The role of environmental, social, and governance (ESG) in accounting conservatism: Evidence from Brazil

ABSTRACT

This study investigates whether there is an association between environmental, social, and corporate governance (ESG) performance and the demand for accounting conservatism in companies listed on the Brazilian Stock Exchange. Recognizing that conservative accounting practices often arise from stakeholders’ concerns about earnings management and information asymmetry - conditions that differ from those of socially responsible companies - this paper explores the relationship between ESG performance and accounting conservatism. This study examines accounting conservatism as a principle of information quality, particularly within the framework of mechanisms such as ESG performance that mitigate information asymmetry. Despite Brazil’s weak enforcement profile, limited general credit availability, and high ownership concentration, the country is globally recognized for its adoption of Sustainability Accounting Standards Board (SASB) reporting standards. These conditions reinforce our assumption and highlight the significance of this study. There is a notable paucity of research investigating the correlation between ESG performance and accounting conservatism in Brazil, especially given the country’s unique regulatory and economic landscape. This study addresses this gap by providing a critical perspective on the interplay between ESG and accounting conservatism. Our findings have substantial practical, social, and environmental implications. Effective disclosure of ESG practices not only enhances the quality of reported information but also increases stakeholder credibility, enabling investors and other stakeholders to make more informed and assertive decisions. We estimate conservatism according to the model proposed by Ball and Shivakumar (2005), and in the main analyses, we use ordinary least squares (OLS) regression and quantile regression methods with robust standard errors and fixed effects for industry and years. Our research indicates that higher ESG performance leads to increased conditional accounting conservatism in the following year. This finding suggests that robust ESG reporting enhances the quality of information available for decision-making, thereby reducing information asymmetry between the company and its stakeholders while bolstering its reputation.

Keywords:
accounting conservatism; ESG performance; CSR; earnings quality; information asymmetry

RESUMO

O objetivo deste estudo é investigar a associação entre o desempenho ambiental, social e de governança corporativa (ESG) e a demanda por conservadorismo contábil em empresas listadas na bolsa de valores brasileira. Considerando que as práticas contábeis conservadoras geralmente decorrem das preocupações dos stakeholders com o gerenciamento de resultados e a assimetria de informações - condições que tendem a se diferenciar em empresas socialmente responsáveis - este estudo explora a relação entre o desempenho ESG e o conservadorismo contábil. O conservadorismo contábil é analisado como um princípio de qualidade da informação, particularmente no contexto de mecanismos, como o desempenho ESG, que contribuem para mitigar a assimetria de informações. Apesar das limitações, no Brasil, quanto ao perfil de fiscalização e à disponibilidade geral de crédito e da alta concentração de propriedade, o país é reconhecido mundialmente pela adoção dos padrões de reporte do Sustainability Accounting Standards Board (SASB). Essas condições reforçam as suposições e destacam a importância do presente estudo. Há escassez de pesquisas que investiguem a correlação entre o desempenho ESG e o conservadorismo contábil no Brasil, sobretudo considerando as especificidades regulatórias e econômicas do país. Esta pesquisa busca preencher essa lacuna ao oferecer uma perspectiva crítica sobre a interação entre as práticas ESG e o conservadorismo contábil. As conclusões da pesquisa têm implicações práticas, sociais e ambientais relevantes. A divulgação eficaz das práticas de ESG melhora a qualidade das informações relatadas e aumenta a credibilidade entre os stakeholders, permitindo que os investidores e outros stakeholders tomem decisões mais informadas e assertivas. A mensuração do conservadorismo foi realizada com base no modelo proposto por Ball e Shivakumar (2005), sendo utilizadas, nas análises principais, regressões por mínimos quadrados ordinários (MQO) e regressões quantílicas com erros-padrão robustos e efeitos fixos por setor e ano. Os resultados da pesquisa indicam que um maior desempenho ESG está associado a um aumento do conservadorismo contábil condicional no ano subsequente. Esse achado sugere que a divulgação robusta de informação ESG aprimora a qualidade das informações disponíveis para a tomada de decisão, reduzindo a assimetria de informações entre a empresa e seus stakeholders, ao mesmo tempo em que fortalecem a sua reputação.

Palavras-chave:
conservadorismo contábil; desempenho ESG; RSC; qualidade dos lucros; assimetria de informações

1. INTRODUCTION

Economics textbooks often state that the primary objective of companies is to maximize profits. However, in today’s advanced economic environment, a sole focus on profit may jeopardize a company’s survival (Hong, 2020). Gray (1990) cautioned that corporate social and environmental issues are not fleeting trends or optional extras but integral to an organization’s core business. Over 30 years later, environmental, social, and corporate governance (ESG) issues have become essential components of investment analysis and decision-making processes, as they can significantly impact the performance of investment portfolios ( Environment Programme - Principles for Responsible Investment, 2024). As sustainable development progresses, there is a growing consensus across all sectors of society that enterprises must broaden their social responsibilities (Wan & Dawod, 2022). Therefore, an increasing number of companies are reporting ESG information (Khemir et al., 2019), and investors are seeking strategic ESG risk management and ethical companies that prioritize sound corporate behavior without sacrificing returns (Hübel & Scholz, 2020; Pedersen et al., 2020).

The reputation of companies among their investors and society, whether positive or negative, may be linked to their ESG performance. In Brazil, following the Brumadinho dam disaster, Vale sought to restore its legitimacy by disclosing detailed information about the incident. Liu et al. (2024) assert that companies with lower ESG ratings experience declines in stock prices, prompting creditors and shareholders to demand more conservative practices. These conservative practices consist of a cautious approach in uncertain situations (International Accounting Standards Board [IASB], 2018). According to Basu (1997), conservative accounting practices reflect accountants’ tendency to require more verification for recognizing good news. Therefore, shareholders may demand “conservative accounting” to reduce information asymmetry (LaFond & Watts, 2008), as with ESG.

ESG disclosure appears to minimize information asymmetry between companies and external investors, enabling shareholders to better assess risks and opportunities in their portfolios, related to environmental, social, and governance aspects (Cho et al., 2013; Cui et al., 2018). The same association was established between information asymmetry and accounting conservatism since conservatism practices arise to mitigate asymmetry with the market (Khan & Watts, 2009; LaFond & Watts, 2008). LaFond and Watts (2008) argue that conservatism restricts managers’ ability to manipulate accounting numbers, thereby reducing information asymmetry between firm insiders and outside investors.

However, accounting literature points to some controversies regarding the relationship between ESG and conservatism. While one approach focuses on ESG and conservatism as positively related, considering them attributes of quality accounting information that reduce information asymmetry (Ball et al., 2000; García Lara et al., 2014), another generally cautions about the causes and consequences of conservatism in accounting practices (Caskey & Laux, 2017; Penman & Zhang, 2002), and considers it inversely related to ESG-engaged companies. For instance, the first approach can be seen in Garanina and Kim (2023). The authors found that companies that disclose more about corporate social responsibility (CSR) in the Russian market tend to have more conservative accounting practices. However, Pan and Zhao (2022) discovered through an experiment conducted in China that the imposition of mandatory CSR policies reduced accounting conservatism, highlighting aspects of the second approach. Similarly, Shankar Shaw et al. (2021) demonstrated a negative relationship between accounting conservatism and CSR spending in Indian firms.

The literature on conservatism and ESG, when applied to the Brazilian context, is still scarce. Most scholars, particularly in studies outside the United States of America, argue that emerging countries often overlook ESG practices (Pan & Zhao, 2022). However, Brazil is a leader in adopting Sustainability Accounting Standards Board (SASB) and Global Reporting Initiative (GRI) reporting standards worldwide (Klynveld Peat Marwick Goerdeler International [KPMG], 2022), signaling clearer ESG information for stakeholders. In addition to the unique characteristics of the Brazilian emerging market context - such as scant oversight, limited credit supply, and a small number of companies with diffuse ownership (Brito & Martins, 2013; Sousa & Galdi, 2016) - we decided to explore the Brazilian case within this controversial literature. Thus, this study aims to examine whether ESG performance is significantly related to the demand for conditional accounting conservatism.

We investigate the relationship between ESG performance and conditional conservatism among 83 non-financial companies listed on the Brazilian Stock Exchange (B3 S.A.) from 2012 to 2022. Utilizing Ball and Shivakumar’s (2005) model as a proxy for conditional conservatism, we capture the recognition of losses relative to gains. Our hypothesis testing employs both ordinary least squares (OLS) and quantile estimation models to elucidate the connection between companies’ ESG performance and conditional conservatism.

Our findings reveal that the previous year’s ESG performance is associated with a higher demand for conditional conservatism. This suggests that companies with better ESG performance indices experience an increased need for conditional conservatism in subsequent years. These insights underscore the potential utility of disclosing ESG practices as a means to mitigate information asymmetry, address potential agency conflicts, and enhance the informativeness of stock prices (Anagnostopoulou et al., 2021; Pan & Zhao, 2022; Schiehll & Kolahgar, 2021). However, this association sometimes diverges when analyzing the current period’s ESG performance, as the contradictory literature approaches suggest. This emphasizes the importance of context and time frame in understanding ESG performance and conditional conservatism dynamics. Furthermore, while exploring different approaches to understanding the impact of conditional conservatism on the quality of accounting information, our research delves into the significance and advantages of ESG performance in both scenarios. Such disclosure appears to enhance the quality of reported information and reduce information asymmetry, empowering investors and other stakeholders to make more informed decisions. Thus, our paper provides both practical and social contributions, emphasizing the value of transparency in ESG practices.

2. RELATED LITERATURE AND RESEARCH HYPOTHESIS

2.1 Accounting Conservatism

Conservative practices have been present in accounting for over 500 years (Ball & Shivakumar, 2005; Basu, 1997; Watts, 2003), and have been extensively explored in academic research (e.g., Ball and Shivakumar [2005], García Lara et al. [2020), and LaFond and Watts [2008]). In the accounting literature, conservatism is conceptually understood in terms of its treatment of profits and losses. It is widely regarded as a prudent response to uncertainty, aiming to ensure that “assets and income are not overstated, and liabilities and expenses are not understated” (IASB, 2018, p. A27). Basu (1997) posits a definition that acknowledges the inherent risks associated with accounting information. According to the author, accounting conservatism involves the asymmetric recognition of good and bad news in earnings and captures the tendency to require a higher degree of verification to recognize good news. Thus, accounting conservatism does not require that all cash flows from revenues must be received before profits are recognized, but rather emphasizes the importance of verifiability in recognizing such cash flows (Watts, 2003).

The literature usually refers to two types of accounting conservatism - unconditional and conditional, differentiated by the dependence (or lack thereof) on new events, either ex-ante or ex-post (Ball & Shivakumar, 2005; Ruch & Taylor, 2015). In line with Basu’s (1997) definition, Ball and Shivakumar (2005) argue that conditional conservatism correlates early registration with potential economic losses. It imposes stringent verifiability criteria for recognizing gains compared to losses, resulting in a more prompt and comprehensive recognition of economic losses relative to gains. However, unconditional conservatism pertains to elements of the accounting process that result in unrecognized goodwill. It involves the application of conservative measurement and recognition standards at the inception of assets and liabilities, leading to a persistent undervaluation of net assets regardless of indications of likely economic losses (Ball & Shivakumar, 2005; Moreira et al., 2010). Since previous studies have predominantly examined conservatism in the context of uncertain events such as valuation and information asymmetry (e.g., García Lara et al. [2020], LaFond and Watts [2008], and Yoon et al. [2019]), our research focuses on conditional conservatism.

The literature on conditional conservatism identifies two primary research approaches. The first approach generally highlights the advantages of accounting conservatism (e.g., Anagnostopoulou et al. [2021], Ball et al. [2000], García Lara et al. [2014], Roychowdhury and Watts [2007], and Watts [2003]). Ball et al. (2000) underscore that conservatism is integral to the transparency and quality of accounting information. They argue that in countries with common-law systems, conservatism aids in monitoring managers and serves as a crucial aspect of corporate governance. Watts (2003) contends that without a proper understanding of the benefits stemming from conservatism, accounting information standards could be compromised. Roychowdhury and Watts (2007) report that conservatism curtails managers’ opportunistic behavior and potentially enhances the company’s value. García Lara et al. (2014) and Pan and Zhao (2022) acknowledge conservatism’s role in enhancing a company’s informational environment and bridging the information gap. Consequently, implementing conservative measures in a company can diminish informational asymmetry in subsequent periods, thereby lowering agency costs and influencing analysts’ and investors’ behavior (Anagnostopoulou et al., 2021; García Lara et al., 2014). Evidence also suggests a positive association between conservatism and the quality of corporate governance, which helps mitigate adverse market reactions to disruptive events (Francis et al., 2013).

The second approach focuses on the causes and consequences of conservatism in accounting practices (e.g., Khan & Watts [2009], LaFond and Watts [2008], and Ruch and Taylor [2015]). The Financial Accounting Standards Board (1980) cautions against applying accounting conservatism indiscriminately, as it can introduce bias to accounting statements, conflicting with essential qualitative characteristics of accounting information, such as representativeness, reliability, neutrality, and comparability. Accounting conservatism also affects the return on accounting information, which can distort the quality of earnings (Caskey & Laux, 2017; Penman & Zhang, 2002). Ruch and Taylor (2015) argue that Research and Development (R&D) expenses may yield future benefits to companies as future income. However, under conservative accounting, immediate expensing of R&D disregards these future benefits by failing to capitalize on these expenses. This approach can impose significant reporting costs for company management. It translates into direct implications for managers, as conservative accounting practices often result in lower reported profits. Consequently, this can lead to reduced compensation for managers, particularly when their pay is tied to performance bonuses (Ball & Shivakumar, 2006; Kim et al., 2013).

Several studies establish a connection between conditional accounting conservatism and information asymmetry in this approach. LaFond and Watts (2008) argue that conservatism emerges in response to information asymmetry, as the capital market demands greater conservatism to mitigate the effects of this asymmetry, which in turn can enhance firm and equity values. Accordingly, Khan and Watts (2009) suggest that companies with longer investment cycles, significant idiosyncratic uncertainties, and greater information asymmetry exhibit more pronounced accounting conservatism.

2.2 ESG Performance and Conditional Conservatism

The voluntary disclosure of socio-environmental information is witnessing a significant rise. Academic research on this topic dates back to the early 1970s, with studies exploring the nexus between socio-environmental factors, governance practices, and corporate financial performance (Friede et al., 2015). In response to societal expectations, companies are increasingly expected to take a stance on their activities and address social and environmental concerns. Embracing social and environmental responsibility practices not only fosters sustainability but also aligns with the ESG criteria outlined by rating agencies (Drempetic et al., 2020).

Stakeholders exert considerable pressure on companies to disclose their ESG practices, driven by a growing interest in “responsible” investments and a desire for transparency. This pressure stems from the need for companies and communities to comprehend corporate actions and their broader societal impacts. Stakeholders seek to minimize informational asymmetry regarding exposure to ESG risks, prompting organizations to undertake initiatives to mitigate such risks (Khan et al., 2016; Schiehll & Kolahgar, 2021). However, critics argue that existing ESG scoring mechanisms may not accurately measure corporate sustainability, as larger companies with greater resources may have inherent advantages. This critique highlights the importance of understanding how sustainability rating agencies evaluate ESG criteria and the metrics they prioritize (Drempetic et al., 2020). Furthermore, from an investor’s perspective, ESG information disclosures continue to lack reliability, comparability, and relevance. Companies often struggle with identifying pertinent ESG data for disclosure and determining their primary target audience (Schiehll & Kolahgar, 2021). Therefore, the development of ESG financial materiality standards by the SASB has emerged as a valuable tool. These standards assist companies in prioritizing material ESG disclosures, thereby enhancing the information’s value for stakeholders’ decision-making processes.

Van Duuren et al. (2016) investigated how managers consider ESG criteria in investments and found that many integrate what can be regarded as a responsible investment into their total assets (TA). They emphasize the use of ESG information as a means to identify “red flags” and manage risks effectively. Consequently, there is a discernible trend wherein investors increasingly prioritize ESG criteria when evaluating companies’ shares and constructing portfolios. Investors favor ethical companies that prioritize good corporate behavior without sacrificing returns (Pedersen et al., 2020). However, Martin and Moser (2016) suggest that while responsible investments may not always translate into immediate improvements in future cash flows, shareholders tend to react positively when managers make and disclose such investments, emphasizing their social benefits over the associated costs.

Several studies have scrutinized the relationship between ESG practices, often measured through CSR indicators, and organizational performance, including the quality of accounting information. For instance, Lys et al. (2015) found a positive association between CSR-related expenditures and future financial performance. Schiehll and Kolahgar (2021) demonstrated that disclosures of ESG information enhance the informativeness of share prices by evaluating the value relevance and financial materiality of companies listed in the Standard & Poor’s and the Toronto Stock Exchange (S&P/TSX) Composite Index. Barbosa and Klann (2023) also concluded that ESG performance is value-relevant, as it is positively related to the market-to-book (MTB) ratio of non-financial companies listed on the B3 S.A. Other studies have investigated the relationship between CSR performance and earnings management using both discretionary accruals and actual activity approaches (e.g., Bozzolan et al. [2015], Kim et al. [2012], and Yoon et al. [2019]). The majority of these studies indicate that companies with superior CSR performance are less inclined to engage in earnings management practices, thereby enhancing the quality of their financial reports.

Previous studies examining the relationship between ESG performance and conditional conservatism have yielded mixed results. Kim et al. (2012) argue that socially responsible companies tend to be more conservative, whereas Burke et al. (2020) and Hong (2020) suggest that higher ESG performance is associated with lower levels of accounting conservatism. Anagnostopoulou et al. (2021) identified a negative association between CSR orientation and conservatism. However, during the post-financial crisis period, the same authors found an inverse trend. When analyzing each sustainability pillar separately, Khlifi et al. (2024) identified a positive effect of environmental performance and a negative impact of the social aspect on the accounting conservatism level. In the context of emerging economies, Pan and Zhao (2022) discovered that enforcing mandatory CSR policies in China reduced accounting conservatism.

By acknowledging these contradictory results from previous literature, we develop a non-directional hypothesis to embrace this divergence. We recognize that accounting conservatism is integral to the transparency and quality of accounting information (Ball et al., 2000), reducing information asymmetry (García Lara et al., 2014), as does reverberating the principles of high-quality ESG engagement. However, we recognize the approach that ESG-engaged companies prioritize meeting shareholders’ demands without incurring financial losses that may arise from accounting conservatism (Basu, 1997) and affect earnings quality (Caskey & Laux, 2017; Penman & Zhang, 2002). Therefore, we articulate the following non-directional hypothesis to allow the data to reveal the nature of this relationship.

H₁: There is a significant relationship between ESG performance and the demand for conditional conservatism.

3. RESEARCH METHOD

3.1 Data and Empirical Design

Our study focuses on publicly listed companies on the B3 S.A. We excluded financial firms and real estate companies due to their unique characteristics, as well as companies with less than 3 years of available data during the analysis period. Consequently, our final sample comprises 83 companies from 2012 to 2022. We collected data on economic and financial information, ESG performance, and board information from the Refinitiv Eikon^® database. Although our study covers the 2012-2022 period, we collected data from 2010 to 2022 to accommodate lagged variables in our model.

Table 1 provides detailed descriptions of the variables employed in estimating the econometric models. The conservatism measure (CON) was adapted from the study by Lee et al. (2015), which estimates the firm-year conditional CON (C_Score) using the Khan and Watts (2009) model. Initially, a proxy for conditional conservatism was estimated using the Ball and Shivakumar (2005) model, highlighted in Equation 1. This model aims to discern the level of conditional conservatism by analyzing the reversal of accounting outcomes among firms (Ball & Shivakumar, 2005), thereby facilitating a precise estimation of conditional conservatism and fostering a deeper comprehension of its implications within our study context:

$${∆NI}_t={\alpha }_0+{\alpha }_1{D∆NI}_{t-1}+{\alpha }_2{∆NI}_{t-1}+{\alpha }_3{D∆NI}_{t-1}*{∆NI}_{t-1}+{\epsilon }_t$$ (1)

where: ${\mathrm{\Delta }\mathrm{N}\mathrm{I}}_{i,t}$ = negative change in earnings in t, ${\mathrm{D}\mathrm{\Delta }\mathrm{N}\mathrm{I}}_{i,t}$ = dummy variable indicating negative earnings variation in t-1, and ${\mathrm{\Delta }\mathrm{N}\mathrm{I}}_{i,t-1}$ = negative change in earnings in t-1.

Table 1
Variables description model

Variables	Description	Operationalization	Previous studies
Dependent variables
CON	Conditional conservatism	C_Score	Ball and Shivakumar (2005), Khan and Watss (2009)
NI	Net income	Net income	Ball and Shivakumar (2005)
Debt	Debt liabilities	Total debt liabilities	Ball and Shivakumar (2005), Khan and Watts (2009)
Independent variables
ESG	ESG combined score	$ESG={ESG}_{Score}-{ESG}_{Controversy}$	Refinitiv Eikon® database Anagnostopoulou et al. (2021)
Control variables
TA	Total assets	Natural logarithm of total assets	Anagnostopoulou et al. (2021)
MTB	Market-to-book	$MTB=\frac{Marketvalue}{Totalequity}$	Anagnostopoulou et al. (2021)
ROA	Profitability	Return operation of total assets	Kim et al. (2012)
LEV	Leverage	$LEV=\frac{Totaldebt}{Marketvalue}$	Anagnostopoulou et al. (2021)
GROWTH	Sales growth	${GROWTH}_i=\frac{({GROWTH}_t-{GROWTH}_{t-1})}{{GROWTH}_{t-1}}$	Burke et al. (2020)
BOARD	Board size	Natural logarithm of total board members	Anagnostopoulou et al. (2021)
FEM	Female board members	Ratio of the number of female members on the board	Anagnostopoulou et al. (2021)
IND	Independent board members	Ratio of independent members on the board	Anagnostopoulou et al. (2021)
CEOB	CEO board member	(1) CEO member of the board; or (0) CEO nonmember of the board	Anagnostopoulou et al. (2021)
CSREA	CSR report external audit	Ratio of the CSR report assured by an external auditor
Industry	TRBC industry	Fixed effect of the industry	Anagnostopoulou et al. (2021)
Year	Year	Fixed effect of the year	Anagnostopoulou et al. (2021)

• CEO = chief executive officer; CSR = corporate social responsibility; ESG = environmental, social, and corporate governance; TRTBC = Thomson Reuters Business Classification.
• Source: Elaborated by the authors.

This model serves to capture the degree to which accounting practices exhibit conditional conservatism by scrutinizing the asymmetric recognition of gains and losses. Second, to examine the effect of good and bad news at the firm-year level, we estimated the proxies G_SCORE, representing good news, and C_Score, representing the asymmetric recognition of bad news, according to the model of Khan and Watts (2009). As shown in Model 2, both measures are expressed as linear functions of firm-specific characteristics, including leverage (LEV), size, and MTB ratio. These proxies facilitate the analysis of conditional conservatism by delineating the company’s response to varying types of news, thereby offering a nuanced understanding of the factors influencing accounting practices.

$${CScoreBS}_{it}={\alpha 3}_{it}={\lambda }_{1t}+{\lambda }_{2t}{SIZE}_{it}+{\lambda }_{3t}{MTB}_{it}+{\lambda }_{4t}{LEV}_{it}$$ (2)

$${GScoreBS}_{it}={\alpha 3}_{it}={\mu }_{1t}+{\mu }_{2t}{SIZE}_{it}+{\mu }_{3t}{MTB}_{it}+{\mu }_{4t}{LEV}_{it}$$ (3)

where SIZE represents the natural logarithm of the market value of equity in period t, MTB denotes the ratio of the market value to the book value of equity in period t, and LEV signifies the ratio of onerous liabilities to the market value of equity in period t.

We revised the equation in accordance with Model 4 because Ball and Shivakumar (2005) propose the hypothesis that firms recognize economic losses more promptly than profits, implying a more conservative behavior:

$${\mathrm{\Delta }NI}_t={\alpha }_0+{\alpha }_1{D\mathrm{\Delta }NI}_{t-1}+{\alpha }_2{\mathrm{\Delta }NI}_{t-1}\left({\mu }_{1t}+{\mu }_{2t}{SIZE}_{it}+{\mu }_{3t}{MTB}_{it}+{\mu }_{4t}{LEV}_{it}\right)+{\alpha }_3{D\mathrm{\Delta }NI}_{t-1}*{\mathrm{\Delta }NI}_{t-1}\left({\lambda }_{1t}+{\lambda }_{2t}{SIZE}_{it}+{\lambda }_{3t}{MTB}_{it}+{\lambda }_{4t}{LEV}_{it}\right)+\left({\delta }_{1t}{SIZE}_{it}+{\delta }_{2t}{MTB}_{it}+{\delta }_{3t}{LEV}_{it}+{\delta }_{4t}{D\mathrm{\Delta }NI}_{t-1}*{SIZE}_{it}+{\delta }_{5t}{D\mathrm{\Delta }NI}_{t-1}*{MTB}_{it}+{\delta }_{6t}{D\mathrm{\Delta }NI}_{t-1}*{LEV}_{it}\right)+{\epsilon }_t$$ (4)

We adopted the estimation using the model of Ball and Shivakumar (2005) because subsequent studies critiqued Basu’s measure (1997) for econometric biases concerning the asymmetric recognition of conservatism (Ball & Shivakumar, 2005; Lee et al., 2015). Therefore, Ball and Shivakumar’s (2005) measure, grounded on the variation of current and lagged profits, is deemed more appropriate for estimating conditional conservatism (Lee et al., 2015).

We measure ESG performance, our explanatory variable, utilizing the ESG combined score sourced from the Refinitiv Eikon^® database, a metric informed by the amalgamation of strengths, weaknesses, and negative concerns linked to ESG performance (Anagnostopoulou et al., 2021; Burke et al., 2020). This score incorporates evaluations of adverse global media coverage concerning business activities associated with companies’ ESG performance. The Refinitiv Eikon^® database delineates this ESG performance variable across 10 principal categories: Emissions, Environmental Innovation, Resource Use, Employees, Human Rights, Community Impacts, Product Responsibility, Governance, Shareholders, Social Responsibility, and Related News. Comprising a total of 630 observed metrics, these categories encompass 186 sets of materiality comparisons tailored to specific sectors. For an exhaustive overview of the ESG measure employed in our analysis (Table 2).

Table 2
Performance measure and degree of environmental, social, and corporate governance (ESG) disclosure transparency

Measure	Description
0.00 to 0.25	Poor performance and insufficient degree of transparency
0.25 ≤ 0.50	Satisfactory performance and a moderate degree of transparency
0.50 ≤ 0.75	Good performance and above-average degree of transparency
0.75 ≤ 1.00	Excellent performance and a high degree of transparency

• Source: Refinitiv Eikon^® database.

Based on the related literature, our study incorporates control variables into the model to test our research hypothesis. These variables are categorized into two groups: (i) economic/financial performance indicators of companies and (ii) characteristics of board composition influencing information quality and accounting conservatism.

In our primary model, we account for variables potentially correlated with conditional conservatism, as underscored by Burke et al. (2020). The first group of control variables includes TA, represented by the natural logarithm of the companies’ TA, as larger companies typically exhibit less information asymmetry in earnings (Anagnostopoulou et al., 2021). The MTB index, reflecting the relationship between market value and book value, is associated with a company’s conditional conservatism due to the accumulation of unconditional conservatism (Burke et al., 2020; Kim et al., 2012). The financial LEV is included as companies with significant LEV levels face pressure from creditors for conservative behavior regarding losses and gains (Anagnostopoulou et al., 2020; Burke et al., 2020). Additionally, variables controlling for companies’ profitability and growth opportunities are incorporated, such as the return on assets (ROA) and the sales growth (GROWTH) determined by the variation in net revenue (Anagnostopoulou et al., 2021; Burke et al., 2020; Kim et al., 2012).

The second group of control variables pertains to board characteristics. Board size (BOARD), measured by the natural logarithm of the number of board members, serves as a proxy associated with safeguarding shareholders and stakeholders, thereby influencing an increase in conservatism when these stakeholders expect the firm to behave more conservatively (Anagnostopoulo et al., 2021; Jizi et al., 2014). Companies with more women on the board (FEM) are anticipated to adopt more conservative accounting practices (Alves, 2023; Anagnostopoulou et al., 2021) due to women’s tendency toward risk aversion. Women directors are more likely to demand managerial conservatism, thereby reducing managerial opportunism and mitigating the risks associated with information asymmetry and the probability of litigation (Alves, 2023). Board independence is measured by the percentage of independent directors (IND). Previous literature (e.g., Anagnostopoulou et al. [2021] and Elshandidy and Hassnein [2014]) suggests that boards with more IND tend to be more conservative. Moreover, we also included a dummy variable indicating whether the CEO is a member of the board of directors (CEOB), considering the CEO’s influence on income information measurement and the effectiveness of the board’s oversight. This is particularly relevant if the CEO’s compensation is linked to performance criteria, as in such cases, the CEO may be less inclined to choose more conservative practices (Yin et al., 2020).

We employed both OLS and quantile estimation models to test the research hypothesis based on the final econometric model outlined in Model 5 (variables are described in Table 1):

$${CON}_{i,t}={\beta }_0+{\beta }_1{ESG}_{i,t-1}+{\beta }_2{TA}_{i,t}+{\beta }_3{MTB}_{i,t}+{\beta }_4{ROA}_{i,t}+{\beta }_5{LEV}_{i,t}+{\beta }_6{GROWTH}_{i,t}+{\beta }_7{BOARD}_{i,t}+{{\beta }_8{FEM}_{i,t}+\beta }_9{IND}_{i,t}+{\beta }_{10}{CEOB}_{i,t}+{\beta }_{11}{CSREA}_{i,t}+{\beta }_{12}{Industry}_i+{\beta }_{13}{Year}_{i,t}+{\epsilon }_{i,t}$$ (5)

In the context of our model, employing both OLS and quantile regression enables a thorough analysis of the relationship between ESG performance and conditional conservatism. OLS provides insights into the average effect of ESG performance on conditional conservatism across the entire sample, while quantile regression allows for the examination of how this relationship varies across different segments of the conditional conservatism distribution. This approach helps capture potential heterogeneity in the relationship between ESG performance and conditional conservatism, providing a more nuanced understanding of the association between these variables.

The results of the diagnostic tests, including the Breusch-Pagan test for heteroscedasticity, and the variance inflation factor (VIF) for multicollinearity, reveal significant findings. Firstly, the Breusch-Pagan test highlights the presence of heteroscedasticity issues (p = 0.0000), suggesting unequal variances across the data. To mitigate the identified heteroscedasticity issues, the regression was rerun using robust and clustered standard errors. These techniques correct for data autocorrelation and address heteroscedasticity problems, ensuring the reliability of the regression results. The VIF of 1.26 does not indicate problems of multicollinearity, as it is below 4, a value that is less than or equal to 4, according to Fávero and Belfiore (2017). Furthermore, to minimize the influence of extreme values, all variables underwent winsorization at the top and bottom 1%, enhancing the robustness of the analysis. Additionally, a yearly control was implemented to account for potential temporal variations in the econometric research model. This control ensures that any time-related factors are adequately addressed in the analysis.

4. RESULTS

4.1 Descriptive Statistics and Correlations

Table 3 presents the descriptive statistics for our firm-year observations. The analysis reveals an average conditional conservatism C_Score of 1.79. The companies exhibit an average ESG performance of 50.63, reflecting a satisfactory relative ESG performance on average. This suggests a moderate level of transparency in the disclosed ESG data when considering only the strengths of ESG performance, excluding the adverse effects of ESG-related exposure in global media coverage.

Table 3
Descriptive statistics

Variable	Observations	Mean	Standard deviation	Minimum	Maximum
CON	658	1.788	19.791	-104.425	365.656
ESG	658	50.635	19.001	5.408	87.201
TA	658	5.04e+10	1.20e+11	1.04e+09	9.87e+11
MTB	658	2.488	3.380	0.002	40.469
ROA	658	4.963	6.656	-15.373	28.977
LEV	658	18.998	137.277	0.000	1348.907
GROWTH	657	0.145	0.328	-0.667	1.935
BOARD	658	9.661	3.482	2.000	27.000
IND	657	40.954	24.016	0.000	100.000
FEM	658	10.651	10.129	0.000	42.857
CSREA	549	44.743	38.486	0.000	82.520

• BOARD = board size; CON = conditional conservatism; CSREA = corporate social responsibility report external audit; ESG = environmental, social, and corporate governance performance; FEM = female board members; GROWTH = sales growth; IND = independent board members; LEV = leverage; MTB = market-to-book; ROA = return on assets; TA = total assets.
• Source: Elaborated by the authors.

Regarding the descriptive analysis of the economic/financial control variables, the companies exhibit an average size (TA) of R$ 5.04 million and an average LEV of 18.99. In terms of market-related indexes presented in Table 4, companies show an average MTB index of 2.48. Profitability measures (ROA) have an average index of 4.96, with an average GROWTH of 15% (Table 3).

Table 4
Spearman’s correlation

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
(1) CON	1.0000
(2) ESG	-0.0131	1.0000
(3) TA	-0.0391*	0.2844*	1.0000
(4) MTB	-0.0577*	-0.0030	-0.2654*	1.0000
(5) ROA	-0.0825*	-0.0061	-0.1687*	0.3751*	1.0000
(6) LEV	0.2181*	0.1049*	0.3645*	-0.5802*	-0.5885*	1.0000
(7) GROWTH	0.0835*	-0.0313*	-0.0335*	0.2444*	0.2800*	-0.0879*	1.0000
(8) BOARD	-0.1082*	0.2847*	0.3622*	-0.1418*	-0.1517*	0.2099*	-0.1289*	1.0000
(9) FEM	-0.0554*	0.1112*	0.0153	0.0633*	0.0322*	-0.1136*	0.0508*	-0.0102	1.0000
(10) IND	0.0379*	-0.0175	-0.1583*	0.1379*	0.0378*	-0.0660*	0.1023*	-0.2407*	0.2281*	1.0000
(11) CEOB	-0.0715*	0.5233*	0.3191*	0.0445*	0.0497*	0.0103	-0.0219	0.3173*	0.0721*	-0.1398*	1.0000
(12) CSREA	-0.0798*	0.1690*	0.0530*	-0.0339*	0.1673*	-0.0305*	-0.0359*	0.1060*	-0.0715*	-0.1026*	0.1491*	1.0000

• BOARD = board size; CEOB = chief executive officer board member; CON = conditional conservatism; CSREA = corporate social responsibility report external audit; ESG = environmental, social, and corporate governance performance; FEM = female board members; GROWTH = sales growth; IND = independent board members; LEV = leverage; MTB = market-to-book; ROA = return on assets; TA = total assets.
• * p < 0.05.
• Source: Elaborated by the authors.

The descriptive analysis of board composition (BOARD) characteristics indicates an average of 9.66 board members, with an average female membership of 10.65% (Table 3). Research suggests that having women on the board enhances the quality of business discussions and leads to higher-quality disclosures and corporate information reporting (Gul et al., 2011). However, our findings align with previous studies, such as Anagnostopoulou et al. (2021), which also note a low percentage of women in board positions. Therefore, the challenge remains to increase female board representation with the expectation of enhancing reporting quality.

In terms of correlation among the research variables, the model reveals weak to moderate correlations. Notably, the MTB ratio exhibits a moderate and statistically significant correlation with CON, as evidenced in Table 4. Conversely, we observe weak but statistically significant correlations between financial LEV, profitability (ROA), and CON. Additionally, the performance variables ESG, MTB, and company size (TA) demonstrate weak yet statistically significant correlations, as outlined in Table 3.

4.2 Discussion and Implications

Table 5 summarizes the results of the statistical inference tests from the model that aimed to determine whether ESG performance in the previous year was associated with a decrease in demand for conditional conservatism.

Table 5
Regression models

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
	OLS	Q10	Q25	Q50	Q75	Q90	OLS	Q10	Q25	Q50	Q75	Q90
ESG_T-1	0.023*	0.005**	0.003*	-0.000	-0.003	-0.003
	(1.72)	(2.47)	(1.94)	(-0.14)	(-1.54)	(-0.47)
ESG							0.011	0.003*	0.001	-0.003**	-0.004**	-0.004**
							(0.95)	(1.87)	(0.68)	(-2.16)	(-3.24)	(-3.24)
TA	-0.370	-0.066***	-0.044	0.069**	0.107***	0.228	-0.290	-0.058**	-0.043**	0.029	0.109***	0.109***
	(-0.68)	(-2.68)	(-1.61)	(2.51)	(3.50)	(0.84)	(-0.60)	(-2.17)	(-2.17)	(1.02)	(4.01)	(4.01)
MTB	-0.103	-0.000	0.028**	0.010***	0.193***	0.240***	-0.058	0.022	0.022	0.094***	0.189***	0.189***
	(-0.43)	(-0.00)	(2.27)	(3.07)	(9.01)	(3.47)	(-0.29)	(0.90)	(1.21)	(3.17)	(8.41)	(8.41)
ROA	0.038	-0.001	-0.017**	-0.014**	-0.010	-0.017	0.019	-0.005	-0.016**	-0.016**	-0.011**	-0.011**
	(0.81)	(-0.17)	(-2.65)	(-2.08)	(-1.64)	(-0.28)	(0.42)	(-0.81)	(-2.22)	(-2.28)	(-2.27)	(-2.27)
LEV	0.046***	0.003***	0.007	0.028	0.072***	0.072	0.046**	0.003**	0.007	0.028	0.072***	0.072***
	(3.22)	(2.91)	(0.26)	(0.89)	(19.09)	(0.15)	(3.23)	(2.25)	(0.26)	(0.87)	(18.78)	(18.78)
GROWTH	-3.526	-0.045	0.164	0.062	-0.003	-0.150	-3.285	-0.035	0.040	0.124	0.018	0.018
	(-1.35)	(-0.39)	(1.57)	(0.64)	(-0.04)	(-0.30)	(-1.34)	(-0.37)	(0.37)	(1.15)	(0.19)	(0.19)
BOARD	0.391	-0.216**	-0.077	-0.0771	-0.245**	-0.221	0.302	-0.162	-0.132*	-0.071	-0.212**	-0.212**
	(0.66)	(-2.37)	(-0.86)	(-1.13)	(-2.76)	(-0.54)	(0.61)	(-1.58)	(-1.92)	(-1.02)	(-2.31)	(-2.31)
FEM	-0.022	-0.003	-0.005**	-0.005*	0.002	-0.004	-0.017	-0.003	-0.005***	-0.003	0.002	0.002
	(-1.03)	(-1.42)	(-2.25)	(-1.67)	(0.57)	(-0.51)	(-0.80)	(-1.22)	(-3.41)	(-1.29)	(0.90)	(0.90)
IND	0.006	0.000	0.001	0.001	-0.006	-0.006	0.009	0.000	0.001	0.001	-0.001	-0.001
	(0.58)	(0.15)	(0.73)	(0.81)	(-0.51)	(-0.09)	(0.78)	(0.39)	(1.06)	(1.04)	(-1.08)	(-1.08)
CEOB	1.109	-0.090*	-0.031	0.066	0.032	0.161	1.109	-0.054	-0.019	0.063	0.012	0.012
	(0.99)	(-1.80)	(-0.55)	(1.24)	(0.67)	(1.09)	(0.98)	(-0.99)	(-0.43)	(1.16)	(0.31)	(0.31)
CSREA	-0.001	-0.000	-0.000	-0.000	0.000	-0.000	0.003	-0.000	0.000	0.001	0.000	0.000
	(-0.27)	(-0.35)	(-0.31)	(-0.11)	(0.20)	(-0.12)	(0.63)	(-0.11)	(0.21)	(0.96)	(0.30)	(0.30)
AES	0	0	0	0	0	0	0	0	0	0	0	0
	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)
BM	0.891	0.094	-0.006	0.067	-0.144	-0.104	1.030	0.011	0.012	0.075	-0.139	-0.139
	(0.58)	(0.15)	(-0.04)	(0.32)	(-0.98)	(-0.24)	(0.65)	(0.02)	(0.07)	(0.56)	(-0.86)	(-0.86)
CC	3.321	0.0912	0.037	0.007	-0.268**	-0.433	2.486	0.077	0.003	-0.069	-0.339**	-0.339**
	(0.95)	(0.15)	(0.23)	(0.03)	(-2.16)	(-0.92)	(0.87)	(0.14)	(0.02)	(-0.49)	(-2.48)	(-2.48)
CNC	2.120	0.021	-0.028	-0.033	-0.270**	-0.016	2.085	-0.035	-0.058	-0.037	-0.278*	-0.278
	(1.19)	(0.03)	(-0.21)	(-0.16)	(-2.02)	(-0.04)	(1.18)	(-0.06)	(-0.32)	(-0.27)	(-1.70)	(-1.70)
ENE	1.351	0.166	0.217	0.024	-0.220	-0.336	1.271	-0.023	0.207	0.004	-0.241*	-0.241
	(0.87)	(0.27)	(1.32)	(0.11)	(-1.46)	(-0.80)	(0.82)	(-0.04)	(1.08)	(0.03)	(-1.82)	(-1.82)
HEAL	0.156	0.094*	0.066	0.020	-0.311	-0.076	0.319	0.038	0.043	-0.037	-0.356*	-0.356**
	(0.16)	(0.15)	(0.41)	(0.09)	(-1.83)	(-0.12)	(0.31)	(0.07)	(0.24)	(-0.25)	(-2.43)	(-2.43)
INDU	1.390	-0.025	-0.015	-0.019	-0.227**	-0.318	1.357	-0.100	-0.019	-0.010	-0.217	-0.217
	(0.75)	(-0.04)	(-0.11)	(-0.09)	(-2.05)	(-0.60)	(0.77)	(-0.18)	(-0.10)	(-0.08)	(-1.60)	(-1.60)
TEC	-0.493	-0.015	-0.135	-0.184	-0.069	-0.238	-0.379	-0.089	-0.178	-0.208	-0.078	-0.078
	(-0.59)	(-0.02)	(-0.86)	(-0.63)	(-0.39)	(-0.36)	(-0.47)	(-0.16)	(-0.92)	(-1.04)	(-0.30)	(-0.30)
UTI	0.204	0.180	0.118	0.003	-0.247**	-0.300	0.438	0.089	0.088	-0.026	-0.211*	-0.211
	(0.21)	(0.29)	(0.88)	(0.01)	(-2.21)	(-0.68)	(0.43)	(0.16)	(0.50)	(-0.20)	(-1.65)	(-1.65)
2012	0	0	0	0	0	0	0	0	0	0	0	0
	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)
2013	4.117	0.832***	0.686***	0.689***	0.805***	1.080**	4.134*	0.845***	0.652***	0.771***	0.835***	0.835***
	(1.64)	(12.29)	(3.83)	(8.74)	(6.50)	(1.98)	(1.69)	(14.95)	(3.57)	(8.73)	(4.50)	(4.50)
2014	6.034**	2.308***	2.504***	2.727***	2.738***	2.360***	5.949**	2.232***	2.427***	2.755***	2.649***	2.649***
	(2.49)	(8.56)	(11.29)	(26.33)	(30.71)	(6.17)	(2.55)	(7.84)	(12.81)	(27.65)	(19.97)	(19.97)
2015	3.391*	0.727	1.221***	1.253***	1.342***	1.369***	3.405	0.704	1.194***	1.365***	1.315***	1.315***
	(1.67)	(0.95)	(6.48)	(9.33)	(12.20)	(3.82)	(1.73)	(1.07)	(5.55)	(14.57)	(8.32)	(8.32)
2016	3.340	0.091	1.592***	1.901***	2.066***	1.791***	3.281	0.029	1.549***	1.890***	1.991***	1.991***
	(1.36)	(0.11)	(3.78)	(10.18)	(17.55)	(4.63)	(1.37)	(0.03)	(3.35)	(11.31)	(13.00)	(13.00)
2017	3.581	1.643***	1.580***	1.443***	1.369***	1.020**	3.505	1.682***	1.487***	1.431***	1.305***	1.305***
	(1.56)	(22.97)	(9.11)	(18.93)	(15.99)	(2.31)	(1.59)	(25.44)	(8.44)	(22.87)	(10.02)	(10.02)
2018	4.624**	1.175***	1.089***	1.040***	1.059***	1.285**	4.517***	1.167***	1.029***	1.047***	0.990***	0.990***
	(2.05)	(17.00)	(6.20)	(15.04)	(8.17)	(2.06)	(2.06)	(16.28)	(5.73)	(14.40)	(6.75)	(6.75)
2019	7.043	-0.211*	-0.225	-0.289***	-0.273	-0.208	5.932	-0.246*	-0.282	-0.235***	-0.204	-0.204
	(1.60)	(-1.92)	(-1.26)	(-3.52)	(-1.39)	(-0.38)	(1.63)	(-2.13)	(-1.55)	(-3.56)	(-1.36)	(-1.36)
2020	4.269**	1.348***	1.259***	1.222***	1.128***	1.206***	4.040*	1.359***	1.183***	1.246***	1.142***	1.142***
	(1.99)	(8.97)	(7.33)	(12.84)	(12.30)	(3.43)	(1.98)	(18.74)	(6.57)	(16.79)	(9.04)	(9.04)
2021	6.294***	2.387***	2.587***	2.728***	2.684***	2.414***	6.157**	2.328***	2.524***	2.716***	2.651***	2.651***
	(2.63)	(22.93)	(11.71)	(24.16)	(27.87)	(5.18)	(2.64)	(15.97)	(11.58)	(23.26)	(21.00)	(21.00)
2022	4.137*	0.638***	0.542***	0.670***	0.748***	1.262	3.920*	0.602***	0.485**	0.705***	0.717***	0.717***
	(1.83)	(7.09)	(3.13)	(4.88)	(8.88)	(1.42)	(1.81)	(7.57)	(2.72)	(4.52)	(5.31)	(5.31)
_cons	1.669	0.273	-0.263	-2.673***	-2.859***	-5.336	0.332	0.050	0.037	-1.627**	-2.865***	-2.865***
	(0.17)	(0.33)	(-0.41)	(-4.27)	(-4.34)	(-1.16)	(0.04)	(0.06)	(0.07)	(-2.73)	(-4.88)	(-4.88)
N	487	487	487	487	487	487	511	511	511	511	511	511
R²	39.36%						38.78%

• Note: Robust standard errors. t statistics in parentheses.
• AES = academic and educational services; BM = basic materials; BOARD = board size; CC = consumer cyclicals; CEOB = chief executive officer board member; CNC = consumer non-cyclicals; CON = conditional conservatism; CSREA = corporate social responsibility report external audit; ENE = energy; ESG = environmental, social, and corporate Performance; FEM = female board members; GROWTH = sales growth; HEAL = healthcare; IND = independent board members; INDU = industrials; LEV = leverage; MTB = market-to-book; ROA = return on assets; TA = total assets; TEC = technology; UTI = utilities.
• *, **, *** = p < 0.10, p < 0.05, p < 0.01, respectively.
• Source: Elaborated by the authors.

Overall, the tests are aligned with the research hypothesis H₁, considering a significant level of 10%. Additionally, the findings show that the demand for conservatism is concentrated in companies with lower C_Score, particularly in the Q10 and Q25 quantiles. Based on these results, it is concluded that previous ESG performance is linked to an increase in companies’ average demand for conditional conservatism in the subsequent period. Furthermore, when evaluating the results of the quantile regression models, we note that companies with lower C_Score levels exert a more pronounced association with the increase in conservatism attributable to past ESG performance in the current period. Our findings build upon existing literature suggesting that heightened ESG scrutiny may bolster conservatism in financial reporting (for example, Garanina and Kim [2023] and Wan and Dawod [2022]). Moreover, higher ESG performance is associated with increased organizational legitimacy, and improved stakeholder relations are associated with increased conservatism, related to the emerging market context (Garanina & Kim, 2023). This phenomenon likely stems from stakeholders’ endeavors to enhance a firm’s information environment and bridge information asymmetries (Pan & Zhao, 2022). Consequently, the positive correlation between superior ESG performance and conservatism offers a mechanism to mitigate information asymmetry and bolster the reputation of organizations before both the market and stakeholders, particularly in emerging markets.

The regression analysis also examined the current period’s ESG performance to assess its association with accounting conservatism. Again, the model was estimated with robust standard errors and included fixed industry and year effects. Table 5 shows that ESG significantly affects accounting conservatism only when analyzed by quantile. This effect is significant at the 10% and 5% significance levels for the quantiles Q10, Q50, Q75, and Q90, respectively, corroborating our hypothesis. However, while Q10 found a positive association between ESG performance and conditional conservatism, in line with our previous result, when analyzing the quantiles Q50, Q75, and Q90, the signal varies from the previous ESG performance results. Those last quantiles demonstrate that companies with elevated ESG levels during the specified period experience a reduction in stakeholder demand for the conditional conservatism of publicly traded companies listed on B3 S/A, which aligns with the contradictory findings in the literature. In other words, most of the current period’s ESG results are consistent with studies that find a significant relationship between ESG performance and a decrease in conditional conservatism, such as Burke et al. (2020), Hong (2020), and Pan and Zhao (2022). In addition, Hong (2020) and Yoon et al. (2019) understand that this negative association improved accounting information quality in Korean firms, aligning with the second conservatism approach whose caution on the negative effects of conservatism practices in the quality of earnings (e.g., Caskey and Laux [2017], Penman and Zhang [2002], and Ruch and Taylor [2015]).

Taken together, these results suggest that firms prioritizing social responsibility disclosures may encounter fewer agency conflicts associated with employing accounting conservatism (Boulhaga et al., 2023; Pan & Zhao, 2022). Our results are consistent with Cui et al. (2018), who found that CSR performance has a significant impact on improving organizational information communication. Enhanced ESG performance, coupled with increased non-financial disclosures, ensures greater availability of information for stakeholders. Consequently, this improved transparency influences decision-making processes, thereby reducing information asymmetry between companies and stakeholders.

Regarding the control variables, the results indicate that LEV is significantly associated with the company’s conditional conservatism at a 1% significance level. These findings are consistent with prior studies (e.g., Burke et al. [2020]). Yoon et al. (2019) also observed a positive association between these variables and improved information quality. While previous research has indicated a negative and significant association between companies’ size, profitability, GROWTH, and losses with conditional conservatism, our empirical model did not find a significant association of these variables (TA, ROA, GROWTH). This contradicts the findings of Burke et al. (2020) and Hong (2020), who identified a significant relationship between these variables and conditional conservatism.

In terms of board characteristics, the size of the board of directors is found to significantly reduce accounting conservatism by a significant 5% in the Q10 and Q75 quantiles, respectively. Conversely, the presence of women on the board shows a significant association only on the Q25 and Q50 quantiles, at the 5 and 10% significance levels, respectively. This suggests that gender diversity tends to lower accounting conservatism, possibly due to more inclusive discussions on firm risks, leading to higher-quality disclosures and reduced information asymmetries. These findings are consistent with Jizi et al. (2014), who observed that larger boards influence conservatism because larger boards are associated with not protecting the interests of shareholders and stakeholders.

Furthermore, our analysis confirms the significance of control variables. LEV, BOARD, and FEM are associated with accounting conservatism at the 10, 5, and 1% significance levels, respectively. These results reinforce the importance of considering various organizational factors in understanding financial reporting practices.

5. CONCLUSIONS

Our investigation delved into whether ESG performance is associated with the demand for conditional accounting conservatism among companies listed on the B3 S.A. Our findings suggest that previous higher levels of ESG performance are significantly associated with increased accounting conservatism in the subsequent period, particularly in the context of emerging markets, as noted in previous studies (Garanina & Kim, 2023; Wan & Dawod, 2022). However, when analyzing the current period’s ESG performance, the relationship varies, and in some quantiles, higher ESG performance is associated with a reduced demand for conservatism. Overall, our findings indicate that while there is a significant association between ESG performance and accounting conservatism, the direction and strength of this relationship can vary. This emphasizes the conflicting literature on the topic while highlighting the importance of context and time frame in understanding these dynamics.

Regarding the LEV and MTB variables, our results indicate an association with companies’ demand for conditional conservatism. These variables are well-documented in the literature on accounting information quality and are known to affect the quality measures reported by organizations. Concerning the board characteristics, the board’s size and FEM tend to affect conservative accounting practices within the company.

This offers both conceptual and empirical contributions to the ongoing discourse surrounding ESG performance and conditional conservatism, particularly within the domain of accounting information quality. It extends the boundaries of research in these areas, especially within emerging markets like Brazil, by incorporating insights from a distinct context. Furthermore, it underscores one of the primary objectives of ESG reporting: enhancing disclosed information for decision-making purposes, thereby mitigating information asymmetry between companies and stakeholders while upholding principles of accounting information quality. In this sense, we provide critical insights into how attributes of accounting information quality may require reconsideration when other mechanisms are employed to diminish information asymmetry, such as ESG performance.

Nevertheless, there are limitations inherent in this study that merit consideration. Firstly, while we touch upon sustainability reporting, we do not specifically address the materiality principle concerning the disclosed ESG information. This represents an area for future research to delve deeper into the materiality of ESG disclosures and their relationship with accounting practices. Secondly, our study primarily focuses on an ESG perspective, potentially limiting its scope by not encompassing all dimensions of sustainability (Larrinaga, 2023). Exploring a broader spectrum of sustainability factors could provide a more comprehensive understanding of their association with accounting conservatism. Thirdly, caution is warranted when generalizing our results, as our sample comprises only 83 Brazilian companies that reported an ESG performance index during the defined study period. Future studies could expand the sample size and include companies from diverse geographic regions to enhance the generalizability of findings. Fourth, we did not control for firm fixed effects. This is to control for the perspective of the ESG variable, which does not vary much over the years and may be correlated with the characteristics of each firm.

Moving forward, future studies could explore the relationship between the reverse causality of conditional conservatism variables and ESG performance. Investigating how changes in ESG performance impact conditional conservatism over time could provide valuable insights into the dynamic nature of this relationship. Additionally, further research could examine the association of ESG performance with other metrics of accounting information quality, such as earnings management and financial reporting transparency. By exploring these avenues, we can deepen our understanding of the intricate interplay between ESG performance and accounting practices, ultimately contributing to the advancement of knowledge in this field.

REFERENCES

• Alves, S. (2023). Do female directors affect accounting conservatism in European Union? Cogent Business & Management, 10(2). https://doi.org/10.1080/23311975.2023.2219088
  » https://doi.org/10.1080/23311975.2023.2219088
• Anagnostopoulou, S. C., Tsekrekos, A. E., & Voulgaris, G. (2021). Accounting conservatism and corporate social responsibility. British Accounting Review, 53(4), 100942. https://doi.org/10.1016/j.bar.2020.100942
  » https://doi.org/10.1016/j.bar.2020.100942
• Ball, R., & Shivakumar, L. (2005). Earnings quality in UK private firms: Comparative loss recognition timeliness. Journal of Accounting and Economics, 39(1), 83-128. https://doi.org/10.1016/j.jacceco.2004.04.001
  » https://doi.org/10.1016/j.jacceco.2004.04.001
• Ball, R., & Shivakumar, L. (2006). The role of accruals in asymmetrically timely gain and loss recognition. Journal of Accounting Research, 44(2), 207-242. https://doi.org/10.1111/j.1475-679X.2006.00198.x
  » https://doi.org/10.1111/j.1475-679X.2006.00198.x
• Ball, R., Kothari, S., & Robin, A. (2000). The effect of international institutional factors on properties of accounting earnings. Journal of Accounting and Economics, 29(1), 1-51. https://doi.org/10.1016/S0165-4101(00)00012-4
  » https://doi.org/10.1016/S0165-4101(00)00012-4
• Barbosa, I. S., & Klann, R. C. (2023). Efeitos do desempenho ambiental, social e de governança na relevância das informações contábeis de empresas brasileiras. Revista Contemporânea de Contabilidade, 20(54), 1-15. https://doi.org/10.5007/2175-8069.2023.e86044
  » https://doi.org/10.5007/2175-8069.2023.e86044
• Basu, S. (1997). The conservatism principle and the asymmetric timeliness of earnings. Journal of Accounting & Economics, 24, 3-37. https://doi.org/10.1016/S0165-4101(97)00014-1
  » https://doi.org/10.1016/S0165-4101(97)00014-1
• Boulhaga, M., Elbardan, H., & Elmassri, M. (2023). The effect of internal control and corporate social responsibility on conditional accounting conservatism: Evidence from France. Journal of Corporate Accounting & Finance, 34(2), 228-241. https://doi.org/10.1002/jcaf.22605
  » https://doi.org/10.1002/jcaf.22605
• Bozzolan, S., Fabrizi, M., Mallin, C. A., & Michelon, G. (2015). Corporate social responsibility and earnings quality: International evidence. International Journal of Accounting, 50(4), 361-396. https://doi.org/10.1016/j.intacc.2015.10.003
  » https://doi.org/10.1016/j.intacc.2015.10.003
• Brito, G. A. S., & Martins, E. (2013). Accounting conservatism and the cost of bank credit in Brazil. Brazilian Business Review, 10(1), 27-48. https://doi.org/10.15728/bbr.2013.10.1.2
  » https://doi.org/10.15728/bbr.2013.10.1.2
• Burke, Q. L., Chen, P. C., & Lobo, G. J. (2020). Is corporate social responsibility performance related to conditional accounting conservatism? Accounting Horizons, 34(2), 19-40. https://doi.org/10.2308/horizons-18-111
  » https://doi.org/10.2308/horizons-18-111
• Caskey, J., & Laux, V. (2017). Corporate governance, accounting conservatism, and manipulation. Management Science, 63(2), 424-437. https://doi.org/10.1287/mnsc.2015.2341
  » https://doi.org/10.1287/mnsc.2015.2341
• Cho, S. Y., Lee, C., & Pfeiffer, R. J. (2013). Corporate social responsibility performance and information asymmetry. Journal of Accounting and Public Policy, 32(1), 71-83. https://doi.org/10.1016/j.jaccpubpol.2012.10.005
  » https://doi.org/10.1016/j.jaccpubpol.2012.10.005
• Cui, J., Jo, H., & Na, H. (2018). Does corporate social responsibility affect information asymmetry? Journal of Business Ethics, 148(3), 549-572. https://doi.org/10.1007/s10551-015-3003-8
  » https://doi.org/10.1007/s10551-015-3003-8
• Drempetic, S., Klein, C., & Zwergel, B. (2020). The influence of firm size on the ESG score: Corporate sustainability ratings under review. Journal of Business Ethics, 167(2), 333-360. https://doi.org/10.1007/s10551-019-04164-1
  » https://doi.org/10.1007/s10551-019-04164-1
• Elshandidy, T., & Hassanein, A. (2014). Do IFRS and board of directors’ independence affect accounting conservatism? Applied Financial Economics, 24(16), 1091-1102. https://doi.org/10.1080/09603107.2014.924291
  » https://doi.org/10.1080/09603107.2014.924291
• Environment Programme - Principles for Responsible Investment. (2024). What are the principles for responsible investment? https://www.unpri.org/about-us/what-are-the-principles-for-responsible-investment
  » https://www.unpri.org/about-us/what-are-the-principles-for-responsible-investment
• Fávero, L. P., & Belfiore, P. (2017). Manual de análise de dados. Elsevier.
• Financial Accounting Standards Board. (1980). Statement of Financial Accounting Concepts No. 2. Qualitative characteristics of accounting information. https://fasb.org/page/ShowPdf?path=aop_CON2.pdf&title=CON%202%20(AS%20AMENDED)
  » https://fasb.org/page/ShowPdf?path=aop_CON2.pdf&title=CON%202%20(AS%20AMENDED)
• Francis, R. N., Harrast, S., Mattingly, J., & Olsen, L. (2013). The relation between accounting conservatism and corporate social performance: An empirical investigation. Business and Society Review, 118(2), 193-222. https://doi.org/10.1111/basr.12008
  » https://doi.org/10.1111/basr.12008
• Friede, G., Busch, T., & Bassen, A. (2015). ESG and financial performance: Aggregated evidence from more than 2000 empirical studies. Journal of Sustainable Finance & Investment, 5(4), 210-233. https://doi.org/10.1080/20430795.2015.1118917
  » https://doi.org/10.1080/20430795.2015.1118917
• Garanina, T., & Kim, O. (2023). The relationship between CSR disclosure and accounting conservatism: The role of state ownership.Journal of International Accounting, Auditing and Taxation, 50(2), 100522. https://doi.org/10.1016/j.intaccaudtax.2023.100522
  » https://doi.org/10.1016/j.intaccaudtax.2023.100522
• García Lara, J. M., García Osma, B., & Penalva, F. (2014). Information consequences of accounting conservatism. European Accounting Review, 23(2), 173-198. https://doi.org/10.1080/09638180.2014.882263
  » https://doi.org/10.1080/09638180.2014.882263
• García Lara, J. M., García Osma, B., & Penalva, F. (2020). Conditional conservatism and the limits to earnings management. Journal of Accounting and Public Policy, 39(4). https://doi.org/10.1016/j.jaccpubpol.2020.106738
  » https://doi.org/10.1016/j.jaccpubpol.2020.106738
• Gray, R. (1990). The greening of accountancy: The profession after pearce. Association of Chartered Certified Accountants.
• Gul, F. A., Srinidhi, B., & Ng, A. C. (2011). Does board gender diversity improve the informativeness of stock prices? Journal of Accounting and Economics, 51(3), 314-338. https://doi.org/10.1016/j.jacceco.2011.01.005
  » https://doi.org/10.1016/j.jacceco.2011.01.005
• Hong, S. (2020). Corporate social responsibility and accounting conservatism. International Journal of Economics and Business Research, 19(1), 1-18. https://doi.org/10.1504/IJEBR.2020.103883
  » https://doi.org/10.1504/IJEBR.2020.103883
• Hübel, B., & Scholz, H. (2020). Integrating sustainability risks in asset management: The role of ESG exposures and ESG ratings. Journal of Asset Management, 21(1), 52-69. https://doi.org/10.1057/s41260-019-00139-z
  » https://doi.org/10.1057/s41260-019-00139-z
• International Accounting Standards Board. (2018). Conceptual framework for financial reporting. IFRS Foundation.
• Jizi, M. I., Salama, A., Dixon, R., & Stratling, R. (2014). Corporate governance and corporate social responsibility disclosure: Evidence from the US banking sector. Journal of Business Ethics, 125 (4), 601-615. https://doi.org/10.1007/s10551-013-1929-2
  » https://doi.org/10.1007/s10551-013-1929-2
• Khan, M., & Watts, R. L. (2009). Estimation and empirical properties of a firm-year measure of accounting conservatism. Journal of Accounting and Economics, 48(2-3), 132-50. https://doi.org/10.1016/j.jacceco.2009.08.002
  » https://doi.org/10.1016/j.jacceco.2009.08.002
• Khan, M., Serafeim, G., & Yoon, A. (2016). Corporate sustainability: First evidence on materiality. Accounting Review, 91(6), 1697-1724. https://doi.org/10.2308/accr-5 1383
  » https://doi.org/10.2308/accr-5 1383
• Khemir, S., Baccouche, C., & Ayadi, S. D. (2019). The influence of ESG information on investment allocation decisions: An experimental study in an emerging country. Journal of Applied Accounting Research, 20(4), 458-480. https://doi.org/10.1108/JAAR-12-2017-0141
  » https://doi.org/10.1108/JAAR-12-2017-0141
• Khlifi, S., Boujelbene, M. A., & Chouaibi, J. (2024). The effect of economic, environmental and social sustainability performance on accounting conservatism: The moderating role of good corporate governance. Review of Accounting and Finance, 23(5), 646-664. https://doi.org/10.1108/RAF-08-2023-0291
  » https://doi.org/10.1108/RAF-08-2023-0291
• Kim, Y., Li, S., Pan, C., & Zuo, L. (2013). The role of accounting conservatism in the equity market: Evidence from seasoned equity offerings. Accounting Review, 88(4), 1327-1356. https://doi.org/10.2308/accr-50420
  » https://doi.org/10.2308/accr-50420
• Kim, Y., Park, M. S., & Wier, B. (2012). Is earnings quality associated with corporate social responsibility?Accounting Review, 87(3), 761-796. https://doi.org/10.2308/accr-10209
  » https://doi.org/10.2308/accr-10209
• Klynveld Peat Marwick Goerdeler International. (2022). Big Shifts, small steps: Survey of sustainability reporting 2022. https://assets.kpmg.com/content/dam/kpmg/xx/pdf/2022/10/ssr-small-steps-big-shifts.pdf
  » https://assets.kpmg.com/content/dam/kpmg/xx/pdf/2022/10/ssr-small-steps-big-shifts.pdf
• LaFond, R., & Watts, R. L. (2008). The information role of conservatism. The Accounting Review, 83(2), 447-478. https://doi.org/10.2308/accr.2008.83.2.447
  » https://doi.org/10.2308/accr.2008.83.2.447
• Larrinaga, C. (2023). Sustainability accounting: ESG approaches are not enough. Revista Contabilidade & Finanças, 34(91), 1-4. https://doi.org/10.15 90/1808-05 7x20239042.en
  » https://doi.org/10.15 90/1808-05 7x20239042.en
• Lee, H. S., Li, X., & Sami, H. (2015). Conditional conservatism and audit fees. Accounting Horizons, 29(1), 83-113. https://doi.org/10.2308/acch-50928
  » https://doi.org/10.2308/acch-50928
• Liu, G., Qian, H., Shi, Y., Zhang, Y., & Wu, F. (2024). Does ESG report greenwashing increase stock price crash risk? China Journal of Accounting Studies, 1-25. https://doi.org/10.1080/21697213.2024.2303070
  » https://doi.org/10.1080/21697213.2024.2303070
• Lys, T., Naughton, J. P., & Wang, C. (2015). Signaling through corporate accountability reporting. Journal of Accounting and Economics, 60(1), 56-72. https://doi.org/10.1016/j.jacceco.2015 .03.001
  » https://doi.org/10.1016/j.jacceco.2015 .03.001
• Martin, P. R., & Moser, D. V. (2016). Managers’ green investment disclosures and investors’ reaction. Journal of Accounting and Economics, 61(1), 239-254. https://doi.org/10.1016/j.jacceco.2015 .08.004
  » https://doi.org/10.1016/j.jacceco.2015 .08.004
• Moreira, R. de L., Colauto, R. D., & Amaral, H. F. (2010). Conservadorismo condicional: estudo a partir de variáveis econômicas. Revista Contabilidade & Finanças, 21(54), 64-84. https://doi.org/10.15 90/S1519-70772010000300006
  » https://doi.org/10.15 90/S1519-70772010000300006
• Pan, Y., & Zhao, R. (2022). Does mandatory disclosure of CSR reports affect accounting conservatism? Evidence from China. Emerging Markets Finance and Trade, 58(7), 1975-1987. https://doi.org/10.1080/1540496X.2021.1949283
  » https://doi.org/10.1080/1540496X.2021.1949283
• Pedersen, L. H., Fitzgibbons, S., & Pomorski, L. (2020). Responsible investing: The ESG-efficient frontier. Journal of Financial Economics, 142(2), 572-597. https://doi.org/10.1016/j.jfineco.2020.11.001
  » https://doi.org/10.1016/j.jfineco.2020.11.001
• Penman, S. H., & Zhang, X.-J. (2002). Accounting conservatism, the quality of earnings, and stock returns . The Accounting Review, 77(2), 237-264. https://doi.org/10.2308/accr.2002.77.2.237
  » https://doi.org/10.2308/accr.2002.77.2.237
• Roychowdhury, S., & Watts, R. L. (2007). Asymmetric timeliness of earnings, market-to-book and conservatism in financial reporting. Journal of Accounting and Economics, 44(1-2), 2-31. https://doi.org/10.1016/j.jacceco.2006.12.003
  » https://doi.org/10.1016/j.jacceco.2006.12.003
• Ruch, G. W., & Taylor, G. (2015). Accounting conservatism: A review of the literature. Journal of Accounting Literature, 34, 17-38. https://doi.org/10.1016/j.acclit.2015.02.001
  » https://doi.org/10.1016/j.acclit.2015.02.001
• Schiehll, E., & Kolahgar, S. (2021). Financial materiality in the informativeness of sustainability reporting. Business Strategy and the Environment, 30(2), 840-855. https://doi.org/10.1002/bse.2657
  » https://doi.org/10.1002/bse.2657
• Shankar Shaw, T., Raithatha, M., Krishnan, G. V., & Cordeiro, J. J. (2021). Did mandatory CSR compliance impact accounting conservatism? Evidence from the Indian Companies Act 2013. Journal of Contemporary Accounting and Economics, 17(3), 100280. https://doi.org/10.1016/j.jcae.2021.100280
  » https://doi.org/10.1016/j.jcae.2021.100280
• Sousa, E. F. de, & Galdi, F. C. (2016). The relationship between equity ownership concentration and earnings quality: Evidence from Brazil. RAUSP Management Journal, 51(4), 331-343. https://doi.org/10.1016/j.rausp.2016.07.006
  » https://doi.org/10.1016/j.rausp.2016.07.006
• Van Duuren, E., Plantinga, A., & Scholtens, B. (2016). ESG integration and the investment management process: Fundamental investing reinvented. Journal of Business Ethics, 138(3), 525-533. https://doi.org/10.1007/s10551-015-2610-8
  » https://doi.org/10.1007/s10551-015-2610-8
• Wan, G., & Dawod, A. Y. (2022). ESG rating and northbound capital shareholding preferences: Evidence from China. Sustainability, 14(15), 1-19. https://doi.org/10.3390/su14159152
  » https://doi.org/10.3390/su14159152
• Watts, R. L. (2003). Conservatism in accounting Part I : Explanations and implications. Accounting Horizons, 17(3), 207-221. https://doi.org/10.2308/acch.2003.17.3.207
  » https://doi.org/10.2308/acch.2003.17.3.207
• Yin, M., Zhang, J., & Han, J. (2020). Impact of CEO-board social ties on accounting conservatism: Internal control quality as a mediator. The North American Journal of Economics and Finance, 52, 101172. https://doi.org/10.1016/j.najef.2020.101172
  » https://doi.org/10.1016/j.najef.2020.101172
• Yoon, B., Kim, B., & Lee, J. H. (2019). Is earnings quality associated with corporate social responsibility? Evidence from the Korean market. Sustainability, 11(15), 1-20. https://doi.org/10.3390/su1115 4116
  » https://doi.org/10.3390/su1115 4116

• 4
  This is a bilingual text. This article has also been translated into Portuguese, published under the DOI https://doi.org/10.1590/1808-057x20242049.pt
• Study presented at the XV Congresso ANPCONT, December 2021, and at the EnAnpad 2022 - XLVI Encontro da Anpad, September 2022.
• Financing
  This study was financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes) - Finance Code 001.

Edited by

• Editor-in-Chief:
  Andson Braga de Aguiar

• Associate Editors:
  Márcia Martins Mendes De Luca and Eduardo da Silva Flores

Publication Dates

• Publication in this collection
  08 Sept 2025
• Date of issue
  2025

History

• Received
  28 Nov 2023
• Reviewed
  26 Dec 2023
• Accepted
  18 Nov 2024