PRICING OF DISCRETIONARY ACCRUALS, STOCK RETURNS AND FIRMS GROWTH: EVIDENCE FROM IRAN
1 Assistant Professor in accounting, university of Kurdistan, Kurdistan, Iran
2Post Graduate Student in Accounting, Azad University, Sanandaj branch, Iran
ABSTRACT
In this study, the relationship between discretionary accruals and stock returns has been investigated by considering the growth variable in companies listed in Tehran Stock Exchanges. The study has been done using the data from 112 companies during 2010-2015 using systematic elimination sampling. For this purpose, three theories have been developed and three types of tests F, t and multivariate regression analysis were used to examine the hypothesis. The results of hypotheses indicate no significant relationship between discretionary accruals and stock returns, but with the addition of firm's growth, the use of discretionary accruals and positive discretionary accruals is more in firms with high growth than firms with low growth.
Keywords:Discretionary accruals, Positive discretionary accruals, Stock returns, Firm's growth.
ARTICLE HISTORY: Received: 17 July 2017, Revised: 21 August 2017, Accepted: 30 August 2017, Published: 11 September 2017
Contribution/ Originality: This study is one of very few studies which have investigated relationship between pricing of discretionary accruals and stock in the case of Iran.
Based on the theoretical concepts of financial reporting, the primary purpose of financial statements is to provide information about the financial situation, financial performance and financial flexibility of business entity to assist the users of financial statements in making economic decisions. Meanwhile, measuring the firm performance and identifying the factors affecting it is important. Out of the performance measurement criteria is the stock returns which is of a special importance, because it shows the amount of management efficacy in using the available resources to gain profit. Earnings are one of the most important items affecting the decision of users of financial statements. The item of financial statements (earnings) can be investigated from different aspects including different classifications in the profit and loss statement in terms of operational and non-operational, frequent and infrequent and/or evaluating the cash and accruals of benefits. Since the possibility of earnings repeatability affects the investors' views toward purchase or sale of securities or maintaining them. One of the factors that the continuity of earnings depends on it is the amount of cash and earnings accruals. Extensive researches have been done in both inside and outside of the country in the context of cash and accrual components of earnings. Sloan and Richard [1] states that the items of accruals with low reliability result in less continuity of earnings, while generally the investors do not expect the less continuity of earnings. Since based on the accepted accounting principles the managers have relative freedom of action about the time of identifying the revenues and expenses, the accruals are the result of aggressive actions of management in identifying, record and reporting the events and the achievement of the executives in distorted financial statements. In choosing between different methods of taking account the financial events within the framework of generally accepted accounting principles, the management has considerable authorities and can schedule the financial activities and delay some sale items and/or the expenditures. This type of manipulation requires that the firm gain much earnings to be able to provide the necessary resources for ordering the flows when needed. Accruals are more easily subjected to change and manipulated than cash flows, thus one of the ways by which the corporate governance can have an important effect on the stability of earnings and quality of accounting accruals is that the strong mechanisms of corporate governance should reduce the incentives or opportunities related to earnings management and thus result in the reduced level of errors in accruals and increased earnings stability [2]. It has been argued that a part of accruals is created in the ordinary procedure of business entity's activity (non-discretional accruals) and only a portion of these items is subjected to the manipulation of management (discretionary accruals). Therefore it seems that the drawback attributed to the accruals accounting system is caused by the latter part. Accordingly it is expected that firms with a high level of accruals incur higher costs to finance and subsequently encounter reduced stock returns in the future. Jones [3] defined the accruals the difference between earnings and cash gained from operation. In fact accruals are said to items that cause the delaying of recording the revenues and expenditures. Sloan and Richard [1] considered the accruals as the changes in working capital to depreciation charges. This definition of accruals does not include the non-current operating assets, non-current operating liabilities, non-cash financial assets and liabilities. Richardson, et al. [4] have provided a comprehensive definition of accruals. They defined the accruals as the difference between accruals earnings and cash earnings. This definition also covers the accruals related to non-current operating assets and liabilities (such as long-term receivable account).
Two bases can be used for the valuation of equity. The first is the ratio of price to earnings and the other is the abnormal earnings of the firm. Abnormal stock returns, based on market data and the ratio of price to earnings, is a measure consisting of the valuation model based on accounting data and the capital market-based model. The common aspect of both bases of price to earnings and abnormal returns, is the stock market value of firms that is at a high level. High ratio of price to earnings in firms represents the market optimism towards the good growth of the firm in the future. In other words, investors with the thought of firms' better performance in the future evaluate the shares and investors to be high, and the unusual high returns indicate that investors expect the repetition of the current opportunities in the future. Therefore, they buy shares with the hope of more returns in the future [5]. And considering the important role of capital markets in the economy of any country, it is important that investors be aware of firm's financial information to cause a good investment and optimum allocation of resources in the capital market. One of the important features of earnings is the earnings quality. Earnings' quality is an important measure of the financial health of business entity and is a multidimensional concept and has no single meaning [6]. Earning's quality means to approaching the cash, ability of growth and continuity [7]. Since the managers have relative freedom of action about the time of recognizing the income and expenses, some believe that accruals can be manipulated by the managers (earnings management) to align the earnings with personal desires [8]. However, understanding the process of identifying those accruals that affect the earnings quality is difficult for the investors and inhibits the precise distinguishing of firm's earnings by them and leads to the incorrect investment and loss of resources. Therefore, it is important to consider the accruals and its components as an index to evaluate the earnings quality [9].
Applying the management view in determining the reported earnings affects the earnings quality and reduces it. Managers adjust the reported earnings by selecting the certain accounting policies, accounting estimates and accruals management. According to Barth, et al. [10] using the estimates in the financial statements affects the earnings and may result in earnings different from the result of actual performance of the firm. Thus, reported earnings are away from the actual earnings and will not have enough objective. In the process of measuring the earnings the accruals accounting system is used in which the accruals adjust the recognition of the cash flow during the period so that the adjusted figures measure the performance of economic units better. Above accruals are based on the estimates that if they are wrong should be modified during the next periods in the future accruals and future earnings. Estimation error and its subsequent amendments, reduces the effectiveness of accruals (accruals' quality). According to Dechow and Dichev [11] accruals quality is reduced with the increased estimation error of accruals. Since the accruals somehow adjust the cash flows and it's sum is zero in the lifetime of an firm, high accruals during a period reverse the accruals in the future periods and considering that the accounting earnings consists of two cash and accruals parts, subsequently the future earnings will also decrease with the reversed accruals Zhang [12]. Richardson, et al. [4] defined the accruals as the difference between accruals earnings and cash earnings. According to Bernard, et al. [13] the accruals anomaly can be divided into two categories:
1 –Risk-based accruals anomaly: The relationship of variables in addition to systematic risk such as the size and the ratio of book value to market value with the stock returns (capital market anomaly) can be caused by lack of complete identification of risk factors Fama and French [3]. Hirshleifer, et al. [14] by examining whether the accruals anomaly is based on the risk or incorrect pricing, found that this anomaly is not based on mere risk but due to the incorrect pricing of accruals.
2. Accruals anomaly based on incorrect pricing of accruals: in the modern financial-behavior theories it is assumed that investors systematically make mistake in pricing the assets and the anomalies are in the result of pricing error [12]. In other words, the presence of inexperienced investors in the market, cause the incorrect pricing of securities; so that the incorrect reaction of investors to changes in accruals make the stock of companies with a different amount of accruals to incorrectly valuated more or less than real that is contrary to the efficient market hypothesis [15].
Chu [16] in her research entitled by accruals, growth and future performance of the firm investigated the role of accruals and growth and cash flows in stability of earnings and future returns in the capital market of America for a period of eleven years. The results show that accruals have inverse relationship with the future return of assets and abnormal stock returns. Papanastasopoulos, et al. [17] in their study examined the relationship between accruals and external financing activities with the future stock returns. Results of their study show that portfolios with high accruals and high volume of external financing activities earn lower returns. Also, the cross-sectional regression results at the firm level indicate that in the presence of working capital accruals, external financing activities have the ability to predict the future returns. Sullivan and Zhang [18] examined the relationship between investment activities and financing activities with future stock returns. Results of their study show that there is no relationship between investment activities anomaly and financing activities anomaly. In addition, the relationship between stock returns and investment activities is weakened after controlling for financing activities, while the relation between stock returns and financing activities remain significant after controlling for investment activities.
Fedyk, et al. [19] in a research examined the tested the stability hypothesis. According to this hypothesis, accruals anomaly based on the incorrect pricing is caused by the focus of investors on earnings regardless of reversible nature of accruals, and with the returned accruals in future period, incorrect pricing of the market is corrected. The study results show that there is a negative relationship between accruals and future stock returns and with the returned discretionary accruals in future periods, this negative relationship is removed.
Wu and Zhang [20] in a study examined the issue of whether the capital market anomalies are caused by the risk factor or incorrect pricing. In this study, they considered variables such as the ratio of book value to market value, size, capital distribution, growth of assets, accruals and the probability of bankruptcy. The results showed that additional returns which caused by these variables, is due to the incorrect pricing of these variables by investors not because of risk factor. Chen [21] in his study seeks to investigate whether the accruals explain the abnormal returns better or cash flows. His research results show that due to the investment duration of cash flows or accruals, each one can report the abnormal returns better. In a long-term investment period, accruals may be a better indicator. But in a short-term investment period, cash flows play the major role.
Resutek [15] has examined the anomaly of accruals and intangible effect of returns on this anomaly. The results of his study show that there is an inverse relationship between accruals and stock returns. Also, by including the intangible returns variable of prior periods into the regression mode, the negative relationship between accruals and stock returns is removed. In other words, intangible returns affect the anomaly of accruals.
This study in terms of type or nature is applied one and of the post-events researches which based on the previous fiscal data examines the hypotheses. The basic information of this research include the audited financial statements data and information related to stock returns, earnings quality, accruals components and earnings cash components that referring to the databases, derived, calculated and the research models will be estimated based on them to provide a basis for hypotheses test. The statistical population of the study is consisted of the companies listen in Tehran Stock Exchange. The study time period was determined taking into account the information close to study time and availability of related information in a period of 8 years from the beginning of 2006 until the end of 2013. The sample study consisted of companies which have been derived based on the following conditions and characteristics of the mentioned statistical population.
Based on the above criteria, 112 companies were identified. Then, all of the information needed for this research was collected from databases of Tehran Stock Exchange, RAHAVARD Novin and Tadbirpardaz softwares and stock publications. It should be noted that the statistical sample in this study was the same statistical population.
In order to calculate the stock return, the stock price at the end of year was subtracted the stock price at the beginning of the year and add the annual dividends and then divide the obtained number by the stock price at the beginning of the year.
In this study, Kaszink [22] model was used. And in order to compatibility of study results with the environment of Iran, the variable of changes in the inventory was added to the model. The main reason to use the Kaznik model was that it has been suggested that this model is more consistent with accounting environment of Iran.
To obtain the time series criteria of accruals for a specific firm, the above equation is fitted for a specific time periods and each estimate leads to remaining of year-firm. In estimates of time series, regression residuals are used to calculate the accruals index and the accruals quality is the standard deviation of estimated residuals of the sample firm.
To obtain the discretionary accruals according to the Kaznik model shown below, the model remaining of model is used, that is, the regression remaining items are discretionary accruals and if the obtained regression remaining is a positive, our discretionary accruals is positive and if the regression remaining items is negative then the discretionary accruals is negative.
Firm growth
At the end of each fiscal year, the total sales of the companies were used to calculate the sales growth. In general, the sales growth rate of firm i is:
In order to test the three hypotheses, we use the following models respectively:
First hypothesis:
Before the first hypothesis test, we selected an appropriate model for regression model. First, using the F-Limer test an integrated data model against combined data model was selected. F Limer statistical probability value is less than 5% significance level and thus for the first hypothesis test, using the method of pooled data is dismissed. Because of not selecting the pooled data model against panel data, Hausman test was performed in order to select the constant effects model against the random effects model. Probability of Hausman statistic is higher than 5% significance level. Therefore, there is no enough reason to reject the random effects model, and the random effects model is used for the first hypothesis test. Combined regression model is not the random effects of discretionary accruals on positive stock returns (0.197) and considering the probability the statistic t (0.5268) is not significance. This suggests that discretionary accruals have no significant effect on stock returns, and investors do not pay attention to the accruals and obtain their information from other sources. Also, firm size, the uniqueness of assets, the ability to generate the internal resources and current ratio, non-debt tax shield and collateral value of assets have no impact on stock returns.
F-test results indicate that the model is generally significant and according to Durbin- Watson statistics (1.82) has not correlation problem. One way that is performed to check the collinearity is to calculate the variance inflation factor. The reported variance inflation factor is less than 5 and more than 5, which indicates that there is no collinearity between the independent variables and there is no need to conversion (performing the natural logarithm of variables). In addition, the results of adjusted coefficient determination show that in the entire study period about 03% of variations in stock returns is caused by discretionary accruals and control variables of the study. Due to the lack of a significant effect of discretionary accruals on stock returns the first hypothesis is rejected and not confirmed.
Table-1. Panel regression model of random effects related to the effect of discretionary accruals on stock returns
Rit = β0 + β1 DAit + β2 Sizeit + β3 CVAit + β4 NDTSit + β5 UNIQit + β6 CGIRit + β7 CRit + εit | ||||
Variables | Regression coefficients |
t-statistic value |
t-statistic probability |
Value (VIF) |
Constant value | 0.293 |
0.174 |
0.8617 |
1.4832 |
Discretionary value | 0.197 |
0.633 |
0.5268 |
1.2429 |
Firm size | -0.039 |
-1.019 |
0.3028 |
1.3381 |
Collateral value of assets | 0.8100 |
0.473 |
0.6361 |
1.4822 |
Non-debt tax shield | -0.021 |
-0.123 |
0.9021 |
1.6736 |
Uniqueness of assets | -0.148 |
-0.555 |
0.5784 |
1.4984 |
Ability to generate the internal resources | -0.0033 |
0.377 |
0.7059 |
1.5847 |
Current ratio | 0.1818 |
0.174 |
0.3206 |
1.6183 |
Determination coefficient | 0.004 |
F-statistical probability |
0.000 |
- |
Adjusted determination coefficient | 0.003 |
Durbin- Watson statistic |
1.82 |
- |
Source: Researchers’ findings
Before testing the second hypothesis, an appropriate model was selected for the regression model. First, using the F-Limer test, pooled data model was selected against the combined data model. Probability of F-Limer statistics in the table was less than 5% significance, therefore, using the pooled data for the second hypothesis test is dismissed. Because of not selecting the pooled data model against the panel data, Hausman test was performed in order to select the constant effects model against the random effects model. Probability of Hausman statistic is less than 5% significance therefore, there is no enough reasons to reject the constant effects model and, the constant effects model is used to test the second hypothesis. Combined regression model of constant effects related to the effect of discretionary accruals on stock returns in firms with high growth than firms with low growth is positive (1.235) and also significant considering the probability of statistic t (0.0005). This suggests that the combination of firm's growth and accruals would result in a significant effect on the stock returns. While, the impact of discretionary accruals alone has no effect on stock returns.
Also, the effect of firm size and current ratio on stock returns is positive and significant on stock returns. Uniqueness of the assets, the collateral value of assets have a negative and significant impact on stock returns and the ability to generate the internal resources and non-debt tax shield have no effect on stock returns.
F-test results indicate that the model in general is significant and according to the Durbin- Watson statistic (2.07) has not the correlation problem. One way to check the collinearity is to calculate the variance inflation factor. Reported variance inflation factor is less than 5 and more than one, which shows that there is no collinearity between the independent variables and there is no need for conversion (performing the natural logarithm of variables). In addition, the results of the adjusted coefficient of determination show that in the entire study period, about 16% of the variations in stock returns is caused by the discretionary accruals of firms with high growth and control variables. Due to the significant effect of discretionary accruals of firms with higher growth than firms with low growth, the second hypothesis is confirmed.
Table-2. Panel regression model of random effects related to the effect of discretionary accruals of firms with higher growth than firms with lower growth on stock returns
Rit = β0 + β1 DAit +β2DAit * SGRit + β3 Sizeit + β4 CVAit + β5 NDTSit + β6 UNIQit + β7 CGIRit + β8 CRit + εit | ||||
Variables | Regression Cofficients |
t-statistic value |
t-statistic probability |
Value (VIF) |
Constant value | 3.563 |
2.709 |
0.069 |
1.5408 |
Discretionary accruals | -0.434 |
-1.73 |
0.083 |
1.5725 |
Accruals and firm's growth | 1.235 |
3.499 |
0.0005 |
1.1567 |
Firm size | 0.065 |
1.693 |
0.0908 |
1.3789 |
Collateral value of assets | -3.905 |
-3.172 |
0.0016 |
1.4893 |
Non-debt tax shield | -0.0705 |
-0.460 |
0.6453 |
1.1052 |
Uniqueness of assets | -0.38 |
-2.082 |
0.0377 |
1.1955 |
Ability to generate the internal resources | 3.91 |
0.008 |
0.993 |
1.1827 |
Current Ratio | 0.1804 |
2.348 |
0.0192 |
1.7927 |
Determination Coefficient | 0.29 |
F-statistic probability |
0.0000 |
- |
Adjusted determination coefficient | 0.16 |
Durbin- Watson statistic |
2.08 |
- |
Source: Researchers’ findings
Before testing the third hypothesis, an appropriate model was selected for the regression model. First, using the F-Limer test, pooled data model was selected against the panel data model. Probability of F-Limer statistics was less than 5% significance, and therefore, using the pooled data for the third hypothesis test is dismissed. Because of not selecting the pooled data model against the panel data, Hausman test was performed in order to select the constant effects model against the random effects model. Probability of Hausman statistic is less than 5% significance therefore, there is no enough reasons to reject the constant effects model and, the constant effects model is used to test the third hypothesis. Combined regression model of constant effects related to the effect of positive discretionary accruals on stock returns in firms with higher growth than firms with low growth is negative (1.143) and also significant considering the probability of statistic t (0.0052). This suggests that the combination of discretionary accruals and firms' growth has a significant negative effect on the stock returns. Also, firm size, uniqueness of assets, ability to generate the internal resources, non-debt tax shield and current ratio have no impact on stock returns, while the collateral value of assets has a significant impact on stock returns.
F-test results indicate that the model in general is significant and according to the Durbin- Watson statistic (2.06) has not the correlation problem.
One way to check the collinearity is to calculate the variance inflation factor. Reported variance inflation factor is less than 5 and more than one, which shows that there is no collinearity between the independent variables and there is no need for conversion (performing the natural logarithm of variables).
In addition, the results of the adjusted coefficient of determination show that in the entire study period, about 20% of the variations in stock returns is caused by the positive discretionary accruals of firms with higher growth than firms with low growth and control variables. Due to the significant impact of positive discretionary accruals on stock returns of firms with higher growth than firms with lower growth, the third hypothesis is confirmed.
Table-3. Panel regression model of random effects related to the effect of positive discretionary accruals of firms with higher growth than firms with lower growth on stock returns
Rit = β0 + β1 PDAit + β2 PDAit * SGRit + β3 Sizeit + β4 CVAit + β5 NDTSit + β6 UNIQit + β7 CGIRit + β8 CRit + εit | ||||
Variables | Regression coefficients |
t-statistic value |
t-statistic probability |
Value (VIF) |
Constant value | 7.77 |
4.413 |
0.000 |
1.1942 |
Positive discretionary accruals | 0.456 |
1.307 |
0.1919 |
1.1393 |
Positive accruals and firm's growth | -1.143 |
-2.81 |
0.0052 |
1.3722 |
Firm size | 0.014 |
0.185 |
0.8527 |
1.4751 |
Collateral value of assets | -7.508 |
-3.124 |
0.0019 |
1.2783 |
Non-debt tax shield | -0.17 |
-0.922 |
0.3571 |
1.3048 |
Uniqueness of assets | -0.81 |
-1.316 |
0.1890 |
1.7483 |
Ability to generate the internal resources | 0.002 |
0.302 |
0.7626 |
1.1089 |
Current ratio | 0.432 |
1.684 |
0.0931 |
1.5172 |
Determination coefficient | 0.33 |
F-statistic probability |
0.0000 |
- |
Adjusted determination coefficient | 0.20 |
Durbin- Watson statistic |
2.06 |
- |
Source: Researchers’ findings
In general, it can be said that managers tend to use the discretionary accruals especially positive discretionary accruals (increases the income) to send the desired information to the outside investors. In fact, in general no significant difference was observed in the use of discretionary accruals and between firms with high and low growth as well as in the use of the positive discretionary accruals. However, further tests showed that discretionary accruals especially positive discretionary accruals are relatively more associated with the performance of firms with high growth. According to the results of Chu [16]; Papanastasopoulos, et al. [17]; Fedyk, et al. [19]; Resutek [15] shows the inverse relationship between accruals and future return on assets and abnormal stock returns, And high accruals portfolios with a high volume of external financing activities, earn lower returns can be stated that the present study has been similar results. Finally, managers in order to enhance the accuracy of predictions, use the two factors of accruals and optional changes in accounting more than other factors. Another important issue in accounting research is the role of accounting earnings in the assessment of securities. Since the earnings is associated with stock returns of firms, there are several propositions in relation to the usefulness of earnings accrual components to evaluate the firm. On the one hand, the Financial Accounting Standards Board believes that accruals are useful for stock assessment. On the other hand, some financial analysts questioned the reliability and relevance of earnings due to its accruals components. They argued that managers tend to manipulate the accruals by using different allowed methods for earnings adjustment. However, in financial decision making process, the role and importance of cash flows is so obvious that without it, the decisions are inefficient or will have some problems in implementing. According to the research findings, investors are recommended to examine the accruals quality and consider in their decision models. And to earn high returns in the future, they should invest in firms that have low accruals and high operating cash flows. Also, it is recommended that investors react more accurately to changes in accruals, and get help from financial analysis in making investment decision.
According to the results of the study suggested that:
1- According to the findings indicating that there is a significant positive correlation between accruals and corporate growth characteristics It is suggested that when the investment decision, accruals considered as a measure of a company's growth.
2- According to the findings regarding the presence of inexperienced investors in the capital market, It is recommended that investors react more precisely to changes in the accrual and use of financial analysts in investment decisions.
3- Requiring companies listed on the stock exchange to provide timely and updated information on accruals separately In the form of computer written reports and financial statements are summarized that can be widely effective in decision-making.
4- To the Audit organization which is Official reference of accounting standards development in the country, it is recommended that When formulating accounting standards ,pay attention to maintain balance between information qualitative characteristics . The inclusion of accounting information less provable, and therefore, less reliable, leads to increase the measurement error and reduce the quality of the profit that is basic criteria of performance evaluation.
In the process of doing scientific research, there is a set of conditions and situations that are beyond the control of the investigator. But can potentially affect research results. This study has been no exception; including restrictions governing it are as follows:
1-In this study, all business units of the sample of manufacturing firms in Tehran Stock Exchange has been selected. Therefore, it is necessary to generalize the results of hypotheses testing to all firms listed in the Tehran Stock Exchange to be cautious.
2- When the realm of hypotheses of this study was the range of 1385 to 1392. Therefore, in generalizing the results to other time periods should be cautious.
3-The data that has been extracted from the financial statements have not been adjusted for the inflation. Due to the different inflation rates in the years studied if the data used were adjusted for, the results may be different from the current results.
4- Measurement error in the accrual of independent random variables were considered. In fact, it’s possible measurement error affected earnings manipulation by managers and Covenants accounting conservatism should be like.
5- The time frame of the study time interval between the years 1385 to 1392; therefore, the financial statements of companies listed on Tehran Stock Exchange have not been used outside of this time period.
6- In this study, it was assumed measurement error in the cash flows of secondary importance. New research suggests that cash flow can be manipulated by managers and Simultaneous modeling of measurement error in the cash flow and accruals can provide improved understanding of sustainable profits.
7- Volatility of stock price in some years it is possible to generalize the results to be limited.
8- Difficult access to required data and stop trading for more than three months has caused many companies to be removed from the sample.
9- Inflation leads to financial statements information fails to properly show the financial position and results of the company's operations. Therefore, taking into account the effects of inflation, different results may be achieved.
Funding: This study received no specific financial support. |
Competing Interests: The authors declare that they have no competing interests. |
Contributors/Acknowledgement: Both authors contributed equally to the conception and design of the study. |
[1] R. Sloan and G. Richard, "Do stock prices fully reflect information in accruals and cash flows about future earnings," Accounting Review, vol. 71, pp. 289-315, 1996. View at Google Scholar
[2] S. L. Gillan, "Recent development in corporate governance: An overview," Journal of Corporate Finance, vol. 12, pp. 381 – 402, 2006.View at Google Scholar | View at Publisher
[3] E. Fama and K. French, "Dissecting anomalies," Journal of Finance, vol. 63, pp. 1653–1678, 2008. View at Google Scholar | View at Publisher
[4] S. Richardson, R. Sloan, M. Soliman, and I. Tuna, "Accrual reliability, earnings persistence, and stock prices," Journal of Accounting and Economics, vol. 39, pp. 437-485, 2005. View at Google Scholar | View at Publisher
[5] E. Houmes and R. Skantz, "Highly valued equity and discretionary accruals," Journal of Business Finance & Accounting, vol. 37, pp. 60-92, 2010. View at Google Scholar | View at Publisher
[6] J. Bellovary, D. Giacomino, and M. Akers, "A review of bankruptcy studies: 1930–present," Journal of Finance Education, vol. 33, pp. 1–42, 2006.
[7] K. Schipper and L. Vincent, "Earning quality," Accounting Horizons, vol. 17, pp. 97–110, 2003.View at Google Scholar
[8] D. Thomakos and T. Wang, "Accruals and the performance of stock returns following external financing activities," British Accounting Review, vol. 43, pp. 214-229, 2011. View at Google Scholar | View at Publisher
[9] D. Hirshleifer, K. Hou, and T. S. Hong, "Accruals, cash flows, and aggregate stock returns," Journal of Financial Economics, vol. 91, pp. 389–406, 2009. View at Publisher
[10] M. E. Barth, W. H. Beaver, J. M. Hand, and W. R. Landsman, "Accruals, accounting-based valuation models, and the prediction of equity values," Journal of Accounting, Auditing, and Finance, vol. 20, pp. 311–345, 2005. View at Google Scholar | View at Publisher
[11] P. M. Dechow and I. Dichev, "The quality of accruals & earnings: The role of accrual estimation errors," Accounting Review, vol. 77, pp. 35-59, 2002. View at Google Scholar | View at Publisher
[12] X. F. Zhang, "Accruals, investment and accrual anomaly," Accounting Review, vol. 82, pp. 1336-1363, 2007. View at Google Scholar | View at Publisher
[13] V. Bernard, J. Thomas, and J. Whalen, "Accounting based stock price anomalies: Separating market inefficiencies from risk," Cotemporary Accounting Research, vol. 14, pp. 89-136, 1997.View at Google Scholar | View at Publisher
[14] D. Hirshleifer, K. Hou, and S. H. Teoh, "Accrual anomaly: Riskor mispricing?," Journal of Finance, vol. 65, pp. 1-51, 2010.
[15] R. J. Resutek, "Intangible returns, accruals, and return reversal: Amultiperiod examination of the accrual anomaly," Accounting Review, vol. 85, pp. 1347-1374, 2010. View at Google Scholar | View at Publisher
[16] J. Chu, "Accruals, growth, and future firm performance, Working Paper." Retrived from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1999685 [Accessed 18 July 2013], 2012.
[17] G. Papanastasopoulos, D. Thomakos, and T. Wang, "Accruals and the performance of stock returns following external financing activities," British Accounting Review, vol. 43, pp. 214–229, 2011. View at Google Scholar | View at Publisher
[18] M. Sullivan and A. Zhang, "Are investment and financing anomalies two sides of the same coin? ," Journal of Empirical Finance, vol. 18, pp. 616–633, 2011. View at Google Scholar | View at Publisher
[19] T. Fedyk, Z. Singer, and T. Sougiannis, "Does the accrual anomaly end when abnormal accrual sreverse?," presented at the The Canadian Academic Accounting Association Conference, 2011.
[20] J. Wu and L. Zhang, "Does risk explain anomalies ? Evidence from expected return estimates," National Bureau of Economic Research, pp. 1-46, 2011.
[21] K. Chen, "Is cash flow really king? earning quality and stock returns," Working Paper, Septamber, 2010.
[22] R. Kaszink, "On the association between voluntary disclosures and earnings management," Journal of Accounting Research, vol. 37, pp. 57-81, 1999.