Tax governance as a social responsibility of the firm: evidence from the Italian cooperative compliance program

Corporate tax payments are a fundamental social responsibility behaviour (Dowling, 2014) that has raised increasing public scrutiny (Dyreng et al., 2016) and customer awareness (Mansi et al., 2020). Despite tax contributions representing the most critical method through which corporations interact with wider societal contexts (Christensen and Murphy, 2004), the combined complexity of financial reporting and tax law provides ample scope for tax planning arrangements, posing substantial moral dilemmas (Lanis and Richardson, 2012a).

As a result, when corporations neglect to pay a “fair” share of taxes to government bodies (Lanis and Richardson, 2012b), consequently undermining public goods financing, they essentially dodge their social obligations by redirecting the tax burden to individuals and smaller enterprises exacerbating social inequality (Freedman, 2003; Landolf, 2006). Thus, tax aggressiveness is regarded as socially irresponsible (Christensen and Murphy, 2004; Erle, 2008; Schön, 2008). As an issue of major public concern, tax aggressiveness is a strategy that is incompatible with community expectations of responsible corporate behaviour and highlights the inconsistencies or decoupling between corporate sustainability commitment and deliberate corporate practices to avoid taxes (Sikka, 2012; Zeng, 2019), with significant and potentially irrecoverable loss to society as a whole (Slemrod, 2004).

In an attempt to foster corporate compliance with tax collection laws, Cooperative Compliance (CC) programs were promoted first by the OECD (2008, 2013) and later by the EU (European Commission, 2020), based on cooperation rather than confrontation. CC programs shed light on governance and social dimensions by transforming tax compliance from a punitive obligation into a collaborative social mechanism that promotes mutual understanding and shared responsibility, promoting transparency, equity and shared responsibility for societal infrastructure and economic stability (OECD, 2013).

To be entitled to enter the CC program, firms are generally required to implement a robust Tax Control Framework (TCF) ^[1] as a tax governance measure that not only ensures compliance but also reflects a broader commitment to full operational transparency, demonstrating their social responsibility and ethical engagement with the broader community (GRI, 2019). Tax administrations, in turn, respond with increased support and guidance, creating a relationship built on trust, reciprocal respect, and shared economic goals. By fostering transparency, reducing uncertainty, and preventing tax litigation, CC programs create a more inclusive and cooperative approach to tax responsibility that benefits both individual organisations and society as a whole. Ultimately, these programs demonstrate that paying taxes is not just a legal requirement but a critical governance mechanism and a firm’s social responsibility, enhancing cohesion, economic development and collective progress with a multi-stakeholder approach (Owens and Pemberton, 2021).

While CC programs have gained traction globally, empirical evidence regarding their effectiveness remains limited and relies mostly on survey data and interviews (De Widt and Oats, 2017; Eberhartinger and Zieser, 2021; Siglé et al., 2022). This study focuses on Italy’s CC program, implemented in 2015 as part of a broader initiative to modernise tax administration and improve tax certainty for large businesses. The study investigates the Italian CC program (Regime di Adempimento Collaborativo), a tax governance initiative designed to foster transparency and collaboration between firms and tax authorities. The research explores two main questions:

The findings suggest that firms are more likely to join the CC program if they are larger (measured by total assets), more profitable (higher return on assets) and less reliant on debt (lower leverage). These characteristics suggest that firms with strong financial health and public visibility are more inclined to adopt the program, likely because they value the benefits of tax certainty and transparency.

The study uses statistical methods to compare firms that joined the CC program with similar firms that did not. After joining the CC program, firms pay more taxes and engage in less tax avoidance. This is measured by an increase in the proportion of income paid as taxes (effective tax rates), indicating that the program discourages aggressive tax-saving strategies. Surprisingly, firms show a drop in their ESG scores after joining the program. This suggests a substitution effect, where firms focus on tax compliance (a governance-related ESG activity) but reduce efforts in other ESG areas, such as environmental or social initiatives.

The contribution of this study to prior literature is twofold. To the best of our knowledge, this is the first study to use financial statement data to show whether firms entering a CC program decrease tax avoidance outcomes. Prior research relied on survey data, interviews, and tax audits. Results suggest that CC programs are effective in curbing tax avoidance and promoting responsible tax governance behaviour. Second, the study contributes to the debate on the relationship between tax avoidance and ESG performance (Hoi et al., 2013; Lanis and Richardson, 2012a; Watson, 2015; Davis et al., 2016; Col and Patel, 2019), providing evidence in support of the substitution effect (Zeng, 2019) and shedding light on the unintended consequences of CC programs on firms’ ESG performance quality.

The remainder of the paper is structured as follows. Section 2 describes the institutional setting. Section 3 discusses extant literature and describes the development of the hypothesis. The methodology is illustrated in Section 4. Section 5 reports the results, which are discussed in Section 6. Section 7 concludes.

In corporate taxation, the interaction between tax authorities and taxpayers is frequently perceived as contentious. However, the primary objective of tax authorities is not to penalise taxpayers but to collect the correct amount of tax in a timely manner, following the rule of law while minimising the burden on taxpayers (Sofrona et al., 2025). Accordingly, the OECD and its member countries have initiated a reassessment of the relationship between tax authorities and taxpayers in recent years. The OECD (2008) draws some features of a correct, mutually beneficial relationship between revenue authorities and taxpayers, initially called an enhanced relationship. The concept of CC develops further (OECD, 2013) as transparency in exchange for certainty. In other words, the taxpayer commits to be fully transparent in exchange for certainty in the tax treatment of its operation. This means that the taxpayer should disclose the information required by the legislation and all the information that the tax administrations could find useful to assess the situations related to the taxpayer. The name itself is intended to state that the approach is based on cooperation rather than confrontation, but without losing the focus on compliance, meaning collecting the right amount of taxes at the right time, as required by the law (OECD, 2013).

A CC program was permanently introduced in Italy in 2015 (Legislative Decree n. 128/2015, following Law n. 23/2014), requiring companies to have an effective TCF in place to participate in the program. The requirements for the TCF are laid out in a provision of the Italian tax authority, which includes criteria such as promoting honesty and fairness, well-documented tax strategy, clear division of roles and responsibilities, effective procedures for managing tax risks, adaptability to internal and external changes and an annual management report. To be eligible to participate in the program, resident and non-resident entities having a permanent establishment in Italy must have a total turnover or operating revenues no lower than 10 billion euros, reduced to 5 billion euros for fiscal years 2020 and 2021 ^[2]. The firm must also possess a proper TCF and provide detailed information about its activities, tax strategy, TCF processes and tax risks. The tax administration verifies whether the company meets the criteria and may visit the company’s facilities to verify the coherence of the informational and control system. If the company meets all the required criteria, it participates in the program, and its name is published on the Italian tax authority website. From a firm’s perspective, the program offers several benefits to participants, including the opportunity to discuss difficult tax issues before filing tax returns and to expedite the inquiry process. This could be particularly advantageous for companies dealing with complex tax situations, such as transfer pricing and cross-border transactions, as it minimises the risk of disagreement with the tax authority on interpreting tax laws after filing. The expedited inquiry process allows participants to seek clarification from the revenue authority on real situations they may encounter in their business activities, with a response time of no more than 45 days. Other advantages are reductions in tax penalties by 50% and no requirement of guarantees to obtain tax refunds.

Other than Italy, several countries have implemented CC programs following similar principles but with variations in design, eligibility criteria, and operational features. Below, key characteristics of the CC program in Italy are compared with other major initiatives in the European Union, namely those of The Netherlands, Denmark and Austria. The evolution and design of CC programs vary across countries, even if influenced by OECD and EU guidelines, reflecting different administrative priorities, tax cultures and implementation timelines. Table 1 presents a comprehensive comparison of key features across four European CC programs, including main studies investigating these regimes (see also Section 3 below).

Some differences emerge when comparing the CC programs across these countries. First, the Italian tax authority is currently the only one in the EU with a public list of firms using the CC program. This characteristic makes it a unique setting suitable for archival research. Second, while the Netherlands pioneered the CC approach with its Horizontal Monitoring program, the Italian program is relatively recent. This temporal difference helped the Italian government align the CC regulation from the beginning with an already established set of guidelines (OECD, 2013) set up at the international level (Manca, 2016). In this respect, the Austrian CC program can be considered closer to the Italian one, as modelled after the most recent OECD standards (Fiala and Ramharter, 2019). Third, Italy’s program initially targeted only the largest corporations with its €10bn threshold, which was significantly higher than in other countries, like Austria and Denmark, having significantly lower thresholds. The subsequent threshold reductions indicate the regulator’s intention to broaden participation, similar to the Dutch model’s evolution towards greater inclusivity. Adoption requirements highlight the higher entry threshold in Italy, meaning that the sample represents the largest corporations, potentially with characteristics different from those of participating firms in countries with lower thresholds. Although this selective approach may affect which firms participate, it allows for collecting more accurate information as large firms’ data are more accessible and reliable. Finally, the approach to risk assessment is slightly different across countries, where the Netherlands and Denmark favour a relationship-based approach (Boll, 2017; De Widt and Oats, 2017) wherein transparency, trust and the continuous, direct dialogue between the tax authority and the firms shape the characteristics of the firm’s internal control system through self-assessment without necessarily referring to a pre-existing TCF, certified according to rigorous guidelines and formal requirements (De Widt, 2017). On the contrary, Austria and Italy, being more recently codified, rely heavily on the OECD’s building blocks of the TCF and on the existence of the TCF itself as a key requirement for initial and subsequent risk assessment (OECD, 2016). These differences are also related to legal framework differences, the degree of tax law complexity (Schipp, 2024) in each jurisdiction, and pre-existing relationships between tax authorities and large firms. However, even if it is impossible to define a one-size-fits-all CC framework valid throughout all countries, all these programs share the same goals and principles. They are – to various extent – all gradually attracted towards the OECD model, as demonstrated by the evolution of the Danish program in recent years (Björklund Larsen et al., 2018). The EU’s cooperative compliance framework – the European Trust and Cooperation Approach – is anticipated to enhance this convergence process (Russo et al., 2022).

Prior research frames tax avoidance as an issue of corporate social responsibility (Bird and Davis-Nozemack, 2018). According to this perspective, tax avoidance is considered a socially irresponsible practice that conflicts with a firm’s societal obligations (Lanis and Richardson, 2015; Avi-Yonah, 2014; Dowling, 2014). Firms are morally obligated to refrain from such behaviour and align their tax compliance with stakeholders’ ethical and social expectations (Scheffer, 2013; Sikka, 2012). This approach, in turn, strengthens society’s ability to apply social pressure for conformity, grounded in a recognised and legitimised framework of corporate accountability.

The view of taxation as an element of corporate socially responsible policies is related to the perception that the most tax-aggressive behaviours are costly to society (Weisbach, 2002). Moreover, taxes are essential for macroeconomic stability, reducing inequalities, and financing the transition to a low-carbon economy (GRI, 2022). Thus, tax avoidance practices are considered inconsistent with good corporate social responsibility practices, as firms should not engage in (unethical) activities negatively affecting society.

CC programs represent governance devices which help firms address the sustainability concerns brought about by aggressive, real or perceived, tax avoidance practices. By establishing an appropriate TCF, firms should be “in control” of their tax position and prevent any practice not in line with their ESG policies ^[3]. The approach transforms tax management from a purely financial exercise into a responsible corporate strategy that proactively prevents practices inconsistent with ethical standards. Through structured internal control systems, CC programs mitigate reputational and legal risks while demonstrating a commitment to transparency and social responsibility. By integrating financial decisions with ethical considerations, CC programs support firms in maintaining a balanced approach to tax management that considers broader societal impacts and stakeholder expectations, ultimately contributing to more sustainable and accountable business practices.

The literature on CC programs predominantly focuses on legal aspects, with empirical studies being relatively scarce, making research on CC programs an emerging field of study. Research from the Netherlands analysing the Dutch CC program (the Horizontal Monitoring Project) presents a nuanced perspective, with Huiskers-Stoop (2015) suggesting that CC programs might formalise pre-existing differences in tax attitudes due to self-selection effects. However, subsequent research by Siglé et al. (2022), incorporating both survey data and tax audit records, indicates that the Dutch CC program fosters improved working relationships, enhances internal controls and transparency and boosts income tax compliance, though effects on VAT compliance remain limited. Complementary research by Goslinga et al. (2019) further underscores the critical importance of increased tax certainty. In the Danish setting, Boll and Brehm Johansen (2018) reveal a nuanced corporate perspective on CC programs, characterised by a complex interplay between supportive attitudes and practical implementation challenges. The Austrian CC program (the Horizontal Monitoring Pilot Project) presents similarly nuanced insights. Elmecker et al. (2016) report substantial goal achievement, while Enachescu et al. (2019) emphasise the significance of legal and planning certainty. Their research also identifies CC as a transformative paradigm shift for tax administration, highlighting the complex organisational implications of such programs. Recent survey-based research in Austria yields promising insights, with Eberhartinger and Zieser (2021) highlighting that firms perceive enhanced legal certainty and planning security as key benefits, alongside reductions in tax risk and compliance costs. Specifically, they use survey data from participating firms and a comparative control group subjected to conventional ex-post audit regimes. Empirical findings reveal significant insights into the transformative potential of such programs. Firms engaged in the CC program consistently report substantially increased tax certainty, with measurable reductions in perceived tax risk and compliance-related expenditures. Notably, while these firms demonstrate superior initial tax risk management capabilities, subsequent improvements do not diverge significantly from the control group.

While extant literature explores the organisational and risk management consequences of entering CC programs, there is still limited evidence on the firms’ fundamentals that affect the decision to enter the CC program. Analysing the financial characteristics that might influence firms’ likelihood of adopting the regime allows us to investigate their motives for such a decision. This leads to the following research hypothesis:

While CC programs are designed to prevent unintentional compliance errors by adopting TCFs, these frameworks might be leveraged to obscure intentional non-compliance and could be strategically employed to enhance firm value through less compliant tax strategies (Chen et al., 2020). In the US setting, Beck and Lisowsky’s (2014) analysis suggests that firms’ decisions to participate in CC programs are influenced by their assessment of tax position uncertainty, with potential implications for subsequent tax risk management and transparency ^[4].

Based on theoretical predictions, managers should be increasingly responsible for a firm’s tax affairs towards the stakeholders by adopting a CC program. In light of this, managers need to balance the risky tax savings from tax avoidance activities against an increased tax certainty. To the best of our knowledge, no prior study has investigated this issue from a quantitative standpoint using financial statement information. This leads to the following research hypothesis:

Firms are expected to show lower levels of tax avoidance after entering the program using multiple tax avoidance metrics (Phillips, 2003; Rego, 2003; Dyreng et al., 2008; Badertscher et al., 2019). The prediction is that the CC program enhances tax enforcement and discourages firms from entering tax-aggressive schemes.

Given the sustainability implications of aggressive tax practices, prior research has investigated the relationship between ESG performance and tax avoidance. Studies on this topic can be divided into those finding that ESG performance and tax avoidance act as complements (Hoi et al., 2013; Lanis and Richardson, 2012a, 2015) and those finding a substitution effect between the two (Watson, 2015; Davis et al., 2016; Col and Patel, 2019). Building on this debate, Chircop et al. (2018) investigate whether the level of social capital in the region where a firm is headquartered affects its tax avoidance activities. Their findings indicate that firms headquartered in areas with high social capital engage significantly less in tax avoidance activities, consistent with the idea that managers view corporate tax payments as a socially responsible action. In an international setting, Zeng (2019) finds strong evidence that Corporate Social Responsibility (CSR) is positively related to tax avoidance. She also finds that firms with higher CSR scores engage in less tax avoidance in countries with weak country-level governance, implying that CSR and country-level governance are substitutes. This leads to the following research hypothesis:

The sample is constructed as follows. First, all Italian companies that entered the program are selected. Since the program began in 2015, the analysis investigates firms that entered the CC program (i.e. the treatment group) in the period from 2015 (the beginning of the program) to 2020 (the last year in the observed data). The sample period for the treatment group (2015–2020) is chosen because the CC program was introduced in Italy in 2015, and firms could join the program at different times during this period. The end of 2020 is selected as the cut-off to ensure sufficient post-treatment observations for firms that joined later in the sample period. This timeframe captures the staggered nature of program adoption and provides a robust basis for analysing its impact ^[5].

The initial sample is composed of 73 firms. Firms in the banking, financial services, and insurance industries are excluded from the analysis, consistent with prior tax avoidance studies, leaving a sample of 40 industrial firms. The analysis period covers the years 2014–2021. The inclusion of 2014 as a pre-treatment year allows for a baseline comparison of firm behaviour before the program’s introduction and for the construction of financial variables requiring lag values. Similarly, the year 2021 is included to allow for the construction of financial variables that require lead values. Thus, the final sample contains 245 firm-year observations, with all non-missing observations for firms that enter the CC program and 1,466 firm-year observations, with all non-missing observations for firms in the treatment and control groups. Financial data is retrieved from AIDA – Bureau Van Dijk; the list of companies admitted to the CC program is hand-collected from the official list published on the Italian Tax Authority website (www.agenziaentrate.it); ESG score is retrieved from the S&P Global Market Intelligence database.

This study employs a quantitative research method to analyse the impact of the CC program on tax avoidance and ESG performance. A quantitative approach is appropriate because it allows for the use of statistical and econometric techniques to measure causal relationships and control for confounding factors. Since the CC program is adopted on a voluntary basis, entities included in the analysis are not randomly selected but instead choose to participate based on their own characteristics, preferences or circumstances. This can lead to a sample that is not representative of the broader population, potentially distorting the results and making it difficult to determine causal relationships (i.e. self-selection bias). To address this issue, the study uses statistical methods (i.e. difference-in-difference approach) that compare firms in the program (treatment group) with a similar group of firms that did not join (control group) based on their financial characteristics.

Specifically, the staggered difference-in-difference (DiD) method is used to isolate the program’s effect by comparing changes over time in firms that adopted the CC program (treatment group) with those that did not (control group). This method is widely recognised for its robustness in evaluating policy interventions. Once firms enter the program, they remain treated throughout the observation period in line with the method’s identifying assumptions (Baker et al., 2022). Furthermore, in robustness checks (untabulated), the parallel trend assumption is tested.

The research adopts a quasi-experimental design, leveraging the staggered implementation of the CC program across firms. This design is justified because it allows for causal inference in a non-randomised setting by comparing treated and untreated firms over time. Matching techniques (i.e. Propensity Score Matching – PSM) are also employed to ensure that the treatment and control groups are comparable based on financial characteristics. This strengthens the validity of the findings by minimising selection bias.

In the first model, a Probit regression is used to investigate the likelihood of firms entering CC programs. Equation (1) displays the regression model:

Y_i,t is the outcome variable (e.g. Adoption and Early Adoption). The dummy variable Adoption is set to 1 for firms under the CC program, and the dummy Early Adoption is set to 1 for firms that adopted the CC program during its initial year of implementation. This design allows us to identify the firms that adopted the CC regime as soon as possible, labelling them the “Early Adopters” and those who adopted the regime in the following years. This distinction is of utmost importance as an early adoption signals a stronger commitment to good tax governance.

The regression equation includes all variables predicted to have a relationship with the firm’s propensity to enter the program, such as size, profitability, growth, leverage, intangible intensity and listing status. To select financial characteristics associated with the likelihood of entering the regime, the study draws on extant literature on tax avoidance. Vector X_i,t includes the following firm financial variables. Size (Size) is the natural logarithm of total assets (Zimmerman, 1983; Plesko, 2003). Profitability (ROA) is measured as the ratio of pre-tax income and total assets (Gupta and Newberry, 1997; Plesko, 2003). Growth (Growth) is measured as the annual change in revenues (Hoi et al., 2013). Firm risk (ROA volatility) is measured as the standard deviation of a firm’s ROA as defined above (Nguyen and Nguyen, 2015). Leverage (Leverage) is the ratio between long-term debt and total assets (Plesko, 2003; Chen et al., 2010). Intangible intensity (Intangibles) is measured as the ratio between intangible assets and lagged total assets (Hoi et al., 2013). Listing status is measured through a dummy variable (Listed) equal to one when the firm is listed on the Italian Stock Exchange. ε_i,t is the error term that captures unexplained variation in the dependent variable, representing factors like omitted variables, random noise or measurement errors that influence the outcome but are not included in the model.

Secondly, this study aims to investigate the relationship between the adoption of the CC program and tax avoidance by resorting to some of the most commonly adopted tax avoidance measures ^[6] (Phillips, 2003; Rego, 2003; Dyreng et al., 2008; Badertscher et al., 2019). To test HP2, the analysis examines how joining the CC program affects firms’ tax avoidance behaviour using a DiD estimation model. This approach compares changes in tax behaviour over time between firms that joined the program (treatment group) and similar firms that did not (control group). By focusing on differences before and after joining the program, the model helps isolate the program’s impact on reducing tax avoidance. The control group is composed of matched non-adopting firms. The matching is conducted on a sample of Italian industrial firms using PSM based on pre-treatment averages of variables Size, ROA, Growth, Leverage, Intangibles, and Listed:

Y_i,t is the outcome variable (e.g. effective tax rate or ESG score) for firm i at time t. Adoption_i,t is a binary variable indicating whether firm i participated in the CC program at time t; X_i,t represents control variables; ε_t captures time fixed effects to control for macroeconomic changes; µ_i captures firm fixed effects to account for unobserved, time-invariant characteristics ^[7]; and ε_i,t is the error term which captures unexplained variation in the dependent variable, representing factors like omitted variables, random noise or measurement errors that influence the outcome but are not included in the model. In line with previous studies on tax avoidance, vector X_i,t includes firm size, profitability, growth, firm risk, leverage and intangible intensity (Zimmerman, 1983; Wang, 1991; Mills et al., 1998; Keating and Zimmerman, 2000; Plesko, 2003; Chen et al., 2010).

To test HP3, equation (2) is rerun, substituting ETR with the ESG Score. The latter analysis is limited to the sample of firms entering the CC program.

Table 2 Panel A shows the descriptive statistics for both the adopting firms and the matched sample of non-adopting firms. Results show that the average current ETR for firms in the sample is 31%, while the median value is 29%. These values are close to the GAAP ETR, measured on taxes accrued (mean value of 32% and median value of 30%). The average firm in the sample has a ROA of 6% and an annual growth rate of 14%. Firms in the sample show relatively low debt levels, 14% of total assets on average and 16% of firms in the sample are listed. The correlation matrix shows no collinearity problems (Table 2 Panel B). Current and GAAP ETR measures show a strong positive and significant correlation. However, the two measures are used interchangeably. Among the other coefficients, size shows a relatively strong and positive (negative) relationship with listing status (ESG score). All other couples of variables show correlation coefficients that are in the range [−0.4, 0.4] or are not statistically significant. However, the maximum VIF values (untabulated) suggest that collinearity does not affect our results.

Table 3 presents the results of probit regressions to evaluate the likelihood of firms entering the CC program [equation (1)] and test HP1. The probit regression in column (1) uses the variable Adoption as a dependent variable, while the regression in column (2) uses the dummy Early Adoption. Both specifications include year fixed effects. These dummies are the terms of interest for testing HP1. Results reveal that both Adoption and Early Adoption are positively associated with the size and profitability of a firm, while they show a negative association with leverage. Furthermore, firms with higher risk profiles (ROA volatility) are more inclined to adopt the CC regime earlier. Notably, the status of being listed on a stock exchange appears to positively influence early adoption, although it has a negative impact on overall adoption rates. Thus, results confirm HP1.

Table 4 columns (1), (2) and (3) presents the results of the staggered difference-in-difference estimation model aimed at evaluating the impact of the CC program on firms’ tax avoidance practices and testing HP2. The regression in column (1) uses the current ETR as the dependent variable, while regressions in columns (2) and (3) use the deferred ETR and GAAP ETR, respectively. All regressions include firm and year fixed effects. The term of interest is the variable Adoption, which accounts for staggered adoption. The coefficient of Adoption is expected to be positive, meaning that firms exhibit lower tax avoidance after entering the CC program.

Results indicate that firms tend to reduce their tax avoidance practices, as evidenced by an increase in the current ETR when considering current income taxes. The decrease in the deferred ETR also suggests that firms should improve tax transparency, reduce tax deferral strategies after entering the CC program and improve temporary book-tax conformity (Hanlon and Heitzman, 2010). This effect is reflected in the GAAP ETR metric coefficient in column (3) (not significant), which includes both current and deferred taxes: the former increases as a reduction of aggressive tax avoidance strategies, and the latter decreases due to the re-alignment of book and tax bases. Hence, HP2 is confirmed since adopting the CC program induces a change in the tax avoidance strategies of participating firms.

Finally, Table 4 column (4) investigates the interplay between ESG performance and tax compliance to test our HP3. The ESG score is used as the dependent variable, and firm and year fixed effects are included. The term of interest is the variable Adoption, which accounts for staggered adoption. The coefficient of Adoption is expected to be positive in case of a complementary relationship between ESG and tax compliance (Hoi et al., 2013; Lanis and Richardson, 2012a, 2015) or negative in case of a substitution effect (Watson, 2015; Davis et al., 2016; Col and Patel, 2019; Zeng, 2019). Results indicate that firms tend to reduce ESG performance, as proxied by the ESG score, after entering the CC program, providing evidence of a substitution relationship in line with prior research (Zeng, 2019).

Untabulated analysis tests the robustness of our specifications to the parallel trend assumptions of the difference-in-difference setting. In line with Autor (2003), lead and lag identification and control variables are included in a time window from time t − 1 to time t + 2. Thus, our sample size is reduced to 1,066 firm-year observations. The insignificant coefficients on the lead and lag identification variables corroborate our previous results. The coefficients on the current Adoption are consistent with our main results in Table 4, columns (1), (2) and (3), statistically significant and robust to the inclusion of firm and year-fixed effects. Moreover, its economic magnitude is consistent with our main results in Table 4.

The study finds that adopting the CC program significantly reduces tax avoidance among participating firms. This is evidenced by an increase in their effective tax rates, indicating they pay a higher share of their income as taxes after joining the program. The reduction in tax avoidance is attributed to the program’s emphasis on transparency, trust and the implementation of robust TCFs. These measures discourage aggressive tax-saving strategies and promote compliance with tax regulations. The results suggest that the CC program successfully aligns corporate tax behaviour with societal expectations of fairness and responsibility.

However, evidence also indicates that firms reduce ESG performance after adopting the CC program (Zeng, 2019). One possible explanation is that firms may reallocate resources previously dedicated to ESG initiatives towards meeting the rigorous requirements of the compliance program, which requires firms to implement robust TCFs and demands significant investments in terms of time, expertise, and financial resources to develop and maintain detailed tax strategies, enhance internal controls to manage tax risks and ensure compliance with extensive documentation and reporting requirements.

Given these demands, firms may face constraints on their capacity to simultaneously invest in ESG-related activities, such as environmental sustainability projects, community engagement or diversity initiatives. The shift in focus towards tax governance might inadvertently deprioritise other ESG efforts, leading to a decline in overall ESG performance. Firms participating in the CC program may prioritise tax compliance as a key aspect of CSR, potentially reducing their focus on other ESG dimensions. Joining the program signals a commitment to transparency and responsible tax governance, which can significantly enhance a firm’s reputation. This reputational benefit may lead firms to strategically substitute tax compliance efforts for broader ESG initiatives, believing that the visibility and measurability of tax governance outweigh the less tangible benefits of other ESG activities.

Additionally, the CC program’s regulatory focus on tax compliance may play a role in the observed ESG performance deterioration. The CC program transforms tax compliance into a collaborative mechanism emphasising transparency and trust between firms and tax authorities. This regulatory emphasis could shift managerial attention and organisational priorities towards tax governance, potentially at the expense of broader ESG objectives.

Moreover, stakeholders, including regulators and investors, increasingly view tax compliance as a critical indicator of social responsibility, further reinforcing this shift. Consequently, governance improvements through tax transparency may come at the cost of environmental or social efforts, reducing overall ESG performance.

Corporate tax contributions support public finances and societal needs. Firms are now evaluated on their good tax governance and the social impact of their tax practices. This emphasis on tax as a social responsibility metric encourages firms to rethink their tax strategies. CC programs help achieve good tax governance by fostering transparency and trust with tax authorities, allowing firms to balance tax optimisation with meeting societal expectations and regulatory requirements.

This study adds to the limited empirical literature on CC by investigating both the antecedents and consequences of entering the CC program. The analysis of the firms’ financial characteristics likely to influence the adoption decision indicates that larger, more profitable, and less leveraged firms are those likely to enter the CC program. Additionally, the listing status of a firm is positively associated with early adoption of the CC program and negatively associated with later adoption. The analysis of the consequences, focused on tax avoidance behaviour, shows that firms increase their tax burden. However, after entering the CC program, firms appear to substitute their ESG performance for tax avoidance, providing evidence of a substitution relationship in line with prior research (Zeng, 2019).

This paper has implications for policymakers, practitioners and other stakeholders, as well as for researchers. Policymakers should develop more integrated regulatory approaches that simultaneously encourage both tax compliance and broader ESG performance, potentially through coordinated incentive structures or by explicitly incorporating tax governance metrics into comprehensive sustainability frameworks. For instance, while the Global Reporting Initiative has explicitly included taxes in ESG reporting (GRI, 2019), the current body of the European Sustainability Reporting Standards lacks a standard specifically devoted to taxes. Managers – who now perceive taxation as a key component of ESG strategies (EY, 2021; PwC, 2022) – should be aware of the potential trade-offs between ESG dimensions when participating in programs like CC. While focusing on tax compliance can enhance transparency and reputation, neglecting environmental and social initiatives may harm long-term ESG performance and stakeholder trust. Stakeholders – increasingly viewing tax contributions as a fundamental measure of a company’s social commitment – should scrutinise a firm’s ESG performance by spotting potential trade-off effects across sustainability dimensions. Our study supports researchers as it adds another theoretical layer of analysis – tax governance – on the relationship between corporate taxation and ESG. Conceptually, the paper further develops the notion of the “substitution effect” between taxation and other ESG dimensions (Zeng, 2019), where firms balance these activities as a risk-management strategy to enhance their reputation (Godfrey, 2005; Minor and Morgan, 2011; Davis et al., 2016; Col and Patel, 2019). The discovered substitution effect between tax compliance and ESG performance opens avenues for further research – also in other settings – on whether this represents a temporary resource allocation adjustment or a fundamental shift in corporate priorities.

This study has some limitations that should be acknowledged. While the focus on the Italian context might limit the study’s external validity, our comparison of CC programs across jurisdictions reveals that the Italian regime shares fundamental principles with similar initiatives in the Netherlands, Denmark and Austria, suggesting broader applicability of our findings. Nevertheless, certain distinctive features of the Italian program – particularly its high entry threshold – may influence participant characteristics differently than in countries with more inclusive eligibility criteria. However, this limitation is partially mitigated by our focus on the largest Italian firms targeted by the CC program. These are subject to high scrutiny for their tax affairs and have more articulated tax transactions, making them comparable to large corporations participating in CC programs internationally.