2. Literature Review
Research conducted in European countries shows that improving the quality of accounting is essential in the context of emerging crises, especially for small enterprises with low liquidity and concentrated ownership structure (
Onali et al. 2017, pp. 455–78). Interesting conclusions also stem from studies on the influence of macroeconomic conditions on income manipulations of European-listed companies during the 2008–2009 financial crisis. The main findings are a decrease in income smoothing and improved accruals quality in the crisis period. CEOs have less incentive to manipulate earnings in crisis times due to a higher market tolerance for weak performance (
Filip and Raffournier 2014, pp. 72–79). We agree with earlier studies’ findings (
Galloppo and Paimanova 2018, pp. 185–213) that Eastern European financial markets did not meet expectations regarding transparency, financial liquidity, and risk in the face of emerging crises. The research above covers a few countries (the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Turkey, and Ukraine). It proves that the European countries’ financial markets are more sensitive to current economic changes, as evidenced by their greater volatility level than well-established financial markets.
European countries have also been classified and divided in terms of the quality of financial information based on a single set of accounting standards (IFRS) into three clusters. Cluster 1 includes Belgium, France, and Sweden; cluster 2 includes Finland, Portugal, and Greece; and cluster 3 includes Austria, Denmark, Germany, Ireland, Italy, the Netherlands, Spain, and the UK. These clusters present a similar pattern of accounting quality reffering to financial information (
Morais et al. 2018, pp. 334–50). It is assumed that improvement in financial transparency also affects earnings management. This statement is criticised by professional observers due to its complexity, as shown by research conducted on the group of European-listed firms (
Cadot et al. 2021, pp. 1628–37). Other studies of cross-national corporate governance conducted in 11 countries in Central and Eastern Europe have demonstrated a causal relationship between good financial reporting that results from good corporate governance. Regarding accounting quality, research revealed problems with Eastern Europe’s transition to western norms. Due to legal heritage, countries were divided into three clusters. The first cluster includes the Baltic countries (Estonia, Latvia, and Lithuania), the second one consists of the Visegrád group (Hungary, Poland, Czech Republic, and the Slovak Republic), and the final cluster covers the countries of Southern Europe (Slovenia, Croatia, Romania, and Bulgaria) (
Lindahl and Schadéwitz 2018, pp. 24–49). Similar conclusions can be drawn from the research on the effect of foreign direct investments (FDIs) on financial reporting quality in transitional economies. Data analysis from 12 transitional economies in Central and Eastern Europe has shown that investment freedom and freedom from corruption increase earnings quality. Moreover, the results show a high level of FDIs is associated with high conditional conservatism (
Hämäläinen and Martikainen 2015, pp. 295–310). In Poland and Eastern European countries, the research confirms the strong impact of the quality of accounting and audit attributes on fluctuations in the prices of companies’ shares. (
Kousenidis et al. 2014;
Mackevičius and Kazlauskienė 2016, pp. 120–37;
Chłodnicka and Zimon 2020).
Therefore, it is crucial to introduce protective and control mechanisms. Investors’ protective mechanisms demand a positive impact on all the firm’s operational processes. These mechanisms tend to establish justice in the capital market via existing structures. Major investors are likely to consume more time and energy to evaluate a particular company’s investment and operating policies to guarantee that CEOs apply proper business strategies and may use their voting rights to encourage individuals for hard working to increase the interest of stockholders. Additionally, effective monitoring by major stockholders may also decrease the opportunistic behavior of CEOs. Therefore, financial statements may lose their effect on reflecting the economic facts of companies and lose their usage to impact the decision-making process. In turn, unreliable financial reports may promote the incidence of SPCR (
Butar and Murniati 2021).
Also, the auditors’ role and audit attributes for guaranteeing the correctness of financial statements and accounting quality have experienced considerable attention from empirical research.
4. Research Methodology
The statistical population comprises all listed companies on the Iraqi Stock Exchange between 2012 and 2018. A systematic elimination method is used for sampling. According to the data collected at the end of 2018, the final statistical sample was determined based on the information in
Table 1.
Basic information and initial data for hypothesis testing are collected using the Iraqi Stock Exchange database. The data analysis method is cross-sectional and year-to-year (data panel). The multivariate linear regression method has been used to test the hypotheses, and descriptive and inferential statistical methods have been used to analyze the obtained data. Thus, the frequency distribution table is used to describe the data. At the inferential level, the F-Limer test, the Hussmann test, the normality test, and the multiple linear regression test are used to test the research hypotheses.
4.1. Research Model
In this study, the following multiple regression model is used to analyze the relationship between accounting quality and audit attributes with the SPCR:
The statistical population is tested separately in pre- and post- ISIS eras for additional analysis. Their results are compared by model (1).
4.2. Research Variables
The dependent variable:
Wi,t: the firm-specific monthly return of firm i in month t;
εi,t: the residual return of the stock of firm
I in month
t;
where:
Ri,t: is the stock return of firm j in month t over the fiscal year;
Rm,t: is the market return in month t. To calculate the monthly market return, the index at the beginning of the month is reduced from the index at the end of the month, and the obtained value is divided into an index at the beginning of the month.
NCSKEW: the negative coefficient of skewness of monthly return of firm j in month t;
Wjt: the firm-specific monthly return of firm j in month t;
N: the number of monthly returns.
Independent variables
Accounting Quality (ACCQ): the absolute error of accruals are used to measure accounting quality, persistent with the research of
Francis et al. (
2005) and
Cornell et al. (
2017). This criterion is obtained from an adjusted Dechow and Dichev model (2002) by
McNichols (
2002). The absolute error of accruals in this model reflects the accruals in the past, current, and future cash flows, and it is more compatible with accrual basis accounting. The
Dechow and Dichev (
2002) model are as follows:
TACC indicates total accruals in the above model and is calculated according to the
Dechow and Dichev (
2002) model.
OCF is the operating cash flows, ∆
sales is the change in sales,
PPE is the property, plant, and equipment, and the model’s estimated residual. The absolute value of the residual in the model is used to measure accounting quality.
- 2.
Audit attributes:
- -
Audit quality (AQ) indicates the size of audit firms. If the audit firms are big, this equals one and otherwise zero.
- -
Auditor industry specialization (AIS) is the auditor’s specialization in the industry i and year t. We used market share as an index for auditor industry specialization. Increased market share increases auditor specialization in industry and experience compared to other rivals. The auditor’s market share is calculated as the total assets of all employees/owners of each auditing firm in a particular industry to all owners’ total assets.
In this study, the industry specialization institutions have market share, i.e., the above ratio is more than [1.2 ∗ (1/the number of all the companies)]. After calculating an audit firm’s market share, the audit firm specialises in that industry if the result is higher than the above equation. Therefore, it equals one and zero otherwise (
Habib and Bhuiyan 2011).
- -
k: the number of auditors in the related industry
s: total audit fee received by the auditor in the related industry
S: total audit fee received by auditors in related industry
- -
Audit fees (LNAFEE): the natural logarithm of audit fees
- 3.
Board attributes:
Corporate governance (CG): in this study, the number of board, audit committee and board specialization, audit committee, and board independence are used as the corporate governance index. Therefore, we obtain the corporate governance variable by exploratory factor analysis of these variables.
Control variables
M-tenure: equals the years that the management has continuously tenured
M-change: If the CEO has changed in the year under review, it equals one, otherwise zero.
Auditor tenure (A-Tenure): equals the years the auditor has continuously audited the unit under consideration.
Auditor change (A-change): If the auditor has changed in the year under review, it equals one; otherwise, zero.
Board of Directors Independence (B-Ind): equals the ratio of non-executive board members to the total number.
Firm size (Size): equals the natural logarithm of the total assets.
Leverage (LEV): variable equals the dividend of total debts to total assets in the current year.
ROE: is the calculation result of dividing the net income by the book value of the shareholder’s equity in the current year.
Return on Assets (ROA): The result of dividing the net profit by the total assets’ book value.
Firm’s age (Age): equals the time since the firm was established to the year under consideration.
Sales Growth (GRW): equals this year’s sales minus last year’s sales divided by last year’s sales.
Market to Book ratio (MTB): dividing the market share by the book equity value.
Loss: if the firm reports a loss, it equals one and otherwise zero.
Year: The dummy variable of the year.
Industry: The dummy variable of industry.
5. Results
This study used one model to analyze the relationship between accounting quality and audit attributes with SPCR. The panel data consists of 35 Iraqi companies from 2012 to 2018. The following variables are used to estimate the models. The variables include accounting quality, audit attributes, SPCR, and other control variables.
Table 2 shows the descriptive statistics.
All the variables are stationary at the unit root test. The obtained LM statistics for each variable are reported in
Table 3.
According to the Collinearity test in
Table 4. According to the VIF statistics, no collinearity is between the variables.
Table 5 presents the correlation coefficients of the variables. This test examines the relationship between the variables in pairs; its output is the above matrix. Since this diameter examines the correlation between the variable and always equals one (i.e., complete correlation). The correlation is higher when the amounts are closer to one, and the closer they are to zero, they lack correlation.
According to the integration test results in
Table 6, the null hypothesis of data integration at the 99% confidence level is rejected. Therefore, a panel data model should be used to estimate the coefficients of these models.
In
Table 7, the Hausman test statistic is 20.59, 23.89, and 32.45 for research models. Since this is lower than the pre- and post-ISIS era tables, and the H
0 (i.e., the proper model is the random effect model) is rejected, the fixed effects model is selected, which is an efficient model. For the primary research model, since the table’s
is higher and the null hypothesis (i.e., the proper model is the random effect model) is not rejected, the efficient model is the random-effects model.
Based on
Table 8, a positive and significant relationship between audit fees and SPCR is clear. This is because its
p-value is 0.039, which is lower than the significance level of 0.05, and its coefficient is a positive amount of 0.069. Further, a negative and significant relationship exists between accounting quality, auditing, and auditor industry specialization with SPCR. This is because their
p-values are 0.007, 0.000, and 0.047, accordingly, which is lower than the significance level of 0.05, and their coefficients are the negative amounts of 0.650, 1.409, and 1.446. Therefore, the null hypothesis is rejected, indicating no significant relationship between these variables. In other words, improving these findings means that higher quality accounting figures will likely send positive signals to the market, resulting in a reduced likelihood of SPCR. Supportively,
Chen et al. (
2001) confirm that the SPCR might result from the firm’s management’s provision of low-quality accounting reports.
Based on the results, there is no significant relationship between corporate governance and SPCR at the 95% confidence level. However, at the 90% confidence level, there is a negative and significant relationship. This is because its p-value is 0.094, higher than the significance level of 5% and lower than 10%. Hence, there is no significant relationship between corporate governance and SPCR at the 95% confidence level. However, at a 90% level, there is a negative and significant relationship between these variables. In addition, there is no significant relationship between audit concentration and SPCR.
According to
Table 8, the model evaluation results are robust. Four classic econometrics estimation hypotheses are analyzed in the panel data, and reliable results are reported. Furthermore, since the model’s significance level is 0.000, lower than the significance level of 5%, the company’s intercept is not significant according to the applied regression. Finally, the model has a sufficient and proper significant level.
The study also examined pre-ISIS data.
Table 9 shows a significant positive relationship between audit fees and SPCR. This is because its
p-value equals 0.013, which is lower than the significant level of 0.05, and its coefficient is a positive value of 0.3 0.376; therefore, the null hypothesis is not rejected.
Table 9 shows a negative and significant relationship between accounting quality, auditing, corporate governance, and audit concentration with SPCR. This is because their
p-values are 0.000, 0.007, 0.002, and 0.000, accordingly, which is lower than the significance level of 0.05. Their coefficients are negative amounts of 0.010, 0.014, 0.021, and 0.079. Therefore, the null hypothesis (i.e., no significant relationship between these variables and SPCR) is rejected. However, the opposite hypothesis indicating a significant relationship is accepted.
The results of post-ISIS data were examined.
Table 10 shows a positive and significant relationship between corporate governance and SPCR during and after ISIS. This is because its
p-value is 0.000 which is lower than the significant level of 0.05, and its coefficient is the positive amount of 6.497. This conveys that ISIS has negative corporate governance to perform its monitoring duties correctly to reduce SPCR. However, this relationship was negative in the pre-ISIS era because of increased corporate governance and reduced SPCR. Furthermore, the relationship between audit fees and SPCR is not confirmed. This is because their
p-value is 0.372, greater than the significance level of 5%. Therefore, the relationship between these two variables is rejected.
Table 10 shows a negative and significant relationship between accounting quality, auditing, auditor industry specialization, and auditor concentration with SPCR during and after the ISIS era. Their
p-value is 0.022, 0.0024, 0.041, and 0.032, accordingly, lower than the significance level of 0.05. Their coefficients are negative amounts of 0.096, 0.037, 2.301, and 0.664. Therefore, the null hypothesis is rejected. However, the opposite hypothesis indicating a significant relationship is accepted.
The research model was tested with different methods to ascertain the results’ robustness and determine whether the results align with the leading results.
Table 11 results based on the t + 1 method show a negative and significant relationship between accounting quality, auditor industry specialization, corporate governance, and audit concentration with SPCR. Their
p-values are 0.009, 0.044, 0.031, and 0.046, accordingly, which is lower than the significance level of 5%. Their coefficients are also negative amounts of 0.639, 0.048, 1.055, and 0.132, indicating a significant negative relationship between these variables and SPCR. However, there is no significant relationship between audit fee and audit quality with SPCR.
Based on the ordinary least square method and
Table 11, there is a significant negative relationship between accounting quality, auditor industry specialization, and corporate governance with SPCR. Their
p-values equal 0.006, 0.028, and 0.000, lower than the significance level of 5%. Their coefficients equal negative amounts of 0.690, 1.369, and 0.473. Based on the ordinary least square method (t + 1 and main method), no significant relationship at the level of 95% is observed between audit fee, audit quality, and audit concentration with SPCR.
Based on the fixed effects method, there is a significant negative relationship between accounting quality, corporate governance, and SPCR. This is because their p-values equal 0.000 and 0.051, lower than the significance level of 5%. Their coefficients equal negative amounts of 0.950 and 0.055. Based on the fixed-effect method, a positive and significant relationship exists between audit fees and audit concentration with SPCR. Their p-values equal 0.016 and 0.009, and their coefficients equal positive amounts of 1.177 and 802.803, hence this equation contradicts previously applied methods. There is no significant relationship at the level of 95% between audit quality and auditor industry specialization with SPCR.
The research model was tested with different methods to ascertain the results’ robustness and determine whether the results aligned with the main results. According to
Table 12 and based on the random effects method, there is a negative and significant relationship between accounting quality, auditing, and audit concentration with SPCR. Their
p-values equal 0.004, 0.026, and 0.000, lower than the significance level of 5%. Their coefficients are negative amounts of 0.012, 0.075, and 0.016. Based on the random-effects model, there is no significant relationship at the level of 95% between auditor industry specialization and corporate governance with SPCR in the pre-ISIS era. However, according to the random effects method, there is a positive and significant relationship between audit fees and SPCR in the pre-ISIS era. This is because its
p-value equals 0.000 and its coefficient equals 0.107, which indicates a positive and significant relationship between these two variables.
According to the ordinary least square method, there is a negative and significant relationship between accounting quality, auditing, corporate governance, and audit concentration with SPCR. Their p-values are 0.000, 0.007, 0.000, and 0.014, lower than the significance level of 5%. Their coefficients equal negative amounts of 0.233, 0.014, 0.010, and 0.018. According to the ordinary least square method, there is no significant relationship at a 95% level between auditor industry specialization and SPCR in the pre-ISIS era. However, according to the ordinary least square method, there is a positive and significant relationship between audit fees and SPCR in the pre-ISIS era. This is because its p-value equals 0.000, and its coefficient is a positive amount of 0.165, which indicates a positive and significant relationship between these two variables.
Table 13 shows a negative and significant relationship between accounting quality, auditing, corporate governance, and auditor concentration with SPCR. Their
p-value is 0.000, 0.000, 0.021, and 0,006, which is lower than the significance level of 5%. Their coefficients are negative amounts of 0.019, 0.104, 0.04, and 0.392. Based on the t + 1 method, there is no significant relationship at a 95% level between auditor industry specialization and audit fee with SPCR in the post-ISIS era.
Based on the ordinary least square method, there is a negative and significant relationship between accounting quality, auditing, auditor industry specialization, and auditor concentration with SPCR. Their p-value is 0.017, 0.000, 0.014, 0.004, and 0.045 accordingly, which is lower than the significance level of 5%. Their coefficients are negative amounts of 0.094, 0.011, 2.299, and 50.555. Based on the ordinary least square method, there is no significant relationship at a 95% level between corporate governance and SPCR in the post-ISIS era. However, according to the ordinary least square model, there is a positive relationship between audit fees and SPCR in the post-ISIS era. This is because its p-value equals 0.021, and its coefficient is 0.030, indicating a positive and significant relationship between these two variables.
Based on the random-effects model, there is a negative and significant relationship between accounting quality, auditing, auditor industry specialization, and corporate governance with SPCR after the ISIS era. Their p-values are 0.030, 0.000, 0.002, and 0.017, accordingly lower than the significance level of 5%. Their coefficients are negative amounts of 0.235, 0.010, 2.255, and 0.094. According to the random-effects model, there is no significant relationship at 95% between auditor concentration and SPCR in the post-ISIS era. Based on the random-effects model, there is a positive and significant relationship between audit fees and SPCR. This is because its p-value equals 0.002 and its coefficient equals 0.005, which indicates a positive and significant relationship between these two variables.
6. Discussion and Conclusions
Recently, the SPCR has drawn the attention of accounting scholars and activists in the markets (
Kousenidis et al. 2014;
Richardson et al. 2005). The financial crisis had an indicative impact on financial markets. The political and economic instability resulting from the entry of ISIS into Iraq has created severe problems for society’s economic, political, security, and performance dimensions. This instability can significantly impact the accounting quality. The auditors’ role and audit attributes in guaranteeing the fairness of financial reports and the quality of accounting figures have received remarkable attention from empirical and analytical efforts. This study addresses the relationship between accounting quality and audit attributes with companies’ SPCRs on the Iraqi Stock Exchange. This study selected audit quality, auditor industry specialization, audit concentration, and audit fees for audit attributes.
Results show a significant positive relationship between audit fees and SPCR.
Further, a negative and significant relationship exists between accounting quality, auditing, and auditor industry specialization with SPCR. Based on the results, there is no significant relationship between corporate governance and SPCR at the 95% confidence level. However, at the 90% confidence level, there is a negative and significant relationship. There is no significant relationship between corporate governance and SPCR at the 95% confidence level. Yet, at a 90% level, there is a negative and significant relationship between these variables. Additionally, there is no significant relationship between audit concentration and SPCR.
In the test of pre-ISIS data, there is a positive and significant relationship between audit fees and SPCR. There is a negative and significant relationship between accounting quality, auditing, corporate governance, and audit concentration with SPCR. In contrast, the results of post-ISIS data show a positive and significant relationship between corporate governance and SPCR during and after ISIS. This conveys that ISIS has negative corporate governance to perform its monitoring duties correctly to reduce SPCR. However, this relationship was negative in the pre-ISIS era because of increased corporate governance and reduced SPCR. Furthermore, the relationship between audit fees and SPCR is not confirmed. Additionally, there is a negative and significant relationship between accounting quality, auditing, auditor industry specialization, and auditor concentration with SPCR during and after the ISIS era.
This study provides deep insights into audit qualities by analyzing four audit attributes’ distinct effects on stock price risk. This paper may contribute to the incremental documents of SPCR. The results from the Iraq developing market add to the understanding of how the financial market participants may use available data to compare with that of developed markets. The findings of this empirical paper may generate helpful information for legislators and capital market analysts, contribute to the development of knowledge in this field, and bridge the gap in the literature.
Our paper also provides some practical implications for equity owners and CEOs. According to the findings of this study, the equity owners are informed that they can preclude the probability of SPCR by improving the corporate governance mechanisms, such as the number of board members, audit committee and board specialization, the presence of audit committee, and board independence. Moreover, they may reduce the SPCR by contracting with auditors possessing particular characteristics, including specialized, market concentration, and high-quality services. More importantly, firms’ authorities are aware that paying greater audit fees to auditors might be translated by the market analysts as hoarding bad news inside the firms by providing auditors with economic incentives. Finally, market policy makers may use our findings to design and implement policies, including improving investors’ rights protection and external corporate governance mechanisms, to rectify the impact of the regional phenomenon, such as ISIS consequences, on the market. The unchanged findings of the pre-ISIS and post-ISIS suggest that strong corporate governance mechanisms such as accounting quality, audit specialization, and board characteristics are likely to perform their governance role effectively even after the unexpected crisis. Thus, emphasizing such issues is likely to improve countries’ business environment.
All the research might be suffered from potential limitations. In this paper, the investigated firms have been decreased by data integration; in case of having further data of Iraqi listed firms, the possibility of having different conclusions would be increased. Additionally, the chosen period and region to conduct the study are limited to the specifications stated in the third section. Therefore, such limitations must be considered to generalise the findings to other ages and geographical regions.