One of the goals that companies should have at every stage of their lifecycle is to achieve profitability. Profitability is one of the key factors in attracting capital from the public and encouraging individuals to invest in companies. Financial statements are important informational resources that provide the necessary information to users for evaluating a company and investing in it. Generally, the purpose of publishing financial reports is to provide essential information about the company’s financial position, operational efficiency and liquidity to individuals who have invested or intend to invest in the company. For this reason, companies must present their information in a completely transparent and accurate manner, free from any bias or Fraud. The primary assumption in financial markets is the provision of accurate information, free from errors and fraud. However, with the occurrence of financial crises in large companies worldwide, such as WorldCom and Enron, the issue of fraud in financial statements has entered the political arena and attracted the attention of lawmakers, academic circles and many researchers. Agency theory argues that, due to the separation of ownership and control in companies, managers are encouraged to act opportunistically at the expense of shareholders [1]. One dimension of the agency problem is the conflict between the preferences of managers and shareholders regarding the disclosure of information. A change or Fraud in financial statements is referred to as financial statement fraud, which leads to intentional false statements to portray the company in a more favorable light, with the goal of misleading and deceiving users. Fraud in financial statements is classified as white-collar crime. White-Collar Crime is a crime committed by an individual who holds a high position in society and their role [2]. Financial fraud is defined by the association of auditors as the misrepresentation of a business's financial condition through the deliberate and incorrect presentation or omission of some disclosed amounts in the financial statements, which leads to the deception of users and the adoption of suboptimal economic decisions. The impact of fraud on corporate profits and stock prices in financial markets is just part of the story. Stock price changes generally occur in two forms: Spikes and crashes. In recent years, sudden stock price changes have attracted significant attention. The phenomenon of stock price crashes, which results in excessive return reductions, is particularly noteworthy since investors place high importance on the return of their investments. The stock price crash, which has a massive impact on capital markets [3,4], has been more widely studied compared to stock price spikes [5]. When stock prices in the market experience a sharp Crash, this adverse event is referred to as a stock price crash. In a given period, when the stock return of a company is lower than the return of the overall market index, the likelihood of a stock price crash increases. It has been shown that financial fraud in companies' financial statements can lead to significant negative outcomes in the capital market [6]. Managers are motivated by factors such as bonuses, career prospects and maintaining their power, which gives them incentives to hide or delay bad news [7,8]. This concealment or delay of bad news leads to an overvaluation of stock prices in the market [8]. When the bad news eventually reaches the market or is identified by market participants, it leads to a stock price crash [9,5]. Given the importance of financial fraud in the information environments of banks, this research seeks to fill this gap in the literature by examining the relationship between financial fraud and stock price crash risk, a topic that appears to have not been explored in studies conducted in Iran and Iraq. In other words, this study seeks to answer the question: Does financial fraud have a significant impact on the risk of stock price crash?

Theoretical Framework

The disclosure of accounting fraud can lead to significant negative consequences in the capital market [10,6]. For example, the stock price of Kraft Heinz on February 22, 2019, Crashd by approximately 28% (or $15 billion in market value) in response to the disclosure of a subpoena by the U.S. Securities and Exchange Commission (SEC) regarding its accounting policies. The company also announced that it would reduce the value of the Oscar Mayer and Kraft brands by $15.4 billion, which would result in a surprising $12.6 billion loss for the fourth quarter of 2018. Finally, on April 22, 2019, Kraft Heinz announced that its CEO would resign on June 30, 2019. This example highlights the severe consequences that companies face for accounting fraud and misreporting.

Managers have incentives to avoid bad news because of concerns about executive compensation contracts, career prospects and building their empires [7,8]. Managers who hide detrimental news or delay its release can lead to overvaluation of stock prices [8]. When the hidden detrimental news reaches its peak, its final disclosure to the capital market leads to a stock price crash [9,5]. The risk of a stock price crash, due to its significant impact on capital markets, has attracted considerable research attention [3,4,11]. However, the role of accounting fraud as a predictor of future stock price crashes has not been examined.

In the capital market, stock prices react to information that market participants deem relevant [12]. The disclosure of adverse news reduces investor confidence, increases selling pressure in the capital market and raises the risk of a stock price crash. Investor attention also plays a vital role in determining capital flows. According to agency theory, as proposed by Jensen and Meckling, with the separation of ownership from control, managers, having full and accurate access to the company’s situation, are incentivized to manipulate financial statement items to present the company’s situation favorably. Financial fraud occurs due to moral hazard issues. The market's discovery of financial fraud can provoke a negative reaction. However, there are competing views regarding how the discovery of financial fraud in financial statements affects the risk of stock price crashes. On one hand, due to the financial and credit costs incurred by companies when disclosing fraud in financial statements, accused companies may seek to strengthen corporate governance to improve the transparency of their information environment, restore public trust and improve disclosure practices [13]. Since stock price crashes occur more in companies with a more ambiguous information environment, actions aimed at improving information transparency and restoring public trust should reduce the risk of future stock price crashes [9]. On the other hand, when companies do not take corrective actions to improve their information environment, the disclosure and detection of accounting fraud in financial statements increase managers’ incentives to hide bad news. Therefore, following the disclosure of accounting fraud, the accumulation of more bad news by employees and the deterioration of disclosure quality led to more ambiguous information environments and a decrease in investor trust in companies, which in turn increases the risk of a future stock price crash. The competitive theories presented above are based on economic theories regarding the information environment of companies as a mechanism through which financial fraud can likely affect the risk of a stock price crash in the future. However, how companies react to past accounting fraud events may reduce the accumulation of bad news and decrease the risk of a stock price crash and vice versa.

Accounting fraud imposes significant consequences on the accused companies. The disclosure of fraud can lead to adverse stock price reactions, the displacement of managers and damage to the reputation of managers [6]. Given the importance of accounting fraud in determining the information environment of companies, we aim to fill this gap in the financial literature by examining the relationship between accounting fraud and future stock price crash risk.

This study examines the effect of accounting fraud on the risk of stock price crashes using a multivariate framework. While previous studies clearly show that the disclosure of accounting fraud can lead to a stock price crash it is still unclear how companies respond to discovered (past) or predicted (future) accounting fraud. The methods that amplify or adjust the future risk of stock price crashes remain poorly understood. In this study, it is assumed that past accounting fraud (fraud that occurred earlier) can influence the risk of a company’s stock price crash by affecting the behavior of adverse news hoarding in the future and its information environment. One perspective is that the disclosure of fraud can lead to corporate governance changes [13], which help restore investor confidence and improve the transparency of companies' information environments, thereby reducing the risk of future stock price crashes. However, a competing view argues that, following the disclosure of fraud, the decrease in investor confidence can intensify managers' incentives to avoid disclosing adverse news which, in turn, increases the risk of a future stock price crash [6]. Given these competing expectations, it is assumed that accounting fraud may be either positively or negatively associated with the future risk of a stock price crash.

Literature Review

Al-Said et al. in a study examining the impact of company characteristics on the risk of future stock price crashes, showed that the audit firm's industry expertise, board size and board independence have a significant negative impact on the risk of future stock price crashes.

Shou et al. investigated the effect of business strategy on the risk of stock price crashes. The results indicated that a forward-looking business strategy increases the risk of stock price crashes and that higher information asymmetry and greater CEO overconfidence exacerbate this relationship.

Wang et al. [14] studied how fluctuations in industry cash flow affect the risk of stock price crashes by examining A-shares of Chinese companies. Their findings showed that an increase in the volatility of industry cash flows significantly raises the risk of stock price crashes. Additionally, a higher level of economic policy uncertainty increases the positive effect of industry cash flow volatility on stock price crash risk and the degree of accounting conservatism moderates the effect of industry cash flow volatility on stock price crash risk.

Richardson et al. [15], in a study titled "The Impact of Accounting Fraud on Stock Price Crash Risk," using 51,495 firm-year observations from 2000 to 2014 of U.S. companies, found that accounting fraud has a significant positive relationship with stock price crash risk. This result supports their theoretical framework, which suggests that accounting fraud affects stock price crashes by reducing the transparency of the company’s information environment.

Wang et al. [14], in a study titled "Examining the Impact of Accounting Conservatism and Corporate Governance on Stock Price Crash Risk" in Shenzhen and Shanghai companies between 2011 and 2016, found that accounting conservatism and corporate governance have a significant negative relationship with stock price crash risk. This means that the more conservative the accounting practices and the stronger the corporate governance, the lower the likelihood of a stock price crash.

Davidson [16-18], in a study titled "Earnings Fraud and Balance Sheet Fraud," found that managers are more likely to engage in earnings Fraud when market sensitivity to earnings is high and the stock price of companies is relatively more sensitive to earnings performance. Additionally, managers are more likely to engage in balance sheet Fraud when market forecast risk is high and companies have greater financial constraints.

Based on the theoretical framework, the research hypotheses are formulated as follows:

• Accounting fraud influences the risk of a stock price crash (negative skewness in stock returns) in commercial banks listed on the stock exchanges of Iran and Iraq

• Accounting fraud influences the risk of stock price crashes (low-to-high volatility) in commercial banks listed on the stock exchanges in Iran and Iraq

Research Methodology

The statistical population of the study consists of all banks listed on the stock and over-the-counter markets of Iran between the years 2014 and 2022 (corresponding to 1393 to 1401 in the Persian calendar) and banks listed on the Iraq Stock Exchange between the years 2013 and 2022. In this study, the systematic removal method has been used to select the sample and the companies included in it must meet the following conditions:

• Their fiscal year should end on March 20th (last day of Esfand) and December 31st of each year

• During the study period, they should not have changed their fiscal year

• All required data for the companies under study should be available and accessible

• They should not have had a symbol suspension for more than three months

Given the above conditions and limitations, a total of 17 banks in Iran and 14 banks in Iraq were selected.

The study's variables are the dependent, independent and control variables, which are explained in the next section.

Dependent Variables:

The dependent variable in this study is the risk of a stock price crash, which is measured using the following two criteria:

Negative Skewness of Stock Returns

To calculate the negative skewness of stock returns, the monthly returns of the company are first calculated using Equation (1):

                                                                      (1)

W_j,t   =  The monthly return of company j in month t during the fiscal year

e_ij      =  The residual return of company j in month t, which represents the residual or error term of the model in Equation (2):

                                    (2)

R_j,t   =  The stock return of company j in month t during the fiscal year

R_m,t =  The market return in month t. To calculate the monthly market return, the index at the beginning of the month is subtracted from the index at the end of the month; the result is then divided by the index at the beginning of the month

Then, using the monthly stock return of the company, the negative skewness of stock returns is calculated using Equation (3):

                          (3)

NCSKEW_j,t  =  The negative skewness of the monthly stock return of company j during fiscal year t

W_j,t              =  The monthly stock return of company j in month t

N                 =  The number of months for which the returns are calculated

Low-to-High Volatility

First, the monthly stock returns of the companies (W_it) are calculated according to Equation (3-1) and the related data are divided into two categories: below the average and above the average. The standard deviation of each category is calculated separately and then DUVOL is calculated using Equation (4):

                                        (4)

Where:

DOWN =   Th standard deviation of observations below the average for the monthly stock return of the company

Up         =   The standard deviation of observations above the average for the monthly stock return of the company

n_u          =   The number of months in which the returns are higher than the average

n_d          =   The number of months in which the returns are lower than the average

A higher DUVOL indicates a greater risk of a stock price crash.

Independent Variable

The independent variable in this study is fraudulent financial reporting. To measure fraudulent financial reporting, the restatement of financial statements criterion is used, which is defined as a dummy variable.

If the financial statements have been restated and the annual adjustments are also reported, the likelihood of fraudulent reporting exists for the previous fiscal period and the value of 1 is assigned for that year. Otherwise, the value 0 is assigned.

Richardson et al. [15] used the following variables as control variables in this study:

• Company Size (Size): The natural logarithm of the total book value of assets

• Leverage (Lev): The ratio of total debt to the total book value of assets

• Return on Assets (ROA): The ratio of net income to the total book value of assets

• Company Age (Age): The natural logarithm of the difference between the founding year and the year under study

• Growth Opportunity (MtB): The ratio of the market value of equity to the book value of equity

• CEO Duality (CEOD): If the CEO is also the chairman or vice-chairman of the board, the value is 1; otherwise, it is 0

• Internal Control Weakness (ICW): If the auditor has pointed out a weakness in internal control, the value is 1; otherwise, it is 0

To test the first hypothesis of the study, following the research by Richardson et al. [15], a panel regression is used as shown in Equation (5):

                    (5)

In the above equation:

NCSKEW_it  =  Negative skewness of stock returns for company i in year t.

Fraud_i,t       =  Fraud for company i in year t

Size_i,t          =  Size of company i in year t

Lev_i,t           =  Leverage of company i in year t

Roa_i,t          =  Profitability of company i in year t

Age_i,t           =  Age of company i in year t

MtB_i,t          =  Growth opportunity of company i in year t

CEOD_i,t        =  CEO duality of company i in year t

IWC_i,t          =  Internal control weakness for company i in year t

The first hypothesis is confirmed if the coefficient β1 is significant; otherwise, it is rejected.

To test the second hypothesis of the study, following the research by Richardson et al. [15], a panel regression is used as shown in Equation (6):

                            (6)

In the above equation:

Duvol_i,t = Low-to-high volatility of company i in year t.

The other variables are as defined in Equation (5).

To confirm the second hypothesis, it is expected that the coefficient b₁ will be significant; otherwise, it will be rejected.

Findings of the Study

Descriptive Statistics: The descriptive statistics related to the research variables for Iran are presented in Table 1.

Table 1: Descriptive Statistics of the Research Variables for Iran

In this section, some concepts of descriptive statistics for variables such as the mean, maximum observations and standard deviation of data related to Iran are presented. The main central indicator is the mean, which represents the equilibrium point and the center of gravity of the distribution and is a good indicator for showing the centrality of the data. For example, the dependent variable in the study, the negative skewness coefficient of stock returns, has a mean of -0.362, the variable for company size has a mean of 20.183 and the profitability variable has a mean of -0.023, indicating that most of the data are concentrated around these points. The standard deviation of financial leverage is 0.428 and the standard deviation of growth opportunity is 6.156, indicating that financial leverage has the least dispersion and growth opportunity has the greatest dispersion.

Table 2 presents the descriptive statistics for the research variables related to the country of Iraq. 

Table 2: Descriptive Statistics of the Research Variables Related to Iraq

This section discusses some concepts of descriptive statistics for variables, including the mean, maximum observations and standard deviation of data related to Iraq. The main central indicator is the mean, which represents the equilibrium point and the center of gravity of the distribution and is a good indicator for showing the centrality of the data. For example, the dependent variable in the study, the negative skewness coefficient of stock returns, has a mean of 1.495, the company size variable has a mean of 19.974 and the profitability variable has a mean of 0.0207, indicating that most of the data are concentrated around these points. The standard deviation of profitability is 0.0332 and the standard deviation of growth opportunity is 4.973, indicating that profitability has the least dispersion and growth opportunity has the greatest dispersion.

Results of the First Hypothesis Test

The results of the first hypothesis test are reported in Table 3. 

Table 3: Results of Estimating the Model for the First Hypothesis

The F-Limer test was used to determine the fit of the model between the pooled model and the fixed effects model and the Hausman test was used to determine which model, between the fixed effects and random effects models, is more appropriate. Based on the F-Limer and Hausman tests, the model fit using panel data with fixed effects is chosen. According to the first hypothesis, it is expected that accounting fraud will have a significant impact on the risk of a stock price crash (negative skewness coefficient of stock returns).

The results in Table 3 show that the significance level of the accounting fraud variable in Iran is 0.0306, which is less than the desired error level of 5%. Therefore, it can be concluded that there is a significant relationship between accounting fraud and stock price crash risk in Iran. On the other hand, the coefficient of the accounting fraud variable in Iran is 0.0138, which is positive. This indicates that accounting fraud expertise has a positive impact on the risk of stock price crashes (negative skewness coefficient of stock returns) for banks listed on the Tehran Stock Exchange. Conversely, the significance level of the accounting fraud variable in Iraq is 0.8920, which is higher than the desired error level of 5%. Therefore, it can be concluded that there is no significant relationship between accounting fraud and stock price crash risk in Iraq. This finding suggests that accounting fraud does not affect the risk of a stock price crash on the Iraq Stock Exchange. Therefore, Iran accepts the first hypothesis of the study, while Iraq rejects it.

On the other hand, the significance level of the F-statistic in both Iran and Iraq is less than the 5% error level, indicating that this model is statistically significant at the 95% confidence level and has high validity. Additionally, the Durbin-Watson statistic indicates that there is no autocorrelation between the residuals in both Iran and Iraq, as this value falls between 1.5 and 2.5. According to the results of the likelihood ratio statistic, relationship (5) in both Iran and Iraq suffers from heteroscedasticity. The Generalized Least Squares method has been used to address the heteroscedasticity issue. Furthermore, the adjusted R-squared of the model in Iran (Iraq) is 0.14 (0.35). This suggests that the set of independent and control variables explains 14% (35%) of the changes in the dependent variable.

Results of the Second Hypothesis Test

The results of the second hypothesis test are reported in Table 4. 

Table 4: Results of the Estimation of the Second Hypothesis Model

The Lagrange Multiplier (LM) F-test was used to determine the fit of the model between the pooled model and the fixed effects model and the Hausman test was used to determine which model between fixed effects and random effects was appropriate. Based on the LM F-test and Hausman test, the model fit using panel data with fixed effects is the most suitable. According to the second hypothesis, it is expected that accounting fraud has a significant impact on the risk of stock price crashes (low-to-high volatility) in commercial banks listed on the Tehran Stock Exchange and the Iraq Stock Exchange.

The results in Table 4 show that the significance level of the accounting fraud variable in Iran (0.0424) is less than the desired error level of 5%. Therefore, it can be concluded that there is a significant relationship between accounting fraud and stock price crash risk in Iran. On the other hand, the coefficient of the accounting fraud variable in Iran is 0.2711, which is positive. This indicates that accounting fraud expertise has a positive impact on the risk of stock price crashes (low-to-high volatility) in commercial banks listed on the Tehran Stock Exchange.

In contrast, the significance level of the accounting fraud variable in Iraq (0.7472) is greater than the desired error level of 5%. Therefore, it can be concluded that there is no significant relationship between accounting fraud and stock price crash risk in Iraq. This finding suggests that accounting fraud does not affect the risk of a stock price crash based on the low-to-high volatility criterion in the Iraq Stock Exchange.

Furthermore, since the significance level of the F-statistic in both Iran and Iraq is less than the 5% error level, the model is statistically significant at the 95% confidence level and has high validity. Additionally, the Durbin-Watson statistic indicates that there is no autocorrelation between the residuals in both Iran and Iraq, as this value falls between 1.5 and 2.5.

According to the results of the likelihood ratio statistic, relationship (6) in Iran shows heteroscedasticity. The Generalized Least Squares method has been used to address this heteroscedasticity issue. Furthermore, the adjusted R-squared of the model in Iran (Iraq) is 0.38 (0.33). This indicates that 38% (33%) of the changes in the dependent variable are explained by the set of independent and control variables.

According to the research literature, stock price crashes are viewed as the primary concern for investors. Therefore, the risk of stock price Crashs in the market is one of the main concerns for investors and evidence related to predicting stock price crashes is of vital importance. The increase in stock price crashes leads to pessimism among investors regarding investment in the stock market, which may eventually result in investors withdrawing their capital from the stock exchange. Furthermore, diversification cannot mitigate this risk, unlike systematic fluctuations. The risk of stock price crashes, defined as an undesirable event, is an infectious phenomenon at the market level. This means that a decrease in the price of a particular stock is not limited to that stock alone but rather affects all stocks in the market. Given the significant economic and social impacts of financial statement fraud and the severe consequences of stock price crashes, the findings of this research are beneficial and have implications for managers, investors and policymakers. The results provide valuable insights into how companies respond to past and future financial statement fraud, with consequences for financial markets. Therefore, the aim of this research is to examine the impact of fraud on the risk of stock price crashes in banks in Iran and Iraq.

The results of the first hypothesis test showed that accounting fraud has a positive impact on stock price crash risk (negative skewness of stock returns) in commercial banks listed on the Tehran Stock Exchange but has no effect in Iraq. This finding is consistent with the study by Richardson et al. [15]. The results of the second hypothesis test showed that accounting fraud has a positive impact on stock price crash risk (low-to-high volatility) in commercial banks listed on the Tehran Stock Exchange but has no effect on the Iraq Stock Exchange. This finding is also consistent with the study by Richardson et al. [15]. Regarding the rejection of the hypotheses in Iraq, it can be said that in Iraq, due to specific economic and political conditions, investors may be more influenced by macroeconomic factors, such as political instability, sanctions, or fluctuations in oil prices (as the country's main source of income), rather than accounting issues. This could cause accounting fraud to have a lesser impact on investor decision-making and, consequently, on the risk of a stock price crash.

Based on the results, investors can better comprehend the importance of a company's information environment, as it appears to be a channel for withholding bad news and the risk of future stock price crashes. This knowledge allows investors to assess the power of a company's information environment when making investment decisions. The findings of this research also help policymakers in two ways. First, policymakers can use our results to develop additional guidelines to improve the information environment and governance mechanisms of companies, promoting transparency and reducing the risk of stock price crashes in the market. Second, our evidence on the market consequences of accounting fraud directly informs policymakers about the level of legal penalties required when drafting laws and regulations to prevent future corporate fraud.

It is suggested to future researchers to examine the impact of corporate governance mechanisms as a supervisory tool on the relationship between accounting fraud and stock price crash risk in future studies. In addition, they should investigate the relationship between accounting fraud and stock price crash risk in light of Environmental, Social and Governance (ESG) disclosures.

1. Jensen, M.C. and Meckling W.H. “Theory of the firm: managerial behavior, agency cost and ownership structure.” Journal of Financial Economics, vol. 3, no. 4, 1976, pp. 305-60.

2. Prasmaulida, S. “Financial statement fraud detection using perspective of fraud triangle adopted by SAS No. 99.” Asia Pacific Fraud Journal, vol. 1, no. 2, 2016, pp. 317-35.

3. Chang, X. et al. “Stock liquidity and stock price crash risk.” Journal of Financial and Quantitative Analysis, vol. 52, no. 4, 2017, pp. 1605-37.

4. Habib, A. et al. “Stock price crash risk: a review of the empirical literature.” Accounting and Finance, vol. 58, 2018, pp. 211-51.

5. Hutton, A.P. et al. “Opaque financial reports, R2 and crash risk.” Journal of Financial Economics, vol. 94, no. 1, 2009, pp. 67-86.

6. Karpoff, J.M. et al. “The consequences to managers for financial misrepresentation.” Journal of Financial Economics, vol. 88, no. 2, 2008, pp. 193-215.

7. Fama, E.F. and Jensen M.C. “Separation of ownership and control.” The Journal of Law and Economics, vol. 26, no. 2, 1983, pp. 301-25.

8. Kothari, S.P. et al. “Do managers withhold bad news?” Journal of Accounting Research, vol. 47, no. 1, 2009, pp. 241-76.

9. Jin, L. and Myers, S.C. “R2 around the world: new theory and new tests.” Journal of Financial Economics, vol. 79, no. 2, 2006, pp. 257-92.

10. Karpoff, J.M. and Lott, J.R. Jr. “The reputational penalty firms bear from committing criminal fraud.” The Journal of Law and Economics, vol. 36, no. 2, 1993, pp. 757-802.

11. Ali, W. et al. “Micro-, meso- and macro-level determinants of stock price crash risk: a systematic survey of literature.” Management Finance, vol. 48, no. 5, 2022, pp. 784-804.

12. Ho, K.C. et al. “Modern health pandemic crises and stock price crash risk.” Economic Analysis and Policy, vol. 74, 2022, pp. 448-63.

13. Farber, D.B. “Restoring trust after fraud: Does corporate governance matter?” The Accounting Review, vol. 80, no. 2, 2005, pp. 539-61.

14. Wang, X. et al. “Local FinTech development and stock price crash risk.” Finance Research Letters, vol. 53, 2023, 103644.

15. Richardson, G. “The effect of accounting fraud on future stock price crash risk.” Economic Modelling, vol. 117, 2022, 106072.

16. Davidson, R.H. Income statement fraud and balance sheet fraud: different frauds, different incentives. Unpublished manuscript, Georgetown University, Washington DC, 2016.

17. Isayed, M.S.H. et al. “The impact of firm characteristics on future stock price crash risk: evidence from Egypt.” Scientific Journal of Financial and Administrative Studies and Research, Faculty of Commerce - Sadat City University, vol. 16, no. 1, 2024, pp. 740-70.

18. Xu, W. et al. “Business strategy and stock price crash risk: international evidence.” Applied Economics, vol. 55, no. 10, 2023, pp. 1098-113.