The Effect of Accounting Accruals on Real Earnings Management: Evidence from the Algerian Companies

Accounting accruals are widely used in measuring the accounting earnings management although the literature showed a negative relationship between the accounting earnings management and the real earnings management. This study explored the effects of the accounting accruals on real earnings management in some of the Algerian companies. The study included a sample of 100 firm-year observations that concern 20 companies during the period 2015 – 2019. The hypothesis testing was based on a leaner regression model that relates real earnings management proxy with accounting accruals. The results indicated a statistically significant negative effect of accounting accruals on real earnings management, which is consistent with the hypothesis and confirmed the many previous studies on the negative relationship between accounting earnings management and real earnings management. The results also provide evidence on the behaviour of managers where many of whom resort to real earnings management in case of non-conformity with the accounting earnings management. The study provides confirmation to users in the real capture of the financial statements quality as well as auditors when certificating financial statements. Users of financial statements in the Algerian companies must carefully use accounting information and employ more than one accounting items when making decisions. Meanwhile, auditors are recommended to focus more on accounting accruals when examining and certificating financial statements.


Introduction
The general objective of financial reporting is to provide useful financial information to users. As such, the financial information quality is the focus of conceptual framework for financial reporting (IASB, 2018). İn addition, the different components of the financial reporting framework have been directed towards the same objective (Kimouche & Charchafa, 2020, pp. 407-408). İn the same manner, managers, auditors, and different parties related to the reporting entity focus on the financial information quality. This quality can be measured through the many indicators developed by accounting and finance literature. İn fact, earnings management was among the widely used measurement indicators (Kimouche & Cherroun, 2020, p. 484).
Earnings management reflects the accounting decisions of managers in terms of selecting and applying accounting policies (accounting earnings management), and real decisions in terms of resources allocation related to operating, financing, and investment activities (real earnings management). Meanwhile, the accrual-based earnings management aims to obscure true economic performance by changing accounting methods or estimates within the generally accepted accounting principles. Real earnings management alters the execution of real business transactions.
Many studies already tackled the different factors affecting earnings management and explained the disparity between companies in terms of earnings management level and the tendency of managers toward different strategies of earnings management. The majority of these studies revealed that the accounting and real earnings management are the more commonly used strategies by managers which are affected by many factors, especially those that relate to the companies and management characteristics. Other studies followed through by investigating the relationship between earnings management and real earnings management to determine whether one of the earnings management practices affects the other as substitutes, complementary or interchangeably. Results of these studies depicted the managers' behavior toward accounting and determined the role of accounting policies when compared with operating, financing, and investment policies.
The present study further explored the relationship between earnings management and real earnings by employing the same concepts with the Algerian companies. İt answers whether the accounting accruals, as a proxy of accounting earnings management, affect the

Theoretical framework
Over the years, several authors and researchers have expressed the concern on earnings management. In fact, the definition alone expresses a negative connotation. As defined by Schipper (1989, p. 92), earnings management is a "purposeful intervention in the external financial reporting process with the intent of obtaining some private gain".
Relatively, Healy and Wahlen (1999, p. 368) added that "earnings management occurs when managers use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers". Since the middle of the 1980s, studies about the managerial incentives to manipulate earnings were primarily based on accounting accruals. For instance, Beneish (2001) traced an explosive growth in accrual-based earnings management studies since accruals are the principal product of the Generally Accepted Accounting Principles (GAAP).
The accrual-based earnings management that is referred also in the literature as "accounting earnings management", is easy to implement because it is the result of accounting decisions related to the selection of accounting policies and the estimations required to apply these accounting policies. However, it is also easy to discover. It includes some techniques conducted in accordance with the accounting rules and principles through the selection of appropriate accounting policies and the use of estimations and judgments in the application of those accounting policies in order to control the level of earnings (Kimouche and Cherroun, 2020).
Several studies argued that the passage of the Sarbanes-Oxley Act in 2002 to increase the transparency and internal control also led to the practice of the real earnings management as an alternative to accrual-based earnings management. In the accounting practice, the real earnings management is within the accounting standards, which according to Cohen et al. (2008) malpractices are difficult to discover by auditors because these activities can be considered legal in response to economic conditions. Cohen et al. (2008) and Liu et al. (2011) added that the introduction of the new corporate governance practices and the international standardization led managers to shift from accounting to real earning │ 5 https://iiari.org/journals/ijafe management because it is less exposed to the scrutiny of auditors and regulators. Hence, managers could prefer real earnings management to accrual earnings management (Roychowdhury, 2006) According to Janin (2000), the real earnings management involves real business activities with direct impact on the operating cash flows. It includes all the actions set by the managers in order to deviate from the normal business practices to meet target earnings. This impact of real earnings management on the performance emanates from the cash flows and not from the accruals, as in the case of accounting earnings management. On the other hand, the accounting accruals are the total adjustments on cash flows permitted by accounting standards (Healy, 1985) through which current cash flow is modified to create a more predictive performance measure, namely earnings (Barth et al., 2016, p. 772). The accruals include items that allow the transition from cash accounting to accrual accounting and transactions that affect the earnings schema over time, permitting managers to transfer results between different periods. It appears when the timing of transaction recognition differs from the timing of cash receipts or payments. As a result, managers use accruals to alter the timing of the cash flows recognition in earnings to enhance performance measurement.
Accounting accruals arise primarily from the accrual basis, which requires the allocation of amounts arising from transactions to more than one period. They are also the results of other accounting principles related to the recognition and measurement, such as conservatism, time period, recognition, and matching. The implementation of these principles requires the general use of judgment and estimation to recognize each transaction to the period in which they occur and allocate financial flows over periods, if they concern more than one period. Accruals represent the non-cash earnings, including the non-paid expenses and non-collected income and other accounting adjustments (Kimouche, 2020(Kimouche, , p. 1417).

Previous studies
There is an extensive literature about earnings management in the last four decades.
Many studies had examined management's choice of accounting methods while others on the accrual management (Wali, 2017, p 378). However, most studies tried to develop measures for earnings management or investigate its techniques and determinants. Therefore, studying 6 │ the relationship between different techniques of earnings management is a new approach recently introduced in accounting. Zang (2012) analyzed the managers' use of real activities manipulation and accrualbased earnings management as substitutes in managing earnings. The data were gathered from the 820 industry-years in the CRSP/Compustat during the period 1987 to 2008. The results indicated that managers trade off the two earnings management methods based on their relative costs and that managers adjust the level of accrual-based earnings management according to the level of real activities manipulation realized. The study documented also that large-sample evidence consistent with managers using real activities manipulation and accrual-based earnings management as substitutes.
Subekti (2012)  The previously summarized studies focused on the nature of the relationship between accounting and real earnings management to understand whether managers used them as substitutes, complementary or interchangeably for the purpose of earnings management.
They compared also accounting and real earnings management in terms of some affecting factors. The present study adopted the same approach while it focused on the effect of accounting accruals on real earnings management. As it was carried out in Algeria, a developing country, the current study can add additional value and comparison to the studies that were carried out in the developed countries.

Model specification
The study model contains an equation in the form of a simple linear regression, which links the real earnings management with the total accounting accruals as shown in equation

REMit = α +βTACCit + ξit
(1) Where: REMit: is the real earnings management for the company i during the period t.
TACCit: is the total accounting accruals for the company i during the period t.
α: is a constant.
β: is the regression coefficient.
ξit: is the error term.

Sample
The study included 20 companies during the period 2015 to 2019 where an unbalanced panel data of 100 firm-year observations has been obtained. The selection of the companies was based on the availability of their financial information because the corporate governance in the Algerian companies is characterized by secrecy and caution.

Real earnings management measurement
The measure of real earnings management is based on the calculations given by Roychowdhury (2006) and using the model of abnormal cash flows followed after Dechow et al. (1998). As shown in equation (2), the model expresses the operating cash flows as a function of sales and change in sales.
At-1: is the total assets for the company i at the end of period t-1.
Sit: is the sales for the company i during the period t.
The residuals of equation 2 represent the abnormal cash flows from operations. Cashbased (real) earnings management is the actual cash flows from operations minus the normal cash flows from operations calculated using the estimated coefficients (α0, α1, β1, and β2).

Total accounting accruals measurement
The calculation of the total accounting accruals was based on equation 3. Where: ΔWCNit: is the variation of working capital needs for the company i during the period t.
CPit: is the non-cash expenses for the company i during the period t.
DOTit: is the amortization and impairment expenses for the company i during the period t. The results of the equation 2 as summarized in Table 1 indicate that it is statistically significant at 1% level. Relatively, the explanatory power of sales and change in sales reaches 23%, which means that 23% of the cash flows are normal and can be expected using 12 │ the sales. However, 77% of the cash flows are abnormal and they are not expectable using the sales. This clearly indicates that managers have a maneuver margin of 77% cash flows to practice real earnings management. The regression coefficients of sales and change in sales are statistically significant at 1% level. These suggest the positive and negative effects of the sales and the change in sales respectively on the operating cash flows.  Table 2 presents the descriptive statistics for 20 companies during the 5-year period 2015 to 2020, which represents the 100 firm-year observations. The results show that the mean of both the real earnings management and the total accounting accruals are negative and are relatively very close. It clearly implies that the two variables are high dispersal with the accounting accruals as more dispersal as indicated by the standards deviation. Moreover, the maximum and minimum values indicate that the two variables contain positive and negative values. Table 3 shows the results of the correlations between the real earnings management and the total accounting accruals. The results suggest a statistically significant and very high negative correlation between the two variables and at 1% level. The correlation reaches more than 83% that provides a primary evidence on the negative effect of the total accounting accruals on the real earnings management as presumed in the hypothesis.

Model selection
Before testing the hypothesis, the appropriate and valid model must be selected. The selection process begins with the comparison between Pooled Regression Model and Fixed Effects Model using the restricted F-test. It is then followed by the comparison between Fixed Effects Model and Random Effects Model using the Hausman test. The restricted Ftest was employed to test the following hypotheses: H0: Pooled Regression Model is valid.

H1: Fixed Effects Model or Random Effects Model is valid.
In applying the restricted F-test, it needs the calculation of the F value (F') at degrees of freedom (N -1) and (NT -N -K) as shown in the equation 3. This is followed by the comparison between the calculated F and the F critical value obtained from the F-distribution table. If the calculated F is more than F critical value at 5% level of significance, H0 must be rejected, and vice versa.     (2012) and Achleitner et al. (2014) that the managers use the real activities manipulation and accrual-based earnings management as substitutes.

Conclusion and recommendation
Earnings management is a determinant attribute of financial accounting quality. The earnings management practices can distort the desired characteristics of the financial statements when used impulsively while it can also improve the presentation of financial statements when used moderately. The accounting earnings management and the real earnings management are the widely used practices to manage earnings. The former is based on the accounting policies and estimations while the latter on the operating, investment, and financing decisions.
The present study analyzed the relationship between the two widely used practices in managing earnings. It specifically explored the effect of accounting accruals, as a proxy to the accounting earnings management, on the real earnings management in the Algerian companies. The study included a sample of 100 firm-year observations that concern 20 companies during the period 2015 to 2019. The hypothesis testing was based on a leaner regression model that relates the real earnings management proxy with total accounting accruals. The results confirm the hypothesis that the total accounting accruals negatively affects the level of real earnings management in the Algerian companies. The results of the current study conform with the the assumptions of the previous studies about the negative relationship between accounting earnings management and real earnings management.
The results imply that the users of financial statements of the Algerian companies must use the earnings prudently when making their decisions. Additionally, they should use other accounting items that improve the earnings quality. Auditors must give more attention to the earnings in order to improve their quality and gain users' confidence. The results also imply the use of more than one proxy for accounting quality because these may provide results in the different trends.
With the explicit importance of this study and its practical implications, it is hereby necessary that future studies use different approaches to measure real earnings management.
The application of various approaches provide relevant and imperative comparisons. Further students can also extend the number of samples and examine the impact of other factors on the real earnings management especially the governance and the company characteristics.