Capturing Disclosure Tone in Saudi Arabia: Do Earnings Management and Accounting Conservatism Matter?

Abstract

This study aimed to analyze the determinants of disclosure tone (DT) in the Saudi business environment during the last nine years. In addition, it tested the impact of earnings management and accounting conservatism on this tone. The study followed a mixed-method approach, “quantitative and qualitative”, to explore the relationships used for the content analysis to analyze the annual reports of a sample of 88 Saudi-listed firms from 2014 to 2022. The results of the study found that there is a positive impact of dividend yield on disclosure tone. Conversely, both firm size and leverage do not have a significant impact. Moreover, earnings management as an accounting practice has a curvilinear effect on disclosure tone, and accounting conservatism as a generally accepted principle positively influences disclosure tone.

1. Introduction

In the current era, the concept of “disclosure” in accounting academic research refers to the formal documents and reports a firm produces to communicate its financial and non-financial performance with stakeholders [1,2]. This includes quarterly reports, annual reports and other official reports [3]. Consequently, “disclosure tone” expresses the comprehensive attitude conveyed by the language used in these reports. It is not just about individual words but the overall impression created by the writing style, word choice and paragraph structure [4]. More precisely, it alludes to firms’ general attitude and phrasing when disclosing information. Therefore, it examines the subjective way these details are presented in addition to the factual numbers [5,6]. This “tone” can be positive (optimistic) or negative (pessimistic).

It is important to consider that there are many factors for comprehending disclosure tone in accounting thought [4,7,8], such as the focus on language, which explores how information is disclosed, paying attention to paragraph structure. In addition, emotional implication attempts to express the feelings that the words imply. Going further, the impact on interpretation and stakeholders’ perceptions of the firm’s financial performance can be mainly influenced by the tone used in disclosure. More deeply, understanding disclosure tone is becoming increasingly important in accounting research as it gives insights into decision-making [9]. It can offer evidence about the management’s attitudes toward the firm’s current and future performance [10]. Moreover, the effectiveness of financial communication can help in assessing how successfully firms are providing stakeholders with objective financial and nonfinancial information.

Luo and Zhou [11] and Bassyouny et al. [12] provided many limitations of disclosure tone research, such as “Correlation & Causation”. While there may be a relation between a firm’s disclosure tone and other factors, it can be hard to prove a direct cause-and-effect relationship between them. For instance, a positive tone in a firm’s tone could suggest real optimism, but it could also be an effort to manipulate stakeholders. One more limitation is “Intentions & Perceptions”. It can be difficult to confirm the intentions behind a particular tone. Managers may not always be intentionally manipulating the tone, as it is the perception of the stakeholders that holds importance. Therefore, it becomes crucial for managers to be mindful of the impact their tone has on others. Henry and Leone [13] highlighted some of the difficulties involved in measuring the tone of disclosure. Two challenges were mentioned: “subjectivity and context dependence”. Firstly, interpreting the emotional tone of language can be subjective, with different people potentially assigning different tones to the same text. Secondly, the meaning of words and phrases can vary depending on the context, so it is important to consider the overall content of a report alongside the language used to get an accurate understanding of the disclosure tone.

More deeply, the disclosure tone is a crucial issue in the Saudi business environment due to the expanding investment landscape following the launch of the Saudi Vision 2030 in 2016. As Saudi Arabia is in the process of diversifying its economy to attract foreign investment, it is crucial to provide clear and transparent financial disclosures with a neutral or positive tone to build trust with international investors. Moreover, The Saudi Capital Market Authority (CMA) emphasizes transparency and fair disclosure practices. Furthermore, the Saudi business culture leans toward relationship-building and fostering trust, and using a professional tone in financial reporting can be seen as a signal of respect toward stakeholders.

Going further, earnings management is the intentional effort made by managers to influence the disclosed earnings by using accounting choices [9]. Such choices can include accelerating or delaying the recognition of expenses and revenues to create a more favorable financial performance [14,15]. On the other hand, accounting conservatism calls for a careful approach to financial reporting, prioritizing the recognition of potential losses over gains [3,16]. This study investigates how these different approaches may affect the disclosure tone used in the Saudi business environment as an emerging stock market. It aims to fill this gap by examining how these accounting practices impact the level of positivity (optimistic) or negativity (pessimistic) in a firm’s financial reports.

The rest of this study is structured as follows: In part two, we will review related literature, theoretical considerations and hypothesis development. This section will mainly focus on the factors that influence disclosure tone, earnings management and accounting conservatism. In part three, we will develop the research models. Part four will present the results, followed by the discussion in part five and the conclusion in part six.

2. Literature, Theoretical Considerations and Hypotheses

2.1. Determinants of Disclosure Tone

The academic literature review underscores the importance of capturing the “determinants of disclosure tone” in studying the firm’s behavior in managing relations with stakeholders through information emotion in any business environment, especially firm size, firm age, leverage, growth rate and dividend yield.

2.1.1. Firm Size (FS)

Positive accounting theory suggests firms might use disclosure tone to influence stakeholders’ perceptions. Smaller firms facing greater information asymmetry and higher contracting costs might be more inclined to use an optimistic disclosure tone to signal their financial health and attract investors [17,18]. Conversely, larger firms with established reputations might not need to manage disclosure tone as much, potentially adopting a more neutral or conservative tone to comply with regulations [3,19]. On the other hand, legitimacy theory suggests that any firm’s main objective is to meet its stakeholders’ expectations. For larger firms, maintaining legitimacy is an important issue. Therefore, using an optimistic tone in its disclosures can serve as a way to show financial health and gain trust and legitimacy from stakeholders.

In terms of empirical evidence, many studies have emphasized the positive impact of firm size on disclosure tone. Allee and Deangelis [20] and Luo and Zhou [11] noted that larger firms tend to avoid risk and focus on mitigating potential legal issues, and using a positive disclosure tone can help reduce the risk of misunderstandings and subsequent lawsuits. Moreover, Alalwani and Mousa [21] and Arslan-Ayaydin et al. [22] argued that larger firms have a wider range of stakeholders, including investors, who are increasingly focused on disclosure language to make decisions. This can lead firms to adopt a more positive and optimistic disclosure tone. From another perspective, there are several reasons why firm size can lead to a negative impact on disclosure tone. Aly et al. [23] and Beretta et al. [24] found that large firms have complex business structures. This complexity can make it difficult to generate and communicate information clearly and concisely, resulting in a more opaque and negative tone in disclosures. Moreover, Chakraborty and Bhattacharjee [10] and Elshandidy and Zeng [1] argued that large firms are concerned about managing their image in front of stakeholders, especially in the case of negative news or performance. This can result in a more defensive or pessimistic disclosure tone.

Based on the above, the impact of firm size (FS) on disclosure tone (DT) remains an area of ongoing academic research, so the following hypothesis can be developed:

H1. 

There is a positive impact of firm size on disclosure tone in Saudi Arabia.

2.1.2. Firm Age (FA)

Stakeholder theory emphasizes that younger firms with fewer stakeholders and less complex stakeholder networks might have greater flexibility in their disclosure tone and might prioritize attracting investors through a more optimistic tone [9,17], whereas older firms with more complex stakeholder networks and established relationships might be more likely to involve a more neutral or conservative tone to maintain trust and legitimacy with diverse stakeholders [11,20].

In terms of empirical evidence, many studies have emphasized the positive impact of firm age on disclosure tone. Aly et al. [23] and Beretta et al. [24] emphasized that over time, firms can refine their disclosure practices, which can result in a clearer and more positive disclosure tone. Moreover, Chakraborty and Bhattacharjee [10] and Elshandidy and Zeng [1] argued that older firms often have a more stable business model and a clearer view of their prospects, and this reduced uncertainty can allow them to adopt a more confident and optimistic disclosure tone. From another perspective, there are several reasons why firm age can lead to a negative impact on disclosure tone, Campbell et al. [25] and Del Gaudio et al. [26] noted that over time, firms may become more risk-averse and focused on preserving the status quo, which can result in a more pessimistic disclosure tone. Moreover, Arena et al. [27] and Pouryousof et al. [28] argued that older firms are subject to greater institutional pressure from regulators, and this pressure can encourage them to adopt more conservative or pessimistic disclosure tones to maintain legitimacy.

Based on the above, the impact of firm age (FA) on disclosure tone (DT) remains an area of ongoing academic research, so the following hypothesis can be developed:

H2. 

There is a positive impact of firm age on disclosure tone in Saudi Arabia.

2.1.3. Leverage (LEV)

Signaling theory suggests that firms that have a high level of debt may use a positive tone when disclosing financial information. This is done to indicate that the firm is financially stable and to reduce concerns about the risk of insolvency [29,30]. By doing so, the firm hopes to reduce the cost of financing debt and to reassure creditors and investors [6,31].

In terms of empirical evidence, many studies have emphasized the positive impact of leverage on disclosure tone. Durnev and Mangen [8] and Rich et al. [2] noted that firms with high levels of debt are being watched more closely by creditors. So, they may use a positive tone to highlight their financial strength and their ability to manage their debt obligations. Moreover, Ataullah et al. [3] and Wang [18] argued that firms use a positive tone in their communications to signal that they are confident in their ability to generate enough cash to pay off their debt in the future. From another perspective, there are several reasons why leverage can lead to a negative impact on disclosure tone. Al Lawati et al. [4] and Li et al. [9] noted that loan agreements typically incorporate financial covenants that a firm must adhere to to remain compliant, and firms with high leverage may be especially worried about violating these covenants. This apprehension may result in a more pessimistic or defensive disclosure tone. Moreover, Bassyouny et al. [12] and Arslan-Ayaydin et al. [22] argued that when a firm has a lot of debt, it may be hesitant to reveal positive information because it fears it will make creditors more optimistic and predict unachievable financial performance.

Based on the above, the impact of leverage (LEV) on disclosure tone (DT) remains an area of ongoing academic research, so the following hypothesis can be developed:

H3. 

There is a positive impact of leverage on disclosure tone in Saudi Arabia.

2.1.4. Asset Growth (AG)

Managerial reputation theory emphasizes that managers of high-growth firms might have a greater incentive to maintain a positive and transparent tone in their disclosures to build a strong reputation [24,32]. This could increase their ability to create strategic partnerships and get appropriate financing terms [10,26].

In terms of empirical evidence, many studies have emphasized the positive impact of growth rates on disclosure tone. Abou-El-Sood and El-Sayed [14] and Tan and Yeo [33] noted that firms with high growth rates are likely to have a positive outlook toward their future prospects, which can be reflected in their disclosures. Such firms may adopt a more optimistic tone, highlighting their achievements, future opportunities and potential for continued growth. Moreover, Lu et al. [34] and Hossain et al. [35] argued that firms with high growth rates tend to focus on creating long-term value. This approach can drive a more strategic communication plan where disclosures highlight the long-term objectives and investments that will support future growth. To create excitement for the firm’s future, a positive tone can be used to emphasize these long-term plans. From another perspective, there are several reasons why growth rate can lead to a negative impact on disclosure tone. Su et al. [36] and Martikainen et al. [30] noted that as a firm grows, it receives more attention from regulators, investors and analysts. This increased scrutiny can make firms more pessimistic when disclosing information. They may be hesitant to reveal positive information that could potentially expose them to litigation if the information is inaccurate. Moreover, El-Deeb et al. [6] and Jain et al. [37] argued that when a firm experiences rapid growth, it can create a situation where the management team has more information about the firm’s internal situation than external stakeholders. Additionally, this growth can lead to greater complexity, which can make it challenging for investors to fully comprehend the firm’s risks and opportunities. As a result, the firm may use more technical or negative language in its disclosures.

Based on the above, the impact of asset growth (AG) on disclosure tone (DT) remains an area of ongoing academic research. So, the following hypothesis can be developed:

H4. 

There is a positive impact of asset growth on disclosure tone in Saudi Arabia.

2.1.5. Dividend Yield (DIV)

Agency theory highlights that managers might use an excessively optimistic tone to justify a more-than-expected dividend payout [36,38]. This could be an attempt to signal their commitment to shareholder value and a sustainable dividend policy [5,39].

In terms of empirical evidence, many studies have emphasized the positive impact of dividends on disclosure tone. Henry et al. [40] and Kang and Lam [41] noted that dividend payments act as a positive signal to stakeholders that a firm has sufficient cash flow to distribute profits. Firms with a history of consistent dividend payments are essentially disclosing in a positive tone or manner. Moreover, Lu et al. [34] and Li et al. [42] noted that firms may use a positive tone to highlight their strong financial performance, future growth prospects and commitment to rewarding shareholders through consistent dividends. From another perspective, there are several reasons why dividends can lead to a negative impact on disclosure tone. Choi [43] and Hamza et al. [15] noted that firms prioritizing dividend payments feel obligated to maintain consistent payouts, leading to pessimistic or negative disclosures. Moreover, Beretta et al. [44] and Albitar et al. [7] noted that focusing on maintaining consistent dividend payouts can lead firms to prioritize short-term financial performance over long-term sustainability, and investors may perceive a negative tone if the firm prioritizes short-term gains over long-term vision.

Based on the above, the impact of dividend yield (DIV) on disclosure tone (DT) remains an area of ongoing academic research. So, the following hypothesis can be developed:

H5. 

There is a positive impact of dividend yield on disclosure tone in Saudi Arabia.

2.2. Disclosure Tone and Earnings Management

Legitimacy theory emphasizes that firms involved in earnings management might use a more pessimistic or conservative disclosure tone to be compliant with regulations and maintain legitimacy. This could be an attempt to avoid scrutiny and potential sanctions while still achieving their desired financial performance [9,14]. Conversely, signaling theory argues that firms engaging in earnings management practices might use an optimistic disclosure tone to cover negative financial performance. This could be an attempt to signal financial health and mislead stakeholders [15,34].

In terms of empirical evidence, many studies emphasize the positive impact of earnings management on disclosure tone. Abou-El-Sood and El-Sayed [14] and Henry et al. [40] highlighted that earnings management refers to the manipulation of financial reports to provide a more favorable image. While firms might use a positive tone to cover these manipulated practices, an overly optimistic tone can raise a signal for potential earnings management, leading to a loss of investors’ trust. Li et al. [9] and Hamza et al. [15] argued that managers mainly use an optimistic tone while engaging in earnings management practices to mislead stakeholders. Such tactics may result in short-term gains in the share price, which could benefit managers who hold stock options. Lu et al. [34] and Al-Shaer and Zaman [45] found that firms that are involved in earnings management may use a very optimistic tone in their disclosures. This is because they could be concerned about attracting scrutiny from regulators and auditors if the tone is not consistent with earnings volume, which could damage their credibility.

Conversely, there are several reasons why earnings management can lead to a negative impact on disclosure tone in financial reports. Arslan-Ayaydin et al. [22] and Rich et al. [2] argued that the main reason is “credibility” concerns, where earnings management undermines the credibility of financial information. This can result in a more pessimistic tone in disclosures as firms attempt to distance themselves from potential accusations of manipulation. Arena et al. [32] and Tan and Yeo [33] highlighted “Loss of Transparency”: firms that resort to earnings management often engage in practices that cover their true financial image. This can result in ambiguous or pessimistic language in disclosures. Elberry and Hussainey [46] and Pouryousof et al. [28] highlighted the concept of “Fear of Scrutiny”: firms that are aware of their earnings management practices may adopt a defensive or pessimistic tone in their disclosures. This could be an attempt to preempt potential criticism or deflect attention from the manipulated figures.

Despite the amount of research in this area, “earnings management and disclosure tone”, scientific gaps remain, especially in the context of the Saudi environment. So, the following hypothesis can be developed:

H6. 

Earnings management positively influences disclosure tone in Saudi Arabia.

2.3. Disclosure Tone and Accounting Conservatism

Signaling theory argues that firms providing conservative financial reporting practices might want to signal their commitment to transparency. This could potentially result in a more neutral or optimistic tone in their disclosures to enhance this image among stakeholders [3,16]. From another perspective, political cost theory emphasizes that firms working in high-political-risk environments might adopt conservative accounting practices. This could be reflected in a more conservative or pessimistic tone in their disclosures to deal with regulators and minimize political costs [42,47].

In terms of empirical evidence, accounting conservatism has a positive impact on disclosure tone in financial reporting. Ataullah et al. [3] and García Lara et al. [48] highlighted “Transparency and Prudence”: firms that use conservative accounting practices may adopt a more neutral or optimistic tone in their disclosures to maintain stakeholders’ trust. D’Augusta and DeAngelis [16] and Hajawiyah et al. [49] confirmed that conservative accounting practices can be beneficial by reducing the risk of misleading investors. These practices help prevent overly optimistic portrayals of a firm’s performance, which allows for a more honest and objective tone in disclosures. Artiach and Clarkson [50] and Garanina and Kim [51] discovered that firms can align their financial reporting approach with the expectations of their stakeholders through the “Aligning with Stakeholder Expectations” strategy. This strategy is useful for stakeholders, especially those who value a conservative approach to financial reporting.

While accounting conservatism—generally—has a positive impact on disclosure tone, there are potential downsides to consider. García Lara et al. [48] and Aprilia and Rahayu [52] argued that “Unduly Pessimistic”, or conservative, accounting practices can provide an overly cautious image, potentially leading to a pessimistic tone in disclosures. This might not accurately reflect the firm’s true financial performance. Pereira et al. [53] and Alia and AbuSarees [54] highlighted that when a firm adopts a conservatism approach, it may downplay good news and limit its ability to signal strong financial health and prospects to potential investors by using a conservative or pessimistic tone. Cho et al. [55] and Noor et al. [56] argued that in some market situations, investors expect an optimistic tone. A firm that applies a highly conservative approach may be perceived as disconnected from market expectations, which could lead to a negative perception and a decrease in its stock price.

Despite the amount of research in this area, “accounting conservatism and disclosure tone”, scientific gaps remain, especially in the context of the Saudi environment. So, the following hypothesis can be developed:

H7. 

Accounting conservatism positively influences disclosure tone in Saudi Arabia.

3. Method

3.1. The Study Sample

Our research population is represented by Saudi firms (listed) during the period from 2014 to 2022, where the total number of firms in the Saudi stock market is 226. We selected the sample based on three criteria: (a) the firm’s financial reports were available, (b) the firm had not been subject to merger or discontinuation during the study period, and (c) financial and banking sectors’ firms were excluded. Thus, the application of our criteria resulted in the selection of 88 firms to be the study sample, equivalent to 38.9% of the Saudi stock market. Therefore, the study sample includes a total of 792 observations (firm year) (

Table 1. The industry classification of the sample according to the Global Industry Classification Standard (GICS) [57].

GICS Sector Name	Firms	Frequency	Percent
Consumer Discretionary	10	90	11.36
Consumer Staples	14	126	15.91
Health Care	6	54	6.82
Industrials	14	126	15.91
Materials	37	333	42.05
Real Estate	7	63	7.95
Total	88	792	100

).

3.2. Variable Measurement

3.2.1. Disclosure Tone (DT)

Li et al. [9] pointed out that there are many methods to measure the tone of disclosures, and the most common in academic research are linguistic analysis, readability and machine learning. In addition, as the Saudi firms’ annual reports were written in both Arabic and English language, we used content analysis to determine the tone of the firm’s disclosures based on the English version only. Following Bassyouny et al. [12] and Alshabibi et al. [29], we relied on the statements in the Management Discussion and Analysis (MD&A) section as units of our measurement.

We analyzed the tone of disclosure in MD&A by applying MAXQDA 11, a program that counts the frequency of positive and negative words. In addition, we used Loughran and McDonald’s dictionaries to determine positive and negative words (lists contain 2329 negative and 354 positive), which are specifically designed for finance-related language [58], and one of the unique features of our study is that we not only used common words but also word combinations to measure disclosure tone. We also collected other information about Saudi-listed firms through document mining from the “TADAWEL” database. Following Henry and Leone [13], we used the following equation:

$$\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{e} \mathrm{T}\mathrm{o}\mathrm{n}\mathrm{e} (\mathrm{D}\mathrm{T})=\frac{\mathrm{P}\mathrm{O}\mathrm{S}\mathrm{W}-\mathrm{N}\mathrm{E}\mathrm{G}\mathrm{W}}{\mathrm{P}\mathrm{O}\mathrm{S}\mathrm{W}+\mathrm{N}\mathrm{E}\mathrm{G}\mathrm{W}}$$

where POSW is the number of positive words, and NEGW is the number of negative words. If the result is positive (negative), the disclosure tone is optimistic (pessimistic).

3.2.2. Determinants of Disclosure Tone

In the table below, we explain the measurement methods and their related hypotheses for each determinant (

Table 2. The measurement of DT determinants.

Variable	Measurement Method	Hypotheses/Potential Impact
Firm Size	Natural logarithm of total assets	H1/+
Firm Age	Natural logarithm of the number of years after the first financial statement date	H2/+
Leverage	Debt-to-equity ratio, total liabilities/total shareholders’ equity	H3/+
Asset Growth	(Current Total Assets—Previous Total Assets)/Previous Total Assets	H4/+
Dividend Yield	Annual Dividend per share/the current share price	H5/+

).

3.2.3. Earnings Management (EM)

Earnings management is a practice used by managers to provide a more favorable financial perspective than reality. Following Alzoubi [59], we measured earnings management by using discretionary accruals, depending on four steps as presented:

We used the cross-sectional “Modified Jones Model”, generated by Dechow and Sloan [60], where the volume of discretionary accruals (NDAC) for a firm was calculated as the difference between the firm’s total accruals and its non-discretionary accruals, as estimated with Equation (1):

$${\mathrm{N}\mathrm{D}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}={\mathsf{\alpha }}_{\mathrm{j}}(1/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1})+{\mathsf{\beta }}_{1\mathrm{j}}({∆\mathrm{R}\mathrm{E}\mathrm{V}}_{\mathrm{i}\mathrm{j}\mathrm{t}}-{∆\mathrm{R}\mathrm{E}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1})+{\mathsf{\beta }}_{2\mathrm{j}}({\mathrm{P}\mathrm{P}\mathrm{E}}_{\mathrm{i}\mathrm{j}\mathrm{t}}/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}-1})$$

(1)

where ${\mathsf{\alpha }}_{\mathrm{j}}$, ${\mathsf{\beta }}_{1\mathrm{j}}$ and ${\mathsf{\beta }}_{2\mathrm{j}}$ are industry-specific coefficients evaluated using the subsequent cross-sectional regression.

$${\mathrm{T}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1}={\mathsf{\alpha }}_{\mathrm{j}}(1/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1})+{\mathsf{\beta }}_{1\mathrm{j}}({∆\mathrm{R}\mathrm{E}\mathrm{V}}_{\mathrm{i}\mathrm{j}\mathrm{t}}/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1})+{\mathsf{\beta }}_{2\mathrm{j}}({\mathrm{P}\mathrm{P}\mathrm{E}}_{\mathrm{i}\mathrm{j}\mathrm{t}}/{\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}1})+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{j}\mathrm{t}}$$

(2)

where ${\mathrm{T}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$ = total accruals for the firm (i) in the industry (j) in the year (t); ${∆\mathrm{R}\mathrm{E}\mathrm{V}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$ = change in revenue for the firm (i) in the industry (j) between the years (t − 1) and (t); ${\mathrm{P}\mathrm{P}\mathrm{E}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$ = gross property, plant and equipment for the firm (i) in the industry (j) in the year (t); ${\mathrm{T}\mathrm{A}}_{\mathrm{i}\mathrm{j}\mathrm{t}-1}$ = total assets for the firm (i) in the industry (j) at the end of previous year; ${∆\mathrm{R}\mathrm{E}\mathrm{V}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$ = the change in receivables for the firm (i) in the industry (j) between (t − 1) and (t).

Having evaluated NDAC from Equation (1), the discretionary accrual (DAC) amount for the firm (i) in the industry (j) for the year (t) was computed as the residual value from the following Equation (3):

$${\mathrm{D}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}={\mathrm{T}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}-{\mathrm{N}\mathrm{D}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$$

(3)

Then, we used the cash flow method to compute total accruals, where TAC was defined as the difference between net income and operational cash flow, as the following Equation (4):

$${\mathrm{T}\mathrm{A}\mathrm{C}}_{\mathrm{i}\mathrm{j}\mathrm{t}}={\mathrm{N}\mathrm{I}}_{\mathrm{i}\mathrm{j}\mathrm{t}}-{\mathrm{O}\mathrm{C}\mathrm{F}}_{\mathrm{i}\mathrm{j}\mathrm{t}}$$

(4)

From the above equations, we used the absolute value of discretionary accruals as a measure of earnings management, following Dakhlallh et al. [61] and Khanh and Thu [62].

3.2.4. Accounting Conservatism (AC)

In accounting research, different methods have been used to measure the degree of accounting conservatism in financial reports. Measures include market-to-book value, earnings to accruals, accruals to cash flows and earnings to return. Following D’Augusta and DeAngelis [16] and Wang [18], we measured accounting conservatism by using the market-to-book ratio (MTB), which compares the current market value of a firm’s equity to the book value of its equity as reported in the financial statements. Thus, we used the following equation:

$${\mathrm{A}\mathrm{C}}_{\mathrm{i},\mathrm{t}}=\frac{{\mathrm{M}\mathrm{V}}_{\mathrm{i},\mathrm{t}}}{{\mathrm{B}\mathrm{V}}_{\mathrm{i},\mathrm{t}}}$$

(5)

where

• MV_i,t is the market value of firm (i) during period (t).
• BV_i,t is the book value of firm (i) during period (t).

A high MTB ratio (greater than 1) indicates the existence of accounting conservatism practices. This implies that the market value of the firm is considerably higher than its book value. The difference between the two could be attributed to the market’s anticipation of future profitability, which is not entirely reflected in the current book value due to conservative accounting practices.

3.3. Research Model

To test Hypotheses 1 to 5, we applied a linear regression model using panel data. The research model is as follows:

$${\mathrm{D}\mathrm{T}}_{\mathrm{i},\mathrm{t}}=\mathsf{\alpha }0+\mathsf{\beta }1 {\mathrm{F}\mathrm{S}}_{\mathrm{i},\mathrm{t}}+\mathsf{\beta }2 {\mathrm{F}\mathrm{A}}_{\mathrm{i},\mathrm{t}}+\mathsf{\beta }3 {\mathrm{L}\mathrm{E}\mathrm{V}}_{\mathrm{i},\mathrm{t}}+\mathsf{\beta }4 {\mathrm{A}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+\mathsf{\beta }5 {\mathrm{D}\mathrm{I}\mathrm{V}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

where

• DT_i,t is the disclosure tone value of firm (i) during period (t).
• FS_i,t is the size of firm (i) during period (t).
• FA_i,t is the age of firm (i) during period (t).
• LEV_i,t is the leverage of firm (i) during period (t).
• AG_i,t is the asset increase percentage of firm (i) during period (t).
• DIV_i,t is dividend yield of firm (i) during period (t).

To test Hypothesis 6, we applied a linear regression model using panel data. The research model is as follows:

$${\mathrm{D}\mathrm{T}}_{\mathrm{i},\mathrm{t}}=\mathsf{\alpha }0+\mathsf{\beta }1 {\mathrm{A}\mathrm{E}\mathrm{M}}_{\mathrm{i},\mathrm{t}}+\sum {\mathsf{\beta }2-6 \mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l} \mathrm{V}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{l}\mathrm{e}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

where

• DT_i,t is the disclosure tone value of the firm (i) during the period (t).
• AEM_i,t is the earnings management degree of the firm (i) during the period (t).
• Control variables are FS, FA, LEV, AG and DIV of the firm (i) during the period (t).

To test Hypothesis 7, we applied a linear regression model using panel data. The research model is as follows:

$${\mathrm{D}\mathrm{T}}_{\mathrm{i},\mathrm{t}}=\mathsf{\alpha }0+\mathsf{\beta }1 {\mathrm{A}\mathrm{C}}_{\mathrm{i},\mathrm{t}}+\sum {\mathsf{\beta }2-6 \mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l} \mathrm{V}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{l}\mathrm{e}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

where

• DT_i,t is the disclosure tone value of the firm (i) during the period (t).
• AC_i,t is the accounting conservatism degree of the firm (i) during the period (t).
• Control variables are FS, FA, LEV, AG and DIV of the firm (i) during the period (t).

4. Results

4.1. Descriptive Statistics

Table 3. Descriptive statistics.

Variable	Obs	Mean	Std. Dev.	Min	Max
DT	792	0.264	0.415	−0.67	0.98
AEM	789	0.041	0.035	0	0.142
AC	792	3.51	3.299	0.822	12.507
FS	792	21.506	1.312	18.969	24.237
FA	792	14.383	5.792	1	23
Lev	792	0.222	0.186	0	0.763
AG	792	0.019	0.117	−0.179	0.396
Div	792	0.036	0.056	0	0.371

presents the statistical summary for disclosure tone, earnings management, accounting conservatism and firm-specific characteristics in our sample (the Saudi Arabia environment), covering the period from 2014 to 2022. The variables are subjected to winsorization, where the extreme values at the top and bottom 3% are replaced with less extreme values to reduce the impact of outliers.

Our main dependent variable, “disclosure tone”, suggests that Saudi firms use both pessimistic and optimistic narrative disclosures, as evidenced by the negative and positive signs of the minimum (−0.67) and maximum (0.98) values, respectively. However, on average, Saudi firms tend to use optimistic narrative disclosures, as evidenced by the positive mean value (0.264). Concerning accounting conservatism (AC), on the one hand, the minimum value of less than one (0.822) indicates that some managers in Saudi firms may opportunistically use aggressive accounting choices and practices to inflate a firm’s reported value. On the other hand, the maximum value (12.507) suggests that some Saudi firms report highly conservative accounting reports. Accordingly, the standard deviation of AC (3.299) shows a wide dispersion around the mean. Nonetheless, the mean of AC (3.51) indicates that the average of our sample adopts a conservative reporting approach.

Regarding accrual-based earnings management (AEM), the minimum value (zero) reflects that there are some Saudi firms that do not exploit the discretionary loopholes in accruals to manipulate accounting reports. On the contrary, the maximum value (0.142) suggests that the reported values of receivables and revenue accounts may be manipulated within the Saudi context. Relatedly, the mean (0.041) indicates that the average of our sample engages in earnings management. Furthermore, AEM shows a widespread variation (0.035) around the mean (0.041) through time and across firms.

Within our Saudi application context, firm size (FS) shows a standard deviation of 1.312, which is very small relative to the mean (21.506) due to applying the natural logarithm on total assets, which caused smoothing in firm size among firms. Accordingly, firm size shows a small range between the minimum (18.969) and maximum values (24.237), which reflects the high concentration around the mean and the homogeneity in firm size among the sample firms. The firm age (FA) of the sampled Saudi firms ranges from 1 to 23 years. Within a business environment, a range of 22 years may be influential enough to differentiate the disclosure tone of the newly established firms from that of the older and well-established ones. However, the average age of our sample firms is around 14 or 15 years.

Concerning firm leverage (Lev), the minimum value (zero) indicates the absence of debts in the capital structure of some Saudi firms. In contrast, the maximum value (0.763) reflects that other firms depend on debt in their capital structure. Additionally, our sample firms exhibit heterogeneity in their reliance on debt relative to equity for financing their assets and operations, as evidenced by the standard deviation of 0.186, which represents around 84% of the mean (0.222). Asset growth (AG) in Saudi firms exhibits a mixture of negative and positive trends, which is apparent from the presence of both negative and positive signs in the minimum (−0.179) and maximum (0.396) values, respectively. Nevertheless, the average asset growth of Saudi firms displays a positive trend, as indicated by the positive mean value (0.019). Furthermore, asset growth has a standard deviation of 0.117, which represents around 616% of its mean (0.019). Implying that the rate of asset growth varies widely in our sample. Dividend yield (Div) shows a wide range between the non-dividend-paying firms and the high-dividend-paying firms, as evidenced by the minimum (zero) and maximum (0.371) values, respectively. Consequently, dividend yield shows heterogeneous values that are widely spread around the mean in our sample (0.036). Nonetheless, the mean value indicates that Saudi firms distribute dividends at a rate of 3.6% on average.

In a nutshell, the descriptive statistics reveal a relative heterogeneity in our sample. As such, the Saudi stock market includes firms with both pessimistic and optimistic disclosure tones, both manipulating and non-manipulating firms, both conservative and aggressive firms, both newly established and well-established firms, both highly leveraged and unleveraged firms, both positive-growth and negative-growth firms and both high-dividend-paying and non-dividend-paying firms. Furthermore, the lack of homogeneity in our firm-year observations is highlighted when comparing the standard deviation value of each variable against its mean value. Nevertheless, it is worth noting that firm size (FS) stands out as the only homogenous variable in our sample. This homogeneity arises due to the utilization of the natural logarithm, which smoothens the data set.

4.2. Correlation

The correlation matrix provides preliminary insight into the association between the dependent and independent variables. We use Pearson’s correlation coefficients to determine the direction and the strength of the linear association between any two variables included in our research. Moreover, Pearson’s correlation coefficients are used to detect the multicollinearity between any two independent variables included in the same regression model, which can lead to inaccurate estimations.

Table 4. Correlation matrix.

Variables	DT	AEM	AC	FS	FA	Lev	AG	Div
DT	1.000
AEM	0.043	1.000
	(0.230)
AC	0.146 ***	0.117 ***	1.000
	(0.000)	(0.001)
FS	0.031	−0.203 ***	−0.244 ***	1.000
	(0.383)	(0.000)	(0.000)
FA	0.060 *	0.017	−0.003	−0.105 ***	1.000
	(0.090)	(0.643)	(0.940)	(0.003)
Lev	0.014	0.091 **	−0.040	0.357 ***	−0.149 ***	1.000
	(0.690)	(0.011)	(0.257)	(0.000)	(0.000)
AG	0.015	0.032	0.065 *	0.098 ***	−0.015	0.034	1.000
	(0.671)	(0.375)	(0.066)	(0.006)	(0.669)	(0.342)
Div	0.081 **	−0.100 ***	0.107 ***	0.138 ***	−0.090 **	−0.345 ***	0.044	1.000
	(0.022)	(0.005)	(0.003)	(0.000)	(0.011)	(0.000)	(0.214)

*, ** and *** denote significance at the 10%, 5% and 1% levels, respectively.

reports the Pearson’s correlation coefficients of the variables in our models. We note that firm age (FA) and dividend yield (Div) have a significant positive correlation with the disclosure tone, while other firm-specific characteristics show an insignificant correlation with the disclosure tone. Consistent with our expectations, accounting conservatism (AC) has a significant positive association with the tone of narrative disclosures in the management’s discussion and analysis.

Variables that do not have a significant linear correlation with each other can be further tested for curvilinearity because Pearson’s correlation coefficients do not capture the potential nonlinear association between variables. Therefore, we will test the curvilinearity of variables that do not have a significant linear correlation with disclosure tones, such as AEM and AG.

Regarding the multicollinearity between the explanatory variables of each model, the results suggest that there is not a multicollinearity problem among all regressors in the study as the highest correlation coefficient equals 0.357, which is found between Lev and FS.

4.3. Hypothesis Testing

The research models are estimated using the ordinary least square (OLS) and the Dynamic Panel corrected standard error (PCSE) techniques, to serve as robustness tests to each other. The PCSE technique is utilized because of its ability to provide precise standard error estimates, generate an estimate that is clear of autocorrelation and reduce susceptibility to outlier estimates. The PCSE technique is employed for analyzing dynamic heterogeneous panel data.

Table 5. The effect of firm-level characteristics on disclosure tone.

Variable	OLS	PCSE
FS	0.006	0.006
FA	−0.0270 **	−0.0252 *
FA2	0.0010 **	0.0009 *
Lev	0.096	0.106
AG	−0.4183 **	−0.4050 **
AG2	2.5871 ***	2.5415 ***
Div	0.6952 **	0.7778 **
L1.DT	0.4012 ***	0.3184 ***
Year Fixed Effects	Included	Included
Obs	704	704
R2	20.8%	14.8%

***, ** and * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

shows the results of the empirical model presented in Equation (1), which tests Hypotheses 1 to 5. The results presented in Table 3 confirm H5. Consistent with our conjectures, we find that dividend yield (Div.) has a significant positive impact on disclosure tone. This suggests that, within the Saudi context, managers tend to use more positive words and fewer negative words when the firm pays higher dividends. Therefore, larger dividend yields increase optimism in management’s discussion and analysis within the Saudi stock market.

Contrary to our conjectures, both firm size (Size) and firm leverage (Lev) do not have a significant impact on disclosure tone according to the results of both the OLS and PCSE models. Therefore, H1 and H3 are rejected.

H2 is rejected because firm age (FA) has a curvilinear effect on disclosure tone (DT), indicating the existence of an optimal level for FA. This optimal level has implications for the appropriate combination of positive and negative words that managers use in their narrative disclosures. Accordingly, deviations from the optimal level will lead to inefficiency in managers’ narrative disclosure tone. The coefficient of FA is significantly negative (less than zero), while the coefficient of FA squared is significantly positive (greater than zero). As a result, the nonlinear effect of FA can be represented by a U-shaped curve. As such, the textual disclosure properties are expected to change as the firm moves from the introduction to the decline stage of its life cycle. Therefore, a 22-year difference between the youngest and oldest firms in our sample indicates that the time factor is powerful enough to change the disclosure tone of firms operating in the Saudi environment.

Similarly, H4 is rejected because asset growth (AG) has a curvilinear effect on disclosure tone (DT), indicating the existence of an optimal level for AG. This optimal level has implications for the appropriate combination of positive and negative words that managers use in their narrative disclosures. Accordingly, deviations from the optimal level will lead to inefficiency in managers’ narrative disclosure tone. The coefficient of AG is significantly negative (less than zero), while the coefficient of AG squared is significantly positive (greater than zero). As a result, the nonlinear effect of AG can be represented by a U-shaped curve.

The turning points for firm age (FA) and asset growth (AG) are 14 years and 7.47%, respectively. Specifically, during the first 14 years of a firm’s operations in the Saudi market, managers tend to use more pessimistic words than optimistic ones. However, after surpassing 14 years of doing business, managers’ narrative disclosure attitude shifts toward optimism against pessimism. Similarly, for Saudi firms with a growth rate of less than 7.47%, a pessimistic tone is more prevalent than an optimistic one. Nevertheless, when Saudi firms exceed the 7.47% growth rate, an optimistic tone becomes dominant over a pessimistic one.

Finally, based on the results of both the OLS and PCSE models, the lagged values of disclosure tone (DT) have a significant positive impact on the DT of the current year. This means that DT is a dynamic phenomenon, indicating that the current DT value can influence future DT values.

Table 6. The effect of earnings management on disclosure tone.

Variable	OLS	PCSE
AEM	−2.186 *	−2.1951 *
AEM_Sq	19.2491 *	19.3065 *
FS	0.004	0.004
FA	−0.0317 **	−0.0318 **
FA2	0.0011 **	0.0011 **
Lev	0.116	0.115
AG	−0.3266 *	−0.3270 *
AG2	2.3090 ***	2.3114 ***
Div	0.7049 **	0.6982 **
L1.DT	0.3936 ***	0.4009 ***
Year Fixed Effect	Included	Included
Industry Fixed Effect	Included	Included
Obs	701	701
R2	22.1%	22.7%

***, ** and * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

shows the results of the empirical model presented in Equation (2), which tests Hypothesis 6. Against our expectations, the results presented in Table 4 reject H6 according to both the OLS and PCSE models.

Earnings management (AEM) has a curvilinear effect on disclosure tone (DT), indicating the existence of an optimal level for AEM. This optimal level has implications for the appropriate combination of positive and negative words that managers use in their narrative disclosures. Accordingly, deviations from the optimal level will lead to inefficiency in managers’ narrative disclosure tone. The coefficient of AEM is significantly negative (less than zero), while the coefficient of AEM squared is significantly positive (greater than zero). As a result, the nonlinear effect of AEM can be represented by a U-shaped curve.

The turning point for earnings management is at 0.06. More specifically, when the levels of AEM do not exceed 0.06, there is an inverse effect on DT, resulting in a higher frequency of negative words compared to positive words in management’s discussion and analysis. However, when AEM exceeds 0.06, the effect of AEM on DT becomes positive, leading to a higher frequency of positive words compared to negative ones in management’s discussion and analysis (Figure 1).

Table 7. The effect of accounting conservatism on disclosure tone.

Variable	OLS	PCSE
AC	0.0102 **	0.0119 **
FS	0.015	0.016
FA	−0.0261 *	−0.024
FA2	0.0009 **	0.0009 *
Lev	0.065	0.070
AG	−0.4422 **	−0.4263 **
AG2	2.5641 ***	2.5206 ***
Div	0.5586 *	0.6216 *
L1.DT	0.3926 ***	0.3115 ***
Year Fixed Effect	Included	Included
Obs	704	704
R2	21.4%	15.6%

***, ** and * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

shows the results of the empirical model presented in Equation (3), which tests Hypothesis 7. Consistent with our expectation, within the Saudi environment, accounting conservatism (AC) has a significant positive effect on disclosure tone (DT), indicating that conservative financial reporting leads to higher optimism and lower pessimism in management’s discussion and analysis. Therefore, the results of both OLS and PCSE models support H7.

5. Discussion

Our research was divided into three main questions: Can operational determinants regulate the disclosure tone? What is the impact of earnings management on disclosure tone? What is the impact of accounting conservatism on disclosure tone? To answer these questions, we conducted an empirical study by using Saudi Arabia’s business environment as a scope of our evidence.

In terms of the first part, “Can operational determinants regulate the disclosure tone?”, this question was our main research gap, especially in Saudi Arabia as an emerging stock market. So, we were trying in this research to “examine the voice of the reports”. Moreover, financial disclosures, once perceived as technical documents, are increasingly recognized as a source of valuable information beyond just the numbers. The language used in these reports, known as disclosure tone, can offer valuable insights into a firm’s financial health and potential red flags. The paper goes over scientific studies analyzing the primary determinants of disclosure tone in financial reports. Going further, the existing academic research presents mixed empirical evidence of how operating determinants impact disclosure tone [2,11,17]. In our study, we fill this gap with empirical evidence from the Saudi business environment, where we found a positive impact of dividend yield on disclosure tone, as well as a curvilinear impact of firm age and asset growth on disclosure tone in firms’ financial reporting. In addition, both firm size and leverage do not have a significant impact on disclosure tone.

In terms of the second part, “What is the impact of earnings management practices on disclosure tone?”, the realm of financial reporting is constantly scrutinizing the relationship between a firm’s financial health and the language used to communicate it. One intriguing area of research explores the impact of earnings management on disclosure tone. Earnings management, the act of manipulating financial statements to present a desired picture, raises concerns about transparency and potential manipulation of investor perceptions. This discussion delves into the scientific research surrounding this complex relationship. The existing academic research presents a mixed picture of how earnings management impacts disclosure tone. Some academic studies find a positive correlation between positive disclosure tone and abnormal stock returns, suggesting investors might react favorably to optimistic language even if it masks underlying issues [9,34]. Moreover, other academic studies highlight how a positive or optimistic tone, especially when inconsistent with actual financial performance, can be a red flag for potential earnings management [15]. In our study, we fill this gap with empirical evidence from the Saudi business environment, where we found a curvilinear impact of earnings management practices on disclosure tone in firms’ financial reporting.

In terms of the third part, “What is the impact of accounting conservatism on disclosure tone?”, financial reporting relies on a delicate balance between transparency and prudence. Accounting conservatism, the practice of anticipating and recognizing potential losses but delaying recognition of gains, embodies this principle. Scientific research is increasingly focused on how accounting conservatism influences the language used in financial disclosures, known as disclosure tone. This discussion explores the scientific research surrounding this intriguing interplay. Going further, empirical evidence in this area presents a mixed landscape of how accounting conservatism impacts disclosure tone [3,16,18]. Some academic studies suggest a negative correlation between positive disclosure tone and accounting conservatism. Firms with more conservative accounting practices tend to use a more neutral or even conservative tone in their disclosures. Other academic studies highlight that accounting conservatism can constrain upward tone management. This suggests that conservative accounting practices might limit managers’ ability to use excessively optimistic language in their disclosures. In our study, we fill this gap with empirical evidence from the Saudi business environment, where we found a positive impact of accounting conservatism on disclosure tone in firms’ financial reporting.

6. Conclusions

The investigation into the level and determinants of disclosure tone, on the one hand, and the impact of earnings management and accounting conservatism on disclosure tone, on the other, provided insights into the intricate dynamics in the relationship between managers and stakeholders. As firms confront an ever-evolving disclosure requirement landscape, understanding the implications of disclosure tone is crucial for sustaining stakeholder trust.

Saudi Arabia is considered an ideal environment for conducting objective research tests since it has a promising financial market with economic consistency and stability, which supports the credibility of the results and their ability to be generalized in the future. Thus, after conducting an applied study from 2014 to 2022, the paper provided evidence that the disclosure tone in the Saudi business environment is mainly positive (optimistic) and argued that the reason for this tone is the financial stability of the Saudi financial market. In the same manner, the operational characteristics have mixed relationships with disclosure tone: there is a positive relationship with dividend yield, while there is no relationship with leverage. Moreover, the paper provided evidence for the curvilinear impact of earnings management practices on disclosure tone in the Saudi business environment. At the same time, there is a positive impact of accounting conservatism on disclosure tone, and we argue that conservatism serves as a guardrail against managers’ manipulation, maintaining a balanced portrayal of a firm’s financial performance.

Finally, our research on disclosure tone in Saudi Arabia can contribute to expanding the body of knowledge on this academic area in the Arab countries’ context which is characterized by a scarcity of applied evidence. Moreover, this paper identifies many challenges that warrant further academic research. The determinants of disclosure tone in each business environment remain a major gap, as the nature of this tone can directly be a reason for capital movement. Thus, future research avenues include investigating the impact of the board of directors’ characteristics, corporate governance mechanisms, political connections, ESG practices and capital structure on disclosure tone.