coefficient on Financial_D is significantly positive, indicating that the presence of financially experienced executives is related to a higher book–tax gap. This finding implies that such executives are more likely to engage in tax strategies that widen the divergence between financial and taxable income—an accepted signal of tax aggressiveness. Similarly, the estimate on Financial_Ratio in Column (6) is significantly positive (0.009, t = 3.584). Given that the standard deviation of Financial_Ratio is 0.106, a one-standard-deviation increase in this variable is associated with a 0.000954 increase in BTD1 (0.009 × 0.106). Considering that the mean of BTD1 is 0.004, this effect represents an absolute change of approximately 0.10% points, or about 23.85% of the sample mean. Turning to BTD2, which isolates the discretionary portion of the book–tax gap by controlling for accrual-based earnings management, the results in Columns (7) and (8) confirm the earlier findings.

Turning to Panel B of Table 3, which examines whether the presence of a CEO with financial experience influences corporate tax behavior, we find consistent evidence of heightened tax aggressiveness. Specifically, firms led by financially experienced CEOs exhibit significantly lower ETRs and higher BTDs. These findings align closely with our central hypothesis that financial expertise within the TMT facilitates more aggressive, yet likely legally defensible, tax positions ^[7]. This behavior is consistent with the theoretical mechanisms we propose: technical tax planning competence, greater risk tolerance shaped by prior financial institutional training, and the strategic integration of tax planning with broader financial and reporting objectives.

In this sub-section, we employ two sets of explanatory variables to test our second hypothesis. We examine whether financial experience acquired in regulatory-oriented institutions versus non-regulatory-oriented financial institutions has differential implications for tax avoidance, to further identify the channel through which financial work experience shapes tax planning behavior. To this end, we replace Financial_D in Eq. (1) with (1) RegulatoryFinancial and NonregulatoryFinancial, and re-estimate the model. We then repeat the analysis using (2) RegulatoryFinancial_Ratio and NonregulatoryFinancial_Ratio as alternative proxies. The estimation results are presented in Table 4.

Taking Column (1), where the dependent variable is ETR1, as an example, the estimate on NonregulatoryFinancial is negative and significant, meaning that companies headed by top executives with experience in non-regulatory financial institutions tend to have lower ETRs. In contrast, the estimate on RegulatoryFinancial is statistically insignificant, suggesting that prior financial experience in regulatory authorities has no impact on ETR outcomes. A similar pattern emerges in Column (2), where we use the continuous measures, RegulatoryFinancial_Ratio and NonregulatoryFinancial_Ratio, to capture the proportion of the TMT with each type of financial background. Again, NonregulatoryFinancial_Ratio is significantly negative, reinforcing the interpretation that a higher concentration of non-regulatory financial expertise within the TMT is associated with more aggressive tax behavior. Meanwhile, RegulatoryFinancial_Ratio remains insignificant, further supporting the view that regulatory-origin experience does not systematically influence firms' tax planning decisions. We observe similar results in Columns 3 and 4, where ETR2 is employed as the outcome variable.

Turning to the BTD measures, Columns (5)–(8) of Table 4 display findings using BTD1 and BTD2 as the explained variables, which capture overall and discretionary tax aggressiveness, respectively. In Column (5), NonregulatoryFinancial is significantly positive, implying larger book–tax gaps for firms with non-regulatory financial experts. RegulatoryFinancial is again insignificant. Columns (6) and (8) reinforce this pattern: NonregulatoryFinancial_Ratio is positively and significantly associated with both BTD1 and BTD2, while RegulatoryFinancial_Ratio has no such effect.

These findings support Hypothesis 2, which posits that while financial expertise enhances technical capacity, only experience gained in market-driven financial institutions is channeled toward tax avoidance, likely due to differing professional norms and risk orientations. Executives from regulatory backgrounds, despite similar technical capabilities, appear less inclined to deploy that expertise in ways that reduce effective tax burdens—possibly reflecting a greater emphasis on compliance and institutional accountability developed in their prior roles.

As shown in Panel A of Table 1, our sample is an unbalanced panel and unevenly distributed across years. Thus, we apply weighted least squares (WLS), assigning each observation a weight proportional to its precision. This approach ensures that more reliable data points exert greater influence on the parameter estimates. The WLS results, presented in Table 5, are consistent with our baseline findings. With respect to Hypothesis 1, the coefficients on Financial_D (Model 1) and Financial_Ratio (Model 2) remain significantly negative, reinforcing the conclusion that firms led by financially expert executives pursue more aggressive tax strategies, reflected in lower ETRs. In Model 3, the estimate on RegulatoryFinancial is insignificant, whereas that on NonregulatoryFinancial is significantly negative. Similarly, in Model 4, RegulatoryFinancial_Ratio attracts an insignificant estimate, whereas NonregulatoryFinancial_Ratio attracts a significantly negative one. These results reaffirm Hypothesis 2.

To mitigate potential selection bias stemming from systematic differences in observable characteristics between firms with and without financially expert top executives, we employ a propensity score matching (PSM) approach. Specifically, we perform nearest-neighbor matching without replacement, applying a caliper of 0.00001. For each firm-year featuring a top executive with a financial background, we identify a matched firm-year lacking such an executive, based on the set of control variables outlined in Eq. (1), along with fixed effects for year, industry, and province. This matching strategy improves comparability between treated and control groups across observable dimensions. We then compute the average treatment effect on the treated (ATT), which quantifies the mean difference in ETRs attributable to the presence of financially expert executives. To further validate our findings, we re-estimate Eq. (1) using the matched sample to test Hypotheses 1 and 2. As shown in Table 6, Panel A, firms with financially expert top executives report significantly lower ETRs relative to their matched counterparts (ATT = 1.33%; t = 2.10), supporting the argument that financial expertise contributes to more aggressive tax behavior ^[8].

In Panel B, the coefficients on Financial_D (Model 1) and Financial_Ratio (Model 2) are significantly negative, reaffirming Hypothesis 1. In Model 3 (and Model 4), the estimates for RegulatoryFinancial (RegulatoryFinancial_Ratio) are statistically insignificant, whereas the coefficients on NonregulatoryFinancial (NonregulatoryFinancial_Ratio) are significantly negative. These findings lend further support to Hypothesis 2.

To further address the concern related to selection bias, we implement the entropy balancing approach, a reweighting technique that achieves covariate balance across treatment and control groups by assigning weights to control observations so that their covariate distributions exactly match those of the treated group (Wilde, 2017). Unlike PSM, entropy balancing retains the full sample and achieves exact mean moment balance, offering greater efficiency and robustness.

As displayed in Panel A of Table 7, pre-treatment differences exist in variables such as SOE status, growth, and foreign experience. After reweighting (Panel B), covariate means are perfectly balanced across groups. Re-estimating our regressions on the weighted sample (Panel C) confirms the robustness of our main findings. Financial_D and Financial_Ratio remain negatively and significantly associated with ETR1. Moreover, only financial experience from non-regulatory financial institutions—not regulatory backgrounds—is linked to lower ETRs, reinforcing both hypotheses.

To further mitigate concerns regarding simultaneity, unobservable heterogeneity, and potential reverse causality, we employ the dynamic system generalized method of moments (GMM) estimator (Arellano and Bover, 1995; Blundell and Bond, 1998). Corporate tax aggressiveness is likely persistent over time, and the financial work experience of top executives may be endogenously determined by unobserved firm characteristics or reverse causality. System GMM addresses these issues by using internal instruments to control for dynamic panel bias, simultaneity, and omitted variable bias, thereby yielding more consistent and reliable parameter estimates. Specifically, we incorporate one-year-lagged ETRs as an explanatory variable in Eq. (1) and apply the dynamic system GMM estimator with year, industry, and province fixed effects (Wintoki et al., 2012). Following the procedures outlined by Kim et al. (2015) and Fan et al. (2019), we use lagged levels of the endogenous variables as instruments for the differenced equation, and lagged differences as instruments for the level equation. In our model, we employ continuous measures—△Financial_Ratio, △RegulatoryFinancial_Ratio, and △NonregulatoryFinancial_Ratio—as the key independent variables, rather than binary indicators such as Financial_D, RegulatoryFinancial, or NonregulatoryFinancial. This choice is methodologically preferable because continuous variables provide richer within-firm and over-time variation, which enhances the strength of internal instruments and reduces weak identification problems commonly encountered with discrete regressors in system GMM frameworks.

Table 8 shows that in Columns (1)–(2), the coefficients on △Financial_Ratio and △NonregulatoryFinancial_Ratio remain significantly negative, supporting both hypotheses. Also, the Hansen test yields a p-value of 1.000 for the difference-in-Hansen test of instrument exogeneity, reinforcing the validity of the selected instruments.