Introduction

In the recent years, climate-related risk disclosure (CRRD) gained prominence among many researchers, regulators, and standard setters. This is due to the crucial role that CRRD is playing in meeting stakeholders’ demand for non-financial disclosure best practices [24]. The Climate Disclosure Project [15], the “Global Reporting Initiative” [34], and the “Task Force on Climate-related Financial Disclosures” [75] have lunched global awareness and guidance for firms to understand the importance of disclosing their climate impact. The TCFD committee has mentioned that CRRD is crucial for financial markets and capital-allocation decisions since stakeholders and different users would understand how climate-related opportunities could affect the firm’s future economic performance, as reflected in its annual reports [78].

In the same vein, the “International Sustainability Standards Board” (ISSB) published an exposure draft IFRS: S2 "climate-related disclosures" in March 2022. This standard requires the firm to report information about its climate-related prospects to enhance different users to assess the effect of these risks on the firm’s value, understand how the firm uses its economic resources to manage such risks and opportunities, and evaluate the extent to which firm could adapt its plans regarding those risks and opportunities. Similarly, the US “Securities and Exchange Commission” [68] issued a rule proposal that, if adopted, would require firms to provide certain climate-related information in their annual reports filed with the SEC. This proposed climate-related disclosure rule encourages firms to disclose certain climate-related risks, governance, opportunities, goals, and other related disclosures that could enhance investors’ assessment of such risks. While in Egypt, starting from the year 2023 all EGX-listed firms would mandatorily provide a separated ESG annual report to declare their environmental performance according to the TFCD provisions.

Theoretically, high level of CRRD increases the firm transparency, credibility of applied accounting policies, reported earnings, reputation, and stockholder’s loyalty [23]. According to legitimacy, stakeholder, signaling, and agency theories, firms should report sufficient climate-related information that could meet stakeholders’ needs in making their decisions and enhance the firm legitimacy [28, 60, 65]. Consequently, firms supposed to be less engaged in unethical practices for legitimacy and agency concerns, and to avoid other social and political issues [84].

Carbon disclosure is voluntarily reported information that subject to the firm management’s discretionary judgements. Thus, CRRD could affect the management accounting choices in preparing the firm financial statements, which, in turn, will be reflected on the firm conservatism level. Accounting conservatism appertains to action that revenues should be delayed, and losses should be quickened [25]. Meanwhile, accounting conservatism is considered a main proxy for financial reporting quality [3]. In addition, LaFond and Watts [52] mentioned that accounting conservatism could alleviate the agency problems among the firm management and different outside groups of stakeholders.

Further, earnings quality (EQ) also used as a major proxy for the firm financial reporting quality, which could affect both a firm CRRD and its accounting conservative practices. Prior studies have documented a positive linkage between EQ and accounting conservatism (e.g., [40, 63, 86]). Hence, according to agency theory, managers should be more selective in their chosen of accounting policies to report more accurate earnings and alleviate the downside consequences of reported climate impacts.

However, from legitimacy, stakeholder, and signaling theories perspective, high-quality earnings would motivate managers to report comprehensive and precise climate-related information to maintain the credibility and legitimacy of the firm performance (e.g., [1, 13, 24]). In the same vein, EQ has a positive association with the firm’s conservative practices [40, 63]. Hence, this study aims to answer the following question:

Does EQ of the Egyptian firms listed in the S&P ESG index has an interactive impact on the association between CRRD and firm conservative level?

This study contributes to the accounting literature related to climate changes disclosure in two ways. First, this study suggests a new list of keywords related to the most frequently CRRD used by Egyptian firms in their annual sustainability reports and other voluntary vehicles. Thus, the suggested list of keywords could help in building an index that would smooth the path for future research related to CRRD in the Egyptian context. Second, this study examines the association between CRRD and accounting conservatism in an emerging market, Egypt. Further, this study explores the effect of EQ on such relationship, which represents a main contribution in the literature.

Literature review and hypotheses development

Theoretical framework

To study the interactive impact of firm EQ on the association between CRRD and firm conservatism level, the literature offers theories that could interpret this interrelationship between CRRD, conservatism, and EQ.

Legitimacy theory is the most prominent theory followed by prior studies on environmental disclosure. Legitimacy theory assesses the firm accountability toward external social and environmental values in which it operates [46]. Suchman [72] defined legitimacy as “a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions”. According to Suchman [72], achieving firm legitimacy involves dynamically recognizing and responding to social changes. Thus, firms make efforts to disclose more climate change information to maintain their stewardship role toward stakeholders’ information needs and protect their interests [32, 37].

Consistent with legitimacy theory, stakeholder theory can explain the firm environmental reporting. According to Freeman [27], stakeholders are “groups and individuals who can influence or be influenced by actions related to value creation”. Based on this theory, stakeholders can be influenced by the firm performance and influence on its upcoming performance. Thus, stakeholder theory implies that a firm should credibly disseminate information that serves the interests of all stakeholders (investors and non-investors), and cope with social and climate changes (e.g., [7, 28, 65]).

Hence, from the perspective of legitimacy and stakeholder theory, firms should align their value and performance with the social and environmental values that they operate in to achieve legitimization and best practices regarding sustainability performance [28, 62, 70].

Additionally, both agency and signaling theories interplay in explaining the firm motives toward climate-related disclosure. The two theories suggest the existence of an information gap between executives and stakeholders [43]. According to agency theory, the executive (agent) is delegated to execute some services on the stakeholders’ behalf (principal), which involves assigning some decision-making authority to the management [45]. Meanwhile, stakeholders have the right to control executives’ performance by incentives and governance structures. However, since the managers have an informational advantage over the outside stakeholders, under the unethical hypothesis, managers may reveal misleading information to withhold their poor performance or achieve self-interest. Thus, reporting more voluntary disclosure will shrink this information gap between insiders and outsiders and maintain the firm stewardship [61].

Similarly, signaling theory suggests that executives have more information that is unknown to outsiders, which has the same level of quality and importance [71]. Therefore, any reported information (positive or negative) holds a useful and meaningful signal for outsiders [85]. With regard to CRRD, firms that have good avoidance/mitigation actions toward climate change effects are willing to signal their good performance in the capital market [14].

From the above-mentioned theories, disclosing more precise climate-related information would reduce the information asymmetry between insiders and outsiders and signaling transparent details enhance stakeholders’ oversight of firm legitimacy and avoid investment risks [11, 32, 37].

The link between CRRD and accounting conservatism

Since CRRD is an unfavorable type of information to be disclosed, it might affect the managers’ behavior regarding their use of accounting policies in preparing the firm’s financial statements. Prior accounting literature provides debatable results regarding the relationship between climate-related disclosures and accounting conservatism.

One stream of prior studies argues that environmental disclosures enhance financial statements’ transparency by making more conservative decisions. For instance, Pereira et al. [64], Wu et al. [83], and Zhang and Kanagaretnam [87] have documented that firms reporting a higher level of environmental information are inclined to utilize more conservative accounting practices and estimations. Similarly, Litt et al. [57] and Gerged et al. (2020, 2021) found that firms that provide more information on environmental impacts are less likely to engage in excessive earrings management practices, rather they tend to make conservative accounting decisions and have better financial performance. Further, Liu et al. [58] have concluded that carbon emissions-intensive firms tend to use more conservative accounting policies to achieve tax incentives and reduce litigation and supervision costs by underestimating asset value and expected profits.

Moreover, Xi and Xiao [84] examined the interactive impact of corporate governance attributes on the association between environmental disclosure, excessive earnings manipulation, and accounting conservatism. The results revealed that firms with high levels of disseminated environmental information rely more on conservative accounting choices and report more precise earnings in their annual reports. In the same vein, Liesen et al. [55] and Saini et al. [66] have stressed that publishing more information about the firm climate risk exposure and managing policies in the firm’s sustainability report supports external users’ predictive power of the stock prices and its future earnings in the capital market.

However, Ding et al. [22] claimed that more CRRD drives managers to use more aggressive accounting policies in their financial reporting, particularly for firms operating in developed countries, within sensitive environmental sectors, and achieving losses. In contrast, Kaya and Akbulut [48] found no association between sustainability reporting and using conservative accounting practices by the firm.

Consequently, based on legitimacy, stakeholder, signaling, and agency theories, firms should respond to climate change effects and show that: (i) it is aligned with stakeholders’ expectations by disclosing more information on how climate-related issues are handled in their annual and sustainability reports [28, 60, 65]. As well as, to reduce socio-political pressures [14, 35]. Thus, according to all the above-mentioned theories, CRRD and accounting conservatism are positively associated, since CRRD is considered an indicator of environmental responsibility that facilitates firms’ access to society’s resources [21]. Further, CRRD is regarded as a reliable information and governance tool from the investor’s viewpoint [64].

The above discussion refers to the scarcity of studies that investigated the relationship between CRRD and accounting conservatism, particularly, in an emerging market. The dearth of such an examination may be justified on the ground that measuring climate-related risks is difficult. Therefore, there is a need for further investigation of the nexus between the CRRD and accounting conservatism in an emerging economy such as the Egyptian context. Thus, the current study aims to fill this gap in the literature by formulating the first hypothesis as follows:

H1

There is a positive association between the CRRD of Egyptian firms and the accounting conservatism level.

The role of earnings quality (EQ) on the association between CRRD and accounting conservatism

Earnings are a reflection of a firm's financial performance, determined through accounting accruals, as highlighted by Dechow in 1994. They hold significant value in decision-making, whether it is for internal purposes like executive compensation plans and debt agreements or for external stakeholders like investors and creditors who use them to gauge the firm's future prospects.

Dechow and Schrand [18] define the 'quality' of earnings as accounting figures that not only represent the firm's current economic performance but also serve as a projection of its future operational performance and an assessment of its intrinsic worth. Furthermore, Dechow et al. [19] emphasize that earnings quality depends on the ability of a firm's financial statements to provide informative insights that are relevant to specific decision-makers and decision models.

In general, earnings quality is associated with various important characteristics such as conservatism, the quality of accounting accruals, consistency, predictability, consistency, relevance to the company's value, and timeliness, as noted by scholars like Schipper and Vincent [67], Dechow and Schrand [18], Christensen et al. [12], Dechow et al. [19], and Carmo et al. [10].

Firms are trying to meet stakeholders’ outlooks by disclosing more information on how environmental, social, and governance issues are handled in their annual and sustainability reports [7, 28, 65]. Based on the legitimacy, stakeholder, and signaling theories, effective managers prefer to signal their social and environmental performance by reporting more voluntary CRRD on the firm’s annual sustainability report or on its website to external users [39]. Thus, high-quality earnings would motivate managers’ inclination to increase the quantity of reported CRRD to meet society’s expectations and legitimize their performances. In turn, this is expected to increase managers’ conservative accounting choices.

In this regard, prior research shows a complementary liaison between voluntary disclosure and EQ to enhance users’ decisions in the capital market (e.g., [26, 54, 80]). Regarding voluntary CRRD, previous studies found that firms with higher EQ tend to disclose more environmental information, which, in turn, enhances the reliability and credibility of the firm’s financial reporting and reduce the firm cost of capital [1, 6, 13, 24].

Clarkson et al. [13] found that voluntary CRRD is value-relevant and enhances a firm’s market value by signaling the management’s proactive environmental tactics to investors. Similarly, Alipour et al. [1] documented a positive linkage between ecological disclosure quality and EQ (proxied by accruals quality). In the same vein, Ellili [24] concluded that compliance with ESG best practices increases reporting quality and improves investment decision efficiency. Further, Bui et al. [7] pointed out a negative association between carbon disclosure and earnings management practices. The results suggested that firms with high-quality governance performance are more committed to their oriented sustainability performance, signaling their carbon management superiority, and thus improving their financial reporting quality.

However, Lee et al. [53] documented that voluntary CRRD has a negative impact on capital market returns since investors perceive this information as bad news and anticipate fewer benefits due to the costs of coping with global warming. The findings concluded that continuing disclosure on management climate changes policies mitigates this negative perception regarding reported carbon disclosure and the market stock price. In the same vein, Lemma et al. [54] claimed a negative association between CRRD and the firm’s EQ, however; this association is partly explained by voluntary CRRD.

In Egypt, the empirical findings of Ismail and Elbolok [44] based on a sample comprising the largest 30 Egyptian listed companies spanning the years from 2005 to 2009 indicate that these firms practiced conservative accounting, and there is a negative correlation between EQ and stock prices concerning conditional conservatism, while no connection exists between unconditional conservatism and EQ. Additionally, Attia et al. [2] conducted an investigation into the impact of BOD characteristics on real earnings management using a sample of Egyptian listed firms. The results lend support to the credibility of financial statements published in the Egyptian Stock Market, as BOD characteristics are demonstrated to play a crucial role in reducing earnings management practices.

However, based on the obfuscation hypothesis, managers may cover their poor performance when the firm’s earnings quality is low by decreasing the level of the reported CRRD, which, in turn, would have an impact on conservative accounting practices. For instance, Jiang et al. [46] provide evidence that managers tend to select specific types of carbon information to disclose in the firm’s annual reports to repair its green image, particularly, in high-intensive carbon sectors. Though, other firms in low-carbon sectors disseminate CRRD as a governance tool. Similarly, Velte [79] found that managers shift from accounting earnings management to real earnings management as a greenwashing technique toward the reported environmental risks. Conversely, Sun et al. [73] found no association between the firm’s CRRD and earnings management practices.

Regarding the association between accounting conservatism and EQ, the accounting literature documented a positive impact of conservative accounting on the firm’s EQ (e.g., [40, 63]). Meanwhile, conservatism is used to proxy a firm’s EQ, “if it is measured by the asymmetric effects of positive and negative changes in earnings” [3].

Penman and Zhang [63] concluded that EQ resulting from a composite of real operations and accounting policies, not only based on changes in accounting practices and estimates. This implies that managers can use the effect of real operations and accounting policies to manipulate the reported earnings. In this regard, Martínez-Ferrero et al. [59] documented that firms engaged with more conservative accounting choices report high-quality earnings and high-quality social and environmental disclosure. Similarly, Hartam and Kresnawati [40] have explored the impact of accounting conservative practices on EQ and determined the interactive effect of the firm’s life cycle. The results indicated a positive association between conservative accounting practices and EQ. In contrast, Zadeh et al. [86] found a negative association between using conservative practices and the reported EQ. Thus, the firm’s accounting conservatism level could not be the sole indicator of its EQ.

Further, as mentioned earlier, Litt et al. [57], Gerged et al. (2020, 2021), and Xi and Xiao [84] found a consistency in the association between the three variables: firm’s high level of CRRD, conservative accounting practices, and reporting accurate earnings. Accordingly, it is expected that EQ has an impact on the association between CRRD and accounting conservatism.

From the above discussion, it is noticed that the results of prior studies are inconsistent and mixed, which represents a gap in the literature. Moreover, no previous studies have investigated the interactive effect of EQ on the association between CRRD and accounting conservatism, particularly in the Egyptian context. Hence, the second hypothesis is formulated as follows:

H2

EQ has a significant impact on the association between CRRD of the Egyptian firms and the accounting conservatism.

Research method

Study variables and measurements

This study aims to explore the association between CRRD and accounting conservatism, in addition, to test the interactive influence for firm EQ on this association. A manual content analysis was carried out to measure the CRRD level of the Egyptian firms.

Measurement of CRRD Prior related studies have identified the following proxies for measuring CRRD, namely absolute level of carbon emissions [9], carbon intensity [13, 16], carbon event-related measures proxied by dummy variable [74], CRRD rating/score [22, 24, 84], CRRD indices [14, 30, 31], Pereira and Monteiro [64], and content analysis [16, 65], Tóth et al. [77].

Due to the fact that disclosing CRRD is a recent practice in the Egyptian context, this study uses manual content analysis to identify the keywords and techniques the Egyptian firms followed to disseminate climate-related information. Our unit analysis is “sentence" to capture a piece of complete information [51, 56]. Each sentence includes at least one of the CRRD keywords (as shown in Appendix 1). The list of keywords is developed based on Clarkson et al.’s [14] and Datt et al. [16] climate-related scoring scheme. Further, we read 20 randomly selected annual sustainability report (narrative environmental section) to figure out the most common CRRD keywords used by Egyptian firms.

Our content analysis also covered firms’ quantitative CRRD reported in tables, where each row counted as one sentence, as well as, each complete numeric information published in a comparative analytics diagram counted as one sentence. Since this numeric CRRD information reflects changes in the firm’s environmental performance yearly. Examples of CRRD sentences reported by the Egyptian firms are presented in Appendix (2).

We define CRRD as all information related to the environmental impacts that the firm published voluntarily through its sustainability report, official website, investor’s presentations, board of directors’ report, annual integrated report, carbon footprint report, and any other environmental-related report. CRRD is information that depicts a firm’s major climate-related perils and their expected impact on the firm economic and environmental future performance. This information includes identifying climate-related risks, assessment techniques, forward-looking strategies, mitigation actions, and previous policies taken to alleviate these risks.

All statements in the narrative sections of the annual sustainability reports and other voluntarily environmental disclosures that contained at least one word from our final climate-related risk keyword list are considered. The CRRD score is the natural log of the number of statements indicating CRRD in the narrative sections of all voluntary environmental reports.

Measurement of accounting conservatism. Researchers have developed various measures to estimate a firm's level of conservatism. Wang et al. [81] categorized common measures, including asymmetric timeliness measure [4], market-to-book ratio [5], cumulative negative non-operating accruals [33], hidden reserves [63], asymmetric cash flow to accruals [3]. C-Score measure of conservatism by Khan and Watts [49], based on Basu's [4] timeliness concept, is widely used in prior studies (e.g., [50, 69]) to estimate conditional conservatism on a firm-year basis.

To estimate Egyptian firm’s accounting conservatism, we followed Kim & Zhang [50] and Shen et al. [69] and used Khan and Watts [49] (C-Score) firm-year conditional conservatism measurement. Based on Basu’s [4] model, earnings are expressed as an asymmetric function of stock returns:

$$E_{i} = \beta_{0} \, + \, \beta_{1} D \, + \beta_{2} D_{i} + \beta_{3} D_{i} *R_{i} \, + \varepsilon_{i}$$
(1)

where E is earnings; R is returns, and D is a dummy variable that is equal to 1 when R is negative, and 0 otherwise; and εi is a random error term. The coefficient β2 indicates the timeliness of earnings good news, the coefficient β3 indicates the incremental effect of the timeliness of bad news relative to good news disclosed (conservatism level), and by adding the two coefficients (β2 + β3), the timeliness of total bad news disclosed is detected. The firm-year level of good news confirmation and accounting conservatism level can be stated as follows:

$$G\_{\text{Score}} = \beta_{2} = \mu_{1} + \mu_{2} {\text{Size}}_{i} + \mu_{3} {\text{MTB}}_{i} + \mu_{4} {\text{Lev}}_{i}$$
(2)
$$C\_{\text{Score}} = \beta_{3} = \omega_{1} + \omega_{1} {\text{Size}}_{i} + \omega_{3} {\text{MTB}}_{i} + \omega_{4} {\text{Lev}}_{i}$$
(3)

where Sizei is the natural logarithm of total assets, MTBi is the market-to-book value ratio of firm’s equity, and LEVi is the firms’ leverage ratio. Equations (2) and (3) are combined into Eq. (1) to estimate the regression Eq. (4), to originate G_Score (annual timeliness of good news) and C_Score (annual additional timeliness of bad news). Firms with higher C_Scores are more conservative in their accounting practices than firms with lower C_Scores. Thus, the annual cross-sectional regression model used to estimate these parameters is as follows:

$$E_{i} = \beta_{{\text{o}}} + \beta_{1} D_{i} + R_{i} (\mu_{1} + \mu_{1} {\text{Size}}_{i} + \mu_{3} {\text{MTB}}_{i} + \mu_{4} {\text{Lev}}_{i} ) + D_{i} *R_{i} (\omega_{1} + \omega_{2} {\text{Size}}_{i} + \omega_{3} {\text{MTB}}_{i} + \omega_{4} {\text{Lev}}_{i} ) + (\nu_{2} {\text{Size}}_{i} + \nu_{3} {\text{MTB}}_{i} + \nu_{4} {\text{Lev}}_{i} + \nu_{2} D_{i} *{\text{Size}}_{i} + \nu_{3} D_{i} *{\text{MTB}}_{i} + \nu_{4} D_{i} *{\text{Lev}}_{i} ) + \varepsilon_{i}$$
(4)

Measurement of earnings quality. The absolute value of short-term discretionary accruals was used to alternate the firm’s EQ. The higher absolute value of short-term discretionary accruals refers to more opportunistic managerial manipulation, implying lower EQ [20, 76]. The following steps show the estimate of the short-term discretionary accruals. First, to calculate the total current accruals (TCAit) for each firm, the following equation was used:

$${\text{TCA}}_{it} = (\Delta {\text{CA}}_{it} - \Delta {\text{Cash}}_{it} ) - (\Delta {\text{CL}}_{it} - \Delta {\text{STD}}_{it} )$$
(5)

where ΔCAit is the change in current assets, ΔCasℎit is the change in cash, ΔCLit is the change in current liabilities, and ΔSTDit is the change in short-term debt. Then, we run the following regression for each firm.

$${\text{TCA}}_{it} /{\text{TA}}_{it - 1} = \alpha_{0} (1 - /{\text{TA}}_{it - 1} ) + \alpha_{1} (\Delta {\text{REV}}_{it} - \Delta {\text{REC}}_{it} /{\text{TA}}_{it - 1} ) + \varepsilon_{it}$$
(6)

where TCAit is the firm’s (i) total short-term accruals at year (t), ΔREVit is the change in net revenues, and ΔRECit is the change in net receivables. After that, each variable was divided by the previous year total assets (TAit − 1), to correct for heteroscedasticity. Third, to split off the innate accruals from the total short-term accruals for each firm, the following equation was used:

$${\text{NDAC}}_{it} = \hat{\alpha }_{0} (1/{\text{TA}}_{it - 1} ) + \hat{\alpha }_{1} (\Delta {\text{REV}}_{it} - \Delta {\text{REC}}_{it} /{\text{TA}}_{it - 1} )$$
(7)

Finally, we deduct the innate accruals from the total short-term accruals to calculate the short-term discretionary accruals for each firm, as follows:

$${\text{DAC}}_{{{\text{it}}}} = {\text{TCA}}_{it} /{\text{TA}}_{it - 1} - {\text{NDAC}}_{it}$$
(8)

Measures of control variables we control some of the firm characteristics that could influence the accounting conservatism level. Referring to previous studies [30, 48, 64, 73, 86], this study uses the following control variables: Sizeit the natural log of total assets, Levit leverage ratio of total debt to total equity MTBit market-to-book ratio, ROAit return on assets for firm profitability; GRWit represents variation in the firm sales, and Lossit to indicate loss firms, which might have incentives to affect its conservative accounting practices. Table 1 shows the study variables and the related measures.

Table 1 Variable definitions and measures
Full size table

Empirical models

This study attempts to explore the association between CRRD and accounting conservatism. Second, to test whether firm EQ has an interactive impact on the association between CRRD and accounting conservatism, we run the following two main models. The first regression model is as follows:

$${\text{Cons}}_{{{\text{it}}}} = \alpha _{0} + \beta _{1} {\text{CRRD}}_{{{\text{it}}}} + \beta _{2} {\text{Size}}_{{{\text{it}}}} + \beta _{3} {\text{LEV}}_{{{\text{it}}}} + \beta _{4} {\text{MTB}}_{{{\text{it}}}} + \beta _{5} {\text{ROA}}_{{{\text{it}}}} + \beta _{6} {\text{GRW}}_{{{\text{it}}}} + \beta _{7} {\text{Loss}}_{{{\text{it}}}} + \varepsilon _{{{\text{it}}}} \quad ({\text{Model}}\;1)$$

where Consit is C- Score of conditional conservatism of the firm (i) in the year (t).CRRDit is the carbon-related risk disclosure of the firm (i) in the year (t). Sizeit is the firm (i) size in the year (t). LEVit, is the firm (i) leverage in the year (t). MTBit is the firm (i) growth opportunity in the year (t). ROAit is the firm (i) return on assets in the year (t). GRWit is the firm (i) growth in sales in the year (t). Lossit, is the firm (i) negative earnings in the year (t), and Ɛit is the Random error.

To test the interaction effect of both CRRD and EQ on the firm accounting conservatism, the second regression model is used as follows:

$${\text{Cons}}_{{{\text{it}}}} = \alpha _{0} + \beta _{1} {\text{CRRD}}_{{{\text{it}}}} + \beta _{2} {\text{EQ}}_{{{\text{it}}}} + (\beta _{3} {\text{CRRD}}_{{{\text{it}}}} *{\text{EQ}}_{{{\text{it}}}} ) + \beta _{4} {\text{Size}}_{{{\text{it}}}} + \beta _{5} {\text{LEV}}_{{{\text{it}}}} + \beta _{6} {\text{MTB}}_{{{\text{it}}}} + \beta _{7} {\text{ROA}}_{{{\text{it}}}} + {\mkern 1mu} \beta _{8} {\text{GRW}}_{{{\text{it}}}} + \beta _{9} {\text{Loss}}_{{{\text{it}}}} + \varepsilon _{{{\text{it}}}} \quad ({\text{Model}}\;2)$$

where EQit is the firm (i) earnings quality in the year (t), (CRRDit * EQit) is the interactive impact of both CRRD and EQ on the firm accounting conservatism. Other variables are as defined in the multiple regression Model (1).

Sample selection and data sources

Our initial sample composed of Egyptian firms listed in SP/EGX ESG index each year over the period from 2018 to 2022, since these firms are considered the top 30 firms that disclose environmental, social, and governance information compared to other firms listed in EGX. Thus, our initial sample consists of 150 observations.

A firm to be included in the sample must meet the following criteria:

  1. (i)

    It is listed in the SP/EGX ESG index.

  2. (ii)

    Has published climate change data available either in its annual sustainability report, integrated report, website, or any other environmental report over the study timescale from the year 2018 to 2022.

  3. (iii)

    Has available financial data from Thomson and Reuters database.

It was noted that not all firms have published sustainability reports or any other voluntary CRRD on their websites, where there are unavailability of carbon disclosure of about 10 firms over the study period; hence, such firms were excluded from the sample. The final sample comprises of 95 firm-year observations, which represent 63% of the initial sample. Table 2 shows details of the sample selection.

Table 2 Sample selection
Full size table

Climate-related information were collected from the firm’s annual sustainability reports, ESG report, carbon footprint report, integrated reports, BOD reports, and their websites, and EGID for unpublished sustainability reports and BOD reports. Other financial data related to accounting conservatism, earnings quality, and control variables were collected from Thomson & Reuters database. The data were statistically analyzed using STATA (14) software.

Empirical findings

This paper employs ordinary least squares (OLS) for testing our models and utilizes generalized method of moments (GMM) regression to address potential endogeneity concerns as a reliability check. We also applied winsorization at the 1% and 99% percentiles for all variables to mitigate outliers' impact. Furthermore, we adjusted the standard errors of our empirical models to account for heteroskedasticity.

Descriptive and univariate analysis

Table 3 shows the descriptive statistics of the study’s variables for the years 2018 to 2022. Our results pointed to some variances in the variables’ distributional characteristics. For example, the accounting conservatism level ranges between − 5.61 to 0.33 with a mean of − 0.34. This is in line with Xi and Xiao’s [84] results.

Table 3 Descriptive analysis of variables
Full size table

Additionally, the CRRD score exhibits a standard deviation of 0.48 and a mean value of 1.77, ranging from a minimum of 0.69 to a maximum of 2.92. These findings suggest that the majority of the Egyptian firms in our sample demonstrate a relatively low level of CRRD. Our results align with earlier research [30, 31, 57, 84]. Moreover, the EQ level varies from − 0.68 to 0.83, with an average value of 0.07, indicating a low level of emotional intelligence (EQ) among the sampled firms.

Table 4 illustrates the correlation analysis of our tested variables. There is no significant correlation between Cons and CRRD or EQ. It correlates positively with firm leverage, gross level, and profitability. However, Cons is correlated negatively with firm size and loss. As appears in Table 4, no multicollinearity issues among independent variables are noticed, since none of the correlation coefficients exceed 0.70. Further, the mean VIF for model (1) is 1.45 and 8.2 for model (2), both are less than 10.

Table 4 Pearson coefficient correlation analysis
Full size table

Multivariate analysis

To assess the impact of CRRD and EQ on accounting conservatism levels, this study conducted two ordinary least square (OLS) models. The results of these models are summarized in Table 5 and depicted in Fig. 1. The first model, examining the relationship between CRRD and accounting conservatism, revealed a negative and statistically insignificant effect of CRRD on firm accounting conservatism. This suggests that CRRD does not account for variations in a firm's conservative financial practices. In other words, the conservative practices of the sampled Egyptian firms appear to be unaffected by CRRD. Consequently, the first null hypothesis is accepted, and the results of Model (1) support the rejection of H1.

Table 5 OLS Results for models (1) and (2)
Full size table
Fig. 1
figure 1

The association between variables

Full size image

However, Model (2) was employed to investigate whether a firm's EQ has an interactive influence on the relationship between CRRD and accounting conservatism. The results indicate a significant and negative effect, significant at the 1% level, for the interaction between CRRD and EQ concerning a firm's conservative financial practices. This implies that when there is more managerial opportunism, as evidenced by higher absolute values of discretionary accruals (which are equivalent to low EQ), it leads to a lower level of accounting conservatism. Furthermore, the R-squared value increased from 22% (as seen in Model 1) to 25% (as seen in Model 2), indicating improvements attributed to the inclusion of EQ in the model. The coefficient of the moderator variable is − 1.608, suggesting that accounting conservatism is more likely to increase when managers reduce their earnings management practices and report high-quality earnings and high-quality CRRD.

Additionally, the results demonstrate a significantly negative correlation between firm size and conservative financial practices in both Model (1) and Model (2). This finding is consistent with the results of Ding et al. [22], Liu et al. [58], and Xi and Xiao [84]. Furthermore, the results indicate a significantly positive relationship between firm profitability and the level of accounting conservatism. This aligns with the findings of Kaya and Akbulut [48] and Ding et al. [22], suggesting that firms achieving high profitability are more likely to employ conservative accounting practices. However, this result contradicts the findings of Xi and Xiao [84], which may be attributed to the use of different control variables in our model. Finally, as observed in Table 5, no notable changes were observed in the relationship between accounting conservatism and other control variables when EQ was introduced as a moderator in the regression model.

Discussion

From the statistical analysis of the regression models that tested the interactive effect of EQ and CRRD on the firm accounting conservative practices, we find contradicting results among the two models. The results of regression Model (1) show an insignificant association between CRRD and firm accounting choices in financial reporting. This result is consistent with Kaya and Akbulut’s [48] results which indicated no impact on environmental risk disclosure and the firm conservatism level.

However, after adding the EQ as a moderator in the regression Model (2), our results reveal a significant negative relationship between CRRD and firm accounting conservatism level. This implies that Egyptian firms with high levels of environmental impacts tend to greenwash their reported earnings through accounting accruals, which, in turn, leads to following less accounting conservative choices. This result aligns with the agency theory assumption of a positive association between environmental risks and management unethical practices. Our finding is consistent with the results of Lemma et al. [54], Ding et al. [22], and Velte [79]. Moreover, our results align with those of Pereira et al. [64], who observed that firms tended to exhibit reduced conservative practices when they disclosed higher levels of climate-related information.

Moreover, in line with legitimacy, agency, stakeholder, and signaling theories, outsiders tend to perceive firms with more voluntary disclosure as having fewer agency issues, which, in turn, reduces their demand for conservative financial statements [8, 36, 64]. Therefore, the negative relationship between CRRD and accounting conservatism is consistent with the literature, as it suggests that firms with more environmental impacts try to mask their poor performance and manipulate their earnings, of maximizing the interest of specific groups of stakeholders [29].

Robustness tests

In this section, we conduct further sensitivity tests to address potential endogeneity concerns and to validate our findings. Specifically, we employ a two-step system GMM estimation technique, which incorporates lagged conservatism estimates into our model. The results, presented in Table 6, affirm that there are no endogeneity problems in our models, and all key assumptions remain consistent when employing the GMM method. The Hansen test of over identification for models (1) and (2) (0.182 and 0.108, respectively) confirms the validity of the GMM instruments, while the AR (2) test for models (1) and (2) (0.333 and 0.286, respectively) indicates the absence of autocorrelation issues in the residuals.

Table 6 System GMM results for models (1) and (2)
Full size table

Further, we employ the static two-stage least squares regression 2SLS estimator, where lagged firm performance (ROAt−1) is used as an instrumental variable. The results in Table 7 reveal the same conclusions presented in Tables 5 and 6, wherein Model (2), the results validate the significant negative association between CRRD and firm accounting conservatism after the moderation impact of the EQ at level 10%. Additionally, we account for industry type (industrial firms worth 1, otherwise worth 0), and the results presented in Table 8 demonstrate that there are no significant changes in the main regression results when this control variable is included.

Table 7 2SLS Results for models (1) and (2).
Full size table
Table 8 Additional analysis: controlling industry type
Full size table

Conclusion, limitations, and suggestions for future research

In recent times, regulatory bodies and industry-driven initiatives have placed a growing emphasis on enhancing CRRD, where the Task Force frameworks have gained widespread adoption among firms globally.

The current study extends empirical research by investigating the moderating effect of the EQ on the association between CRRD and accounting conservatism in an emerging setting of Egypt. The study analysis based on both univariate and multivariate statistical analysis using 95 firm-year observations over the period 2018–2022. The results reveal no liaison between CRRD and accounting conservatism; however, after incorporating the EQ in the regression model, the results show a significantly negative liaison between CRRD and the firm’s conservative practices.

Our findings have some theoretical and practical implications. For academics, this study extends prior literature that examined the consequences of climate-related disclosures on firm accounting practices. Further, this study provides insights for researchers toward a deeper understanding of current regulations and mandatory reporting of climate impacts by Egyptian firms. In addition, for managers, our results show that EQ is crucial in explaining the association between CRRD and conservative practices in the Egyptian setting, as well as, our results show low levels of both CRRD and EQ. Thus, managers must endeavor to improve the quality of the reported climate-related risks and reported earnings to enhance conservative practices in preparing the annual financial reports. This, in turn, will increase the credibility of the firm’s financial reporting and social legitimacy. For investors and other stakeholders, identification of frequent expressions used by Egyptian firms regarding CRRD voluntary channels could enhance them in making their investment decisions. Moreover, our results have some implications for policymakers in Egypt. The content analysis detects that approximately 40% of the Egyptian firms listed in the SP/EGX ESG index do not publish any voluntary climate disclosure, even though all listed firms published their ESG report for the year 2022 only. Hence, it would be more effective to issue stricter regulations and penalties to force listed firms to report some details on CRRD in annual financial statements’ narratives, to consistently describe more about what they mention in their published ESG reports.

This study, however, has some limitations. First, the small sample size may add cautions in explaining and generalization of the results. Further, we may consider the possibility of endogeneity issues existing even in our additional GMM results due to the small size. Hence, sample size can be extended by future research to include all the Egyptian firms listed on EGX 100 after the year 2023, as all listed firms will be obliged to publish an ESG report with the regular annual financial statements. Second, our models consider only the quantity of CRRD and ignore its quality. Thus, future research might examine whether the quality of the reported climate-related information matters to a firm conservative accounting choices. Third, this study only focuses on climate change risk disclosures (environmental risks) and does not consider other social and governance factors related to the ESG matrices. Further, examining the extent to which managerial power influences climate-related risk dissemination could also be investigated by future research.