The primary objective of this study is to examine the correlation between financial statement comparability and conditional and unconditional conservatism within companies listed on the Tehran Stock Exchange (TSE).
To achieve this, a sample of 193 companies, resulting in 1,546 firm-year observations, were listed on the TSE between 2014 and 2021. The study’s research hypotheses are assessed by applying multiple regression models.
The findings reveal a notable positive association between financial statement comparability and conditional conservatism. Additionally, the research results indicate a significant and negative connection between financial statement comparability and unconditional conservatism.
According to the findings, corporate managers may prioritize financial statement comparability to enhance conditional accounting conservatism, which might be translated as a suitable benchmark for competitors. Equity owners may decrease the agency problems associated with CEOs by emphasizing comparable financial reports, as it improves the quality of financial figures and facilitates stakeholders’ evaluation and comparison of various companies’ performances.
A review of the relevant literature underscores the absence of research focusing on the relationship between financial statement comparability and conditional and unconditional conservatism within emerging markets. Consequently, this study aims to address this gap by investigating this relationship in the context of emerging markets and contributing to the existing body of literature in this field.
1. Introduction
Accounting conservatism remains a contentious subject of debate in academic and policy-making circles. Despite the International Accounting Standards Board (IASB) and the Financial Accounting Standards Board (FASB) increasingly emphasizing information neutrality, some policymakers underscore the continuing significance of conservatism, primarily driven by concerns stemming from the recent financial crisis. Given conservatism’s vital role in accounting, this study examines the connection between financial statement comparability and different forms of conservatism, conditional and unconditional. Although conservatism has far-reaching implications for the book value of net assets and net income, accountants frequently concentrate on specific manifestations of conservatism. For instance, they may opt for accelerated depreciation methods over the straight-line depreciation method, choose to capitalize on research and development costs rather than recognize them as expenses, or adopt the lower-of-cost or market (LCM) method for inventory valuation, all of which are elective conservative accounting practices. However, empirical research typically employs a broader conservatism framework encompassing the cumulative impact of managers' accounting choices, incorporating factors such as information asymmetry, underlying assumptions, and estimates (Ahmed and Duellman, 2007).
Conservatism can be classified into two distinct categories: conditional and unconditional. Conditional conservatism arises from asymmetric verifiability requirements, leading to the more prompt recognition of losses compared to gains. As LaFond and Watts (2008) posit, conditional conservatism is essential in information, particularly in uncertain circumstances. Firstly, it is a governance mechanism that curbs managerial incentives and abilities to manipulate accounting income. Secondly, conservative accounting measures current corporate performance and complements other information sources by offering reliable insights. While the concept of financial statements influencing other information sources is not new, it has received limited empirical attention, and there is scant evidence regarding companies' commitment to varying levels of information disclosure in private contexts. Building upon prior research (such as Kim et al., 2013; Choi et al., 2019; Habib et al., 2020; Kim et al., 2021), it can be inferred that conservatism can enhance the comparability of financial statements. Comparability, a crucial qualitative characteristic of financial information, empowers users to make more informed and rational decisions (FASB, 2010). According to FASB Statement No. 8, improving financial statement comparability is one of the primary drivers behind financial reporting standards. Furthermore, per the Iranian financial reporting conceptual framework (2014), even if the information is reliable and data-driven, its utility remains questionable without understandability and comparability (Mehrvarz and Marfou, 2016). To gain a comprehensive grasp of distinctions and similarities among investment opportunities, FASB (2010) underscores the significance of accounting information comparability, emphasizing its necessity for informed decision-making. Accounting information comparability is paramount for external stakeholders, including investors, analysts, and creditors (Kim et al., 2013; Choi et al., 2019) and internal users within a company (Chen et al., 2018). De Franco, Kothari and Verdi (2011) suggest that the comparability of accounting information is contingent upon the proximity of two companies' accounting systems and can significantly influence the companies' accounting practices, including conservatism.
Conservatism and comparability stand as prominent attributes in financial reporting. While they may not be explicit accounting principles, they wield considerable influence over the valuation of assets and the determination of net income. Comparability of financial statements, in particular, empowers investors to discern distinctions and commonalities among the accounting information of various companies. Therefore, one of the highly sought-after qualities of financial statements is information comparability, which not only reduces the cost of information acquisition but also enhances the informational landscape of the company (De Franco, Kothari and Verdi, 2011). Consequently, it facilitates the efficient allocation of capital and bolsters investor confidence. Recognizing the significance of accounting comparability, a multitude of studies have documented its favorable impact on various outcomes (Chauhan and Kumar, 2019).
Considering the theoretical framework, we anticipate that financial statement comparability affects accounting conservatism. A comprehensive review of prior research encompassing comparability and conservatism reveals a body of work examining conservatism, comparability, corporate performance, and information quality (Daryaei et al., 2022); conservatism about executive compensation (Li et al., 2020); conservatism’s impact on earnings quality (Nassir Zadeh, Askarany and Arefi Asl, 2022); comparability and its influence on managers' use of corporate resources (Kim et al., 2021); comparability and its connection to liquidity risk (Kim et al., 2016); comparability and stock return volatility (Habib et al., 2020). Surprisingly, none of these studies have delved into the association between financial statement comparability and conservatism, whether conditional or unconditional (Chauhan et al., 2022). Thus, this current study endeavors to bridge this research gap and contribute to the pertinent literature. As such, the primary objective of this study is to explore the relationship between financial statement comparability and both conditional and unconditional conservatism. This paper aims to address the fundamental question of whether a noteworthy and statistically significant correlation exists between comparability and conditional conservatism.
The findings of this paper are important for corporate managers. They are aware that prioritizing financial statement comparability-a pivotal qualitative characteristic crucial for decision-makers relying on financial statements-is likely to enhance conditional accounting conservatism, which might be translated as a suitable benchmark for competitors. Additionally, given the identified negative and significant relationship between financial statement comparability and unconditional conservatism, equity owners may decrease the agency problems associated with CEOs by emphasizing comparable financial reports since comparability of financial reporting can improve the quality of financial figures and facilitate stakeholders' evaluation and comparison of various companies' performances. Our findings also have the potential for policymakers in that implementing standardized accounting procedures may enhance comparability and transparency, impacting how conservatively reporting finances is conducted.
2. Theoretical framework and hypothesis development
2.1 Theoretical underpinnings
According to the proposed theory by the literature, agency theory, developed by Jensen and Meckling (1679), has the strongest power to explain the impact of financial reporting comparability on accounting conservatism. High quality accounting figures will reduce the managerial ability to inflate earnings or mask poor performance, leading to more comparability. As a result of comparable financial information, greater accounting conservatism is expected.
The next applicable theory in this study is the information quality theory proposed by Wang and Strong (1996). Generally, it has been observed that higher comparable information is related to higher information quality. When firms practice more comparability, real information is more readily recognized, and gains have to meet more stringent verification standards. Conservatism can be achieved due to more reliable and comparable financial statements across firms.
Finally, the components of regulatory environment theory may underscore the relationship between financial reporting comparability and accounting conservatism. Based on this theory, the outcome of financial statements may be impacted by the regulatory environment, which concerns accounting standards and disclosure requirements. More conservative accounting and higher comparability are related to better regulatory environments. This suggests stronger regulatory environments are related to more conservative accounting and higher comparability. On the one hand, the positive theories, emphasized by Levi-Faur (2011) and Jordana and Levi-Faur (2004), address the requirements for regulations, including market failures, controlling information asymmetry, and meeting special interests of different groups. On the other hand, normative theories deal with how regulations should be designed to work optimally; this theory mainly concentrates on efficiency, fairness, and minimum burdens on business (Stigler, 1971).
2.2 Existing literature
The two related accounting concepts are conservatism and financial statement comparability. Conservatism may offset managerial optimism, overstatement of earnings, and resultant overvaluation; it also enhances earnings quality, reduces manipulation, and promotes financial comparability. According to Daryaei et al. (2022), verifiability or, more importantly, asymmetric verifiability does play a significant role in the recognition of revenues and expenses and therefore leads to the high quality of financial reporting. For example, Hsieh et al. (2019) considered conservative accounting to be a reaction to ambiguity and ambiguous choice that is conservative favors unfavorable outcomes. Their results confirm that the company experiencing higher ambiguity issues and more conservative reports. Hansen et al. (2018) focused specifically on the corporate life cycle and found that unconditional conservatism decreases across life cycle stages but conditional conservatism does not.
Comparability of the financial statements is considered one of the major determinants of financial reporting quality by an accounting standard-setting body like FASB. Kim et al. (2013) insist that with higher comparability, there is less asymmetry in information; thus, the ability to detect financial fraud and misstatement is higher. Those firms that have low comparability tend to show more fraudulent characteristics than other firms do. Comparability enables users to gauge the performance of entities against their peers; hence, these would provide greater visibility of differences. Therefore, comparability, as mentioned by Choi et al. (2019), may be inversely related to unconservative reporting. These observations are also shared in a study by Zhang (2008) and Blanco and Dhole (2017), who observe the relation of comparability and unconservative reporting to be negative. The latter adds that with lower comparability, the risk of fraud for an entity also heightens. This paper tries to examine empirically the relationship between financial statement comparability and financial reporting quality.
2.3 Hypothesis development
On the one hand, conditional accounting conservatism is often called “news-dependent” conservatism as it entails realizing profits later than losses. This conservatism is reactive, contingent on the emergence of particular news or events that impact the company’s financial situation. In other words, it ensures that negative news appears in the financial statements faster than positive news, which may encourage more circumspect and careful financial reporting. On the other hand, unconditional accounting conservatism, known as “balance sheet” conservatism, entails consistently understating net assets and earnings, even in the absence of particular circumstances. This proactive conservatism is ingrained in the company’s accounting procedures and standards. Over time, it consistently undervalues assets and incomes, acting as a safety net against unforeseen events. A perspective on conservatism posits that adopting conservative accounting policies heightens the likelihood of income underreporting. Furthermore, investors perceive conservative management policies as valuable insider information, owing to the informative nature of conservative accounting policies and their role in reducing investors' uncertainty regarding the company’s economic prospects (Gietzmann and Trombetta, 2003).
2.3.1 Comparability and conditional accounting conservatism
The two key factors that help in increasing the accuracy of financial reporting are accounting conservatism and financial statement comparability. Comparability increases transparency; hence, stakeholders would be able to benchmark companies more effectively. Firms, knowing that comparability may make them scrutinized for their misrepresentations in financial reporting, use more conservative accounting in order to minimize comparability-related risks. Arianpoor and Efazati (2024) illustrated that firms with comparable statements are perceived as highly transparent, which motivates conservative behavior. Also, board independence reinforces the relationship between comparability and CEO compensation, whereas Ramineh (2024) proves that comparability and conservatism have a positive effect on profit reaction coefficients.
Another channel through which comparability might impact conservatism is through the reduction in information asymmetry. The sharper the financial comparisons become, the lesser the extent to which managers can manipulate information; thus, comparability supports conservative accounting. Hashemi Dehchi et al. (2020) have found that comparability increases the relevance of financial information especially for low information asymmetry. Islam et al. (2023) find evidence that greater comparability even decreases the possibility of financial vulnerability as firms with poor comparability are likely to create greater financial vulnerabilities. Moreover, it is also argued that comparability promotes conservative accounting by reducing earnings management. As comparability has a better chance of uncovering financial manipulations, it pushes firms toward prudence in their accounting practices. Comparability reduces earnings manipulations that force CEOs to be more conservative. Syahputri and Nawirah (2023) suggests that conservatism decreases inefficiency and restricts the opportunity to manipulate earnings.
Decision-making increases constitute another result developed from comparability, whereby standardized financial information contributes to better decisions by stakeholders. Therefore, such expectation encourages the application of conservative accounting by firms. Rustiarini et al. (2021) indicated that board size and diversity positively contribute to conservatism while overconfidence reduces conservatism. Kim (2023) found that business firms with similar segments have increased comparability hence inducing conservative accounting practices. Comparability thus reinforces conservatism by facilitating more transparency and less asymmetry in information, resulting in better decision-making and less risk of financial manipulations. Such results make for better and more consistent financial reporting, which is helpful for investors and managers.
According to the provided discussions, we expect that improving accounting comparability is likely to enhance conditional accounting conservatism. To assess this expectation, the first hypothesis is developed:
Financial statement comparability positively and significantly impacts conditional accounting conservatism.
2.3.2 Comparability and unconditional accounting conservatism
Conditional conservatism is a way of improving the quality of accounting information, especially in contractual settings. That is, firms with high-quality disclosure practices are more likely to provide reliable information about losses and adverse events than firms with poor quality disclosure, which may use conditional conservatism as a strategy to achieve the goals set by the management. On the other hand, unconditional conservatism has negative implications on the relevance of the financial reports through the symmetrical undervaluation of assets and income, and through building reserves that enable earnings management. Still, this reduces the decision usefulness of the financial reports according to Kaldirim (2019).
Prior research suggests that unconditional conservatism is associated with stock returns. Kale and Villupuram (2024) identify that investors' risk aversion determines the pricing of conservatism, where the more focused downside risk is related to unpredictability in company operations. Not all other studies find a significant relationship between unconditional conservatism and product market competition (Tavakoli and Daemi, 2024; Salehi et al., 2020a, b, c). The general view seems to be that unconditional conservatism increases the opacity of financial reports (Khalaf and Hussein, 2023).
Unconditional conservatism also reduces financial report readability, a proxy for the financial reports' interpretational difficulty for its stakeholders (Kian et al., 2021). Busy directors, especially female directors, are related to unconditional conservatism; thus, they can potentially reduce financial report quality (Le et al., 2023). However, unconditional conservatism may be said to have improved reliability if it is at a moderate level (Sebrina and Taqwa, 2019). Avabruth et al. (2024) report that share pledging by dominant owners is positively associated with conditional and negatively with unconditional conservatism, more so in emerging markets and particularly for family businesses.
Comparability of financial statements is expected to reduce unconditional conservatism, as it improves the quality of financial reporting. Greater comparability promotes more transparency and allows for better monitoring of CEO performance. Therefore, unconditional conservatism would have less room to operate as an earnings manipulative technique. For example, Daryaei et al. (2022) find that by enhancing comparability, firms improve the linking of executive compensation to firm performance by structuring incentives for CEOs in a manner that aligns their interests with those of the shareholders.
Finally, referring to the specification of regulatory environment theory, unconditional conservatism may be affected by adopting international financial reporting standards and other regulatory measures to enhance comparability. For example, Iwasaki et al. (2018) dispute that adopting IFRS may lead to increased comparability, reducing the cost of equity and demand for unconditional conservative accounting; firms are less prone to rely on conservative accounting practices to convey information to investors about financial stability. Concerning provided discussions, the second hypothesis is conducted as follows:
Financial statement comparability negatively and significantly impacts unconditional accounting conservatism.
3. Methodology
The current study is characterized by its applied research purpose and employs a descriptive-correlational research method. The research design is semi-empirical and adopts a post-event approach by utilizing past datasets. The post-event approach is chosen when the researcher examines the subject after the events have transpired, and there is no opportunity to manipulate independent variables.
3.1 Sample and data collection procedure
The statistical population for this research encompasses all companies listed on the TSE from 2014 to 2021. Such a period is investigated for the following reasons: (1) several economic events and activities require high-quality financial reporting quality, which might be reflected in financial reporting comparability. For example 2014, Iran and the global powers struck a framework agreement on the Iran nuclear deal. Following that, the Joint Comprehensive Plan of Action (JCPOA), which lifted certain sanctions on Iran in exchange for limits on its nuclear program, was formally signed in 2015. Such an agreement was supposed to remove some of the boundaries of expanding foreign direct investment in Iran, which could be operationalized through firm-level businesses. Accordingly, improved financial reporting quality may provide a clearer picture of Iran’s financial markets' current condition, and (2) data availability is another reason that encouraged us to conduct our investigation in such a period; the results are preseted in Table 1.
The sample selection criteria
| Description | Excluded companies | The total number of companies |
|---|---|---|
| Total number of the companies listed on the TSE | 474 | |
| Financial intermediaries, financial supply, insurance and investment companies | 88 | |
| Companies whose shares did not trade on the TSE for more than 6 months | 114 | |
| Companies that have been listed on the TSE during the research period | 24 | |
| Companies whose required information is unavailable | 55 | |
| Statistical sample | 193 |
Source(s): Created by authors
Employing a systematic elimination sampling (screening) method, a sample of 193 companies from all listed firms on TSE are selected based on the following criteria:
- (1)
Availability of audited financial information for the companies.
- (2)
Active listing on the TSE throughout the research period.
- (3)
Exclusion of banks and financial institutions, including investment companies, financial intermediaries, and holdings.
3.2 Research variables measurement
The operational definition of all variables is provided in Appendix 1.
3.3 Data analysis techniques and model specification
The following two regression models test the research hypotheses using the Ordinary Last Square (OLS) statistical approach. Model 1 is used to test the first hypothesis, which is as follows:
Model 1
Model 2 is used to test the second hypothesis, which is as follows:
Model 2
4. Data analysis
4.1 Descriptive statistics
Descriptive statistics encompass summarizing, organizing, and presenting a dataset and computing key measures like mode, mean, and median, which provide insights into various population characteristics. Table 2 provides descriptive statistics for the research variables, including minimum and maximum values, mean, median, standard deviation, skewness, and kurtosis. Additionally, Table 3 displays the frequency distribution for the study’s only discrete variable, “Busy.”
Descriptive statistics for the research variables
| Variable symbol | Variable name | Mean | Median | Standard deviation | Skewness | Kurtosis | Min | Max |
|---|---|---|---|---|---|---|---|---|
| CFO | Operating cash flow | 0.198 | 0.077 | 1.337 | 14.779 | 234.161 | −2.140 | 22.700 |
| COMPACC | comparability | −0.376 | −0.375 | 0.005 | −1.741 | 9.153 | −0.400 | −0.353 |
| LEV | Financial leverage | 0.552 | 0.546 | 0.234 | 0.827 | 7.212 | 0.000 | 2.364 |
| MTB | Market-book value | 4.443 | 2.690 | 6.475 | 3.694 | 24.986 | −23.400 | 65.200 |
| NI | Net income | 0.137 | 0.074 | 0.564 | 8.674 | 83.144 | −1.650 | 5.660 |
| ROA | Return on assets | 0.125 | 0.099 | 0.153 | 0.541 | 4.594 | −0.581 | 0.902 |
| SALE | Sales growth | 0.615 | 0.258 | 1.898 | 6.624 | 61.145 | −1.000 | 25.549 |
| SIZE | Firm size | 14.724 | 14.551 | 1.628 | 0.612 | 3.999 | 8.915 | 20.821 |
Source(s): Created by authors
Frequency of the variable fiscal year-end
| Title | Frequency | Frequency percentage |
|---|---|---|
| Yes | 1,485 | 96.200 |
| No | 59 | 3.800 |
| Sum | 1,544 | 100.000 |
Source(s): Created by authors
In Figure 1, a histogram is depicted alongside a density curve. Above the diagram, the Pearson correlation between pairs of variables is displayed, while below it, a distribution chart for pairs of variables is presented. Based on the visual representation in the figure, it can be concluded that the research-independent variables are not interrelated, and it is safe to say that there is no collinearity among the research variables.
4.2 Panel regression modeling
Prior to hypothesis testing, an assessment is made to ascertain whether the models are fixed or random effects, and the findings are briefly summarized in the table below.
The F Limmer test determines the choice between pooled and panel data methods. The probability levels associated with the research models reveal that the null hypothesis advocating using the pooled data method is rejected in all instances. Meanwhile, the Hausman test selects between a panel with random effects and a panel with fixed effects. When the null hypothesis is rejected, the preferred model is a panel with fixed effects. The outcomes of both the F Limmer and Hausman tests are detailed in Table 4.
Determining the fixed or random effects of the research models
| Number | Dependent variable | F Limmer test | Hausman test | Result | ||
|---|---|---|---|---|---|---|
| Statistic | p-value | Statistic | p-value | |||
| Model 1 | NI | 457.648 | <0.001 | 5.584 | 0.589 | Panel with random effects |
| Model 2 | MTB | 484.885 | <0.001 | 114.619 | <0.001 | Panel with fixed effects |
| Model 3 | NI (with a time lag) | 487.266 | <0.001 | 11.604 | 0.114 | Panel with random effects |
| Model 4 | MTB (with a time lag) | 530.479 | <0.001 | 9.068 | 0.248 | Panel with random effects |
Source(s): Created by authors
4.2.1 Model 1: the dependent variable (NI)
Table 5 shows that the p-value is less than the 0.05 probability level (p < 0.001), which means that the model itself is statistically significant. R square shows, the coefficient of determination value, which accounts for 0.281, implying that 28% of dependent variable variations are explained by the independent ones. Furthermore, financial statements comparability (COMPACC) is less than the 5% level and thus confirms the hypothesis that comparability enhances conditional accounting conservatism. When companies have comparable financial statements, their reporting is more conservative, in a way consistent with prior literature arguing that transparency and benchmarking promote conservatism because they facilitate the comparison of firms by various stakeholders. Comparability reduces information asymmetry (Hashemi Dehchi et al., 2020), thereby decreasing managers' incentives and opportunities to manipulate financial information, and promoting more conservative accounting. Comparability also facilitates investors' ability to compare firm performance with competitors and industry peers, which promotes informed investment decisions and capital allocation that foster conservatism due to the desire to avoid adverse market reactions (Kapellas and Siougle, 2017).
The summary of the regression coefficients of model 1
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | 2.305 | 1.104 | 2.087 | 0.037 | |
| COMPACC | 7.995 | 2.926 | 2.732 | 0.006 | |
| SIZE | 0.073 | 0.020 | 3.621 | 0.000 | |
| ROA | 0.460 | 0.134 | 3.440 | 0.001 | |
| LEV | 0.001 | 0.094 | 0.015 | 0.988 | |
| SALE | −0.004 | 0.008 | −0.542 | 0.588 | |
| CFO | 0.015 | 0.012 | 1.205 | 0.228 | |
| BUSY | −0.308 | 0.079 | −3.905 | 0.000 | |
| Coefficient of determination | Adjusted coefficient of determination | Durbin-Watson statistic | F-statistic | p-value | |
| 0.281 | 0.175 | 0.912 | 2.645 | <0.001 | |
Source(s): Created by authors
Further analyses also show that among control variables, SIZE and ROA have a positive and significant impact on the level of conditional accounting conservatism due to the coefficients of 0.073 and 0.460 and p-values of 0.000 and 0.001, respectively. However, other control variables, including LEV, SALE, CFO, and BUSY, show no statistically significant impact on adopting conditionally conservative approaches.
4.2.2 Model 2: the dependent variable (MTB)
Table 6 also shows a p-value below 0.05 (p < 0.001), confirming the regression model’s statistical significance. The coefficient of determination is 0.355, meaning 35% of the variance in the dependent variable is explained by the independent variables. The negative coefficient (−168.058) and p-value (0.000) for COMPACC confirm a significant negative relationship between financial statement comparability and unconditional conservatism, supporting the second hypothesis. Meaning, that companies with highly comparable financial statements are less likely to use unconditional accounting conservatism, aligning with expectations that comparability mitigates unconditional conservative practices like systematic underestimation of assets and income. This transparency reduces management opportunism and limits their ability to obscure financial information. Comparability may also be linked to more performance-based CEO compensation, reducing incentives to manipulate accounting figures through unconditional conservatism. Thus, enhanced financial reporting comparability can curb the application of unconditional conservatism (Kaldirim, 2019; Khalaf and Hussein, 2023; Daryaei et al., 2022).
The summary of the regression coefficients of model 2
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | −55.989 | 14.853 | −3.770 | 0.000 | |
| COMPACC | −168.058 | 39.583 | −4.246 | 0.000 | |
| SIZE | −0.529 | 0.102 | −5.210 | 0.000 | |
| ROA | 4.446 | 1.187 | 3.747 | 0.000 | |
| LEV | 2.355 | 0.747 | 3.153 | 0.002 | |
| SALE | 0.155 | 0.080 | 1.936 | 0.053 | |
| CFO | −0.149 | 0.115 | −1.294 | 0.196 | |
| BUSY | 3.210 | 0.854 | 3.757 | 0.000 | |
| Coefficient of determination | Adjusted coefficient of determination | Durbin-Watson statistic | F-statistic | p-value | |
| 0.355 | 0.255 | 2.327 | 3.576 | <0.001 | |
Source(s): Created by authors
Additionally, ROA, LEV, SALE, and BUSY are among the control variables positively impacting the unconditional accounting conservatism, resulting from the p-values of 0.000, 0.002, 0.053, and 0.000, respectively.
4.3 The robustness of the results
The effect of independent variables in year t on dependent variables in year t+1 is examined to check for the robustness of the results.
4.3.1 Model 1: the dependent variable (NI) (with a time lag)
As illustrated in Table 7, the p-value registers below 0.05 (p < 0.001), establishing the significance of the regression model. The coefficient of determination, which gauges the relationship between the independent and dependent variables while accounting for control variables, is computed as 0.310. This indicates that approximately 31% of the variance in the dependent variable can be attributed to the independent variables. Furthermore, Table 7 tests the significance of the independent variable’s effect or the regression coefficients. In the case of financial statement comparability, considering its associated p-value, which is less than 0.05, the hypothesis regarding the relationship between the research variables is validated. This affirms a significant positive relationship between financial statement comparability and conditional accounting comparability while accounting for a year lag. These findings mean that preparing comparable financial statements may improve the informational environment of companies in subsequent years in terms of adopting conditional accounting conservatism. Thus, our findings are still robust, considering the one-year lag.
The summary of the regression coefficients of model 1 (with a time lag)
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | −5.642 | 1.330 | −4.243 | 0.000 | |
| COMPACC | 10.889 | 3.285 | 3.315 | 0.001 | |
| SIZE | 0.092 | 0.030 | 3.071 | 0.002 | |
| ROA | 0.320 | 0.158 | 2.028 | 0.043 | |
| LEV | 0.194 | 0.113 | 1.724 | 0.085 | |
| SALE | 0.007 | 0.008 | 0.891 | 0.373 | |
| CFO | 0.035 | 0.019 | 1.836 | 0.067 | |
| BUSY | 0.184 | 0.567 | 0.325 | 0.745 | |
| Coefficient of determination | Adjusted coefficient of determination | Durbin-Watson statistic | F-statistic | p-value | |
| 0.310 | 0.191 | 0.975 | 2.606 | <0.001 | |
Source(s): Created by authors
4.3.2 Model 2. the dependent variable (MTB) (with a time lag)
As depicted in Table 8, the p-value falls below 0.05 (p < 0.001), establishing the significance of the regression model. The coefficient of determination, which assesses the relationship between the independent and dependent variables while controlling for other variables, is computed at 0.457. This indicates that the independent variables can account for approximately 46% of the variance in the dependent variable. Furthermore, within Table 8, the significance of the independent variable’s effect, as reflected in the regression coefficients, is rigorously tested. Considering its associated p-value, which is less than 0.05, the hypothesis concerning the relationship between the research variables is upheld in the case of financial statement comparability. This confirms a negative and significant relationship between financial statement comparability and unconditional accounting comparability while accounting for a year lag. These findings mean that preparing comparable financial statements may improve the informational environment of companies in subsequent years in terms of avoiding unconditional accounting conservatism. Thus, our findings are still robust, considering the one-year lag.
The summary of the regression coefficients of model 2 (with a time lag)
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | −77.028 | 14.814 | −5.200 | 0.000 | |
| COMPACC | −229.626 | 38.985 | −5.890 | 0.000 | |
| SIZE | −0.632 | 0.121 | −5.226 | 0.000 | |
| ROA | 2.357 | 1.467 | 1.606 | 0.109 | |
| LEV | 2.883 | 0.948 | 3.042 | 0.002 | |
| SALE | 0.105 | 0.096 | 1.095 | 0.274 | |
| CFO | −0.206 | 0.213 | −0.968 | 0.334 | |
| BUSY | 2.561 | 2.783 | 0.920 | 0.358 | |
| Coefficient of determination | Adjusted coefficient of determination | Durbin-Watson statistic | F-statistic | p-value | |
| 0.457 | 0.407 | 1.613 | 9.188 | <0.001 | |
Source(s): Created by authors
4.4 Additional analyses
In this section, we conduct an analysis utilizing robust regression and M-estimators. In robust regression, apart from addressing errors, the optimization of regression coefficients is also carried out. This optimization helps to constrain the regression coefficients, preventing them from taking exceedingly large values.
4.4.1 The dependent variable (NI) (the robust regression)
As revealed in Table 9, the p-value falls below 0.05 (p < 0.001), affirming the significance of the regression model. The coefficient of determination, which measures the relationship between the independent and dependent variables while considering control variables, is determined to be 0.379. This indicates that the independent variables can explain approximately 38% of the variance in the dependent variable. Furthermore, within Table 9, the significance of the independent variable’s effect, as reflected in the regression coefficients, is rigorously tested. Considering its associated p-value, which is less than 0.05, the hypothesis regarding the relationship between the research variables is confirmed in the case of financial statement comparability. This establishes a positive and significant relationship between financial statement comparability and the dependent variable.
Summary of model 1 regression coefficients (the robust regression)
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | 1.671 | 0.137 | 12.230 | 0.000 | |
| COMPACC | 4.267 | 0.366 | 11.652 | 0.000 | |
| SIZE | 0.011 | 0.001 | 10.053 | 0.000 | |
| ROA | 0.415 | 0.013 | 32.105 | 0.000 | |
| LEV | −0.005 | 0.008 | −0.599 | 0.549 | |
| SALE | −0.002 | 0.001 | −1.996 | 0.046 | |
| CFO | −0.002 | 0.001 | −1.915 | 0.055 | |
| BUSY | −0.191 | 0.009 | −21.169 | 0.000 | |
| Coefficient of determination | Adjusted coefficient of determination | Akaike information criterion (AIC) | F-statistic | p-value | |
| 0.379 | 0.377 | 1848.941 | 2328.388 | <0.001 | |
Source(s): Created by authors
4.4.2 Model 2: the dependent variable (MTB) (the robust regression)
As presented in Table 10, the p-value falls below 0.05 (p < 0.001), signifying the statistical significance of the regression model. The coefficient of determination quantifies the relationship between the independent and dependent variables while considering control variables stands at 0.469. This implies that the independent variables can account for approximately 47% of the variance in the dependent variable. Furthermore, Table 10 rigorously examines the significance of the independent variable’s effect or the regression coefficients. Considering its associated p-value, which is less than 0.05, the hypothesis concerning the relationship between the research variables is supported in the case of financial statement comparability. This confirms the presence of a negative and significant relationship between financial statement comparability and the dependent variable.
Summary of the regression coefficients of model 2 (the robust regression)
| Regression model | Unstandardized coefficient | t-statistic | p-value | ||
|---|---|---|---|---|---|
| B | Standard error | ||||
| C | −101.042 | 3.963 | −25.498 | 0.000 | |
| COMPACC | −274.901 | 10.622 | −25.880 | 0.000 | |
| SIZE | −0.223 | 0.031 | −7.124 | 0.000 | |
| ROA | 5.928 | 0.375 | 15.822 | 0.000 | |
| LEV | 2.780 | 0.239 | 11.623 | 0.000 | |
| SALE | 0.015 | 0.026 | 0.570 | 0.569 | |
| CFO | −0.003 | 0.037 | −0.079 | 0.937 | |
| BUSY | 1.826 | 0.262 | 6.975 | 0.000 | |
| Coefficient of determination | Adjusted coefficient of determination | Akaike information criterion (AIC) | F-statistic | p-value | |
| 0.469 | 0.430 | 2655.299 | 1138.952 | <0.001 | |
Source(s): Created by authors
5. Discussion and conclusion
Conservatism is a pivotal concept in accounting, which delays the recognition of revenues while recognizing the costs in a timely manner, and undervalues the assets and overstates liabilities. Conventionally identified with the expression “anticipate no profits but realize all possible losses” and may appear in different shapes such as behavioral, temporal, or evaluative. Respecting the importance of conservatism in the informativeness of accounting figures, this paper examines how financial statement comparability, one of the qualitative characteristics of financial reporting, affects conditional and unconditional conservatism.
The first hypothesis of the study indicates that there is a positive and significant relation between financial statement comparability and conditional conservatism, thereby meaning that with increased comparability, conditional conservatism for companies listed on the TSE will be improved, which aligns with the findings of (Hansen et al., 2018). The second hypothesis also indicates a negative association between financial statement comparability and unconditional conservatism, meaning higher comparability may reduce unconditional conservatism, thus supporting Chen and Gong’s (2019) arguments.
The results propose critical implications for corporate managers, investors, auditors, and regulators. Managers should focus on comparability in financial statements, as this may enhance conditional conservatism and could even improve a firm’s relative market position by signaling alleviated agency problems. In addition, regulators may consider that improved financial reporting comparability and reduced unconditional conservatism are drivers of enhanced relevance of financial information. Consequently, focusing on these two elements and developing regulations in this regard may significantly help decrease information asymmetry and increase market efficiency. For example, implementing standardized accounting procedures may enhance comparability and transparency, impacting how conservatively reporting finances is conducted. Macroeconomists and analysts may also benefit from our findings by providing more accurate predictions and economic plans, as it is well-documented in the literature that firm-level data has predictable power in allocating macroeconomic indicators such as the unemployment rate (Salehi et al., 2020a, b, c) and GDP growth dispersion (Daemigah, 2020).
Limitations acknowledged in the study include possible selection bias due to inconsistent or incomplete data and legislative and cultural contexts that could influence reported results. In addition, economic conditions prevailing at the time of research, such as booms or recessions, may affect relative levels of conservatism and comparability. All these factors should be considered in interpreting the findings. Apart from these limitations, the study develops valuable insights into the association of conservatism with financial statement comparability as to its importance in the improvement of the quality of financial reporting.Table A1
References
Further reading
Appendix
The operational definition of variables
| No. | Variable | Type | Definition |
|---|---|---|---|
| 1 | Conditional conservatism | Dependent | One of the dependent variables in this study is conditional conservatism, which is assessed using Basu’s (1997) conservatism model. In this model, positive returns signify positive news, while negative returns indicate negative news. Basu’s model posits that earnings respond more promptly to negative than positive news. The model is as follows: Where: NIi,t: the company’s accounting net income in year t to the equity market value at the end of the year (the beginning of year t). Di,t: a dummy variable that equals 1 if there is bad news and 0 otherwise. In this model, if β3 , it shows the degree of conservatism, which is calculated separately for each firm and year. (β3+ β2) is the reaction of earnings to bad news, and because (β3+ β2) is greater than β2, then β3 is positive, and in fact, it is the earnings asymmetric timeliness coefficient, which is the measure of conservatism. And β2 is the reaction of earnings to good news. Ri,t: the stock return in year t, which is defined as the difference between the stock price at the end of the period and the stock price at the beginning of the period plus adjustments (such as including dividends, bonus shares, etc.) divided by the stock price at the begging of the period |
| 2 | Unconditional conservatism | Dependent | According to Ryan (2006) and Beaver and Ryan (2000), the paramount measure of unconditional conservatism is the market-to-book value ratio. They contend that costs such as research and development and advertising, when expensed rather than capitalized, do not find representation in book values. At the same time, the market assigns a value to these expenses. Consequently, the application of unconditional conservatism creates a disparity between market and book values, leading to a higher market-to-book value ratio. Thus, the MTB measure, denoting the market-to-book ratio, is chosen as the gauge of unconditional conservatism MTB: the market-to-book ratio of equity |
| 3 | Comparability | Independent | The model developed by De Franco, Kothari and Verdi (2011) is employed to assess financial statement comparability, with the accounting outputs of companies serving as the basis for evaluating the comparability of their financial figures. When the accounting systems of two companies are similar, the reported accounting results tend to be alike. In this study, following the approach of De Franco, Kothari and Verdi (2011) and Kim et al. (2016), economic events and accounting outputs (figures) are measured using the companies' stock returns and earnings, respectively. Financial statement comparability is calculated for each firm-year observation, and the estimation of parameters βi and αi for the most recent 12 quarters of firm i in year t is conducted using the regression equation presented below: Earningsit: the adjusted quarterly net earnings to the market value of equity at the beginning of the period for firm i and year t. Returnit: the quarterly stock return for firm i and year t. αi,βi: the estimated coefficients of the accounting function for firm i. Likewise, this procedure is reiterated to estimate the parameters αj and βj for firm j in year t. When employing the quarterly stock return of firm i in year t within both accounting systems, the projected earnings for both firms i and j, corresponding to the same economic events (i.e., the quarterly stock return of firm i in year t), are determined through the following regression equations. Essentially, the proximity of accounting functions between two companies serves as an indicator of comparability between them. To quantify the disparity between the accounting functions of firms i and j the anticipated earnings of the two companies are calculated using the following models: E(Earnings) iit: the predicted earnings of firm i considering the function of firm i and the return of firm i in year t. E(Earnings) ijt: the predicted earnings of firm j considering the function of firm j and the return of firm i in year t. Finally, the accounting comparability between firms i and j is defined as follows: CompAcctijt: the comparability of accounting figures between firms i and j, whose greater value shows higher comparability. Then, to determine the firm-year comparability measure for each firm i, all the values of CompAcctijt are sorted from the greatest to the lowest. Then, the mean of CompAcctijt for the four firms j with the highest comparability to firm i in year t is calculated as the comparability measure for firm i, which is used as the variable M4-CompAcctit in the model |
| 4 | Size | Control | Firm size is defined as the logarithm of total assets at the end of the fiscal year. According to Yuliarti and Yanto (2017), we control for the firm size effect on the level of accounting conservatism |
| 5 | ROA | Control | Return on assets is calculated by dividing operating income by total assets. ROA and sales are added to control profitability’s impact on the conservatism level, as suggested by Khalilov and Osma (2020) |
| 6 | Lev | Control | Financial leverage represents the ratio of total debts to total assets. This variable is also employed to control for the impact of indebtedness on conservatism (Dang and Tran, 2020) Sales: Sales growth is calculated as the difference between the current year’s and the previous year’s sales, divided by the previous year’s sales |
| 7 | CfO | Control | Operating cash flow is the firm’s operating cash flow in a given year. This variable is added to control the impact of operation cash flow on accounting conservatism (Martani and Dini, 2010) |
| 8 | Busy | Control | This variable takes the value 1 if the fiscal year ends on the 29th Esfand, and 0 otherwise Industry: A dummy variable representing the industry |
| 9 | Year | Control | A dummy variable indicating the year |
Source(s): Created by authors
